Skip navigation

Abt Associates report Governments' Management of Private Prisons 2003

Download original document:
Brief thumbnail
This text is machine-read, and may contain errors. Check the original document to verify accuracy.
The author(s) shown below used Federal funds provided by the U.S.
Department of Justice and prepared the following final report:
Document Title:

Governments' Management of Private Prisons

Author(s):

Douglas McDonald ; Carl Patten

Document No.:

203968

Date Received:

January 2004

Award Number:

1998–CE–VX–0002

This report has not been published by the U.S. Department of Justice.
To provide better customer service, NCJRS has made this Federallyfunded grant final report available electronically in addition to
traditional paper copies.

Opinions or points of view expressed are those
of the author(s) and do not necessarily reflect
the official position or policies of the U.S.
Department of Justice.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

Governments’
Management of
Private Prisons

Grant # 98-CE-VX-0002
Cambridge, MA
Lexington, MA
Hadley, MA
Bethesda, MD
Washington, DC
Chicago, IL

September 15, 2003

Prepared for
National Institute of Justice
810 Seventh Street, N.W.
Washington, DC 20531
Prepared by
Douglas McDonald, Ph.D.
Carl Patten, Jr., Esq.
With the assistance of
Elizabeth Fournier
Stephen Crawford
Abt Associates Inc.
55 Wheeler Street
Cambridge, Massachusetts 02138

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

This project was supported by Grant No. 98-CE-VX-0002, awarded by the National Institute of Justice, Office of
Justice Programs, U.S. Department of Justice. Points of view in this document are those of the authors and do
not necessarily represent the official position or policies of the U.S. Department of Justice.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

Contents
Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . i
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xix
One View: The Market is Most Efficient . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xix
A Contrary View . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xx
Failures at Privately Operated Prisons: Who Gets Blamed? . . . . . . . . . . . . . . . . . . xxi
Exercising “Buying Power”Strategically . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxv
The Challenge of Developing New Methods of Managing Mixed Public and
Private Organizations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxvi
What This Report Is About . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxvii
Data Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxvii
A Roadmap to This Report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxx
1.
Why Contract? Governments’ Objectives in Turning to Private
Imprisonment Firms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Turning to the Private Sector to Acquire Needed Resources Expeditiously . . . . . . . 1
Two Private Prisons Markets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.
The Contractual Structure: Selected Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Facility Ownership and the Risks of Entrenchment . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Payment Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Designing Performance Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
The “Making Government Accountable” Movement and Its Implications for
Contracting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Performance Standards in Criminal Justice Agencies . . . . . . . . . . . . . . . . . . . . . . . . 22
Performance-Based Contracting for Criminal Justice Services . . . . . . . . . . . . . . . . . 24
Operational and Non-Operational Goals in Contracts . . . . . . . . . . . . . . . . . . . . . 25
3.
Monitoring Contractors’ Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
How Closely are Facilities Monitored? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
The Purpose of Monitoring and the Monitors’ Role . . . . . . . . . . . . . . . . . . . . . . . . . 31
Monitoring Outcomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
Designating Outcomes to be Monitored . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
Exactly How is Such Performance Measured and Monitored? . . . . . . . . . . . . . . 39
Assessing Achievement of Strategic Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
Comparing Costs of Government Operation to Contracting . . . . . . . . . . . . . . . . 40
Comparing Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
4.
Texas: Going Private to Expand Capacity Quickly . . . . . . . . . . . . . . . . . . . . . . . 45
The Legislature Decides to Contract for New Prisons . . . . . . . . . . . . . . . . . . . . . . . . 46
The Legislature Defines Its Objectives for Privatization . . . . . . . . . . . . . . . . . . . . . . 48
Implementing the Privatization Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
Structuring What Texas Seeks in the Market: The RFP . . . . . . . . . . . . . . . . . . . . . . 53

Abt Associates Inc.

Contents

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

5.

6.

Operations, Not Ownership . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Defining the Contractors’ Obligations: Striking a Balance Between
Specificity and Flexibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Daily Living Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Healthcare . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
Safety, Security and Discipline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
Personnel Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
Promoting Prisoner Rehabilitation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
Selecting Contractors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
Monitoring Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Did Texas Achieve Its Objectives in Privatizing Prisons? . . . . . . . . . . . . . . . . . . . . 63
Cost Savings? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
Florida: Seeking More Cost-Effective Performance . . . . . . . . . . . . . . . . . . . . . . 67
The Legislature’s First Try . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
The Legislature Tries Again . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
The Legislature’s Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
Specifying How the DOC and the CPC and Private Firms Are to Cooperate . . . 71
Establishing the Chain of Accountability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
Translating Policy Objectives into Action in the RFP and the Contract . . . . . . . . . . 74
Lower Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
Obtaining Capable Partners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
Effective Programming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Writing the Contract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
Determining If Objectives Were Achieved . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
Were Savings Achieved? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
Did Privatization Result in Less Criminality? . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
Have the Private Firms Introduced Innovations into the State’s
Prison System? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
Oklahoma: Managing the Risks of Dependence . . . . . . . . . . . . . . . . . . . . . . . . . 85
Too Many Prisoners and Not Enough Money . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
Finding Rather than Building Prison Beds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
A Private Prisons Industry Emerges in Oklahoma . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
Contracting Strategies to Manage Risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
Protection of The State’s Bargaining Power Through Regulation . . . . . . . . . . . . 94
Sharing the Risks Associated with Under-Use . . . . . . . . . . . . . . . . . . . . . . . . . . 97
Ensuring Adequate Quality and Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
Statutory Checkpoints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
Liquidated Damages Clause . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
Has the Privatization Program Met Its Objectives? . . . . . . . . . . . . . . . . . . . . . . 102

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

Foreword
This is the report of a project supported by grant number 98-CX-VX-002 awarded by the
National Institute of Justice, Office of Justice Programs, U.S. Department of Justice. Points
of view in this document are those of the authors and do not necessarily represent the official
position or policies of the U.S. Department of Justice. Nor do they represent the position or
policies of Abt Associates Inc. Abt Associates is a non-partisan research corporation that
takes no position as a corporation on the matters examined here.
Voncile Gowdy, Ph.D. served as the monitor for this grant at the Institute, and Ronald
Everett, Ph.D. assumed this responsibility following Dr. Gowdy’s retirement. We appreciate
their support and assistance.
The sources of information used for this report were several. A mail survey was sent to
directors of correction in state and federal governments to inquire about several aspects of
contracting practices as of the end of 1997. Additional information was collected by means
of telephone interviews with selected directors in several states, which were conducted by
Tom Rolfs, the retired director of the Washington State Department of Corrections’
Institutional Division. Mr. Rolfs served as a consultant to this project. Richard Crane, an
attorney and an expert in legal and contracting issues surrounding privatization, also served
as a consultant to the study. He advised us on various aspects of contracting.
Carl Patten, Jr., an Abt Associates staff member at the time, also conducted on-site visits to
prisons, public officials, and private prison administrators in Texas, Oklahoma, and Florida.
These trips were taken to develop detailed information about contracting for prison
operations in each of these states. Many persons were generous with their time. These
included, in Texas: Former State Senator Ray Farabee, now Vice Chancellor and General
Counsel, The University of Texas System; Craig Washington, Former Texas State Senator
and U.S. Representative; Ron Champion, Assistant Director, Texas Department of Criminal
Justice, State Jail Division, now of Wackenhut Corrections Corporation; Terri Wilson,
Administrator of Contract Facility Operations for the Institutional Division of the TDCJ;
Frank Inmon, Assistant Director, Purchasing and Leases, Contracts Branch; and Sharon
Keilin, Assistant Director for Operational Support, Texas Department of Criminal Justice.
We are also grateful to Elaine Cummins, who wrote her dissertation on Texas privatization,
for advising us about whom to interview.
In Oklahoma, those interviewed included James Saffel, Director, Oklahoma Department of
Corrections; Dennis Cunningham, Administrator, Private Prisons Administration, Oklahoma
Department of Corrections; Cal Hobson, Oklahoma State Senator (District 16); Richard W.
Kirby, Deputy General Counsel to Governor Frank Keating; and Linda Allen, Contract
Monitor, State of Oklahoma Department of Correction. In Florida, they included James
Biddy, Deputy Assistant Secretary, Florida Department of Corrections; Damon Smith,
Abt Associates Inc.

Foreword

i

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

Partner; Mirabella, Smith & McKinnon, Inc.; C. Mark Hodges, Executive Director, Florida
Correctional Privatization Commission; David McIntyre, Contract Monitor; Florida
Correctional Privatization Commission; and La'Tera D. Osborne, Operations Coordinator at
the Commission.
In addition to the information developed in the course of interviewing these persons, a
number of others have contributed to this in ways not always visible, as they have been
valuable colleagues to Douglas McDonald on other studies of privately operated prisons.
These include Gerald Gaes, Director of the Federal Bureau of Prisons’ Office of Research
and Evaluation; Scott Camp, of the same office; Julianne Nelson, an economist who consults
with the Bureau on privatization issues, and Malcolm Russell-Einhorn of Abt Associates.
The senior author is also indebted to Mary Levick Owens, formerly of Abt Associates, who
conducted with him another earlier study of contracting practices in selected states.
Several staff at Abt Associates were engaged in this project. These included Elizabeth
Fournier and Stephen Crawford, both of whom assisted in the mail survey of correctional
officials, in the development of the databases containing responses to the survey, and in the
analyses of these data. Joan Gilbert also assisted in the administration of the mail survey and
in other clerical tasks. Mary-Ellen Perry assumed clerical responsibilities at the end of the
project.
Two reviewers commission by the National Institute of Justice read the final report to this
project and made a number of excellent suggestions, some of which have been incorporated
into this revised and now final version.
I am grateful to all of these persons who contributed to this report.
Douglas McDonald
Cambridge, Massachusetts
September 2003

ii

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

Summary
This report examines state and federal governments’ practices of contracting with private
firms to manage prisons, including prisons owned by state and federal governments and those
owned by private firms. Its focus is on contracting for imprisonment services in secure
facilities, rather than for low-security or non-secure community-based facilities. Focus is
also limited to facilities for convicted adult offenders, rather than facilities that serve as local
jails or immigrant detention facilities.
Several aspects on the contracting process are examined: reasons for deciding to contract, the
structure of the solicitations and contracts, and the procedures for monitoring the
performance of contractors. This contracting process is important because it is the link
between public agencies and private firms that do the public’s work. The performance of
these privately operated prisons depends in large part upon how governments structure their
contracting programs and how they manage private firms through the contract procurement
and administration processes.
When governments turn to the private sector for prison beds, they are consumers in a
marketplace. As consumers, they can exercise power over the kind and level of services to be
provided and at what cost. Contract procurement and administrative procedures can either
strengthen the government’s hand in accomplishing its ultimate goals or weaken it. The
challenge is to devise effective procedures for contracting with private firms. The state’s
strategic objectives will most likely be achieved if the reasons for contracting are reinforced
by the design of the work to be delivered under contract, how financial incentives are
structured, how risks are distributed among contractor and government agency, how bids are
evaluated and winners selected, how contract compliance is monitored, and how good
performance by the contractor is encouraged and rewarded.
Data Sources

Information was obtained for this study from three principal sources. These included:
•

A mail survey of correctional agencies in all states governments, the Federal Bureau
of Prisons, the District of Columbia, the Commonwealth of Puerto Rico, and the U.S.
Virgin Islands. This was undertaken to get information about each agency’s practices
and experience with contracting for privately delivered imprisonment as if December
31, 1997. The first inquired about the agencies’ practices and plans regarding
privatization as of December 31, 1997. The second section was designed to obtain

Abt Associates Inc.

Summary

iii

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

information about each contract, its administration, and monitoring. Of the 55
surveyed government agencies, 1 all but two responded to the survey.
•

Site visits and interviews were conducted by Abt Associates staff in Texas, Florida,
and Oklahoma to examine contracting practices. These particular states conduct a
disproportionately large amount of the prison privatization business in this country.
At the end of 1997, there were 37,651 prisoners housed in privately operated facilities
for state or federal departments of correction. Forty-two percent of these, or
approximately 15,700 persons, were sent there by these three states. These agencies
are also among those that house the largest percentage of their prisoners in privately
operated facilities. Texas had the second-largest number of their prisoners in
privately operated facilities—7,223, or five percent of its total population at year-end
1997 (the Federal Bureau of Prisons had the largest number: 9,951). Oklahoma
reported having 4,588 prisoners in contract facilities, or 23 percent of its total state
prisoner population—the highest proportion in the country. Florida had 3,877 in such
facilities.

•

Information was also obtained from review of statutory laws, requests for proposals,
contracts, evaluations and other assessments of contracting.

Structure of the Report

The report has two parts. The first three chapters examine selected issues pertaining to
contracting for imprisonment services in all jurisdictions. The second part includes three
separate case studies of contracting experiences in Texas, Florida, and Oklahoma. These are
not meant to be exhaustive accounts but rather focused examinations of why the states chose
to engage the private imprisonment industry and how it has implemented this choice.
Attention is also given to assessing, briefly, the extent to which the state has achieved its
objectives.
The Prevalence of Contracting

Twenty-three states reported having contracts with private firms on December 31, 1997, to
house prisoners, as did the District of Columbia, the Federal Bureau of Prisons, and the
Commonwealth of Puerto Rico. Two other states reported placing prisoners in private
facilities located in other states through an agreement between two public agencies (hereafter,
an “intergovernmental agency agreement” or “IGA”). These were reported by the two
respondents as equivalent to contracts, and we have included them in our analysis. These 28
jurisdictions reported having a total of 91 active contracts on that date, with 84 different

1

iv

Two agencies contract for private prisons in Florida: the state Department of Corrections and the Florida Privatization Commission.
Both responded to the survey.

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

private facilities. (The number of contracts exceeded the number of private facilities because
some facilities contracted with more than one political jurisdiction.) These 84 facilities held
a total of 37,651 prisoners at year-end 1997, or 4.3 percent of the nation’s 1.2 million
prisoners held by state or federal correctional agencies.
Two Markets

Contracting with private prisons in this country has followed two different dynamics, and
each poses different challenges to managing prison privatization. The dominant mode is for a
government agency to decide to contract for some of its needed state prison beds, and then to
seek a contractor willing to provide these beds in-state. In some instances, the state owns the
facility and signs a contract with a private firm to manage and operate it, while in other
instances the government elects to contract out the design, construction and operation of a
new facility. The most usual result, regardless of who owns the facility, is the creation of
one-to-one relationship between prison operator and the state prison system. That is, the state
prison system is the contractor’s sole client at the facility; the only prisoners held in the
facility are those under the jurisdiction of the client state agency. Moreover, the prison is in
the same state as the publicly operated prisons, which creates at least some of the conditions
supportive of a close integration between the publicly operated facilities and the privately
operated prisons.
The second general pattern of contracting for prison beds poses different challenges for state
management. Rather than waiting for the states to issue a call for service, some private firms
take the risk of building facilities without first being assured of any prisoners from a
particular correctional department. (These are often called “spec” prisons, built as
speculative ventures by private correctional firms.) Once built and staffed, they advertise
their availability to correctional and law enforcement agencies anywhere in the country that
are in need of prison beds. Not all firms succeed in attracting prisoners. Those that do may
hold prisoners from a variety of different agencies, both out-of-state as well as from the state
in which they are located. As such, these facilities are oriented to what is essentially a
national market, in contrast to those facilities that are brought into being as a result of a state
or federal government's issuance of a request for proposals and subsequent awards of
contracts. Many of these facilities that are oriented to the national market may not have any
prisoners at all from the correctional agencies in the states in which they are located. Indeed,
they may have no relationship with the state governments in these states, other than an
obligation to pay corporate income taxes. Owners of private property do not need licenses
from state correctional agencies to build and operate imprisonment facilities and, until
recently, most state legislatures have not established regulatory systems to govern private
prison operations.
Of the 84 facilities having some sort of contractual or quasi-contractual relationship with
state and federal governments on December 31, 1997, 15 were not in the same state as the
contracting government and served the national market. There were probably other out-ofAbt Associates Inc.

Summary

v

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

state or in-state facilities in this latter category which were not identified by survey
respondents because the questionnaire asked about all contracts with private firms, which
was read by at least some respondents as not encompassing intergovernmental transfers or
interagency agreements that resulted in prisoners being housed ultimately in privately
operated facilities.
The national and local in-state markets differed in a number of significant ways. Two-thirds
of the agreements between state governments and nationally-oriented facilities were noncompetitive interagency agreements authorizing prisoner transfers. The rest resulted from
competitive bidding. In contrast, 70 percent of the contractual agreements between state
governments and privately operated prisons located within their states’ boundaries involved
competitive bidding. None of the agreements with the nationally-oriented facilities specified
whether inmates shall be housed in single-bed or double-bunked facilities, whereas half of
the agreements with in-state facilities did so. Moreover, half of these nationally-oriented
facilities were smaller and held fewer than 200 inmates; the rest held between 201-600
inmates. In contrast, only 20 percent of the facilities serving their host state governments had
few than 200 inmates, and a third had more than 600. The majority of agreements between
states and out-of-state facilities had terms of two years or less, where most of the contracts
between state governments and in-state facilities were longer—between two and five years,
and some longer. The per diem rates paid for out-of-state facilities were also higher than for
in-state contracts. All agreements between state governments and out-of-state facilities that
sold services on the national market specified payment rates in excess of $35 a day, whereas
55 percent of the contracts between states and in-state facilities specified lower per diem
rates. Nationally-oriented facilities having agreements with more than one client government
sometimes charged different per diem rates to each client. There were also significant
differences in how states monitored out-of-state and in-state facilities, discussed below.
Governments’ Strategic Objectives for Contracting

Even though questions about costs and savings associated with contracting appear paramount
with analysts who have compared private and government-operated prisons, a review of
recent history suggests that these considerations have been of secondary importance to
legislators’ and correctional administrators’ interests in obtaining needed beds quickly. This
is not to say that the prospects or, at least, hopes for savings have been insignificant. Claims
and beliefs about the cost-effectiveness of privately delivered services have certainly
provided important ideological supports for decisions to turn to the private sector rather than
meeting the increased demand for prisons with more government-run facilities. But
contracting offers other advantages, and may serve governments’ interests regardless of
whether taxpayers’ monies are saved in the process.
Correctional administrators who had active contracts with private imprisonment firms at
year-end 1997 were asked to report their government’s objectives for contracting and to rank
the relative importance of these objectives. According to them, it is clear that contracting
vi

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

was undertaken in most states primarily to reduce overcrowding and to acquire beds quickly.
(An important dynamic was the demand of federal courts to improve conditions of
confinement in the public facilities.) Lowering the cost of operations or construction was
reportedly of secondary importance in all but eight states. (In these eight states, cost savings
were paramount.) In all, 86 percent of the 28 responding jurisdictions cited reduction of
overcrowding as an objective of any rank, whereas about half (57 percent) cited hopes of cost
savings. Improving service quality by contracting was given as the most important reason in
only one state, although nearly half (43 percent) of the contracting jurisdictions reported
interests in improving service quality, even though this objective was ranked most often as
being somewhere between the fifth and ninth most important reason.
The survey recorded the correctional directors’ representations of the reasons, and these may
not correspond precisely with answers that legislators or governors would have given to the
question. How legislators and governors would have answered the question is especially
relevant, because these were reportedly the most active initiators of efforts to contract for
private imprisonment. The Abt Associates survey of state and federal correctional
administrators asked who initiated the decision to contract for private imprisonment services
in their jurisdiction. In jurisdictions that contracted for the management and operations of
entire prisons, the initiatives to do so were most often taken by legislatures or governors, and
not by correctional administrators. That is, of the 28 jurisdictions that had active contracts in
the closing days of 1997, the legislature was reportedly the source of the initiative in 11
jurisdictions and the governor in five. Correctional administrators were identified as the
initiating agents in seven, but this may not be a true reflection of their role. In at least one
jurisdiction where the chief executive initiated privatization, the director of the correctional
agency took credit for the initiative to contract out. In three states, the federal courts were
identified as the source of the initiative, but this was probably not an accurate depiction of
how contracting came about there. The courts had found conditions in these states’ prisons to
be in violation of constitutional standards, but judges do not typically determine the means by
which these are to be remedied.
Does Contracting with Privately Owned Facilities Risk Entrenchment and Dependence Upon
the Private Firm?

Contracting relationships differ considerable in what is being purchased. Of the 91 different
contracts that were active on December 31, 1997, 41 were with facilities owned by private
firms rather than the contracting government. In theory, at least, there is reason to suspect
that facility ownership gives a contractor an upper hand in subsequent competitions and may
therefore serve to restrict competitiveness.
Among these 91 contracts, only 17 had been recompeted—awarded following expiration of a
previous contract. The remainder were either still under the first contract signed, or were
procured using non-competitive means (e.g., sole source procurement). Of the 17,
incumbency provided an overwhelming advantage: all but one were awarded to the
Abt Associates Inc.

Summary

vii

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

incumbents. (The single contract that was not awarded to an incumbent was one where the
facility was publicly owned, where eight different firms competed, and the award went to the
lowest bidder.) Whether the contracts were awarded to the lowest bidder was known by
respondents for 13 of the competitions. Incumbents won 12 of the 13, and submitted the
lowest price bid in all but one. This indicates that incumbency itself is a powerful advantage.
There was no evidence that private ownership of the facilities made any difference in the
outcomes of these procurements. Only four of the facilities were privately owned; all four
firms won new contracts. Only two of these clearly resulted from competitive procurements
following issuance of requests for proposals, however. The others resulted from unspecified
other forms of procurement.
Payment Structures

Several different contractual arrangements exist to compensate private prison operators,
which are discussed briefly here. Each strikes a different balance between the contractor’s
interest and the government’s. These include, at one end, indefinite delivery/indefinite
quantity contracts that permit the government to pay only for those beds used and, at the other
end, fixed price contracts whereby the government pays the contractor a specified amount
regardless of how many prisoners it sends to the facility. There exist several variants of these
arrangements designed to allocate different financial risks.
Performance-Oriented Contracts

In principal, a distinction can be drawn between two different types of performance
objectives for contractors (or any other agency). On one hand, desired performance can be
defined as compliance with procedures and standards–essentially, conforming to specified
processes or activities. Or performance objectives can be defined as the outcomes that a
particular service aims to accomplish (such as improved prisoner or staff safety, or improved
public safety, or prisoner rehabilitation, or improvements in prisoners’ health status). Most
state and federal governments adopt the former approach and ask that private contractors
perform like their public sector counterparts. Contractual statements of work generally
specify that compliance will be required with the same procedural rules, regulations, and
standards that are in force in the public facilities. For example, correctional administrators
reported that 57 of the 91 contracts in force at the end of 1997 required that facilities achieve
ACA accreditation within a specified time. In addition, administrators reported that 61
contracts explicitly required compliance with conditions established in consent decrees or
other court-mandated standards.
Achieving ACA accreditation is not an outcomes-based performance goal. Rather, ACA
standards primarily prescribe procedures. The great majority of ACA standards are written
in this form: “The facility shall have written policies and procedures on . . . .” The standards
emphasize the important benefits of procedural regularity and effective administrative control
viii

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

that flow from written procedures, and careful documentation of practices and events. But,
for the most part, the standards prescribe neither the goals that ought to be achieved nor the
indicators that would let officials know if they are making progress toward those goals over
time.
An alternative orientation is to establish performance objectives that focus on outcomes to be
achieved other than simple compliance with procedures and standards. This orientation is
encouraged by initiatives in the federal and some state governments to emphasize and
evaluate the performance of public services on the basis of outcomes rather than procedural
compliance. This is consistent with the broader movement in public administration circles
(but not yet in correctional circles) to develop performance-based contracts. For example, the
Federal Acquisition Regulations, supplemented by guidance provided by the Office of
Federal Procurement Policy, encourage the use of performance-based statements of work.
OFPP states in its best practices guide for performance-based service contracting that
“performance-based service contracting emphasizes that all aspects of an acquisition be
structured around the purpose of the work to be performed as opposed to the manner in which
the work is to be performed or through broad, imprecise statements of work which preclude
an objective assessment of contractor performance.”2
Examination of selected contracts for private imprisonment indicates that such outcomesoriented performance objectives are rare. An exception is the contract between the Federal
Bureau of Prisons and the Wackenhut Corrections Corporation at the Taft Correctional
Institution. This contract gives the contractor substantial freedom in structuring its own
procedures but requires that particular outcomes be achieved. Procedures to monitor progress
toward such achievement have been created, and financial incentives to accomplish the
objectives are built into the contract. Twice a year, a special review is conducted to
determine the extent to which the contractor has performed above and beyond strict
compliance with contractual requirements. Accomplishing the stated performance objectives
can earn the firm an award fee amounting to a maximum of five percent of total payments.
The effectiveness of such contractual structures in corrections has not yet been studied, other
than a current study by Abt Associates of the Taft Correctional Institution.
Monitoring Performance

Governments’ practices of monitoring privately operated facilities vary widely. Nearly all
contracts active on the last day of 1997 reportedly received some oversight from the
contracting agency. There were significant differences in how governments monitored instate as opposed to out-of-state facilities that served the national market. Ninety-percent of
the contracts/ intergovernmental agreements with these nationally-oriented and mostly out-ofstate facilities were monitored fewer than 20 hours per month. The other ten percent were

2

Office of Federal Procurement Policy, A Guide to Best Practices for Performance-Based Service Contracting, p. 3.

Abt Associates Inc.

Summary

ix

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

monitored at the rate of between 20 and 80 hours per month. This contrasts sharply with the
monitoring of in-state facilities having (nearly always) exclusive contracts with the
government in the state where they are located. Fifty-two percent of those contracts or
agreements were monitored more than 80 hours per month. Indeed, almost half (48 percent)
had full-time contract monitors, compared to fewer than 10 percent of the “national”
facilities. States that contracted with facilities in-state were much more likely to employ staff
especially trained for the task of monitoring. Sixty-three percent of the contracts with in-state
facilities employed trained monitors, compared with only 14 percent of the
contracts/agreements with out-of-state facilities.
Not surprisingly, the most publicly visible troubles in privately operated prisons have
occurred most often in these arrangements whereby governments contract with out-of-state
facilities to hold prisoners. State contract administrators and monitors rated their
performance below that observed at in-state facilities with which states had (mostly)
exclusive relationships. In 38 percent of all contracts or agreements with out-of-state
facilities, the monitors or administrators rated the quality of the service as below that of
comparable facilities in their own department of correction, compared with 7 percent of the
contracts with in-state facilities.
Assessing Whether or Not the State’s Strategic Objectives in Contracting Are Achieved

Beyond monitoring contractor performance is the question of whether the privatization
initiative taken as a whole–including both the government’s decisions and the contractor’s
performance–are accomplishing the state’s strategic objectives for contracting. Most
governments have not sought such assessments (although Texas and Florida, as discussed in
this report, require them by law). Those that have done so have generally focused on
comparing costs of contracting with the cost of the government operating the facility in
question. Very few have compared the performance of the privately operated facilities
(Florida is the exception, as it commissions studies of post-release recidivism to estimate the
extent to which privately operated facilities improve prisoner rehabilitation). Most
assessments of performance rely on the contract officers’ judgments regarding contract
compliance. There are a number of methodological issues that must be resolved when
assessing whether or not objectives are being achieved. These apply to assessments of costs
and savings and to assessments of performance, and are discussed briefly.
Case Studies of Texas, Florida, and Oklahoma Privatization Experiences

These three states have among the longest and most extensive experience with contracting for
privately operated imprisonment. The states’ objectives for engaging the private sector have
varied, as have the states’ approaches to contracting. Three case studies examine why the
state governments chose to contract, what objectives they sought, what kinds of contracting
procedures they developed, and how successful they were in accomplishing their objectives.

x

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

Texas: Going Private to Expand Capacity Quickly

When the State of Texas contracted out the management and operation of four 500-bed
prisons in 1989, it initiated what was at the time the most expansive prison privatization
project in the nation, moving beyond specialized niches to big medium security facilities and
facilities in the State Jails Division (which house convicted felons) and in the Parole Division
(which are prisons for parole violators). By mid 2,000, 14,339 Texas state prisoners were
being held in private facilities, by far more inmates and private facilities than any other state
(even though other states held a larger proportion of its prisoners in privately operated
prisons). This program was initiated by the state’s legislature principally to expand prison
system capacity quickly and to improve correctional performance. This was needed to avert
heavy fines that could be imposed by a federal court that had declared the conditions and
overcrowding in Texas prisons to be unconstitutional. To carry out these objectives, the
state’s government undertook its privatization initiative by writing comprehensive legislation,
developing detailed contracts and assigning ample staff to oversee and monitor them, and
assigned an independent commission the task of determining if the initiative achieved the
financial objectives that the legislature and the executive branch sought (a secondary
objective of privatization was to deliver equivalent services with at least a 10 percent savings
to the state).
Although state officials initially sought to rely on private firms to construct and to operate
new facilities, they ultimately determined that their objectives would be better served by
owning the facilities and contracting only for operation. This was thought to strengthen the
state’s hand in the event of contractor failure to comply with contractual obligations and by
reducing the risks of entrenchment. To increase the likelihood that the facilities would
operate according to the state’s mandates and the federal court’s demands, contracts were
written that blended mandatory requirements with performance objectives and incentives that
left some discretion to the contractor in some areas. The state did not take the approach that
private firms should be given wide latitude in determining how the service was delivered.
Requirements and specific performance objectives were spelled out for all aspects of prison
administration–an approach that mirrored in many respects the approach that the federal court
had taken in forcing the state to comply with its orders. This included requirements for daily
living conditions, health care, safety, security, disciplinary procedures, personnel
requirements, including training, and requirements for prisoner rehabilitation programming,
with corresponding performance objectives.
The state sought to improve the likelihood of good performance by allowing competition at
time of bidding while restricting the playing field to capable players. To be allowed to bid,
firms had first to demonstrate substantial experience and capacity. Procedures for evaluating
bids were designed to favor the bidder proposing not the least costly service but the one
offering the best operational plan–especially the best plans for rehabilitation programming
(again, one of the federal court’s principal concerns).

Abt Associates Inc.

Summary

xi

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

Contracts also built in strong provisions for monitoring to determine if conditions complied
with contractual requirements and to determine if performance objectives were being met.
As of December 31, 1997, monitors of ten of the 20 facilities were on site full-time. Four of
these facilities were state jails and six were institutional units. The TDCJ reported that
monitors spent approximately 25 hours, or less, of their time monitoring five of the 20
privately operated facilities.
The most elaborate of the performance measurement procedures were erected to determine if
the required 10 percent savings objective was achieved with each of the contracts. The task
of determining this was given to the Sunset Advisory Commission. The statute specified that
the commission should analyze the cost and quality of services in the private prisons as
compared to the cost and quality of any similar state services. However, as the commission
noted in its 1991 report, the development of a cost estimate was complicated by the fact that
the TDCJ did not operate a comparable facility. The commission ultimately overcame this
challenge and developed an estimation approach, concluding that contracting saved the state
14–15 percent—more than the required 10 percent. This computation involved some unusual
accounting assumptions, however. The commission counted payments by each of the firms
to local governments in amounts that approximated the property taxes the firms would have
paid if the prisons were privately owned. Because the state owned the facilities, and the firms
contracted only to operate them, the firms were not legally liable for any such taxes. Nor was
the state liable for such taxes. Consequently, these “payments in lieu of taxes” appear to be
no more than a fee paid by the private firms to governments so that the net cost of contracting
to Texas exceeds the 10 percent threshold. Not counting these payments, private operation
appears to cost the state 8.8 or 9.7 percent less than direct TDCJ operation, depending upon
the facility—or slightly less than the savings required by statute.
In short: the evidence from Texas suggests that the private firms are delivering a service that
would cost the government approximately 9–10 percent more if the state’s corrections
department operated the facilities directly. This assumes that the estimates of the
department’s costs of direct operation are accurate, of course. Lacking more information
about how these costs were estimated, it is not possible to evaluate them.
Florida: Seeking More Cost-Effective Performance

Florida’s correctional privatization program did not bloom in an atmosphere of governmental
cooperation, as the legislature and the state’s correctional agancy were not in agreement
about what was needed. Legislative authorization to contract out prison operations had been
on the books since 1985, but the Department of Corrections’ leaders felt fully competent to
manage any needed prisons. Consequently, in 1993, the legislature passed a second law, the
Correctional Privatization Act, which created a separate agency, the Florida Privatization
Commission, to carry out its privatization directives. This is the only such agency in the
nation dedicated to privatization that is independent of a department of corrections. The

xii

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

commission adopted a more aggressive stance regarding privatization, in accordance with its
mission, and undertook to contract for several prisons. By the end of 1998, the state had
3,877 of its prisoners held in five privately operated prisons, including South Bay
Correctional Facility, which held the highest proportion of maximum-security inmates of any
privately operated facility in the country.
Establishing a separate agency to expedite privatization may have solved some of the
legislature’s problems, but it created others. The CPC was not given executive authority to
run the state's prisons, and the courts could not remand convicted offenders directly or
indirectly to the commission’s custody. The CPC and the DOC were going to have to
coordinate their activities if privatization was to work in Florida. The law was therefore
written in such a way to bind the DOC to utilize any privately operated facility that the CPC
brings under contract.
As in many other states, Florida’s legislators sought to engage private firms to improve the
cost effectiveness of imprisonment in the state. Written into the law is a provisions that any
contract with a private imprisonment firm must “result in a cost savings to the state of at least
7 percent over the public provision of a similar facility.” The law also requires that
contractors provide “for the same quality of services as that offered by the department.”
What sets Florida’s law apart from others, however is the explicit goal of reducing
recidivism: it requires that "work and education programs must be designed to reduce
recidivism." As part of its annual review of the commission and the privatization program,
the commission is directed to compare recidivism rates of the private and public facilities.
Also distinctive is the legislature’s explicitly stated goal of turning to private contractors to
spur “innovation.”
In accord with the legislature’s intent, the commission designs its RFPs to obtain bids from
firms that: (1) are well qualified to deliver the service based on their prior experience, and
which offer a capable management team; (2) offer plans for effective educational, training
and substance abuse programs; and (3) offer the state an opportunity to reduce expenditures
for imprisonment. To facilitate bidder’s meeting the financial savings targets set by law, the
CPC’s approach is to tell bidders what the cost of state operations would be and, by
extension, what the bid price would have to be in order to produce the minimum necessary
savings to the state. To determine if contracting has actually achieved these savings, Florida
law mandates that an independent oversight agency, the Auditor General’s Office of Program
Policy Analysis and Government Accountability (OPPAGA), review the actual performance
of all privately operated correctional facilities under contract with either the CPC or the
DOC.

Abt Associates Inc.

Introduction

xiii

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

The CPC’s approach, consistent with the legislature’s interest in innovation, is to allow firms
a good deal of freedom to design their procedures for delivering imprisonment services.
Private firms are also not bound to follow procedures prevailing in the state’s publicly
operated facilities. Moreover, the CPC’s RFPs request that the bidders take advantage of this
freedom and design a regimen that will improve offenders’ chances of doing well and
conforming to the law when they leave prison.
To assure that private firms deliver education, training and work programs, CPC contracts set
a daily per inmate spending target for these programs that must be met. To encourage
private firms to design and deliver rehabilitative programming, the CPC has designed its
proposal evaluation procedures to reward bidders that offer good rehabilitative programming.
Has Florida’s privatization program accomplished the legislature’s objectives? With respect
to financial savings, the answer has been the subject of considerable controversy. In a 1998
report, OPPAGA reviewed the actual performance of all privately operated correctional
facilities under contract with either the CPC or the DOC and compared costs at two privately
operated facilities–the Bay and Moore Haven Correctional Facilities, operated by CCA and
Wackenhut, respectively–with a similarly sized public facility, the Lawtey Correctional
Institution. It concluded that the Moore Haven facility was more expensive than Lawtey to
operate, and that the Bay facility was marginally less expensive (0.2 percent) than Lawtey.
Both therefore failed to produce the minimum seven percent savings required by law.
The CPC and the two management firms took exception to several accounting assumptions
that were made by OPPAGA analysts and produced their own analyses, which showed that
these two privately operated facilities saved the state money. Moreover, in yet another
analysis, the DOC found that both facilities were more expensive than state-run facilities.
Some of the differences in findings stem from the choice of comparison facilities. The DOC
did not chose a single publicly run facility for the comparison but rather averaged the cost of
nine other public facilities. This biased the case against the private facilities because all but
one of these facilities were larger than both of the privately operated ones. (Larger prisons
are generally less costly to operate because of economies related to scale.) The OPPAGA
study made a more reasonable decision, electing to compare the two privates to Lawtey,
which was about the same size and of the similar security level.
The studies also differed in how certain categories of costs were treated. These categories
included costs associated with monitoring, taxes paid and not paid to the state, credits for
revenues from inmate medical co-payments retained by the private firms, and retirement fund
surcharges. The comparability of medical expenditures was also disputed. The CPC and the
contractors argued that medical spending at the two private facilities and at Lawtey were not
comparable because the prisoners at Lawtey had significantly different medical needs (an
assertion rejected by OPPAGA).

xiv

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

This indicates how the seemingly simple comparison of costs to the state of either direct
operation or of contracting is not a straightforward matter. What matters here, however, is
that the legislature established a system for monitoring the performance of its privatization
program that included an assessment of whether it hit its intended targets.
To determine if privately operated facilities produced more law abiding persons than the
publicly operated prisons, the CPC commissioned a study by independent analysts. The first
such study, released in 1998, compared recidivism rates of inmates released from prisons
operated by the Florida Department of Corrections to those of inmates released from one
prison operated by the Corrections Corporation of America–the Bay Correctional
Facility—and another operated by Wackenhut Corrections Corporation, the Moore Haven
Correctional Facility. The study concluded that a smaller proportion (17 percent) of the
sample of releases from the privately operated facilities had recidivated within twelve months
of their release, compared to a sample of releases from public facilities (24 percent).
However, studies of the rehabilitative effects of different prisons, or different prison
programs, nearly always face significant difficulties that frustrate one’s ability to draw strong
conclusions from the comparisons. Unless prisoners are assigned randomly to the various
prisons under study, it is difficult to sort out the possible effects attributable to differences in
prisons or prison programs from differences in the groups of prisoners that are created by
purposive selection. For example, the RFP for at least one of the private facilities (Bay
Correctional Facility) permits private prison administrators to request that prisoners be
transferred to public facilities if they are “considered by the chief health officer to be
medically, physically or mentally incapable of participating in the programmatic activities
(which have been specifically designed to reduce recidivism) for greater than two weeks.” If
the facility administrators exercised this right, it would result in collecting at this facility a
group having stronger odds of success, regardless of what programming they ultimately
received.
In addition, there were systematic differences in the types of prisoners held in the private and
public institutions, which may have affected the differences in post-release criminality.
Thirty-five percent of the prisoners in the private prison sample spent significant periods of
time in the public correctional facilities prior to being transferred to the private facilities.
These prisoners were thus exposed to both public and private facilities. And finally, there are
significant questions about how the study subjects were selected from among all prisoners
released from these facilities for the purpose of matching.
There has been no formal assessment of whether the private firms introduced innovations
into the state’s prison system. The legislature may have assumed that any innovation of
significance would result in more effective prisons (i.e., lower recidivism) and less costly
ones. The existing studies of comparative costs and of comparative recidivism rates to not

Abt Associates Inc.

Introduction

xv

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

support any strong inferences about the state’s obtaining more innovative imprisonment from
the private sector.
Oklahoma: Managing the Risks of Dependence

As in several other states, Oklahoma turned to privately operated prisons to relieve
overcrowding in its state facilities. Oklahoma’s practice is distinctive, however, because it
sought to rely entirely on private facilities for the additional beds that it needed. It contracted
with out of state facilities and, later, with firms that built prisons “on spec” within the state.
As a result, by the late 1990s, it led the nation in the proportion of state prisoners in privately
operated facilities. Although the state’s policy makers may not have intended to go further
than other governments in developing a hybrid public/private correctional system, various
incremental policy decisions had this result. Oklahoma’s management of this hybrid system
is of interest, consequently, because the state developed a number of strategies to maximize
the benefits of relying heavily upon the private sector while minimizing the risks of
dependence.
The state has long had one of the highest incarceration rates in the country, and the demand
for prison cells has grown consistently since the mid-1970s. The capacity of the state’s
prisons failed to keep pace with the growing numbers of inmates and became overcrowded.
The state’s fortunes were tied closely to oil, and the oil revenues collapsed during the 1980s.
Oklahomans were averse to their government spending more money for prisons, which made
it difficult for the state to build its way out of the crunch. The state first turned to emergency
releases of inmates to reduce overcrowding, but public support for these release programs
had disappeared by the mid-1990s.
Then-governor Frank Keating hoped that the year-to-year increases in demand for prison cells
might be a short-run problem that could be accommodated by means other than prison
construction, so he sought to avoid spending money on prisons that might not be needed in
the future. Beds in other states’ prisons were in short supply, so the DOC turned to privately
operated facilities to house its prisoners. Beginning in 1995, the state began contracting with
privately owned and operated facilities, many of which were in neighboring Texas.
Entrepreneurially-minded private firms and even public officials of Oklahoma towns and
cities saw opportunities in this and began building facilities “on spec” in Oklahoma, betting
that Oklahoma correctional officials would ultimately chose to rely on local facilities rather
than out-of-state ones. This bet paid off, and the state began contracting with these in-state
facilities.
By mid-May, 2000, Oklahoma had 6,204 beds under contract, and 93 percent of those beds
were filled. It was contracting with seven facilities—all of which were located in Oklahoma.
Prisoners in these privately operated facilities constituted about 28 percent of the total
population of state prisoners. This ratio has stayed about the same since then, at least through
mid-2003.

xvi

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

Relying so heavily upon privately owned and operated prisons creates certain risks for the
state. These are substantially different from the risks borne by correctional departments in
Texas, Florida, and elsewhere, which enter into contracts only to manage and operate
government-owned facilities. Absent special agreements, they cannot resume control of the
facilities in the event of trouble. Like all governments that contract with private firms, they
retain responsibility and liability for correctional officers’ performance, but do not have direct
control and authority over them. To minimize the negative effects of the department’s
dependence upon private suppliers, the legislature amended in 1997 the enabling statute to
regulate the state’s relationship with the private prison industry and to protect to the extent
possible the state’s position vis-a-vis this industry. Provisions were written to assure
competitive procurements; procurements that limit the financial impacts on the state budget;
established rights in contract to purchase the facility at a predetermined price so that the state
could acquire more prison properties quickly and with minimized disturbance to the inmate
population; and provisions for comprehensive liquidated damages to create incentives for
contractors to maintain a constitutionally acceptable environment for inmates.
Bringing its inmates back to in-state private facilities has enabled the department to monitor
its contractors much more aggressively than was convenient when prisoners were housed in
out-of-state facilities. By 1998, contract monitors were spending 60 percent to 80 percent
FTE at the in-state facilities.3 Departmental administrators believed this to be sufficient.
The department also conducts regulatory audits.
The principal objectives of Oklahoma’s privatization program have been to reduce
overcrowding, to avoid capital investments in prisons, to avoid developing long-term
obligations to larger numbers of state government employees, to retain flexibility in
obligations in the event of a slackened demand for prison beds, and to do all of these without
spending more money than would be spent if the state had expanded its publicly-operated
prison system. No requirements were established in law to evaluate these outcomes, but
several conclusions can be drawn without complicated study.
•

•

•

3

Use of contractor-operated facilities has enabled the state to meet the demand for beds
and thereby reduce overcrowding.
Oklahoma has succeeded in housing substantial numbers of offender in prisons
without spending a penny on new facilities. For fiscal year 1998, $0 was allocated to
the Department of Corrections' budget for capital construction or debt service.
By developing a mixed public and private correctional system in the state, the
privatization program has allowed the state to expand capacity without increasing
proportionately the numbers of state employees and the long-term financial
obligations associated with such employees.

Interview with Dennis December 7, 1998.

Abt Associates Inc.

Introduction

xvii

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

xviii

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

Introduction
This report examines state and federal governments’ practices of contracting with private
firms to manage prisons, including prisons owned by state and federal governments and those
owned by private firms. Its focus is on contracting for imprisonment services in secure
facilities, rather than for low-security community-based facilities from which prisoners leave
during the day or evening for work, school, or other rehabilitative services. Moreover, we
focus primarily upon facilities for convicted adult offenders, rather than local jails that may
hold both sentenced and unsentenced prisoners, or detention facilities that hold immigrants
pending deportation hearings.
We examine several aspects on the contracting process: reasons for deciding to contract, the
structure of the solicitations and contracts, and the procedures for monitoring the
performance of contractors subsequent to receiving the awards. This contracting process is
important because it is the link between public agencies and private firms that do the public’s
work. Our working hypothesis is that the performance of these privately operated prisons
depends in large part upon how governments manage them through the contract procurement
and administration processes.
This hypothesis is offered as a corrective to much of current thinking in research and policy
circles about privatization. That is, there is a tendency to ascribe the advantages or
disadvantages of privatization to inherent features of privately operated firms or, more
broadly, to features of the private market as opposed to the top-down command system that
characterizes government operations. To some advocates of privatization and contracting,
private firms and the markets have halos, while government is seen as inherently inefficient.
To opponents of privatization, private firms and the market have horns, and the profitmaximization incentives that operate in private markets are seen to encourage sacrificing
public purposes to private enrichment. This study seeks to explore how contract procurement
and administration practices further or hinder public agencies in their efforts to accomplish
public objectives through contracting.

One View: The Market is Most Efficient
Research on the relative costs of private and public facilities is generally animated by the
effort to prove or disprove a number of claims that are made about the inherent superiority of
the private market with respect to delivering services more cost-effectively. Such claims
include the following:
•

managers in private firms are less encumbered by “red tape” and can procure
needed goods and services faster;

Abt Associates Inc.

Introduction

xix

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

•

managers in private firms can manage labor more firmly because protections
against firing and disciplining are fewer, especially in the absence of contracts
with organized labor;

•

in the private sector, managers and the firms they manage bear greater financial
risks than public managers, which spurs greater attention to both cost control and
service quality; and, among other claims, that

•

competition among providers creates incentives for cost effectiveness that are
absent in “public monopolies.”

Corresponding to these various claims are various hypotheses about government managers’
behavior, which are sometimes read as explanations of why government is inherently less
cost-conscious and cost-effective than the private for-profit sector. These include:
•

incentives to grow agency budgets and to spend allocated budgets are more
intensely felt than are incentives to conserve money;

•

government managers are not at risk financially for their decisions, and the
opportunity for financial rewards in addition to salary do not exist;

•

risk-taking in government is discouraged because the potential costs to career
advancement are greater for failing to achieve an objective than for by-the-books
performance; and

•

civil service regulations severely constrain government managers’ ability to
manage, discipline, and fire staff.

A Contrary View
Countering these arguments for the superiority of the private sector are more negative views.4

4

xx

•

Contracting out often results in higher costs to states and localities, especially
when “hidden” costs of contracting are tallied;

•

contractors often cut corners by hiring inexperienced, transient staff at low wages,
by ignoring the requirements of contracts, and by providing insufficient
supervision, all of which result in poor service;

See, for example, Passing the Bucks: The Contracting Out of Public Services (American Federation of State, County and Municipal
Employees, AFL-CIO, 1983, Chapter 1.

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

•

contracting has often been subject to corruption, in the form of kickbacks, bribery,
and collusive bidding;

•

contractors may “lowball” to get the bid and then come back for more money, or
fail to deliver the promised level of service;

•

contractors go out of business, leaving the public sector high and dry; and

•

private firms are subject to strikes and work stoppage.

In addition to these concerns about contracting generally, contracting for imprisonment
services raises a number of special concerns. Perhaps the most prominent is a heightened
risk of violating the rights of prisoners, which may lead to litigation, protests, prisoner
uprisings, and threats to livelihood and property.
Concerns about labor strikes and stoppages were especially worrisome to federal officials
when the Clinton administration first raised the possibility of contracting out the operations
of secure federal prisons in its FY 1996 budget proposal.5 The Bureau of Prisons and others
in the Department of Justice greeted this proposal with considerable reluctance, and the
Department ultimately registered its concerns about the risks of strikes or walk-outs by
private corrections officers and, to a lesser extent, the risks of other disruptions and inmate
disturbances.6 This slowed privatization efforts in the Bureau for several months, but
ultimately Congress mandated a “privatization demonstration” in the federal prison system at
a new low-security federal prison in Taft, California.

Failures at Privately Operated Prisons: Who Gets Blamed?
There are several examples of the tendency to see inherent problems in privately operated
prisons when things go wrong. For example, on June 18, 1995, there was a “disturbance” at a
privately-operated detention center for illegal immigrants in Elizabeth, NJ. This 300-bed
facility had been operated for almost a year under contract with the Immigration and
Naturalization Service by ESMOR Inc. During this period, there were reportedly numerous
complaints and allegations of abuse and inappropriate conditions of confinement, as well as
complaints that both ESMOR and INS officials had failed to take appropriate corrective
actions.7 Detainees rioted, the contractor lost control of the facility, and outside law
enforcement officials had to be called in to retake it and to gain control. The congressman

5
3

7

Gerth, Jeff and Stephen Labaton. “The Pitfalls of Private Penitentiaries.” The New York Times, November 24, 1995, A1.
Letter from Stephen R. Colgate, Assistant Attorney General for Administration, to Hon. Harold Rogers, Chairman, Subcommittee on
the Departments of Commerce, Justice and State, the Judiciary, and Related Agencies, Committee on Appropriations, U.S. House of
Representatives, June 5, 1996.
U.S. Immigration and Naturalization Service, “The Elizabeth, New Jersey, Contract Detention Facility Operated by ESMOR, Inc.,”
1995.

Abt Associates Inc.

Introduction

xxi

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

who represented that district raised questions about the contract and the contractor, and the
story made the first page of The New York Times. The facility was closed, the contract was
transferred to another firm (the Correctional Corporation of America), and ESMOR changed
its name to the Correctional Services Corporation. Critics argued that privately operated
facilities did not adequately protect the public or the inmates’ rights.
The publicity surrounding this incident paled in significance to the uproar that was caused by
troubles at the Northeast Ohio Correctional Facility in Youngstown, Ohio. The Corrections
Corporation of America (CCA) entered into an agreement with this economically depressed
city to build a facility, promising to hire locally to relieve unemployment. The facility, which
was designed as a strong prison capable of holding 1,500 medium security inmates, opened in
May, 1997. The District of Columbia, under pressure by the federal courts to reduce
overcrowding, contracted with the facility to take minimum and medium custody prisoners.
Soon after the facility opened and after the DC government transferred 900 prisoners to it,
prisoners assaulted each other with homemade knives, two were killed, and six dangerous
prisoners escaped. Youngstown residents were alarmed and rightly concerned for their
safety, and the governor called upon the Ohio Department of Rehabilitation and Corrections
to assist, even though no agency of the state government, including the department, had any
relationship to the facility. Recognizing that the state lacked any regulatory authority over
private prisons that were located within its jurisdictional boundaries, legislators passed a law
in March, 1998 that established a measure of state authority over this and any other private
prison that might be located in Ohio.8 The report of an investigation ordered by the
Department of Justice cited a variety of conditions that contributed to breakdown, including
insufficiently trained staff, absence of necessary policies, high staff turnover, inadequate
safety procedures, and mismanagement by the government client agency.9
Wackenhut’s turn at getting negative publicity came in New Mexico in August 1999. The
state had not built a maximum or medium-security prison since 1987, and it lacked beds
needed to accommodate the growing numbers of inmates in the state prison system. After
relying upon out-of-state facilities to handle the overflow for several years, the state
government entered into an agreement in 1995 with Wackenhut Corrections Corporation to
build and operate two prisons, one near the town of Hobbs and the Guadalupe County
Correctional Facility near Santa Rosa. The former opened in January 1999, the latter in May
1998. In addition, the state also negotiated to place prisoners in a prison located in the town
of Estancia, operated by CCA.
In 1999, four inmates were murdered by other inmates in the Wackenhut-operated facilities.
In August of that year, inmates in the Santa Rosa facility rioted for three hours, taking control

8
9

xxii

Ohio Revised Code §009.07(G)(2).
John C. Clark, Report to the Attorney General: Inspection and Review of the Northeast Ohio Correctional Center (Washington,
D.C.: Office of the Corrections Trustee for the District of Columbia, Nov. 25, 1998).

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

of two housing units, and a guard was stabbed to death. These incidents, plus the injuries of
two guards at the CCA-operated facility in Torrence County, provoked intense media
attention and calls by state officials for investigations. Shortly afterwards, 109 of the most
troublesome prisoners in the Santa Rosa facility were taken out in the middle of the night and
taken to an airport in Albuquerque under extremely heavy guard, where they were flown to
Virginia and placed in that state’s “supermax” prison.
Critics of privatization refer to these incidents and others like them as evidence that privately
operated prisons cannot be operated as well as public ones, that they are especially ill suited
to more dangerous prisoners, and that public safety is compromised when governments rely
upon them to house prisoners. New Mexico’s Department of Corrections Commissioner
Robert Perry, who supported contracting with private prisoners, was reported in the press as
saying that hard-core inmates take “special management,” something that private prisons “are
not really designed to do.”10
Attributing these incidents to private prisons, and seeing them as risks inherent in
privatization, is perhaps an oversimplification, however. Publicly operated prisons have their
share of killings, assaults, escapes, and uprisings. But more to the point: investigations of at
least the first two of the incidents above do not lay the fault solely at the doorstep of the
private firms. Rather, the troubles stemmed in some measure from the governments’
management of these contractors.
An internal assessment report by the INS identified a number of ways that government
actions contributed to the troubles at the Elizabeth, NJ detention center.11 The poor
conditions in the facility resulted in large part from its INS-defined mission as a short-term
detention center to hold illegal immigrants until their cases have been heard and resolved by a
separate agency, the Executive Office for Immigration Review (EOIR). EOIR case
processing was slow, so that detainees spent months and even years in facilities lacking any
programming. Detainees were anxious about their hearings and their possible deportation,
which no doubt created tensions. The private contractors provided only the detention
services, and had no control over the pace of case processing, as that was done entirely by
INS and EOIR officials.
ESMOR’s staffing of the facility was found to be deficient and of low quality. At the first
whiff of smoke, staff abandoned the facility rather than risk injury to themselves. But this is
a direct consequence of the specifications that INS issued in its request for proposals. In its
solicitation, the INS specified that the contractor would pay detention officers at a rate
established by the Department of Labor for “rent-a-cops” charged with guarding property
rather than higher-cost security officers trained for guarding persons. Although one of the

10
11

Loie Fecteau and Leslie Linthicum, “Riot Suspects Sent to Va.” The Albuquerque Journal 9/4/1999.
U.S. Immigration and Naturalization Service, The Elizabeth, New Jersey, Contract, 1995.

Abt Associates Inc.

Introduction

xxiii

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

bidders reportedly challenged this and proposed that higher paid staff be required, the INS
accepted a bid that offered what it had requested. It is not surprising, consequently, that the
detention officers were unwilling or unable to quell a disturbance. Moreover, the facility was
sometimes understaffed, but this resulted in part from the government taking months to
complete security clearances for applicants, which made it difficult for the contractor to fill
vacancies.
The government’s role in contributing to the troubles at Youngstown were also detailed in an
investigation undertaken at the request of the U.S. Attorney General by the Office of the
Trustee for the District of Columbia.12 Authors of the investigation cited numerous examples
of how the DC government mismanaged its privatization initiative. For example, rather than
taking months to bring the facility’s population up to capacity, as wise correctional practice
dictates, the DC government shipped, and CCA received, large numbers of prisoners into the
facility within a matter of weeks, including many who were among the worst troublemakers
in the DC system. Even though it had contracted with CCA to imprison medium and
minimum security prisoners, it shipped many who had been classified as maximum security.
The breakdown at the Youngstown facility put a spotlight on another widespread problem:
that government regulatory controls over the private imprisonment business are poorly
developed. In general, governments exercise their powers to regulate and monitor private
firms’ imprisonment practices through their contracts with them. But some firms are
constructing facilities in one state while contracting with an out-of-state government to house
their prisoners. In Youngstown, neither the Ohio Department of Correction nor any other
Ohio state agency had any relationship with the Corrections Corporation of America. A
similar situation exists in nearly all other states where private facilities house prisoners from
another out-of-state correctional agency. This has prompted much discussion about whether
and how this emergent industry should be regulated or licensed.13
In the finger-pointing that followed the uprising in the New Mexico facility and the death of
the guard, some state legislators raised questions about the adequacy of the department’s
classification procedures. For example, these procedures did not incorporate information
about gang affiliation, which would provide correctional administrators guidance about who
to separate from whom. Because the prisoner violence in these prisons was gang-related,
information about gang affiliation would have given both the department and the Wackenhut
administrators better capacity.
As in the case of Youngstown, there was also some dispute about whether the state erred in
sending dangerous prisoners to these private facilities. Wackenhut’s CEO reportedly

12
13

xxiv

William C. Collins, Privately Operated Speculative Prisons and Public Safety: A Discussion of Issues (The Corrections Program
Office, Office of Justice Programs, U.S. Department of Justice, undated).
Clark, Report to the Attorney General.

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

complained to the department a month before the deadly uprising that the state was sending
maximum security prisoners to the medium-security prison in Hobbs, even though they had
not agreed to this.14 The state’s commissioner replied that all were classified as “medium
security” prisoners at the time of their transfer (although the subsequent transfer of 109 of
them to a “supermax” facility indicates that there was good reason to consider them as more
dangerous and risky than the medium-security designation would lead one to suspect).
Critics also charged that Wackenhut’s correctional officers were inexperienced and
insufficiently trained, a commonly heard allegation about private prisons. Like many new
facilities, both public and private, they no doubt had a higher proportion of “green” officers,
but all staff were required to undergo the same eight-week long training course at the state
training academy, where new staff in the state’s department of correction are trained.

Exercising Buying Power Strategically
These examples point to the importance of how the relationship between government
correctional departments and private contractors is structured and managed. When
governments turn to the private sector for prison beds, they are consumers in a marketplace.
As consumers, they can exercise power over the kind and level of services to be provided and
at what cost. Contract procurement and administrative procedures can either strengthen the
government’s hand in accomplishing its ultimate goals or weaken it. For example, public
agencies may elect to contract in hopes of saving money, but then issue statements of work in
solicitations that specify in great precision exactly how the private firm must deliver the
service or require bidders to follow precisely the same rules and procedures that prevail in the
government-run prisons of that jurisdiction. This procurement approach may work against
the strategic objective of contracting to save money or to improve cost effectiveness.
Similarly, how governments structure their contracting arrangements may stimulate or inhibit
the longer-term development of a competitive industry. Over the long run, a competitive
industry affords consumers (governments, in this case) greater leverage over suppliers and,
consequently, greater power to buy at more advantageous prices.
The challenge is to devise effective procedures for contracting with private firms and
administering the contracts. Strategic objectives will most likely be achieved if the reasons
for contracting are reinforced by the design of the work to be delivered under contract, how
financial incentives are structured, how risks are distributed among contractor and
government agency, how bids are evaluated and winners selected, how contract compliance is
monitored, and how good performance by the contractor is encouraged and rewarded.

14

Michael Coleman and Loie Fecteau, “Perry: Let’s Toughen Inmate Restrictions,” The Albuquerque Journal 9/5/1999.

Abt Associates Inc.

Introduction

xxv

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

The Challenge of Developing New Methods of Managing Mixed
Public and Private Organizations
Not all states rely upon privately operated facilities to house their prisoners, but several have
transformed what was once an entirely governmental operation to one in which private
operators as well as public officials cooperate to provide prison beds. By the end of 1997, 25
states as well as the federal government, the District or Columbia, and the Commonwealth of
Puerto Rico used privately operated facilities, located either within their geographical
boundaries or in other states.15 Some had placed significant proportions of their prisoners in
them. Texas, for example, reported having 14 active contracts at the end of 1997 to hold
7,223 state prisoners in privately operated prisons (or five percent of its total prisoner
population), not including contracts to hold state jail inmates. Oklahoma reported holding 23
percent of its state prisoners in privately operated facilities, under 9 different contracts.
Privately operated facilities also held 23 percent of New Mexico’s prisoners, 22 percent of
the District of Columbia’s, 21 percent of Colorado’s, 20 percent of Tennessee’s, 19 percent
of Louisiana’s, 17 percent of Puerto Rico’s, 16 percent of Mississippi’s, and 10 percent of the
Federal Bureau of Prison’s.
In some instances, governments engaged private correctional firms without creating thick
contractual relationships, and without seeking to transform their public correctional systems
into hybrid ones. In such cases, correctional departments were typically looking for shortterm relief from overcrowding, by “renting beds,” often in out-of-state facilities. Beds were
sometimes obtained in these facilities through contractual agreements with private firms, but
also through intergovernmental agreements between the sending state and a county sheriff’s
department in another state. This allowed states to transfer prisoners to these local jails, and
the sheriffs would, in turn, send them to a privately operated facility with which they had an
established contract to house prisoners.
In other instances, the partial “privatization” of the state’s correctional system reflected a
policy decision by the legislature or the governor, rather than being a temporary expedient.
Enabling legislation was passed in many states explicitly authorizing private prisons;
appropriations were made to contract for such facilities; competitive bidding was conducted;
and detailed contracts were negotiated. In some places, the public correctional agency’s
leaders sought to integrate the private facility’s management team into the agency’s team.
Although the emergence of a private imprisonment industry seems to pose new
administrative challenges to public correctional executives, public administrators in other
domains have been faced with similar problems. In the United States and in a number of
other countries, governments have been experimenting with new forms of public

15

xxvi

This was determined by the Abt Associates survey of state and federal correctional agencies.

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

management for the purpose of increasing the efficiency and accountability of public
services. With increasing complexity in economic and social life, and with citizens’ rising
expectations of what governments can and should do, demands on governments have
increased. In nations with market-based economies, these expanded responsibilities have
been met not only by enlarged government agencies but also by a growing network of private
organizations that deliver services that governments pay for in part or fully. Resistence to
paying for bigger governments with tax revenues and borrowing has been mounting,
reinforcing the trend toward more expansive reliance on networks of private organizations.
Privatization is but one of a number of related reforms that have been occurring in
governments around the world during the past two decades. These other reform initiatives
include efforts to refocus governments on outcomes-based performance rather than rewarding
compliance with procedures, to break down bureaucratic administrative hierarchies, to
introduce market-based incentives and competition, to redefine the public being served as
consisting of “customers,” and to “empower” managers by freeing them from constraints on
their authority. Although these various reforms do not necessarily involve the privatization
of public services, they generally seek to change the traditional culture in public
organizations. In that traditional culture, management operates vertically, through
hierarchical authority. Still, the development of traditional hybrid public/private services, or
interdependent networks of private firms and government agencies, creates new managerial
challenges. For example, in governmental organizations, top managers do not usually
articulate performance objectives for the agencies that they operate, whereas they are forced
to do so when writing contracts with private parties. When structuring contracts, they also
have opportunities to create incentives and mechanisms for accountability that are more
difficult to implement in existing public organizations.

What This Report Is About
This report explores how governments structure and manage their prison privatization
projects. We are especially interested in exploring the interconnections between how
governments as consumers structure and obtain the services of private providers, and the
extent to which procedures for implementing privatization initiatives are consistent with
governments’ strategies or objectives for turning to the private sector. These objectives may
not necessarily be limited to cost savings.
Data Sources

The data we acquired for this study were developed from several sources. One was a mail
survey of federal and state correctional agencies undertaken to get information about each
agency’s practices and experience with contracting for privately delivered imprisonment. A
second source was telephone interviews conducted with a number of correctional
administrators following the mail survey. A third was the site visits and interviews

Abt Associates Inc.

Introduction

xxvii

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

conducted by our staff in several states. These data have been augmented by information
from other sources, including published and unpublished reports and articles, personal
interviews and other communications, and reviews of various documents (contracts, requests
for proposals, evaluations, among others).
Mail Survey

The mail survey was limited to privately operated “prisons,” rather than jails or detention
centers. Because some private facilities do not necessarily identify themselves as “prisons”
as opposed to any other type of “confinement facility,” this required making several
distinctions. We included only those that are most equivalent to secure confinement facilities
in state or federal prison systems and which contract (or otherwise agree) with the state or
federal correctional agencies to provide prison space. In contrast with jails, prisons are
designed to hold sentenced inmates for longer terms and have a variety of programs for the
inmates. We therefore excluded from our purview all privately operated facilities that
function as jails (mostly with county or municipal clients); all detention centers for illegal
immigrants or others; all facilities operating under contract with the U.S. Marshals Service,
the U.S. Immigration and Naturalization Service, or local governments; all privately operated
non-secure facilities; and all juvenile facilities.
To identify the subset of privately operated facilities that most closely correspond to state or
federal prisons, we mailed questionnaires to the heads of all correctional agencies in all
states, the Federal Bureau of Prisons, the District of Columbia, the Commonwealth of Puerto
Rico, and the U.S. Virgin Islands. (Hereafter, these governmental entities are referred to as
“jurisdictions.”) This questionnaire had two sections. The first inquired about the agencies’
practices and plans regarding privatization as of December 31, 1977. The second section was
designed to obtain information about each contract, its administration, and monitoring. Of
the 55 surveyed government agencies, 16 all but two responded to the survey. This survey
was conducted in early 1988 and is hereafter referred to as the “1998 survey.”
Twenty-three states reported having contracts with private firms on December 31, 1997, to
house prisoners, as did the District of Columbia, the Federal Bureau of Prisons, and the
Commonwealth of Puerto Rico. In addition, two other states reported placing prisoners in
private facilities located in other states through an agreement between two public agencies
(hereafter, an “intergovernmental agency agreement” or “IGA”). These were reported by the
two respondents as equivalent to contracts, and we have included them in our analysis. (In
the discussion that follows, we call all reported agreements “contracts,” even if this is not
technically accurate in the case of these two jurisdictions.) These 28 jurisdictions reported
having a total of 91 active contracts on that date, with 84 different private facilities. (The
number of contracts exceeded the number of private facilities because some facilities

16

Two agencies contract for private prisons in Florida: the state Department of Corrections and the Florida Privatization Commission.
Both responded to the survey.

xxviii

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

contracted with more than one political jurisdiction.) These 84 facilities held a total of
37,651 prisoners at year-end 1997, or 4.3 percent fo the nations 1.2 million prisoners held by
state or federal correctional agencies.
In addition, surveyed agencies reported having active contracts on the same day with 529
community-based facilities, such as work-release or educational-release facilities or half-way
houses. For the purposes of our survey, “community-based facilities” were defined as they
are in the periodic Bureau of Justice Statistics census of state and federal prisoners. That is, a
community-based facility is one in which 50 percent or more of the inmates are regularly
permitted to depart unaccompanied for work or study release or for other rehabilitation
programming. These community-based facilities were not sent surveys, as indicated above.
In addition to the 84 privately operated secure facilities that received prisoners under contract
directly from correctional agencies in the surveyed jurisdictions, others held these
jurisdictions’ prisoners via a more circuitous route. That is, some state or federal correctional
agencies sent prisoners to a local government agency using an intergovernmental agency
agreement. These local governments in turn contracted with private companies to house the
prisoners. By requesting information about only those facilities with which correctional
agencies in the surveyed jurisdictions contracted, private facilities receiving prisoners
through the intermediary of a local government were not represented in our survey.17 The
numbers of prisoners in these facilities can be estimated from the information provided in the
survey, however. All surveyed jurisdictions reported a total of 52,370 prisoners housed in
privately operated facilities on December 31, 1997, 14,719 more than they accounted for in
the 91 contracts they identified as being active on that date. We assume that those prisoners
housed in privately operated facilities through intergovernmental agency agreements were
included in the larger number. We cannot assume that all of these nearly 15,000 prisoners
were held in secure facilities, however. Because of some ambiguity in the wording of the
question posed to correctional administrators, some respondents may also have counted
among these 14,719 inmates those held in non-secure community-based private facilities.
Case Studies

Case studies were conducted to examine practices in a small number of selected correctional
agencies. These include the state correctional departments in Oklahoma, Texas, and Florida.
(A separate agency exists in Florida, the Florida Privatization Commission, which is
independent of the Department of Corrections.) These particular states and the federal
government conduct a disproportionately large amount of the prison privatization business in
this country. At the end of 1997, there were 37,651 prisoners housed in privately operated
facilities for state or federal departments of correction. Forty-two percent of these, or
approximately 15,700 persons, were sent there by these three states. These agencies are also

17

The two exceptions to this were the two states that reported such IGA agreements as though they were contracts; their responses are
included in this report.

Abt Associates Inc.

Introduction

xxix

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

among those that house the largest percentage of their prisoners in privately operated
facilities. Texas had the second-largest number of their prisoners in privately operated
facilities—7,223, or five percent of its total population at year-end 1997 (the Federal Bureau
of Prisons had the largest number: 9,951). Oklahoma reported having 4,588 prisoners
contracted out, but this represented 23 percent of its total state prisoner population—the
highest proportion in the country. Florida had 3,877 in such facilities.
These jurisdictions also have the most experience with contracting for privately operated
facilities. Of the 84 privately operated facilities that were reported to be contracting at yearend 1997 with one or more state or federal correctional agencies, 28 of them were contracting
with one or more of the three agencies examined here.
The choice of these jurisdictions does not reflect a judgment that their practices represent the
best or worst of those found in the United States. Rather, we selected these largely because
they had more contracting experience than most, because they made greater use of the private
sector than most, and because their contracting practices have evolved over the years.
A Roadmap to This Report

Although not labeled as such, the following consists of two parts. The first, encompassing
Chapters 1 through 3, examines some general issues pertaining to contracting for
imprisonment services. Chapter 1 addresses reasons for contracting (or “strategic
objectives”), and distinguishes between two different private imprisonment markets that exist
to serve these objectives. Chapter 2 examines selected issues involving how contracts are
structured. These include how the scope of the contract is defined and whether it includes
management services only for government-owned facilities or involves purchasing beds in
privately-owned facilities. The implications of these different types of contracting
arrangements are explored in Chapter 3. That chapter also discusses how different contracts
structure payments to private firms and examines briefly the differences in incentives and
risks that flow from these arrangements. Considerably more attention is then given to the
development of performance-oriented contracts that aim to accomplish specified outcomes
rather than contractual agreements that require compliance with procedural rules and
regulations. Throughout these first three chapters, references are made to specific
developments in various states, but the focus is on contracting practices generally.
The second part, including Chapters 4 through 6, examines the experiences of three states in
contracting for imprisonment–Texas, Florida, and Oklahoma, respectively. Each chapter
discusses why the states chose to engage the private imprisonment industry and how it has
implemented this choice. Attention is then given to assessing, briefly, the extent to which the
state has achieved its objectives.

xxx

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

1.

Why Contract? Governments’ Objectives in
Turning to Private Imprisonment Firms

Even though questions about costs and savings associated with contracting appear paramount
with analysts who have compared private and government-operated prisons, a review of
recent history suggests that these considerations have been of secondary importance to
legislators’ and correctional administrators’ need to rely on the private sector to provide
needed beds quickly. This is not to say that the prospects or, at least, hopes for savings have
been insignificant. Claims and beliefs about the cost-effectiveness of privately delivered
services have certainly provided important ideological supports for decisions to turn to the
private sector rather than meeting the increased demand for prisons with more governmentrun facilities. But contracting offers other advantages, and may serve governments’ interests
regardless of whether taxpayers’ monies are saved in the process.

Turning to the Private Sector to Acquire Needed Resources
Expeditiously
Although entrusting the care and keep of prisoners to private businessmen had a long and
controversial history in the Nineteenth Century, the contemporary version of this began in the
late 1970s. With little notice, correctional agencies had began, during the late 1960s, to enlist
small, generally not-for-profit, organizations to operate halfway houses, work release
facilities, and other “community based” facilities. In 1979, the U.S. Immigration and
Naturalization Service began to contract with private organizations to house illegal aliens
being detained for deportation hearings. From this seedbed emerged the contemporary
private prison industry, and several of the most important players got their start there.
This was not seen as “privatization,” in part because that term and the ideological baggage
that accompanied it did not become current until partway into the Reagan Administration, in
the 1980s. Instead, contracting out for whole facilities of this sort was seen as little different
than contracting for a wide variety of other discrete services—which was a longstanding
practice in most jurisdictions. There are numerous reasons why “buying” rather than
“making” particular services makes sense to public managers, beyond any potential for cost
savings, and these by and large explained why contracting emerged. Contracting, for
example, provides a way for managers to focus their attention and other valuable resources in
their core business activities rather than on the details of producing all needed goods and
services.1 Thus, school boards have found it sensible to contract with private food service

James W. Culliton, Make or Buy: A Consideration of the Problems Fundamental to a Decision Whether
to Manufacture or Buy Materials, Accessory Equipment, Fabricating Parts, and Supplies, Harvard
University Graduate School of Business Administration, Business Research Studies, vol. 24, no. 4 (Dec.
1942); Michael O’Hare, Robert Leone, and Marc Zegans, “The Privatization of Imprisonment: A Managerial
1

Abt Associates Inc.

Why Contract?

1

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

providers to operate school cafeterias so that principals can devote their attention to
educational matters.
Community-based organizations had the resources to provide to correctional departments,
and it was simply easier in many instances to rely on these facilities than to build or to find
and renovate facilities in urban centers. These resources could also be abandoned easily if
the political tides changed with respect to these halfway-out facilities. Moreover, provision
of work and educational release in these halfway house settings appeared somewhat
peripheral to the more central mission of correctional agencies—the secure imprisonment of
convicted offenders. Managers could retain their focus on their core mission by delegating
such peripheral services to others in the private sector.
For the INS, the speed of acquiring needed detention space was the dominating motive for
contracting. The cycle of requesting capital construction in the federal government and
finally getting a new facility “on line” is a long one, lasting years. Faced with increasing
demands for space to detain illegal immigrants, the INS chose to contract with private
providers who would build facilities quickly, using private sources of capital. Because
payments to these firms could be charged against INS budgets for operations rather than for
capital investments, even though some portion of those payments went to pay off the firm’s
capital investment, the INS could sidestep federal requirements for procuring new
construction. Indeed, within a few months of signing a contract with INS to detain prisoners
near Denver, Wackenhut Corrections Corporation built and opened a 150-bed facility.
In the early 1980s, state governments began to be interested in contracting with privately
operated prisons, in many instances because contracting overcame similar constraints on
speedy expansion of government-owned prisons. By the mid to late 1970s, the demand for
prison cells began to intensify. State and federal legislatures began, in the mid-1970s, to
reform criminal sentencing laws to stiffen punishments meted out to criminals. At roughly
the same times, lawmakers also declared a “war” on drug abuse, and drug users and
traffickers began to be send to prison in large numbers, and for a long time. This reversed
what had been a decade-long downward trend in prison populations. In 1972, the number of
prisoners in state and federal prisons declined to 200,000, and the incarceration rate that year
dropped to a level not seen in this country since the 1920s. Within ten years, the number of
prisoners behind bars in state and federal prisons had doubled. By 1998, the nation’s prisoner
population had grown to 1.2 million.
Governments undertook a massive prison building campaign to supply the needed cells and
beds, but capacity often lagged behind need. By the mid-1980s, for example, state and
federal prisons held more offenders than they were built to accommodate. By 1995,

Perspective,” in Douglas McDonald (ed.), Private Prisons and the Public Interest.
2

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

governments had still not caught up with the demand for prison cells. Crowding made it
difficult to manage prisoners and jeopardized the fragile peace that exists in these institutions.
Federal courts also found crowding and various other conditions of confinement in many
prisons throughout the country to be so deficient as to constitute a violation of the Eighth
Amendment’s prohibition of “cruel and unusual punishment.” The courts demanded that
governments remedy these shortcomings. By 1988, prisons and jails in 39 states, plus the
District of Columbia, US Virgin Islands, and Puerto Rico were under such court orders. The
one-two punch of swelling prisoner populations and federal court orders has had huge fiscal
implications for states.
Expenditures of this magnitude were not sufficient to meet the demand for prison cells,
however, which fostered a private market in imprisonment services to emerge. The ability of
governments in many states to build all the prison cells they needed was constrained by the
various expenditure controls and revenue restrictions that were passed during the “taxpayer
rebellion” that began in California with the passage of Proposition 13 in 1978. Voters were
rejecting requests by governments to borrow for prison construction, and constitutional debt
ceilings were being reached in many states. Faced with a swelling sea of prisoners, lawsuits
claiming unconstitutional conditions, and a restricted ability to build prisons fast enough,
governors, legislators, and correctional administrators began turning to the private sector for
needed beds. As in the federal government, state governments were allowed to contract with
firms that provided both the buildings and the management and could charge these
expenditures against their operations budgets. Requirements for capital construction projects
and taking on public debt were thereby averted.
The importance of contracting as a means of expanding needed prison capacity is evident in
the answers state correctional directors gave in the Abt Associates survey regarding their
states’ reasons for contracting. Those correctional administrators who had active contracts
with private imprisonment firms at year-end 1997 were also asked to report their
government’s objectives for contracting and to rank the relative importance of these
objectives. From their responses, it is clear that contracting was undertaken in most states
primarily to reduce overcrowding and to acquire beds quickly (Table 1.1). Lowering the cost
of operations or construction was reportedly of secondary importance in all but eight states.
(In these eight states, cost savings were paramount.) In all, 86 percent of the 28 responding
jurisdictions cited reduction of overcrowding as an objective of any rank, whereas about half
(57 percent) cited hopes of cost savings. Improving service quality by contracting was given
as the most important reason in only one state, although nearly half (43 percent) of the
contracting jurisdictions reported interests in improving service quality, even though this
objective was ranked most often as being somewhere between the fifth and ninth most
important reason.
The survey asked the 25 jurisdictions that reported not having contracts with private
imprisonment firms on December 31, 1997 their reasons for not doing so; seventeen provided

Abt Associates Inc.

Why Contract?

3

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

reasons. Four reported not having contracts because their prison systems either already had
sufficient space or because planned construction was expected to be sufficient to handle the
demand. Three others cited legal prohibitions that prevented them from contracting; four
reported that contracting was not under consideration because of concerns about labor
relations or because of labor opposition; two reported that the issue of contracting was under
study or that a decision was pending; two were not convinced that cost savings would result;
one reported that no funding was available for such contracting; and one reported having
concerns about accountability and the quality of private management.
Table 1.1
Reported Objectives, by Rank, for Contracting with Private Correctional Firms

1st

2nd

3rd

4th

5th-8th

Number (and Percent) of
States Citing this
Objective

Reducing overcrowding

14

2

3

3

2

24 (86%)

Speed of acquiring additional beds

2

9

4

1

4

21 (75%)

Gaining operational flexibility

1

0

8

2

5

17 (61%)

Operational cost savings

8

3

1

3

1

16 (57%)

Construction cost savings

0

6

3

4

3

16 (57%)

Improving caliber of services

1

0

0

1

9

12 (43%)

Reducing legal liability exposure

0

0

1

2

7

11 (39%)

Other

3

2

0

0

1

6 (21%)

Rank

Note:

One state listed several objectives without ranking them. Its responses are counted in the column on the right summing the
responses, but not in the columns indicating rank. The sum total reported here does not, therefore, always equal the total of
the ranked objectives.

Source:

Abt Associates survey of state and federal correctional administrators.

These data may oversimplify the reasons why state or federal departments of correction
sought to rely on private firms rather than on expanding government-operated facilities.
Discerning any particular government’s principal reason for contracting is often problematic.
Public agencies and public programs often lack clear statements of mission and purpose, and
their stated goals are often multiple, ambiguous, and even contradictory. This reflects in part
the fact that decisions often result from agreements and conflicts among different political
actors who are motivated by different political interests. The survey recorded the correctional
directors’ representations of the reasons, and these may not correspond precisely with
answers that legislators or governors would have given to the question.
How legislators and governors would have answered the question is especially relevant,
because these were reportedly the most active initiators of efforts to contract for private
4

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

imprisonment. The Abt Associates survey of state and federal correctional administrators
asked who initiated the decision to contract for private imprisonment services in their
jurisdiction. In jurisdictions that contracted for the management and operations of entire
prisons, the initiatives to do so were most often taken by legislatures or governors, and not by
correctional administrators. That is, of the 28 jurisdictions that had active contracts in the
closing days of 1997, the legislature was reportedly the source of the initiative in 11
jurisdictions and the governor in five. Correctional administrators were identified as the
initiating agents in seven, but this may not be a true reflection of their role. In at least one
jurisdiction where the chief executive initiated privatization, the director of the correctional
agency took credit for the initiative to contract out. In three states, the federal courts were
identified as the source of the initiative, but this was probably not an accurate depiction of
how contracting came about there. The courts had found conditions in these states’ prisons to
be in violation of constitutional standards, but judges do not typically determine the means by
which these are to be remedied.
These findings are not surprising, in that the “privatization” efforts in this country, at least
with respect to prisons, have rarely involved conversion of government-run facilities to
privately run ones. If cost savings were indeed the most important consideration, it would be
logical to consider such conversions. Instead, in nearly all instances, the private sector has
been called upon to staff or to build and staff new facilities.

Two Private Prisons Markets
Contracting with private prisons in this country has followed two different dynamics, and
each poses different challenges to managing prison privatization. The dominant mode is for a
government agency to decide to contract for some of its needed state prison beds, and then to
seek a contractor willing to provide these beds in-state. In some instances, the state owns the
facility and signs a contract with a private firm to manage and operate it, while in other
instances the government elects to contract out the design, construction and operation of a
new facility. The most usual result, regardless of who owns the facility, is the creation of
one-to-one relationship between prison operator and the state prison system. That is, the state
prison system is the contractor’s sole client at the facility; the only prisoners held in the
facility are those under the jurisdiction of the client state agency. Moreover, the prison is in
the same state as the publicly operated prisons, which creates at least some of the conditions
supportive of a close integration between the publicly operated facilities and the privately
operated prisons. In these circumstances, the state’s management of the contractor is
typically close. The procurement and contracting processes provide an opportunity to spell
out quite precisely the specific rights and responsibilities of the contractor and the state. The
process of monitoring the contract creates the opportunity for close and ongoing involvement
by the state’s managers and their representatives, although not all states elect to conduct close
monitoring.

Abt Associates Inc.

Why Contract?

5

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

The second general pattern of contracting for prison beds poses different challenges for state
management. Rather than waiting for the states to issue a call for service, some private firms
take the risk of building facilities without first being assured of any prisoners from a
particular correctional department. (These are often called “spec” prisons, built as
speculative ventures by private correctional firms.) Once built and staffed, they advertise
their availability to correctional and law enforcement agencies anywhere in the country that
are in need of prison beds. Not all firms succeed in attracting prisoners. Those that do may
hold prisoners from a variety of different agencies, both out-of-state as well as from the state
in which they are located. As such, these facilities are oriented to what is essentially a
national market, in contrast to those facilities that are brought into being as a result of a state
or federal government's issuance of a request for proposals and subsequent awards of
contracts. Many of these facilities that are oriented to the national market may not have any
prisoners at all from the correctional agencies in the states in which they are located. Indeed,
they may have no relationship at all with the state governments in these states, other than an
obligation to pay corporate income taxes. Owners of private property do not need licenses
from state correctional agencies to build and operate detention facilities and, until recently,
most state legislatures have not established regulatory systems for the states or localities to
govern private prison operations.
The emergence of these latter types of facilities is not new. During the mid-1980s, a number
of small correctional firms were buying or building detention facilities with an eye to offering
beds to a regional or national market. These included Joe and Charles Fenton’s firm in
Pennsylvania, as well as Philip Tack, who created the 268 Center in Cowansville, a small
town down on its luck in economically depressed western Pennsylvania. Tack arranged with
the District of Columbia’s correctional agency to take fifty-five of the District’s prisoners to
relieve overcrowding in the D.C. jails. This alarmed the townspeople, who had initially
welcomed the center and the jobs it brought, and they began to patrol the streets with
shotguns, fearing escapes. This caught the eye of the press and, consequently, organized
labor, which spurred the state legislature into passing a moratorium on private prisons in the
state.2 This constellation of events foreshadowed those in Youngstown, OH over a decade
later.
Texas has been home to the largest number of these facilities. At the end of 1997, there were
23 privately owned and/or operated correctional facilities in the state, not including six
privately managed jails under contract with the state and some others not reported because
state or federal government agencies relied on intergovernmental agency agreements to place
prisoners in them. As of that date, the state had no contractual or any other type of
relationship with nine of them.3 All of these nine housed prisoners from other states.

2
3

6

See Aric Press, “Private Prisons in the 1980s,” in Douglas McDonald (ed.), Private Prisons and the Public Interest (New Brunswick,
NJ: Rutgers Univ. Press, 1990), pp. 31-32.
Abt Associates survey of state and federal correctional agencies, conducted in 1998.

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

Including those in Texas, the 1998 Abt Associates survey identified 84 facilities that were
under contract with state correctional agencies to hold prisoners outside the contracting
states’ boundaries. In addition, as mentioned above, there were other such facilities that were
not so identified because state and federal governments were asked to list only those with
which these governments contracted to hold prisoners. Because some governments did not
contract directly with these facilities, they chose not to include them in their lists. Rather
than entering into a contractual agreement with the private prison operator, state and federal
agencies have used other available means of securing beds in private prisons. One such
method is an intergovernmental agency agreement, negotiated between a state correctional
agency that elects to send its prisoners to a private facility located in another state and a
sheriff of the county in which the facility is located. Another method is to use the authority
of interstate compacts that have been established to transfer legal jurisdiction over inmates
from one government entity to another. Once legal jurisdiction over inmates is transferred to
local sheriff, the sheriff then transfers the prisoners to the privately operated facility by means
of a contractual arrangement between the local sheriff and the facility. These sheriffs thereby
act as intermediaries between out-of-state correctional agencies and private prisons.
Table 1.2 shows the names and host states of facilities identified in the Abt survey that held
prisoners from out-of-state on December 31, 1997. In addition, it shows the states from
which these inmates were sent. As shown, most of these facilities served only one out-ofstate government, but five held prisoners sent by more than one state government.
The national and local in-state markets differed in a number of significant ways. Two-thirds
of the agreements between state governments and nationally-oriented facilities were noncompetitive interagency agreements authorizing prisoner transfers. The rest resulted from
competitive bidding. In contrast, 70 percent of the contractual agreements between state
governments and privately operated prisons located within their states’ boundaries involved
competitive bidding. None of the agreements with the nationally-oriented facilities specified
whether inmates shall be housed in single-bed or double-bunked facilities, whereas half of
the agreements with in-state facilities did so. Moreover, half of these nationally-oriented
facilities were smaller and held fewer than 200 inmates; the rest held between 201-600
inmates. In contrast, only 20 percent of the facilities serving their host state governments had
few than 200 inmates, and a third had more than 600.
The majority of agreements between states and out-of-state facilities had terms of two years
or less, where most of the contracts between state governments and in-state facilities were
longer—between two and five years, and some longer. The per diem rates paid for out-ofstate facilities were also higher than for in-state contracts. All agreements between state
governments and out-of-state facilities that sold services on the national market specified
payment rates in excess of $35 a day, whereas 55 percent of the contracts between states and

Abt Associates Inc.

Why Contract?

7

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

Table 1.2
Privately Operated Facilities Servicing a National Market, by Location and Governments
Contracting with Them (Includes only facilities contracting with state or federal correctional
agencies on December 31, 1997)
Facility
Basile Detention Center
Central Arizona Detention
Center

Location

Client States

Louisiana

Idaho

Arizona

Contracting Partners/Operator

Louisiana Corrections Services

Montana Corrections Corporation of America
New Mexico
Oregon

Central Texas Parole Violator
Facility

Texas

Texas
Oklahoma

Crystal City Correctional Facility

Texas

Hawaii

Bobby Ross Group

Dickens County Correctional
Facility

Texas

Hawaii

Bobby Ross Group

Frio Detention Center

Texas

Idaho

Correctional Services Corporation

Hardeman County Correctional
Center

Tennessee

Wackenhut Corrections Corporation

Tennessee Corrections Corporation of America
Indiana

Karnes County Correctional
Facility

Texas

Oklahoma

Bobby Ross Group

Limestone County Detention
Facility

Texas

Oklahoma

Capital Correctional Resources, Inc.

Mansfield Law Enforcement
Center

Texas

Oklahoma

Mansfield Public Finance Authority

Newton County Correctional
Facility

Texas

Hawaii
Oklahoma

Bobby Ross Group

Odessa Detention Center

Texas

Oklahoma

Gil R. Walker, Inc.

Prairie Correctional Facility

T. Don Hutto Correctional Center
West Tennessee Detention
Source:

8

Minnesota

Colorado Corrections Corporation of America
Idaho
Minnesota
North Dakota

Texas

Wyoming

Corrections Corporation of America

Tennessee

Montana

Corrections Corporation of America

Abt Associates Inc. 1998 Survey of State and Federal Correctional Agencies

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

in-state facilities specified lower per diem rates. Nationally-oriented facilities having
agreements with more than one client government also charged different per diem rates to
each client. One facility–the Hardeman County Correctional Center in Tennessee–charged
the state of Indiana $45/day, while Tennessee paid $32.47/day. The State of Tennessee did
contract for a much larger number of inmates, however: 1,521 on December 31, 1997, as
opposed to Indiana’s 69 inmates.
There were also significant differences in the states’ monitoring practices at these out-of-state
facilities. These are discussed in Chapter 3.

Abt Associates Inc.

Why Contract?

9

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

10

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

2. The Contractual Structure: Selected Issues
Designing the contractual relationship between governments and private imprisonment firms
involve myriad considerations. This chapter does not attempt to address all such
considerations; more comprehensive examinations are available elsewhere.1 The discussion
here is limited to issues associated with (1) the risks of contracting with private firms to
provide facilities as well as management services, (2) risks associated with different payment
provisions, and (3) performance-oriented contracts.

Facility Ownership and the Risks of Entrenchment
Many of the purported benefits of contracting depend upon the competitiveness of the private
imprisonment market. To the extent that firms compete with one another for contracts, there
will be a more pronounced tendency toward cost control, and buyers in his marketplace (i.e.,
governments) will have more options for picking the most attractive and advantageous bid. If
governments have the option of canceling or not renewing a contract and signing up another
provider, firms will be under pressure to perform effectively or risk going out of business.
Moreover, a firm’s performance will be measured not only against standards agreed upon in
contracts but also against what the competition offers. On the other hand, if monopoly
conditions prevail or if a single provider is entrenched in a particular state, the government
will lose freedom of action and may become excessively dependent upon the private
provider. The ostensible advantages of privatization may thereby evaporate. (Governments
may, of course, elect to turn back to full reliance upon the public correctional agency and
thereby withdraw from the market.) This suggests that how governments structure the
privatization program, and decisions that they make about allocating ownership and
management responsibilities, have an effect on the competitiveness of the marketplace and,
by extension, the government’s leverage over future suppliers.
There is substantial variation in how governments have structured the overall contracting
arrangement with respect both to ownership of the facility and to the obligations to manage
and operate it.
•

1

The narrowest contracts are for selected services such as contracting with food service
companies to operate prison cafeterias, or with private firms to deliver prison health
care services. These specific-service contracts are common and are not examined
here.

See, for example, William Collins, “Contracting for Corrections Services Provided by Private Firms”; also see Malcolm RussellEinhorn, “Legal Issues Relevant to Private Prisons” in Douglas McDonald, Elizabeth Fournier, Malcolm Russell-Einhorn and
Stephen Crawford, Private Prisons in the United States: An Assessment of Current Practice (Abt Associates Inc., 1998, at
www.abtassoc.com.)

Abt Associates Inc.

The Contractual Structure

11

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

•

Governments also contract for full-service management and operation, but retain
ownership of the facility.

•

Another approach is to contract for beds in facilities operated by private firms and
owned by entities other than the client government, either by the operating firms
themselves, or by other private entities that partner with operating firms. These
facilities typically operate in the second tier of the private prisons market, which we
have termed the “national” market; these facilities offer beds to any and all client
agencies. As such, any single government does not have exclusive rights to use of the
facility.

•

Still another approach is for a government to contract with a firm to finance,
construct, and operate a facility that will exist for the (largely) exclusive use of that
government. For example, the INS contracted with the Wackenhut Corrections
Corporation to finance, build, and operate a detention center outside of Denver to
hold illegal immigrants. This type of arrangement is similar to a contractor-owned
and operated facility that serves client governments in a national market, but differs in
that facilities come into being not as a speculative investment but as a consequence of
a contract with a client government that has exclusive use of the facility during the
term of the contract (although some provision might exist for renting some unused
beds to other agencies’ prisoners).

•

Another approach is for a government to create a private corporation that exists
entirely to serve the interests of that government, and which assumes responsibility
for financing and constructing a correctional facility. These corporations are
effectively controlled by public officials as they sit on the boards of directors,
although the entity is legally independent of the government. The facility is then
leased to the government. The government’s correctional agency can chose either to
staff it directly, or the government can contract with a private management firm to
operate it. Governments have created these nominally private corporations to own
and finance facilities because these corporations can sell corporate bonds in the
capital markets and are not subject to the debt restrictions and voter-approval
requirements that many state governments face.

There exist some variants of each of these broad types. For example, town or county
governments may decide to get into the private prison business by creating special purpose
authorities to finance and construct prisons as speculative investments, partnering with an
established management firm to operate the facility, and then offer its beds to the national
market. As discussed in Chapter 6, public officials in a small Oklahoma town took this route
to develop its local economy.

12

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

In the Abt Associates 1998 survey of correctional state and federal correctional agencies,
respondents reported that 34 of the 84 facilities that were under contract with state and
federal governments on December 31, 1997 were owned by governments. Fifty of the
facilities were owned by the management firms that operated the facilities or by other private
entities. Seven of the facilities were owned by entities that were only nominally private, as
they were created by governments to own the facility on behalf of that government. For
example, the Industrial Development Authority of Brunswick County owns the Lawrenceville
Correctional Center on behalf of Brunswick County.
In theory, at least, contracting with entities that own facilities as well as operate them may
run the risk of allowing the contractor to become entrenched and thereby to minimize or
eliminate subsequent competitions.2 Economist Oliver Williamson has theorized that when
services are provided through contractual arrangements, competition among providers is
severely constrained if the assets specific to a contractual relationship are substantial and not
easily redeployed.3 To the extent that such assets exist, the original contractor will have an
advantage over other bidders and parity among them will not exist. Incumbents will already
have a facility in place at the end of the first contract period, and revenues from the first
contract may have paid off a substantial part of the initial investment. Ownership of the
needed asset may pose a barrier to entry to other potential bidders. The number of suppliers
willing to offer their services at competitive prices will be fewer than in a market where no
transactional-specific assets have been developed. Buyers may thereby become more
dependent upon the original contractor and may find it more difficult to exit in the event of
inadequate performance. In these circumstances, it is hypothesized that the power of the
buyer is reduced, the cost to the buyer of ending the contractual relationship increases, and
the supplier’s ability to exact higher fees for its services is consequently strengthened.
The survey conducted by Abt Associates of state and federal correctional agencies was
designed to test this hypothesis, but the results are not conclusive one way or the other.
Agencies that contract for privately operated prisons were asked if the facilities were owned
by their governments or by private entities, and were also asked if the incumbents won or lost
the most recent recompetitions for the contracts. Among the 91 contracts that were active at
year-end 1997, only 17 had been awarded following expiration of a previous contract. The
remainder were either still under the first contract signed, or were procured using noncompetitive means (e.g., sole source procurement). Of the 17, incumbency provided an
overwhelming advantage: all but one of these were awarded to the incumbents. (The single
contract that was not awarded to an incumbent was one where the facility was publicly
owned, where eight different firms competed, and the award went to the lowest bidder.)

2

Douglas McDonald, “When Government Fails: Going Private as a Last Resort,” in McDonald (ed.), Private Prisons and the Public
Interest (New Brunswick, NJ: 1990), pp. 194-196.

3

Oliver E. Williamson, “What is Transaction Cost Economics?” Working Paper No. 1014 (New Haven, CT: Institute for Social and
Policy Studies, Yale University, May 1984).

Abt Associates Inc.

The Contractual Structure

13

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

Whether the contracts were awarded to the lowest bidder was known by respondents for 13 of
the competitions. Incumbents won 12 of the 13, and submitted the lowest price bid in all but
one. This indicates that incumbency itself is a powerful advantage.
There was no evidence that private ownership of the facilities made any difference in the
outcomes of these procurements. Only four of the facilities were privately owned; all four
firms won new contracts. Only two of these clearly resulted from competitive procurements
following issuance of requests for proposals, however. The others resulted from unspecified
other forms of procurement.

Payment Structures
Various methods have been devised to compensate contractors for their services. The
principal issue is how to structure payments for varying numbers of prisoners under custody.
Different types of contracts and payment structures distribute risks differently between the
government and to the contractor, and thereby create different opportunities for the contractor
to profit.
For the contractor, the principal financial risk is not having enough prisoners and consequent
revenues to offset costs. The costs of operating prisons are largely fixed, as the staff,
equipment, and utilities required to operate a prison remain relatively constant even if the
numbers of prisoners are substantially below or above planned capacity. Therefore, it is in
the contractor’s interest to maintain a stable number of prisoners, either from a single or
multiple government agencies, or to have financial guarantees regardless of the numbers of
prisoners sent to them. Governments have different interests. Some may want to have the
resource available and to pay only when used; others may want to protect themselves against
having to pay more than they have budgeted for. There exist a number of different
arrangements for paying private firms for their services that seek to balance the different
financial interests of contractors and governments.
A common approach that affords the government great flexibility is to use a fixed-price,
indefinite delivery/indefinite quantity (ID/IQ) contract so that prison beds are available on an
as-needed basis. These contracts provide for an indefinite quantity of a service or supply,
within stated limits, to be furnished during a fixed period, with deliveries to be scheduled by
placing orders with the contractor. The government defines its requirements but puts
contractors on notice that the delivery dates required and the quantity of the services are
unknown at the time the contract is issued. The contractor is paid a fixed price per prisoner,
per person/day used. This price may include all the contractor's costs associated with housing
and safeguarding a prisoner, although limits might be placed on the contractor’s liabilities for
health care costs, as discussed below. Contractors may also be allowed to bill separately for
specified unpredictable costs, such as transporting prisoners to and from airports, or to and
from off-site health care facilities.
14

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

These pure ID/IQ contracts serves the government’s interests well because the government
will not be obligated to pay for services that it does not use (unlike a fixed-price contract).
But the contractor bears substantial risk in this arrangement. Its profits and financial health
are at risk if the government fails to request a sufficient level of services to offset the
contractor’s costs of creating the resource. Most of the costs of providing imprisonment
services are fixed, as the numbers of staff, equipment, and utilities required to operate a
correctional service remain relatively constant even if the numbers of prisoners are
substantially below or above planned capacity.
To minimize the risk to the contractor of insufficient demand, contracts may specify a
guaranteed minimum number of person/days to be ordered during a contract period, as well
as a guaranteed minimum dollar amount that the government is obligated to pay. The
contractor is thereby entitled to receive an order of the specified minimum quantity, even if
the government does not subsequently order that quantity of services. The higher the
minimum order specified in the contract, the lower the financial risk to the contractor.
Setting a high minimum quantity creates a higher risk of unnecessary government spending,
however. If the government is unwilling or unable to use the contractor’s imprisonment
services at the level guaranteed, it is effectively paying the contractor for services not used.
In the event that the government requests considerably more imprisonment services than
planned, the contractor’s opportunities for profit increase substantially if the contract
obligates the government to pay the full per diem for these higher numbers of prisoner/days
consumed. The government is thereby liable to pay considerably more for a contracted
imprisonment service than it would if the facility were operated directly by the government.
This is because the marginal per diem costs of the government imprisoning in its own
facilities 10 or 20 or even 30 percent more prisoners than planned and budgeted are low,
relative to the average cost per diem. If the government pays a per diem rate that
approximates the contractor’s average per diem cost at planned levels, the difference between
the contractor’s actual marginal costs and the paid per diem rate can be taken as profit.
If the government is able to regulate its need for prison space well, the opportunity for such
“windfall profits” reaped by a private contractor will be slim. In some instances, however,
governments may not be able to regulate demand for cells. For example, Hamilton County,
Tennessee, contracted in the mid-1980s with the Correctional Corporation of America to
operate a 412-bed penal farm at Chattanooga. The county agreed to reimburse CCA at a
fixed rate of $21 per day for each inmate in custody. Shortly after signing the contract, the
number of prisoners rose dramatically—about 40 percent—because county judges began
taking a sterner line on drunk driving offenses, and because intake of new prisoners by the
state's prisons slowed, backing prisoners up in the county facility. CCA's billings were far

Abt Associates Inc.

The Contractual Structure

15

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

higher than anyone expected, as a consequence, and the county's expenditures were $200,000
over budget in less than a year.4
One method of distributing financial risks posed by unpredictable and varying demand for
detention services is to establish a base per diem rate to be paid for a certain specified level of
services, and different rates to be paid under conditions where demand is lower or higher than
this specified level. In the event of greater demand for services than specified, establishing a
considerably lower per diem reimbursement rate for the additional prisoner/days may benefit
the government substantially. That is, a lower rate that approximates the contractor’s actual
marginal costs plus a reasonable margin for profit will eliminate the possibility of a large
windfall profit and a correspondingly large “windfall expenditure” by the government.
Fixed-price contracts allocate risks differently. The contractor will not be able to be
reimbursed for any marginal costs associated with high levels of use, but will be able to keep
revenues and profit from lower-than-expected levels of use. The financial risk to the
contractor can be protected by establishing a firm upper limit on the number of prisoner/days
to be provided during the contract period. As the government is obligated to pay the
contractor the same amount regardless of demand, this contractual arrangement places the
burden on the government to keep demand for imprisonment services high.
The Federal Bureau of Prisons’ contractual arrangement for management and operation of a
prison in Taft, California is structured as a fixed price contract with an option for a variable
per diem rate above a specified population level. The facility was constructed by the
government for a total of 2,000 adult male inmates, including 275 in the Institution Hearing
Program under INS deportation orders. The solicitation sought a firm fixed price, award fee
contract. The fixed price was to include all costs for operation and maintenance of the
facility, including medical services, associated with an average daily population of 1,946
inmates. The Bureau requested industry comments when it was planning the procurement,
and industry representatives voiced a preference for the fixed price contract over an
indefinite quantity contract based on a unit price per inmate day. In this instance, the Bureau
agreed with industry, recognizing that the contractor must operate the institution regardless of
the variation in the inmate daily count.5
The contractor was allowed to bid, in addition to the fixed price, a fixed incremental unit
price per inmate day that will be paid if the inmate population exceeds 1,946, up to a
maximum of 2,355. In the contract awarded to The Wackenhut Corrections Corporation, this
“incremental unit” price is substantially lower than the average per prisoner/day price for the

4

Aric Press, “The Good, the Bad, and the Ugly: Private Prisons in the 1980s,” in McDonald (ed.) Private
Prisons, page 29.

5

Transcript of Pre-Proposal Conference, December 11–12, 1996, pp. 12–13.

16

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

first 1,946 prisoners. How closely this approximates the marginal cost to the contractor of
housing these additional prisoners is unknown. The much lower unit price does offer
significant protection to the government against high costs associated with unanticipated
increases in demand for bedspace, however.
Examples of other payment structures include the following:
•

A Texas RFP for one-to-three 500-bed prisons offers a fixed price contract but
allows for an adjustment to the price if fewer than a minimum number of
inmates are assigned to the facility.6 Texas pays the contractor the greater of the
“monthly operator” payment or the minimum monthly payment. The monthly
operator payments equals the operator daily rate times the number of inmates
who occupied the facility times the number of days. The minimum monthly
payment is the operator per diem rate (a fixed daily rate) times the minimum
number of inmates agreed upon for the facility times the number of days in the
preceding calendar month during which the facility was available for use and
occupancy.

•

Kentucky reimburses its contractor at a negotiated price per inmate/day. The
price per day includes all costs associated with operation of the facility with the
exception of health care costs. Payments are made monthly within thirty days
after receipt of an itemized invoice which details the names and identification
numbers of the inmates assigned to the facility and the number and listing of
calendar days in which the inmates resided in the facility. No payments are
made for inmates who are absent from the facility, for any reason, in excess of
three days.7

•

North Carolina reimburses the contractor on a per diem inmate cost which
includes all operating costs. The North Carolina RFP examined here includes
provision for potential cost sharing between the state and the contractor for
inmate health care.8

•

Tennessee’s payment terms are quarterly allocations of the fixed annual contract
cost. If fewer than 160 out of the 180 detainees contracted for are housed in the
facility, the price is reduced by $7 per day.9

6

Texas Department of Criminal Justice Facilities. “Request for Proposals for the Operation of from One to Three Facilities,” 1995.

7

Commonwealth of Kentucky Department of Corrections, December 1993, Contract for the Operation of Three Minimum Security
Facilities.

8

North Carolina Department of Correction, December 1995, Request for Proposals for Two 500-Cell Medium Security Institutions.

9

State of Tennessee Department of Corrections, October 1986, Request for Proposal to Manage and Operate Carter County Work
Camp.

Abt Associates Inc.

The Contractual Structure

17

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

•

California reimburses the contractor based on an approved per diem rate which
is expected to be the not-to-exceed-cost ceiling divided by the number of
anticipated inmates.10

Determining which payment structure best suits the government should turn on an
identification of the various risks that the government faces during the contracting period,
and its ability to manage the demand for prison cells. Decisions to lessen the contractor’s
financial risks are likely to produce lower cost bids (assuming adequate competition among
bidders). Health care costs, for example, may be difficult for a contractor to predict in the
absence of detailed information about the inmate population to be housed. If a contractor is
expected to cover all costs of prisoner health care, the bid price is likely to be higher than if
there is a cap on such liabilities. A small number of prisoners with AIDS, for example, can
require large expenditures for health care.

Designing Performance Requirements
In general, state and federal governments demand in their contracts that privately operated
facilities perform like their public sector counterparts. Statements of work in the contracts
generally specify that compliance will be required with the same procedural rules,
regulations, and standards that are in force in the public facilities. Although the Abt
Associates survey did not explicitly ask if compliance with departments of corrections’ rules
and regulations was required of contractors, monitors for eleven contracts volunteered that
such compliance was indeed specified as a performance objective in the contract. An
examination of a selected number of contracts also found this to be true. Correctional
administrators also reported that 57 of the 91 contracts in force at the end of 1997 required
that facilities achieve ACA accreditation within a specified time. In addition, administrators
reported that 61 contracts explicitly required compliance with conditions established in
consent decrees or other court-mandated standards.
There is no doubt that inclusion of such procedural requirements in contracts is necessary, if
only to protect the state against lawsuits. As to the matter of safeguarding inmate rights, it is
generally accepted that private prisons are treated as “state actors” for purposes of civil rights
suits, and that all relevant constitutional requirements apply with equal force to private as
well as public correctional facilities.11 Moreover, private prison employees are not covered
by the “qualified immunity” laws that shield public correctional authorities who reasonably

10

State of California Department of Corrections, October 1995, Community Correctional Facility Male Bed Expansion Request for
Proposals.

11

See West v. Atkins, 487 U.S. 42 (1988). See also, e.g., Street v. Corrections Corp. of America, 102 F. 2d 810, 814 (6th Cir. 1996);
Payne v. Monroe County, 779 F. Supp. 1330, 1335 (M.D. Fla. 1991).

18

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

believe that their discretionary actions are lawful.12 Finally, private prisons and officials are
not protected by other governmental immunities that may otherwise limit the monetary
damages available to inmates suing over prison conditions.13 Compliance with national
standards and laws provides governments a measure of protection against lawsuits.
Contracts that go beyond requiring compliance with procedural rules, regulations, and
standards, and which specify performance outcomes, are rare. One example is the contract
between the Federal Bureau of Prisons and Wackenhut Corrections Corporation for the latter
to operate the government-owned facility in Taft, California. Before describing this
particular contract, some background to performance standards and performance-based
contracting follows.

The “Making Government Accountable” Movement and Its
Implications for Contracting
The current emphasis in public administration circles on performance standards owes its
origins in the late 1970s and early 1980s to several sources. One is the durable belief in this
country, intensified during the past twenty-five years, that government is ineffectual, that
government employees are indifferent to agency performance or to citizens’ interests, and
that waste is pervasive. This belief was given official voice in the 1984 report of the
President’s Private Sector Survey on Cost Control, generally known as the Grace
Commission after its head, J. Peter Grace, the chief executive officer of the W.R. Grace and
Company. This report paraded example after example of real and apparently wasteful
practices by governments, and concluded that more than $400 billion could be saved by the
federal government during a three-year period simply by eliminating waste, without affecting
the quality or quantity of services. Although the analysis and the conclusions were incorrect,
as Steven Kelman’s critique shows, this report added fuel to demands to “run the
government like a business.”14
This dovetailed with the movement to control government spending by means of limiting
taxing authority, initiated in the contemporary era by the passage of Proposition Thirteen in
California in 1978. Within five years, fifty-one new expenditure controls or relevant

12

Richardson v. McKnight, 117 S. Ct. 2100 (1997).

13

The Eleventh Amendment and state sovereign immunity statutes may prevent inmates from suing federal or state correctional
agencies and supervisory officials in their official capacities. In these cases, plaintiffs may be limited to suits for monetary damages
against officials in their personal capacity.

14

For a critique of the Grace Commission’s report, see Steven Kelman, Making Public Policy: A Hopeful View of American
Government (New York: Basic Books, Inc., 1987), Chapter 12, and Kelman, “The Grace Commission: How Much Waste in
Government?” The Public Interest 78 (Winter, 1985).

Abt Associates Inc.

The Contractual Structure

19

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

restrictions were placed on state and local governments’ spending powers.15 This stimulated
public officials’ interest in adopting different approaches to improving government costeffectiveness by means of contracting out government responsibilities as well as by adopting
more performance-based management.
This intersected with a similar trend in the business community. Economic performance of
the U.S. began declining in the 1970s. Japanese manufacturers captured substantial shares of
the American market for highly visible consumer durables such as automobiles and electronic
equipment, with products competitive in performance, superior in manufacturing quality and
lower in cost than their American counterparts. This shocked Americans, who had grown
accustomed to technological and economic superiority during the postwar period. Lessons
began to be drawn from analyses of the Japanese industrial performance. One is that industry
in Japan has benefitted from strategic planning, and that Japanese manufacturers have
incorporated the concepts of “total quality management” and “continuous process
improvement” into their operations. Performance matters; dimensions relevant to the product
or the service are measured and monitored closely; and results are used to re-engineer further
improvements in the product or service design.
These various trends towards performance-oriented government culminated in this country in
the Government Performance and Results Act of 1993 (P.L. 103-62). GPRA aims to
improve the operations of federal government programs by changing management emphasis
from inputs and processes to performance and results. It mandates that each federal agency
develop a strategic plan describing the agency’s goals, objectives and operations. Each
agency is in turn to derive from its strategic plan an annual performance plan for each of its
program activities, defining program-level goals, operations and performance indicators.
These performance plans were to be the basis for annual performance reports to Congress.
Over the long run, it is the goal of Congress to integrate these annual performance reports
into the budgeting process.
In the words of the act, the purposes of GPRA are to:

15

20

•

improve the confidence of the American people in the capability of the federal
government by systematically holding federal agencies accountable for
achieving program results;

•

initiate program performance reform with a series of pilot projects in setting
program goals, measuring program performance against those goals, and
reporting publicly on their progress;

Thomas R. Swartz, “A New Urban Crisis in the Making,” Challenge (September-October 1987), pp.35-37.

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

•

improve federal program effectiveness and public accountability by promoting a
new focus on results, service quality, and customer satisfaction;

•

help federal managers improve service delivery, by requiring that they plan for
meeting program objectives and by providing them with information about
program results and service quality;

•

improve congressional decision making by providing more objective
information on achieving statutory objectives, and on the relative effectiveness
and efficiency of federal programs on spending; and

•

improve internal management of the federal government.

GPRA establishes a performance measurement system the elements of which are strategic
plans, annual performance plans, and annual performance reports. The act also introduces
refinements to this basic approach. The first is a “managerial accountability and flexibility”
option allowing agencies to escape existing administrative procedural requirements and
controls in return for accountability to specific performance goals. The second points to a
future extension of the concept which links program performance goals directly to budgeting.
Finally, in recognition of the extent of the change in current agency practices represented by
the provisions of GPRA, the act specifies that implementation should begin with a limited
number of pilot projects, with extension to the full range of federal agencies following
thereafter.
The Senate report accompanying the GPRA legislation emphasized that measures of program
performance should, to the extent possible, concentrate on outcomes, rather than outputs.
Outcomes refers to “the actual results, effects, or impact of a program activity compared to its
intended purpose, ” whereas outputs refer to actual levels of activity or effort that are
realized. For example: eligible clients completing a job training program are outputs, but an
increase in their rate of long-term employment is an outcome. “While recognizing that
outcome measurement is often difficult, and is infeasible for some program activities,” notes
the report, “the Committee views outcome measures as the most important and desirable
measures, because they gauge the ultimate success of government activities.”
The committee recognized that “not all governmental programs lend themselves easily to
measurable goals,” but cautioned that “managers should resist the temptation to decide too
quickly that a particular program is unsuitable for measurable goals.” The report goes on to
restate the central concern of the Congress:
The fundamental question is, what is the difference between a successful program and
a failure? Between a well-run program and one that is mismanaged? How can we tell

Abt Associates Inc.

The Contractual Structure

21

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

the difference, and how should that be defined? If the difference cannot be defined,
then is that not just an invitation to waste, inefficiency, and ineffectiveness?
Whether through precise quantitative measures or otherwise, Congress seeks explicit
judgements of whether programs have met their goals or not. If agencies provide services by
means of contracts with private providers, assessment of contractors’ performance vis-a-vis
agency goals is also needed.
Emphasis on performance objectives and performance measurement has not been limited to
the federal government. Several states have passed similar laws (e.g., Florida, Minnesota,
and Oregon ),16 a preference for performance-based contracting has been voiced by many
organizations, as discussed below; and the emphasis on seeing performance as outcomes
rather than process has seeped into public administration generally.

Performance Standards in Criminal Justice Agencies
These developments suggest that we are at a major crossroads in how we administer, manage
and control public agencies. But the application of performance-standards to criminal justice
organizations has a very short track record, with a few important exceptions.
In the late 1970s, Wildhorn, Lavin, and Pascal wrote an important study on measuring the
performance of criminal justice agencies.17 In 1981, Burt wrote a similar pioneering article
on measuring correctional agency performance.18 In 1982, the National Institute of Justice
published a series of reports on performance measurement in all areas of criminal justice.19
In 1993 the Bureau of Justice Statistics (BJS) published papers from the BJS/Princeton
Project, in which a study group of academics and practitioners commissioned and reviewed

16

Section 216, Florida Statutes, the Government Performance and Accountability Act of 1994, requires state agencies to submit
performance-based budget requests, programs, and performance measures. Chapter 186, Florida Statutes, requires each agency to
identify the measurable objectives that will be used to judge the achievement of the goals and objectives in their Agency Strategic
Plans. Also see the Minnesota Milestones and Performance and Outcomes Reporting and Monitoring System (PERFORMS)
(http://www.finance.state.mn.us/budget/bis/performs/reports.html); and Oregon Benchmarks
(http://www.econ.state.or.us/opb/os_cont.htm).

17

Sorrel Wildhorn, Marvin Lavin, and Anthony Pascal, Indicators of Justice: Measuring the Performance of Prosecution, Defense, and
Court Agencies Involved in Felony Proceedings; a Guide to Practitioners. (National Institute of Law Enforcement and Criminal
Justice, Law Enforcement Assistance Administration, U.S. Department of Justice, May 1977).

18

Martha R. Burt, Measuring Prison Results: Ways to Monitor and Evaluate Corrections Performance. (National Institute of Justice,
June 1981).

19

This series was published by NIJ in July 1982, and included the following works: Gordon P. Whitaker, Stephen Mastrofski, Elinor
Ostrom, Roger B. Parks, and Stephen J. Percy, Basic Issues in Police Performance; Joan E. Jacoby, Basic Issues in Prosecution and
Public Defender Performance; Thomas J. Cook and Ronald W. Johnson, with Ellen Fried, John Gross, Mary Wagner, and James
Eisenstein, Basic Issues in Courts Performance; and Gloria Grizzle, Jeffrey Bass, J. Thomas McEwen, Deborah Galvin, Ann G.
Jones, Harriet D. Mowitt, and Ann D. Witte, Basic Issues in Corrections Performance.

22

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

papers on performance-based standards for criminal justice agencies.20 These were efforts to
design and/or to discuss performance standards.

In 1983, Norman Carlson, then the director of the Federal Bureau of Prisons, began a
program to develop and implement performance measures in the agency. Since Carlson’s
retirement in 1987, the work has been continued by his successors. As originally conceived,
the Key Indicators/Strategic Support System (KI/SSS) involved defining performance goals
for the agency and for facilities (outcomes which were to be achieved); developing sub-goals
for all major aspects of facilities’ operation, the attainment of which would contribute to the
overall facility and agency goals; defining those sub-goals in operational terms; identifying
measures for each; monitoring facilities’ performance relative to the operational goals (which
involves data collection, analysis, and frequent reporting to managers); and modifying
practice to move the agency closer to its goal. As it developed, the scope of KI/SSS has been
narrowed. Performance monitoring data are routinely collected and made available in an
easily useable format. However, it is distributed only to top facility administrators, rather
than to managers and supervisors generally.
In 1987 the National Center for State Courts, with funding from the Bureau of Justice
Assistance, started the three-year Trial Court Performance Standards Project. It appointed a
commission which decided to focus on standards development in five areas: (1) access to
justice, (2) expeditiousness and timeliness, (3) equality, integrity, and fairness, (4)
independence and accountability, and (5) public trust and confidence. Eventually, the
commission developed twenty-two standards in these areas, devised measures for their
attainment, field tested the standards, and promoted their adoption by key judicial
organizations.
Outcomes-oriented performance monitoring systems have also been developed for policing.
COMPSTAT (Computerized Statistics) is a relatively new police management technique that
relies heavily on the presentation and analysis of reported crime data and other quantitative
police performance indicators. The process originated in New York City in the mid-1990s,
when the New York Police Department (NYPD) began holding regularly scheduled
command staff meetings in which area commanders would discuss crime trends,
interventions underway, and plans for reducing crime in their assigned areas. Computerized
crime maps (e.g., maps showing the location of reported robberies in the past 30 days) and
other crime indicators (e.g., a graph showing the number of robberies by month over the past
3 years) are displayed on large screens so that all attendees can see and comment on the data.
The COMPSTAT meetings also provide a forum for creating short- and long-term objectives

20

See John J. DiIulio et al., Performance Measures for the Criminal Justice System. (Bureau of Justice Statistics, United States
Department of Justice, October 1993).

Abt Associates Inc.

The Contractual Structure

23

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

and for holding area commanders accountable for crime levels, case clearance rates, and other
crime and quality of life indicators.
When New York City began to experience significant reductions in crime in the mid and late
1990s, many high-level city officials pointed to the COMPSTAT process as a key – or
perhaps even the key – factor. The perceived success of COMPSTAT in New York City led
many other cities to implement similar programs in their police departments. For example,
COMPSTAT-like programs have been implemented in Seattle (where it is referred to as
SeattleWatch), Baltimore (Crimestac), Minneapolis (Codefor), Boston (CAM Meetings), and
Los Angeles (Fastrac). The extent to which COMPSTAT actually contributes to crime
reduction and improvements in quality of life is unclear. The Police Foundation is currently
conducting a national evaluation of COMPSTAT processes, with funding from the National
Institute of Justice, that may address some of these performance issues.

Performance-Based Contracting for Criminal Justice Services
Consistent with this broader movement to promote performance standards in government
agencies is the preference for performance-based contracts. Thus, Federal Acquisition
Regulations, supplemented by guidance provided by the Office of Federal Procurement
Policy, encourage the use of performance-based statements of work. OFPP states in its best
practices guide for performance-based service contracting that “performance-based service
contracting emphasizes that all aspects of an acquisition be structured around the purpose of
the work to be performed as opposed to the manner in which the work is to be performed or
through broad, imprecise statements of work which preclude an objective assessment of
contractor performance.”21 Furthermore,
The statement of work...establishes the standard for measuring performance
effectiveness and achievement both during contract performance and upon
contract completion. The work description establishes goals that become the
standards against which contract performance is measured. The SOW is not
complete unless it describes both the work requirements and the criteria for
determining whether the work requirements are met.22
One of the commonly voiced concerns with performance-based contracting, however, is that
the specification of objectives may be too vague to ensure satisfactory performance, and that
more detailed statements of work are needed. As Peter Cole argues in his How to Write a
Statement of Work, “A purely performance-based description is rarely appropriate; usually

21

Office of Federal Procurement Policy, A Guide to Best Practices for Performance-Based Service Contracting, p. 3.

22

Cole, How to Write a Statement of Work, (Vienna, VA: National Contract Management Association, 1995), p. 1–5.

24

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

there are certain constraints that must be described to ensure that the contractor produces a
useful end product.”23 This may be especially true of high-risk activities, such as imprisoning
criminals.
The need for specificity in statements of work has been recognized by others as well. For
example, the Sunset Advisory Commission was tasked by the Texas Legislature to review
and evaluate the performance of the Texas Department of Criminal Justice in contracting for
correction facilities and services. It concluded that “a greater degree of specificity can help
avoid any misunderstanding between the department and the vendors as to what the
department has actually contracted with the vendors to provide.”24
A monograph on developing requests for proposals prepared by the American Correctional
Association under a grant from the Office of Juvenile Justice and Delinquency Prevention
also recommends that “Authors of requests for proposals should understand that their primary
responsibility is to communicate the agency’s needs, requirements, and expectations to an
external audience as effectively and clearly as they possibly can. Authors of RFPs should
never rely on ambiguous or general language when precision is called for. It is equally
inappropriate to provide specific language when general guidance is more appropriate.”25
Performance objectives, performance standards, and performance indicators are susceptible
to precise and detailed specification, although such specification is not easy. In his book on
source selection, Don Edmunds writes that it is a struggle to write “objectives” rather than
“solutions.”26 The struggle may be worth it, however. Focusing on contractors’ performance,
and measuring their compliance with performance standards, while permitting flexibility in
how the service is delivered (within specified constraints) may enhance the contractors’
ability to meet the state’s objectives. (This is, admittedly, a hypothesis rather than an
established fact.)
Operational and Non-Operational Goals in Contracts

Although concerned with public agency goals, John DiIulio and James Q. Wilson’s
distinction between “operational” and “non-operational” goals is relevant to performance
objectives in contracting. An operational goal is “an image of a desired future state of affairs

23

Ibid., pp. 2–10, 2–11.

24

Sunset Advisory Commission, Recommendations to the Governor of Texas and Members of the Seventy Second Legislature, Final
Report (March 1991), Chapter 4: “Information Report on Contracts for Correction Facilities and Services,” p.20.

25

Office of Juvenile Justice and Delinquency Prevention, Monograph: Developing a Request for Proposals and a Proposal Review
Process, (Office of Justice Programs, U.S. Department of Justice, prepared under Grant No. 90-JS-CX-K003 by the American
Correctional Association), p. 6.

26

Don L. Edmunds, Source Selection: A Seller’s Perspective, (Holbrook & Kellogg, Inc., 1993), p. 201–202.

Abt Associates Inc.

The Contractual Structure

25

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

that can be compared unambiguously to an actual or existing state of affairs,” whereas a nonoperational goal cannot be so compared.27 An example of a non-operational goal is
“improving the quality of prisoner rehabilitation programs,” while an operational goal is
“increasing the average verbal and math scores on a particular standardized test of prisoners
by 20 percent by the second year of the contract.” DiIulio observes that most private sector
managers have operational goals—expanding “bottom line” profits, for example—while
public mangers often work in the context of multiple and contradictory non-operational
goals.28 This results, in part, from the fact that the political consensus needed to support or
alter broad areas of public policy is defined and formulated in terms of non-operational and
largely symbolic goals.
In these terms, the goals established in most private imprisonment contracts are essentially
non-operational. Furthermore, the goal of achieving ACA accreditation of detention facilities
is not an outcomes-based performance goal. Rather, as Parent has observed, ACA standards
primarily prescribe procedures.29 The great majority of ACA standards are written in this
form: “The facility shall have written policies and procedures on . . . .” The standards
emphasize the important benefits of procedural regularity and effective administrative control
that flow from written procedures, careful documentation of practices and events, etc. But,
for the most part, the standards prescribe neither the goals that ought to be achieved nor the
indicators that would let officials know if they are making progress toward those goals over
time.
The accreditation movement has been driven by an implicit assumption that facilities that
conform to ACA standards will indeed be “better”—less violent, more manageable, more
efficient, more humane, etc.—than facilities that do not conform. Two studies, both
involving juvenile facilities, question this presumed link. David Roush studied juveniles’
perceptions of well-being in ten Midwestern detention centers, five of which were accredited
by ACA and five of which were not. He found that residents’ perceptions of well-being were
higher in the non-accredited facilities.30 Dale Parent and colleagues at Abt Associates
examined rates of suicidal behavior, escapes and attempted escapes, and injuries to staff and
juveniles among more than 900 juvenile facilities, comparing those which conformed to key
ACA standards with those that did not conform. They found no differences in the rates of

27

John J. DiIulio, Jr., “Measuring Performance When There is No Bottom Line,” in DiIulio et al., Performance Measures for the
Criminal Justice System, p. 145.

28

Ibid., p. 146.

29

Dale Parent, Valerie Leiter, Stephen Kennedy, Lisa Livens, Daniel Wentworth, and Sarah Wilcox. Conditions of Confinement: A
Study to Evaluate Conditions in Juvenile Detention and Corrections Facilities. (Washington, D.C.: Office of Juvenile Justice and
Delinquency Prevention, United States Department of Justice, May 1993).

30

David W. Roush, “Exploring institutional quality of life: A study of the relationship between conditions of confinement and residents
perceptions of well-being.” Dissertation Abstracts International, 59, 1011–A (Doctoral dissertation, Michigan State University,
September 1990).

26

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

negative incidents among conforming and non-conforming facilities.31 Hence, conforming to
ACA standards did not appear to be related to superior conditions of confinement, as
measured by rates of these critical incidents.
The authors of both studies explain this apparent lack of close conjunction between selected
performance indicators and conformity with ACA standards in the nature of the standards
themselves. The authors of both studies explain this apparent lack of conjunction between
selected performance indicators and conformity with ACA standards in the nature of the
standards themselves. Using training as an example, Roush observes that ACA standards
require only that certain levels of training occur. Whether or not the skills acquired in that
training are used by staff, or whether their use is monitored or evaluated, is not assessed.32
Roush also notes that formal compliance with ACA standards is a dichotomous
measure—that is, it indicates simply whether the facility does or does not comply.
Assessment of written policy is relatively easy: the policy either exists or does not exist. It
either does or does not contain the elements required by the ACA standard. Roush argues
that the process of implementing policy is more usefully conceived of as a continuous
process, and that a more useful indicator measures degrees of intensity or frequency of
compliance or implementation.33 Hence, while two facilities may formally comply with an
ACA standard, practice in the two may vary greatly. Roush observes further that ACA
accreditation depends more on review of records than on observation of and feedback from
staff and residents. This places high value on the adoption of written policies and low value
on the extent to which they are actually implemented.34
Parent also found that most of the individual ACA standards deal with procedures, not
outcomes. Such standards generally are worded as follows: “The facility should have a
written policy governing . . . . .” Two facilities could conform equally to an ACA standard by
having a written policy on a particular issue, yet they could have diametrically opposite
practices and outcomes on that issue. For example, both could have a written policy on the
use of isolation, each of which covers topics like when isolation may be used, how long
inmates can be isolated, and how reviews should be conducted to extend periods of isolation.
One facility, however, might use isolation rarely, if at all, while the other could use it
prodigiously.

31

Parent et al., Conditions of Confinement.

32

Roush, “Exploring Institutional Quality of Life”, p. 80.

33

Ibid., p. 83.

34

James Q. Wilson, “The Problem of Defining Agency Success”, in DiIulio et al., Performance Measures for the Criminal Justice
System, p.161.

Abt Associates Inc.

The Contractual Structure

27

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

In Parent’s Conditions of Confinement Study, members of the advisory board recognized the
importance of procedural standards. They argued that regularity in procedures is an essential
foundation of good facility management. Hence, by promoting procedural regularity in its
standards, ACA has promoted greatly improved facility management. However, the advisors
also emphasized that critical need to build upon procedural regularity by developing
performance-based standards that define the outcomes that facilities should achieve, and
which provide benchmarks against which facility progress can be measured. We think this
observation is relevant to all correctional and detention operations, and is not limited to
juvenile corrections.
The commitment to performance standards and monitoring differs importantly from a
commitment to being accredited. In accreditation an agency typically engages in a short
period of intense activity as it prepares for an accreditation audit—it drafts new policies to
conform to standard requirements, it organizes its records to show auditors that it conforms to
standards, etc. This effort leads to a state of being accredited. Typically, however, the level
of activity and attention to standards drops sharply once the audit has been completed and
accreditation is awarded, and later picks up again when the facility prepares for reaccreditation.
In contrast, performance monitoring involves an agency commitment to continuous
monitoring of practice to measure improvement over time. Once performance standards are
established, administrators conduct an initial assessment to determine where their practices
fall in relation to performance objectives. They then decide what modifications they will
make in practice, structure, organization, or training to achieve improved results.
Administrators implement those changes and allow them to operate for an appropriate
interval. They then reassess the facility’s performance to determine if targeted practices have
improved. If so, they may select other practices for improvement during the next cycle. The
process of assessment, modification, implementation, and reassessment goes on
continuously. Facilities may never reach some goals—but they may continue to make
incremental progress toward them. In other words, performance standards create a state of
becoming rather than a state of being.

28

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

3.

Monitoring Contractors’ Performance

Although statements of work are crafted with an eye to averting problems before they occur,
they cannot be expected to produce problem-free operations. After the contract is signed, the
keys to minimizing the risks associated with contracting are effective procedures for
monitoring performance and for resolving problems once identified. This chapter addresses
some selected issues associated with monitoring the extent to which the objectives of the
government’s privatization program are being achieved.

How Closely are Facilities Monitored?
The 1998 Abt Associates survey of state and federal correctional agencies found that
governments’ practices of monitoring privately operated facilities vary widely. Nearly all
contracts active on the last day of 1997 reportedly received some oversight from the
contracting agency. This ranged from minimal attention by a centrally located contract
administrator to a combination of a contract administrator and one or more on-site monitors.
Of the 28 state and federal government agencies that reported having active contracts, twenty
reported using monitors in addition to contract administrators. Several states required a
strong background and knowledge of a government agency or corrections as prerequisites for
monitoring correctional contracts. Others required experience as an internal auditor or
investigator. Most governments required at least some level of formal training (up to 40
hours annually) in areas such as auditing, establishing performance standards and measures,
ethics, and reporting disputes and conflicts. Contract monitors surveyed by Abt Associates
often reported receiving more training than was required. One reportedly received more than
120 hours of training as a contract monitor. However, only 12 states reported providing
training specifically geared towards the monitoring of privately operated correctional
facilities.
The amount of time and labor spent on monitoring privately operated facilities also varied
widely, from full-time monitors who spend a substantial proportion of their time on-site to
almost no monitoring at all. Of the 89 contracts for which we have information, 46 of them
(or 52 percent) reported having monitors on-site on a daily basis. These monitors generally
worked full-time on these assignments, although some had responsibilities for more than one
contract. A smaller number of contracts (16) had part-time monitors, who averaged about
one day a week, visiting the facilities on a monthly basis. Ten contracts had monitors who
devoted about the same level of attention to the contract, but visited only quarterly.
There were significant difference in how governments monitored in-state as opposed to outof-state facilities. As discussed in Chapter 1, the national market for beds in privately
operated facilities works differently than the market between purchasing states and in-state
providers. Governments tend to use facilities in the national market with shorter-term

Abt Associates Inc.

Monitoring Contractors’ Performance

29

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

agreements to purchase needed capacity. Most of these facilities were out-of-state, although
not always. Respondents to the 1998 Abt Associates survey of correctional administrators
reported different practices in monitoring these mostly out-of-state facilities, compared with
their practices for monitoring contracts with in-state and mostly exclusive providers (Table
3.1).
Table 3.1
Governments’ Monitoring of In-State Contractors and Out-of-State
Facilities
Agreements with Facilities

Average number of
hours/month
devoted to monitoring
<20 hours
20-80
>80

Frequency of on-site inspections
Daily/Full-time monitoring
Weekly
Monthly
Quarterly
Other
Are contract monitors trained for
the task?
Yes
No

In-State

Out-Of-State

40%
8
52
100%

90%
10
0
100%

48%
6
4
4
38
100%

10%
0
38
19
33
100%

63%
37
100%

14%
86
100%

Source: Abt Associates Inc. 1998 Survey of State and Federal Correctional Administrators

Ninety-percent of the contracts/intergovernmental agreements with these “nationallyoriented” and mostly out-of-state facilities were monitored less than 20 hours per month. The
other ten percent were monitored at the rate of between 20 and 80 hours per month. This
contrasts sharply with the monitoring of in-state facilities having (nearly always) exclusive
contracts with the government in the state where they are located. Fifty-two percent of those
contracts or agreements were monitored more than 80 hours per month. Indeed, almost half
(48 percent) had full-time contract monitors, compared to fewer than 10 percent of the
“national” facilities. In another 6 percent of the contracts/agreements, monitors of the

30

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

in-state facilities conducted on-site inspections or visits weekly. The pattern at the nationals
was significantly different. The majority of these contractors were inspected on-site monthly
(38 percent) or quarterly (19 percent). Administrators reported “other” schedules for site
visits/inspections, which may indicate irregularly scheduled visits of unknown frequency.
States that contracted with facilities in-state were also much more likely to employ staff
especially trained for the task of monitoring. Sixty-three percent of the contracts with in-state
facilities employed trained monitors, compared with only 14 percent of the
contracts/agreements with out-of-state facilities.
Not surprisingly, the most publicly visible troubles in privately operated prisons have
occurred most often in these arrangements whereby governments contract with out-of-state
facilities to hold prisoners. State contract administrators and monitors also rated their
performance below that observed at in-state facilities with which states had (mostly)
exclusive relationships. In 38 percent of all contracts or agreements with out-of-state
facilities, the monitors or administrators rated the quality of the service as below that of
comparable facilities in their own department of correction, compared with 7 percent of the
contracts with in-state facilities.1 This suggests that monitoring is a crucial element in
achieving a privatization program’s objectives.

The Purpose of Monitoring and the Monitors’ Role
Although surveyed state and federal governments were not asked to specify the objectives of
their monitoring activities, it is likely that the most common approach is a narrowly
conceived one: to ensure compliance with contract requirements, state laws, rules, policies
and procedures. This orientation appears generally consistent with the primary objective of
contracting for privately operated facilities in the first place: simply to obtain beds in
facilities that meet minimum constitutional and professional standards. Where financial
savings are sought as an objective, measuring the state’s success can be done in government
offices by accountants without setting foot in the facility. Contract monitors therefore have
no role in monitoring progress towards cost savings. States that seek other than these
minimalist objectives in their contracts–such as “prisoner rehabilitation”–are few in number.
There are other purposes that can be served beyond monitoring strict compliance. One is to
ensure the safety and security of the public, staff, and inmates, which may require attention to
matters other than specific contractual provisions. Another purpose is to facilitate problem
solving. Accordingly, the American Correctional Association’s monograph on monitoring
and evaluating contracts states, “The “purpose of contract monitoring, besides ensuring
compliance, is to detect problems and work together to solve them. The contracting agency

1

Abt Associates 1998 survey of state and federal correctional agencies.

Abt Associates Inc.

Monitoring Contractors’ Performance

31

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

and the provider must feel free to work openly and honestly to provide the best possible
service.”2 Still another purpose beyond ensuring compliance is to maintain the credibility of
the policy and practice of contracting in the eyes of various constituencies such as the
legislature, the public at large, or inmates’ families.3
The federal government requires that contract monitoring and administration go beyond
assessments of strict compliance. According to the Federal Acquisitions Requirements,
contracting officers are responsible for ensuring performance of all necessary actions for
effective contracting, ensuring compliance with the terms of the contract, and safeguarding
the interests of the government in its contractual relationships. To perform these
responsibilities, contracting officers are allowed wide latitude to exercise business judgment.4
Among other duties, the contracting officer must ensure that contractors receive impartial,
fair, and equitable treatment; and must obtain the advice of specialists in audit, law,
engineering, transportation, and other fields, when appropriate. To exercise these
responsibilities effectively, contract officers sometimes share monitoring responsibilities with
contract monitors, who act as their eyes and ears.
The contract monitoring role, therefore, has a number of potentially different orientations,
which can produce tensions. At one end is the legalistic approach, focused on strict contract
compliance, which may involve on-site “white gloves” inspections. At the other is the
problem-solving partnership role.
In its monograph on monitoring and evaluating contracts prepared for the Office of Juvenile
Justice and Delinquency Prevention, the ACA points out that “Ideally, contract monitoring is
not a process of finding fault or blame and threatening the provider with penalties. This
approach is counter-productive because it focuses only on the negative, creates anxiety and
distrust, and causes the provider to be secretive or to withhold critical information for fear of
losing the contract or appearing to be deficient.”5 Lester Edelman and colleagues observe
that “the adversarial management relationship jeopardizes the ability of either side to realize
its expectations. The result is increased costs for the taxpayer and declining profit margins
for the contractor.”6

2

Office of Juvenile Justice and Delinquency Prevention, Monograph: Monitoring and Evaluating Contracts (Office of Justice
Programs, U.S. Department of Justice, prepared under Grant No. 90-JS-CX-K003 by the American Correctional Association, March
31, 1992), p. 1.

3

Richard Crane, “Monitoring Correctional Services Provided by Private Firms” a report for the Association of State Correctional
Administrators, undated, p. 2.

4

FAR 1.602–2.

5

Office of Juvenile Justice and Delinquency Prevention, Monograph: Monitoring and Evaluating Contracts, p. 4.

6

Lester Edelman (Chief Counsel, U.S. Army Corps of Engineers) et al., Partnering: Alternative Dispute Resolution Series, Pamphlet
4, December 1991, IWR Pamphlet 91-ADR-P-4, p. 1.

32

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

The Army Corps of Engineers has developed a formal procedure, called “partnering,” to
minimize disputes under its architect-engineering contracts.
Partnering is the creation of an owner-contractor relationship that promotes
achievement of mutually beneficial goals. It involves an agreement in
principle to share the risks involved in completing the project, and to establish
and promote a nurturing partnership environment . . . . Partnering seeks to
create a new cooperative attitude in completing government contracts. To
create this attitude, each party must seek to understand the goals, objectives,
and needs of the other—their win situation—and seek ways that these
objectives can overlap.7
A form of partnering has also been used in environmental restoration contracting. This
program, called Total Environmental Partnering (TEP), is used to help coordinate activities
of the Corps of Engineers, the architect/engineer firm which is to design a restoration project,
the remedial action contractor that is to perform the restoration, the state and federal
regulators and the landowners. Jeffrey Hills states that “The innovative use of partnering can
have substantial effects. The TEP concept resulted in remarkable success with previously
unheard of levels of cooperation. TEP’s use of partnering synergized the efforts of all
interested parties while improving the performance and lowering the cost of standard
contracting instruments . . . .”8
This broad conception of the monitors’potential roles and responsibilities suggests a number
of associated issues.
1.
2.
3.
4.

5.

Have the state’s objectives regarding monitoring been defined? Have these objectives
been translated for monitors into specific guidelines for action?
Have the criteria (e.g., cost, quality of service, responsiveness to governments’
requests) been identified?
Have procedures for measuring performance according to these established criteria
been devised?
Are procedures for rewarding and sanctioning contractors developed and is there
consensus among the government’s representatives in how these rewards and
sanctions will be applied?
What procedures exist for identifying and training contract monitors, and for
reviewing their performance?

7

Ibid., p. 2.

8

Maj. Jeffrey W. Hills, “Total Environmental Partnering,” Contract Management, September 1996, p.9.

Abt Associates Inc.

Monitoring Contractors’ Performance

33

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

Monitoring Outcomes
As discussed in Chapter Two, performance-based contracts are optimally focused on
achieving designated outcomes other than complying with procedural rules, regulations, and
standards. Monitoring such contracts raises a number of issues. What outcomes are to be
established as indicators of performance, and how susceptible are they of measurement?
Can outcomes be attributed to the actions of the agencies being monitored–in this case, the
privately operated correctional firm? Can effective incentives be tied to accomplishment of
the objectives to encourage good performance? Can these outcomes be monitored by
contract monitors, or must other resources be called upon to perform these assessments?
Designating Outcomes to be Monitored

Defining institutional performance as outcomes often imposes a substantial challenge to
monitors. Contract monitors are best able to assess information that they can obtain directly
from review and observation of practices and procedures. If the objective of the privatization
initiative (or of the public correctional agency in toto) is to accomplish goals such as
preventing or reducing the subsequent criminality of released prisoners, contract monitors
will not be able to assess a prison’s success in accomplishing this mission. Determining
whether released offenders commit more crimes is a difficult task. The behaviors that one is
trying to affect are those of individuals rather than institutions; they are hidden and usually
difficult to detect; and they are subject to influences other than the experience of
imprisonment or treatment. Moreover, success in these types of services must realistically be
measured not as perfect conformity to the law but rather the improved likelihoods of
avoiding recidivism or relapse. This poses not only a measurement challenge but also a
methodological one of estimating the extent of any observed changed that can be attributed to
the experience of imprisonment. These kinds of assessments require costly data collection
and complex statistical analyses, which are outside of the domain within which contract
monitors operate. Not surprisingly, few correctional privatization programs seek such
objectives. (The exception is Florida, as discussed below. Assessment of the state’s
achievement of this rehabilitation objective is undertaken periodically by social scientists
who conduct statistical analyses of recidivism, comparing prisoners released from privately
operated prisons and from prisons operated by the Florida Department of Corrections.9 )
Another approach is to designate outcomes that are within the ambit of what correctional
facilities do and that can be measured more directly. An example of this is the Bureau of
Prison’s contract for the Taft Correctional Institution. The solicitation and subsequent

9

34

See Lonn Lanza-Kaduce and Karen Parker, “A Comparative Recidivism Analysis of Releasees from Private and Public Prisons in
Florida,” University of Florida, 1997. For a critique of this study, its approach and findings, see Gerald Gaes, Scott Camp, and
William Saylor, “The Performance of Privately Operated Prisons: A Review of Research,” appendix 2 to Douglas McDonald,
Elizabeth Fournier, Malcolm Russell-Einhorn and Stephen Crawford, Private Prisons in the United States: An Assessment of Current
Practice (Abt Associates Inc., 1998).

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

contract was clearly designed to advance the Bureau’s mission and the strategic objectives.
These objectives derive from the Bureau’s system for managing all federal prisons, and not
just privately-operated ones. This performance management system, first developed in the
mid-1980s, involves:
•

defining performance goals for the agency and for facilities in terms of outcomes to be
achieved;

•

developing sub-goals for all major aspects of facilities’ operation, the attainment of which
would contribute to the overall facility’s and agency’s goals;

•

defining those sub-goals in operational terms that are susceptible to being measured; and

•

specifying the functions, or activities, that have to be carried out to accomplish goals (the
Bureau calls these “vital functions”).

The Bureau’s mission is to protect public security, to protect inmates and staff, to provide
inmates with appropriate and humane levels of care, and to give inmates opportunities to
become better citizens. (Note that these are more immediate objectives than “producing lawabiding citizens” who do not recidivate.) How these are to be accomplished is set forth in the
Bureau’s “vision statement”:
The Bureau provides for public safety by assuring that no escapes and no
disturbances occur in its facilities. The Bureau ensures the physical safety of
all inmates through a controlled environment which meets each inmate’s need
for security through the elimination of violence, predatory behavior, gang
activity, drug use, and inmate weapons. Through the provision of health care,
mental, spiritual, educational, vocational and work programs, inmates are well
prepared for a productive and crime free return to society.10
The contract for the Taft Correctional Institution sets the same performance goals as the
Bureau sets for all other facilities. Moreover, it adopts the same “vital functions” identified
as necessary to accomplish the Bureau’s strategic objectives at all federal prisons. These
vital functions are not designed to provide an instruction manual in how to accomplish the
objectives. For example, the following describe requirements pertaining to institutional
security, control, and inmate accountability.
Vital Function #1: Provide a safe and secure environment for staff and
inmates through effective communication of operational concerns. This

10

Bureau of Prisons, Vision Statement, undated.

Abt Associates Inc.

Monitoring Contractors’ Performance

35

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

includes verbal and written instructions, post orders, institution supplements,
information dissemination, training, and crisis prevention.
Vital Function #2: Gather intelligence information related to security
concerns for dissemination to appropriate contractor and Federal Bureau of
Prisons staff.
Vital Function #3: Provide an adequate security inspection system to meet the
needs of the institution.
Vital Function #4: Maintain an adequate level of emergency readiness to
respond to institution emergencies.
Vital Function #5: Maintain a level of occurrence for the following listed
incidents at, or below, the average rate of occurrence at other BOP facilities of
the same security level.
Assaults without weapons on other inmates
Assaults with weapons on staff
Assaults with weapons on other inmates
Homicides
Suicides
Escapes
Whereas the Bureau relies on self-monitoring to keep performance aligned with the agency’s
mission, it devised a different set of instruments to manage the performance of a contractor.
First, it designed an innovative performance-based contract. Second, it created financial
incentives to encourage the contractor to perform well. Third, it gave Wackenhut
responsibility for managing its own quality control system at the Taft Correctional Institution.
Performance-Based Contract Provisions
Rather than specifying in detail precisely how the contractor will deliver all its services, the
Bureau established objectives (vital functions) that the managers at the Taft facility are
contractually obliged to accomplish. This accords with Federal Acquisition Regulations and
the Office of Federal Procurement Policy (OFPP), which encourage the use of performancebased contracts.11 This emphasis on performance rather than process is reinforced by the
statement of work in the contract. Whereas some contracts for imprisonment services,
written by state or local governments, go to great lengths to detail exactly how the private

11

36

Office of Federal Procurement Policy, A Guide to Best Practices for Performance-Based Service Contracting (Washington, D.C.:
Executive Office of the President’s Publications Office, 1996), p. 3 (Italics added).

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

operator is to structure its operations, the Taft contract leaves such decisions largely to the
contractor. What is specified in the statement of work is quite short, essentially listing
activities that are deemed absolutely essential to the Bureau of Prisons, and which clarify
where the contractor’s responsibilities begin and end relative to the Bureau’s. (The statement
of work also incorporates by reference, however, a more extensive list of standards for
facility operations established by the American Correctional Association.12) For example, the
section in the statement of work that specifies required institutional security procedures is
quite short—slightly longer than three pages. Insofar as maintaining security and control are
the most important objectives for a prison, the brevity of the prescribed procedures is quite
striking.
Financial Incentives to Perform Well
The contract establishes financial incentives to encourage Taft’s managers to achieve
performance goals that go beyond simple compliance in such areas as quality, timeliness,
technical ingenuity and cost effective management. Twice a year, Bureau officials rate Taft’s
performance to determine if and how much additional money could be awarded to the
contractor. These bonuses can be substantial—equaling as much as five percent of the
amounts paid to the contractor for its services. The size of the award is determined by the
government’s judgmental evaluation of the contractor’s performance according to criteria
stated in the contract. These criteria are:
Quality of Work
•

Results of quality assurance inspections and observations of government personnel

Contractor Responsiveness
• Timeliness, effectiveness and appropriateness of response to both routine and unusual
institution events
• Timely response to BOP concerns
• Reaction to changing service requirements
Management of Quality Control Program
• Effectiveness of quality control program
• Self-initiated service improvements
Determination of the fee is based on the findings reported periodically by the contract
monitors. These reports evaluate the contractor’s performance in each of fifteen different
“departments,” or functional areas, such as institutional security, safety, mail, educational
services, laundry, etc. For each department, monitors are required to evaluate aspects of
12

Taft Correctional Institution Contract, Statement of Work, p. 7.

Abt Associates Inc.

Monitoring Contractors’ Performance

37

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

performance that pertain to the quality of the work (not just whether it complied with the
terms of the contract), the contractor’s “responsiveness,” and the sufficiency of the
contractor’s own quality control management. These areas are assigned grades ranging from
“unsatisfactory” to “outstanding.” Because the Bureau intends to award bonuses only for
performance that goes beyond simple compliance with the contract, only that performance
rated “good” or better contributes to earning an award. How the Bureau defines and
interprets these ratings, and how much weight it assigns to each, is shown in Table 3.2. If,
for example, the contractor’s performance is rated “outstanding” in all aspects, it would be
eligible for 80 to 100 percent of the entire award fee. In contrast, if it received ratings of
“good” in all functional areas, the bonus could range between 1 and 39 percent of the award
pool.

Table 3.2

Performance Evaluation Rating Used to Determine Award Fees
Range of
Points

Rating

Performance Description

Outstanding

Superlative level of performance; achievement of distinguished results
and effectiveness. No deficiencies
80-100%

Excellent

Of exceptional merit; exemplary performance in a timely, effective and
professional manner. Very minor deficiencies. No effect on overall 40-79%
performance.

Good

Very efficient performance; fully responsive to contract requirements;
more than adequate results; reportable deficiencies but with little 1-39 %
identifiable effect on overall performance.

Fair

Effective performance; responsive to contract requirements; adequate
results. Reportable deficiencies with identifiable, but not substantial 0%
effects on overall performance.

Marginal

Meets minimum acceptable standards; useful levels of performance
but suggested remedial action; reportable deficiencies which adversely 0%
affect overall performance.

Unsatisfactory

Below minimum acceptable standards; poor performance; inadequate
results; requires prompt remedial action; significant deficiencies.
0%

Source: Taft Correctional Institution Contract, Section J, Attachment F, pp. 6-7.

Monitors do not make the decisions about the size of the awards. Rather, they provide the
ratings that are considered by the Performance Evaluation Board and the Fee Determination
Officer.
In addition, contract officers rely upon contract monitors’ reporting to assess whether the
contractor’s performance is sufficient to earn the full monthly fee or if the fee is to be

38

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

reduced. The contract establishes broad guidelines to assist the contract officer in
determining how much money to deduct from these monthly fees. It defines performance in
nine different areas as well as the relative value the Bureau places on performance in these
areas (see Table 3.3). Not surprisingly, institutional security and control are seen as being the
most important, and are assigned 20 percent of the total value of what the contractor is
required to do. For example, if Wackenhut failed utterly to comply with the various
contractual requirements pertaining to institutional security, control, and inmate
accountability, it would be liable to lose up to ten percent of its monthly fee for services not
rendered. Complete failure to perform rarely occurs, however. More often, service is more
deficient by degrees. This requires Bureau officers to make a subjective judgment to
determine the amount of the deduction, guided by the upper limits shown in Table 3.3.

Table 3.3
How the Bureau of Prisons’ Contract for the Taft Correctional Institution
Defines the Relative Value of the Contractor’s Functional Responsibilities
Contract Requirements

Value Attributed

Security/control/inmate accountability

20%

Inmate admission, classification, programs and transfer

20%

Facility maintenance and repair

15%

Personnel

10%

Health services

10%

Food services

10%

Quality control

10%

Inmate services

5%

Source: Taft Correctional Institution Contract, Section J, Attachment C, pp. 2-6.

Exactly How is Such Performance Measured and Monitored?

The design of an outcomes-oriented performance assessment regimen poses a challenge to
monitors. Precisely how do they translate their task into operational terms? When asked
how evaluating performance differs from assessing compliance, the former contract
administrator answered, “Judging compliance is like deciding whether somebody passes or
fails a course. Judging performance is like giving them grades.”14 But are the measures used
to assess performance essentially the same as if compliance with standards is being assessed,
except that shades of grey are distinguished? How the Taft facilities performance is being

14

Ray Marshall, the Bureau’s first Contract Administrator at Taft, in a telephone conversation with Douglas McDonald, in 1999.

Abt Associates Inc.

Monitoring Contractors’ Performance

39

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

evaluated is the subject of a research study, which is examining how monitors have
implemented the performance-as-outcomes design of the contract.15 Because this is one of
the few performance-oriented contracts in corrections (the other being recent contracts for
state jails in Texas), how monitors conceive of performance and measure it will provide some
indication of how feasible such contracts are for imprisonment services, and if they have
value beyond that afforded by more conventional procedures-oriented contracts.

Assessing Achievement of Strategic Objectives
Beyond monitoring contractor performance is the question of whether the privatization
initiative taken as a whole–including both the government’s decisions and the contractor’s
performance–are accomplishing the state’s strategic objectives for contracting. Interestingly,
most governments have not sought such assessments. Those that have done so have
generally focused on comparing costs of contracting with the cost of the government
operating the facility in question. Very few have compared the performance of the privately
operated facilities. Most assessments of performance rely on the contract officers’ judgments
regarding contract compliance.
There are a number of methodological issues that must be resolved when assessing whether
or not objectives are being achieved. These apply to assessments of costs and savings and to
assessments of performance.
Comparing Costs of Government Operation to Contracting

Estimating costs and/or savings associated with contracting requires estimating how much
the state spends to contract and comparing that to what the state would have spent to provide
the service directly. Governments have not generally followed consistent accounting
procedures in making such comparisons, although there exist templates that can be followed.
One is that specified in the U.S. Government’s Office of Management and Budget Circular
A-76. Although developed to guide estimation before a decision is made to contract, the
general methodology can be adapted to retrospective analyses of costs and savings resulting
from contracting.
Whereas the A-76 guidelines provide assumptions for estimating various costs that will be
incurred for contracting in the future, actual costs can be obtained from state expenditure
accounts. These include payments to the contractor for services, expenditures for contract
administration and monitoring, and other costs associated with the facility’s operation that are
not covered by the contractor. These might include, for example, payments by the state to
cover certain inmate health care costs (such as those exceeding a contractually-established

15

40

This is being conducted by Abt Associates Inc., with Douglas McDonald as principal investigator.

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

cap). Costs need not be included in this analysis if they would have been incurred regardless
of whether a decision was made to contract. For example, when the government contracts
only for the management of a state-owned facility, the costs associated with facility
maintenance and debt service can be ignored, as the state will incur these costs under either
option.
Whether state government overhead expenditures should or should not be allocated to
contractor-operated facilities is controversial. One school of thought argues that
governmental indirect/overhead expenditures should not be assigned to the contractor facility
because the cost of administering the contract is captured in the direct cost of contract
administration and monitoring. Another school argues that governmental overhead
expenditures should be assigned to all operations, whether provided under contract or directly
by government employees. The OMB Circular A-76 adopts the first argument (i.e., that such
overhead costs should not be assigned to contractor-operated facilities, although they should
properly be assigned to government operated ones). A decision to do otherwise results in
computing a higher total cost of contracting. Because governmental overhead costs can run
as high as 10–15 percent of directly attributable operating costs, deciding one way or another
about this accounting rule will have a large impact on any comparison of costs and will
strongly favor government provision.
An assumption also has to be made about whether to include an estimated cost to the
government of self-insurance against lawsuits involving the contractor-operated facilities.
Under existing law, governments cannot shield themselves completely from liability claims
because private firms are considered to be acting on behalf of the state. Following this logic,
it would make sense to estimate an annualized expenditure for self-insurance–a kind of
insurance premium equivalent. (Governments typically self-insure, paying out claims as they
are settled, rather than taking out policies with private insurers and paying them annual
premiums.) Whether this makes sense depends upon the details of the contract, however.
Governments may design their contracts to require contractors to bear full financial
responsibility for all liabilities and damages up to a specified amount (one such contract
specifies $75 million). This gives the state much protection. In such a case, the annualized
cost of the state’s self-insurance for contractor operations is effectively zero, and should not
be included as a cost of contracting.
Estimating what the government would have spent to imprison persons in a facility operated
by the department of correction is more difficult, if only because this is the course not taken.
The most sensible approach is for a disinterested party to estimate, prior to making a decision
about contracting, what it would cost the government to have the department of correction
provide the service. Some states (e.g., Texas, as discussed in Chapter 4) charge an agency
other than the department of correction to make this assessment. Federal OMB A-76
guidelines do not go this far but instead require a high official not involved in developing the
cost estimate to certify that the estimate is realistic. Moreover, the circular provides specific

Abt Associates Inc.

Monitoring Contractors’ Performance

41

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

accounting assumptions to be followed when developing these estimates so as to create, to
the extent possible, a level playing field for both the public and private sector.
In practice, most states do not develop disinterested estimates of government operation before
making a decision to contract. If the contracting decision were left entirely to public
correctional administrators, and if decisions about privatization were not so politically
contentious, rational administrators would undertake such cost analyses in the same way that
private sector managers do—to identify the most advantageous method of obtaining a good
or service. But most prison privatization decisions are initiated not by correctional
administrators but by their political masters–the governors or legislatures. And these
decisions are often surrounded by much politicking by private firms and public employee
union representatives who feel their jobs may be at stake.
Estimating the costs of a decision not taken is difficult when it is long after a contract has
been let. The fundamental problem is deciding what to base this estimate upon. If the prison
under contract had been operated previously by the public correctional authority, a
before/after comparison would be informative. This rarely happens in this country, however.
Nearly all contracts have been awarded to provide new space.
A commonly considered solution to this problem of estimation is to examine the operating
costs of other government-operated prisons and to compare those costs, with some
adjustments, to the calculated total cost of contracting. This may not be appropriate,
however, because labor costs drive correctional facility budgets, and labor costs vary
considerably from one facility to another, depending upon their architecture. What may be
compared in this instance is not public versus private management but the results of
architectural design decisions. Moreover, many of the facilities managed by contractors serve
specialized functions (e.g., for housing women, or for providing drug abuse treatment
programming) and no comparable government-operated facility exists.
Because estimation of what the government would have spent if it had not contracted can be
so methodologically difficult, a persuasive analysis may be nearly impossible to make. Given
this, a state that will seek ultimately to evaluate the costs of contracting as opposed to
government provision should develop best estimates of what direct provision would cost in
advance of moving forward with efforts to solicit contractors’ bids. An accurate assessment
will include not only costs to the department of correction (which would be carried on its
budget) but all other costs to the state (such as the cost of self-insurance against liability
claims). The federal OMB Circular A-76 specifies a number of different cost categories that
should be included in such estimates.
If one of the state’s objectives in contracting is to save money (or to obtain more
cost/effective service at the same level of expenditure), developing such estimates in advance
of soliciting bids serves another important function. It not only provides the state with a basis

42

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

for future cost analyses but also grounds for determining if any of the bids make good
financial sense for the state. In short, a coherent strategy for contracting when the cost
savings or cost efficiency is an objective is to build in procedures for early estimation of what
the government would spend to provide the service directly and to require subsequent cost
comparisons using defensible accounting procedures.
Comparing Performance

If the performance objective to be achieved by contracting is simply to house prisoners in
facilities that meet specified standards, rules and regulations, then a comparison of the
facility’s operation with those standards, rules and regulations will suffice. If, on the other
hand, the state hopes to obtain some level of performance relative to that found in other
government-operated facilities, the task of assessing performance will be more difficult.
A comparative analysis can be simplified by making the rulebook to be followed by the
contractor the same as that used by government-operated facilities. (Requiring contractors to
replicate what the government already does may not be desirable for other reasons, however.)
The extent to which the privately-operated facility complies with the rulebook could thereby
be compared to compliance in government-operated facilities. This assumes that one could
institute an unbiased and reliable method for assessing compliance uniformly in both the
privately and government operated facilities—not a simple task.
Assessment of performance outcomes other than procedural compliance poses other
methodological challenges. Apart from defining the outcomes to be sought, the real practical
difficulties involve acquiring data and analyzing them. If the desired objectives are to
achieve in the private facilities comparable or superior levels of public security, or safety with
respect to staff and prisoners, or certain levels of inmate educational accomplishment, or
prisoner health, information about these dimensions need to be collected in both the public
and privately operated facilities. Government agencies may not collect on a routine basis the
information needed for such comparisons, and the contractor will probably not implement
costly data collection and reporting unless contractually obliged. A forward-looking
privatization strategy will consider whether it is possible to assess accomplishment of
performance-based objectives and what kinds of information must be developed by
government agencies and the contractor to support assessment.
The data demands for such assessment go beyond measuring the objectives being sought.
Prisons will generally differ in the types of prisoners sent to them, and these differences
might affect one or another facility’s ability to achieve the performance objectives,
independently of how it is managed. For example, a private facility that has received a
disproportionate number of the state correctional agency’s troublemakers will be expected to
perform more poorly on measures of staff and prisoner safety and perhaps public security (if
they are more likely to escape, for example). Or, conversely, if the state is unwilling to risk

Abt Associates Inc.

Monitoring Contractors’ Performance

43

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

sending its more difficult prisoners to the privately operated facility, that facility’s
performance may look superior if judged simply on the basis of aggregate numbers of
assaults on staff or inmates, or other similar measures. Even if one suspects that no such
dramatic “dumping” or screening occurs, one has to assume that systematic differences
among prisoner populations will occur from one facility to another. Prisons go to great
lengths to classify and sort offenders based on a variety of characteristics, and offenders’
subsequent assignment to facilities is purposeful. Comparisons of institutional performance
therefore need to take into account differences in the types of prisoners housed in one facility
as compared to another. In the absence of random assignment, the best strategy for doing so
is to measure as many differences among prisons and their prisoners as are theoretically
reasonable and to use multivariate statistical methods to estimate the relationship between
observed performance and any other dimension of interest–in this case, public or private
management.
There exist a number of other methodological challenges that may threaten one’s ability to
draw strong inferences about differences in performance associated with private or public
management. They cannot all be addressed here. The principal lesson here is that evaluation
of whether contracting accomplishes the various strategic objectives the state seeks is a task
that goes beyond mere monitoring and requires trained social scientific analysts. If such
evaluations are to be sought, the demands of such evaluations need to be specified early and
procedures should be established to support the required data collection and analyses.

44

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

4. Texas: Going Private to Expand Capacity
Quickly
When the State of Texas contracted out the management and operation of four 500-bed
prisons in 1989, it initiated what was at the time the most expansive prison privatization
project in the nation. Previously limited to specialized niches in the state=s correctional
system (community based pre-release facilities, for example), contracting extended to main
line state prisonsbig medium security facilities. Since then, the number of contracts has
grown, as has the number of facilities under private management, and Texas now leads the
country in both the number of private facilities under contract and the numbers of prisoners
held in them. By mid 2,000, 14,339 Texas state prisoners were being held in private
facilities, by far more inmates and private facilities than any other state.1 Moreover, as a
consequence of completely independent developments, the state also became home to many
more privately operated facilities that housed prisoners sent by other states. These two
developments occurred more or less simultaneously, but developed on two separate tracks.
While the state government was developing its own approach to contracting with the private
sector to hold Texas inmates, private entrepreneurs were building facilities and offering their
beds to correctional and law enforcement agencies anywhere, but none contracted with the
state government to hold prisoners held under its jurisdiction.
Unlike some other states, the Texas state correctional agency did not back into using private
facilities as a short-term expedient. Rather, the state government undertook its privatization
initiative by writing comprehensive legislation, developing detailed contracts and assigning
ample staff to oversee the program. An independent commission was tasked with
determining if the initiative delivered the financial objectives that the legislature and the
executive branch sought.
This chapter examines why the state turned to contracting and how it has structured its
procedures to accomplish its strategic objectives. It first discusses the events that provided
the impetus for turning to the private sector. It then turns to a discussion of how the
legislature initiated its privatization program, defining its objectives and creating the legal
framework for its implementation. Then it examines how correctional administrators sought
to accomplish these objectives through contract procurement procedures (including design of
the RFP and the contract), how they monitor contractors’ performance, and how the state
evaluates the entire privatization initiative to see if these objectives are indeed being
accomplished.

1

Allen J. Beck and Jennifer C. Karberg, Prison and Jail Inmates at Midyear 2000 (U.S. Department of Justice, Bureau of Justice
Statistics, March 2001), Table 3. This probably excludes prisoners held in state jails.

Abt Associates Inc.

Texas: Going Private to Expand Capacity Quickly

45

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

The Legislature Decides to Contract for New Prisons
The chain of events that led the Texas legislature to enlist the services of the private prisons
industry began in the early 1970s at the desk of Judge William Wayne Justice, United States
District Court judge for the Eastern District of Texas. At the culmination of a federal lawsuit
by a prisoner against the entire Texas Department of Corrections (TDC), the judge ordered a
vast reorganization of the prison system. This culminated, several years later, in the
legislature's decision to embrace privatization as a partial solution to the overcrowding and
deficient conditions that had resulted in the finding of unconstitutionality.
For decades, prisoners had worked from sun-up to sun-down, six days per week to support
the Texas Department of Corrections’ industry operations.2 Inmates were housed and fed
cheaply and were guarded by “building tenders,” armed inmates who were given special
privileges in exchange for their services. Between the late 1940s and 1958, under the
direction of Oscar Ellis, the former commissioner of the Texas Department of Corrections,
the daily cost of supporting an inmate fell from $4 to $1.25.3 George Beto, Ellis=s successor,
extended this tradition of Aefficiency” and Aorderliness,”4 as did Beto=s successor, James
Estelle, who was appointed commissioner of the TDC in 1972. However, it was under
Estelle=s watch that these practices would come to an end.
During the early 1970s, petitions filed against the Texas prison system by inmates had been
accumulating at a fast rate. Many of these came before United States District Court Judge
William Wayne Justice. In 1972, Justice decided to develop a prisoner=s rights case, and he
chose David Ruiz=s petition, among others, to develop the issues. In his petition, Ruiz
alleged that prison officials had harassed him and had denied him his constitutional rights to
medical services and access to the courts.5 The trial began in 1978 as a class action suit on
behalf of all prisoners in Texas against all of the Texas prisons. It lasted 159 days, and the
jurors heard the testimony of 349 witnesses. Finally, on December 12, 1980, Judge Justice
issued an 118-page opinion in which he held that the TDC had violated the constitutional
rights of its prisoners in the following six areas: 1) space per inmate; 2) security and
supervision; 3) health care; 4) disciplinary procedures; 5) access to legal services; and 6)
sanitation and safety.6 The court ordered the TDC to bring its operations in the areas listed
above up to constitutional standards. A consent decree was drawn up in which the parties
agreed to what improvements would be made and the process for making those
improvements. The directives of the consent decree were couched in very general language
for the sake of brevity and to address the similarly broad violations found by the court.

2
3
4
5
6

46

Feely and Rubin, Judicial Policy Making and the Modern State, (Cambridge University Press, 1998), p. 83.
Ibid., p. 83.
Ibid., p. 83.
Ibid., p. 81.
Ibid., p. 84.

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

The parties agreed to the appointment of Vincent Nathan as special master to monitor the
planning and execution of the TDC=s effort as outlined in the consent decree. The order
appointing Nathan as special master allowed him to
observe, monitor, find facts, report or testify as to his findings, and make
recommendations to the court concerning steps which should be taken to achieve
compliance. The special master may and should assist the defendants in every
possible way, and to this end he may and should confer informally with the
defendants and their subordinates on matters affecting compliance.7
Furthermore, the special master had the power to conduct confidential interviews; examine
TDC files; attend all formal meetings of TDC officials; require written reports from any TDC
staff member; and order and conduct hearings on the defendant=s compliance with the court=s
order.8 On the other hand, the order provided that the special master could not Aintervene in
the administrative management of the Texas Department of Corrections,” or Adirect the
defendants or any of their subordinates to take or to refrain from taking any specific action to
achieve compliance.”9 These powers proved to be significant and in turn resulted in
sweeping changes.
Before the Ruiz litigation, the Texas prison system was characterized as stern and one in
which the use of income-producing labor and inmate trustees allowed, in some cases, the
TDC to refund money to the state.10 After the Ruiz decision, Judge Justice and Nathan
implemented vocational programs; increased expenditures on food and health care
significantly; replaced inmate trustees with salaried guards; and demanded the enactment of
legislation that required more due process.11 These initiatives increased the prison system=s
need for funding significantly.
While the state was facing the need to increase spending to make the mandated
improvements, the numbers of prisoners behind bars in the state system were growing. In
1972, when Judge Justice first considered Ruiz's petition, Texas held 15,000 inmates.12 By
1983, that number had more than doubled, growing to more than 35,000.13 Citing frustration
with the demands of the court on one side and the legislature=s reluctance to increase funding
on the other, James Estelle resigned from his position as commissioner in 1983.14 Judge
Justice had also grown frustrated with the pace at which the state was changing its prisons

7
8
9
10
11
12
13
14

Ibid. p. 86.
Ibid. p. 86.
Ibid. p. 86. p. 86 quoting Ruiz v. Estelle (S.D. Tex. 1981) (Memorandum Opinion, July 24).
Ibid. at p. 83.
Ibid. p. 86. 89-90.
Ibid. pp. 92.
Bureau of Justice Statistics, “Prisoners in custody of state and federal correctional authorities” (spreadsheet), at
www.ojp.usdoj.gov/bjs.
Feely and Rubin, Ibid. p. 91.

Abt Associates Inc.

Texas: Going Private to Expand Capacity Quickly

47

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

system and threatened to impose fines if the state did not move faster in meeting the terms of
the consent decree that had been signed years before.
Seeking to comply with the requirements of the Ruiz consent decree and to manage spending
for corrections, Texas State Senator Ray Farabee sponsored Senate Bill 251later enacted as
Chapter 495 of the Texas Government Code in 1987. Chapter 495 authorizes the Texas
Board of Criminal Justice to contract with a private vendor or with a county for the financing,
construction, operation, maintenance, or management of a secure correctional facility.15
Chapter 495 allows the Institutional Division to contract for private construction and
operation of Texas prisons. (In 1989, the TDC was renamed the Texas Department of
Criminal Justice (TDCJ) to reflect the reorganization of Texas= criminal justice system.)
Further enabling legislation was passed to authorize two other new divisions of TDCJthe
State Jails Division and the Parole Divisionto contract as well with private firms to house
prisoners.

The Legislature Defines Its Objectives for Privatization
In the mid-1980s, as the Texas legislature debated the Senate Bill 251 that ultimately became
law, advocates on the national and international stages touted privatization as a means to
more efficient and innovative government. Competition with privately operated facilities,
advocates argued, would spur the development of more effective and more cost-efficient
ways of delivering services in these Apublic monopolies.” In the context of the ideological
battle being waged on both sides of the Atlantic by the Thatcher and Reagan administrations,
privatization was seen as a means of Aturning back the State,” reversing the trend toward
bigger and bigger governments, thereby expanding the scope of the marketplace. This had
greater relevance in post-World War II Great Britain than in this country, but conservative
advocates here nonetheless sought to Aprivatize” the relatively few areas of public service
delivery that remained in the confines of government, including prisons.
Although State Senator Farabee thought that the presence of private managers might
introduce a competitive environment among public correctional managers in Texas, this was
not his primary concern.16 Rather, he sought first to alleviate overcrowding, to comply with
federal court orders, to improve correctional performance (as demanded by Judge Justice),
and to avoid heavy fines.17 By the mid-1980s, Judge Justice had become so impatient with
the pace at which the state was changing its prison system that he demanded that the state pay
a daily fine in excess of $800,000 if it did not improve its efforts to comply with the
mandates of the decision. This caught the attention of policy makers in the legislature. It

15
16

Tex. Gov’t. Code Ann. '495.001 (West 1998).
Interview on December 4, 1998 with former State Senator Ray Farabee, now Vice Chancellor and General Counsel, the University of

17

Texas.
Ibid.

48

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

was clear that Texas needed more beds quickly if it was to avoid paying such fines. It
decided to turn to the private sector to acquire those beds in a speedy fashion and to operate
facilities according to constitutionally acceptable standards.
The activist roles played by Judge Justice and his special monitor, Vincent Nathan,
significantly influenced the creation of the state=s correctional privatization objectives. For
example, the court had found Texas to be deficient in its rehabilitative programming, and
required prison authorities to give more than lip service to this correctional objective.
Consequently, the legislature drafted Chapter 495.001 to direct the Texas Board of Criminal
Justice to Agive priority to entering contracts . . . that will provide . . . facilities designed to
successfully integrate inmates into society through preparole, prerelease, work release, and
prison industries programs.”18 Moreover, the legislation established procedures to ensure that
the private sector provide services that meet the various constitutional standards set by the
court. For example, Chapter 495.003(b) requires that a contractor demonstrate that it has Athe
qualifications and the operations and management experience to carry out the terms of the
contract;@19 and Athe ability to comply with the standards of the American Correctional
Association and with specific court orders.20
However, other factors besides Ruiz influenced the drafting of Chapter 495 and affected how
the law=s objectives were defined. Several legislators and special interest groups were
opposed to privatization and ultimately introduced provisions that shaped the privatization
program. Former Texas State Senator Craig Washington was opposed to privatization for
three reasons. First, he believed that the practice is unconstitutional and that the incarceration
of inmates is a non-delegable state duty. Second, he argued that it is immoral for a private
entity to profit from crime in any manner. Third, he believed that the creation of private
prisons creates the potential for an industrial complex that will create the demand for more
prisoners to feed an increasing supply of prison beds. However, he knew that, given the
severe fines the TDC faced as a consequence of overcrowding, the passing of Senate Bill 251
was inevitable. Consequently, he sought various provisions that would limit the impact of
privatization.21
Unable to win the day on whether incarceration in toto can be delegated, he and others
succeeded in getting into the law provisions that limited those duties that can be delegated to
a private contractor or county operating under a contract authorized by Chapter 495.22 Under
this provision, the vendor may not

18
19
20
21
22

Tex. Gov=t. Code Ann. '495.001(e).
Ibid. at '495.003(b)(1).
Ibid. at '495.003(b)(2).
Interview with Craig Washington, Former Texas State Senator and U.S. Representative,
March 25, 1999.
These provisions differ from the provisions of other states that define which duties are not delegable to the contractor. Namely,
subsection (2) and (4) above prohibit a contractor from executing a task in manner that would favor the inmate. However, there is no
provision that prevents a private contractor from executing a task in a manner that would be against an inmate’s interest. Similar
provisions of other states prohibit the performance of the task without reference to its effect.

Abt Associates Inc.

Texas: Going Private to Expand Capacity Quickly

49

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

(1)
(2)
(3)

compute inmate release and parole eligibility dates;
award good conduct time;
approve an inmate for work, medical, or temporary furlough or for preparole
transfer; or
classify an inmate or place an inmate in less restrictive custody than the custody
ordered by the institutional division.

To protect inmates against what were seen to be profit-maximization incentives (which might
result in cheapening of prisoner care), he and other opponents sought to insert provisions and
language that raised as high a standard as possible for accountability and responsibility of the
private contractors.
Unions representing correctional employees opposed the bill, and their opposition also
affected how the privatization program=s objectives were framed. The Communications
Workers of America, realizing that it could not stop the implementation of privatization,
sought to limit the effect privatization might have on its membership. It stated its opposition
to Senate Bill 251, but voiced its support of measures that would: 1) require comparable pay
for both private sector employees and public sector employees; 2) limit the number of private
prisons; and 3) require cost comparisons based on the operation of facilities housing similar
inmates with similar programs.
The efforts of politicians and special interest groups opposed to privatization were not in
vain. Chapter 495 was written to authorize holding only minimum or medium security
inmates in a private facility.23 This established firewalls in the prison system to contain the
threat of privatization. The legislation also prohibits the Board from undertaking true
privatization: converting to private operation any facility already operated by public
employees or already being built for public operation.24 Contracting out would thereby slow
the growth of the public employment but would not eliminate current employees= jobs.
The law also included a provision that any contractor must offer programs of equal level and
quality offered by state-operated facilities housing similar inmates at a cost of at least 10
percent less.25 A separate governmental entity—the Sunset Advisory Commission—was
assigned responsibility for making the comparison between the costs of publicly operated and
privately operated facilitieswhether direct or indirect.26 According to Senator Farabee, this
provision was inserted to help all concerned to Aswallow the pill.”27 This was especially
significant. Even though it is commonly assumed that the best measure of privatization=s
success is whether it provides more cost-effective service, adding this objective to the Texas

23
24
25
26
27

50

Tex. Gov’t. Code Ann. '495.002.
Ibid. at '495.006.
Ibid. at '495.003(c)(4).
Ibid. at '495.003(e).
Interview with Ray Farabee, December 3, 1998.

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

legislature=s mission statement for privatization was done to win over opponents. Cost
efficiency therefore appeared to be a concern of secondary importance, compared to acquiring
more prison beds speedily.
In summary: the state=s privatization program was undertaken to accomplish several different
objectives. The impetus for the program was the need to acquire additional prison capacity
quickly and to satisfy the federal court=s demands for constitutionally compliant
imprisonment. The speedy acquisition of prison capacity was therefore one of the primary
objectives, as was the goal of having these new prisons meet constitutional standards. The
goal of providing more cost/effective services than the TDCJ was providing was also added,
but this was not the dominant objective.

Implementing the Privatization Program
The Texas legislature passed Senate Bill 251 in the 70th Legislative Session, in 1987, and
TDC was given authority to contract with private vendors for the construction, management
and operation of prison facilities. TDC quickly began to translate the legislature’s mandate
into action. Requests for proposals were written and issued in mid-1987, offering private
firms opportunities to design, locate, construct and manage operations of four medium
security 500-bed prisons, at Venus, Kyle, Bridgeport, and Cleveland, Texas. Although the
state had requested bids to build and operate these prisons, it subsequently determined that it
would be less costly if the state managed the construction rather than contracted for it. The
contracts were therefore limited management and operation responsibilities. Ultimately,
awards were made to two vendors, the Wackenhut Corrections Corporation and the
Corrections Corporation of America, which were responsible for operating two 500-bed
facilities each.. These prisons first opened their doors to inmates during the summer of 1989.
House Bill 93, 72nd Legislative Session (1991) extended further TDCJ=s options for
privatization by allowing the construction of an additional 2,000 beds. Contracts were signed
with US Corrections Corporation, Concept, Inc., Wackenhut Corrections Corporation and
Corrections Corporation of America for three additional 500-bed facilities and a 500-bed
expansion at an existing 500-bed facility. These additional facilities began operation in the
summer of 1989.28
In 1993, six years after passing the original enabling legislation, the legislature created the
State Jail Division within the TDCJ and extended to it the authority to contract for private
correctional services. This division’s purpose is to manage relatively short-term, low custody
offenders without having to house them in state prisons.29 State jails house two types of
offenders: 1) fourth degree felons and 2) those in transition from county jails to state

28
29

Internet: http://www/tdcj.tx.us/opsupport/cont-fac-pvt.prison.htm
Interview with Ron Champion, then the Assistant Director, Texas Department of Criminal Justice, State Jail Division, April 20, 1998.

Abt Associates Inc.

Texas: Going Private to Expand Capacity Quickly

51

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

prison.30 Fourth degree felonies generally include drug, alcohol and/or property offenses.
The sentences are two years, with no chance for parole or good time.31 The construction of
these state jails is a relatively inexpensive dormitory style; however, the legislature commits
substantial funding towards programmatic needs. Consequently, each inmate receives an
individualized treatment plan to assess his or her programmatic needs during his sentence.
The thinking is that these are offenders who present low security risk but relatively high
potential for reintegration into society.32
In the first paragraph of Chapter 507, which authorizes the State Jail Division, the law grants
the division authority to contract out for correctional facilities. It provides that “The state jail
division, with the approval of the board, may contract with the institutional division, a private
vendor, a community supervision and corrections department, or the commissioners court of
a county for the construction, operation, maintenance, or management of a state jail felony
facility.”33 State Jail Division facilities must comply with the rules and regulations set forth
by the TDCJ; however, it is free of certain requirements that were imposed on the
institutional division, such as the 10 percent savings provision.
In addition to the Institutional Division and the State Jails Division, the Parole Division of
the TDCJ was also authorized to contract for privately operated imprisonment services. This
division is responsible for supervising offenders released from prison who are serving out
their sentences in the community.34 Persons who violate their parole conditions are returned
not to the state prisons operated by the Institutional Division but instead to separate parole
violator facilities under the authority of the Parole Division. Some of these facilities are
operated by contractors.
By the end of 1997, the State Jail Division and the Institutional Division were contracting
with 5 firms to operate 20 facilities, holding a total of 13,314 inmates.35 By mid-2000, the
numbers had grown even largerB14, 339 in the Institutional Division alone, and an
undetermined number in the state jails and in parole violator facilities.36
Privatization consequently played a significant role in the state=s attempts to solve its
overcrowding problem and to improve the conditions of confinement in its prisons. During
the mid-1980s, when the privatization initiative was launched by the legislature, the Texas
prison system was already one of the biggest in the nation, and it has grown even bigger in
subsequent years. By mid-2000, the numbers of Texas state prison inmates had grown to
163,500, only a thousand fewer than California=s, the largest correctional system in the U.S.

30
31
32
33
34
35
36

52

Ibid.
Ibid.
Ibid.
Tex. Gov’t. Code Ann. '507.
http://www.tdcj.state.tx.us/parole/parole-home.htm
From survey of state correctional agencies, Abt Associates Inc., 1998.
Beck and Karberg, Prison and Jail Inmates at Midyear 2000, Table 3.

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

at the time.37 The TDCJ=s budget had grown to $2.3 billion by fiscal year 2000.38 Even
though government-operated facilities held the vast majority of these 163,500 prisoners, the
state has relied significantly upon the private sector to handle some of the increase in demand
for prison beds. By mid-2000, 9 percent of the state=s prisoners were housed in privately
operated facilities.39
Texas has gone further than most states in developing a hybrid public/private prison system,
and has over the years refined procedures to achieve its strategic objectives for contracting.
These procedures include specifying what the state wants from a contractor and how the
public/private partnership is to function, establishing procedures whereby contractors bid and
are selected, monitoring the contractors= ongoing performance, and evaluating whether the
state=s strategic objectives with respect to contracting are being achieved.

Structuring What Texas Seeks in the Market: The RFP
During the decade following the state=s first effort to contract with private prison operators in
1987, each of the divisions designed their own bidding documents and followed their own
contracting procedures. In 1998, TDCJ centralized the process with the creation of the
Purchasing and Leases Department within TDCJ=s Financial Services Division. Any of the
operating divisions seeking to contract specifies the goods and services it wants, and passes
the request to the Purchasing and Leases Department. This department caries out the rest of
the process to ensure that it is consistent with the laws and regulations of the state.40 This
centralization of procurement duties has helped the contracting process maintain consistency
from contract-to-contract throughout the TDCJ, has helped enforce the state’s rights by more
accurately transforming the objectives of the requesting division into the effective language
in the RFP/contract, and has eliminated redundancy between the RFP and contract. A
common problem in early contracting arrangements throughout the U.S. was a lack of
consistency in the contract documents. Contracts often incorporated, by reference, the
statements of work that were issued in the request for proposal as well new provisions that
were negotiated by the winning bidder and the state. Because contractual provisions and the
referenced statements of work were not always Aharmonized@ before the parties signed the
documents, ambiguities were sometimes created that led to confusion about rights and
responsibilities.
In 1998, the Purchasing and Leases Department developed a standardized RFP and contract
for private operational and management services in prisons.

37
38
39
40

Beck and Karberg, Prison and Jail Inmates at Midyear 2000, Table 2.
http://www.tdcj.state.tx.us/finance/budget/finbud-goal.htm.
Beck and Karberg, Prison and Jail Inmates at Midyear 2000, Table 3.
http://www.tdcj.state.tx.us/content/fin%20servic.../fin%20svcs%20prchsg%20leas%20home.htm.

Abt Associates Inc.

Texas: Going Private to Expand Capacity Quickly

53

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

Operations, Not Ownership

As mentioned above, the first RFP issued by TDC in 1987 solicited bids for both
construction and operation of the first four 500-bed prisons. This was thought to be the
fastest way to acquire the additional capacity–the principal motivation for contracting.
Between the time of issuing the request and writing the contracts, however, the state
determined that it would be less costly for the state to build and own the facilities, and to
have private contractors assume only the responsibility for managing them. The state also
recognized other advantages of ownership. Owning the buildings would make it easier for
the state to assume control of the facilities in the event of a private firm’s failure, or in the
event that the state or the contractor elected to terminate the contract. It was thought that
state ownership would also minimize the risk of entrenchment by incumbent contractors,
thereby increasing the likelihood that subsequent recompetitions remain competitive. Since
then, the state has adopted this as a policy, and has contracted only for the management of
government-owned facilities.
Defining the Contractors’ Obligations: Striking a Balance Between Specificity and Flexibility

A tension has often existed in contracting practices between the state setting forth not only
the objectives to be accomplished but also the precise methods by which the contractor is to
accomplish these. Some of the early RFPs for privately operated prisons amounted to
handbooks on how to operate a prison, usually modeled after the rules, regulations, and
procedures established for the government-run facilities in the jurisdiction. An RFP issued
by the State of Tennessee in the 1980s, for example, ran to hundreds of pages. No contractor
chose to bid on it.
The argument often heard against such detailed specification of procedures is that the
contractor should be given the opportunity to devise more effective and efficient ways of
accomplishing the state=s objectives. For example, the Federal Bureau of Prisons consciously
sought to encourage innovation by limiting specification of procedures to those deemed to be
absolutely essential in its solicitation for operation and management services in its Taft, CA
facility. Similarly, Florida’s RFPs leave room to contractors for a similar reason (see the
following chapter).
Texas did not undertake privatization for the purpose of encouraging innovation, however.
As discussed above, its principal objectives were and have been to acquire needed capacity
while ensuring, to the maximum extent possible, that the conditions of confinement in
privately operated facilities met the standards established by a vigilant federal court. The
state=s experience with federal court oversight had been difficult enough with the stateoperated prisons. Faced with what the court saw as TDC=s/TDCJ’s insufficient compliance,
the judge and his special master issued orders setting forth, in precise detail, operational
procedures to be followed in many areas of prison administration. These included, for
example, lists of toiletries that prisoners would be given. (For this and other reasons, critics
54

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

of federal court intervention decried what they saw as an increasingly Aadministrative”
judiciary, overstepping what they saw as the boundary separating properly executive and
judicial functions.)
Because state government remain legally liable for contractors= failures to meet constitutional
standards of care, the state sought to minimize its risks. Thus, when considering how to
define the contractors= obligations and how to encourage desired performance, it chose to
blend mandatory requirements, on the one hand, with performance objectives and incentives,
on the other. However, by setting standards that have to be met rather than specifying precise
procedures to be followed (by means of an operations manual, for example), the state is
allowing the contractor to devise its own procedures as long as they conform to the
established standards of practice.
The Ruiz litigation arose, in great part, because Texas inmates had been confined in
conditions that were found by the court to be below a constitutionally acceptable level.
TDCJ=s standardized RFP and contract therefore create explicit and obligatory standards of
confinement in several important areas, including daily living conditions, security and
discipline, health care, and legal access. These are often very detailed and leave the
contractor little discretion in how the requirement is to be met.
Daily Living Conditions

The standardized contract covers everyday personal services such as food services,
commissary, hygiene services (laundry, barber shop), telecommunications, recreation, and
mail.41 Although some provisions are more comprehensive than others, all defer to some
official, external standard—either ACA standards, TDCJ standards, court order standards, or
any combination of the three.
Provisions addressing hygiene services, commissary services and telecommunications are
more comprehensive. In addition to meeting the standards of court orders, the ACA and the
TDCJ, the contract spells out how these services are to be delivered, mirroring the kinds of
specific directives imposed by the federal court on the state’s prisons administrators. For
example, these provisions require the contractor to:

41

•

Provide a list of essential items provided to offenders.

•

Provide a quantity and distribution schedule of essential items.

One of the performance measures in the appendix requires the contractor to obtain ACA accreditation within 18 months. Although
ACA standards apply to everyday living conditions, the focus of these standards is very broad and apply to many other issues.

Abt Associates Inc.

Texas: Going Private to Expand Capacity Quickly

55

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

•

Post essential item type and distribution schedule in all offender living areas
and the offender orientation handbook.

•

Establish an accountability system to ensure that all offenders receive essential
items.

•

Establish minimum inventory levels based on demand, distribution schedule
and order/ship time to ensure an adequate quantity on hand to meet facility
demands.

Health Care

Claims of inadequate health care were central features of the Ruiz litigation and have been
salient elements of many other large and successful class action suits against prison systems.
Beyond heightened federal court interest, prisoner health care poses other risks as well: it can
be very expensive and unpredictably so. In its contracts with private contractors, TDCJ seeks
to minimize the risks associated with the contractors’ delivery of health care services.
The standardized contract makes the contractor generally responsible for all medical care and
costs while the offender is in the custody of the contractor. That is, the contractor is required
to provide complete on-site care services including medical, dental, mental health,
pharmaceutical, medical records, emergency care and sick call services.42 The contract also
specifies who has responsibility when inmates are transferred from the TDCJ to private
facilities. If the TDCJ has initiated a procedure involving a medical or dental prosthesis prior
to transfer, TDCJ will be responsible for completing the procedure. Some ambiguity exists
as to when and if a procedure has been “initiated.” If there is a dispute over whether a
procedure has been initiated prior to transfer, the contractor may appeal the decision to the
Associate Division Director of Health Services.43 If an offender is offered specialty services
while in TDCJ units and refuses evaluation and/or treatment, the contract does not consider
this “initiating” the service, and any future requests for such specialty services become the
responsibility of the contractor.44
The contract establishes limits on the contractor’s liabilities for prisoner health care. For
example, the contractor is responsible for the costs associated with the first 48 hours of
hospitalization—including transportation. If the inmate requires longer hospitalization, the
TDCJ assumes responsibility for the costs.45 Moreover, the contractor is responsible for
providing security for any offender admitted to the hospital during the first 96 hours of

42
43
44
45

56

TDCJ RFP/Contract for Venus Pre-Release Center, p. 34 (June 29, 1998).
Ibid. p. 35.
Ibid. p. 34
Ibid. p. 35

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

hospitalization. After such time, plus a reasonable time for the TDCJ to assume security
duties, the TDCJ is responsible for securing the inmate.46
In the standardized contract, the TDCJ reserves the right to provide complete on-site health
care services if it believes that it can provide the required services at or below the cost offered
by the contractor.47 For this reason, the RFP asks bidders to prepare an alternative pricing
schedule that offers a per diem rate excluding the price for health care services. In the event
that the TDCJ decides to take over the facility=s health care operations, this pricing schedule
is activated.48
The contract also requires the contractor to maintain offender health files in accordance with
TDCJ Health Services policies and procedures.49 All offender medical records and files are
the property of TDCJ, according to the contract. Prescribing uniformity and property rights
of the records allows for a relatively smooth transition in the event the TDCJ decides to
assume the health care operations of a privately operated facility.
The contract establishes performance measure for health care and financial incentives to meet
them. It requires the contractor to provide acceptable medical treatment in accordance with
TDCJ and National Commission on Correctional Health Care (NCCHC) standards. Failure
to do so and failure to correct the problems within 30 days can result in a daily fine of
$1,000.50 Second, it requires that the contractor to obtain NCCHC accreditation within 14
months of opening the facility. Failure to do so can lead to a decrease in unit monthly per
diem payments.51
Safety, Security and Discipline

In the area of prison safety, security and inmate control, the TDCJ contracts allow the
contractor some room to innovate or, at least, to determine how it wants to perform the
required functions. The state does so by establishing standards and policies rather than
enumerating when and how the task must be executed, which allows the contractor more
operational options. For example:
Safety:

Contractor shall operate and maintain the facility in compliance with all
applicable federal, state, and local safety and fire codes and applicable court orders, TDCJ
policy statements, and ACA standards relative to safety. Contractor shall establish a safety
program.52

46
47
48
49
50
51
52

Ibid. p. 35.
Ibid., 6 (June 29, 1998).
Ibid., p. 36 (June 29, 1998).
Ibid., p. 36 (June 29, 1998).
Ibid., p. 88 (June 29, 1998).
Ibid., p. 88 (June 29, 1998).
Ibid., p. 44.

Abt Associates Inc.

Texas: Going Private to Expand Capacity Quickly

57

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

Contractor shall adequately secure buildings and provide other security
equipment necessary to maintain control of offenders. Contractor shall provide
reimbursement to TDCJ of costs incurred by TDCJ in the event of escape(s) or other
extraordinary events at the facility or from any location where the contractor is responsible
for the offender.53
Disciplinary Rules and Regulations: Contractor shall impose discipline through rules,
regulations and orders pursuant to an offender disciplinary system meeting or exceeding
ACA standards, court orders and TDCJ policy.54
Security:

Whereas the contracts allow private firms some latitude in designing procedures for
accomplishing these objectives, provisions governing the use of force are much more
prescriptive. This is an area where the state’s liabilities are most keenly felt, and the
contracts leave little room for ambiguity or invention.
Limitation as to
location

Use of Non-Deadly Force

! on the grounds of the
facility

! cases of self-defense against an
offender

! transporting offenders

! protection of a third party or property
where the third party is subject to
imminent and immediate threat by
the offender

! pursuing escapees
from the facility

! regaining control of inmates in the
event of a mutiny, rebellion, riot or
disturbance— only under the control
of the assistant facility administrator
or higher authority, except in cases of
emergency

Use of Deadly Force

! to prevent serious
injury to an
individual
! to quell an inmate
uprising in which
serious bodily harm
is imminent and
lesser means have
failed
! prevention of escape

! temporary isolation or other forms of
confinement where lesser means have
failed
Source: TDCJ RFP/Contract for Venus Pre-Release Center, p. 46-47 (June 29, 1998).
Personnel Requirements

Although concrete, bars, locks, and guard towers are the most obviously recognizable
characteristics of prisons, the conditions of confinement turn more on the character of the
services carried out within a prison than on the hardware. The prison staff is the most

53
54

58

Ibid., p. 44.
Ibid., p. 44.

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

important determinant of these conditions. The TDCJ contracts aim to maintain acceptable
conditions by means of several provisions governing the makeup of the contractor’s labor
force. First, they enumerate the minimum qualifications and amount of training key
employees must posses in order to work in the proposed facility. Second, the contracts
impose requirements on the staffing practices of private contractors to ensure that the
contractor maintains that level of quality throughout its staff.
The standardized contract requires the contractor to provide training in accordance with
TDCJ policy. That is, new security staff must complete a 160-hour pre-services training
academy and 80 hours of on-the-job training for new security staff. New non-security staff
must complete a 120 hour pre-service training program approved by the TDCJ. Both
categories of personnel must complete a TDCJ approved annual 40 hour in-service training
program.55 The contract also requires that new security staff complete a minimum of 160
hours of pre-service training.
Provisions of the model contract that focus on the staffing practices are even more
comprehensive. The contract requires that qualifications and job descriptions correspond
with like positions in the TDCJ system. If not, all exceptions must be noted.56 Furthermore,
if a position title corresponds with a TDCJ position title, the job description must also follow
the TDCJ job description.57
The contract requires rigorous procedures for screening applicants. It requires the contractor
to complete a background investigation and to make the results available to the TDCJ. This
provision applies to the contractor’s employees as well as the employees of independent
consultants, independent contractors and volunteer workers who work at the facility on a
routine basis. Furthermore, security employees employed by private contractors are subject
to the same criteria as TDCJ-employed security employees.58 The contract also requires that
job applicants who currently are employed by TDCJ or who have previously been employed
by TDCJ must release to the contractor information concerning all disciplinary actions taken
during their employment with TDCJ as well as their TDCJ re-hire eligibility status.”59 The
TDCJ has encountered situations where discharged TDCJ personnel have found employment
at private firms.60 This requirement serves to prevent such situations.
The contract also requires the contractor to maintain a staff of sufficiently qualified
employees. The bidder must submit a staffing plan showing the number of security
supervisors/security administrative staff as well as correctional officers. The contractor must

55
56
57
58
59
60

Ibid., p. 12.
Ibid., p. 13.
Ibid., p. 13.
Ibid., p. 15.
Ibid., p. 13.
Interview with Sharon Keilin, Assistant Director for Operational Support, Texas Department of Criminal Justice, March 25, 1999.

Abt Associates Inc.

Texas: Going Private to Expand Capacity Quickly

59

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

ensure that vacancies will be filled within 30 days of the vacancy occurring. If the position
remains open more than 30 days, TDCJ=s monthly payment is reduced by an amount equal to
the salary and fringe benefits for each day the position was vacant. If the contractor fails to
fill a vacant position within 90 days from the date of vacancy, the TDCJ=s monthly contract
payment is reduced by the base salary (which includes company fringe benefits) plus an
administrative penalty equal to the base salary of the vacant position for each day that the
position remains vacant in excess of 90 days.61 Ron Champion, formerly of the State Jail
Division, commented that some contractors were slow in filling vacant positions and were
able to pocket the amounts that would have been paid to an employee if the position had been
filled.62 This provision allows the contractor time to search for a replacement, but it provides
an incentive for the contractor to act swiftly. Furthermore, the provision imposes sanctions in
escalating levels. If the contractor fills the position on the 29th day, then there is no penalty
and it pockets the amount not paid to an employee who would have filled the position.
Between the 30th and 90th day, the provision simply deducts the amount that the contractor
saves. Essentially, this is a wash for the contractor. It is not until the 90th day that sanctions
are actually imposed and the contractor incurs a penalty.
Promoting Prisoner Rehabilitation

The absence of adequate rehabilitation services and programming to inmates was a central
element of the Ruiz litigation, and Judge Justice ordered the state to remedy this. TDCJ’s
contracts therefore include a number of provisions to assure sufficient services in the
privately operated facilities. These provisions include specification of procedures to follow,
outcomes measures to be achieved, and financial incentives.
The contract requires the contractor to provide life skills counseling, substance abuse
education, and a “full range of academic and vocational programs.@63 This includes programs
for the illiterate through college level courses, as well as a properly stocked library. To
encourage the delivery of adequate services and high levels of inmate participation, the
contract establishes various performance outcomes. Some refer to procedures. For example,
some reward the timely evaluation and enrollment of inmates into educational and
rehabilitation programs (see chart).
Similarly, procedural performance objectives are established for educational programs:

61
62
63

60

•

At least 60 percent of the offender population will be enrolled in education
programs.

•

Students will participate at least three hours per day five days per week.

RFP/Contract for Venus Pre-Release Center, p. 14.
Interview with Ron Champion, Assistant Director, Texas Department of Criminal Justice, December 3, 1998.
RFP/Contract for Venus Pre-Release Center, p. 18.

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

•

Approximately 20 percent of the offender population will be involved in
vocational training.

•

70 percent of vocational participants will complete the course and earn a
certificate of completion.64
Duty

Acceptable
Standard

Penalty

Incoming eligible offenders placed
into education programs (academic,
vocational, pre-release) within 10
days of arrival.

97-100%

Number of days past the 10 day
requirement times the per diem for
delinquent offenders.

Percentage of eligible offenders
completed or enrolled in Pre-release
or life skills

75%

Total percentage below 75% times
the combined salaries of all
educational staff for the period of
that particular semester.

Evaluation of ITP (Individualized
Treatment Plan) within the required
10 day time period.

97-100%

Number of days past the 10 day
requirement times the per diem for
delinquent offenders.

Source: TDCJ RFP/Contract for Venus Pre-Release Center, p. 87-88 (June 29, 1998).

Procedural targets and threats of sanctions for low levels of participation do not, in
themselves, do anything to encourage the development of effective programming, however.
For that purpose, the contract includes provisions designed to encourage such development.
These measure educational outcomes of students. For example:
•

There will be an average student growth gain of 50 percent between testing
periods.

•

65 percent of test takers who take the entire GED battery will pass the battery.

•

The pass rate on individual GED tests will average 75 percent.

•

The average numerical course grade for all vocation participants will be 85.65

Selecting Contractors
Texas seeks to limit its risks by contracting only with firms that meet certain criteria. For
example, the solicitation for the 1,000-bed pre-release center at Venus, Texas, establishes the
following capabilities that must be demonstrated by any bidder: 1) a minimum of two years

64

Ibid., p. 32.

65

Ibid., p. 32.

Abt Associates Inc.

Texas: Going Private to Expand Capacity Quickly

61

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

experience managing and operating a secure adult correctional facility housing
minimum/medium security felons with at least a 500 bed capacity; 2) the ability to obtain the
minimum commercial insurance required; 3) the ability to commence operations without
financial assistance from TDCJ; and 4) provide upper level management personnel to operate
and manage the solicited facility that meet or exceed TDCJ minimum standards for like
positions.66
These requirements balance the state’s interest in allowing competition while restricting the
playing field to capable players. The latter three requirements aim to weed out those firms
that appear to pose the greatest risk of being financially or operationally incapable of
executing the terms of the contract. At the same time, the competition is not necessarily
limited to the biggest two or three firms. Allowing bids from firms that have not yet operated
prisons of the same size (but half the size) opens the door wider than it would be if the entry
requirement was experience having run a 1,000-bed facility.
The evaluation procedures are designed to favor the bidder proposing not the least costly
service but the one offering the best operational plan–especially the best plans for
rehabilitation programming. (All bidders must meet the threshold 10 percent cost savings
requirement, however, except for proposals submitted to the State Jails Division.) This
emphasis on operations and quality of programming reinforces and is consistent with the
overall privatization program’s mission to obtain constitutionally acceptable conditions of
confinement and programs supporting prisoner rehabilitation. Accordingly, bidders’ offers
are given separate scores for each of four main areas. These are, in descending order of
importance, 1) the operational plan, 2) cost, 3) general, and 4) historically underutilized
businesses. The score given to the operational plan is aggregated from scores on four
subcategories. These are, in descending order of importance, the proposed rehabilitation
programs, security, administration and quality control.
In response to the 1998 Abt Associates survey of state and federal correctional agencies,
TDCJ reported that of the 20 facilities under contract on January 1, 1998, the state procured
18 contracts competitively, having issued a request for proposal. The other two were solesource procurements. As would be expected given the stated criteria for evaluating
proposals, the TDCJ reported that it considered factors other than price. However, it is
interesting that winning bids in 12 of the 18 competitions were the lowest priced bids. 67

Monitoring Performance
Monitoring the private firm’s performance following the award of the contract is an essential
element of the state’s privatization program. The standardized contract used by TDCJ
requires both self-review by the contractor as well as review by TDCJ staff. To support the

66
67

62

Ibid., p. 119.
Abt Associates Survey, 1998.

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

state monitors’ duties, the contractor must provide a variety of data that will enable the state
to monitor achievement of the established performance objectives, such as those discussed
above for educational programs. Similar performance outcomes are established for health
care, maintenance, staff qualifications and turnover/vacancies, among other areas. (A
substantial proportion or these outcomes pertain to rehabilitation programming and health
care.) The contract requires the private firm to submit up to 47 different reports from the
contractor. A little more than half of these reports are periodic—weekly, monthly, quarterly
or annually. The other half occur prior to the commencement of operation or upon some
triggering event.
In addition to documenting performance for the state, the TDCJ conducts audits at the
facility—a right that is explicitly established in the standard contract. These audits are
conducted by the contract monitor assigned to the facility, by other TDCJ staff, and by
authorized representatives of the local school district. These auditors interview staff and
inspect the services and facilities. By contractual right, such audits can occur at any time.68
As of December 31, 1997, monitors of ten of the 20 facilities were on site full-time. Four of
these facilities were state jails and six were institutional units. The TDCJ reported that
monitors spent approximately 25 hours, or less, of their time monitoring five of the 20
privately operated facilities. These five facilities were under contract with the Parole
Division. Monitors of the Institutional Division and State Jail Division facilities received
either daily or weekly written reports. Monitors of the Parole Division facilities received
monthly written reports. All of the facilities underwent security audits.

Did Texas Achieve Its Objectives in Privatizing Prisons?
As discussed above, the legislative authors of the prison privatization program sought three
principal objectives: to alleviate overcrowding in the state’s prison system, to improve the
conditions of confinement in that system, and to avoid having to pay stiff fines levied by the
federal courts. Did contracting accomplish these objectives?
Initially, the legislature thought that relying on the private sector would speed acquisition of
new prison beds, thereby accomplishing one of the objectives (relieving overcrowding). But
the state quickly backtracked and decided instead to build and own the prisons, and to
contract only for operations and management. Financing and construction thereby followed
the pace set by other state facilities. Contracting for management services alone doesn’t
seem to have relieved overcrowding any faster than would have happened if the state built
and then operated the facilities.

68

RFP/Contract for Venus Pre-Release Center, p. 53.

Abt Associates Inc.

Texas: Going Private to Expand Capacity Quickly

63

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

The second objective of the privatization program–to improve conditions of confinement in
the prison system—may have been advanced somewhat by reliance upon contractual
obligations to obtain constitutionally acceptable conditions. But did contracting achieve this
goal any faster than was achieved in the state-operated facilities? Probably not.
An indication of how public correctional administrators evaluated the overall performance of
the privately operated facilities under contract can be seen in their responses to the 1998 Abt
Associates survey. Contract monitors in TDCJ were asked about the responsiveness of
contractors in reporting to the correctional agency. The TDCJ reported that three facilities
were “responsive,” 13 were “very responsive” and four were “fully responsive.” This survey
also asked to compare the vendors= performance to the requirements of the contract. The
TDCJ reported that, in one case, the contractor exceeded the requirements of the contract. In
14 facilities, the contractor equaled the requirements of the contract. The TDCJ also reported
that the performance of these same 15 facilities was equal to comparable publicly operated
facilities.
Cost Savings?

The most elaborate of the performance measurement procedures were erected to assess the
extent to which the legislatively required 10 percent savings objective was achieved with
each of the contracts. The task of determining if the private firms succeeded in meeting the
savings target was given to the Sunset Advisory Commission. The statute specified that the
commission should analyze the cost and quality of services in the private prisons as
compared to the cost and quality of any similar state services. However, as the commission
noted in its 1991 report, AThe development of a cost estimate was complicated by the fact that
the TDCJ did not, and still does not, operate a comparable facility.@69
The commission therefore requested that the Texas Department of Criminal Justice develop
two cost estimates: (1) the total cost to the state of contracting for the operation of the
privately operated prisons, and (2) the total cost to the state if TDCJ operated the facilities
directly. The estimate of the former cost included actual contract costs plus a number of
other direct and overhead costs borne by the state in support of this contract. The second
estimate included all direct and indirect costs that would be associated with direct
government operation of the CCA facilities and, separately, the Wackenhut facilities. The
commission concluded, from a comparison of the actual costs and the estimated costs, that
contracting saved the state 14–15 percent—more than the required 10 percent.
This computation includes a strange accounting of savings, however. The Sunset Advisory
Commission counts payments by each of the firms to local governments in amounts that
approximated the property taxes the firms would have paid if the prisons were privately

69

64

The State of Texas, Sunset Advisory Commission, Recommendations to the Governor of Texas and Members of the 72nd Legislature:
Final Report (March 1991).

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

owned. Because the state owned the facilities, and the firms contracted only to operate them,
the firms were not legally liable for any such taxes. Nor was the state liable for such taxes.
These “payments in lieu of taxes” appear to be no more than a fee paid by the private firms to
governments so that the net cost of contracting to Texas exceeds the 10 percent threshold.
Not counting these payments, private operation appears to cost the state 8.8 or 9.7 percent
less than direct TDCJ operation, depending upon the facility—or slightly less than the
savings required by statute.
An examination of the commission=s report indicates that the principal sources of savings
associated with contracting, apart from this payment, were lower operating costs and a lower
overhead expenditure by the state. Operating costs were $1.53 to $1.96 lower on a daily per
prisoner basis, state overhead costs were $1.36 less for the private facilities. The precise
reason for the lower operating costs cannot be identified because the report provided only
aggregate-level expenditure data. That is, we cannot determine the extent to which lower
operating costs reflected lower spending for staff salaries, fringe and retirement benefits, for
fewer staff, or for non-labor costs.
In January of 1999, the Criminal Justice Policy Council conducted another cost study.70
Among other things, this report compared the operational costs per inmate in publicly
operated facilities and privately operated facilities. In calculating these costs, the council
focused only on operational costs; costs for construction and major renovations were
excluded as well as fringe benefits paid by the state.71 (The report does not reveal whether
payments by private operators to the state in lieu of property taxes were treated as part of the
savings to the state as they were in the Sunset Commission report.) The council estimated
the average cost per inmate/day during fiscal year 1998 for publicly-operated Institutional
Division facilities was $38.71. The equivalent cost to the state of inmate care in privately
operated facilities under contract with this division was $34.39. During FY 1997, the costs
were $39.28 and $34.04, respectively.72 Thus, the privately operated facilities as a whole
delivered a savings of 11.2 percent compared to publicly operated facilities during FY
1998—beating the 10 percent savings mandate. During FY 1997, the savings were
greater–13.3 percent.
The study also estimated that the that the costs per inmate per day at publicly operated state
jails were $32.21 and $31.07 for 1997 and 1998 respectively. This compared to $29.37 and
29.69 during 1997 and 1998, respectively, for the privately operated facilities.73 These
represent less than a 10 percent savings, but this specific savings mandate does not apply to
privately operated state jails.

70
71
72
73

Criminal Justice Policy Council, “Oranges to Oranges: Comparing The Operational Costs of Juvenile and Adult Correctional
Programs,” January, 1999.
Ibid., p. 5.
Ibid., p. 13.
Ibid., p. 12.

Abt Associates Inc.

Texas: Going Private to Expand Capacity Quickly

65

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

In short: the evidence from Texas suggests that the private firms are delivering a service that
would cost the government approximately 9–10 percent more if the state=s corrections
department operated the facilities directly. This assumes that the estimates of the
department=s costs of direct operation are accurate, of course. Lacking more information
about how these costs were estimated, it is not possible to evaluate them.

66

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

5.

Florida: Seeking More Cost-Effective
Performance

Florida was a seedbed for some of the earliest correctional privatization projects in the
nation. One of the first efforts to transfer an existing public correctional facility to a private
agency occurred in 1982, when a contract was given to the Eckerd Foundation, a nonprofit
arm of a major drugstore chain and drug manufacturer, to operate the Florida School for Boys
in Okeechobee.1 A few years later, in 1985, one of the first privatizations of a secure adult
correctional facility occurred when the Bay County commissioners transferred responsibility
for managing the county jail from the local sheriff to the Corrections Corporation of America.
In that same year, the legislature passed Chapter 944.105 to authorize the Florida Department
of Corrections “to enter into contracts with private vendors for the provision of the operation
and maintenance of correctional facilities and the supervision of inmates.” It was not until
almost a decade later that the state began contracting to hold state prisoners, however.
Although the legislature had signaled its interest in contracting with private correctional firms
to operate prisons, the Department of Corrections did not exactly rush to undertake such
contracting. The department’s leaders felt fully competent to manage any needed prisons.
Consequently, in 1993, the legislature passed a second law, the Correctional Privatization
Act, which created a separate agency, the Florida Privatization Commission, to carry out its
directives. This is the only such agency in the nation independent of a department of
corrections dedicated to privatization. The Commission adopted a more aggressive stance, in
accordance with its mission, and undertook to contract for several prisons. By the end of
1998, the state had 3,877 of its prisoners held in five privately operated prisons, including
South Bay Correctional Facility, which held the highest proportion of maximum-security
inmates of any privately operated facility in the country. This chapter examines Florida’s
experience with contracting for private imprisonment.

The Legislature’s First Try
In passing Chapter 944.105 into law in 1985, the legislature articulated its objectives
regarding privatization. First, the statute requires the contract between the Department of
Corrections and a private firm “to offer a substantial savings to the department, as determined
by the department.”2 Second, Chapter 944 requires the contract to provide “for the same

1

Douglas McDonald, "Private Penal Institutions," in Michael Tonry (ed.), Crime and Justice: A Review of Research (University of
Chicago Press, 1992), pp. 374. A decade earlier, Massachusetts closed its publicly operated training schools and turned to smaller
privately operated facilities, which amounted to "privatization" by another name ("deinstitutionalization"). See Robert B. Coates,
Alden D. Miller, and Lloyd E. Ohlin, Diversity in a Youth Correctional System: Handling Delinquents in Massachusetts (Cambridge,
MA: Ballinger, 1978).

2

Fla. Stat. ch. 944.105(1)(a) (1997).

Abt Associates Inc.

Florida: Seeking More Cost-Effective Performance

67

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

quality of services as that offered by the department.”3 In other words, the Florida legislature
sought more cost effective services than it thought it was getting in the state operated prisons.
The new law contained provisions to support this objective. These included making private
vendors liable in tort for the care and custody of inmates and liable to the DOC for any
breach of contract4–which would, arguably, create a strong disincentive for private firms to
sacrifice quality of conditions for higher profits. Furthermore, it required that private
contractors employ certified correctional officers and that all other employees of the private
vendor receive, at a minimum, the same quality of training as required of state employees.5
The Department of Corrections did not see the need to call upon the private sector to operate
its facilities. It continued to expand its public facilities in the face of a growing inmate
population. Under pressure, the department finally issued its first request for proposal almost
five years after the legislature passed Chapter 944.105. This first solicitation was
unsuccessful. As mentioned above, the statute required that the contract offer “a substantial
savings to the department, as determined by the department.”6 At the time of appropriating
the funds for the contract (which the Florida law required), the legislature defined how
“substantial savings’ was to be interpreted: at least a 10% savings.7 However, the department
designed the procurement so as to require two separate bids, one for constructing the prison
and a second for operating it, and it determined that each proposal had to meet the 10%
savings requirement.8 None of the bids met the cost savings requirement for both categories.
Damon Smith, a lobbyist for Wackenhut Corrections Corporation, stated that Wackenhut’s
bid for operations offered a savings of approximately 17 percent. However, the firm was
disqualified because its construction bid was only 5 percent to 7 percent cheaper.9 In 1991,
there was another round of bidding. U.S. Corrections Corporation was the only contractor to
submit a proposal, which was accepted. Consequently, in 1994, nine years after Chapter
944.105 passed, Florida’s first privately operated prison—the Gadsden Correctional
Institution—came on line.10

3

Fla. Stat. ch. 944.105(1)(b).

4

Fla. Stat. ch. 944.105(2).

5

Fla. Stat. ch. 944.105(7).

6

Fla. Stat. ch. 944.105(1)(a). (Emphasis added.)

7

Interview with Damon Smith, Partner; Mirabella, Smith & McKinnon, Inc. December 9, 1999.

8

Ibid.

9

Ibid.

10

Interview with James Biddy, Deputy Assistant Secretary, Florida Department of Corrections, April 27, 1999.

68

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

It is fair to say that the legislature, the governor, and the Department of Corrections were not
on the same page when it came to correctional privatization. In this respect, conditions in
Florida differed from those that were found in Texas and Oklahoma (discussed in chapter 6).
In Texas, both the legislative and executive branches were under pressure to act quickly to
avoid heavy fines for failing to comply with the Ruiz court order, and they agreed to contract
for prison operations. In Oklahoma, the goal for the governor and the Department of
Corrections was to expand capacity quickly without incurring the public debt required to
build more prisons, a policy that the legislature concurred with. Even though the federal
courts had put pressure on the State of Florida to remedy conditions in its prisons that were
found unconstitutional,11 the Department of Corrections chose to solve its problems without
having to “go private.” The department was also under considerable pressure to expand
quickly in the face of a mounting tide of prisoners, but correctional administrators felt that
they could manage this on their own, without having to turn to the private corrections
industry.

The Legislature Tries Again
In 1993, frustrated by the unwillingness of the Department of Corrections to engage the
private correctional firms, the Florida legislature passed Chapter 957, the Correctional
Privatization Act.12 Chapter 957 created the Florida Correctional Privatization (CPC) within
the state’s Department of Management Services to expedite privatization. The law authorizes
the CPC to enter “into contracts with contractors for the designing, financing, acquiring,
leasing, constructing, and operating of private correctional facilities.”13 Those sympathetic to
the DOC saw the passing of Chapter 957 as the legislature and special interest groups
shoving privatization down the throats of the DOC. Those sympathetic to privatization saw
the Florida government taking the necessary measures to make sure that its intentions were
implemented.14 Regardless of the politics surrounding the issue, one thing is clear: from
1985 to the eve of the enactment of Chapter 957, there was little agreement between the
legislature and the DOC regarding privatization. Passage of Chapter 957 overcame this
impasse.
Establishing a separate agency to expedite privatization may have solved some of the
legislature’s problems, but it created others. The CPC was not given executive authority to
run the state's prisons, and the courts could not remand convicted offenders directly or

11

Celestino and Costello v. Singletary, U.S. District Court, Middle District of Florida, Jacksonville Division, Case Nos. 72-109-CIVJ14. The parties entered into several settlement agreements on food services, overcrowding and health care.

12

Fla. Stat. ch. 957 (1997).

13

Fla. Stat. ch. 957.03(1).

14

Interview with Damon Smith, December 9, 1999.

Abt Associates Inc.

Florida: Seeking More Cost-Effective Performance

69

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

indirectly to the commission’s custody. The CPC and the DOC were going to have to
coordinate their activities if privatization was to work in Florida.
The Legislature’s Objectives

The new law signaled the legislature’s objectives in creating the commission.
•

Cost savings were considered important: "A contract entered into under this chapter
for the operation of private correctional facilities shall maximize the cost savings of
such facilities...."15 Specifically, a contract with a private imprisonment firm must
“result in a cost savings to the state of at least 7 percent over the public provision of a
similar facility.”16 To establish that target, the law directs the Auditor General to
calculate the inmate per diem costs in a comparable facility in a similar location
operated by the DOC. At the end of each contract term, the legislation requires the
Auditor General to evaluate the costs and benefits of the contract, to examine the
performance of the contractor, and to make a recommendation to the legislature about
whether to continue the contract.17

•

Chapter 957 also emphasizes the goal of reducing recidivism. The law requires that
"work and education programs must be designed to reduce recidivism."18 As part of
its annual review of the commission and the privatization program, the commission is
directed to compare recidivism rates of the private and public facilities. This is the
only legislature in the nation to declare prisoner rehabilitation a mission of a
privatization program.

•

The legislature also identified innovation as a goal of the contracts to be overseen by
the CPC. Chapter 957 states: “In its request for proposals, the CPC shall invite
innovation and shall not require the use of prototype designs of state correctional
facilities specified or designed by or for the department.”19 Moreover, “the
commission may waive any rule, policy, or procedure of the department related to the
operations standards of correctional facilities that are inconsistent with the mission of
the commission to establish cost-effective, privately operated correctional facilities.”20
The wish that the CPC be independent of the DOC’s influences is also evident in

15

Fla. Stat. ch. 957.04(1).

16

Fla. Stat. ch. 957.07.

17

Fla. Stat. ch. 957.11.

18

Fla. Stat. ch. 957.04(1)(f).

19

Fla. Stat. ch. 957.03(4)(b).

20

Fla. Stat. ch. 957.04(1)(e).

70

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

provisions specifying the required qualifications of the commission’s five members.
These members, appointed by the governor, must not be employees of the DOC or the
Department of Juvenile Justice (DJJ), and four must be employed in the private
sector.21 They may not have had any business relationship with the DOC or the DJJ
during the two years prior to appointment, and may not have any such relationship for
2 years following their service as commissioner.22 The same applies to the executive
director of the CPC and consultants.23
Some supporters of the law see innovation as the primary objective.24 Asked in the Abt
Associates survey to list the state’s objectives regarding privatization in order of importance,
CPC administrators ranked innovation as the most important. (The DOC’s administrators
had a different interpretation: they reported that cost reduction was the principal objective,
and the second was "reducing recidivism through innovative rehabilitation programs.")25
Regardless of whether the pursuit of innovation was of foremost or secondary importance in
legislators’ minds, the fact that innovation was considered an important objective of
privatization distinguishes the Florida experience from that in Oklahoma and Texas, as well
as with that elsewhere in the country.
Specifying How the DOC and the CPC and Private Firms Are to Cooperate

Creation of the CPC, and the intention that the CPC would contract with private firms, raised
questions about how a number of essential functions would be allocated among the DOC, the
CPC and private contractors. The CPC is not a state correctional agency and does not have
correctional authority apart from the right to initiate and monitor contracts with private
facilities. Because the CPC does not receive prisoners sent by the courts, the possibility
existed that the CPC could contract with private firms for prisons but not have any prisoners
to put in them. The law was therefore written in such a way to bind the DOC to utilize any
privately operated facility that the CPC brings under contract. It provides that
The department shall transfer and assign prisoners, at a rate to be determined by the
commission, to each private correctional facility opened pursuant to this chapter in an

21

Fla. Stat. ch. 957.03(2).

22

Fla. Stat. ch. 957.03(2).

23

Fla. Stat. ch. 957.03(3)(e).

24

Interview with David McIntyre, Contract Monitor; Florida Correctional Privatization Commission, April 20, 1999.

25

This divergence of opinion between the DOC and the CPC is instructive. The law did not set forth a hierarchy of objectives, so these
assessments depend upon interpretations of what the legislature intended. The DOC saw cost reduction as the principal rationale for
privatization. Absent evidence that contracting would produce substantial savings, it chose not to contract. In contrast, the CPC
gives primacy to inviting innovation, which makes it easier to justify privatization, if only because the outcome (more innovative
services) cannot be judged until some time in the future. Put another way, the pursuit of innovation is a license to experiment and to
try different ways of delivering a service.

Abt Associates Inc.

Florida: Seeking More Cost-Effective Performance

71

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

amount not less than 90 percent or more than 100 percent of the capacity of the
facility pursuant to the contract with the commission.26
Moreover, the law requires that the private firm, the DOC, and the CPC develop and honor a
cooperative agreement that specifies exactly how prisoner transfers are to occur.27 In the
event of bankruptcy or violations that pose a threat to public safety or health, the law gives
the DOC the right to assume temporary control of the facility “with the approval of the
commission.”28
To protect the interests of prisoners and the state, the law also limits a number of
responsibilities that can be delegated to private firms. These include:
•

Making final determinations on the custody classification of an inmate. This remains
a DOC responsibility.29

•

Developing or adopting disciplinary rules or penalties that differ from the disciplinary
rules and penalties that apply to inmates housed in correctional facilities operated by
the department.30

•

Making a final determination on a disciplinary action that affects the liberty of an
inmate.31

•

Making a decision that affects the sentence imposed upon or the time served by an
inmate, including a decision to award, deny or forfeit gain-time.32

•

Making recommendations to the Parole Commission with respect to the denial or
granting of parole, control release, conditional release, or conditional medical
release.33

26

Fla. Stat. ch. 957.08.

27

Fla. Stat. ch. 957.06(2).

28

Fla. Stat. ch. 957.14.

29

Fla. Stat. ch. 957.06(1).

30

Fla. Stat. ch. 957.06(3).

31

Fla. Stat. ch. 957.06(4).

32

Fla. Stat. ch. 957.06(5).

33

Fla. Stat. ch. 957.06(6).

72

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

•

Developing and implementing requirements that inmates engage in any type of work
unless approved by the CPC.34

•

Determining the eligibility for any form of conditional, temporary, or permanent
release from a correctional facility.35

Establishing the Chain of Accountability

The legislature had to define lines of accountability for this new commission and the facilities
it would contract with. Chapter 957 establishes three approaches to assure accountability.
•

The first is that the CPC is to report annually to the legislature and the Auditor
General (and not the DOC) on the “status and effectiveness of the facilities under [the
CPC’s] management.”36 This report must include a comparison of recidivism rates
for inmates of privately operated facilities to those rates of comparable publicly
operated facilities.37

•

Second, the law requires oversight of the contractor by the CPC and third party
experts. The act requires the appointment of a full-time CPC-employed monitor with
office space on site and unlimited access to the facility.38 The contractor must also
obtain and maintain accreditation by the ACA, thereby gaining third party oversight.39

•

Third, the law assigns liability to the contractor. The contractor must also indemnify
the state and the Department of Corrections “against any and all liability ....”40 Thus,
if a court finds that a contractor violated the constitutional rights of an inmate or
committed a tort against an inmate or free citizen, it is accountable and pays for that
violation. This provision makes it difficult for the contractor to avoid paying for its
mistakes, and thereby creates an incentive to perform in accordance with applicable
law and standards.

34

Fla. Stat. ch. 957.06(7).

35

Fla. Stat. ch. 957.06(8).

36

Fla. Stat. ch. 957.03(4)(c).

37

Fla. Stat. ch. 957.03(4)(c).

38

Fla. Stat. ch. 957.03(4)(c).

39

Fla. Stat. ch. 957.04(c).

40

Fla. Stat. ch. 957.04(1)(b).

Abt Associates Inc.

Florida: Seeking More Cost-Effective Performance

73

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

Translating Policy Objectives into Action in the RFP and the
Contract
In accord with the legislature’s intent, the commission designs its RFPs to obtain bids from
firms that: (1) are well qualified to deliver the service based on their prior experience, and
which offer a capable management team; (2) offer plans for effective educational, training
and substance abuse programs; and (3) offer the state an opportunity to reduce expenditures
for imprisonment. How the commission scores proposals indicates the relative importance of
these three aspects of each bidder’s offer. The qualifications and capacity of the offered,
taken together, are the most important, and are worth 30 percent of the maximum number of
points that can be earned. The design and implementation plan for education, training, and
substance abuse programs is highly ranked, but slightly less important than overall capacity
and qualifications–worth a maximum of 27 percent of all points. The offer’s plan for
reducing costs is of third-most importance: 25 points. The remaining 18 points can be
awarded based on the offer’s plans for complying with ACA requirements, proposed staffing
pattern, compliance with health care standards established by the federal courts and the
department, and plans for interacting with the DOC (the least important part of the plan,
worth only 1 point).
What of the legislature’s interest in “innovation”? This objective is difficult to implement in
some respects. That is, asking for innovative prisons is something of a hollow call on its
own. The CPC’s approach, consistent with the legislature’s intent, is to communicate that
proposers have a good deal of freedom to design their plan to deliver the service, and that
they are not bound to follow the procedures prevailing in the state’s publicly operated
facilities. Moreover, the RFP requests that the bidders take advantage of this freedom and
design a regimen that will improve offenders’ chances of doing well and conforming to the
law when they leave prison. Bidders are not explicitly evaluated according to the extent to
which their plans are innovative, however.
Lower Costs

Even though proposed cost savings are awarded fewer points in the course of evaluating
proposals, bidders cannot even get to the table if they do not propose to provide the minimum
savings of seven percent. This suggests that reducing expenditures is the most important
objective. Once this objective is achieved (or, at least, the promise of the savings is offered),
other factors are considered in choosing one bidder over others. Among those bids that meet
the threshold seven percent savings, one offering a more capable organization and a superior
plan for programming will be chosen over a lower-priced bid that came from a less qualified
and experienced company offering an inferior plan for rehabilitation programs.
The CPC’s approach to implementing this legislative objective is to tell bidders what the cost
of state operations would be and, by extension, what the bid price would have to be in order
74

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

to produce the minimum necessary savings to the state. As mandated in Chapter 957, an
independent party, the State’s Auditor General, determines what the DOC would spend for an
additional prison, both for construction and for operations, separately, and the seven percent
savings target is calculated from these amounts. Although determination of actual costs to
the state of public operation is not a straightforward matter (see the discussion at the end of
this chapter), the commission establishes in the RFP some assumptions to be adopted by the
bidders.
Scoring Bidders’ Cost Proposals
1

Operating cost evaluation

12%

2

Construction cost evaluation

8%

3

Total cost evaluation

3%

4

Financing plan

2%

Total

25%

First, the request separates out the accounting of proposed construction and operating costs.
Unlike the DOC’s first RFP, the offer does not have to meet the cost savings target for each.
Rather, the savings resulting from both construction and ongoing operations, taken together,
has to meet the target.41 If the total cost exceeds this figure, the CPC rejects the proposal; if
the proposal meets the figure, or is lower, the proposal receives three points.42
Estimates of imprisonment costs per inmate/day are dependent not only on what either a
private or public entity spends to deliver the service but also on the numbers of prisoners that
are being served. Other things being equal, per diem costs for an underutilized facility will
be higher than for a facility filled to capacity. This is because a facility that is three-quarters
full requires the same staffing levels as one in which an inmate occupies every bed (and
spending for staff constitutes the largest share by far of all operating costs in prisons).
Because a private facility cannot control the extent to which its available capacity is to be
used by the state, the commission asks bidders to estimate its costs assuming a 95 percent
occupancy rate.
Costs of construction, expressed on an annualized or per diem basis, also depend upon
assumptions about the length of time over which these costs are to amortized. Spreading a
cost incurred in one year over a 20 year period will result in an amortized per diem cost that

41

"State of Florida Correctional Privatization CPC Request for Proposal for the Designing, Financing, Acquiring, Leasing,
Constructing, and Operating of one 750 Bed, Medium Custody, Secure Correctional Facility for Adult Males," addendum 1 (1995).
Answer to question 45 (1995).

42

State of Florida Correctional Privatization CPC Request for Proposal for the Designing, Financing, Acquiring, Leasing, Constructing,
and Operating of one 750 Bed, Medium Custody, Secure Correctional Facility for Adult Males, p. 44 (1993).

Abt Associates Inc.

Florida: Seeking More Cost-Effective Performance

75

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

is 35 percent lower than if one assumes the cost will be spread over a ten-year period. Firms
may seek to capture their capital investments in shorter or longer periods. They may also
make different assumptions about how much they hope to realize from a sale of the property
at the end of that period. This may affect calculations they make about how much they will
seek to recover from per diem or per annum “rental charges” to the state. So that all bidders
adopt common assumptions about recovery of capital investments, the commission requires
that offerers assume that equipment expenditures are amortized over a ten year period, with
other capital investments amortized over 20 years.
The RFP instructs the proposers to provide first, second and third year per diem bids to
account for inflation. Contractors must give per diem bids on inmates up to 90 percent
occupancy and the rates charged for inmates above the 90 percent occupancy mark.43 The
second and third year rates, however, are maximums, and the CPC reserves the right to
negotiate these rates downward so as to not exceed the system-wide major facility per
prisoner operating costs of the DOC.44
Obtaining Capable Partners

The commission’s interest in obtaining qualified and established contractors is evident in
how bids will be evaluated (see the chart below). The seven sub-factors fall into to two main
categories: (1) organizational ability (factors 1 thru 4 ), worth 19 of the 30 points, and (2)
financial resources and backing, worth another 8 points. The bidders’ experience offering
equal opportunity employment is worth three points.
Scoring Bidders’ Qualifications
1

Experience in operation of correctional facilities.

5%

2

Organizational size, structure and history.

5%

3

Experience of all facility administrators and assistants.

5%

4

Proof of ability to comply with applicable court orders and ACA
standards.

4%

5

Financial strength and proof of insurance.

4%

6

Financial statements.

4%

7

Equal opportunity.

3%

Total

30%

43

Ibid., p. 1. These two rates vary substantially to reflect savings due to economies of scale. For example, in a contract drafted in 1998,
the 90 percent occupancy rate for the first year was $50.90 per diem, and the rate for each inmate above the 90 percent occupancy
mark was $14.55 per day.

44

Ibid., p. 39.

76

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

Effective Programming

In the Bay Correctional Facility RFP, the CPC states that “programming, designed to reduce
recidivism, is to be a vital operational aspect of these facilities....”45 This proposal requires
that bidders provide education, training and work programs and that
programs must be designed to reduce recidivism. The vocational education programs
offered by the proposer shall include an analysis concerning the career outlook for the
job training being offered. This analysis shall include, as a minimum, specific
information including: type of jobs the vocational training will prepare the inmates
for; the estimated job growth, in the State of Florida, in the fields of training being
offered; the salary range of the jobs available; and the qualifications necessary for the
jobs.46
The RFP sets a per diem amount to be spent for such programs, which has the effect of
assuring the CPC that programs will be provided at a certain level and that bidders will not
sacrifice programming for the sake of developing a more competitive price. In the Bay
Correctional Facility RFP the per diem amount was $2.28 for educational programs and
$2.67 for substance abuse programs. These amounts are mandatory minimum amounts,
however. In the spirit of inviting innovation, the CPC allowed proposers to submit an
education and substance abuse program that would cost more than the per diem amounts
specified. Upon review by the CPC, if the legislature is interested in appropriating additional
funds for the additional programs, the CPC will accept the alternative program proposal.47
The CPC gives further support to this mission by permitting private prison administrators to
request transfer of prisoners who are “considered by the on-site Chief Health Officer to be
medically, physically or mentally incapable of participating in the programmatic activities
(which have been specifically designed to reduce recidivism) for greater than two weeks.”48
The relative importance of well-designed training, educational, and substance abuse
programming in the proposers’ plans is made clear by the evaluation criteria that are specified
in the RFP. The design and implementation plan for educational and substance abuse
programs can earn the bidder up to 15 points, or 15 percent of the total maximum allowable
score. This is the single most important factor in this particular proposal (although certain
other factors, added together–such as various aspects relating to organizational
capacity–count for more, as discussed above).

45

Ibid., p. 11.

46

Ibid., p. 1.

47

State of Florida Correctional Privatization CPC Request for Proposal., p. 31-32.

48

Ibid.

Abt Associates Inc.

Florida: Seeking More Cost-Effective Performance

77

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

Scoring Bidders’ Proposed Rehabilitative Programs, Design, and Project Schedule
1

Education and substance abuse

15%

2

Minority Participation

5%

3

Description of proposed facility

3%

4

Project schedule

3%

5

Ability to proceed (steps taken with the community)

1%

Total

27%

Writing the Contract

The contracts written by the CPC contain a number of provisions to allocate risks and
responsibilities in ways designed to protect both the state and the contractor. These support
the state’s objective in receiving good performance at the agreed-upon price. Illustrative of
these design features are the provisions in the contract, signed in 1998, for managing and
operating the Bay Correctional Facility.
In this contract, the contractor makes eight representations. All of these representations, for
the most part, are a product of common sense. Examples are authorization to execute the
contract, compliance with necessary laws, and delivery of financial statements. However, the
“no litigation” representation required of the contractor is of particular interest. Other than
that disclosed to the CPC before the drafting of the contract, the contractor represents that
there is no current litigation, pending litigation or labor dispute with the contractor’s
employees that might adversely change the contractor’s ability to perform its obligations.49
The contractor must also represent that there is no action, suit, or proceeding related to
environmental or civil rights matters.50 These provisions serve two purposes. First, they give
the CPC an “out” if the contractor has lawsuits, or potential lawsuits, that could materially
affect the firm’s ability to provide the agreed upon services to the CPC. Second, the
provision addressing environmental and civil rights matters operates as an incentive for
contractors to run tight ships with respect to their other contracts with other states.
The contract also enumerates certain obligations of the contractor with respect to alterations
or improvements as well as recordkeeping. Subject to prior approval of the CPC, the
contractor may remodel the facility or make substitutions, alterations, additions,
modifications and improvements to the facility from time to time provided that the contractor

49

Bay Correctional Facility Operations & Management Contract, p. 10-11 (1998).

50

Ibid., p. 11.

78

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

does not use lesser quality.51 (The contractor, however, may make minor improvements
without prior approval.52) The burden of cost and quality is on the contractor.53 The
contractor also requires the contractor to maintain a computerized and manual records and
reporting system,54 and requires the contractor to perform self-monitoring.55
Healthcare for inmates is an expensive component of correctional operations, and assuming
responsibility for that care poses some financial risk to contractors. Those risks are
especially pronounced for prisoners because of their often-extensive histories of drug and
alcohol abuse. However, the contracts have negotiated provisions to manage these financial
risks. For example, the number of HIV affected inmates at the facility shall not exceed 3.4
percent of the facility’s design capacity.56 Contractors are also able to limit their
responsibilities to the first $7,500 or the first five days of an inmate’s hospitalization
expense—whichever comes first. In each subsequent year, the monetary portion of the cap
rises three percent to accommodate inflationary pressures.57
The contract places restrictions on the contractor’s use of subcontractors and employees. No
subcontract over $150k in payments by contractor within a 12 month period may be entered
into unless the commission reviews and approves it.58 Arrangements with an affiliate or
member company of the contractor are subject to this provision as well.59 With respect to
employees, the contractor must conduct criminal and employment history check, drug test.
Criminal history checks are performed by the state and the FBI.60 Furthermore, part time
hours are limited to 450 hours per week.61
As mentioned above, Chapter 957 requires that private facilities run at least at 90 percent of
capacity and that bidders must submit two per diem rates—one for a facility filled to 90
percent or less and another for more than 90 percent. However, the contract provides that,

51

Ibid., p. 16.

52

Ibid., p. 17.

53

Ibid., p. 17.

54

Ibid., , p. 21.

55

Ibid., p. 64.

56

Ibid., p. 19.

57

Ibid., p. 21.

58

Ibid., p. 36.

59

Ibid., p. 37.

60

Ibid., p. 40.

61

Ibid., p. 38-39.

Abt Associates Inc.

Florida: Seeking More Cost-Effective Performance

79

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

regardless of the number of inmates incarcerated at the facility, a guaranteed amount paid to
the contractor equal to 90 percent occupancy times the 90 percent per diem rate subject to
legislative appropriations.62
This contract establishes several options to protect the state’s interest in the event of nonperformance: 1) reduce the claim to judgment; 2) take action to cure the breach of contract;
3) assessment of liquidated damages—not to exceed $500 per day during the breach; and 4)
termination and removal of the contractor.63 The existence of these different options allows
the CPS to match the remedy to the severity of the breach.
Florida also places itself in a favorable position in the case of a transition from one contractor
to another. The contract requires the contractor to work with the commission and/or the
DOC for 60 days for an efficient transition in management to either of those parties or to a
third party.64

Determining If Objectives Were Achieved
The legislature sought privatization in the hopes that certain objectives would be achieved–to
reduce expenditures for prisons, to produce more law-abiding citizens following their release,
and to introduce innovations into the state’s prison system. To assess the effectiveness of the
privatization program that would be implemented following the passage of the law, it
required a systematic assessment of the program’s effectiveness, at least with respect to
savings and recidivism.
Were Savings Achieved?

To determine if contracting accomplished the goals set by the legislature with respect to
spending, Florida law mandates that an independent oversight agency, the Auditor General’s
Office of Program Policy Analysis and Government Accountability (OPPAGA), review the
actual performance of all privately operated correctional facilities under contract with either
the CPC or the DOC. In 1998, this agency released a report comparing costs at two privately
operated facilities–the Bay and Moore Haven Correctional Facilities, operated by CCA and
Wackenhut, respectively–with a similarly sized public facility, the Lawtey Correctional
Institution. It concluded that the Moore Haven facility was more expensive than Lawtey to
operate, and that the Bay facility was marginally less expensive than Lawtey–0.2 percent.
Both therefore failed to produce the minimum seven percent savings required by law.

62

Ibid., p. 42.

63

Bay Correctional Facility Contract, p. 61.

64

Bay Correctional Facility Contract, p. 67.

80

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

The CPC and the two management firms, CCA and Wackenhut, took exception to several
accounting assumptions that were made by OPPAGA analysts and produced their own
analyses, which showed that these two privately operated facilities saved the state money.
Moreover, in yet another analysis, the DOC found that both facilities were more expensive
than state-run facilities.65
Some of the differences in findings stem from the choice of comparison facilities. The DOC
did not chose a single publicly run facility for the comparison but rather averaged the cost of
nine other public facilities. This biased the case against the private facilities because all but
one of these facilities were larger than both of the privately operated ones. (Larger prisons
are generally less costly to operate because of economies related to scale.) The OPPAGA
study made a more reasonable decision, electing to compare the two privates to Lawtey,
which was about the same size and of the similar security level.
The studies also differed in how certain categories of costs were treated. These categories
included costs associated with monitoring, taxes paid and not paid to the state, credits for
revenues from inmate medical co-payments retained by the private firms, and retirement fund
surcharges. The comparability of medical expenditures was also disputed. The CPC and the
contractors argued that medical spending at the two private facilities and at Lawtey were not
comparable because the prisoners at Lawtey had significantly different medical needs (an
assertion rejected by OPPAGA).
This indicates how the seemingly simple comparison of costs to the state of either direct
operation or of contracting is not a straightforward matter. What matters here, however, is
that the legislature established a system for monitoring the performance of its privatization
program that included an assessment of whether it hit its intended targets. (For a systematic
comparison of public and private imprisonment, and a discussion of cost accounting issues
involved, see Douglas McDonald’s forthcoming comparison of federal prisons and the Taft
Correctional Institution, operated under contract to the Federal Bureau of Prisons.)
Did Privatization Result in Less Criminality?

As required by law, the CPC commissioned a study to determine if privately operated
facilities produced more law abiding persons than the publicly operated prisons. LanceKaduce and Parker conducted the study, released in 1998, which compared the recidivism
rates of inmates released from prisons operated by the Florida Department of Corrections to
those inmates released from one prison operated by the Corrections Corporation of

65

These estimates are reproduced and analyzed in Florida Department of Corrections, Privatization in the Florida Department of
Corrections (1998). Available at web site (time stamped April 28) www.dc.state.fl.us/administrative/reports/privatize/index.html.

Abt Associates Inc.

Florida: Seeking More Cost-Effective Performance

81

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

America–the Bay Correctional Facility—and another operated by Wackenhut Corrections
Corporation, the Moore Haven Correctional Facility.66
The study concluded that a smaller proportion (17 percent) of the sample of releases from the
privately operated facilities had recidivated within twelve months of their release, compared
to a sample of releases from public facilities (24 percent). Recidivism was measured
variously, including returns to prisons for new arrests and technical violations of parole. A
scale was developed using these various indicators of recidivism to measure seriousness of
recidivism. The study found that public sector releases were more likely to commit
drug/weapon possession offenses, property offenses, and violent offenses. The mean level of
seriousness on this 0 to 5 point scale was 3.43 for the public sector releases and 2.32 for the
private sector releases.
Studies of the rehabilitative effects of different prisons, or different prison programs, nearly
always face significant difficulties that frustrate one’s ability to draw strong conclusions from
the comparisons.67 Unless prisoners are assigned randomly to the various prisons under
study, it is difficult to sort out the possible effects attributable to differences in prisons or
prison programs from differences in the groups of prisoners that are created by purposive
selection. For example, as discussed above, the RFP for at least one of the private facilities
(Bay Correctional Facility) permits private prison administrators to request that prisoners be
transferred to public facilities if they are “considered by the chief health officer to be
medically, physically or mentally incapable of participating in the programmatic activities
(which have been specifically designed to reduce recidivism) for greater than two weeks.” If
the facility administrators exercised this right, it would result in collecting at this facility a
group having stronger odds of success, regardless of what programming they ultimately
received.
In addition, there were systematic differences in the types of prisoners held in the private and
public institutions, which may have affected the differences in post-release criminality.
Moreover, 35 percent of the prisoners in the private prison sample spent significant periods
of time in the public correctional facilities prior to being transferred to the private facilities.
These prisoners were thus exposed to both public and private facilities. And finally, there are
significant questions about how the study subjects were selected from among all prisoners
released from these facilities for the purpose of matching.
In short: that Florida seeks to conduct a serious evaluation of whether one of privatization’s
most important objectives is being achieved is laudable. Absent a stronger methodology,

66

L. Lanza-Kaduce and K.F. Parker, “A Comparative Recidivism Analysis of Releasees from Public and Private Prisons in Florida”
(Gainsville, FL: Private Corrections Project, Center for Studies in Criminology and Law, 1998).

67

For a review and criticism of this study, see Gerald G. Gaes, Scott D. Camp, and William G. Saylor, “The Performance of Privately
Operated Prisons: A Review of Research,” in Douglas McDonald , Private Prisons in the United States, Appendix 2.

82

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

however, the study does not provide indisputable evidence that privately operated prisons are
doing any better or any worse than the public prisons.
Have the Private Firms Introduced Innovations into the State’s Prison System?

There has been no formal assessment of this. The legislature may have assumed that any
innovation of significance would result in more effective prisons (i.e., lower recidivism) and
less costly ones. The existing studies of comparative costs and of comparative recidivism
rates to not support any strong inferences about the state’s obtaining more innovative
imprisonment from the private sector.

Abt Associates Inc.

Florida: Seeking More Cost-Effective Performance

83

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

84

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

6. Oklahoma: Managing the Risks of Dependence
As in several other states, Oklahoma turned to privately operated prisons to relieve
overcrowding in its state facilities. Oklahoma’s practice is distinctive, however, because it
sought to rely entirely on private facilities for the additional beds that it needed. It contracted
with out of state facilities and, later, with firms that built prisons “on spec” within the state.
As a result, by the late 1990s, it led the nation in the proportion of state prisoners in privately
operated facilities. Unlike several other states, it also relies more heavily upon privately
owned facilities. Although the state’s policy makers may not have intended to go further than
other governments in developing a hybrid public/private correctional system, various
incremental policy decisions had this result. Oklahoma’s management of this hybrid system
is of interest, consequently, because the state developed a number of strategies to maximize
the benefits of relying heavily upon the private sector while minimizing the risks of
dependence.
This chapter examines how and why Oklahoma developed its distinctive practice of
contracting for private imprisonment. It also discusses some of the consequences of
Oklahoma’s heavy reliance upon privately operated facilities, the risks of dependence, and
the particular approaches it has taken to designing, procuring, and administering private
imprisonment services to manage this risk.

Too Many Prisoners and Not Enough Money
Oklahoma, like many other states, experienced a tremendous increase in the demand for
prison cells during the 1980s and 1990s. Between 1978 and 1988, for example, the
population in its state prisons grew from 4,200 to 8,900.1 During the next ten years, it rose
another 5,900, to 14,800.2 Not only did the number of prisoners behind bars grow during this
period, but the burden of imprisonment fell heavily on Oklahoma’s taxpayers because judges
there rely on imprisonment more than in most other states. As the chief administrator of
Oklahoma’s privatization program says, “Incarceration rates for Oklahoma men and women
are consistently among the highest in the world.”3 For example, as of June 30, 1999, the state
had an incarceration rate of 653 per 100,000 citizens, compared to 468 for the nation as a

1

Bureau of Justice Statistics, Prisoners in custody of State and Federal Correctional Authorities (corpop10.wk1 spreadsheet), at
www.ojp.usdoj.gov/bjs.

2

Bureau of Justice Statistics, Prisoners in custody of State and Federal Correctional Authorities (corpop05.wk1 spreadsheet), at
www.ojp.usdoj.gov/bjs; also Prisoners in Custody, 1977-1998.

3

Dennis Cunningham, The History of Private Prison Contracting in the Oklahoma Department of Corrections,
http://www.doc.state.ok.us/docs/history_of_private_prison_contracrts.htm.

Abt Associates Inc.

Oklahoma: Managing the Risks of Dependence

85

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

whole.4 This was not a new development, as the state’s incarceration rate has been among
the highest in the country for some time.5
The capacity of the state’s prisons failed to keep pace with the growing numbers of inmates
and became overcrowded. The state’s fortunes were tied closely to oil, and the oil revenues
collapsed during the 1980s.6 Oklahomans were averse to their government spending more
money for prisons, which made it difficult for the state to build its way out of the crunch.
Choosing not to build expensive prisons and take on more government employees, the state’s
lawmakers chose instead to relieve the demand for prison beds. In 1984 and again in 1993,
the legislature passed laws that placed caps on the numbers of prisoners that the state’s
system could hold.7 These laws required the Department of Corrections to issue a request to
the governor to declare a state of emergency whenever the inmate population exceeded 95
percent of capacity for thirty consecutive days.8 Following the governor’s declaration of
such, the DOC’s director was given authority to grant emergency time credit to inmates
incarcerated for nonviolent offenses and those classified as medium security or lower.9 These
credits were awarded in sixty-day blocks, up to 360 days, which made more inmates eligible
for immediate release. These emergency release powers could be used until the prison
population no longer exceeded 95 percent of capacity.10 The state also implemented
diversion programs and other early release programs—involving privately-operated halfway
houses and community placement programs—to reduce demand. These strategies worked to
relieve population pressures to some extent, but not entirely. During the decade that was
bracketed by passage of the two emergency powers laws, the state’s prison population
doubled—from 7,900 in 1984 to 16,400 in 1993.11 And the prisons were still
overcrowded—inmate population in 1995 was in excess of 120 percent of the Board of
Corrections-approved capacity.12 Early release was not enough.

4

Allen Beck and Christopher Mumola, Prisoners in 1998, Bureau of Justice Statistics, August 1999, table 3.

5

Between 1993 and 1998, the state ranked either third or fourth among U.S. jurisdictions in terms of rates of incarceration of adult
inmates and has surpassed the national average by at least 130 inmates per 100,000 citizens during each of these years. A decade
earlier, in 1983, it had the tenth highest incarceration rate. Bureau lot Justice Statistics, “Prisoners at Midyear 1983,” (Washington,
DC: October 1983), Table 2.

6

Interview with Cal Hobson, Oklahoma State Senator (District 16), December 7, 1998.

7

Ibid.

8

Okl. Stat. Ann. tit. 57, sect. 572 (A) (West 1997).

9

Ibid.

10

Ibid.

11

Bureau Justice Statistics, “Prisonsers in 1984” (Washington, DC: 1985), Table 2; Darrell K. Gilliard and Allen Beck, “Prisoners in
1993,” (Wasjomgtpm,

12

Cunningham, The History of Private Prison Contracting. in the Oklahoma Department of Corrections.

86

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

Public support of these programs had also grown thin by the mid-1990s. Some beneficiaries
of these early-out programs committed crimes—including murders.13 An appetite was
growing among citizens and their political representatives for solutions to Oklahoma’s
overcrowding problems that would eliminate the need for early release.14 Emergency releases
were finally stopped in August 1996.

Finding Rather than Building Prison Beds
In the mid-1990s, the state shifted course. Instead of releasing convicted criminals to free up
prison beds, the department began to search for facilities to hold them. By the mid-1990s, the
state’s economy was improving and a prison construction campaign could have been afforded
more easily than a decade earlier. But Governor Frank Keating thought that year-to-year
increases in demand for prison cells might be a short-run problem that could be
accommodated by other means. (The run-up in prisoners was not driven primarily by parallel
increases in numbers of serious crimes committed in the state, but by changes in sentencing
policy and judicial sentencing practices, neither of which were immutable.) The governor
therefore wanted to avoid spending money to build prisons that might not be needed in the
future.15 Prison beds would have to be found elsewhere if the state did not build them.
Other prison systems could not offer much help. Although some states had made excess
capacity available to other states in earlier years, unused capacity generally dried up as prison
populations grew each year throughout the nation. Between 1990 and 1995, for example, the
number of prisoners in state and federal prisons increased 45 percent, from 743,000 to 1.1
million.16 The state therefore began contracting in 1994 with county jails to house prisoners.
Although these jails were of sufficient quality to house the state’s inmates, this solution had
limited impact because many counties had overcrowding problems of their own.17
The state then turned to the private prison industry. By 1994, departmental administrators
felt that the only way to comply with capacity rates without acquiring more prisons was to
contract with privately operated facilities.18 As discussed elsewhere in this report, private
entrepreneurs in Texas had been building a number of privately owned and operated

13

Ibid.

14

Ibid.

15

Ibid.

16

Bureau of Justice Statistics, corpop05.wk1, at www.ojp.usdoj.gov/bjs. These numbers refer to populations at year-end.

17

Cunningham, Dennis. The History of Private Prison Contracting. in the Oklahoma Department of Corrections.

18

Cunningham, Dennis. The History of Private Prison Contracting in the Oklahoma Department of Corrections, Internet:
http://www.doc.state.ok.us/docs/history_of_private_prison_contracrts.htm.

Abt Associates Inc.

Oklahoma: Managing the Risks of Dependence

87

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

correctional facilities that they advertised to any local, state, or federal government that
needed prison beds.
On November 1, 1995, the department issued its first RFP to house Oklahoma prisoners in
privately operated prisons. The procurement process was completed within a month, and an
agreement was written and executed two weeks later. By December 15, 1995, Oklahoma had
acquired 560 medium security beds in four privately-operated facilities in Texas: the Odessa
Detention Center operated by Gil R. Walker Corporation; the Limestone County Detention
Center operated by Capitol Correctional Resources, Inc.; the Mansfield Law Enforcement
Center operated by the City of Mansfield; and the Central Texas Parole Violator facility
operated by Wackenhut Corrections Corporation.19
For such swift procurement, the department used its authority to develop intergovernmental
agency agreements rather than negotiating direct contracts with the private imprisonment
firms. Many of these private firms had negotiated standing contracts with a correctional
authority in the county in which they were located—typically, the local sheriff’s office.
Oklahoma and other governments could use their authority to enter into intergovernmental
agreements to transfer custody over prisoners to these local sheriffs, who immediately
transferred custody to the private firm under contract.

A Private Prisons Industry Emerges in Oklahoma
Seeing business go to Texas gave some Oklahomans the idea of getting a piece of this market
for themselves and their towns. Jack Barrett, the mayor of Holdenville, was one such person.
The town of Holdenville, located ninety miles southeast of Oklahoma City and even closer to
the state’s main penitentiary at McAlester, had suffered a severe economic decline as a result
of the 1980s oil bust. During that decade, Holdenville lost 16 percent of its population.20
Barrett observed that the towns which had survived the oil bust were those with either a
college or a prison. To see if Holdenville could get into the prison business, he went to the
Department of Corrections to sell them on the idea of putting prisoners in a facility that the
town would build. The department had the authority to contract directly with private firms,
as laws authorizing this had been on the books since 1980. (This authority had been used to
develop agreements with half-way houses and other small community-based operators in
Oklahoma as part of the earlier strategy to get prisoners out of overcrowded facilities.) DOC
officials made no guarantees to Barrett, however. First, it is contrary to the state’s
constitution to guarantee obligations lasting over one year. Second, there was serious doubt

19

Ibid.

20

Christopher Swope, “The Inmate”, Governing, p. 18, 21 (October 1998).

88

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

as to whether a town the size of Holdenville could complete a $34 million project—the cost
of building the proposed prison.21
Although the Department of Corrections made its doubts known, it did not tell Barrett that it
would not place prisoners in his proposed facility. He took this as a positive sign. According
to Barrett, he knew that although the Department did not confirm they needed the space, it
was apparent that they needed 4,000 beds.22 The state had already exported prisoners to
Texas, so they could either put prisoners in Holdenville or continue paying Oklahoma dollars
to Texas facilities.
Barrett thought that the team could hedge its bet: if he could not get Oklahoma inmates, he
could certainly get some of the 1,700,000 inmates being held throughout the United States.23
Indeed, he knew that the town of Hinton, Oklahoma had built a prison a few years earlier and
filled it with North Carolina inmates. So he decided to press ahead. Getting financing
proved to be a challenge until he formed a partnership with the Corrections Corporation of
America and proposed that CCA manage the facility.24 Barrett then employed the same
design and construction companies that had designed the DOC’s last prison and asked them
to build one like it in Holdenville.25 A private entity was created, the Holdenville
Correctional Authority, which raised the capital and built the Davis Correctional Facility.
Barrett’s bet was a good one. In 1996, the department created a special unit, the Private
Prisons Administration, to manage its privatization program. A request for proposals was
issued in January of 1996, which resulted in contracts with two Oklahoma firms, the
Holdenville Correctional Authority for use of its Davis Correctional Facility, and the Hinton
Economic Development Authority, owner of the Great Plains Correctional Facility. These
contracts were for 250 and 960 medium security beds for men, respectively.
In 1997, the department continued to pursue both in-state and out-of-state contracting
strategies. A third RFP issued in February of 1997 resulted in three contracts—one in
Oklahoma and two in Texas. The Department awarded this round of contracts to facilities
operated by CCA and the Bobby Ross Group. By April 15, 1997, the Department of
Corrections held 2,708 inmates in eight different secure private facilities. Just over 55
percent of these inmates were housed in six out of state facilities.26 Table 6.1 lists the

21

Ibid., p. 18.

22

Ibid., p. 18.

23

Ibid., p. 18.

24

Ibid., p. 21.

25

Ibid., p. 22.

26

Cunningham, The History of Private Prison Contracting.

Abt Associates Inc.

Oklahoma: Managing the Risks of Dependence

89

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

facilities, location, population and operator of the privately operated facilities the state had
contracts with as of April 15, 1997.

Table 6.1
Private Facilities Contracting with the State Department of Corrections in April, 1997
Facility

Prisoners

Location

Operator

Central Texas Parole
Violator Facility

Texas

Limestone County Detention
Center

Texas

560 (men)

Capital Correctional Resources, Inc.

Mansfield Law Enforcement
Center

Texas

215 (men)

Mansfield Public Finance Authority

Odessa Detention Center

Texas

95 (women)

Crystal City Correctional
Center

Texas

224 (women) Bobby Ross Group, Inc.

Newton County Correctional
Center

Texas

160 (men)

Bobby Ross Group, Inc.

Davis Correctional Facility

Oklahoma

960 (men)

CCA

Great Plains Correctional
Facility

Oklahoma

250 (men)

CCA

Source:

140 (men)
Wackenhut Corrections Corporation
104 (women)

Gil R. Walker, Inc.

Cunningham, Dennis. The History of Private Prison Contracting in the Oklahoma Department of Corrections, Internet:
http://www.doc.state.ok.us/docs/history_of_private_prison_contracrts.htm.

In 1997, the Oklahoma legislature passed comprehensive amendments to the state’s enabling
statute. These amendments laid out a comprehensive process for procuring private
correctional services and also served to regulate the relationship between the Oklahoma
Department of Corrections and the private prison industry.27 In June of 1997, the department
became concerned about the treatment of its inmates at the Limestone Facility, and decided to
move them to the Great Plains Facility in Hinton, Oklahoma. However, the department
continued to send inmates to Texas. It executed another IGA agreement on January 29, 1998
with Karnes County, Texas to house male inmates in the Karnes County Correctional
Center.28 Wackenhut operated this facility. This was the last out of state contract executed
by Oklahoma. As the year passed, more problems surfaced at the Texas facilities. By

27

A more comprehensive description of these amendments appears below.

28

Cunningham, The History of Private Prison Contracting.

90

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

October of 1998, the Department had removed all of its inmates from Texas and placed them
in private prisons in Oklahoma.29
The distance of the out-of-state facilities presented major challenges to contract monitors.
Monitors would visit these facilities monthly and stay one to three days.30 These visits were
supplemented with frequent written and verbal reports. However, the department found that
the quality of the services of these facilities was substandard.31 In April of 1998, Oklahoma
held contracts with five out-of-state facilities. One facility had three fires within a six-week
period in the restricted housing unit. Other problems included insufficient inmate clothing
issue; excessive inmate disciplinary reports; insufficient medical and dental staff; and
inmates working outside the fence and in control of dangerous weapons (chainsaws).32
To accommodate the return of Oklahoma inmates from Texas, the department contracted for
562 additional beds with the Great Plains Correctional Facility, operated by CCA. It also
contracted for 700 male beds at Northfork Correctional Center; 1,200 male beds at Lawton
Correctional Facility; and 550 female beds at the Central Oklahoma Facility in McCloud.33
By December of 1998, all of the inmates at Northfork were transferred to Lawton and the
contract with the Central Oklahoma Correctional Facility was expanded by another 100
beds.34 On July 30, 1999, the department signed a contract with CCA to house 250 inmates
at the Diamondback Correctional Facility in Watonga, Oklahoma.35
By mid-May, 2000, Oklahoma had 6,204 beds under contract, and 93 percent of those beds
are filled. It was contracting with seven facilities—all of which are located in Oklahoma
(Table 6.2). Prisoners in these privately operated facilities constituted about 28 percent of
the total population of state prisoners. This ratio has stayed about the same since then, at
least through mid-2003.36
In short: Oklahoma turned to privatization because the governor and his staff believed that it
would be more cost efficient than building new prisons and operating them into the distant
future. Rather than sink capital into expensive facilities that have long lives and no

29

Cunningham, The History of Private Prison Contracting.

30

Abt Associates survey of private correctional facilities. December 31, 1997.

31

Ibid.

32

Interview with Dennis Cunningham, April 2, 1998.

33

Ibid.

34

Ibid.

35

Ibid.

36

Oklahoma Department of Corrections. Facts at a Glance. May 30,2003.

Abt Associates Inc.

Oklahoma: Managing the Risks of Dependence

91

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

alternative uses, it would be smarter public policy, the governor believed, to rely on shortterm agreements with contractors. When (or if) the demand for prison beds diminished,
contracts could simply not be renewed when they ended. Even better for the state, Oklahoma
wrote into its contracts the right to terminate in the event that its inmate population decreases.

92

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

Table 6.2
Private Facilities Contracting with the State Department of Corrections on November 13, 2000
Security
Level

Contract
Capacity

Prisoners

Cimarron Correctional
Facility

medium

960 (men)

849

owned and
operated by CCA

Davis Correctional Facility

medium

960 (men)

964

owned and
operated by CCA

Great Plains Correctional
Facility

medium

802 (men)

807

was operated by
CCA, now it is
owned by Hinton
Development
Authority and
operated by Cornell
Corrections
Corporation.

Great Plains Correctional
Facility

Minimum

10

0

Central Oklahoma
Correctional Facility

medium

620 (women)

574

owned by McCloud
Correctional
Services, LLC,
operated by
Correctional
Services
Corporation.

Diamond Back Correctional
Facility

medium

950 (men)

960

owned and
operated by CCA

Diamond Back Correctional
Facility

minimum

10

0

Lawton Correctional Facility

medium

1,872 (men)

1,883

owned and
operated by
Wackenhut.

Lawton Correctional Facility

minimum

20 (men)

0

owned and
operated by
Wackenhut.

Facility

Source:

Owner/Contractor

same as above

same as above

Internet: http://www.doc.state.ok.us/DOCS/privatep.htm. As of November 13, 2000.

Abt Associates Inc.

Oklahoma: Managing the Risks of Dependence

93

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

Costs
The governor believed that reliance upon privately operated facilities would save money over
the long term, but in the short term, the state paid contractors on a per prisoner basis more
than it was costing the state to house them in public facilities. At the end of 1997, the state
was spending between $39.00 and $45.95 per prisoner/day to house inmates in contract
facilities (Table 6.3). This does not include some other costs to the Department associated
with contracting, including state monitoring and administration, a healthcare allowance, and
some allocated indirect costs–all of which increase the cost to the state by about seven to
eight percent.37 During that year, the average cost per prisoner for all prisons in
Oklahoma–both government operated and privately operated–was $37.62. If this were
adjusted by excluding the higher costs paid to contractors, the average for facilities operated
by the Department of Corrections would be still lower. The more telling comparison, though,
is what the state would have spent to house prisoners in new government-operated facilities
instead of putting them in contract facilities. The Department estimated that a new 1,000 bed
medium security facility would cost $45.21 per day in FY2000, including the annualized
costs of debt service on financing the $44 million required of construction and startup costs
(almost $11 per prisoner/day). This was higher than the average cost of existing stateoperated medium security facilities, largely because there was no debt to service for these
older facilities.38 This indicates that the real choice was between making a long-term
commitment to a $45 per prisoner/day prison (in FY2000 dollars) and paying about the same
amount in short-term contracts.
By FY2003, the difference between spending for new state-run prison and contracting
appeared to favor contracting. The department estimated that the cost per day during FY2003
of a newly constructed 1,000 bed medium security prison would have been $51.91–higher
than the average cost per day of operating already-built medium security prisoners ($46.55).
The daily rates for beds in contractor-operated facilities during this same fiscal year ranged
from $40.82 at the Wackenhut Corrections Corporation’s Lawton Correctional Facility to
$43.99 at the Cimarron Correctional Facility, owned by the Cushing Municipal Authority.
Adding the costs incurred by the department in connection with these contracts raised the
cost to the state to $44.19 at Lawton and $47.37 at Cimarron.39 These costs were less than
what the state would have spent to build and operate another prison.

37

Accounting of other contract related costs are reported in Oklahoma Department of Corrections, Projected FY2003 Cost of DOCOperated Medium Security Beds Compared to Private Prison Contracts,” undated.

38

Oklahoma Department of Corrections, “Projected FY2000 Cost of DOC Operated Medium Security Beds Compared to Private Prison
Contracts,” undated, at www.doc.state.ok.us/DOCS/rates.htm.

39

Oklahoma Department of Corrections, “Projected FY2003 Cost of DOC Operated Medium Security Beds Compared to Private Prison
Contracts,” undated.

94

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

From a financial standpoint, these comparisons suggest that the aggressive contracting
strategy makes sense for the state. The state avoided taking on debt for decades when the
need for prisons might diminish long before the debt was due. Oklahoma thereby shifted the
risks associated with large capital investments required of prisons to the private sector.
Paying private firms their per capita prices did not seem unreasonable compensation for
bearing the risk of building and maintaining properties that may not be needed.40
Table 6.3
Per Inmate/Day Fees Paid by Oklahoma for Privately Operated Prisons
(Contracts Active on December 31, 1997)
Facility

Per Diem

Central Texas Parole Violator Facility

$39.00

Cimarron Correctional Facility

$43.59

Davis Correctional Facility

$42.85

Great Plains Correctional Facility

$45.95

Karnes County Correctional Facility

n/a

Limestone County Detention Facility

n/a

Mansfield Law Enforcement Center

$42.00

Newton County Correctional Facility

$43.00

Odessa Detention Center

$39.00

Average for all OK DOC and Contract Facilities

$37.62

Source:

Contract facility rates from Abt Associates Inc. survey of state and federal correctional agencies,
1998; average DOC per diems from Camille Graham Camp and George Camp, The Corrections
Yearbook:1998 (Middletown, CT: Criminal Justice Institute, Inc., 1998), p. 90; per diem cost
refers to that in effect on 1/1/98.

Contracting Strategies to Manage Risk
Relying so heavily upon privately owned and operated prisons creates certain risks for the
state. These are substantially different from the risks borne by correctional departments in

40

This is not the only way that Oklahoma has shifted risk. Like Florida and Texas, the contractor is responsible for on-site medical and
dental care. In Texas, the contractor is responsible for all medical care and transportation during the first 48 hours of hospitalization.
In Florida, the contractor is responsible for the first $7,500 -- not to exceed five days of hospitalization. In Oklahoma, however, the
contractor is responsible for the first $70,000 and the State of Oklahoma pays 70 percent of the rest of the costs. Clearly, contractors
take on more medical cost risk in Oklahoma compared to those in Texas and Florida. (Source: Abt Associates survey of private
correctional facilities, 1998.)

Abt Associates Inc.

Oklahoma: Managing the Risks of Dependence

95

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

Texas, Florida, and elsewhere, which enter into contracts only to manage and operate
government-owned facilities. Absent special agreements, they cannot resume control of the
facilities in the event of trouble. Like all governments that contract with private firms, they
retain responsibility and liability for correctional officers’ performance, but do not have direct
control and authority over them. To manage these risks, the state has developed a number of
different legal strategies.
Protection of The State’s Bargaining Power Through Regulation

As Oklahoma moved through its options of early release programs, opening condemned cell
blocks and contracting with county jails and out-of-state facilities, the state was a sitting duck
in the eyes of prison bed suppliers such as CCA and Holdenville. The state needed the beds
and it needed the beds in Oklahoma. To minimize the negative effects of the department’s
dependence upon private suppliers, the legislature amended in 1997 the enabling statute
(Title 57, Section 561) to regulate the state’s relationship with the private prison industry and
to protect to the extent possible the state’s position vis-a-vis this industry.
Several provisions aim to assure competitive procurement. After the department’s initial
review, it must select at least three, but no more than five, contractor proposals for further
review.41 If there is more than one contract up for bid, the department must select at least
twice the number of bids contemplated to clear the first cut.42 In an attempt to avoid
agreements disadvantageous to the state, the law requires the department to compare the
bidders’ proposed operating and capital costs to those that would be incurred if the
department provided the service and/or construction.43 Title 57 also requires that each
contractor’s proposal demonstrate a cost benefit to the state when compared to the level and
quality of programs provided by similar state- operated facilities.44 These provisions serve to
provide a competitive atmosphere throughout two stages of the contracting process and fair
prices for the services proposed.
Once the state executes a contract with a contractor, more regulations come into play. The
contractor must offer a price that will not increase at a rate higher than the increase of the
Consumers Price Index for urban consumers.45 This protects the state from large rate hikes
from suppliers upon whom the department has become dependent. It also requires that the
state bear the risk to private firms of inflationary cost increases. The law states that any

41

Okl. Stat. Ann. tit. 57, sect. 561(F).

42

Ibid.

43

Ibid., sect. 561.1(G).

44

Ibid., sect. 561.1(D).

45

Ibid.

96

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

contract must also grant the department the option to purchase the facility at a predetermined
price.46 Thus, if it becomes advantageous for the state to own more facilities, it has an
avenue to acquire more prison properties quickly and with minimized disturbance to the
inmate population.
Sharing the Risks Associated with Under-Use

Firms that make investments to create prison beds face the risk that they will not be used at a
level sufficient to earn a profit (or break even). This creates an incentive to sell beds to any
correctional agency that needs them. This can have negative consequences for Oklahoma.
Beds may not be available when the need them. If beds are available, Oklahoma inmates may
be held with inmates sent there by other correctional agencies. Even though all may be
classified similarly with respect to security, differences in how governments classify
prisoners may result in some bad matches.
Oklahoma has developed procedures for sharing the risks of under-use of facilities with the
private providers, which still protect the state’s interests. In a contract between the State of
Oklahoma and the Wackenhut Corrections Corporation, for example, the state contracted to
place inmates in the Lawton Correctional Facility. Wackenhut agreed to make 750 cells or
1,500 beds available to Oklahoma.47 For each day an Oklahoma inmate resides in the Lawton
Facility, Oklahoma agreed to pay Wackenhut $40.00, although the contract capped the total
amount of payments to at $22,500,000 during the 1999 fiscal year.48
To eliminate the risk associated with the beds being leased to someone else when not being
used, some jurisdictions contract for a minimum number of beds and pay for them whether
filled or not (e.g., the Federal Bureau of Prisons’ arrangement with Wackenhut for a facility
at Taft, California). Oklahoma made no such agreement, in part because its general objective
is not to carry more capacity than it needs or uses. What Oklahoma and Wackenhut agreed
to, therefore, was that the firm would make 1,500 beds available to the state, but if the state
did not use them for a continuous 30 day period, Wackenhut could offer them to another
jurisdiction. Wackenhut agreed to give the Department written notice to offer such beds to
another jurisdiction. Upon receiving written notice, the Department has 5 business days to
house inmates in the reserved beds or pay the per diem to maintain their availability.
Moreover, if Wackenhut contracts to provide those beds to another jurisdiction at a higher
price than that paid by Oklahoma, upon the reliance of the removal of Oklahoma inmates,

46

Ibid., sect. 561.1(B).

47

Oklahoma/Wackenhut Contract, Contractual Agreement.

48

Ibid., Section 7.1.

Abt Associates Inc.

Oklahoma: Managing the Risks of Dependence

97

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

and there is a delay in removing those inmates, the State of Oklahoma must pay the higher
price until Oklahoma removes those inmates.49
Ensuring Adequate Quality and Performance

Not owning the privately operated facilities removes an enforcement option that both Texas
and Florida retain in their contracts—that is, the right to remove the contractor from a
publicly-owned facility. Thus, Oklahoma was forced to explore options that would balance
its need to achieve its policy objectives with its need to ensure that its inmates were kept in a
constitutionally acceptable environment.
The State of Oklahoma always has the option of pulling its inmates out of a facility—and it
has done this on more than one occasion. This is a drastic response, and it is more
complicated and risky task to the state than removing a contractor from a state-owned facility.
However, Oklahoma has built into its procedures less drastic, yet effective, enforcement
options to minimize the need to remove inmates. First, the enabling legislation created
several checkpoints to ensure that the department deals with reputable contractors. Second,
the contracts contain a comprehensive liquidated damages clause that acts as an incentive to
contractors to maintain a constitutionally acceptable environment for inmates.
Statutory Checkpoints

The statute requires the Department of Corrections to keep a current file of contractors that
are interested in and capable of operating and/or building a secure facility.50 This requires the
Department of Corrections to maintain intelligence on these organizations—such as
evaluations of past performance and lists of contracts each contractor holds in and outside of
Oklahoma.51 Only contractors on this list are eligible to bid on RFPs issued by the Oklahoma
Department of Corrections. (However, a private prison firm may submit a request at any
time to be included in this bidders lists.) Thus, when the department decides to issue an RFP,
the department has information that indicates that each potential bidder has the resources and
experience to deliver a constitutionally acceptable environment to Oklahoma inmates.
After the department issues an RFP there is an initial review by the department. The
legislation requires the department to consider: 1 ) the requirements of the project; 2) the
capacity of the contractor; and 3) the past performance of the contractor as indicated on the
performance evaluation form.52 After the initial screening, the department submits its

49

Ibid., Section 10.9.

50

Okl. Stat. Ann. tit. 57, sect. 561(C).

51

Ibid.

52

Ibid.

98

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

recommendations to the Board of Corrections. The Board of Corrections must consider the
contractor’s qualifications and experience; ability to provide qualified personnel; financial
condition; ability to comply with applicable court orders; and ability to acquire
accreditation.53 Furthermore, the legislation requires that proposals allow for on-site
monitoring and termination of the contract for cause.54 These provisions provide another
checkpoint for contractor quality and capacity during the procurement process. Furthermore,
after the department executes the contract, it has avenues of monitoring performance and the
enforcement power to terminate the contract in the case of a contractor breach.
Liquidated Damages Clause

When Oklahoma terminates a contract, it must remove its inmates from the contractor’s
prison.55 If the contractor breaches the contract, the state may seek damages allowed by law
or equity, liquidated damages as set forth in the contract, and/or termination of the contract
for cause. Oklahoma’s liquidated damages clause is somewhat unique, relative to those in
Texas and Florida contracts. It specifies that if a contractor fails to perform after 45 days
notice, the state may seek liquidated damages. The contractor must pay the amounts listed
for each day the breach occurs, and the amounts are calculated according to a formula:
payments=V x B x $25.00, where V = the relative value of service area and B = the relative
value of the breach. There are four service areas and four categories of breach.
Each service area contains a group of responsibilities of the contractor. The service areas and
their values are listed below along with the designated value of each breach.
Service Area 1 includes: security and control, ACA accreditation, health services, use of
force, escapes and contract monitoring.

V

B

Total Value
of Breach

Failure to Provide Service

5

5

$625

Failure to Document

5

2

$250

Failure to Report

5

2

$250

Failure to Comply with Other
Applicable Requirements

5

5

$625

Operator Breach

Source:

Oklahoma/Wackenhut Contract, Appendix C.

53

Ibid., sect. 561(L).

54

Ibid., sect. 561.1(D).

55

Correctional Services Contract Between Wackenhut Corrections Corporation and the State of Oklahoma Department of Corrections,
April 22, 1998, Section 10.4C.

Abt Associates Inc.

Oklahoma: Managing the Risks of Dependence

99

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

Service Area 2 includes: sanitation and hygiene, food service, mail, religion, access to court,
inmate discipline, grievance, visitation, records and reports, and employee qualifications &
training.

Operator Breach

V

Total Value
of Breach

B

Failure to Provide Service

4

4

$400

Failure to Document

4

2

$200

Failure to Report

4

2

$200

Failure to Comply with Other
Applicable Requirements

4

4

$400

Source:

Oklahoma/Wackenhut Contract, Appendix C.

Service Area 3 includes: operating standards, transportation, maintenance, repairs and
replacements, inmate work, academic & vocational training, sentence computation data,
classification & case management, commissary, policies/procedures/post orders, and inmate
management fund/bank accounts.

Operator Breach

V

Total Value
of Breach

B

Failure to Provide Service

3

3

$225

Failure to Document

3

1

$75

Failure to Report

3

1

$75

Failure to Comply with Other
Applicable Requirements

3

3

$225

Source:

Oklahoma/Wackenhut Contract, Appendix C.

Service Area 4 includes: laundry and inmate clothing, telecommunications,
supplies/perishables, and recreation.

100

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

Operator Breach

V

Total Value
of Breach

B

Failure to Provide Service

2

3

$150

Failure to Document

2

1

$50

Failure to Report

2

1

$50

Failure to Comply with Other
Applicable Requirements

2

3

$150

Source:

Oklahoma/Wackenhut Contract, Appendix C.

Although the Texas contract has a liquidated damages clause, these terms bear more
resemblance to the performance measures in the Texas contracts. That is, these terms
identify specific duties in which the failure to perform will result in a monetary penalty.
Monitoring

When all of Oklahoma’s contracts were for out-of-state facilities, contract monitors visited
the sites once per month.56 Such low levels of monitoring activity exposed the state to some
risk. Bringing its inmates back to in-state private facilities enabled the department to monitor
its contractors much more aggressively. Contract monitors now spend 60 percent to 80
percent FTE at the in-state facilities.57 Departmental administrators believe this to be
sufficient.58 The department also conducts regulatory audits. The intensity of monitoring
that is required depends, in great part, upon the experience of the private contractor’s staff.
In general, newer facilities require more monitoring. During these beginning stages, the
monitor assists the contractor in “getting up to speed.” As Linda Allen, an Oklahoma
contract monitor, stated, “once they [the contractors] know what to do, they will do it.”
According to Allen, startup periods are very sensitive times because inmates can become
frustrated with “breaking in new management,” and that inmates may know the protocol
better than the facility operators. For this reason, many monitors like having former
Oklahoma DOC staff in place.59

56

Abt Associates Inc. survey of state and federal correctional agencies, 1998.

57

Interview with Dennis December 7, 1998.

58

Ibid.

59

Interview with Linda Allen, Contract Monitor, State of Oklahoma Department of Corrections, December 9, 2000.

Abt Associates Inc.

Oklahoma: Managing the Risks of Dependence

101

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

Conclusion
The primary motivation behind Oklahoma’s decision to privatize was to prevent the
acquisition of prison properties. This motivation was a product of the economic difficulties
resulting from the oil bust during the 80's. According to Governor Frank Keating, by
absorbing increasing numbers of inmates through private contracts, the state would have
more flexibility when numbers begin to decrease. Oklahoma has been able to avoid the long
term obligations of servicing debt and maintaining prison properties. But has it been able to
avoid locking itself into political obligations to support the private prison industry in its
state? For now, the answer seems to be “yes.”
Has the Privatization Program Met Its Objectives?

The principal objectives of Oklahoma’s privatization program have been to reduce
overcrowding, to avoid capital investments in prisons, to avoid developing long-term
obligations to larger numbers of state government employees, to retain flexibility in
obligations in the event of a slackened demand for prison beds, and to do all of these without
spending more money than would be spent if the state had expanded its publicly-operated
prison system. No requirements were established in law to evaluate these outcomes, but
several conclusions can be drawn without complicated study.
Reducing Overcrowding
Use of contractor-operated facilities has enabled the state to meet the demand for beds and
thereby reduce overcrowding. Whereas the state's prisons were filled to 120 percent of
capacity prior to engaging private correctional facilities, they were operating slightly below
capacity (96 percent) at the end of 2001.60
Averting capital investments
Oklahoma has succeeded in housing substantial numbers of offender in prisons without
spending a penny on new facilities. For fiscal year 1998, $0 was allocated to the Department
of Corrections’ budget for capital construction or debt service.61 At the end of 2000, 6,000 of
the state’s prisoners were being held in privately owned and operated correctional facilities.
Had the state chosen to build six 1,000 bed medium security prisons, the cost of construction
and startup would have been in excess of $200 million (based on the estimates for what it
would cost to build and equip one 1,000 bed facility in FY2000).

60

Oklahoma Department of Corrections, “Capacity compared to facility count,” at www.doc.state.ok.us/charts/capcount.htm.

61

Camp, Camille Graham and Camp, George M. The Corrections Yearbook: 1998 (Middletown, CT: Criminal Justice Institute, Inc.,
1998), p. 86.

102

Governments’ Management of Private Prisons

Abt Associates Inc.

This document is a research report submitted to the U.S. Department of
Justice. This report has not been published by the Department. Opinions or
points of view expressed are those of the author(s) and do not necessarily
reflect the official position or policies of the U.S. Department of Justice.

Avoiding growth of government workforce
By developing a mixed public and private correctional system in the state, the privatization
program has allowed the state to expand capacity without increasing proportionately the
numbers of state employees. Had the state built enough prisons to house those now held in
privately operated facilities, thereby achieving the current levels of utilizing capacity in the
state’s prisons, the DOC workforce would have increased substantially. Shrinking that
workforce by means other than attrition is difficult. Public employees are represented by
union representatives who wield substantial political power in state elections.
Avoiding the development of a constituency to preserve high rates of incarceration
Governor Keating’s strategy was to avoid making long term financial commitments by
relying upon contracts that could be ended when demand slackened. Senator Hobson did not
believe that contracting with in-state facilities would be as flexible as proponents purported.
He argued that regardless of whether the operating agency is the public DOC or a private
firm, a constituency dependent upon the prison will develop, which will make it difficult for
the legislature to choose a policy that would result in shrinking the demand for prisons. This
argument has long been made by several critics of privatization, worrying about the
development of a “corrections industrial complex” that will distort policymaking to the
advantage of private contractors.
Shutting down a major employer in a rural area can be politically difficult, whether the
employer is a state or public agency. Employees are represented by legislators who strive to
protect their interests. Nonetheless, it is probably easier to abandon a privately owned and
operated prison than a publicly operated one in the event of a downturn in numbers of
prisoners. Given a choice between making a commitment to a larger public employee
workforce and to private contractors and their employees, it is likely that the latter will afford
the state’s governor, legislators, and correctional administrators more flexibility to adjust to
the hoped for decline in demand for prison space.

Abt Associates Inc.

Oklahoma: Managing the Risks of Dependence

103