GAO
September 2012

United States Government Accountability Office

A Report to Congressional Requesters

BUREAU OF
PRISONS
Growing Inmate
Crowding Negatively
Affects Inmates, Staff,
and Infrastructure

GAO-12-743

September 2012

BUREAU OF PRISONS
Growing Inmate Crowding Negatively Affects
Inmates, Staff, and Infrastructure
Highlights of GAO-12-743, a report to
congressional requesters

Why GAO Did This Study

What GAO Found

BOP operates 117 federal prisons to
house approximately 178,000 federal
offenders, and contracts with private
companies and some state
governments to house about another
40,000 inmates. BOP calculates the
number of prisoners that each BOPrun institution can house safely and
securely (i.e., rated capacity). GAO
was asked to address (1) the growth in
BOP’s population from fiscal years
2006 through 2011 and BOP’s
projections for inmate population and
capacity; (2) the effects of a growing
federal prison population on operations
within BOP facilities, and the extent to
which BOP has taken actions to
mitigate these effects; and (3) actions
selected states have taken to reduce
their prison populations, and the extent
to which BOP has implemented similar
initiatives.

The Department of Justice’s Bureau of Prisons’ (BOP) 9.5 percent population
growth from fiscal years 2006 through 2011 exceeded the 7 percent increase in
its rated capacity, and BOP projects continued population growth. Growth was
most concentrated among male inmates, and in 2011, 48 percent of the inmates
BOP housed were sentenced for drugs. From fiscal years 2006 through 2011,
BOP increased its rated capacity by about 8,300 beds as a result of opening 5
new facilities and closing 4 minimum security camps, but because of the
population expansion, crowding (or population in excess of rated capacity)
increased from 36 to 39 percent. In 2011 crowding was most severe (55 percent)
in highest security facilities. BOP’s 2020 long-range capacity plan projects
continued growth in the federal prison population from fiscal years 2012 through
2020, with systemwide crowding exceeding 45 percent through 2018.

GAO analyzed BOP’s inmate
population data from fiscal years 2006
through 2011, BOP’s 2020 long-range
capacity plan, and BOP policies and
statutory authority. GAO visited five
federal prisons chosen on the basis of
geographic dispersion and varying
security levels. The results are not
generalizable, but provide information
on the effects of a growing prison
population. GAO selected five states
based on actions they took to mitigate
the effects of their growing prison
populations—and assessed the extent
to which their actions would be
possible for BOP. GAO makes no
recommendations in this report. BOP
provided technical clarifications, which
GAO incorporated where appropriate.

According to BOP, the growth in the federal inmate population has negatively
affected inmates, staff, and infrastructure, but BOP has acted within its authority
to help mitigate the effects of this growth. BOP officials reported increased use of
double and triple bunking, waiting lists for education and drug treatment
programs, limited meaningful work opportunities, and increased inmate-to-staff
ratios. These factors, taken together, contribute to increased inmate misconduct,
which negatively affects the safety and security of inmates and staff. BOP
officials and union representatives voiced concerns about a serious incident
occurring. To manage its growing population, BOP staggers meal times and
segregates inmates involved in disciplinary infractions, among other things.
The five states in GAO’s review have taken more actions than BOP to reduce
their prison populations, because these states have legislative authority that BOP
does not have. These states have modified criminal statutes and sentencing,
relocated inmates to local facilities, and provided inmates with additional
opportunities for early release. BOP generally does not have similar authority. For
example, BOP cannot shorten an inmate’s sentence or transfer inmates to local
prisons. Efforts to address the crowding issue could include (1) reducing the
inmate population by actions such as reforming sentencing laws, (2) increasing
capacity by actions such as constructing new prisons, or (3) some combination of
both.
A Triple-Bunked Cell in a BOP Facility

View GAO-12-743. For more information,
contact David C. Maurer at (202) 512-9627 or
maurerd@gao.gov.
United States Government Accountability Office

Contents

Letter

1
Background
BOP’s Population Grew More than Systemwide Capacity, and BOP
Projects Continued Population Growth through 2020
BOP’s Population Growth Has Negatively Affected Inmates, Staff,
and Infrastructure, but BOP Has Acted to Help Mitigate These
Effects
States Have Taken Broader Actions Intended to Reduce Prison
Populations than Those Taken at the Federal Level
Concluding Observations
Agency Comments

6
12
18
32
39
41

Appendix I

Objectives, Scope, and Methodology

42

Appendix II

BOP’s Population Growth

48

Appendix III

Effects of a Growing Inmate Population

64

Appendix IV

GAO Contact and Staff Acknowledgments

84

Related GAO Products

85

Tables
Table 1: Site Visits to BOP Facilities
Table 2: Site Visits to State Correctional Facilities
Table 3: Offense Composition of the Inmate Population in BOP
Facilities, by Year, from Fiscal Years 2006 through 2011
Table 4: Offense Composition of the Non-U.S. Citizen Inmate
Population in BOP Facilities from Fiscal Years 2006
through 2011
Table 5: BOP’s Rated Capacity and Crowding, by Facility Security
Level as of September 2011

Page i

44
47
50
51
52

GAO-12-743 Federal Prison Crowding

Table 6: BOP Population, Rated Capacity, and Percentage
Crowding from Fiscal Years 2006 through 2011
Table 7: BOP Projected Population, Rated Capacity, and
Percentage Crowding from Fiscal Years 2012 through 2020
Table 8: Rated Capacity and Temporary Bed Space by Institutional
Security Level From Fiscal Years 2006 through 2011
Table 9: Rated Capacity and Temporary Bed Space of Selected
BOP Facilities from Fiscal Years 2006 through 2011
Table 10: Systemwide Inmate Participation Rates in Selected BOP
Programs in September 2011
Table 11: BOP’s Drug Education Programs in Male Facilities:
Inmate Participation Levels, Waiting List Numbers, and
Average Waiting Time, by Institutional Security Level from
the End of Fiscal Years 2006 through 2011
Table 12: BOP’s Nonresidential Drug Treatment Programs in Male
Facilities: Inmate Participation Levels, Waiting List
Numbers, and Average Waiting Time, by Institutional
Security Level from the End of Fiscal Years 2006
through 2011
Table 13: BOP’s Residential Drug Abuse Programs in Male
Facilities: Inmate Participation Levels, Waiting List
Numbers, and Average Waiting Time, by Institutional
Security Level from the End of Fiscal Years 2006
through 2011
Table 14: BOP Inmate to Total BOP Staff Ratios from Fiscal Years
1997 through 2011
Table15: BOP’s Inmate to Total Institutional Staff Ratios from
Fiscal Years 2006 through 2011
Table 16: Snapshots of BOP’s Inmate to Correctional Officer
Ratios, by BOP Region, from Fiscal Years 2006
through 2011
Table 17: Guilty Findings for Prohibited Acts by Severity Level,
Fiscal Years 2006 through 2011

83

Figure 1: Illustration of a 20-Cell Medium Security Facility with 45
Inmates and 50 Percent Crowding and a 20-Cell High
Security Facility with 39 Inmates and 56 Percent
Crowding

10

55
60
65
66
69

71

72

73
78
79
80

Figures

Page ii

GAO-12-743 Federal Prison Crowding

Figure 2: BOP Systemwide Population, Rated Capacity, and
Percentage Crowding from Fiscal Years 2006 through
2011
Figure 3: Examples of BOP’s Use of Temporary Bed Space
Figure 4: Percentage Crowding in Male Long-Term Facilities from
Fiscal Years 2006 through 2011 by Institutional Security
Level
Figure 5: BOP’s Projections for Population, Rated Capacity, and
Percentage Crowding from Fiscal Years 2012 through
2020
Figure 6: Projected Percentage Crowding in Male Long-Term
Facilities from Fiscal Years 2012 through 2020 by
Institutional Security Level

Page iii

15
19
53
57
58

GAO-12-743 Federal Prison Crowding

Abbreviations
ACA
ADP
ADX
BOP
DHO
DOJ
ESL
GED
IG
NIC
OMB
PHS
RDAP
RICO
RRC
S&E
UDC

American Correctional Association
average daily population
Administrative Maximum
Bureau of Prisons
disciplinary hearing officer
Department of Justice
English as a Second Language
General Educational Development
Inspector General
National Institute of Corrections
Office of Management and Budget
Public Health Service
Residential Drug Abuse Program
Racketeer Influenced and Corrupt Organizations Act
Residential Reentry Center
Salaries and Expenses
Unit Discipline Committee

This is a work of the U.S. government and is not subject to copyright protection in the
United States. The published product may be reproduced and distributed in its entirety
without further permission from GAO. However, because this work may contain
copyrighted images or other material, permission from the copyright holder may be
necessary if you wish to reproduce this material separately.

Page iv

GAO-12-743 Federal Prison Crowding

United States Government Accountability Office
Washington, DC 20548

September 12, 2012
The Honorable Elijah E. Cummings
Ranking Member
Committee on Oversight
and Government Reform
House of Representatives
The Honorable Robert C. Scott
Ranking Member
Subcommittee on Crime, Terrorism,
and Homeland Security
Committee on the Judiciary
House of Representatives
As of December 2010, federal and state correctional facilities
incarcerated more than 1.6 million persons (about 1 in 200 U.S.
residents), according to the most recently available data from the U.S.
Department of Justice (DOJ). 1 At the federal level, DOJ’s Bureau of
Prisons (BOP) is responsible for approximately 218,000 of these inmates,
with a fiscal year 2012 operating budget of about $6.6 billion—the second
largest budget within DOJ. 2 BOP’s population has increased by more
than 400 percent since the late 1980s, and by about 50 percent since
2000. According to BOP, this growth is primarily attributed to an increase
in inmates’ sentence length over time. 3 At the end of fiscal year 2011,
BOP housed nearly 178,000 inmates in the 117 institutions that it owns

1

See Department of Justice, Bureau of Justice Statistics, Prisoners in 2010 (Washington,
D.C.: December 2011). The Bureau of Justice Statistics is the statistical agency of DOJ.

2

The Federal Bureau of Investigation has the largest budget within DOJ at $8 billion.
BOP’s $6.6 billion includes $6.551 billion for salaries and expenses and $90 million for
buildings and facilities.

3
The Sentencing Reform Act of 1984, Pub. L. No 98-473, 98 Stat. 1987, changed the
federal sentencing structure. The act was effective for offenses committed on or after
November 1, 1987. The act abolished parole, and subsequent legislation established
mandatory minimum sentences for many federal offenses, which limits the authority that
BOP has to affect the length of the sentence or the size of the inmate population.

Page 1

GAO-12-743 Federal Prison Crowding

and operates, and it contracted with other correctional facilities, such as
those of states and private companies, to house nearly 40,000 inmates. 4
BOP calculates the number of prisoners a given prison facility is built to
house safely and securely and calls this its rated capacity. 5 Crowding, as
defined by BOP, is the extent to which a facility’s inmate population level
exceeds its rated capacity. 6 Systemwide, BOP prisons exceed their rated
capacity by 39 percent, and crowding has been a significant ongoing
concern. Assessments conducted through the Federal Managers’
Financial Integrity Act have identified prison crowding as a material
weakness since 2006. 7 Further, since that same year, DOJ’s Inspector
General (IG) has included detention and incarceration among DOJ’s top
10 management and performance challenges departmentwide. In its 2011
list of challenges, the IG noted its concerns regarding the impact of
federal prison crowding and the related stresses on BOP’s prison staffing
regarding BOP’s ability to safely manage the increasing federal inmate
population. 8 The Attorney General has reported both of these findings
annually in his performance and accountability report for the department.
While federal inmate population growth has been steady, the overall
growth of the state inmate population began to decline in 2009. DOJ
reports that the overall state prison population increased from calendar

4

BOP has established performance-based contracts with four private corrections
companies. The private prisons in which BOP houses federal inmates operate in
accordance with BOP policies.

5

Rated capacity is the maximum population level at which an institution can make
available basic necessities, essential services (e.g., medical care), and programs (e.g.,
drug treatment, basic education, and vocational education). According to BOP officials, by
contract with BOP, privately run prisons cannot exceed 15 percent overcapacity. BOP also
has agreements with state and local governments and contracts with privately operated
facilities for the detention of federally adjudicated juveniles and for the secure detention of
some short-term federal inmates.

6

Unless noted otherwise, the term “crowding” in this report refers to BOP’s definition.

7

The Federal Managers’ Financial Integrity Act, Pub. L. No. 97-255, 96 Stat. 814 (1982),
assessment process evaluates the effectiveness of internal controls to support effective
and efficient program operations, reliable financial reporting, compliance with applicable
laws and regulations, and whether financial management systems conform to financial
system requirements.

8

Since 1998, the IG has prepared lists of top management challenges for the department.
By statute, the Attorney General is required to include each year’s list in the annual
performance and accountability report.

Page 2

GAO-12-743 Federal Prison Crowding

years 1977 through 2008, with the first decline of 0.2 percent (2,857
prisoners) occurring in 2009. DOJ reported a second-year decline in 2010
of 0.8 percent (10,881 prisoners), with 25 states reporting decreases in
their prison populations. 9 While not all states experienced reductions in
their prison populations, in those states that did, changes in those states’
policies and practices are potential contributing factors to this decline.
According to a 2008 study by the Pew Center on the States, a state may
reduce its prison population growth while protecting public safety by (1)
diverting a greater number of low-risk offenders from prison, (2) reducing
the time that low-risk offenders are in prison, or (3) a combination of these
approaches. 10 In contrast to the prison populations of the states, the
federal prison population has continued to grow. BOP is required by
statute to provide for suitable housing and the safekeeping, care, and
subsistence of all persons charged with or convicted of offenses against
the United States. 11 Thus, while the size of a prison population is, in part,
a function of crime rates, sentencing laws, and law enforcement policies,
these factors are all beyond BOP’s control. BOP’s population is expected
to continually increase, given current incarceration rates, and safety and
security concerns will remain paramount.
We have previously reported on BOP’s population projections, the
security and safety of inmates and staff, and inmate programs. 12 For
example, in November 2009, having assessed how BOP developed its
population projections and compared its projections with its actual inmate
population growth from fiscal years 1999 through August 20, 2009, we
concluded that BOP’s projections were accurate, on average, to within 1
percent of the actual inmate population growth during this time period. 13
In February 2012, we reported on BOP’s use of its discretionary authority

9

See Department of Justice, Bureau of Justice Statistics, Prisoners in 2010 (Washington,
D.C.: December 2011), and Prisoners in 2009 (Washington, D.C.: December 2010).

10
The Pew Center on the States, One in 100: Behind Bars in America 2008 (Washington,
D.C.: February 2008). The Pew Center on the States provides nonpartisan reporting and
research, advocacy, and technical assistance to help states deliver better results. Among
the issues it addresses are heath, the economy, revenue and spending, and public safety.
11

18 U.S.C. § 4042(a)(2).

12

See Related GAO Products at the end of this report.

13

GAO, Bureau of Prisons: Methods for Cost Estimation Largely Reflect Best Practices but
Quantifying Risks Would Enhance Decision Making, GAO-10-94 (Washington, D.C.:
Nov.10, 2009).

Page 3

GAO-12-743 Federal Prison Crowding

to reduce a prisoner’s period of incarceration. 14 We recommended that
BOP establish a plan, including time frames and milestones, for requiring
contractors to submit separate prices of beds in residential reentry
centers—also known as halfway houses—and home detention services.
BOP concurred and has actions under way to address the
recommendation.
You asked us to review the impact of crowding in BOP facilities and any
related lessons that BOP can learn from selected states. Specifically, this
report addresses the following questions:
1. What was the growth in BOP’s population from fiscal years 2006
through 2011, and what are BOP’s projections for inmate population
and capacity?
2. What is known about the effects of a growing federal prison
population on operations (i.e., inmates, staff, and infrastructure) within
BOP facilities, and to what extent has BOP taken actions to mitigate
these effects?
3. What actions have selected states taken to reduce their prison
populations, and to what extent has BOP implemented similar
initiatives?
To address the first question, we analyzed policies and procedures that
may affect the increased federal prison population (e.g., BOP’s inmate
classification policy) and BOP’s statutory authority affecting its capacity
and conditions of confinement. We also analyzed BOP’s inmate
population data and crowding percentages by institutional security level
from fiscal years 2006 through 2011 and BOP’s 2020 long-range capacity
plan, which was issued in January 2012. We assessed the reliability of
BOP’s inmate population and crowding data by reviewing relevant
documentation, interviewing knowledgeable agency officials about how
they maintain the integrity of their data, and updating previous
assessments that we did for previously issued reports. We found BOP’s
inmate population and crowding data to be sufficiently reliable for the

14
For GAO reports on federal prisons, see, for example: GAO, Bureau of Prisons:
Eligibility and Capacity Impact Use of Flexibilities to Reduce Inmates’ Time in Prison,
GAO-12-320 (Washington, D.C. Feb. 7, 2012).

Page 4

GAO-12-743 Federal Prison Crowding

purposes of this report. We also interviewed BOP headquarters officials
to discuss how BOP’s population has grown.
To address the second question, we analyzed BOP’s statutory authority,
policies, and procedures pertinent to the effects of the growing population
on operations in BOP facilities (i.e., effects on inmates, staff, and
infrastructure) and BOP’s ability to mitigate the effects of a growing
population. We also analyzed BOP studies on the effects of population
growth and prison crowding on BOP operations, as well as BOP data
from fiscal years 2006 through 2011 on available bed space, inmate
program participation and waiting lists, inmate-to-staff ratios, and
available infrastructure costs. We also present systemwide BOP staffing
ratios from fiscal years 1997 through 2011 because officials believed that
presenting the ratios for a longer period better illustrates the effect of
BOP’s population growth relative to the number of staff. 15 We assessed
the reliability of BOP’s inmate, staff, and infrastructure data by
interviewing knowledgeable agency officials to determine how BOP
collects and maintains the integrity of these data. We found these data to
be sufficiently reliable for the purposes of this report. We visited 5 of
BOP’s 117 prisons that are located in four of BOP’s six regions, which we
chose on the basis of varying security levels and to ensure geographic
dispersion. 16 Because we did not randomly select the prisons we visited,
our results are not generalizable to all BOP prisons; however, they
provided important insights into BOP’s operations. We interviewed BOP
headquarters officials and all six regional directors. Further, we discussed
the effects of BOP’s population growth on correctional staff with officials
from the Council of Prison Locals, the union that represents all
nonmanagement staff working in BOP facilities.
To also address the second as well as the third question, we compared
and contrasted BOP’s actions to mitigate the effects of its increased
population and attempt to reduce its prison population with similar actions
taken by five selected states––Kansas, Mississippi, New York, Ohio, and
Wisconsin––that had experienced prison population growth and had

15

According to BOP officials, BOP also includes this information in its annual
congressional budget request.
16

We selected five federal prisons of different security levels, including one that was a
complex and on whose grounds there was a low and a medium security facility. Thus, our
five selected sites included six BOP facilities––one low, three medium, one high, and one
administrative. We describe the distinctions among security levels later in this report.

Page 5

GAO-12-743 Federal Prison Crowding

taken actions to mitigate its effects or reduce their prison populations. To
select the states, among other things, we reviewed the Bureau of Justice
Statistics’s (BJS) report on state prison inmate populations in 2010. 17 We
also reviewed relevant reports on actions that states have taken to
mitigate the effects of their prison population growth, published from 2006
through 2011 (e.g., those from the Pew Center on the States). Further,
we assessed the extent to which actions in these selected states would
be possible for BOP to undertake within its statutory authority. We also
conducted site visits to three facilities in two of these five states.
Dissimilarities between federal and state prison systems—legally,
structurally, and in how crowding calculations are determined––limit the
comparability between federal and state correctional systems, but we
mitigated this limitation by the criteria we used to select the states in our
sample (e.g., size of the prison population and diverse approaches to
addressing increased prison populations). We are unable to generalize
about the types of actions states have taken to mitigate the effects of
state prison population growth or reduce their prison populations, but the
information we obtained provides examples of state responses to prison
population growth. Appendix I includes more details about our scope and
methodology.
We conducted this performance audit from September 2011 to
September 2012 in accordance with generally accepted government
auditing standards. Those standards require that we plan and perform the
audit to obtain sufficient, appropriate evidence to provide a reasonable
basis for our findings and conclusions based on our audit objectives. We
believe that the evidence obtained provides a reasonable basis for our
findings and conclusions based on our audit objectives.

Background

There are specific state and federal laws that define, prohibit, and
penalize criminal behavior. Various factors, such as the nature and type
of the crime committed and the relevant law, may determine whether the
state or federal justice system is responsible for the prosecution,
sentencing, and incarceration of an individual accused and found guilty of
a crime. State and federal laws also define the potential sentences for
those crimes to be imposed by judges and methods for reducing the
period of incarceration, such as parole, probation, or good conduct time

17

See BJS, Prisoners in 2010 (Washington, D.C.: Dec. 2011).

Page 6

GAO-12-743 Federal Prison Crowding

credit. These laws and policies affect the growth of their respective prison
populations and the level of crowding in state and federal prison
populations.

Federal Prison System

To carry out its responsibility for the custody and care of federal offenders,
BOP currently houses inmates across six geographic regions in 117 federal
institutions, 15 privately managed prisons, 185 residential reentry centers,
and home detention. 18 BOP’s central office consists of eight divisions that
provide oversight of major BOP program areas and operations, such as
correctional programs and health services, as well as the National Institute
of Corrections (NIC). 19 BOP has six regional offices, each led by a regional
director, covering the Mid-Atlantic, North Central, Northeast, South Central,
Southeast, and Western regions of the United States.
BOP generally houses sentenced inmates in its long-term institutions.
Male long-term institutions include four security level designations––
minimum, low, medium, and high––and female institutions include three
security designations––minimum, secure, and high. 20 The security level
designation of a facility depends on the level of security and staff
supervision that the institution is able to provide, such as the presence of
security towers; perimeter barriers; the type of inmate housing, including
dormitory, cubicle, or cell-type housing; and inmate-to-staff ratio.
Additionally, BOP designates some of its institutions as administrative
institutions, which specifically serve inmates awaiting trial, or those with
intensive medical or mental health conditions, regardless of the level of
supervision these inmates require. 21 From fiscal years 2006 through

18

According to BOP officials, privately managed contract facilities are low security and
primarily house non-U.S. citizens convicted of crimes while in this country legally or
illegally. Home detention describes all circumstances under which an inmate is serving a
portion of his or her sentence while residing in his or her home.

19
NIC, a component of BOP, provides training, technical assistance, information services,
and policy/program development assistance to federal, state, and local corrections
agencies.
20

In this report, data presented by institutional security level include information for male
inmates by the four security levels and for females by the three security levels. Unless
noted, these data do not include information on detention, medical, administrative, or
Witness Security Program housing.
21

The Administrative Maximum (ADX) facility in Florence, Colorado, houses offenders
requiring the tightest controls.

Page 7

GAO-12-743 Federal Prison Crowding

2011, the distribution of facilities by security designation remained
relatively constant. In fiscal year 2011, there were 7 stand-alone minimum
security camps, 29 low security facilities, 46 medium security facilities, 16
high security facilities, and 19 administrative facilities. 22
Rated capacity reflects the number of
prisoners that the institution was built
to house safely and securely.

BOP establishes a rated capacity for each of the facilities that it owns and
operates. 23 A facility’s rated capacity reflects the number of prisoners that
it was designed to house safely and securely and in which BOP can
provide inmates adequate access to services, necessities for daily living,
and programs designed to support their crime-free return to the
community. 24 In determining a facility’s rated capacity, BOP considers
American Correctional Association (ACA) occupancy and space
requirements. 25 Since 1990, ACA has required 35 square feet of
unencumbered space per inmate to ensure that each inmate has
sufficient movement or exercise space within the inmate’s personal living
space, whether in a cell, room, or open dormitory. In essence, rated
capacity is the measure of inmate housing space and, therefore, does not
include housing used for medical and special housing purposes (e.g.,
disciplinary segregation and administrative detention space). BOP also
does not include in its rated capacity additional beds placed in areas such
as a facility’s halls, gyms, mezzanines, or television rooms to address
crowding. BOP excludes this use of space from its rated capacity
calculation because it considers these beds to be temporarily converted
housing space that is to be restored to its original purpose when
circumstances permit. Further, for such temporary space to become
permanent space and thus included in a facility’s rated capacity, the

22
BOP has 7 stand-alone minimum security camps that are not colocated with higher
security level facilities. BOP also has 73 minimum security satellite camps that are
colocated with a secure institution or complex. Stand-alone camps usually have a rated
capacity of 256 inmates, and colocated camps usually have a rated capacity of 128
inmates. Female secure facilities are included in the low security level facility figure.
23

BOP does not include privately contracted beds as part of its systemwide rated capacity,
because the capacity of each of these facilities is based on its contract.
24

Basic necessities include safety, living space, and access to toilets, showers, and food.
Essential services include medical care, visitation, and telephones to allow contact with
family and other members of the community. Programs include drug treatment, work,
education, vocational training, anger management, and parenting to prevent idleness and
enable inmates to develop skills needed to return to the community following release.

25
ACA’s mission includes the development and promotion of effective standards for the
care, custody, training, and treatment of offenders.

Page 8

GAO-12-743 Federal Prison Crowding

facility would require infrastructure changes, such as additional toilet or
shower facilities, to meet ACA standards.
Crowding in BOP’s facilities is the extent to
which a prison’s inmate population
exceeds the institution’s rated capacity.

According to BOP, rated capacity is the basis for measuring prison
crowding and is essential to both managing the inmate population and
BOP’s budget justifications for capital resources. BOP’s formula for
calculating rated capacity has changed over time. Until 1991, the rated
capacity of a facility was equivalent to the total number of cells, because
the rated capacity was based on one inmate being housed in each cell.
As a result of the growth in BOP’s population during the 1980s, BOP
began to double-bunk (i.e., house two inmates in each cell) in many of its
facilities, particularly those at the lower security levels. In 1991, BOP
established a new rated capacity formula that allowed for stratified
bunking across all security levels. BOP’s current rated capacity guidelines
account for
•

25 percent double bunking and 75 percent single bunking of cells
within high security facilities,

•

50 percent double bunking and 50 percent single bunking of cells
within medium security facilities, and

•

100 percent double bunking of cells in low and minimum security
facilities.

By way of illustration, figure 1 shows crowding in a medium security
facility and a high security facility, each with 20 cells. The rated capacity
of the medium security facility, which includes 50 percent double bunking,
is 30 beds. With a population of 45 inmates, 67 percent of the inmates are
double bunked, 33 percent are triple bunked, and the facility’s percentage
crowding is 50 percent. The rated capacity of the high security facility,
which includes 25 percent double bunking, is 25 beds. With a population
of 39 inmates, 97 percent of the inmates are double bunked, 3 percent
are single bunked, and the facility’s percentage crowding is 56 percent.

Page 9

GAO-12-743 Federal Prison Crowding

Figure 1: Illustration of a 20-Cell Medium Security Facility with 45 Inmates and 50
Percent Crowding and a 20-Cell High Security Facility with 39 Inmates and 56
Percent Crowding

BOP initially classifies an inmate to a particular institution based on
•

the level of security and supervision the inmate requires;

•

the level of security and staff supervision the institution is able to
provide;

Page 10

GAO-12-743 Federal Prison Crowding

•

the inmate’s program needs, such as residential drug treatment or
intensive medical care; 26

•

where the inmate resides when sentenced;

•

the level of crowding in an institution; and

•

any additional security measures to ensure the protection of victims,
witnesses, and the public.

In most cases, BOP’s Designation and Sentence Computation Center
staff calculates a point score for the inmate and then matches the inmate
with a commensurate security level institution. 27
As of December 31, 2011, BOP had a total staff of about 38,000,
including correctional officers 28 and administrative, program, and support
personnel responsible for all of BOP’s activities nationwide. 29 BOP’s
philosophy is that all employees are correctional workers first, whether or
not they serve as correctional officers. Accordingly, BOP trains all
employees in basic correctional duties to secure the facility in the event of
a disturbance and to provide inmate supervision. BOP also requires them
to participate in annual refresher training. 30 When circumstances warrant,
a warden will require program and administrative staff members to serve
in the capacity of a correctional officer––a practice that BOP calls
augmentation. For example, under augmentation, a vocational education
teacher or a psychologist may provide escort services for an inmate

26

For prior work related to BOP’s implementation of Second Chance Act provisions, which
affect programming needs related to preparing inmates for eventual reentry into society,
see GAO, Federal Bureau of Prisons: BOP Has Mechanisms in Place to Address Most
Second Chance Act Requirements and Is Working to Implement an Initiative Designed to
Reduce Recidivism, GAO-10-854R (Washington, D.C.: July 14, 2010).

27

See GAO-12-320.

28
Correctional officers enforce the regulations governing the operation of a correctional
institution, serving as both supervisors and counselors of inmates.
29

These staff included all staff on-board funded under BOP’s appropriation for Salaries
and Expenses (i.e., headquarters, regional, institutional, and Public Health Service staff),
as well as Buildings and Facilities, Commissary, and Federal Prison Industries staff.

30

See GAO, Bureau of Prisons: Evaluating the Impact of Protective Equipment Could Help
Enhance Officer Safety, GAO-11-410 (Washington, D.C.: Apr. 8, 2011).

Page 11

GAO-12-743 Federal Prison Crowding

leaving the facility for specialized medical care or provide ancillary
supervision in a recreational yard.

State Prison Systems

Selected state departments of corrections included in our review share
similarities and exhibit differences with BOP. For example, both state
departments of corrections and BOP are required to house, clothe, and
feed inmates in a safe and secure setting, but selected states determine
rated capacity and measure crowding differently. Several of the selected
state departments of corrections’ methods are different from BOP’s
methods. For example, New York calculates rated capacity using
standards set forth by the New York State Commission of Correction and
by subtracting temporary beds from the number of general population
beds. In contrast, Wisconsin does not calculate rated capacity but instead
uses design capacity and operational capacity. 31 Furthermore, differences
in state and federal authorities affect the types of actions that are taken to
mitigate the effects of crowding. For example, state departments of
corrections may have been granted certain state statutory authorities that
are not currently available at the federal level (i.e., states may transfer
inmates to county and local jails, but BOP does not have this option).

BOP’s Population
Grew More than
Systemwide Capacity,
and BOP Projects
Continued Population
Growth through 2020

From fiscal years 2006 through 2011, the inmate population in BOP-run
facilities grew 9.5 percent, while capacity grew less than 7 percent. As a
result, BOP’s overall crowding increased during this period from 36
percent to 39 percent. BOP projects an additional 15 percent increase in
its inmate population by 2020.

31
Design capacity is the number of inmates that planners intended for a facility.
Operational capacity is the number of inmates that can be accommodated based on a
facility’s staff, existing programs, and services.

Page 12

GAO-12-743 Federal Prison Crowding

BOP’s Population Grew
Steadily because of a
Variety of Factors

The inmate population housed in BOP-run facilities steadily increased
from 162,514 to 177,934 inmates—or 9.5 percent—from fiscal years 2006
through 2011. 32 A variety of factors contribute to the size of BOP’s
population. These include national crime levels, law enforcement policies,
and federal sentencing laws, all of which are beyond BOP’s control.
During the 6-year period, growth occurred in BOP’s male, female, and
both its U.S. citizen and non-U.S. citizen populations. Specifically,
•

The number of male inmates housed in BOP institutions increased
about 10 percent (151,003 to 165,595).

•

The number of female inmates housed in BOP institutions increased
about 7 percent (11,511 to 12,339).

•

The relative proportion of non-U.S. citizen to U.S. citizen inmates
housed in BOP facilities remained constant (about 26 percent),
although the approximately 16 percent (46,369 to 53,733) growth in
the noncitizen inmate population surpassed the approximate 13
percent growth in the U.S. citizen inmate population (135,074 to
152,581). 33
•

Non-U.S. citizen inmates are housed in BOP-run low, medium,
and high security level facilities, as well as in private contract
facilities. 34

•

The largest number of these inmates are housed in low security
facilities.

32

These data include only those U.S. inmates housed in BOP-run facilities, not privately
contracted facilities.

33
These data include only those U.S. citizen or non-U.S. citizen inmates housed in BOPrun facilities. The total number of non-U.S. citizen inmates, including those housed in
BOP-run and private contract facilities, increased about 13 percent (50,275 to 56,933)
from fiscal years 2006 through 2011, and constituted about 26 percent of the total BOP
population during this time period.
34

BOP does not send non-U.S. citizen inmates to minimum security facilities because of
their risk of flight.

Page 13

GAO-12-743 Federal Prison Crowding

•

Drug, weapons/explosives, and immigration offenses constituted the
largest number of offenses for which all BOP inmates were
incarcerated in each year from fiscal years 2006 through 2011. 35
Specifically, in fiscal year 2011,
•

48 percent of the inmates BOP housed were serving sentences
for drugs,

•

16 percent for weapons/explosives, and

•

12 percent for immigration.

Appendix II provides additional information on the growth of the federal
inmate population from fiscal years 2006 through 2011 and the other
offense categories for which BOP inmates have been sentenced.

Rated Capacity Grew Less
than 7 Percent,
Contributing to Crowding

BOP’s 9.5 percent population growth from fiscal years 2006 through 2011
among inmates housed in BOP facilities exceeded the increase in its
rated capacity, which grew less than 7 percent (from 119,510 beds to
127,795). BOP’s rated capacity during this 6-year period grew because it
opened five new facilities and closed four stand-alone minimum security
camps, which BOP officials told us were less efficient to operate. As
shown in figure 2, however, because the inmate population in BOP-run
facilities grew at a faster rate than the growth in rated capacity, crowding
in BOP-run institutions increased from 36 to 39 percent systemwide.

35

BOP officials explained that for reporting purposes, they categorize inmates according to
the offense for which an inmate is serving the longest sentence (dominant sentence
offense). For example, an inmate may be serving sentences for both drug and immigration
offenses, but BOP will categorize the inmate by the offense having the longer sentence
(e.g., the drug offense).

Page 14

GAO-12-743 Federal Prison Crowding

Figure 2: BOP Systemwide Population, Rated Capacity, and Percentage Crowding from Fiscal Years 2006 through 2011

From fiscal years 2006 through 2011, the percentage crowding in male
medium security facilities increased from 37 percent to 51 percent and
from 53 percent to 55 percent in high security level facilities. Table 5 in
appendix II illustrates the range of crowding across BOP institutions of
different security levels as of December 2011, and the double, triple, and
quadruple bunking that has resulted. For example, the population in
BOP’s high security population was about 21,000 in December 2011—or
about 7,000 more than its rated capacity—resulting in 97 percent double
bunking and a 55 percentage crowding.
According to BOP, BOP’s ability to increase rated capacity is directly
affected by funding appropriated for new prison construction and to
support contracts with private prison providers for additional inmate bed
space. In fiscal year 2005, the Office of Management and Budget (OMB)
placed a moratorium on all new BOP prison construction. To address
BOP’s bed space needs, OMB focused on contracting with private

Page 15

GAO-12-743 Federal Prison Crowding

prisons. BOP officials stated that, because of this moratorium, the yearly
presidential budget submissions for BOP’s Buildings and Facilities
account for each of fiscal years 2006 through 2011 did not include
requests to begin construction on any new facilities. Instead, the Buildings
and Facilities requests included what BOP considers to be baseline
funding36 for ongoing expenses, which were generally about $25 million
during each of the 6 fiscal years in this period.37 Congress, however,
provided about $1.1 billion specifically to aid in the site selection, design,
and construction of new BOP facilities—in addition to funding the baseline
that BOP requested. According to BOP officials, the time from receiving
the funding appropriation to building and opening—or activating—a facility
is generally 3 to 5 years.
In addition, according to BOP officials, funding was requested and
provided to contract for private bed space during this period—with the
exception of fiscal years 2008 and 2011. BOP officials explained that only
low security inmates can be housed in privately managed facilities; thus,
in years when BOP has received related funds, they have been able to
move these lower security inmates to the contracted facilities. However,
since they have not consistently received this money, BOP officials told
us they designated some low security inmates to medium security
facilities. As a result, BOP is currently housing 4,500 low security inmates
in medium security facilities, contributing to crowding at that security level.
BOP officials said that housing low security
inmates in medium security institutions is
contributing to crowding in medium
security institutions, but BOP’s low security
facilities are “just plain full.”

According to BOP data, 81 percent of male inmates housed in low
security facilities were triple bunked at the end of 2011. Officials noted
that if they were able to add more contract beds they could reduce
crowding in medium and low security facilities by moving (1) non-U.S.

36
BOP’s Building and Facilities budgetary account includes two “decision units”—one for
“new construction” and one for “maintenance and repair.” Within the “new construction”
decision unit, baseline funding includes about $10 million annually to support the lease
payments on BOP’s federal inmate transfer center, as well as other costs associated with
considering potential construction sites, studying environmental impact, and any facility
expansion and conversion projects. These baseline funds also cover the salaries and
administrative costs of architects, project managers, and procurement and other staff
necessary to carry out the efforts.
37
During this period, the fiscal year 2008 budget submission was the outlier, when the
request included $115 million to support construction that had already begun at the
Mendota facility. Congress ultimately appropriated these funds, but did so as part of the
fiscal year 2007 budget. Because of protracted budget negotiations in fiscal year 2007, the
budget that year passed after the fiscal year 2008 budget had been submitted.

Page 16

GAO-12-743 Federal Prison Crowding

citizen inmates from low security facilities to contract facilities and (2) low
security inmates from medium security facilities to low security facilities.
(See table 6 in app. II for additional data on BOP’s population growth,
rated capacity, and percentage crowding data for fiscal years 2006
through 2011.)

BOP Projects Continued
Population Growth

BOP’s 2020 long-range capacity plan assumes continued growth in the
federal prison population from fiscal years 2011 through 2020, with about
15 percent growth in the number of inmates BOP will house.38 To address
some of this growth, BOP expects to activate five newly constructed
prisons by 2014, adding about 6,720 beds.39 In addition, BOP is
budgeting for additional contracted bed space—1,000 beds in 2013 and
1,500 the next year, but the addition of these contracted beds is subject
to future appropriations. Despite its plans to add capacity through 2014,
given the expected inmate population growth, BOP projects crowding will
increase from the current rate of 39 percent to 44 percent by 2015.
Beyond 2015, BOP projects it will be able to bring crowding in BOP
facilities down to 35 percent by 2020. BOP’s projections assume that
BOP will receive additional funding for constructing new facilities.
Specifically, BOP assumes an overall increase of over 17,500 beds from
fiscal years 2016 through 2020, generally as a result of opening new high
and medium security facilities, none of which is under construction. BOP
has not requested funding for this additional bed space, and as a result,
its plans are contingent on the budget development and appropriations
processes and are subject to change. Appendix II provides additional
information on BOP’s population growth from fiscal years 2006 through
2011 and projections from fiscal years 2012 through 2020.

38

These projections are from BOP’s 2020 capacity plan dated January 10, 2012. In
November 2009, we concluded that BOP’s projections at the time were accurate, on
average, to within 1 percent of the actual inmate population growth from fiscal year 1999
through August 20, 2009. See GAO-10-94.
39

According to BOP, these facilities include two medium security/camp facilities in
Mendota, California (1,152 beds) and Berlin, New Hampshire (1,280 beds), that will open
from 2012 through 2013; one female secure/low security camp facility in Aliceville,
Alabama (1,792 beds), opening from 2012 through 2014; and one medium security/camp
facility in Hazelton, West Virginia (1,280 beds); and one high security facility in Yazoo City,
Mississippi (1,216 beds), scheduled to open from 2013 through 2014.

Page 17

GAO-12-743 Federal Prison Crowding

BOP’s Population
Growth Has
Negatively Affected
Inmates, Staff, and
Infrastructure, but
BOP Has Acted to
Help Mitigate These
Effects

According to BOP and our observations, the growth of the federal inmate
population and related crowding have negatively affected inmates housed
in BOP institutions, institutional staff, and the infrastructure of BOP
facilities, and have contributed to inmate misconduct, which affects staff
and inmate security and safety. Nevertheless, BOP officials said that it is
difficult to demonstrate or isolate the effects of crowding, per se, as
distinguished from population growth or other factors such as staffing
levels.

Impacts of Population
Growth
Inmates

The growth in the inmate population affects inmates’ daily living
conditions, program participation, meaningful work opportunities, and
visitation. Appendix III describes each in greater detail, and we present
some highlights here.

Daily Living

To increase available bed space, BOP reports double bunking in excess
of the percentages included in a facility’s rated capacity; triple and
quadruple bunking; or converting common space, such as a television
room, temporarily to housing space. As a result of BOP actions to
increase available bed space in its institutions to accommodate the
growing federal inmate population, more inmates are sharing cells and
other living units, which brings together for longer periods of time inmates
with a higher risk of violence and more potential victims. 40 Table 8 in
appendix III illustrates the use of temporary beds by institutional security
level, and shows, for example, that temporary beds, not including those
used for disciplinary purposes, composed about 29 percent of the bed
space in male high security facilities in fiscal year 2011. According to
BOP headquarters officials, wardens have discretion to provide temporary
beds by adding a third bunk within cells, converting a television room to
bed space, or using both approaches. The facility’s infrastructure also

40

BOP. The Effects of Changing Crowding and Staffing Levels in Federal Prisons on
Inmate Violence Rates—Executive Summary (Washington, D.C.: October 2005).

Page 18

GAO-12-743 Federal Prison Crowding

affects the approach the warden may implement. For example, the
smaller cells in older BOP facilities make it more difficult to add a third
bed, while the larger cells in newer facilities can be triple-bunked. The
officials noted, however, that triple-bunking all cells in a unit presents a
challenge to staff who have to manage the large number of inmates.
Additionally, a regional director may have a preferred approach to
providing temporary beds within his or her region. Figure 3 illustrates
some of the options used.
Figure 3: Examples of BOP’s Use of Temporary Bed Space

All of the BOP facilities we visited reported using temporary beds from
fiscal years 2006 through 2011 and continue to do so. At the time of our
site visits in 2011 and 2012, these facilities continued to use temporary
space. For example, we observed triple-bunked cells in a low security
facility and a converted television room that housed 10 inmates in a
medium security facility. In addition to experiencing crowding in a facility’s
housing and common areas, inmates may experience crowded bathroom
facilities, reductions in shower times, shortened meal times coupled with
longer waits for food service, and more limited recreational activities
because of the increased inmate population.

Program Participation

According to BOP officials, the growth in the inmate population affects the
availability of program opportunities, resulting in waiting lists and inmate
idleness. BOP provides programs including education, vocational training,

Page 19

GAO-12-743 Federal Prison Crowding

drug treatment, and faith-based reentry programs that help to rehabilitate
inmates and support correctional management. 41 BOP officials said that
two long-term benefits of inmate programming are (1) public safety,
attributable to enhanced inmate skill sets that can reduce future crime
and inmate rearrest rates, and (2) institutional safety and security
because of reduced inmate idleness.
According to BOP officials, facility staff provide a range of education
programs, including mandatory General Educational Development (GED)
courses; 8- to 10-week nonmandatory courses on topics such as
parenting, word processing, and conversational Spanish; occupational
training; and computer-based self-paced courses such as English. BOP
reported that overall inmate participation in one or more programs was 36
percent in September 2011 (see table 10 in app. III). 42 BOP also offers
residential drug abuse treatment in more than half of its facilities and
nonresidential drug abuse and drug education programs in all of its
facilities. The percentage of participation, number of inmates on waiting
lists, and length of the average waiting time varied by program. As tables
11 through 13 in appendix III illustrate, all of the drug treatment and drug
education programs had waiting lists from fiscal years 2006 through 2011.
For example, as of the end of fiscal year 2011, about 2,400 inmates in
male medium security institutions participated in residential drug
treatment, almost 3,000 more inmates were on the waiting list to
participate, and the average wait for enrollment exceeded 3 months.
According to BOP officials, if BOP cannot meet the substance abuse
treatment or education needs of inmates because it does not have the
staff needed to meet program demand, some inmates will not receive
programming benefits. As we reported in February 2012, long waiting lists
for BOP’s Residential Drug Abuse Program (RDAP), which provides
sentence reductions for eligible inmates who successfully complete the
program, 43 constrained BOP’s ability to admit participants early enough to

41

See GAO-01-483.

42

BOP’s Monthly Participation Reports provide a snapshot of program participation levels
of inmates within BOP facilities. Figures for overall inmate participation do not duplicate.
That is, if an inmate is enrolled in more than one program area (for example GED and
parenting), the inmate’s participation is counted only once.
43

28 C.F.R. § 550.53(b) outlines the RDAP eligibility criteria. 28 C.F.R. § 550.55 outlines
eligibility for early release.

Page 20

GAO-12-743 Federal Prison Crowding

earn their maximum allowable reductions in times served. 44 From fiscal
years 2009 through 2011, BOP expanded RDAP capacity by 400 slots.
Though wait times for enrollment have declined, the program continues to
experience long waiting lists.

Meaningful Work
Opportunities

According to BOP headquarters officials, the growth in the federal inmate
population has also affected inmate work opportunities, as it is difficult to
find meaningful work for all inmates, even though generally all inmates
are required to have a job. 45 BOP inmates participate in a variety of jobs.
For example, at facilities we visited, we observed inmate workers
preparing meals under the supervision of staff, sweeping the floors, and
working in a factory that produced and printed a variety of government
publications. BOP officials explained, however, that with the growth of the
prison population, fewer opportunities exist to engage in meaningful work.
This makes it difficult for staff to keep inmates busy, resulting in inmate
idleness, which can lead to additional tension and fighting between
inmates. For example, officials at one facility told us that more inmates
than needed may be assigned to a task and paid the same wage, but
consequently, not everyone is engaged and equally busy.

Visitation

According to BOP headquarters officials, the quality of the interaction
between an inmate and family can positively affect an inmate’s behavior
in prison and aids an inmate’s success when returning to the community;
however, crowded visiting rooms make it more difficult for inmates to visit
with their families. Each BOP facility has visiting space to accommodate
the number of inmates that the facility was designed to house and a
visitor capacity to enable staff to manage the visitation process. The
infrastructure of the facility may not support the increase in visitors as a
result of the growth in the prison population. For example, at one older
facility we visited, officials said that the number of visitors was so great
and the visiting room was so small that visitors had to wait hours to get
into the visiting room.

44

For more on RDAP, see GAO-12-320.

45

The Crime Control Act of 1990, Pub. L. No. 101-647, § 2905, 104 Stat. 4789, 4914
(codified at 18 U.S.C. § 4121 note) established a mandatory work requirement for all
federal prisoners. A prisoner may be excused from this requirement only as necessitated
by security considerations; disciplinary action; medical certification of disability; or a need
to work less than a full schedule in order to participate in literacy training, drug
rehabilitation, or similar programs in addition to the work program.

Page 21

GAO-12-743 Federal Prison Crowding

Staff

The increased inmate-to-staff ratio can
increase staff overtime and stress and
reduce inmate and staff communication.
This can affect the safety and security of
the institution as a whole.

BOP headquarters officials said that with the primary exception of hiring
staff when a new facility opens, the number of staff positions generally
has not increased as BOP’s population has grown, affecting staff stress
and overtime hours worked. BOP officials explained that BOP is required
to feed, clothe, and provide medical care for inmates. After these costs
are met, BOP funds staffing levels to the extent possible. As a result,
BOP headquarters officials confirmed that overall staffing in BOP facilities
systemwide is on average less than 90 percent of authorized levels,
varying by the facility’s location. For example, a warden may staff
correctional programs at the 97 percent level and the business office at
the 60 percent level. In addition to funding, BOP officials identified
recruitment challenges that affected staffing levels. For example, one
regional director said that finding qualified staff to hire was an ongoing
issue in his region because generally applicants did not meet BOP’s
requirements. At two facilities we visited, officials noted that hiring
professional staff (e.g., psychologists or medical staff) was difficult
because BOP salaries were less than those paid in the community for the
same position.
From fiscal years 1997 through 2011, BOP reported that the systemwide
ratio of inmates to all BOP staff (i.e., at BOP headquarters, regional
offices, institutions, and training centers) increased from 3.57:1 to
4.94:1. 46 This is not the only staffing ratio BOP calculates however. In
addition, BOP calculates a ratio of inmates and institutional staff within its
facilities. 47 From fiscal years 2006 through 2011, the inmate to total
institutional staff ratio for all facilities systemwide and for all male facilities
was generally 5.2:1. In fiscal year 2011, this ratio was lower in high
security facilities (4.1:1) and higher in low and minimum security facilities
46

BOP calculates the ratio for inmates and all BOP staff at the systemwide level; not by
security level. The inmate population is the actual population at each facility on the last
day of the fiscal year. The staffing level is the total number of staff on board as of the last
pay period of the fiscal year and includes all staff funded by BOP’s Salaries and Expenses
appropriations and Public Health Service staff at BOP headquarters, regional offices,
institutions, and, training centers.
47

In fiscal year 2005, BOP began calculating overall institutional staffing ratios on the
basis of (1) all staff under BOP’s Salaries and Expenses appropriations, including
correctional officers, noncustody staff (e.g., teachers, psychologists, and administrative
staff); and Public Health Service staff on board as of the last pay period of the fiscal year;
and (2) the average inmate daily population at each BOP facility. These ratios exclude
BOP Buildings and Facilities, Federal Prison Industries, Commissary, regional office,
training center, and central office staff, as well as staff at facilities that were being
activated.

Page 22

GAO-12-743 Federal Prison Crowding

(6.1:1 and 8.0:1, respectively). Further, BOP calculates a ratio of inmates
to correctional officers. According to BOP, this ratio is 10:1 systemwide,
but it varies depending on security level and mission of the facility. For
example, in one Special Management Unit we visited, officials told us that
there were about 6 inmates to each correctional officer. 48 This contrasted
to a medium security facility where officials told us the ratio of inmates to
correctional officers was 14:1.
According to an August 2010 DOJ study of BOP’s staffing, 49 nearly all
BOP facilities had fewer correctional staff on board than needed, with a
BOP-wide staffing shortage in excess of 3,200. 50 Moreover, even if BOP
filled all authorized positions, the study reported that the shortage would
exceed 1,800. The study team observed that the institutional staff was
very lean, highly functional, and adept at managing large numbers of
inmates at a time, but there was also anecdotal evidence that
understaffing was stressing the workforce. Thus, the study concluded that
the systemwide inmate-to-staff ratio in BOP institutions—5.3:1 in 2009
when the study was prepared—must, at a minimum, be maintained.
Nevertheless, the study stated that BOP’s use of a systemwide ratio had
not been sufficiently effective in justifying additional annual budget
requests, because the ratio did not convey operational realities at the
institutional level. Specifically, the study found that there were variances
in the number of daily correctional officer shifts based on the time of day
and the day of the week that the overall ratio was not incorporating. Our
observations illustrated this point. At one medium security facility we
visited, officials reported a population of about 1,300 inmates, 56 percent
crowding, an inmate to total staff ratio of 6.0:1. Facility officials explained,

48

A Special Management Unit operates as a more controlled and restrictive environment
for inmates whose interaction requires greater management to ensure the safety, security,
or orderly operation of BOP facilities, or protection of the public.
49

Justice Management Division, DOJ, BOP Staffing Study (Washington, D.C.: August
2010).

50
According to the study, BOP uses a staffing roster to determine the number of
correctional officers needed to fill custody posts at its facilities. The roster process
identifies a clear need of correctional services personnel that is typically higher than the
numbers of both funded and authorized positions. The study states that there appears to
be a disconnect between (1) establishing authorized and funded staff positions and (2)
determining the number of correctional officers needed to ensure institutional safety and
security through BOP’s staffing roster process.

Page 23

GAO-12-743 Federal Prison Crowding

however, that about 17 correctional officers were on duty during nights
and evenings to supervise the general population. Thus, in contrast to the
reported ratio, the actual inmate-to-staff ratio during these occasions was
about 76:1. The DOJ study goes on to state that an inmate-to-staff ratio
can provide a valuable perspective when it is used to show how staffing
varies during specific shifts at specific institutions. When ratios are used
in this context, decision makers can more effectively determine the
appropriate number of institution staff needed to safely manage an
institution. Accordingly, for BOP to justify its staffing levels plus additional
resources for increased staff—as the inmate population grows—the study
recommended, among other things, that BOP set a minimum inmate-tostaff ratio that is required to run a safe, secure, and efficient prison
system given operational realties. Partly in response to the DOJ report,
BOP officials said that they had developed minimum staffing guidelines,
and as of June 2012, BOP was in the process of applying these
guidelines at each facility. Tables 14-16 in appendix III illustrate the
various inmate-to-staff ratios and trends over time.
BOP headquarters officials and the union representatives said that
correctional staff worked more overtime hours to meet additional staff
needs as a result of the larger inmate population. 51 Alternatively, in lieu of
paying overtime, facility management may divert other professional or
administration staff, as trained correctional officers, from their primary
duties to supervise other aspects of inmate care and confinement, such
as meal times or medical trips. According to BOP headquarters officials,
this practice, known as augmentation, affects programming. For example,
if a teacher has to fill a correctional post, then the class does not occur or
another teacher may be required to supervise the course. Headquarters,
regional, and facility officials said that they generally used augmentation
during annual correctional officers’ refresher training. For example, during
our site visit to a Special Management Unit, we observed an
administrative staff member serving as a correctional officer in the unit,
replacing a correctional officer who was attending BOP-required annual

51

For fiscal years 2009, 2010, and 2011, BOP reported a total of 1,480,713; 1,416,269.50;
and 1,381,129.50 overtime hours charged by correctional officers in all facilities,
respectively. BOP also maintains data on overtime costs for all BOP staff. Institutional
overtime costs for all BOP regions totaled $102,877,891; $102,352,434; and $89,035,146
for each respective year. BOP officials said that the decline in fiscal year 2011 was due, in
part, to reduced appropriations that year.

Page 24

GAO-12-743 Federal Prison Crowding

staff refresher training. 52 BOP facilities and regional offices have tracked
the use of augmentation, but headquarters did not review or analyze this
information centrally until February 2012, at the direction of the new BOP
Director. Therefore, during the course of our audit work, BOP could not
provide any trend analysis on the use of augmentation systemwide.

Infrastructure

The increased population taxes the infrastructure that was designed for a
smaller inmate population, affecting use of toilets, showers, water, and
electricity, and wear and tear on food service equipment (e.g., freezer
units). According to BOP headquarters and regional officials, crowding
affects the general usage and upkeep of the facility, which affects the
facility itself, the environment, and the local community.
BOP has also experienced increased maintenance and repair costs, with
51 facilities over 30 years old and newer facilities also in need of
maintenance and repair. 53 BOP reported systemwide maintenance and
repair costs of about $228 million in fiscal year 2006 and $262 million in
fiscal year 2011—approximately a 15 percent increase. BOP
headquarters officials stated that they are most concerned with “life safety
issues,” such as ensuring that sprinkler systems work properly in the
event of fire in the facility. These officials said that requests for repairs are
often put off when BOP does not receive funding. (See app. III for
additional information on the effects of BOP’s population growth on
infrastructure.)

Security and Safety

BOP officials said the increasing inmate population and staffing ratios
negatively affect inmate conduct and the imposition of discipline, thereby
affecting security and safety. A 2005 BOP report on the effects of
crowding and staffing levels in federal prisons on inmate violence rates
concluded that population pressures on both staffing levels and inmate
living space have an upward impact on serious prison violence.
Nevertheless, the study also found that systemwide violence rates
remained stable, although measures of both percent rated capacity and

52
BOP established a Special Management Unit at Lewisburg Penitentiary in fiscal year
2008, and subsequently converted the entire facility to a Special Management Unit, with
the exception of a unit housing general population high security inmates.
53

In our previous work (GAO-10-94), we reported that BOP’s methods for estimating costs
in its annual budget requests largely reflect the four best practices outlined in GAO, Cost
Estimating and Assessment Guide: Best Practices for Developing and Managing Capital
Program Costs, GAO-09-3SP (Washington, D.C.: March 2009).

Page 25

GAO-12-743 Federal Prison Crowding

inmate to correctional officers ratios rose in federal prisons during the
latter part of the study.54 The study posits that this stability may stem from
prison managers employing some operational practices, such as
augmentation, that in the short term countered the negative effect of
increased crowding. BOP officials told us that a follow-up to the 2005
study is not necessary because they did not believe that the findings
would change.
BOP generally reported increases in the number of guilty findings for
inmate misconduct from fiscal years 2006 through 2009, but the number
of findings for misconduct of the greatest severity (e.g., killing, serious
assault, and possession of weapons) began to decline in fiscal year 2010.
Additionally, from fiscal years 2006 through 2011, BOP systemwide
imposed almost 4,000 lockdowns––a temporary situation in which all
inmates are confined to their living quarters/cells until staff are able to
assess the situation following a critical incident (e.g., a disturbance,
assaults on staff by several inmates, or a food or work strike) and can
safely return the institution to normal operations. Similar to the inmate
misconduct data, the number of lockdowns increased from fiscal years
2006 through 2009, and then began to decline. Appendix III provides
information on BOP’s disciplinary system and data on inmate misconduct
and lockdowns.
BOP places inmates in Special Housing
Units for disciplinary or administrative
reasons when their presence in the
general inmate population would otherwise
threaten the safety, security, or orderly
operation of the facility or potentially cause
harm to the public. BOP places inmates in
Special Management Units when
inmates need an even more restrictive and
controlled environment—for even longer
terms—than can be offered in a Special
Housing Unit or among the general inmate
population.

BOP officials at all levels told us that they believe the establishment of
Special Management Units beginning in fiscal year 2008 had contributed
to the decrease in misconduct in the general population and the decline in
the use of lockdowns, but that these facilities are too new to evaluate.55
Nevertheless, BOP officials stated that its Special Management Units are
now crowded and experiencing waiting lists.56 Specifically, BOP reported
that from March 1, 2012, through April 20, 2012, 231 inmates were
approved for Special Management Unit placement and were awaiting a

54

The study used calendar quarter data for 73 federal low, medium, and high security all
male prisons from July 1996 through December 2004, a period of increased prison
crowding and increased inmate to correctional officer ratios.
55
We have ongoing work that focuses on BOP’s Special Management Units and Special
Housing Units. We expect to publish our results in early 2013. As part of our review, we
are analyzing their effects on inmate misconduct.

56

As of January 31, 2012, BOP reported housing 1,664 male inmates in its Special
Management Units.

Page 26

GAO-12-743 Federal Prison Crowding

bed. The average wait time for placement in a Special Management Unit
bed was 110 days.
Officials said that waiting lists for transfers to a Special Management Unit
contribute to crowding in the facility Special Housing Unit. According to
BOP officials, without space for disciplinary segregation, they are limited
in how they can address inmate misconduct. Officials further stated that
when a facility has no Special Housing Unit space available, the regional
office may move the inmate to a Special Housing Unit in another facility of
a different security level—a practice referred to as trans-segregation.
Alternatively, headquarters officials said that disciplinary hearing officers
may dispense shorter time in segregation or use other sanctions or a
combination of nonsegregation sanctions. As a result, the officials said
that the imposed sanctions may not be as much of a deterrent with the
inmates, which affects the security and safety of inmates and staff.
BOP officials said they have both
experienced and effective staff, but that
they are reaching the highest crowding
rates ever and have increasingly
unfunded repair requests. They said that
BOP cannot keep operating as it is
without new capacity.

Additionally, BOP headquarters officials and union representatives we
spoke with expressed concerns about future effects of increased inmate
population growth. First, officials raised concern about the possibility of a
serious incident occurring, especially at a high security or medium level
facility. A serious incident could occur in a high security facility because
these facilities are extremely crowded and house the most serious
inmates (i.e., those who have committed the most serious crimes in
society or in prison). A union representative also said that medium
security facilities were at risk of an incident because these facilities lack
the better lockdown procedures found in high security facilities.
Nevertheless, BOP officials did not discount an incident happening at a
low security facility because of the high gang presence in these facilities.
They said that although the criminal histories of low security inmates
suggest that they are not a “high risk” for violence, these inmates may still
be a high risk for problems because of frustrations resulting from crowded
conditions. Second, BOP officials were also concerned that the federal

Page 27

GAO-12-743 Federal Prison Crowding

courts might require BOP to address conditions related to crowding or
that ACA might revoke the accreditation of BOP institutions. 57

BOP Has Worked to
Increase Inmate and Staff
Safety and Security and
Has Reported Utilizing
Resources More Efficiently
Efforts to Increase Inmate and
Staff Safety and Security

BOP has taken actions to manage a growing population within its
facilities—and its approaches were similar to those in selected states we
reviewed. These have generally been aimed at
•

increasing inmate and staff safety and security and

•

utilizing resources efficiently.

Controlled inmate movement. BOP has implemented controlled
movement for inmates, which is a practice that officials from one of the
five states we reviewed also reported using, specifically to deal with
crowded conditions. For example, because of crowded conditions, one
way that BOP restricts inmates’ movement in high and medium security
facilities is by instituting earlier in-cell hours at night for inmates. 58 Further,
BOP has a system of inmate movement in place to reduce potential
tension and fighting and allow staff to better supervise inmates (i.e.,
staggering activities or meal times so that one cell block or unit of inmates
proceeds at a time). Like officials at BOP, officials from Mississippi’s
Department of Corrections stagger recreational activities to curtail inmate
fighting so that only one cell block or unit is released to the yard at a time.
Disciplinary housing. As previously discussed, because escalating
tensions in crowded facilities can cause increased security concerns,
BOP utilizes Special Housing Units to segregate inmates involved in

57

In May 2011, the United States Supreme Court held in the case of Brown, Governor of
California, v. Plata, 131 S.Ct. 1910, that a court-mandated prison inmate population limit
was necessary to remedy the violation of a federal right, specifically the severe and
unlawful mistreatment of prisoners through grossly inadequate provision of medical and
mental health care. The Court recognized that for years the medical and mental health
care provided by California’s prisons had fallen short of minimum constitutional
requirements and had failed to meet prisoners’ basic health needs with needless suffering
and death being the well-documented result. The Court stated that overcrowding had
overtaken the limited resources of prison staff, imposed demands well beyond the
capacity of medical and mental health facilities, and created unsanitary and unsafe
conditions that made progress in the provision of care difficult or impossible to achieve.
58

Controlled movement in high security facilities is often called restricted movement.

Page 28

GAO-12-743 Federal Prison Crowding

disciplinary infractions. 59 Officials in Kansas and New York also reported
using disciplinary housing. BOP officials stated that the use of Special
Housing Units has resulted in a decrease in inmate misconduct because
those inciting tension within the general population have been removed. 60
According to a 2006 New York State Department of Correctional
Services’ report on prison safety, the department believed that the
certainty of facing Special Housing Unit confinement for misbehavior
contributed generally to improved inmate conduct, as reflected in inmates
spending less time in these units without reductions in time to make room
for other inmates. 61
Preferential housing. To increase the safety and security of inmates and
staff, BOP has encouraged positive behavior from inmates by rewarding
them with preferential housing, and this was a practice we observed
directly in a New York State Department of Corrections and Community
Supervision facility. BOP officials described preferential housing as cells
or dormitory rooms with fewer inmates than the facility’s general
population housing and located close to the phones or showers. BOP
union officials stated that, at some facilities, preferential housing is given
to inmates directly in response to their good behavior, and officials
explained that inmates highly coveted these rewards and modified their
behavior accordingly. For example, in one facility we visited, inmates in
the cleanest unit were rewarded by getting to eat meals before inmates in
other units. At one state facility we visited in New York, officials rewarded
well-behaved inmates by allowing them to live in the “honor block”—a
preferential housing unit that allows inmates more freedom of movement
and additional personal decision making (e.g., inmates are allowed to
decide when they wish to shower and do their laundry).
Expanded program options and incentives. To accommodate growing
inmate populations, reduce inmate idleness, and help inmates prepare for
life outside of prison—all of which relate to institutional safety—BOP

59
As of January 31, 2012, BOP reported housing 11,624 male inmates and 179 female
inmates in Special Housing Units.
60

As part of our review on BOP’s Special Management Units and Special Housing Units,
we are analyzing their effects on inmate misconduct. We expect to publish our results in
early 2013.

61

New York State Department of Correctional Services, Prison Safety in New York
(Albany, New York: April 2006).

Page 29

GAO-12-743 Federal Prison Crowding

officials have expanded inmates’ program options. Officials from all five of
the states we selected reported similar activities. Generally GED classes
are held during the day, but, for example, two BOP facilities we visited
have begun offering evening GED classes to accommodate the increase
in inmates who are required to receive a GED education.62 Also, one BOP
facility we visited had expanded its vocational training capacity by
combining woodshop learning with classroom study, so that one group of
inmates could be learning in the woodshop while another group of
inmates could be participating in classroom lessons. According to the
program director, when he ran the program from 2001 through 2005,
about 30 inmates received certification each year for completing the
program; currently, about 60 to 80 inmates receive certification annually.
At one facility we visited, to encourage program participation, inmates in
the Special Management Unit were given cash incentives for completing
psycho-educational programs, such as stress/anger management classes
or those designed to improve interpersonal relationships and help
inmates focus on personal goals and maintaining positive conduct.
Specifically, an inmate may earn $25 for completing these types of
classes.63 Like BOP, both Ohio and Wisconsin have offered additional
programming in the form of expanded program hours and expanded
reentry programs; for example, classes on financial literacy, housing, and
personal health care to teach inmates who are about to be released how
to manage their daily lives in the community.

Efforts to Utilize Resources
Efficiently
BOP’s Federal Correctional Complex in
Florence, Colorado, comprises three
secure facilities: an Administrative
Maximum U.S. Penitentiary, a high
security U.S. Penitentiary, and a medium
security Federal Correctional Institution.
The complex also includes a minimum
security satellite camp

Correctional complexes. BOP has established correctional complexes
over the last 15 years to better leverage its staff. Officials from one state
in our sample—Kansas—told us they employ this practice as well. BOP
correctional complexes are institutions that are located on the same
grounds and may include low, medium, and high security facilities.
According to BOP officials, the use of correctional complexes helps in
particular with the leveraging of medical services and supplies. It also

62
According to BOP Program Statement 5350.28, generally an inmate confined in a
federal institution who does not have a verified GED credential or high school diploma is
required to attend an adult literacy program for a minimum of 240 instructional hours or
until a GED is achieved, whichever occurs first.

63

This program is funded by the BOP Inmate Trust Fund, which is maintained by profits
from inmate purchases of commissary products, telephone services, and the fees inmates
pay for using the inmate computer system. See GAO, Bureau of Prisons: Improved
Evaluations and Increased Coordination Could Improve Cell Phone Detection,
GAO-11-893 (Washington, D.C.: Sept. 6, 2011).

Page 30

GAO-12-743 Federal Prison Crowding

helps BOP manage crowding by sharing staff resources across the
correctional complex, if one facility has greater needs than another for
certain programs.
Energy conservation. BOP has also taken actions to minimize the
burden that crowding places on facilities’ infrastructure, and officials from
one state in our sample—Ohio—acknowledged similar activities.
According to BOP officials, BOP has aggressively pursued energy
conservation following a 2009 governmentwide executive order to reduce
energy usage.64 BOP officials stated that their energy-saving efforts have
prevented BOP from experiencing a dramatic increase in energy usage
despite the growing prison population. Examples of energy-saving actions
that BOP officials reported include the installation of slower-flowing
shower heads, which use 2 rather than 5 gallons of water per minute and
flushing toilets every 5 to10 minutes rather than after each use. According
to BOP officials, most BOP facilities have recycling programs. One BOP
facility that we visited began a recycling program that, in addition to
efficiently utilizing resources, created inmate jobs and benefitted the
environment. Ohio officials told us that they have reduced utility costs with
similar efficiency initiatives.
Visitor accommodations. BOP has taken a variety of actions to
accommodate the increased number of visitors within existing
infrastructure, which is similar to the steps officials from two of the five
states we reviewed. For example, at one BOP facility we visited, because
the facility did not have money to enlarge the visiting room, it shortened
the length of visits from 4 hours to 2 hours and changed visiting hours
from an open schedule (i.e., where visitors can come during any visiting
hours) to a rotating basis (i.e., visitors for a particular inmate may be
allowed to visit on certain days and between certain hours). In an effort to
supplement face-to-face visits, BOP has permitted the increased use of email between inmates and their loved ones.65 Ohio and Wisconsin have
implemented scheduled visitation times rather than open visitation hours.

64
Federal Leadership in Environmental, Energy, and Economic Performance, Exec. Order
No. 13514, 74 Fed. Reg. 52,117 (Oct. 5, 2009).
65
An inmate is permitted to exchange electronic messages only with persons who have
accepted the inmate’s request to communicate. For more information on BOP’s use of
electronic messaging, see GAO-11-893.

Page 31

GAO-12-743 Federal Prison Crowding

In addition, Ohio officials told us that the state offers e-mail as a way to
supplement visitation.

States Have Taken
Broader Actions
Intended to Reduce
Prison Populations
than Those Taken at
the Federal Level

While BOP and the five selected states have taken a variety of similar
actions to manage the growing number of inmates they incarcerate, these
states have been able to take broader actions than BOP to reduce their
prison populations because these states have legislative authority that
BOP does not have. 66 These states’ actions can be grouped into three
general categories:
•

modifying criminal statutes and sentencing,

•

relocating inmates (e.g., moving them from state to local facilities or
community corrections, whereby their release is supervised at halfway
houses or in-home detention), or

•

providing inmates with good time credit or adjusting inmates’
sentences based on participation in certain programs or
demonstration of positive behaviors.

To take these actions, these state departments of corrections have
generally worked with their state legislatures to propose and pass
legislation that effects these changes. Officials from three of the five
states we spoke with—Kansas, Ohio, and Wisconsin—told us their states
also embarked on justice reinvestment efforts to facilitate legislative or
other changes to their corrections approaches. 67 For example, in 2006,
Kansas sought technical assistance through the Justice Reinvestment

66

Actions taken by states and discussed in this section have been intended to reduce
prison populations. However, a variety of factors and circumstances (e.g., new drug
sentencing laws and a drop in drug-related crime) may also have contributed to decreases
in prison populations.
67

The Justice Reinvestment Initiative is administered by DOJ’s Bureau of Justice
Assistance, Office of Justice Programs, in coordination with related efforts supported by
independent organizations (e.g., the Pew Center on the States). It provides technical
assistance and competitive financial support to states and localities engaged in or well
positioned to consider different investments in their justice and law enforcement dollars.
When considering reinvestment, states and localities collect and analyze data on drivers
of criminal justice populations and costs, identify and implement changes to increase
efficiencies, and measure both the fiscal and public safety impacts of any changes.
Wisconsin participated in the initiative in 2008, but was no longer participating in the
initiative at the time of our review.

Page 32

GAO-12-743 Federal Prison Crowding

Initiative to avert an increase of 700 new inmates in its prison population
that it projected between 2007 and 2010. Using the justice reinvestment
approach, as described below, Kansas (1) relocated inmates from staterun facilities by diverting them to nonprison alternatives and transferring
them to county jails and community corrections facilities and (2) offered
inmates credit for positive behavior thereby reducing inmates’ time in
prison. As a result, the state experienced a net increase of 10 inmates
rather than the 700 inmates it had anticipated during this period.
Officials in the five selected states generally believed that the actions
taken had helped them to reduce their prison populations; however,
because these initiatives were recent, empirical data showing the impact
of these initiatives were generally not available. In contrast, federal law
does not provide BOP with the authority to implement many of these
measures and generally requires BOP to provide for suitable housing and
the safekeeping, care, and subsistence of all persons charged with or
convicted of offenses against the United States. 68 Unlike certain states’
laws, federal law does not provide BOP with the authority to transfer
inmates to local prisons or move inmates to community corrections or
supervised release beyond what current federal law permits. 69
Additionally, because of the mandatory minimum sentences required for
many federal offenses and the absence of parole for most federal inmates
in the federal system, BOP generally does not have the authority to
significantly modify an inmate’s period of incarceration. 70

Modifying Criminal
Statutes and Sentencing

Two of the five states we reviewed have changed their sentencing
statutes or guidelines. For example, in 2009, New York implemented

68

18 U.S.C. § 4042(a)(2). See GAO-12-320.

69
The Second Chance Act of 2007, Pub. L. No. 110-199, § 251(a), 122 Stat. 657, 692-93,
amended 18 U.S.C. § 3624(c) to enable BOP to place inmates in community corrections
for up to 12 months and home detention for the shorter of 10 percent of the term of
imprisonment or 6 months.
70
According to the U.S. Parole Commission, offenders who are under the supervision of
the commission and eligible for parole include inmates currently incarcerated for federal
offenses committed prior to the Sentencing Reform Act of 1984; all District of Columbia
offenders, as of August 5, 2000; the U.S. military prison population that has been
transferred to federal correctional institutions; a few cases of Americans who have
committed a crime in a foreign country; and offenders in the federal witness protection
program.

Page 33

GAO-12-743 Federal Prison Crowding

changes to its drug statutes, which affected the sentencing of some drug
felony offenders. These changes included revising the ranges for state
prison sentences by lowering the minimum sentence allowable for certain
nonviolent drug felony offenders. New York has reported decreases in its
total custody population since 1999, when the population reached 71,472
and drug offenders constituted 31.2 percent of this population. From the
end of calendar year 2009 through 2011, New York reported a decrease
in its total custody population from 58,378 to 55,090. This decrease
included not only a decline in the number of drug offenders from 10,319
to 7,509 but also in the percentage of drug offenders in custody from 17.7
percent to 13.6 percent. In 2011, Ohio revised its sentencing laws to
eliminate the differences between the penalties for crack and powder
cocaine violations. Generally, the change provides for a uniform
determination of the penalty for drug offenses based upon the amount of
any type of cocaine (powder cocaine or any compound, mixture,
preparation, or substance containing cocaine) an individual possesses. In
effect, this change resulted in an increasing of the amount of crack
needed to those of powder cocaine for lower-level offenses and a
decreasing of the amounts of powder cocaine needed to crack levels for
higher-level offenses.
While states may change their sentencing statutes or guidelines to reduce
their prison populations, at the federal level, BOP does not determine
which offenders are sentenced to prison and what the length of their
sentences should be. On May 1, 2007, the U.S. Sentencing Commission
submitted to Congress amendments to the federal sentencing
guidelines. 71 These guidelines became effective on November 1, 2007. 72
One of the amendments modified the drug quantity thresholds for crack

71
Under 28 U.S.C. § 994(a), the commission is to promulgate and distribute to all of the
courts of the United States and to the United States Probation System guidelines for use
of a sentencing court in determining the sentence to be imposed in a criminal case,
including (1) whether to impose a sentence to probation, a fine, or a term of imprisonment
and (2) the appropriate amount of a fine or the length of a term of probation or
imprisonment, among other things. Pursuant to 28 U.S.C. § 994(p), generally, the
commission is to submit to Congress amendments to the guidelines and modifications to
previously submitted amendments that have not taken effect.
72
The federal sentencing guidelines provide federal judges with a set of consistent
sentencing ranges to consult when determining a sentence. The guidelines consider both
the seriousness of the criminal conduct and the defendant’s criminal record. Federal
courts must consult the sentencing guidelines and take them into account when
sentencing, but are not bound to apply the guidelines.

Page 34

GAO-12-743 Federal Prison Crowding

cocaine offenses. Generally, the commission lowered the sentencing
guidelines for certain crack cocaine offenses. Subsequently, the
commission made the amendment apply retroactively. As a result, some
incarcerated offenders were eligible to receive a reduction in sentence
under 18 U.S.C. § 3582(c)(2). 73 The effect of this change was realized
almost fully in fiscal year 2008. 74

Relocating Inmates

Selected states have also relocated inmates to relieve the crowding of
their state prison facilities, such as through use of nonprison alternatives
(e.g., drug treatment programs), expanded use of parole, the movement
of inmates to county or local jails, or the wider reliance on community
corrections (e.g., halfway houses). For example, Mississippi expanded
use of house arrest and Kansas and New York expanded drug
rehabilitation programs as an alternative to incarceration for certain lowlevel drug offenders. Other selected states have passed legislation that
allows some inmates to be paroled sooner and made parole available to
more inmates. For example, Mississippi has passed legislation that
extended parole eligibility to (1) all nonviolent offenders irrespective of the
offender’s first-time offender status and (2) certain drug sale offenders.
New York allows well-behaved drug and other nonviolent offenders to
appear before a parole board earlier. Additionally, Kansas and Wisconsin
have moved some low-risk inmates to county jails for more localized
management and relief for state prison crowding.
Three of the five selected states also reported using community
corrections—also known as supervised release at halfway houses or inhome detention—as a way to either divert offenders from prison or move
more inmates out of prison. For example, Ohio uses halfway houses, as
is typical, to provide supervision and treatment services to inmates who
are released from state prison or are sentenced to halfway houses by

73

See GAO-12-320.

74

The change to the sentencing guidelines for crack cocaine went into effect on November
1, 2007. As of June 2011, the Sentencing Commission reported that of the 25,736 inmate
applicants for a sentence reduction, 16,511 (64.2 percent) had been granted their
requests. Eligible inmates received an average sentence reduction of 26 months. The
Sentencing Commission was able to determine the origin of the motion for 15,016 of the
inmates who were granted a sentence reduction. U.S. Sentencing Commission, U.S.
Sentencing Commission Preliminary Crack Cocaine Retroactivity Data Report, June 2011
(Washington, D.C.: June 2011).

Page 35

GAO-12-743 Federal Prison Crowding

courts for an offense or as a result of violating parole. 75 Like Ohio, Kansas
has diverted parole and probation violators to supervised release.
Similarly, Mississippi uses an intensive supervised release program as an
alternative to incarceration for low-risk and nonviolent offenders.
At the federal level, BOP uses Residential Reentry Centers (RRC) to help
inmates reintegrate into the community, as well as to reduce crowding in
prisons, but is limited by what federal law allows as well as the capacity of
these facilities. 76 BOP may place inmates there in the final months of their
sentences (not to exceed 12 months) under conditions that will afford the
inmate a reasonable opportunity to adjust to and prepare for reentry into
the community. BOP officials stated that BOP maximizes the use of
RRCs to the extent possible but that there are not enough beds in RRCs
to accommodate all eligible inmates. In February 2012, we reported that
as of November 2011, BOP estimated 8,859 available RRC beds under
contract. 77 For each available RRC bed, BOP can transfer one inmate to
the RRC for a maximum of 12 months, or BOP could send multiple
inmates for shorter placements (e.g., three inmates for 4 months each).
To provide all eligible inmates with the maximum allowable 12 months in
an RRC, BOP would require about 29,000 available beds annually. 78
Further, BOP places inmates according to a court’s or judge’s sentence,
which may require some probation and supervised release violators to
serve terms in community corrections.

Providing Inmates with
Credit for Positive
Behaviors

Selected states have taken a variety of actions that reduce inmates’ time
in prison by providing inmates with credit for positive behaviors. For
example, in Mississippi, certain inmates may be eligible to receive a trusty
time allowance of 30 days’ reduction of sentence for each 30 days’
participation during any calendar month in approved programs while in

75
In Ohio, generally parole is used for inmates convicted of crimes prior to July 1, 1996,
when Ohio was under an indefinite sentencing structure. Most offenders who committed
crimes after July 1, 1996, serve definite sentences, with a period of postrelease
supervision for certain crimes upon their release from prison.
76

BOP refers to halfway houses as RRCs.

77

GAO-12-320.

78
GAO, Federal Bureau of Prisons: Methods for Estimating Incarceration and Community
Corrections Costs and Results of the Elderly Offender Pilot, GAO-12-807R (Washington,
D.C.: Jul. 27, 2012).

Page 36

GAO-12-743 Federal Prison Crowding

trusty status. 79 According to Mississippi officials, these programs include
alcohol and drug treatment, GED classes, faith-based programs, and
vocational education. New York and Kansas allow inmates who complete
certain rehabilitative programs to be released earlier, either through
earned compliance credits in Kansas or by receiving a parole hearing
earlier in New York. New York officials stated that since 1998,
approximately 37,000 inmates have been released because of this policy.
Ohio has an earned credit program that rewards an inmate for productive
participation in educational programs, vocational training, prison
industries work, substance abuse treatment, or any other constructive
program with specific performance standards. 80 Additionally, in Ohio,
there is a mechanism for the possible release with sentencing court
approval of certain department of corrections inmates who have served at
least 80 percent of their prison terms. 81
Officials from two of the five selected states also reported that their states
award inmates credit toward the service of their sentence for good
behavior—compliance with institutional disciplinary regulations—as a way
to relieve prison crowding. For example, Kansas’s department of
corrections offers good time credits and is authorized to adopt rules and
regulations providing for a system of good time calculations. The system
provides circumstances under which an inmate may earn good time
credits and for the forfeiture of earned credits. 82
We have previously reported on BOP’s use of its sentence reduction
authority, noting that it is affected by both inmate eligibility and BOP

79
Pursuant to Miss. Code Ann. § 47-5-138.1, a trusty time allowance is a reduction in
sentence that may be granted in addition to any other administrative reduction in sentence
to an offender in trusty status as defined by the classification board of the Department of
Corrections. In this instance, “trusty” refers to those inmates who are eligible to receive an
allowance of 30 days reduction of sentence for each 30 days of their participation during
any calendar month in an approved program
80

Ohio Rev. Code Ann. § 2967.193.

81

Ohio Rev. Code Ann. § 2967.19.

82
Pursuant to Kan. Stat. Ann. § 21-6821, generally, the good time credit, which can be
earned by an inmate and subtracted from any sentence, is limited to 15 percent of the
prison part of the sentence for a crime committed on or after July 1, 1993, and increased
to 20 percent of the prison part of the sentence for certain crimes committed on or after
January 1, 2008.

Page 37

GAO-12-743 Federal Prison Crowding

capacity. 83 Specifically, BOP’s RDAP offers sentence reductions of up to
1 year to inmates convicted of a nonviolent offense who successfully
complete the program. BOP officials told us that they strive to maximize
RDAP programs as a management tool to reduce recidivism and because
of the program’s ability to reduce an inmate’s sentence. RDAP programs
are full, however, and BOP cannot keep up with demand for RDAP
enrollment, which limits BOP’s ability to fully leverage this program.
Additionally, because of long waiting lists, those eligible for a sentence
reduction are generally unable to complete RDAP in time to earn the
maximum reduction. Furthermore, federal law provides for the amount of
time awarded for each inmate who successfully completes the program.
Moreover, according to BOP officials, given BOP’s staffing shortages and
in the absence of additional funding for the program, BOP has generally
been unable to increase the number of RDAP’s staff to accommodate
more inmates.
Further, as we reported in February 2012, BOP is authorized to award up
to 54 days of good conduct time credit each year (which vests on the date
the inmate is released). 84 Good conduct time credit may be given to an
inmate serving a sentence of more than 1 year, but less than life. BOP’s
method of awarding good conduct time credit at the end of each year an
inmate serves results in a maximum of 47 days earned per year of
sentence imposed. 85 From fiscal years 2009 through 2011, BOP data
show that about 87 percent of inmates had earned all of their available
good conduct time credit by the end of each year, and an additional 3
percent of inmates earned at least 90 percent of the maximum available
good conduct time credit. 86
Some inmates have contested BOP’s methodology in court, maintaining
that allowing inmates 54 days per year of sentence imposed was the

83

GAO-12-320.

84

GAO-12-320.

85
As authorized in statute, 18 U.S.C. § 3624(b), BOP awards “up to 54 days at the end of
each year of the prisoner’s term of imprisonment,” or 54 days per year of sentence served.
As applied by BOP, this results in 47 days earned per year of sentence imposed because
inmates do not earn good conduct time for years they do not ultimately serve because of
being released early.
86

BOP tracks inmates’ earned good conduct time credit throughout their terms of
imprisonment.

Page 38

GAO-12-743 Federal Prison Crowding

original intent of the statute, 87 but the U.S. Supreme Court upheld BOP’s
approach. BOP officials told us that the agency was supportive of
amending the statute related to good conduct time credit, and legislation
pending before Congress would allow for 54 days to be provided for each
year of the term of imprisonment originally imposed by the judge, which
would result in inmates serving 85 percent of their sentence. 88 BOP
provided us with estimates in December 2011 showing that if the good
conduct time credit allowance was increased from 47 to 54 days, as
proposed, BOP could save over $40 million in the first fiscal year after the
policy change from the early release of about 3,900 inmates. As of July
2012, the legislative proposal has been introduced in the Senate but not
the House. 89 BOP officials told us that they are examining initiatives that
would allow for the restoration of good conduct time, but that they are
reluctant to pursue them. They explained that loss of good conduct time is
one of the most powerful sanctions in BOP’s inmate discipline program,
which helps ensure the safety, security, and orderly operation of
correctional facilities.

Concluding
Observations

Over the last 25 years, BOP’s population has grown more than 400
percent, and BOP projects future growth through 2020. With more
inmates, BOP’s spending to secure, feed, and provide services to a
growing population has also been rising. BOP’s annual appropriation now
exceeds $6.6 billion, and represents nearly a quarter of DOJ’s annual
budgetary authority. Despite the continued growth in inmates and related
expenses, in recent years, BOP has been adding capacity and staff at a
lower rate than the inmate population has been growing. As a result, both

87

Under the Sentencing Reform Act, the U.S. Sentencing Commission established
sentencing guidelines with the understanding that inmates would receive good conduct
time credit so that their actual time served would be 85 percent of the length of the
sentence imposed by the judge, assuming good behavior. BOP’s method of awarding
good conduct time, however, results in inmates serving more than 85 percent of their
imposed sentences, even after earning the maximum good conduct time credit.
88

The additional credit would be awarded retroactively to inmates sentenced under the
Sentencing Reform Act prior to the legislative change. For the hypothetical inmate with a
10-year sentence, the inmate would receive a total of 540 days of good conduct time.
Thus the inmate would serve 3,110 days (85 percent) of the 3,650 days sentence.
89

Second Chance Reauthorization Act of 2011, S.1231, 112th Cong. § 4(f) proposes to
amend certain statutory provisions related to good conduct time in 18 U.S.C. § 3624(b)(1).
As of August 2012, the legislative proposal has been introduced in the Senate but not the
House.

Page 39

GAO-12-743 Federal Prison Crowding

individual facilities and the federal prison system as a whole are
experiencing increased crowding. Crowding has implications for inmates,
staff, and infrastructure—as well as safety and security, and the potential
for inmate disruptions or an even more serious security incident is a
significant concern. BOP has taken steps to help mitigate the implications
of crowding in the federal system, but does not have the authority to
implement many of the reforms that several states have adopted to
reduce crowding and, in some states, the size of their prison populations.
BOP also requires congressional approval and appropriated funds to
expand capacity in the federal system. As such, BOP has limited ability to
address crowding in the federal prison system. We are not taking a
position on matters of policy such as how crowding in the federal system
should be addressed. However, as policy makers weigh whether and how
to address crowding in the federal system, options that will be important
to consider include (1) reducing the size of the projected inmate
population by reforming sentencing laws, allowing alternatives to
incarceration, and/or providing BOP greater sentencing flexibility; (2)
increasing capacity in the federal system by constructing new prisons,
contracting for additional private capacity, and adding additional staff; or
(3) taking some combination of both approaches.

Page 40

GAO-12-743 Federal Prison Crowding

Agency Comments

We provided a draft of this report to DOJ for official review and comment.
BOP provided technical clarifications, which we incorporated where
appropriate.
We are sending copies of this report to the Attorney General, selected
congressional committees, and other interested parties. In addition, this
report is available at no charge on the GAO website at
http://www.gao.gov.
If you or your staff have any further questions about this report, please
contact me at (202) 512-9627 or maurerd@gao.gov. Contact points for
our Offices of Congressional Relations and Public Affairs may be found
on the last page of this report. Key contributors to this report are listed in
appendix IV.

David C. Maurer
Director
Homeland Security and Justice

Page 41

GAO-12-743 Federal Prison Crowding

Appendix I: Objectives, Scope, and
Methodology
Appendix I: Objectives, Scope, and
Methodology

Our objectives for this report were to address the following questions:
1. What was the growth in the Bureau of Prison’s (BOP) population from
fiscal years 2006 through 2011, and what are BOP’s projections for
inmate population and capacity?
2. What is known about the effects of a growing federal prison
population on operations (i.e., inmates, staff, and infrastructure) within
BOP facilities, and to what extent has BOP taken actions to mitigate
these effects?
3. What actions have selected states taken to reduce their prison
populations, and to what extent has BOP implemented similar
initiatives?
To address the first question, we analyzed BOP’s statutory authority and
policies and procedures (e.g., BOP’s inmate classification policy) that
potentially affect growth in the federal prison population and conditions of
confinement in BOP facilities. We also analyzed BOP’s (1) inmate
population data (e.g., demographics and offenses), (2) 2020 long-range
capacity plan based on inmate population projections and future capacity
estimates depending on funding, 1 (3) percentage crowding at all
institutional security levels, (4) staff-to-inmate ratios, and (5) available
infrastructure costs (e.g., water and electricity costs). Unless otherwise
noted, all of these data covered the period from fiscal years 2006 through
2011. We also reviewed Department of Justice (DOJ) and BOP reports
describing BOP’s population and staffing during this period. We assessed
the reliability of BOP’s inmate population data and crowding data by
reviewing relevant documentation, interviewing knowledgeable agency
officials about how they maintain the integrity of their data, and updating
assessments that we did for previously issued reports. We found BOP’s
inmate population and crowding data to be sufficiently reliable for the
purposes of this engagement.
To determine how BOP developed its population and capacity projection
estimates, we analyzed BOP’s program statements, performance goals,
and congressional budget submissions for fiscal years 2011 through 2013.
We also interviewed BOP headquarters officials to discuss the extent to
which BOP’s population has grown; the reasons for this growth; how BOP
1

We analyzed BOP’s 2020 long-range capacity plan dated January 10, 2012.

Page 42

GAO-12-743 Federal Prison Crowding

Appendix I: Objectives, Scope, and
Methodology

calculates the percentage crowding in its facilities; and how BOP develops
its population growth and capacity projections, including any changes to
this process since our November 2009 report. 2 In that report, we assessed
how BOP developed its population projections and capacity plans. We
compared BOP’s projections with its actual inmate population growth from
fiscal years 1999 through August 20, 2009, and concluded that BOP’s
projections were accurate, on average, to within 1 percent of the actual
inmate population growth within this time period. We also reviewed
government and academic studies on federal incarceration determinants.
To address the second question, we analyzed BOP’s statutory authority,
policies, and procedures pertinent to the effects of the growing prison
population on BOP operations, including inmates, staff, and infrastructure,
and that may affect BOP’s ability to mitigate the effects of a growing
population. Further, we analyzed BOP studies on the effects of population
growth and crowding on BOP operations. We also analyzed data
provided by BOP on available bed space including temporary bed space
for all security levels, 3 inmate program participation and waiting lists,
inmate-to-staff ratios, and available infrastructure costs. Unless otherwise
noted, these data covered the period from fiscal years 2006 through
2011. We also present systemwide BOP staffing ratios from fiscal years
1997 through 2011 because officials believed that presenting the ratios
for a longer period better illustrates the effect of BOP’s population growth
relative to the number of staff. 4 We assessed the reliability of BOP’s
inmate, staff, and infrastructure data by interviewing knowledgeable
agency officials to determine how BOP collects and maintains the
integrity of these data. We found these data to be sufficiently reliable for
the purposes of this report. In addition, we reviewed BOP’s fiscal year
2011 congressional budget submission to identify past actions to address
federal prison population growth and crowding, including any proposed
legislative changes. To observe some of the effects of a growing federal
prison population and crowding on current BOP inmates, staff, and
2

See GAO-10-94.

3

BOP reports the use of additional cots in areas such as an institution’s halls, gyms,
mezzanines, or television rooms—to address crowding in an institution—as temporary
housing because this temporary living space is to be restored as program space when
circumstances permit. Thus, such temporary use of space is not factored into BOP’s rated
capacity calculation.

4

According to BOP officials, BOP also includes this information in its annual congressional
budget request.

Page 43

GAO-12-743 Federal Prison Crowding

Appendix I: Objectives, Scope, and
Methodology

infrastructure, we conducted visits to 5 of BOP’s 117 institutions. We
chose these prisons on the basis of varying security levels and to ensure
geographic dispersion. As shown in table 1, the five prisons we visited
were located in four of BOP’s six regions. Additionally, to identify
variations in the effects of increased prison populations, we selected
facilities of different security levels (i.e., low, 1; medium, 3; high, 1; and
administrative, 1); the Petersburg Complex included a medium and a low
security facility.
Table 1: Site Visits to BOP Facilities
Facility name

Region

Security level

Petersburg Complex

Mid-Atlantic Region

Low, medium

SeaTac

Western Region

Administrative—mixed security levels
including men and women

Lewisburg

Northeast Region

High, Special Management Unit

Schuylkill

Northeast Region

Medium

Leavenworth

North Central Region

Medium (previously high)

Source: GAO analysis of BOP information.

During each site visit, we interviewed institutional management officials
and toured the facility to observe inmate housing, recreational areas, food
service, medical services, and educational and vocational programming.
Because we did not visit all BOP facilities and did not randomly select the
facilities we visited, our results are not generalizable to all BOP facilities.
Nevertheless, these results provided us with examples of the effects of
BOP’s population growth on a facility’s inmates, staff, and infrastructure,
as well as examples of actions taken at the facility level to mitigate these
effects. We also interviewed the six regional directors to obtain their
perspectives on the increased prison population and the effects of this
growth and crowding on BOP institutions within each region. Further, we
discussed the effects of BOP’s population growth on correctional officers
with officials from the Council of Prison Locals, the union that represents
all non-management staff working in BOP facilities. Additionally, we
analyzed American Correctional Association’s (ACA) standards (e.g.,
minimum inmate space standards) and reviewed ACA audits from fiscal
years 2009 through 2011 (the most recent audits available) of the BOP
institutions we visited to try to identify potential effects of growing

Page 44

GAO-12-743 Federal Prison Crowding

Appendix I: Objectives, Scope, and
Methodology

populations in these facilities. We also met with ACA officials to identify
any areas where BOP might not be meeting these standards. 5
Further, to determine the extent to which BOP has taken actions to mitigate
the effects of a growing federal prison population, we analyzed BOP’s
statutory authority to identify provisions that affect BOP’s ability to mitigate
the effects of the growth of the prison population. We also analyzed BOP’s
policies, DOJ and BOP studies, and BOP’s fiscal year 2011 congressional
budget submission to identify actions BOP has taken, including any
proposed legislative changes that could mitigate the effects of the growth of
the prison population. We interviewed BOP headquarters officials to obtain
information on these actions and proposals as well as to discuss the extent
of their statutory authority, which affects their ability to mitigate the effects
of the increased population. During our BOP site visits and interviews with
BOP regional directors, we asked officials to identify any actions taken at
the facility or regional level to mitigate the effects of the growth of the
federal prison population and crowding in facilities in the region. Also, we
discussed actions taken by BOP to mitigate the effects of prison population
growth with officials from the Council of Prison Locals. We also interviewed
corrections experts from DOJ’s National Institute of Corrections (NIC), 6
ACA, Pew Center on the States, and academia. We selected these experts
from our review of the corrections literature and on the recommendation of
BOP officials and other experts. While the views of these experts are not
representative, they provided us with perspectives on BOP’s actions and
ability to mitigate the effects of its increased prison population.
To also address the second as well as the third questions, we compared
and contrasted BOP’s actions to (1) mitigate the effects of its increased
population and (2) attempt to reduce its prison population with similar
actions taken by five states––Kansas, Mississippi, New York, Ohio, and

5
ACA’s mission includes the development and promotion of effective standards for the
care, custody, training, and treatment of offenders. As part of its accreditation process, a
visiting committee of ACA auditors (1) audits the corrections agency or correctional facility
against standards and expected practices documentation and (2) evaluates the quality of
life or conditions of confinement. An acceptable quality of life is necessary for an agency
to be eligible for accreditation. The quality of life in a facility includes staff training, cell size
and time inmates spend outside the cells, current population, adequacy of medical
services, offender programs, recreation, food service, classification, sanitation, use of
segregation, crowding, and reported and/or documentation of incidents of violence.
6

NIC, a component of BOP, provides training, technical assistance, information services, and
policy/program development assistance to federal, state, and local corrections agencies.

Page 45

GAO-12-743 Federal Prison Crowding

Appendix I: Objectives, Scope, and
Methodology

Wisconsin––that had experienced prison population growth and had
taken actions to mitigate its effects or reduce their prison populations. To
select these states, we analyzed DOJ’s Bureau of Justice Statistics’s
(BJS) report on 2010 state prison inmate populations. 7 We also reviewed
relevant governmental and nongovernmental reports on state prison
population growth and states’ actions taken to reduce or mitigate the
effects this growth, which were published from 2006 through 2011 (e.g.,
the Pew Center on the States, Vera Institute, and Council of State
Governments). We also interviewed (1) BOP, NIC, and BJS officials; (2)
stakeholder interest groups (e.g., Pew Center on the States, Council of
State Governments, ACA, and Association of State Correctional
Administrators); and (3) academic corrections experts to obtain their
perspectives on state efforts to mitigate the effects of prison population
growth or reduce prison populations.
We selected these five states because they (1) reflected a range of prison
population sizes (e.g., New York and Ohio have two of the largest state
prison populations, with 56,656 and 51,712 inmates, respectively); (2)
were involved in addressing prison crowding issues (e.g., Kansas and
Ohio are working with the Council of State Governments and the Pew
Center on the States to address prison crowding issues as part of the
Justice Reinvestment Initiative); and (3) had taken actions to address
population growth, including actions similar to BOP’s actions. 8
For each of the five selected states, we obtained available data (e.g., BJS
state correctional population) and reviewed relevant studies on prison
conditions for context. We then interviewed state corrections officials

7

See BJS, Prisoners in 2010 (Washington, D.C.: December 2011) and Prisoners in 2009
(Washington, D.C. December 2010). BJS is the statistical agency of DOJ.

8

The Justice Reinvestment Initiative, which is administered by DOJ’s Bureau of Justice
Assistance in the Office of Justice Programs in coordination with related efforts supported
by independent organizations (e.g., the Pew Center on the States), provides technical
assistance and competitive financial support to states and localities engaged in or well
positioned to undertake justice reinvestment. The purpose of justice reinvestment is to
manage and allocate criminal justice populations more cost-effectively, generating savings
that can be reinvested in evidence-based strategies that increase public safety while
holding offenders accountable. States and localities engaging in Justice Reinvestment
Initiative activities collect and analyze data on drivers of criminal justice populations and
costs, identify and implement changes to increase efficiencies, and measure both the
fiscal and public safety impacts of those changes. According to Wisconsin officials,
Wisconsin participated in the initiative in 2008, but was no longer participating in the
initiative at the time of our review.

Page 46

GAO-12-743 Federal Prison Crowding

Appendix I: Objectives, Scope, and
Methodology

regarding actions taken in their states to reduce the prison population, as
well as to mitigate the effects of prison population growth. We analyzed
correctional statutes and policies identified by these officials during our
interviews. As shown in table 2, we also conducted site visits to three
facilities in two states.
Table 2: Site Visits to State Correctional Facilities
State

Facility name

Security level

Kansas

Lansing

Maximum, medium, and minimum

New York

Edgecombe

Low (parole violators; substance abuse)

New York

Sing Sing

Maximum

Source: GAO, based on state data.

During our site visits, we discussed with correctional administrators and
observed actions taken within these facilities to mitigate the effects of the
growth of the prison populations.
Additionally, to further address the third question, we assessed the extent
to which actions implemented in the five states to reduce their prison
populations would be possible under current federal law for BOP to
implement. Dissimilarities between federal and state prison systems—
legally, structurally, and in how crowding calculations are determined––
limit the comparability between federal and state correctional systems.
We mitigated the effects of these limitations by the criteria used to select
the five states. We are unable to generalize about the types of actions
states have taken to mitigate the effects of state prison population growth
or reduce their prison populations. Nevertheless, the information we
obtained through these visits provided examples of state responses to
prison population growth. Further, to determine the extent to which BOP
has implemented similar initiatives from states’ experiences, we analyzed
BOP documentation describing BOP initiatives to address the growth of
federal prison populations. We also discussed with BOP officials the state
actions we identified, including the extent to which these actions would be
possible under current federal law for BOP to implement.
We conducted this performance audit from September 2011 to
September 2012 in accordance with generally accepted government
auditing standards. Those standards require that we plan and perform the
audit to obtain sufficient, appropriate evidence to provide a reasonable
basis for our findings and conclusions based on our audit objectives. We
believe that the evidence obtained provides a reasonable basis for our
findings and conclusions based on our audit objectives.

Page 47

GAO-12-743 Federal Prison Crowding

Appendix II: BOP’s Population Growth
Appendix II: BOP’s Population Growth

This appendix presents additional information and elaboration on BOP’s
population growth. Specifically, it addresses

Factors Contributing to
Growth

•

factors contributing to growth,

•

offense composition of BOP’s population,

•

BOP’s rated capacity and percentage crowding trends, and

•

BOP’s long-range capacity plan.

Among the factors that contribute to the size of the federal prison
population are national crime levels, law enforcement policies, and
federal sentencing laws, all of which are beyond BOP’s control. According
to BOP officials, the length of the sentences that federal inmates serve is
one of the single most important factors in prison population growth. Prior
to the passage of the Sentencing Reform Act of 1984, most federal
statutes provided only for broad maximum terms of imprisonment and
federal judges had broad discretion in sentencing. 1 Federal law outlined
the maximum sentence, federal judges imposed a sentence within a
statutory range, and federal parole officials eventually determined the
actual duration of incarceration. The Sentencing Reform Act of 1984
abolished parole, and subsequent legislation established mandatory
minimum sentences for many federal offenses; these federal laws limit
the ability of BOP to affect the length of the sentence or the size of the
inmate population.
According to BOP, the increase in sentence length is the primary reason
for the growth in the federal inmate population from 42,000 in 1987 to
over 218,000 today. Drug offenses constitute the largest component of
admissions to BOP. The average time an inmate served for drug offenses
increased 250 percent after 1987, when the U.S. Sentencing Commission
revised the U.S. Sentencing Guidelines in order to implement the

1
The Sentencing Reform Act of 1984, Pub. L. No 98-473, 98 Stat. 1987, changed the
federal sentencing structure. The act was effective for offenses committed on or after
November 1, 1987.

Page 48

GAO-12-743 Federal Prison Crowding

Appendix II: BOP’s Population Growth

Sentencing Reform Act of 1984. 2 Most recently, the Administrative Office
of the U. S. Courts reported to BOP that for the year ending March 31,
2012, federal courts had ordered over 8,000 sentences of 6 or more
years, and nearly 3,000 of those prison sentences were for 12 or more
years. 3 Beyond drug offenses, BOP reported that length of sentence for
almost all federal offenders had increased since the implementation of the
act. For example, immigration offenders currently serve much longer
sentences than they did in the 1980s.

Offense Composition of
BOP’s Population

As shown in table 3, drug, weapons/explosives, and immigration offenses
composed the largest number of offenses for which all BOP inmates were
incarcerated in each year from fiscal years 2006 through 2011. 4

2

The U.S. Sentencing Commission is an independent agency in the judicial branch of
government. Its principal purposes are (1) to establish sentencing policies and practices
for the federal courts, including guidelines to be consulted regarding the appropriate form
and severity of punishment for offenders convicted of federal crimes; (2) to advise and
assist Congress and the executive branch in the development of effective and efficient
crime policy; and (3) to collect, analyze, research, and distribute a broad array of
information on federal crime and sentencing issues.
3

The Administrative Office of the U.S. Courts is the central support entity for the judicial
branch. It provides a wide range of administrative, legal, financial, management, program,
and information technology services to the federal courts.

4

BOP officials explained that for reporting purposes they categorize inmates according to
the offense for which an inmate is serving the longest sentence (dominant sentence
offense). For example, an inmate may be serving sentences for both drug and immigration
offenses, but BOP will categorize the inmate by the offense having the longer sentence
(e.g., the drug offense).

Page 49

GAO-12-743 Federal Prison Crowding

Appendix II: BOP’s Population Growth

Table 3: Offense Composition of the Inmate Population in BOP Facilities, by Year, from Fiscal Years 2006 through 2011
Fiscal year
Offense

2006

2007

2008

2009

2010

2011

Drugs

91,690

95,321

94,456

96,415

96,094

97,164

Weapons/explosives

27,036

28,901

30,256

31,454

32,188

33,136

Immigration

18,992

19,862

20,165

22,298

21,767

24,800

Fraud/bribery/extortion

8,426

9,268

10,229

10,493

10,930

11,462

Sex offenses

3,824

4,754

5,989

7,308

8,610

9,653

Robbery

9,428

9,099

8,965

8,712

8,493

8,289

Burglary/larceny

6,381

6,417

6,499

6,694

6,747

7,136

Homicide/aggravated assault

4,895

4,942

4,940

4,894

4,900

5,114

Court/corrections

2,174

2,209

2,192

2,180

2,183

2,200

Miscellaneousa

2,374

2,298

2,305

2,281

2,064

2,006

Counterfeiting/embezzlement

1,066

1,016

1,017

945

928

948

Continuing criminal enterpriseb

430

416

412

390

374

364

National security

108

103

105

98

97

95

176,824

184,606

187,530

194,162

195,375

202,367

Total offender population

Source: GAO analysis of BOP data.

Notes: Sentencing information is not available for all inmates, for example, pretrial inmates whose
information has not been entered into the data system.
a

Miscellaneous offenses include criminal civil rights violations; food and drug violations; economic
espionage; destruction of an energy facility; District of Columbia offenses, such as driving while
intoxicated, malicious mischief, and contributing to the delinquency of a minor; and violations of fish
and game laws.
b

Continuing criminal enterprise refers to sentencing under the Racketeer Influenced and Corrupt
Organizations Act (RICO) Act (codified at 18 U.S.C. § 1961), for example, drug kingpins.

BOP explained that the offense composition of BOP’s population
generally shows a higher number of drug offenses than immigration
offenses, because drug offenses carry longer sentences than immigration
offenses. For example, BOP data show that the number of drug offenses
in fiscal year 2011 was four times greater than the number of immigration
offenses that year, although the yearly admission to BOP for each of
these offenses is usually about the same. Further, the number of
admissions for weapons offenses is generally about one-fourth that of
immigration offenses, but the number of weapons offenders in the BOP
population is significantly higher than the number of immigration offenders
because of the much longer sentences for weapons offenses.
The distribution of offenses varied by institutional security level, although,
with the exception of high security facilities, drug violations constituted the

Page 50

GAO-12-743 Federal Prison Crowding

Appendix II: BOP’s Population Growth

largest number of offenses. In medium and low security facilities, drugs,
weapons/explosives, and immigration were among the three largest
offense categories. In minimum security facilities, drug offenses were
followed by fraud/bribery/extortion and weapons/explosives offenses. 5 In
high security facilities, weapons/explosives offenses constituted the
largest number of offenses, followed by drugs, robbery,
homicide/aggravated assault, and immigration.
As shown in table 4, drug and immigration offenses accounted for the
largest numbers of offenses among non-U.S. citizen inmates in each year
of the 6-year period. Violent and property offenses increased during the
same period, but were a smaller number of the total offenses.
Table 4: Offense Composition of the Non-U.S. Citizen Inmate Population in BOP Facilities from Fiscal Years 2006 through
2011
Year
Dominant offense

2006

2007

2008

2009

2010

2011

Drug offenses

19,980

21,450

21,501

21,690

20,931

22,135

Immigration offenses

15,509

16,819

16,570

18,227

17,978

21,451

Violent offenses

1,908

2,086

2,288

2,387

2,481

2,696

Property offenses

1,047

1,190

1,470

1,423

1,429

1,532

Miscellaneous

1,505

1,615

1,656

1,621

1,655

1,746

Source: GAO analysis of BOP data.

Note: Drug offenses include the distribution of narcotics to a minor and possession of narcotics.
Immigration offenses include illegal entry and illegal reentry into the United States. Violent offenses
include assault, homicide, bank robbery, and firearms/weapons/explosive violations. Property
offenses include embezzlement, auto theft, larceny, and destroying government property. Other
offenses include sex offenses (e.g., obscene mailing and white slavery); justice system offenses (e.g.,
perjury, obstruction of justice, and jumping bail); racketeering; general offenses (e.g., bribery,
extortion, and failure to pay child support); and regulatory offenses (e.g., customs law violations,
espionage, sabotage, and violations of national defense laws), among other things.

During each year of the 6-year period, the largest number of non-U.S.
citizen inmates (ranging from about 28,000 to 33,000) were incarcerated in
low security facilities for drug offenses. Among non-U.S. citizen inmates
incarcerated in medium and high security facilities, immigration, followed by
drugs, constituted the largest number of offenses. As to why these inmates
were housed in medium and high security facilities, BOP officials explained

5

BOP does not send non-U.S. citizen inmates to minimum security facilities because of
their risk of flight.

Page 51

GAO-12-743 Federal Prison Crowding

Appendix II: BOP’s Population Growth

that offenders who were convicted of violent offenses, had a history of
violent behavior, or had been found guilty of serious misconduct while in
BOP custody were usually sent to higher security level facilities. These
officials said that over 45 percent of non-U.S. citizen inmates who were
placed in higher security level facilities had a history of violence.

BOP’s Rated Capacity and
Percentage Crowding
Trends

Table 5 below shows BOP’s rated capacity and double, triple, and
quadruple bunking levels as of September 30, 2011.

Table 5: BOP’s Rated Capacity and Crowding, by Facility Security Level as of September 2011

Facility
type

Actual inmates
double bunked

Actual inmates
triple bunked

Actual inmates
quadruple
bunked

Percentage
crowding

37%

0

14%

81%

0

37%

Rated
capacity

Actual
population

Minimum

18,476

21,091

13,246

63%

7,845

Low

32,242

44,174

8,384

19%

35,790

Medium

41,039

61,908

40,338

65%

21,570

35%

0

51%

Highb

13,570

20,978

20,244

97%

0

0%

0

55%

Minimum

4,207

5,086

2,449

48%

2,637

52%

0

21%

Secure

3,808

5,866

0

0%

5,250

89%

616

54%

Number Percentage

Number Percentage

Malea

Femalec

Source: GAO analysis of BOP data.

Notes: According to BOP, rated capacity is the maximum population level at which an institution can
make available basic necessities, essential services (e.g., medical care), and programs (e.g., drug
treatment, basic education, and vocational education). BOP calculates rated capacity only for the
prisons that it operates; therefore, private institutions are excluded. Further, BOP does not calculate
rated capacity for residential reentry centers or inmates in home confinement because BOP does not
assign security levels to these confinement arrangements. The capacity figures used to calculate
percentages and determine double and triple occupancy for this table are from BOP end of fiscal year
2011 historical information from BOP’s 2020 long range capacity plan.
a

Male long-term institutions include four security level designations––minimum, low, medium, and
high.
b

The remaining 3 percent of inmates in high security facilities are single-bunked.

c

Female facilities include three security designations––high, secure, and minimum, but female high
security facilities are single bunked.

Page 52

GAO-12-743 Federal Prison Crowding

Appendix II: BOP’s Population Growth

As a result of the growth in the inmate population in BOP-run facilities
relative to the increased rated capacity, crowding in BOP-run institutions
increased from 36 to 39 percent systemwide from fiscal years 2006
through 2011. Nevertheless, within male facilities, the percentage
crowding varied by security level, as shown in figure 4. For example, the
percentage crowding in male medium security facilities increased from 37
percent to 51 percent, and from 53 percent to 55 percent in high security
level facilities (see table 6). Additionally, the percentage crowding in
minimum security facilities more than doubled from fiscal year 2009
through 2010 because of a population increase of more than 1,400
inmates while capacity increased by 69 beds.
Figure 4: Percentage Crowding in Male Long-Term Facilities from Fiscal Years 2006
through 2011 by Institutional Security Level

Page 53

GAO-12-743 Federal Prison Crowding

Appendix II: BOP’s Population Growth

During the 6-year period, the overall percentage crowding in female longterm facilities decreased from 39 to 36 percent, but there were some
variations by facility security level (i.e., minimum, secure, and high).
Specifically, the percentage crowding in secure facilities decreased from
73 to 54 percent and the percentage crowding in minimum security
facilities increased from 17 to 21 percent. BOP headquarters officials
explained that BOP was able to increase the number of female secure
facility beds by converting a male facility at Waseca, Minnesota, to a
female facility. 6
Table 6 shows BOP’s male and female populations, rated capacity, and
percentage crowding data by institutional security level for fiscal years
2006 through 2011.

6

BOP officials explained that BOP was able to convert the male facility because it had
opened a new male facility and converted older male high security facilities to medium
security facilities, thereby increasing the number of beds available for male inmates.

Page 54

GAO-12-743 Federal Prison Crowding

Appendix II: BOP’s Population Growth

Table 6: BOP Population, Rated Capacity, and Percentage Crowding from Fiscal Years 2006 through 2011
2006

2007

2008

2009

Security
level of
facility
Total
inmates in
BOP-run
institutions
Males

Pop
162,514

Cap
119,510

151,003

111,067

36%

155,254

113,111

37%

153,992

113,288

36%

160,315

Long-Term

133,331

97,439

37%

137,195

100,170

37%

136,138

100,361

36%

142,005

Minimum

20,046

16,963

18%

20,003

17,271

16%

18,556

17,353

7%

18,904

Low

43,723

31,037

41%

41,646

30,791

35%

41,825

31,081

35%

45,153

%Cro
36%

Pop
167,323

Cap
122,189

%Cro
37%

Pop
165,964

Cap
122,366

%Cro
36%

Pop
172,423

Cap
125,778

2010

2011

%Cro
37%

Pop
173,289

Cap
126,713

%Cro
37%

Pop
177,934

Cap
127,795

%Cro
39%

116,050

38%

161,295

117,171

38%

165,595

118,596

40%

102,867

38%

143,059

104,003

38%

148,151

105,327

41%

17,768

6%

20,329

17,837

14%

21,091

18,476

14%

32,294

40%

44,264

32,302

37%

44,174

32,242

37%

Medium

51,972

37,911

37%

56,492

39,680

42%

55,543

38,479

44%

57,904

39,375

47%

57,708

40,294

43%

61,908

41,039

51%

High

17,590

11,528

53%

19,054

12,428

53%

20,214

13,448

50%

20,044

13,430

49%

20,758

13,570

53%

20,978

13,570

55%

462

490

-6%

476

490

-3%

476

490

-3%

458

490

-7%

444

490

-9%

451

490

-8%

Florence
a
ADX
Marion

0

804

0%

0

0

0%

0

0

0%

0

0

0%

0

0

0%

0

0

0%

2,715

2,646

3%

2,642

2,657

-1%

2,622

2,604

1%

2,683

2,604

3%

2,505

2,487

1%

2,501

2,517

-1%

Detention

14,085

9,349

51%

14,521

9,455

54%

14,335

9,494

9,721

52%

14,877

9,823

51%

14,085

9,878

43%

410

339

21%

420

339

24%

421

339

51%
24%

14,741

Witness
b
security
Females

428

368

16%

410

368

11%

407

384

6%

11,511

8,443

36%

12,069

9,078

33%

11,972

9,078

32%

12,108

9,728

24%

11,994

9,542

26%

12,339

9,199

34%

Long-Term

10,060

7,253

39%

10,691

7,847

36%

10,690

7,847

36%

10,730

8,497

26%

10,691

8,311

29%

10,971

8,039

36%

Minimum

5,192

4,429

17%

5,362

4,429

21%

5,134

4,429

16%

5,026

4,479

12%

5,003

4,479

12%

5,086

4,207

21%

Secure

4,853

2,800

73%

5,311

3,394

56%

5,537

3,394

63%

5,690

3,994

42%

5,671

3,808

49%

5,866

3,808

54%

15

24

-38%

18

24

-25%

19

24

-21%

14

24

-42%

17

24

-29%

19

24

-21%

393

378

4%

574

378

52%

466

378

378

38%

481

378

27%

435

378

15%

812

30%

804

853

-6%

816

853

23%

521

1,058

857

853

0%

822

853

-4%

933

782

19%

Medical

High
Medical
Detention

-4%

Legend:
Pop refers to inmate population.
Cap refers to rated capacity of the facility.
% Cro refers to the percentage crowding.
Source: BOP.

Notes: BOP designates some of its institutions as administrative institutions, which specifically serve inmates awaiting trial, or those with intensive medical or
mental health conditions, regardless of the level of supervision these inmates require.
a

The Administrative Maximum (ADX) facility in Florence, Colorado, houses offenders requiring the tightest controls.

b

Witness security refers to BOP housing for inmates in the federal Witness Security Program.

Page 55

GAO-12-743 Federal Prison Crowding

Appendix II: BOP’s Population Growth

BOP’s 2020 Long-Range
Capacity Plan

BOP’s 2020 long-range capacity plan projects continued growth in the
federal prison population from fiscal years 2012 through 2020. 1 The plan
relies on multiple approaches to house the increased federal prison
population, including contracting with the private sector for certain inmate
populations; expanding existing institutions where infrastructure permits;
and acquiring, constructing, and activating new facilities as funding
permits. BOP officials explained that BOP changes its capacity plan
several times each year. For example, rated capacity figures may change
as a result of the reclassification of a facility to address population needs.
BOP also adjusts its actual capacity and population figures each year.
As shown in figure 5, BOP expects the overall inmate population in BOPrun institutions to continue to grow from approximately 182,600 inmates in
fiscal year 2012 to about 204,410 inmates in fiscal year 2020. The plan
also projects an increase in systemwide capacity from 128,433 beds in
2012 to 151,895 beds in 2020, with a projected reduction in crowding
from 42 percent to 35 percent.

1

These projections are from BOP’s 2020 capacity plan dated January 10, 2012. In
November 2009, we concluded that BOP’s projections were accurate, on average, to
within 1 percent of the actual inmate population growth from fiscal year 1999 through
August 20, 2009. See GAO-10-94.

Page 56

Appendix II: BOP’s Population Growth

Figure 5: BOP’s Projections for Population, Rated Capacity, and Percentage Crowding from Fiscal Years 2012 through 2020

Specifically, BOP’s 2020 capacity plan projects the male inmate
populations in long-term institutions at all security levels to continue to
grow or remain stable; however, BOP projects variations in the changes
in rated capacity across the security levels, resulting in respective
differences in crowding percentages. For example, as shown in figure 6,
from fiscal years 2017 through 2020, BOP’s plan projects a decrease in
crowding in male medium security facilities from 71 percent to 58 percent
and in high security facilities from 55 percent to 12 percent, as a result of
the projected increased capacity. 2 Congressional budget requests have
2
BOP’s 2020 long-range capacity plan, dated January 10, 2012, projects an increased
capacity of 5,808 (43,239 to 49,047) beds and increased population of 3,812 (73,737 to
77,549) inmates in male medium security facility between fiscal years 2017 and 2020. In
male high security facilities, the plan projects an increased capacity of 7,200 (15,485 to
22,685) beds and increased population of 1,320 (24,012 to 25,332) inmates from fiscal
years 2017 and 2020.

Page 57

Appendix II: BOP’s Population Growth

not included funding for this additional bed space, and as a result BOP’s
plans are contingent on the budget development and appropriations
processes and are subject to change. During the same period, BOP also
projects crowding in low security facilities to be about 35 percent.
Figure 6: Projected Percentage Crowding in Male Long-Term Facilities from Fiscal
Years 2012 through 2020 by Security Level

Page 58

Appendix II: BOP’s Population Growth

BOP’s plan also projects an increase in the female long-term inmate
population from fiscal years 2012 through 2020, but anticipates variations
in percentage crowding across security levels. Specifically, the
projections show an increase in the percentage crowding in female
minimum security facilities from 24 to 37 percent and a decrease in the
percentage crowding in female secure facilities from 58 to 32 percent.
BOP noted that this reduction in crowding is contingent on appropriations
to fund the opening of a secure female facility in Aliceville, Alabama, in
fiscal years 2013 and 2014. According to BOP officials, BOP’s projections
do not include any additional long-term facilities for women, because the
percentage crowding in female facilities is well below the percentage
crowding in, for example, male medium security facilities. Nevertheless,
officials said that BOP will review and adjust capacity needs as it
develops its plan beyond fiscal year 2020.
Table 7 shows BOP’s rated capacity, population, and percentage
crowding projections from fiscal years 2012 through 2020. According to
BOP officials, BOP’s long-term population projections are always
conservative; therefore, the actual number of inmates would likely be
higher than the projections. Additionally, they said that other factors that
may affect the accuracy of these projections include legislation affecting
federal crimes and sentencing as well as new law enforcement initiatives
leading to more arrests, prosecutions, and convictions.

Page 59

Appendix II: BOP’s Population Growth

Table 7: BOP Projected Population, Rated Capacity, and Percentage Crowding from Fiscal Years 2012 through 2014
Security level of
facility
Total BOP inmates
BOP institutions
Males
Long-term
Minimum
Low
Medium
High
Florence ADXa
Medical
Detention
Witness securityb
Females
Long-term
Minimum
Secure
High
Medical
Detention
Contract
Males
Juveniles
Residential Reentry
Centers
Long-term
(criminal aliens)
Taft
D.C. felons
Jail-detention
Home Confinement
Females
Juveniles
Residential Reentry
Centers
Long-term
D.C. felons
Jail-detention
Home Confinement

2012
Pop
222,768
182,624
169,921
152,414
21,795
43,105
65,993
21,521
441
2,566
14,085
415
12,703
11,322
5,285
6,017
20
448
933
40,144
38,330
144
7,609

2013

Cap

% Cro

128,433
119,177
105,923
18,604
32,211
40,583
14,525
490
2,552
9,828
384
9,256
8,096
4,264
3,808
24
378
782

42%
43%
44%
17%
34%
63%
48%
-10%
1%
43%
8%
37%
40%
24%
58%
-19%
19%
19%

Pop
229,268
188,051
174,958
157,353
22,597
44,045
68,501
22,210
444
2,647
14,085
429
13,093
11,697
5,470
6,207
20
462
933
41,218
39,334
148
7,609

2014

Cap

% Cro

131,435
120,973
107,719
18,796
32,211
42,137
14,575
490
2,552
9,828
384
10,462
9,302
4,470
4,808
24
378
782

43%
45%
46%
20%
37%
63%
52%
-9%
4%
43%
12%
25%
26%
22%
29%
-16%
22%
19%

Pop
233,765
190,187
176,873
159,202
22,816
43,824
69,873
22,690
444
2,704
14,085
438
13,315
11,909
5,550
6,339
21
472
933
43,578
41,630
152
8,215

24,186

25,186

26,686

2,347
0
1,783
2,261
1,814
21
1,219

2,347
0
1,783
2,261
1,883
21
1,288

2,347
0
1,783
2,448
1,948
22
1,317

0
0
208
366

0
0
208
366

0
0
208
401

,,,

Page 60

Cap

% Cro

134,175
123,177
109,923
18,988
32,211
43,239
15,485
490
2,552
9,828
384
10,998
9,838
4,470
5,344
24
378
782

42%
44%
45%
20%
36%
62%
47%
-9%
6%
43%
14%
21%
21%
24%
19%
-14%
25%
19%

Appendix II: BOP’s Population Growth

Table 7: BOP Projected Population, Rated Capacity, and Percentage Crowding from Fiscal Years 2015 through 2017
Security level of
facility
Total BOP inmates
BOP institutions
Males
Long-term
Minimum
Low
Medium
High
Florence ADXa
Medical
Detention
Witness securityb
Females
Long-term
Minimum
Secure
High
Medical
Detention
Contract
Males
Juveniles
Residential Reentry
Centers
Long-term
(criminal aliens)
Taft
D.C. felons
Jail-detention
Home Confinement
Females
Juveniles
Residential Reentry
Centers
Long-term
D.C. felons
Jail-detention
Home Confinement

2015
Pop
237,884
192,567
179,074
161,344
23,260
43,755
71,198
23,131
443
2,755
14,085
446
13,493
12,079
5,598
6,459
21
481
933
45,317
43,336
155
8,371

2016

Cap

% Cro

134,175
123,177
109,923
18,988
32,211
43,239
15,485
490
2,552
9,828
384
10,998
9,838
4,470
5,344
24
378
782

44%
45%
47%
22%
36%
65%
49%
-9%
8%
43%
16%
23%
23%
25%
21%
-13%
27%
19%

Pop
242,003
194,947
181,220
163,430
23,704
43,687
72,467
23,572
444
2,806
14,085
454
13,727
12,304
5,703
6,580
21
490
933
47,055
45,042
157
8,527

2017

Cap

% Cro

134,675
123,677
109,923
18,988
32,211
43,239
15,485
490
2,552
10,328
384
10,998
9,838
4,470
5,344
24
378
782

45%
47%
49%
25%
36%
68%
52%
-9%
10%
36%
18%
25%
25%
28%
23%
-11%
30%
19%

Pop
246,122
197,327
183,366
165,517
24,149
43,619
73,737
24,012
443
2,858
14,085
463
13,962
12,529
5,807
6,700
22
499
933
48,794
46,747
160
8,683

28,186

29,686

31,186

2,347
0
1,783
2,495
1,981
22
1,342

2,347
0
1,783
2,541
2,014
22
1,367

2,347
0
1,783
2,588
2,047
23
1,392

0
0
208
408

0
0
208
416

0
0
208
424

Page 61

Cap

% Cro

135,175
124,177
109,923
18,988
32,211
43,239
15,485
490
2,552
10,828
384
10,998
9,838
4,470
5,344
24
378
782

46%
48%
51%
27%
35%
71%
55%
-10%
12%
30%
20%
27%
27%
30%
25%
-9%
32%
19%

Appendix II: BOP’s Population Growth

Table 7: BOP Projected Population, Rated Capacity, and Percentage Crowding from Fiscal Years 2018 through 2020
Security level of
facility
Total BOP inmates
BOP institutions
Males
Long-term
Minimum
Low
Medium
High
Florence ADXa
Medical
Detention
Witness securityb
Females
Long-term
Minimum
Secure
High
Medical
Detention
Contract
Males
Juveniles
Residential Reentry
Centers
Long-term
(criminal aliens)
Taft
D.C. felons
Jail-detention
Home Confinement
Females
Juveniles
Residential Reentry
Centers
Long-term
D.C. felons
Jail-detention
Home Confinement

2018
Pop
250,241
199,710
185,514
167,605
24,593
43,551
75,010
24,452
443
2,909
14,085
471
14,196
12,755
5,912
6,821
22
508
933
50,533
48,453
163
8,840

2019

Cap

% Cro

138,111
127,113
112,859
19,884
32,211
43,839
16,925
490
2,552
10,828
384
10,998
9,838
4,470
5,344
24
378
782

45%
46%
49%
24%
35%
71%
44%
-10%
14%
30%
23%
29%
30%
32%
28%
-8%
34%
19%

Pop
254,360
202,088
187,659
169,691
25,037
43,482
76,279
24,893
442
2,961
14,085
479
14,430
12,980
6,016
6,941
23
517
933
52,272
50,159
166
8,996

2020

Cap

% Cro

145,871
134,873
120,619
21,932
32,211
46,191
20,285
490
2,552
10,828
384
10,998
9,838
4,470
5,344
24
378
782

39%
39%
41%
14%
35%
65%
23%
-10%
16%
30%
25%
31%
32%
35%
30%
-6%
37%
19%

Pop
258,479
204,470
189,805
171,777
25,482
43,414
77,549
25,332
443
3,012
14,085
488
14,665
13,205
6,121
7,062
23
526
933
54,010
51,864
169
9,152

32,686

34,186

35,686

2,347
0
1,783
2,634
2,080
23
1,417

2,347
0
1,783
2,681
2,113
24
1,442

2,347
0
1,783
2,727
2,146
24
1,467

0
0
208
431

0
0
208
439

0
0
208
446

Page 62

Cap

% Cro

151,895
140,897
126,643
22,700
32,211
49,047
22,685
490
2,552
10,828
384
10,998
9,838
4,470
5,344
24
378
782

35%
35%
36%
12%
35%
58%
12%
-10%
18%
30%
27%
33%
34%
37%
32%
-4%
39%
19%

Appendix II: BOP’s Population Growth

Legend:

Pop refers to inmate population.
Cap refers to rated capacity of the facility.
% Cro refers to the percentage crowding.
Source: BOP.

Notes: BOP designates some of its institutions as administrative institutions, which
specifically serve inmates awaiting trial, or those with intensive medical or mental health
conditions, regardless of the level of supervision these inmates require.
a
The Administrative Maximum (ADX) facility in Florence, Colorado, houses offenders
requiring the tightest controls.
b
Witness security refers to BOP housing for inmates in the federal Witness Security
Program.
Assumptions:
1. The population projections for fiscal year 2012 and beyond have been adjusted.
2. BOP will activate Federal Correctional Institution Mendota, California, and Federal Correctional
Institution Berlin, New Hampshire, during fiscal year 2012.
3. BOP will activate Federal Correctional Institution Hazelton, West Virginia, during fiscal year 2013.
4. BOP will activate U.S. Penitentiary Yazoo City, Mississippi, during fiscal year 2013.
5. No additional contract beds are projected to be added in fiscal year 2012.
6. One thousand additional contract beds are projected in fiscal year 2013, and 1,500 additional
contract beds annually in fiscal year 2014 and beyond.
7. No increase to Residential Reentry Centers and Home Confinement is projected from fiscal years
2011 through 2013.
8. Future capacity increases are dependent upon future funding to construct new prisons and to
expand the use of private prison contracts.

Page 63

Appendix III: Effects of a Growing Inmate
Population
Appendix III: Effects of a Growing Inmate
Population

This appendix presents additional information and elaboration on the
effects of a growing inmate population on
•

inmates,

•

staff,

•

infrastructure, and

•

security and safety.

Inmates

The growth in the inmate population has affected inmates’ daily living
conditions, program participation, meaningful work opportunities, and
visitation.

Daily Living

BOP uses double bunking in excess of the percentages included in a
facility’s rated capacity; triple bunking or converting common space (e.g.,
a television room) temporarily to house its growing population. BOP
counts these additional beds as temporary space rather than increased
rated capacity. 1 As a result of these actions to increase available bed
space, more inmates are sharing cells and other living units, bringing
together for longer periods of time inmates with a higher risk of violence
and more potential victims. 2 Table 8 shows the temporary bed space
BOP added from fiscal years 2006 through 2011 by security level and
male and female facilities. During the 6-year period, the use of temporary
space generally increased in male medium, and high security facilities
and in female minimum security facilities, but the number of temporary
beds fluctuated with changes in the number of general population beds
(e.g., rated capacity).

1

BOP reports the use of additional cots in areas such as an institution’s halls, gyms,
mezzanines, or television rooms—to address crowding in an institution—as temporary
housing because (1) this temporary living space is to be restored as program space when
circumstances permit and (2) the additional bathrooms and other facilities required to
meet permanent housing space specifications are not added to the infrastructure. Thus,
such temporary use of space is not factored into BOP’s rated capacity calculation.

2

BOP. The Effects of Changing Crowding and Staffing Levels in Federal Prisons on
Inmate Violence Rates (Washington, D.C.: October 2005).

Page 64

GAO-12-743 Federal Prison Crowding

Appendix III: Effects of a Growing Inmate
Population

Table 8: Rated Capacity and Temporary Bed Space by Institutional Security Level from fiscal years 2006 through 2011
Year
Facility
type

2006
RC

2007

2008

2009

2010

Tem

RC

Tem

RC

Tem

RC

Tem

RC

2011
Tem

RC

Tem

18,476

2,562

a

Male

Minimum 16,963

3,030

17,271

2,679

17,353

1,150

17,768

1,083

17,837

2,439

Low

31,037

11,083

30,791

9,252

31,081

9,078

32,294 11,306

32,302

10,409

32,242 10,329

Medium

37,911

10,178

39,680 12,894

38,479 13,146

39,375 14,611

40,294

13,376

41,039 16,831

High

11,528

4,367

12,428

4,781

13,448

4,830

13,430

4,678

13,570

5,252

13,570

5,472

Femaleb
Minimum

4,429

737

4,429

862

4,429

679

4,479

521

4,479

498

4,207

853

Secure

2,800

1,947

3,394

1,811

3,394

2,081

3,994

1,540

3,808

1,707

3,808

1,902

Legend:
RC refers to rated capacity.
Tem refers to temporary bed space, not including disciplinary housing beds.
Source: GAO analysis of BOP data.

Notes: Total bed space, not including disciplinary housing beds, is the sum of RC plus Tem.
a

Male long-term institutions include four security level designations––minimum, low, medium, and
high. Female facilities include three security designations––high, secure, and minimum.
b

Female high security facilities did not experience crowding during the 6-year period.

BOP officials told us that they were aware of the use of temporary beds in
BOP’s 117 institutions, but BOP does not track whether a facility uses
television rooms or triple bunks in a cell. Instead, when temporary beds
are added, BOP generally refers to this as triple bunking. As shown in
table 9, all of the BOP facilities we visited reported using temporary beds
during the period from fiscal years 2006 through 2011. At the time of our
visits in 2011 and 2012, these facilities continued to use temporary space,
with the exception of SeaTac.

Page 65

GAO-12-743 Federal Prison Crowding

Appendix III: Effects of a Growing Inmate
Population

Table 9: Rated Capacity and Temporary Bed Space of Selected BOP Facilities from Fiscal Years 2006 through 2011
Year
2006

2007

2008

2009

2010

2011

Facilitya

RC

Tem

RC

Tem

RC

Tem

RC

Tem

RC

Tem

RC

Tem

Petersburg (low)

858

356

834

414

834

381

834

413

834

225

834

371

1,152

364

1,152

481

1,108

686

1,108

759

1,108

596

1,108

557

736

152

768

179

768

56

768

39

722

(59)

722

(71)

72

(21)

72

(11)

72

25

528

237

720

281

720

429

Lewisburg
(general
population)

698

607

698

648

698

615

242

(40)

160

(47)

160

41

Schuylkill
(medium)

720

473

848

382

848

379

848

446

848

293

848

400

1,193

476

1,193

372

1,193

556

1,193

516

1,193

515

1,193

505

Petersburg
(medium)
SeaTac
Lewisburg
(Special
Management
Unit)b

Leavenworth
(medium)

Legend:
RC refers to rated capacity.
Tem refers to temporary bed space, not including disciplinary housing beds.
Source: GAO analysis of BOP data.

Notes: The parentheses represent beds available above the rated capacity, but not used. The beds
were available because of specific institutional changes, such as the conversion of Lewisburg to a
Special Management Unit. During fiscal years 2010 and 2011, the temporary female beds at SeaTac
were not used.
a

The Petersburg, Lewisburg, Schuylkill, and Leavenworth facilities are male long-term institutions.
Male long-term institutions include four security level designations––minimum, low, medium, and
high. SeaTac is an administrative facility that specifically serves inmates awaiting trial, or those with
intensive medical or mental health conditions, regardless of the level of supervision these inmates
require.
b

A Special Management Unit operates as a more controlled and restrictive environment for inmates
whose interaction requires greater management to ensure the safety, security, or orderly operation of
BOP facilities, or protection of the public. BOP established a Special Management Unit at Lewisburg
Penitentiary in fiscal year 2008, and subsequently converted the entire facility to a Special
Management Unit, with the exception of a unit housing general population high security inmates.

According to all of the regional directors and wardens in the two facilities
we visited, different regions and facilities used different approaches to
temporary bed space. For example, one regional director said that all
facilities in his region used some temporary space to house inmates, but
generally, institutions were no longer using television rooms to house
inmates. He said that it is safer to manage 3 inmates in one cell through
triple bunking than to manage 16 inmates in a converted television room.

Page 66

GAO-12-743 Federal Prison Crowding

Appendix III: Effects of a Growing Inmate
Population

Alternatively, at a medium security facility we visited in another region,
officials told us that all inmates were double bunked in cells and the
facility had converted four former television rooms to temporarily house 8
to 10 inmates in each. In each housing unit, all televisions were relocated
to a single common room. BOP headquarters officials noted that having a
single television room is a common cause of disciplinary incidents
because fighting may erupt among groups of inmates who want to watch
different programs.
In addition to crowding in a facility’s housing and common areas, inmates
may experience crowded bathroom facilities, reductions in shower times,
shortened meal times coupled with longer waits for food service, and
more limited recreational activities because of the increased inmate
population. For example, with more inmates, it takes longer for
correctional officers to escort inmates to the dining hall and for each
inmate to be served in the food service line. According to BOP officials,
extended wait times at meals in particular can be problematic because
BOP attempts to keep inmates on strict schedules and extended waits
may cause inmates to arrive late for vocational classes or work
assignments, which can delay the start of the class or assignment. The
increased inmate population also affects recreation space and activity
time. 3 For example, according to staff at one BOP facility we visited, in a
crowded arts room, inmates may accidentally bump elbows, resulting in
tension or friction, which may lead to a security incident. At another BOP
facility, with a rated capacity of 850 but housing 1,300 inmates, officials
said that crowding affects accessibility to recreational activities such as
team sports, especially during warmer weather, when 500 inmates may
be in a recreational area supervised by one or two staff.

3

Non-U.S. citizens or deportable aliens can participate in recreation programs (e.g.,
leisure, fitness, wellness, or sports activities). A deportable alien is an alien in the United
States and admitted to the United States subject to any grounds of removal specified in
the Immigration and Nationality Act, 8 U.S.C. § 1227. This includes any alien illegally in
the United States, regardless of whether the alien entered the country by fraud or
misrepresentation or entered legally but subsequently violated the terms of his or her
nonimmigrant classification or status.

Page 67

GAO-12-743 Federal Prison Crowding

Appendix III: Effects of a Growing Inmate
Population

Program Participation

BOP provides programs including education, vocational training, drug
treatment, and faith-based reentry programs that help to rehabilitate
inmates and support correctional management. 4 According to BOP
officials, two benefits of inmate programming are (1) public safety, and (2)
institutional safety and security because of reduced inmate idleness.
These officials said, however, that the growth in the inmate population
had increased program waiting lists, contributing to inmate idleness. 5
BOP officials said facility staff offer a variety of education programs, such
as mandatory General Educational Development (GED) courses; 8- to 10week nonmandatory courses on topics such as parenting, word processing,
and conversational Spanish; occupational training; and computer-based
self-paced courses (e.g., English). 6 According to BOP data, overall inmate
participation in one or more programs ranged from 35 to 37 percent from
fiscal years 2006 through 2011. 7 The percentage participation, number of
inmates on waiting lists, and length of the average waiting time varied by
program. For example, BOP snapshots from fiscal years 2008 through
2012 of the total population at BOP-run facilities showed that between 13
and 14 percent of inmates were enrolled in literacy programs, while

4

See GAO-01-483.

5

BOP also houses low security non-U.S. citizens in private contract facilities. BOP’s
contracts require the private providers to provide all programs (1) in accordance with the
contract, which requires compliance with ACA standards and (2) as outlined in the
contractors’ technical proposals. According to BOP, most private contract facilities provide
work and self-improvement opportunities to inmates.

6

According to BOP, non-U.S. citizens or deportable aliens housed in BOP facilities are
exempt from the required participation in the GED program, but are encouraged to attend
these courses. These inmates can also participate in English as a Second Language
courses, adult continuing education classes (e.g., typing and computer literacy), a release
preparation program, and parenting classes. Additionally, these inmates may participate in
BOP’s occupational education programs if BOP resources permit after meeting the needs
of other eligible inmates.

7

BOP’s Monthly Participation Reports provide a snapshot of program participation levels
of inmates within BOP facilities. Figures for overall inmate participation do not duplicate.
That is, if an inmate is enrolled in more than one program area (for example GED and
parenting), the inmate’s participation is counted only once.

Page 68

GAO-12-743 Federal Prison Crowding

Appendix III: Effects of a Growing Inmate
Population

between 11 and 12 percent remained on waiting lists. 8 Table 10 shows
systemwide participation rates in selected BOP education and training
programs as of September 2011.
Table 10: Systemwide Inmate Participation Rates in Selected BOP Programs in September 2011

Inmate participation rateb

At least one
program

GED classes

Adult continuing
education

Occupational
training

Parent
education

Postsecondary
programsa

36%

13%

11%

7%

2%

1%

Source: GAO analysis of BOP data.
a

Post Secondary Education is a program category that is funded by inmates’ personal funds rather
than through BOP’s salaries and expenses budget account. The curriculum extends through an
associate’s degree.

b

Some programs are not offered year-round (for example parenting), and if the participation snapshot
occurs when a program is not in session, the participation level will be zero.

BOP provides inmates with the opportunity to participate in a variety of
drug treatment programs. In more than half of its facilities, BOP offers a
residential drug abuse treatment program. In all of its facilities, BOP offers
nonresidential drug abuse and drug education programs. All of the drug
treatment and drug education programs had waiting lists from fiscal years
2006 through 2011. According to BOP officials, if BOP cannot meet the
substance abuse treatment or education needs of inmates because it
does not have the staff needed to meet program demand, some inmates
will not receive programming benefits. As we reported in February 2012,
long waiting lists for BOP’s Residential Drug Abuse Program (RDAP),
which provides sentence reductions for eligible inmates who successfully

8

Enrollment figures include participation in GED classes and programs BOP is piloting but
has not yet implemented on a wider scale. BOP defines the literacy wait list to include
those inmates who are capable of participating and willing to participate in the literacy
program and who are not enrolled in the GED program, English as a Second Language
(ESL) program, or a literacy pilot program. Snapshots of the literacy program waiting lists
in January and February 2012 at the facilities we visited were Petersburg medium (250),
Petersburg low (93), Lewisburg (13), Schuylkill (36), Leavenworth (187), and SeaTac (78).

Page 69

GAO-12-743 Federal Prison Crowding

Appendix III: Effects of a Growing Inmate
Population

complete the program, 9 constrained BOP’s ability to admit participants
early enough to earn their maximum allowable reductions in times
served. 10 From fiscal years 2009 through 2011, the number of slots for
inmates to participate in RDAP increased by 400. According to BOP
officials, as RDAP capacity has increased, BOP has reduced waiting lists
even with continued growth in the inmate population, thereby enabling
inmates to enter the program sooner and increasing the number of
inmates (from 14 to 25 percent) who complete the program and receive a
sentence reduction. Nevertheless, the program continues to experience
long waiting lists, although the average wait has declined. For example, in
low security facilities in fiscal year 2006, 3,547 inmates participated in the
RDAP program, 3,378 inmates were on the waiting list, and the average
waiting time was about 205 days. In contrast, in low security facilities in
fiscal year 2011, 3,082 inmates participated in the program, 3,723 were
on the waiting list, but the average waiting time was approximately 80
days. Tables 11-13 provide additional data on participation, waiting lists,
and average waiting time for BOP drug education and treatment
programs in male facilities.

9

Under 28 C.F.R. § 550.53(b), to be admitted into RDAP, inmates must meet the following
criteria: (1) inmates must have a verifiable substance use disorder; (2) inmates must sign an
agreement acknowledging program responsibility; and (3) when beginning the program, the
inmate must be able to complete all components of the program. Under 28 C.F.R. § 550.55,
inmates may be eligible for early release by a period not to exceed 12 months if they meet
the following criteria: (1) were sentenced to a term of imprisonment under either (i) 18 U.S.C.
Chapter 227, Subchapter D for a nonviolent offense (i.e., an inmate who committed a federal
offense on or after November 1, 1987, after the effective date of the Sentencing Reform Act
of 1987, also known as “new law.”); or (ii) D.C. Code § 24-403.01 for a nonviolent offense,
meaning an offense other than those included within the definition of “crime of violence” in
D.C. Code § 23-1331(4); and (2) successfully complete a RDAP during their current
commitment. To receive the full-sentence reduction of 12 months, inmates are required to
participate in the program for 27 months. “Old law” inmates who are parole eligible, may, at
the U.S. Parole Commission’s discretion, be considered for an advanced release date
through an award of Superior Program Achievement.

10

For more on RDAP, see GAO-12-320.

Page 70

GAO-12-743 Federal Prison Crowding

Appendix III: Effects of a Growing Inmate
Population

Table 11: BOP’s Drug Education Programs in Male Facilities: Inmate Participation Levels, Waiting List Numbers, and Average
Waiting Time, by Institutional Security Level from the End of Fiscal Years 2006 through 2011
Year
Program participation by security level

2006

2007

2008

2009

2010

2011

Participation

2,578

2,394

2,378

3,716

3,195

3,198

Waiting list

2,984

3,700

3,808

6,268

5,129

4,681

Average wait in days

324.5

333.6

298.1

209.2

190.5

105.0

9,016

9,159

9,344

11,448

14,074

12,523

10,436

12,193

12,467

17,948

17,474

18,231

230.2

228.6

192.0

159.5

123.7

82.6

Participation

7,804

7,826

8,560

10,490

11,263

10,130

Waiting list

9,892

9,200

9,749

17,909

18,527

19,992

Average wait in days

223.4

197.7

160.4

168.7

143.1

84.4

Participation

4,959

4,872

4,084

5,423

5,291

5,183

Waiting list

6,269

5,857

5,124

7,104

7,043

7,400

Average wait in days

141.5

136.8

130.4

102.4

90.5

63.2

833

813

646

746

628

769

High

Medium
Participation
Waiting list
Average wait in days
Low

Minimum

Administrative
Participation
Waiting list
Average wait in days

789

756

669

1,088

1,118

1,295

180.5

150.7

153.2

130.9

111.2

73.7

Source: GAO analysis of BOP data.

Note: Drug education programs are distinct from residential and nonresidential drug treatment
programs.

Page 71

GAO-12-743 Federal Prison Crowding

Appendix III: Effects of a Growing Inmate
Population

Table 12: BOP’s Nonresidential Drug Treatment Programs in Male Facilities: Inmate Participation Levels, Waiting List
Numbers, and Average Waiting Time, by Institutional Security Level from the End of Fiscal Years 2006 through 2011
Year
Program participation by security level

2006

2007

2008

2009

2010

2011

Participation

493

585

554

818

566

946

Waiting list

160

197

397

465

699

1,388

253.4

264.7

310.4

395.0

199.1

101.3

2,480

3,281

4,195

3,411

3,446

4,852

High

Average wait in days
Medium
Participation
Waiting list
Average wait in days

568

774

997

2,723

3,688

4,707

299.2

266.1

214.1

204.4

178.0

98.5

Low
Participation

4,088

4,747

4,127

3,717

3,538

4,931

Waiting list

1,070

1,615

1,704

2,691

3,570

3,842

Average wait in days

160.9

184.4

188.2

211.4

169.9

79.8

2,475

2,820

2,559

3,040

3,055

3,984

Minimum
Participation
Waiting list

596

676

980

1,989

2,531

3,115

223.0

217.5

165.8

171.7

151.4

85.3

Participation

224

239

247

218

195

759

Waiting list

111

184

172

202

200

523

121.8

268.0

244.3

226.2

125.6

59.3

Average wait in days
Administrative

Average wait in days
Source: GAO analysis of BOP data.

Page 72

GAO-12-743 Federal Prison Crowding

Appendix III: Effects of a Growing Inmate
Population

Table 13: BOP’s Residential Drug Abuse Programs in Male Facilities: Inmate Participation Levels, Waiting List Numbers, and
Average Waiting Time, by Institutional Security Level from the End of Fiscal Years 2006 through 2011
Year
Program participation by security level

2006

2007

2008

2009

2010

2011

a

High

Participation

63

1

0

2

0

0

354

298

305

277

226

245

377.0

322.5

324.9

292.8

231.3

131.1

Participation

1,772

1,884

2,008

2,114

2,220

2,379

Waiting list

2,577

2,770

3,054

2,745

2,546

2,928

Average wait in days

242.3

226.5

213.6

194.8

147.1

92.8

Waiting list
Average wait in days
Medium

Low
Participation

3,547

3,326

3,256

2,987

3,398

3,082

Waiting list

3,378

3,409

3,773

3,185

3,264

3,723

Average wait in days

205.3

174.6

178.0

166.2

125.3

80.2

Participation

3,704

3,443

3,789

3,764

3,719

3,231

Waiting list

4,104

3,931

3,998

3,440

3,379

3,758

Average wait in days

258.8

237.6

223.9

202.9

145.6

83.8

Minimum

Administrative
Participation
Waiting list
Average wait in days

64

31

92

56

75

65

145

87

149

108

92

106

279.6

224.2

199.8

187.7

156.6

87.1

Source: GAO analysis of BOP data.
a

According to BOP officials, high security facilities currently do not offer a Residential Drug Abuse
Program, but because the program is offered near an inmate’s release date, high security inmates
may apply for the program and may be able transfer to a lower security level facility that offers the
program by the time they are ready for release. BOP plans to activate Residential Drug Abuse
Programs in high security facilities in fiscal year 2013 and has requested funding for this program in
its fiscal year 2013 budget submission.

BOP also implements two faith-based reentry programs through its
Religious Services Branch. Life Connections is BOP’s 18-month
residential program, begun in 2002, that offers a core curriculum taught
by spiritual guides hired from different faiths. The number of enrollments
in the program ranged from 345 to almost 400 inmates systemwide. The
program waiting lists for each security level were generally equal to or
greater than the number of participants. Threshold is BOP’s
nonresidential faith-based program, which began in 2008. Over 550
inmates were enrolled in the program as of January 2012; maintaining a

Page 73

GAO-12-743 Federal Prison Crowding

Appendix III: Effects of a Growing Inmate
Population

waiting list for Threshold program participation is at the discretion of
participating institutions’ wardens.

Meaningful Work
Opportunities

According to BOP headquarters officials, the growth in the federal inmate
population has also affected inmate work opportunities, as it is difficult to
find meaningful work for all inmates in a crowded facility, even though
generally all inmates are required to have a job. 11 They said that with the
growth of the prison population, fewer opportunities exist to engage in
meaningful work. This makes it difficult for staff to keep inmates busy,
resulting in inmate idleness, which can lead to additional tension and
fighting between inmates. Inmate discord can then affect the security and
safety of other inmates and staff. For example, BOP headquarters
officials and the warden at one facility we visited explained that inmate
wages vary with the job. Specifically, wages may range from 12 cents per
hour for sweeping the facility to $1.15 per hour for some jobs in factories
that Federal Prison Industries, also known as UNICOR, runs on the
prison grounds. Tensions may arise because inmates want to be
reassigned from lower- to higher-paying jobs. According to BOP officials,
BOP has tried to develop an index of idleness; however, measuring
idleness is difficult because inmates are usually engaged in some activity,
even though that activity may not be meaningful to their development
(e.g., filling salt shakers in the cafeteria).
Facility officials underscored that the most desirable jobs in the facilities
were those in UNICOR factories. 12 These factories (1) produce items
such as furniture, office supplies, and uniforms for sale to government
customers and (2) perform services for both government and private
sector purchase. Officials at one medium security facility we visited told
11

The Crime Control Act of 1990, Pub. L. No. 101-647, § 2905, 104 Stat. 4789, 4914
(codified at 18 U.S.C. § 4121 note) established a mandatory work requirement for all
prisoners. Specifically, this section provided that in general, it is the policy of the federal
government that convicted inmates confined in federal prisons, jails, and other detention
facilities shall work. The type of work in which they will be involved shall be dictated by
appropriate security considerations and by the health of the prisoner involved. A federal
prisoner may be excused from the requirement to work only as necessitated by security
considerations; disciplinary action; medical certification of disability such as would make it
impracticable for prison officials to arrange useful work for the prisoner to perform; or a
need for the prisoner to work less than a full work schedule in order to participate in
literacy training, drug rehabilitation, or similar programs in addition to the work program.
12
Non-U.S. citizens who are currently under an order of deportation, exclusion, or
removal from the United States are precluded from participating in the UNICOR program,
under 28 C.F.R. § 345.35.

Page 74

GAO-12-743 Federal Prison Crowding

Appendix III: Effects of a Growing Inmate
Population

us that these jobs teach inmates valuable reentry skills, such as coming
to work on time. 13 Such jobs can also teach money management and
budgeting, and because the earnings are higher than those for other
prison jobs, inmates have the opportunity to send money home to their
families. As a result, officials said there is less idleness among inmates
with UNICOR jobs, which helps to support security and safety in the
facility, and that the recidivism rate for participants is lower. 14 Facility
officials also noted that the decline in the number of UNICOR jobs has
resulted in waiting lists and challenged staff to create jobs to support
industrial work programming. Systemwide, the number of UNICOR
factories peaked at 110 in 2007, declining to 88 in 2011. These factories
employed over 23,000 inmates in 2007, declining to 14,200 inmates in
2011. In May 2010 the waiting list for UNICOR jobs was over 26,000
inmates, with an average waiting time of 16 months. 15 According to BOP
officials, a UNICOR waiting list includes inmates from the facility where
the factory is located. An inmate transferring to a facility may be placed
on the waiting list, but an inmate cannot be transferred to a facility to
participate in the UNICOR program. At one facility we visited, officials told
us that approximately 200 inmates were on the waiting list. They
explained that the waiting list consisted of three groups: inmates with
financial responsibilities, inmates with prior UNICOR experience, and
inmates on the general waiting list. When slots opened up, inmates were
selected equally from each of the three lists.

13

Reentry refers to the transition of inmates from prisons or jails back into the community.
It is BOP’s philosophy that preparation for reentry begins on the first day of an inmate’s
incarceration.
14
William G. Saylor and Gerald G. Gaes. PREP: Training Inmates through Industrial Work
Participation, and Vocational and Apprenticeship Instruction, Federal Bureau of Prisons
(Washington, D.C.: Sept. 24, 1996), and Saylor and Gaes. The Differential Effect of
Industries and Vocational Training on Post Release Outcome for Ethnic and Racial
Groups, Federal Bureau of Prisons (Washington, D.C.: Sept. 8, 1999). Recidivism
generally refers to a former inmate’s relapse into criminal behavior, and although agencies
may measure recidivism in different manners, recidivism measures can include the
rearrest, reconviction, or re-incarceration of former inmates.

15
The waiting lists by security level were minimum (638), low (2,549), medium (13,154),
high (9,439), and maximum (569). The source for these data is Marketing Research &
Corporate Support.

Page 75

GAO-12-743 Federal Prison Crowding

Appendix III: Effects of a Growing Inmate
Population

Visitation

Crowded visiting rooms make it more difficult for inmates to visit with their
families. BOP headquarters officials said the quality of the interaction
between an inmate and family can positively affect an inmate’s behavior
in prison and aids an inmate’s success when returning to the community.
Each BOP facility has visiting space to accommodate the number of
inmates that the facility was designed to house and a visitor capacity to
enable staff to manage the visitation process. The infrastructure of the
facility may not support the increase in visitors as a result of the growth in
the prison population. Further, with more inmates, the visitation process
requires more staff resources. BOP officials explained that the visiting
process requires at least four staff––one in the front lobby to process
visitors, one to escort inmates in and out of the visiting room and search
the inmate for contraband following the visit, one to document and search
visitors to prevent the introduction of contraband, and one to walk around
the visiting area supervising interactions.
Limited visiting capacity and the larger numbers of inmates can lead to
frustrations for inmates and visitors, such as when visits are shorter or
visitors are turned away because there are too many visitors on a
particular day. Five of the regional directors and officials at four of the
facilities we visited reported that the effect of the population growth on
visitation varied by region and facilities within the region because of a
number of factors, including proximity of the facility to inmates’ families. If
a large percentage of the inmate population is from the area where the
facility is located, the visiting room is used at a greater frequency.
Conversely, if inmates’ families do not live near the facility, the increase in
the number of inmates does not have a similar effect on visitation. For
example, one regional director told us that the increase in the number of
inmates had not affected overall visitation in the region because of the
large number of inmates who were non-U.S. citizens and whose families
did not visit. In another region, the regional director told us that some
facilities have problems with visitation only on holidays, when families
wish to be together, while others have problems regularly. According to
officials, allowing all inmates rather than none or some inmates to have
visitors helps inmates’ morale and facility management.
Additionally, the larger number of inmates also limits inmate access to the
telephone to call home and computer to e-mail family members and other

Page 76

GAO-12-743 Federal Prison Crowding

Appendix III: Effects of a Growing Inmate
Population

contacts. 16 For example, at one facility we visited, each housing unit had
three telephones for about 156 inmates. We reported in September 2011
that BOP provided a variety of options to its inmates for making phone
calls to friends and families; nevertheless, the number of contraband cell
phones in prisons had risen. Given the potential that these phones
provide for furthering criminal activity (e.g., selling drugs), the illicit use of
cell phones can pose a danger to staff and inmates, as well as the public
at large. 17

Correctional Staff

Facility officials told us that because of the large prison population,
correctional officers do not have time to use their interpersonal
correctional skills and maintain communication between staff and
individual inmates. Representatives of the correctional workers’ union and
officials at three of the five prisons we visited specifically emphasized the
importance of interaction and communication between inmates and
correctional staff for purposes of inmate reentry and facility management.
During one site visit, we observed that facility department heads and unit
managers stand along the cafeteria serving line during meals to provide
the opportunity for inmates to speak with staff about problems and
concerns. 18 A union representative observed that inmates used to tell
correctional staff about problems (e.g., where other inmates were storing
contraband), but with more than one inmate in a cell, inmates may not
want to talk to the correctional officer in front of the other inmates.
Tables 14 through 16 provide historical data on inmate to total BOP staff
ratios, inmate to BOP institutional staff ratios, and inmate to BOP
correctional staff ratios.

16
Through BOP’s system for e-mail, inmates can communicate with a list of contacts, but
they cannot access the Internet. Both inmates and persons in the community with whom
they correspond must consent to having all incoming and outgoing electronic messages
monitored and retained by BOP staff.
17

GAO, Bureau of Prisons: Improved Evaluations and Increased Coordination Could
Improve Cell Phone Detection, GAO-11-893 (Washington, D.C.: Sept. 6, 2011).

18
According to BOP, this practice also affords security additional to that provided by
uniformed correctional officers, as these administrators and staff are also trained
correctional officers.

Page 77

GAO-12-743 Federal Prison Crowding

Appendix III: Effects of a Growing Inmate
Population

Table 14: BOP Inmate to Total BOP Staff Ratios from Fiscal Years 1997 through
2011
Year

Total S&E staff
onboard

BOP institution
population

Inmate to staff ratio by
fiscal year

1997

28,302

101,091

3.57

1998

28,870

108,207

3.75

1999

29,176

117,295

4.02

2000

30,382

125,560

4.13

2001

31,806

130,327

4.10

2002

31,823

137,527

4.32

2003

32,265

146,212

4.53

2004

32,746

152,518

4.66

2005

32,735

159,501

4.87

2006

33,114

162,514

4.91

2007

33,994

167,323

4.92

2008

34,139

165,964

4.86

2009

34,914

172,423

4.94

2010

35,972

173,289

4.82

2011

35,987

177,934

4.94

Source: BOP data.

Note: The President’s annual budget justification to Congress for BOP includes a systemwide inmate
to total BOP staff ratio. This ratio is systemwide rather than by facility security level. BOP calculates
this ratio using (1) the inmate population at each facility on the last day of the fiscal year and (2) the
total number of BOP staff on board as of the last pay period of the fiscal year. The total number of
BOP staff includes all staff at BOP institutions, regional offices, training centers, and the central office
(i.e., staff funded by BOP’s Salaries & Expenses (S&E) appropriation and Public Health Service
(PHS) staff).

Page 78

GAO-12-743 Federal Prison Crowding

Appendix III: Effects of a Growing Inmate
Population

Table15: BOP’s Inmate to Total Institutional Staff Ratios from Fiscal Years 2006
through 2011

Systemwide
Administrative
Complex

2006

2007

2008

2009

2010

2011

5.2

5.2

5.2

5.2

5.2

5.2

6

5.7

5.5

5.1

5

4.9

5.1

5.4

5.4

5.3

5.1

5.3

Detention

4.6

4.4

4.6

4.6

4.5

4.6

High

4.5

4.5

4.2

4.1

4.1

4.1

Low

6.6

6.1

6

6.1

6.2

6.1

Medical

3

2.7

2.8

2.9

2.8

2.8

Medium

5.4

5.5

5.6

5.5

5.5

5.6

Minimum

8

8

7.6

7.6

7.6

8

Female facilities

5.7

5.7

5.7

5.5

5.2

5.2

Male facilities

5.2

5.2

5.2

5.2

5.2

5.2

Source: GAO analysis of BOP data.

Note: For purposes of this table, a ratio of “x:1” is expressed more simply as “x.” These ratios were
calculated using the following information. The inmate population used to calculate the ratio is based
on the Fiscal Year Average Daily Population (ADP) for each facility. ADP is calculated by totaling the
total inmate days recorded at each facility for the fiscal year and dividing that number by the number
of days in the fiscal year. The staffing level used to calculate the ratio is based on the fiscal year end
onboard staffing level at each facility as of the last pay period of the fiscal year. S&E and PHS staff
onboard at year end were included in this calculation. The calculation excludes Buildings and
Facilities, UNICOR, and Trust Fund staff as well as all staff at regional offices, training centers, and
central office locations and facilities that were in the activation process.

Page 79

GAO-12-743 Federal Prison Crowding

Appendix III: Effects of a Growing Inmate
Population

Table 16: Snapshots of BOP’s Inmate to Correctional Officer Ratios, by BOP Region, from Fiscal Years 2006 through 2011
Region
Mid-Atlantic

North Central

Ratios

South Central

Southeast

Western

2007

2008

2009

2010

2011

Low ratio

7.8

3.6

8.1

7.4

7.9

7.5

High ratio

24.4

26

23.3

22.5

24.6

25.3

Average ratio

11.5

10.5

11

10.8

10.3

10.6

Low ratio

1.6

4.9

5.2

5.3

5.0

5.3

High ratio

37

26

25.8

28.2

26.7

28.6

9

8.8

9.1

9.3

8.8

9.3

Average ratio
Northeast

2006

Low ratio

6.9

5.5

4.6

5.9

5.5

5.9

High ratio

21.7

17.2

17.8

24.6

19.8

14.6

Average ratio

11.8

10.4

10.3

10.6

10.3

10.1

Low ratio

7.6

7.5

6.7

7.1

7.3

7.8

High ratio

31.7

29.6

24.2

24.1

22.3

24.4

Average ratio

13.2

12.9

11.7

12.5

12.5

12.7

Low ratio

9.3

9.4

8.4

9.7

9.2

9.9

High ratio

34.9

28.1

26.2

27.4

25.6

30.4

Average ratio

12.4

12

11.8

11.8

11.6

12.2

Low ratio

5.9

4.7

5.9

6.5

5.4

5.9

High ratio

16.5

18.1

17

17.8

14.6

16.5

Average ratio

10.4

10

9.7

9.7

9.4

10.3

Source: GAO analysis of BOP data.

Note: Correctional officer refers to those BOP staff who enforce the regulations governing the
operation of a correctional institution. All values imply an inmate to correctional officer ratio of “x:1.”
For purposes of this table, we have omitted the “1” from the ratio and just reported the “x.” BOP
includes facilities not yet fully activated in its data. Thus, where a facility had an inmate-to-staff ratio of
some number less than 1:1, we omitted that facility from our analysis of the high, low, and average
ratios for the respective region. For example, in fiscal year 2010 and 2011, BOP listed a facility called
Mendota in its listing for the Western Region. In fiscal year 2010, BOP recorded the inmate to
correctional officer ratio for Mendota as 00:1 and in fiscal year 2011, BOP reported the ratio as 0.6:1.
This facility was not fully activated in either year, so we omitted it from our table.

Page 80

GAO-12-743 Federal Prison Crowding

Appendix III: Effects of a Growing Inmate
Population

Infrastructure
Systemwide, water costs were over $25
million in fiscal year 2005 and more than
$37 million in fiscal year 2011, an increase
of about 48 percent. Water usage
increased from 7.9 billion gallons in 2005
to 9.9 billion gallons in 2011.

More inmates results in greater water usage for heating, laundry,
showers, toilets, sanitation, and food service. As a result, BOP is the
largest energy and water consumer in DOJ. Specifically, BOP estimated
that each inmate uses approximately 150 gallons of water per day. BOP’s
electricity costs also increased about 35 percent, from about $79 million
in fiscal year 2005 to more than $107 million in fiscal year 2011.
According to BOP, the main reason for this increase is the rising cost of
electricity over the last 6 years, although more inmates with more laundry
also have had a marginal impact. BOP regional and facility officials said
that with the increased population, the food service equipment is used to
prepare more meals, thereby shortening the life of the equipment.
In February 2012, BOP reported 150 major unfunded repair projects (i.e.,
each project had an estimated repair cost of $300,000 or more) for a total
projected repair cost of almost $346 million. These included unfunded
repairs totaling about $30 million for four of the five prisons we visited, such
as replacement of a roof, repairs to the perimeter wall, replacement of the
perimeter fence, and upgrading a sewer system. One regional director
observed, however, that when BOP does not have funding for repairs, staff
find a way to “make do.” For example, at one older facility in that region,
staff built a second roof over the computer room, rather than undertake the
more costly roof repair, because the main roof of the facility leaked.
BOP officials said facilities are so crowded that closing older facilities is
not an option. These officials explained that BOP’s goal is to reduce
crowding to 15 percent systemwide. Because BOP projects continued
population growth and does not anticipate a large increase in facility
capacity, a significant decrease in the population would have to occur
before BOP would be able to consider facility closures. Officials said that
before closing BOP-run facilities, BOP would reduce its private prison
contracts, but reductions in facility space would also depend on the
security level of the population where reductions occurred.

Security and Safety

According to BOP officials, the increasing inmate population and staffing
ratios negatively affect inmate conduct and the imposition of discipline,
thereby affecting security and safety. BOP maintains an inmate discipline
program for all inmates in BOP custody, which is to help ensure the
safety, security, and orderly operation of correctional facilities, as well as
the protection of the public, by allowing BOP staff to impose sanctions on

Page 81

GAO-12-743 Federal Prison Crowding

Appendix III: Effects of a Growing Inmate
Population

inmates who commit prohibited acts. 19 BOP classifies prohibited acts into
four levels according to the severity of the offense and provides a range
of sanctions. 20 Available sanctions are based on the severity of the
offense and include disallowance and forfeiture of good conduct time
credit, disciplinary segregation, loss of privileges (e.g., visitation,
telephone, and commissary). 21 BOP disciplinary hearing data show an
increase from fiscal years 2006 through 2010 and a decline in fiscal year
2011. 22 Table 17 shows the trends in the number of guilty findings for
each level of prohibited acts from fiscal years 2006 through 2011.
According to BOP officials, moderate severity (300-level) prohibited acts
include less serious but more frequently committed types of inmate
misconduct, such as insolent behavior toward staff, thus explaining the
large number of findings reported for this category.

19

BOP’s inmate discipline program is authorized by 18 U.S.C. § 4042(a)(3).

20
BOP has four levels of offenses, ranging from level 100 to level 400: greatest severity
level (100) offenses (e.g., killing, serious assault, and possession of weapons), high
severity level (200) offenses (e.g., fighting or threatening bodily harm), moderate severity
level (300) offenses (e.g., being in an unauthorized area, refusing an order, or insolence),
and low severity level (400) offenses (e.g., feigning illness).
21

Generally, BOP is authorized to award up to 54 days of good conduct time credit for
each year served (which vests on the date the inmate is released). Good conduct time
credit may be awarded to an inmate serving a sentence of more than 1 year, but less than
life. The credit may be disallowed for an inmate found to have committed a prohibited act.
Loss of good conduct time credit is a mandatory disciplinary sanction for Violent Crime
Control and Law Enforcement Act of 1994 violent inmates, Prison Reform Litigation Act
inmates, and D.C. Code offenders pursuant to 28 C.F.R. § 541.4. The amount of good
conduct time disallowed is based upon the severity level of the offense and is a sanction
that may only be imposed by the disciplinary hearing officer (DHO). The DHO may also
sanction an inmate to forfeiture of good conduct time, if available, as good conduct time
does not vest until the inmate is released.
22
BOP has two types of disciplinary hearings. One is conducted by the Unit Discipline
Committee (UDC), which consists of two or more institution staff. The other type of
hearing is conducted by the DHO, who works for the regional director. Initially, the UDC
reviews the incident report. Depending on the severity of the prohibited act, the UDC may
make a finding that the inmate did or did not commit the prohibited act or the UDC may
refer the report to the DHO for a hearing. The DHO must make a final disposition on all
greatest (100) and high (200) severity level offenses. The total number of UDC hearings
increased from about 67,000 in fiscal year 2006 to about 72,700 in fiscal year 2010,
declining to about 70,800 in fiscal year 2011. The total number of DHO hearings increased
from about 53,000 in fiscal year 2006 to over 60,600 in fiscal year 2010 and then declined
to about 59,600 in fiscal year 2011.

Page 82

GAO-12-743 Federal Prison Crowding

Appendix III: Effects of a Growing Inmate
Population

Table 17: Guilty Findings for Prohibited Acts by Severity Level, Fiscal Years 2006
through 2011
Guilty findings
for 100-level
Fiscal years prohibited acts

Guilty findings
for 200-level
prohibited acts

Guilty findings
for 300-level
prohibited acts

Guilty findings
for 400-level
prohibited acts

2006

7,711

23,731

54,219

1,929

2007

8,361

24,112

51,519

2,025

2008

9,552

24,980

50,633

1,933

2009

11,021

27,386

54,043

2,125

2010

10,677

28,915

59,462

2,042

2011

10,195

27,168

60,269

1,731

Source: GAO analysis of BOP data.

Note: This table reflects only the most serious prohibited act for each incident report. One
hundred-level prohibited acts are of the greatest severity (e.g., killing or assaulting a
person) and 400-level acts are of the lowest severity (e.g., feigning illness).

During the 6-year period, BOP data indicated that the most frequently
imposed sanctions for guilty findings were loss of privileges, disallowance
of good time credit, and segregation.
In addition to maintaining individual discipline, BOP facility management
may lock down a facility––a temporary situation in which all inmates are
confined to their living quarters/cells until staff are able to assess the
situation following a critical incident (e.g., a, assaults on staff by several
inmates, or a food or work strike) and can safely return the institution to
normal operations. According to BOP data, BOP systemwide imposed
almost 4,000 lockdowns from fiscal years 2006 through 2011. Similar to
the inmate misconduct data, the number of lockdowns increased from
fiscal years 2006 through 2009, peaking at 1,042 that year and then
declining to 824 in fiscal year 2011.

Page 83

GAO-12-743 Federal Prison Crowding

Appendix IV: GAO Contact and Staff
Acknowledgments
Appendix IV: GAO Contact and Staff
Acknowledgments

GAO Contact

David C. Maurer, (202) 512-9627 or maurerd@gao.gov

Staff
Acknowledgments

In addition to the contact named above, Joy Booth, Assistant Director;
Pedro Almoguera; Willie Commons; Eric Hauswirth; Lara Miklozek; Linda
Miller; Meghan Squires; Barbara Stolz; and Greg Wilmoth made key
contributions to this report.

Page 84

GAO-12-743 Federal Prison Crowding

Related GAO Products
Related GAO Products

Federal Bureau of Prisons: Methods for Estimating Incarceration and
Community Corrections Costs and Results of the Elderly Offender Pilot.
GAO-12-807R. Washington, D.C.: July 27, 2012.
Bureau of Prisons: Eligibility and Capacity Impact Use of Flexibilities to
Reduce Inmates’ Time in Prison. GAO-12-320. Washington, D.C.
February 7, 2012.
Bureau of Prisons: Improved Evaluations and Increased Coordination
Could Improve Cell Phone Detection. GAO-11-893. Washington, D.C.:
September 6, 2011.
Bureau of Prisons: Evaluating the Impact of Protective Equipment Could
Help Enhance Officer Safety. GAO-11-410. Washington, D.C.:
April 8, 2011.
Bureau of Prisons: Methods for Cost Estimation Largely Reflect Best
Practices, but Quantifying Risks Would Enhance Decision Making.
GAO-10-94. Washington, D.C.: November 10, 2009.
Prison Construction: Clear Communication on the Accuracy of Cost
Estimates and Project Changes is Needed. GAO-08-634. Washington,
D.C.: May 29, 2008.
Cost of Prisons: Bureau of Prisons Needs Better Data to Assess
Alternatives for Acquiring Low and Minimum Security Facilities.
GAO-08-6. Washington, D.C.: October 5, 2007.
Prisoner Releases: Trends and Information on Reintegration Programs.
GAO-01-483. Washington, D.C. June 18, 2001.
State and Federal Prisoners: Profiles of Inmate Characteristics in 1991
and 1997. GAO/GGD-00-117. Washington, D.C.: May 24, 2000.

(441010)

Page 85

GAO-12-743 Federal Prison Crowding

GAO’s Mission

The Government Accountability Office, the audit, evaluation, and
investigative arm of Congress, exists to support Congress in meeting its
constitutional responsibilities and to help improve the performance and
accountability of the federal government for the American people. GAO
examines the use of public funds; evaluates federal programs and
policies; and provides analyses, recommendations, and other assistance
to help Congress make informed oversight, policy, and funding decisions.
GAO’s commitment to good government is reflected in its core values of
accountability, integrity, and reliability.

Obtaining Copies of
GAO Reports and
Testimony

The fastest and easiest way to obtain copies of GAO documents at no
cost is through GAO’s website (http://www.gao.gov). Each weekday
afternoon, GAO posts on its website newly released reports, testimony,
and correspondence. To have GAO e-mail you a list of newly posted
products, go to http://www.gao.gov and select “E-mail Updates.”

Order by Phone

The price of each GAO publication reflects GAO’s actual cost of
production and distribution and depends on the number of pages in the
publication and whether the publication is printed in color or black and
white. Pricing and ordering information is posted on GAO’s website,
http://www.gao.gov/ordering.htm.
Place orders by calling (202) 512-6000, toll free (866) 801-7077, or
TDD (202) 512-2537.
Orders may be paid for using American Express, Discover Card,
MasterCard, Visa, check, or money order. Call for additional information.

Connect with GAO

Connect with GAO on Facebook, Flickr, Twitter, and YouTube.
Subscribe to our RSS Feeds or E-mail Updates. Listen to our Podcasts.
Visit GAO on the web at www.gao.gov.

To Report Fraud,
Waste, and Abuse in
Federal Programs

Contact:
Website: http://www.gao.gov/fraudnet/fraudnet.htm
E-mail: fraudnet@gao.gov
Automated answering system: (800) 424-5454 or (202) 512-7470

Congressional
Relations

Katherine Siggerud, Managing Director, siggerudk@gao.gov, (202) 5124400, U.S. Government Accountability Office, 441 G Street NW, Room
7125, Washington, DC 20548

Public Affairs

Chuck Young, Managing Director, youngc1@gao.gov, (202) 512-4800
U.S. Government Accountability Office, 441 G Street NW, Room 7149
Washington, DC 20548

Please Print on Recycled Paper.