Congressional Research Federal Prison Population Buildup Jan 2013

• Topics: Crime/Demographics, Statistics/Trends

Download original document:
Brief thumbnail

This text is machine-read, and may contain errors. Check the original document to verify accuracy.

The Federal Prison Population Buildup:
Overview, Policy Changes, Issues,
and Options
Nathan James
Analyst in Crime Policy
January 22, 2013

Congressional Research Service
7-5700
www.crs.gov
R42937

CRS Report for Congress
Prepared for Members and Committees of Congress

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

Summary
Since the early 1980s, there has been a historically unprecedented increase in the federal prison
population. Some of the growth is attributable to changes in federal criminal justice policy during
the previous three decades. An issue before Congress is whether policymakers consider the rate of
growth in the federal prison population sustainable, and if not, what changes could be made to
federal criminal justice policy to reduce the prison population while maintaining public safety.
This report explores the issues related to the growing federal prison population.
The number of inmates under the Bureau of Prisons’ (BOP) jurisdiction has increased from
approximately 25,000 in FY1980 to nearly 219,000 in FY2012. Since FY1980, the federal prison
population has increased, on average, by approximately 6,100 inmates each year. Data show that
a growing proportion of inmates are being incarcerated for immigration- and weapons-related
offenses, but the largest portion of newly admitted inmates are being incarcerated for drug
offenses. Data also show that approximately 7 in 10 inmates are sentenced for five years or less.
Changes in federal sentencing and correctional policy since the early 1980s have contributed to
the rapid growth in the federal prison population. These changes include increasing the number of
federal offenses subject to mandatory minimum sentences; changes to the federal criminal code
that have made more crimes federal offenses; and eliminating parole.
There are several issues related to the growing federal prison population that might be of interest
to policymakers:
•

The increasing number of federal inmates, combined with the rising per capita
cost of incarceration, has made it increasingly more expensive to operate and
maintain the federal prison system. The per capita cost of incarceration for all
inmates increased from $19,571 in FY2000 to $26,094 in FY2011. During this
same period of time, appropriations for the BOP increased from $3.668 billion to
$6.381 billion.

•

The federal prison system is increasingly overcrowded. Overall, the federal
prison system was 39% over its rated capacity in FY2011, but high- and mediumsecurity male facilities were operating at 51% and 55%, respectively, over rated
capacity. At issue is whether overcrowding might lead to more inmate
misconduct. The results of research on this topic have been mixed. One study
found that overcrowding does not affect inmate misconduct; but the BOP, based
on its own research, concluded that there is a significant positive relationship
between the two.

•

The inmate-to-staff ratio has increased from 4.1 inmates per staff member in
FY2000 to 4.9 inmates per staff member in FY2011. Likewise, the inmate to
correctional officer ratio increased from 9.8 inmates per correctional officer in
FY2000 to 10.2 inmates per correctional officer in FY2011, but this is down from
a high of 10.9 inmates per correctional officer in FY2005.

•

The growing prison population is taking a toll on the infrastructure of the federal
prison system. The BOP reports that it has a backlog of 154 modernization and
repair projects with an approximate cost of $349 million for FY2012. Past
appropriations left the BOP in a position where it could expand bedspace to
manage overcrowding but not reduce it. However, reductions in funding since

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

FY2010 mean that the BOP will lack the funding to begin new prison
construction in the near future. At the same time, it has become more expensive
to expand the BOP’s capacity.
Should Congress choose to consider policy options to address the issues resulting from the
growth in the federal prison population, policymakers could choose options such as increasing the
capacity of the federal prison system by building more prisons, investing in rehabilitative
programming, or placing more inmates in private prisons.
Policymakers might also consider whether they want to revise some of the policy changes that
have been made over the past three decades that have contributed to the steadily increasing
number of offenders being incarcerated. For example, Congress could consider options such as
(1) modifying mandatory minimum penalties, (2) expanding the use of Residential Reentry
Centers, (3) placing more offenders on probation, (4) reinstating parole for federal inmates, (5)
expanding the amount of good time credit an inmate can earn, and (6) repealing federal criminal
statutes for some offenses.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

Contents
Introduction...................................................................................................................................... 1
Federal Prison Population ................................................................................................................ 2
Conviction Offense for Federal Inmates.................................................................................... 3
Length of Sentences for Federal Offenders ............................................................................... 6
Policy Changes that Contributed to Prison Population Growth....................................................... 7
Mandatory Minimum Sentences................................................................................................ 8
Federalization of Crime ............................................................................................................. 9
Eliminating Parole for Federal Inmates ................................................................................... 10
Issues Related to Prison Population Growth.................................................................................. 10
Cost of Operating the Federal Prison System .......................................................................... 10
Prison Overcrowding ............................................................................................................... 20
Inmate-to-Staff Ratio ............................................................................................................... 26
Prison Construction and Maintenance ..................................................................................... 30
Select Policy Options ..................................................................................................................... 32
Continuing or Expanding Current Correctional Policies ......................................................... 32
Expanding the Capacity of the Federal Prison System ..................................................... 32
Investing in Rehabilitative Programs ................................................................................ 33
Placing More Inmates in Private Prisons .......................................................................... 34
Changing Existing Correctional and Sentencing Policies to Reduce the Prison
Population............................................................................................................................. 36
Changes to Mandatory Minimum Penalties ...................................................................... 37
Alternatives to Incarceration ............................................................................................. 37
Early Release Measures .................................................................................................... 43
Modifying the “Safety Valve” Provision ........................................................................... 49
Repealing Federal Criminal Statutes for Some Offenses .................................................. 50
Conclusion ..................................................................................................................................... 51

Figures
Figure 1. Federal Prison Population, FY1980-FY2012 ................................................................... 3
Figure 2. Conviction Offenses of Inmates Entering Federal Prison, FY1998-FY2010 ................... 5
Figure 3. Inmates in Federal Prison at the End of the Fiscal Year, by Major Offense Type,
FY1998-FY2010 ........................................................................................................................... 6
Figure 4. Length of Sentence for Inmates Entering Federal Prison, FY1998-FY2010 ................... 7
Figure 5. Appropriations for the BOP, FY1980-FY2012 ............................................................... 12
Figure 6. The BOP’s Appropriation as a Share of the DOJ’s Discretionary Budget
Authority, FY1980-FY2012 ....................................................................................................... 13
Figure 7. Difference Between Appropriations and the Administration’s Request for the
BOP’s S&E and B&F Accounts ................................................................................................. 14
Figure 8. Per Capita Cost ($) of Medical Care, Utilities, and Food, FY2000-FY2012 ................. 17
Figure 9. Per Capita Cost ($) of Rehabilitative Programs, FY2000-FY2012 ................................ 19

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

Figure 10. Overcrowding in the Federal Prison System, FY1981-FY2012................................... 21
Figure 11. Growth in Funding for Contract Confinement, Inmate Care and Programs,
Institutional Security and Administration, and Overall Salaries and Expenses .......................... 23
Figure 12. Change in BOP Staff, Correctional Officers,
and Institutional Inmate Population ............................................................................................ 29
Figure 13. The Cost of Expanding Rated Prison Capacity by One Inmate.................................... 31
Figure 14. Proportion of Sentenced Defendants in Federal Courts Placed on Probation .............. 40

Tables
Table 1. Per Capita Cost of Incarceration in the Federal Prison System, FY2000-FY2012 .......... 15
Table 2. Overcrowding in All BOP facilities and Low-, Medium-, and High-Security
Male Facilities, FY1995-FY2012 ............................................................................................... 22
Table 3. Inmate-to-Staff and Inmate-to-Correctional Officer Ratios for the Federal Prison
System ........................................................................................................................................ 27
Table A-1. Appropriations for the BOP by Account; Number of Inmates Under the BOP’s
Jurisdiction; and the Number and Capacity of and Overcrowding in BOP Facilities ................ 53
Table A-2. Appropriations for the BOP, by Decision Unit, FY1999-FY2012 ............................... 55

Appendixes
Appendix. Select BOP Data ........................................................................................................... 53

Contacts
Author Contact Information........................................................................................................... 55

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

Introduction
The Bureau of Prisons (BOP) is the largest correctional agency in the country in terms of the
number of prisoners under its jurisdiction.1 The BOP currently operates 118 correctional facilities
in 35 states and Puerto Rico.2 The BOP was established in 1930 to house federal inmates,
professionalize the prison service, and ensure consistent and centralized administration of the
federal prison system.3
Since the early 1980s, there has been a historically unprecedented increase in the number of
inmates incarcerated in the federal prison system. The number of inmates under the BOP’s
jurisdiction has increased from approximately 25,000 in FY1980 to nearly 219,000 in FY2012. In
comparison, the federal prison population increased by approximately 12,000 inmates between
1930 and 1980. Since FY1980, the federal prison population has increased, on average, by
approximately 6,100 inmates each fiscal year.
Some of the growth in the federal prison population is attributable to policy changes over the
previous three decades, including
•

increasing the number of federal offenses subject to mandatory minimum
sentences,

•

changes to the federal criminal code that have made more crimes federal
offenses, and

•

eliminating parole.

The BOP faces several challenges resulting from the increasing number of inmates placed under
its supervision. The first is the increasing cost of operating the federal prison system. Data show
that with each passing fiscal year it is increasingly more expensive to incarcerate an inmate in a
federal prison, yet the BOP must operate the federal prison system within the annual
appropriation approved by Congress. Second, the federal prison system is becoming more
overcrowded, especially in high- and medium-security male prisons. Research conducted by the
BOP suggests that there might be a link between higher levels of overcrowding and inmate
misconduct. Third, the federal inmate population is increasing at a rate whereby the gap between
the number of inmates and the number of staff and correctional officers is slowly starting to
widen. Finally, the rising federal inmate population is starting to place a strain on the
infrastructure of the federal prison system. The BOP has not been able to expand prison capacity
at a rate that would allow it to close older prisons and it has also had to defer hundreds of millions
of dollars in maintenance costs, which might result in either direct or indirect security problems.
There are a number of policy avenues lawmakers could consider should Congress choose to
address the growth in the federal prison population. Several options—such as expanding the
capacity of the federal prison system, continued investment in rehabilitative programs, and
1

E. Ann Carson and William J. Sabol, Prisoners in 2011, U.S. Department of Justice, Office of Justice Programs,
Bureau of Justice Statistics, NCJ 239808, Washington, DC, December 2012, p. 21, http://www.bjs.gov/content/pub/
pdf/p11.pdf, hereinafter, “Prisoners in 2011.”
2
Data provided by the U.S. Department of Justice, Bureau of Prisons.
3
U.S. Department of Justice, Bureau of Prisons, About the Bureau of Prisons, p. 1, http://www.bop.gov/about/
index.jsp, hereinafter “About the Bureau of Prisons.”

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

placing inmates in private prisons—either continue or expand current correctional policies.
However, Congress might also consider changing some existing correctional or sentencing
policies as a means of addressing some of the issues related to the growth of the federal prison
population. Some of these options include placing some inmates in alternatives to incarceration,
such as probation, or expanding early release options by allowing inmates to earn more good time
credit or allowing inmates to be placed on parole once again. Congress could consider reducing
the amount of time inmates are incarcerated in federal prisons by repealing mandatory minimum
penalties for some offenses or reducing the length of the mandatory minimum sentence. Finally,
policymakers could consider repealing federal criminal statutes for some offenses.

Federal Prison Population
At the end of 1930, the BOP operated 14 facilities that held approximately 13,000 inmates.4 By
the end of 1940, the BOP had expanded to 24 facilities that held approximately 24,000 inmates.5
The number of inmates in the federal prison system, with a few fluctuations, remained at
approximately 24,000 for the next four decades.6 Then, as shown in Figure 1, beginning in
FY1980 the federal prison population started an unabated, three-decade increase. The total
number of inmates under the BOP’s jurisdiction increased from approximately 25,000 in FY1980
to nearly 219,000 in FY2012. Between FY1980 and FY2012, the federal prison population
increased, on average, by approximately 6,100 inmates annually. The growth in the federal prison
population was much higher between FY1990 and FY2009 compared to the period of FY1980
through FY1989. On average, the federal prison population increased by approximately 3,700
inmates per fiscal year between FY1980 and FY1989. In contrast, the average increase per fiscal
year between FY1990 and FY1999 was approximately 7,600 inmates and between FY2000 and
FY2009 it was approximately 7,500 inmates. The growth in the federal prison population for the
first few years of the current decade has been erratic. The federal prison population only grew by
nearly 1,500 inmates between FY2009 and FY2010, but in FY2011, it grew by more than 7,500
inmates, which is more in line with previous trends. The total number of inmates in federal prison
increased by approximately 900 prisoners in FY2012, the lowest level of annual growth in any
fiscal year since FY1980.
Recent trends in the federal prison population stand in contrast to overall incarceration trends.
The Bureau of Justice Statistics (BJS) reports that the total number of inmates under the
jurisdiction of state correctional authorities decreased between 2009 (1,407,369) and 2010
(1,404,032) and between 2010 and 2011 (1,382,418).7 However, while the number of state
inmates has decreased, the federal prison population has continued to increase.
The data in Figure 1 also show that most of the federal prison population is incarcerated in a BOP
facility, as opposed to a contract facility.8 However, over the years the BOP has had to rely
increasingly on contract facilities to help manage the federal prison population. In FY1980, less
4

U.S. Department of Justice, Bureau of Prisons, A Brief History of the Bureau of Prisons, http://www.bop.gov/about/
history.jsp.
5
Ibid.
6
Ibid.
7
Prisoners in 2001, p. 2.
8
Contract facilities include bedspace the BOP contracts for in privately operated prisons, Residential Reentry Centers
(i.e., halfway houses), and state and local correctional facilities.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

than 2% of federal inmates were housed in a contract facility. The number of federal inmates in
contract facilities increased to nearly 11% in FY1990, approximately 14% in FY2000, and nearly
18% in FY2010.
Figure 1. Federal Prison Population, FY1980-FY2012
Number of inmates in thousands
250

Number of Inmates

200

150

100

2012

2010

2008

2006

2004

2002

2000

1998

1996

1994

1992

1990

1988

1986

1984

1982

1980

Fiscal Year
Institution

Contract

Total

Source: Presentation of data provided by the U.S. Department of Justice, Bureau of Prisons.

The following discussion of some of the demographics of federal inmates uses data from the BJS
rather than the BOP. BJS data on federal prisoners are only available for FY1998 through
FY2010. Therefore, the BJS data cannot be used to show how these select demographics changed
since the federal prison population started its sustained growth in the early 1980s. The proceeding
discussion is intended to provide context for the discussion later in the report of potential policy
options for addressing federal prison population growth.

Conviction Offense for Federal Inmates
As shown in Figure 2, in FY1998 approximately 18% of inmates entering the federal prison
system were convicted for an immigration offense. There was a slight increase in the proportion
of such inmates being sent to federal prison in both FY1999 and FY2000, but this trend was
reversed by FY2002. However, in FY2003, the proportion of inmates entering the federal prison
system for immigration offenses started an unabated increase. By FY2010, immigration offenders
accounted for approximately 30% of all inmates entering the system that fiscal year. There was

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

also a noticeable increase in the number of inmates entering the federal prison system for
weapons-related convictions between FY1998 and FY2010, but it was not as pronounced as the
increase in the number of inmates convicted for immigration offenses. Also, unlike the
immigration offenders, the number of inmates entering federal prisons for weapons-related
offenses has leveled-off. One other noticeable trend is the decrease in the number of inmates
being sent to federal prison for violent and property crimes. In FY1998, violent and property
offenders comprised approximately 9% and 18%, respectively, of all inmates entering federal
prison. By FY2010, these offenders accounted for approximately 4% and 11% of prison
admissions. The proportion of offenders entering federal prison for public order offenses9
remained relatively consistent between FY1998 and FY2010.
Despite the increase in the number of inmates entering the federal prison system for immigration
and weapons-related offenses, drug offenders still constitute the largest portion of inmates
entering federal prisons. The number of inmates being sent to federal prison for drug offenses has
decreased somewhat since FY1998 (when 41% of inmates entering federal prison were convicted
for drug offenses). In every fiscal year between FY1998 and FY2010, drug offenders constituted
the largest proportion of prison admissions, though in FY2009 and FY2010, immigration
offenders were a close second. The vast majority of sentenced drug offenders, more than 95%,10
were sent to federal prison for trafficking offenses.11

9
Public order offenses include tax law violations; bribery; perjury; national defense; escape; racketeering and
extortion; gambling; liquor; mailing or transporting of obscene materials; traffic; migratory birds; conspiracy, aiding
and abetting, and jurisdictional offenses; violations of regulatory laws and regulations in agriculture, antitrust, labor
law, food and drug, motor carrier, and other regulatory offenses.
10
Data downloaded from U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics, Federal
Criminal Case Processing Statistics, http://bjs.ojp.usdoj.gov/fjsrc/index.cfm.
11
“Trafficking offenses” include an offense where an offender knowingly and intentionally imported or exported any
controlled substance in schedule I, II, III, IV, or V (as defined by 21 U.S.C. §812). It includes manufacturing,
distributing, dispensing, selling, or possessing with intent to manufacture, distribute, or sell a controlled substance or a
counterfeit substance; exporting any controlled substance in schedules I-V; manufacturing or distributing a controlled
substance in schedule I or II for purposes of unlawful importation; or making or distributing any punch, die, plate,
stone, or any other thing designed to reproduce the label upon any drug or container, or removing or obliterating the
label or symbol of any drug or container. It also includes knowingly opening, maintaining or managing any place for
the purpose of manufacturing, distributing, or using any controlled substance (for example, 19 U.S.C. §1590; 21 U.S.C.
§§333(e), 825(a)-(d), 830(a), 841(a)-(b) (d)(e)(g), 842(a), 843(a)(b), 845, 846, 848, 854, 856, 858, 859(a)(b), 860(a),
861(c)(f), 952(a)(b), 953(a)(e), 957, 959, 960(a)(b)(d), 961, 962, and 963; and 46A U.S.C. §§1903(g) and (j)). U.S.
Department of Justice, Office of Justice Programs, Bureau of Justice Statistics, Compendium of Federal Justice
Statistics, 2004, NCJ 213476, Washington, DC, December 2006, p. 119, http://www.bjs.gov/content/pub/pdf/
cfjs04.pdf.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

Figure 2. Conviction Offenses of Inmates Entering Federal Prison, FY1998-FY2010
45.0%
40.0%

Proportion of new admissions

35.0%
30.0%
25.0%
20.0%
15.0%
10.0%
5.0%
0.0%
1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Violent

Property

Drug

Public order

Weapons

Immigration

Source: Presentation of data from the U.S. Department of Justice, Bureau of Justice Statistics, Federal Criminal
Case Processing Statistics.
Notes: Percentages were calculated excluding offenders whose conviction offense was classified as “unknown.”

As shown in Figure 3, in FY1998, weapons and immigration offenders were 8.3% and 7.1%,
respectively, of all federal inmates. By FY2010, weapons and immigration offenders comprised
15.5% and 11.6% of all federal inmates. By FY2010, approximately 8 out of every 10 inmates in
federal prison were incarcerated for a drug, weapons, or immigration offense. While a growing
proportion of federal inmates were incarcerated for drug, weapons, or immigration offenses,
fewer inmates were incarcerated for violent offenses. In FY1998, nearly 12% of federal inmates
were incarcerated for a violent offense; by FY2010, the proportion of federal inmates incarcerated
for violent offenses decreased to 6.4%.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

Figure 3. Inmates in Federal Prison at the End of the Fiscal Year,
by Major Offense Type, FY1998-FY2010
100%
90%

Proportion of all inmates

80%
70%
60%
50%
40%
30%
20%
10%
0%
1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Property

Public order

Violent

Weapon

Immigration

Drug

Length of Sentences for Federal Offenders
As shown in Figure 4, in any given fiscal year between FY1998 and FY2010, more than 7 in 10
inmates entering the federal prison system were sentenced to a term of incarceration that was five
years or less. The data indicate two distinct trends in the sentencing of federal inmates. First,
since FY1998 approximately three of every 10 inmates entering federal prisons were sentenced to
a term of incarceration that was less than one year, though fewer inmates entered the federal
prison system in FY2010 with a sentence of one year or less compared to FY1998. Second, since
FY1998 there has been a slow, but steady, growth in the proportion of inmates sentenced to
between three and five years of incarceration, and to a lesser extent, the proportion of inmates
sentenced to between five and 10 years of incarceration.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

Figure 4. Length of Sentence for Inmates Entering Federal Prison, FY1998-FY2010
40.0%

Proportion of new admissions

35.0%
30.0%
25.0%
20.0%
15.0%
10.0%
5.0%
0.0%
1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
1 month to < 1 year

1 year to < 3 years

3 years to < 5 years

5 years to < 10 years

10 years to < 20 years

20 years to < Life

Life
Source: Presentation of data from the U.S. Department of Justice, Bureau of Justice Statistics, Federal Criminal
Case Processing Statistics.
Notes: Percentages were calculated excluding offenders whose sentence was classified as “unknown.”

Policy Changes that Contributed to Prison
Population Growth
A confluence of changes to federal sentencing and correctional policy since the early 1980s—
including the expanded use of mandatory minimum penalties, the federalization of crime, and the
abolition of parole for federal inmates—have contributed to the growing federal prison
population. The expanded use of mandatory minimum penalties has resulted in offenders being
sentenced to longer terms of imprisonment than they were 20 years ago. At the same time, the
expanding federal criminal code, combined with greater enforcement of federal criminal statutes,
has resulted in more people entering the federal criminal justice system. Thus, while more
offenders are being arrested by federal law enforcement, tried in federal courts, and sentenced to
incarceration in federal prisons for increasingly longer periods of time, the abolition of parole
ensures that most inmates will serve all or nearly all of their sentences.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

Mandatory Minimum Sentences
In a recent report, the United States Sentencing Commission (USSC) found that the enactment of
a greater number of federal mandatory minimum sentences has, in part, contributed to the
growing federal prison population. Mandatory minimum penalties have contributed to federal
prison population growth because they have increased in number, have been applied to more
offenses, required longer terms of imprisonment, and are used more frequently than they were 20
years ago.12
The number of mandatory minimum penalties in the federal code expanded as Congress made
more offenses subject to such penalties. The USSC reported that the number of mandatory
minimum penalties in the federal criminal code nearly doubled from 98 in 1991 to 195 in 2011.13
Not only has there been an increase in the number of federal offenses that carry a mandatory
minimum penalty, but offenders who are convicted of offenses with mandatory minimums are
being sent to prison for longer periods. For example, the USSC found that, compared to FY1990
(43.6%), a larger proportion of defendants convicted of offenses that carried a mandatory
minimum penalty in FY2010 (55.5%) were convicted of offenses that carried a mandatory
minimum penalty of five years or more.14
While only offenders convicted for an offense carrying a mandatory minimum penalty are subject
to those penalties, mandatory minimum penalties have, in effect, increased sentences for other
offenders.15 The USSC has incorporated many mandatory minimum penalties into the sentencing
guidelines, which means that penalties for other offense categories under the guidelines had to
increase in order to keep a sense of proportionality.16 Research by the Urban Institute found that
increases in expected time served contributed to half of the prison population growth between
1998 and 2010.17 The increase in amount of time inmates were expected to serve was probably
partially the result of inmates receiving longer sentences and partially the result of inmates being
required to serve approximately 85% of their sentences after Congress eliminated parole for
federal prisoners (this is discussed in greater detail in the “Eliminating Parole for Federal
Inmates” section).
However, the increase in the federal prison population is not solely attributable to the increased
use of mandatory minimum penalties. The USSC reported that the number of inmates in the
12

U.S. Sentencing Commission, Report to Congress: Mandatory Minimum Penalties in the Federal Criminal Justice
System, Washington, DC, October 2011, p. 63, http://www.ussc.gov/Legislative_and_Public_Affairs/
Congressional_Testimony_and_Reports/Mandatory_Minimum_Penalties/20111031_RtC_Mandatory_Minimum.cfm,
hereinafter “Mandatory Minimum Penalties in the Federal Criminal Justice System.”
13
Ibid., pp. 71-72.
14
Ibid., p. 76.
15
Erik Luna and Paul G. Cassell, “Mandatory Minimalism,” Cardozo Law Review, vol. 32, no. 1 (September 2010), pp.
16-17; James E. Felman, on behalf of the American Bar Association, statement before the United States Sentencing
Commission in the Hearing on Mandatory Minimums, May 27, 2010, p. 9, http://www.ussc.gov/
Legislative_and_Public_Affairs/Public_Hearings_and_Meetings/20100527/Testimony_Felman_ABA.pdf, hereinafter
“Felman testimony.”
16
Ibid.
17
Kamala Mallik-Kane, Barbara Parthasarathy, and William Adams, Examining Growth in the Federal Prison
Population, 1998 to 2010, The Urban Institute, Washington, DC, September 2012, p. 10, http://www.urban.org/
UploadedPDF/412720-Examining-Growth-in-the-Federal-Prison-Population.pdf, hereinafter “Examining Growth in the
Federal Prison Population.”

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

federal prison system who were convicted of an offense that carried a mandatory minimum
penalty increased 178%, from approximately 40,000 in FY1995 to nearly 112,000 in FY2010.18
Of these offenders, nearly 30,000 in FY1995 and approximately 80,000 in FY2010 were actually
subject to a mandatory minimum penalty.19 However, over the same time period there was a
similar rate of growth in the number of inmates in federal prison who were not convicted of an
offense that carried a mandatory minimum. In FY1995, nearly 32,000 inmates in federal prison
were convicted of an offense that did not carry a mandatory minimum.20 This increased 152%, to
approximately 80,000 inmates, by FY2010.

Federalization of Crime
While the increase in the number of federal criminal statutes carrying a mandatory minimum
sentence has contributed to the escalating federal prison population, the USSC also identified the
federalization of crime21 as another contributing factor to prison population growth. Over the past
four decades the federalization of crime resulted in more people entering the federal criminal
justice system as federal law enforcement agencies and the U.S. Attorneys Office started to
enforce a broader array of federal offenses. The Urban Institute concluded that increased federal
law enforcement activity contributed to about 13% of the growth in the federal prison population
between 1998 and 2010, though the effects were not consistent across offense types and time.22
For example, heightened immigration enforcement and increased investigation of weapons
offenses contributed to approximately one-tenth of the population growth, but the growth in the
prison population resulting from investigating more weapons offenses mainly occurred between
1998 and 2005.23 However, decreased drug investigations reduced the federal prison population
from what it might have been assuming that federal law enforcement priorities and practices had
remained as they were in 1998.24

The USSC limited its analysis of the number of inmates in federal prisons who were convicted of or subject to a
mandatory minimum penalty to FY1995-FY2010 because the commission’s analysis relied on combining USSC data
with BOP data and there were limitations with the data prior to FY1995. Mandatory Minimum Penalties in the Federal
Criminal Justice System, p. 81.
19
Even though a defendant might be convicted for an offense that carries a mandatory minimum penalty there are
mechanisms whereby the court may impose a term of imprisonment that is below the mandatory minimum (i.e., the
defendant can be convicted for an offense carrying a mandatory minimum penalty, but the defendant is not subject to a
mandatory minimum penalty when sentenced). For example, under 18 U.S.C. §3553(e), the court “[u]pon motion of the
Government…shall have the authority to impose a sentence below a level established by statute as a minimum sentence
so as to reflect a defendant’s substantial assistance in the investigation or prosecution of another person who has
committed an offense.” Section 3553(e) also requires the sentence to be imposed in accordance with the federal
sentencing guidelines.
20
Mandatory Minimum Penalties in the Federal Criminal Justice System, p. 82.
21
The USSC defined “federalization of crime” as the transformation of traditional state and local criminal offenses into
federal crimes. Ibid., p. 63.
22
Examining Growth in the Federal Prison Population, p. 10.
23
Ibid.
24
Ibid., p. 11.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

Eliminating Parole for Federal Inmates
The BOP has identified the abolition of parole for federal inmates as one cause of the growing
federal prison population.25 The Comprehensive Crime Control Act of 1984 (P.L. 98-473)
abolished parole for federal inmates and modified how much good time credit an inmate could
earn. Anyone sentenced to incarceration for a federal crime committed after November 1, 1987, is
not eligible for parole. Abolishing parole in the federal correctional system means that the BOP
has not only had to confine a growing number of inmates, but it also has to confine them until
they serve all, or nearly all, of their sentences. The BOP reported that as of September 30, 2012,
only about 3% of inmates in federal prisons were still eligible to be paroled.26 The remainder of
federal inmates will have to serve their entire sentence, minus any good time credit they
might earn.27

Issues Related to Prison Population Growth
The growth of the federal prison population has given rise to several issues of interest to
policymakers. These include
•

the increasing cost of operating the federal prison system;

•

overcrowding in federal prisons;

•

an increasing inmate-to-staff ratio; and

•

a growing need for capital investment in correctional facilities.

Analysis of these issues is provided below.

Cost of Operating the Federal Prison System
The burgeoning prison population has contributed to mounting operational expenditures for the
federal prison system. Congress funds BOP’s operations through two accounts: Salaries and
Expenses (S&E) and Buildings and Facilities (B&F).28 The S&E account (i.e., the operating
budget) provides for the custody and care of federal inmates and for the daily maintenance and
operations of correctional facilities, regional offices, and BOP’s central office in Washington, DC.
It also provides funding for the incarceration of federal inmates in state, local, and private
facilities. The B&F account (i.e., the capital budget) provides funding for the construction of new
facilities and the modernization, repair, and expansion of existing facilities.

25
U.S. Department of Justice, Bureau of Prisons, A Brief History of the Bureau of Prisons, http://www.bop.gov/about/
history.jsp.
26
Data provided by the U.S. Department of Justice, Bureau of Prisons.
27
Each prisoner serving a term of imprisonment of more than one year, but not prisoners serving a life sentence, can
receive a good time credit of up to 54 days per year to count toward serving the sentence. The amount of the credit is
subject to the determination of the BOP. 18 U.S.C. §3624(b).
28
For a more in-depth analysis of the BOP’s appropriations, see CRS Report R42486, The Bureau of Prisons (BOP):
Operations and Budget, by Nathan James.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

As shown in Figure 5, the BOP’s appropriations increased more than $6 billion from FY1980
($330 million) to FY2012 ($6.641 billion). Between FY1980 and FY2012, the average annual
increase in the BOP’s appropriation was approximately $197 million. The data show that, by and
large, growth in the BOP’s appropriation is the result of ever-growing appropriations for the S&E
account. This is not surprising considering the constant growth in the federal prison population
and the fact that the S&E account provides funding for the care of federal inmates. Also, it has
been argued that even though appropriations for the BOP’s S&E account are discretionary, they
are effectively mandatory because “[b]y law, the BOP must accept and provide for all [f]ederal
inmates, including but not limited to inmate care, custodial staff, contract beds, food, and medical
costs. The BOP cannot control the number of inmates sentenced to prison, and unlike other
[f]ederal agencies, cannot limit assigned workloads and thereby control operating costs.”29
Appropriations for the B&F account have not grown as steadily as appropriations for the S&E
account. This is, in part, explained by how funding for the B&F account is used; namely,
Congress typically provides marked increases in this account only when there is a decision to
expand prison capacity. For example, the noticeable spike in appropriations for the B&F account
in FY1990 paved the way for an increase in BOP’s prison capacity in the mid- to late 1990s.

29
U.S. Congress, Senate Committee on Appropriations, Subcommittee on Commerce, Justice, Science, and Related
Agencies, Departments of Commerce and Justice, and Science, and Related Agencies Appropriations Bill, 2013, report
to accompany S. 2323, 112th Cong., 2nd sess., April 19, 2012, S.Rept. 112-158 (Washington: GPO, 2012), p. 64.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

Figure 5. Appropriations for the BOP, FY1980-FY2012
Appropriations in billions of dollars
7
6
5
4
3
2
1

2012

2010

2008

2006

2004

2002

2000

1998

1996

1994

1992

1990

1988

1986

1984

1982

1980

Fiscal Year
S&E

B&F

Total

Source: Presentation of data provided by the U.S. Department of Justice, Bureau of Prisons.
Notes: Between FY1980 and FY1995, appropriations for the National Institute of Corrections (NIC) were
provided in a separate account. After FY1995, the operating expenses for NIC are paid out of the BOP’s S&E
account. Therefore, to make appropriations for the S&E account as comparable as possible, appropriations for
the NIC for FY1980-FY1995 were added to appropriations for the S&E account. Between FY1996 and FY2000,
the BOP received an amount from the Violent Crime Reduction Trust Fund, which, according to the BOP, was
used for substance abuse treatment. As such, these amounts were also added to the S&E account Appropriations
include all supplemental and reprogrammed appropriations and rescissions to current year budget authority.

The BOP’s expanding budget is starting to consume a larger share of the DOJ’s overall annual
appropriation. Figure 6 shows what proportion of the DOJ’s discretionary appropriation was for
the BOP’s overall appropriation and for the BOP’s S&E appropriation, which has grown steadily
along with the prison population. The BOP’s overall budget is more susceptible to changes in
year-to-year appropriations for the B&F account. The trend lines show that since FY1980 both
the BOP’s total budget and appropriations for the BOP’s S&E account have, in general,
encompassed a growing share of the DOJ annual appropriation. The noticeable spike in the
BOP’s share of the DOJ’s annual appropriation in FY1990 was the result of Congress
appropriating more than $1 billion for the B&F account. In addition, the decrease in the BOP’s
share of DOJ’s appropriation observed in FY2009, a break in a general upward trend that started
in FY2000, was the result of Congress appropriating an additional $4 billion for DOJ under the
American Recovery and Reinvestment Act of 2009 (P.L. 111-5).

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

Figure 6. The BOP’s Appropriation as a Share of the DOJ’s Discretionary Budget
Authority, FY1980-FY2012
40.0%
35.0%
30.0%
25.0%
20.0%
15.0%
10.0%
5.0%

Total BOP appropriations

BOP S&E appropriations

Linear (Total BOP appropriations)

Linear (BOP S&E appropriations)

2012

2010

2008

2006

2004

2002

2000

1998

1996

1994

1992

1990

1988

1986

1984

1982

1980

0.0%

Source: The BOP’s annual appropriation data was provided by the U.S. Department of Justice, Bureau of
Prisons. Annual discretionary budget authority data for the Department of Justice were taken from Table 5.4
from the FY2013 Budget of the United States Government.

A comparison of the BOP’s annual appropriations for its S&E and B&F accounts to the
Administration’s request for both accounts shows that Congress has been more likely to fund the
Administration’s request for prison construction and less likely to fully fund the Administration’s
request for the upkeep and care of the prison population. The requested appropriation indicates
what the BOP believed it would need to properly manage the growing prison population each
fiscal year. The data suggest that in many fiscal years the BOP operated with a budget below what
it felt was adequate given the growing number of inmates under its jurisdiction.
The data presented in Figure 7 show that between FY1980 and FY2012, Congress appropriated
less than the Administration’s request for the B&F account 14 times. Over the same time period
Congress appropriated less than the Administration’s request for the S&E account 20 times. In
contrast to this general trend, however, the amount appropriated for the S&E account between
FY2007 and FY2010 actually exceeded the Administration’s request. The additional amounts, as
noted by the House Committee on Appropriations, were to compensate for underfunding the BOP,
which resulted in inadequate staffing levels and shortfalls in inmate programs.30 Both the House
30
U.S. Congress, House Committee on Appropriations, Omnibus Appropriations Act, 2009, committee print, 111th
Cong., 1st sess., March 2009 (Washington: GPO, 2009), p. 274.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

and Senate Committee on Appropriations reported that they felt that the Administration’s requests
for the BOP were inadequate for several years, which did not allow the bureau to meet its basic
operational needs.31
Figure 7. Difference Between Appropriations and the Administration’s Request for
the BOP’s S&E and B&F Accounts
Amounts in millions of dollars
1,200
1,000
800
600
400
200
0
-200

2012

2010

2008

2006

2004

2002

2000

1998

1996

1994

1992

1990

1988

1986

1984

1982

1980

-400

Fiscal Year
S&E

B&F

Source: Appropriated amounts were provided by the U.S. Department of Justice, Bureau of Prisons. The
Administration’s requested amounts were taken from the appendix to the Budget of the United States Government,
for FY1980-FY2012.
Notes: Between FY1980 and FY1995, appropriations for the National Institute of Corrections (NIC) were
provided in a separate account. After FY1995, the operating expenses for NIC are paid out of the BOP’s S&E
account. Therefore, to make appropriations for the S&E account as comparable as possible, appropriations for
the NIC for FY1980-FY1995 were added to appropriations for the S&E account. Between FY1996 and FY2000,
the BOP received an amount from the Violent Crime Reduction Trust Fund, which, according to the BOP, was
used for substance abuse treatment. As such, these amounts were also added to the S&E account.

Ibid. See also, U.S. Congress, House Committee on Appropriations, Departments of Transportation and Housing
and Urban Development, and Related Agencies Appropriations Act, 2010, Conference Report to Accompany H.R.
3288, 111th Cong., 1st sess., December 8, 2009, H.Rept. 111-366 (Washington: GPO, 2009), p. 671; U.S. Congress,
Senate Committee on Appropriations, Subcommittee on Commerce, Justice, Science, and Related Agencies,
Departments of Commerce and Justice, and Science, and Related Agencies Appropriations Bill, 2013, report to
accompany S. 2323, 112th Cong., 2nd sess., April 19, 2012, S.Rept. 112-158 (Washington: GPO, 2012), p. 65.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

While it is not surprising that the BOP’s annual appropriations would increase along with the
prison population—after all, more inmates require more care and supervision, which requires
additional funding—recent per capita expenditure data from the BOP indicate that it is getting
more expensive each year to incarcerate an inmate in the federal system. As shown in Table 1, the
overall per capita cost of incarcerating an inmate in the federal system has steadily increased from
FY2000 to FY2012. Over this time period, the cost of incarceration rose from approximately
$22,000 per inmate to more than $29,000 per inmate, an increase of 34.4%.
Table 1. Per Capita Cost of Incarceration in the Federal Prison System,
FY2000-FY2012
Security Level
Fiscal Year

All of BOP

High

Medium

Low

Minimum

Federal
Correctional
Complexesa

2000

$21,603

$26,518

$21,417

$18,407

$17,452

$21,360

2001

22,175

26,135

21,806

18,846

17,788

20,543

2002

22,518

27,456

21,473

19,228

18,770

21,538

2003

23,180

26,461

21,946

19,480

18,136

21,948

2004

23,267

26,951

21,896

19,242

17,647

21,764

2005

23,431

26,377

21,718

19,193

17,478

22,458

2006

24,439

25,398

23,648

20,834

17,291

23,152

2007

24,923

26,109

23,492

21,922

17,812

22,804

2008

25,895

27,924

24,065

23,373

19,635

23,958

2009

27,251

32,119

25,442

24,087

20,772

25,750

2010

28,282

33,858

26,248

25,377

21,005

27,267

2011

28,894

34,629

26,852

26,853

21,286

27,516

2012

29,027

34,046

26,686

27,166

21,694

27,683

Source: U.S. Department of Justice, Bureau of Prisons.
Notes: Per capita costs include support costs. The per capita cost of incarceration for all of BOP includes direct
costs for federal detention centers, administrative security facilities, medical referral centers, privately operated
institutions, residential reentry centers, and contracts with state and local institutions. It also includes support
costs for federal detention centers, administrative security facilities, and medical referral centers.
a.

Federal correctional complexes (FCC) contain two or more facilities with different security levels on the
same grounds. For example, FCC Allenwood (PA) contains a high, medium, and low security facility.

However, the per capita cost of incarceration decreases as inmates are moved into lower-security
level institutions. For example, in FY2012 it cost the BOP approximately $34,000 to house an
inmate in a high-security facility. In comparison, it cost the BOP approximately $27,000 to house
an inmate in either a medium- or low-security facility and nearly $22,000 to house an inmate in a
minimum-security facility. This is partly because lower-security facilities do not require as many
correctional officers; hence their operating expenditures are lower. In addition, the rated capacity
for a facility decreases as the security level increases, meaning that higher-security facilities hold
fewer inmates, which results in a higher per capita cost of incarceration for higher-security
facilities. For example, even if the total operating expenditures for a high- and a low-security
facility were the same for any given fiscal year, the per capita expenditure in the high-security
facility would be greater because it held fewer inmates.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

The BOP has identified rising utility, food, and medical care costs as three of the primary drivers
of the increasing cost of the federal prison system.32 The BOP’s expenditures on utilities, food,
and medical care have generally increased each fiscal year since FY2000.33 Moreover, as shown
in Figure 8, the per-inmate cost for utilities, food, and medical care has increased, although, the
per capita increase in the cost of food and utilities has not been as pronounced as the increase in
the per capita cost of inmate medical care. The medical costs for the BOP are most likely
increasing as the result of the general upward climb of health care costs in the United States, with
annual increases in health care costs outstripping inflation.34 The BOP reported that an increasing
number of federal inmates require medical care, primarily as a result of the expanding inmate
population. According to the BOP, conditions such as diabetes, hypertension, and infectious
diseases have a slightly higher rate of incidence in the incarcerated population.35 In addition, the
federal prison population is aging—BJS data shows that at the end of FY2010 approximately
14% of federal inmates were over the age of 50 and nearly 4% were over the age of 60—and in
general, older individuals require more medical care.

U.S. Department of Justice, Bureau of Prisons, FY2013 Performance Budget, Congressional Submission, Salaries
and Expenses, pp. 22, 25, and 42, http://www.justice.gov/jmd/2013justification/pdf/fy13-bop-se-justification.pdf;
hereinafter, BOP’s FY2013 S&E Budget Justification.
33
Data provided by the Bureau of Prisons.
34
Alex Nussbaum, “Health Care Costs Rise Faster Than U.S. Inflation Rate,” Bloomberg, May 21, 2012,
http://www.bloomberg.com/news/2012-05-21/health-care-costs-rise-faster-than-u-s-inflation-rate.html.
35
BOP’s FY2013 S&E Budget Justification, p. 21.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

Figure 8. Per Capita Cost ($) of Medical Care, Utilities, and Food, FY2000-FY2012
6,000

5,000

4,000

3,000

2,000

1,000

0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
Fiscal Year
Medical

Utilities

Food

Source: Based on a CRS analysis of data provided by the U.S. Department of Justice, Bureau of Prisons.
Notes: Per capita costs were calculated using figures on the number of inmates held in BOP facilities at the end
of the fiscal year.

In addition to the rising cost of utilities, food, and medical care, the cost of providing
rehabilitative programs for inmates has generally increased since FY2000 (see Figure 9). On a
per capita basis, the BOP spends the most on education programs for inmates. In FY2012, the
BOP spent nearly $275 more per inmate for education programs than it did in FY2000. The BOP
had similar per capita costs for substance abuse treatment programs and psychological services
between FY2000 and FY2005. However, after FY2005, the per capita costs for substance abuse
treatment started to grow at a rate that exceeded that of psychological services.
The growing cost of providing education and substance abuse treatment programs might be the
result of the BOP’s need to expand access to programming in order to meet increasing demand.
Under current law, the BOP is required to provide a functional literacy program for all mentally
capable inmates who are not functionally literate and to offer literacy/General Equivalency
Diploma (GED) programs for inmates who have not earned a high school diploma or its
equivalent.36 In addition, federal inmates are required to make satisfactory progress toward
earning a high school diploma or a GED in order to earn their full allotment of good time credit.37
36

See 18 U.S.C. §3624(f)(1) and 18 U.S.C. §3624(b)(3).
Each prisoner serving a term of imprisonment of more than one year, but not prisoners serving a life sentence, can
(continued...)
37

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

The BOP reports that since these requirements went into effect, demand for literacy programs has
increased.38 Also, current law (18 U.S.C. §3621) requires BOP to provide, subject to
appropriations, residential substance abuse treatment39 and appropriate aftercare40 for all eligible
prisoners.41 Prisoners who are convicted of nonviolent crimes and who successfully complete a
residential substance abuse treatment program are eligible to have their sentence reduced by not
more than one year.42
The BOP reports that in FY2007 and FY2008, it was not able to provide substance abuse
treatment to all eligible prisoners due to inadequate funding for program expansion; but, after an
investment in expanding drug treatment services, the BOP reports that since FY2009 it has been
able to provide substance abuse treatment to all eligible inmates.43 Obligations for psychology
services include funding for most rehabilitative programming (e.g., mental health and sex
offender treatment) other than education programs and substance abuse treatment. Per capita cost
for psychology services increased starting in FY2007, which roughly coincides with provisions in
the Adam Walsh Child Protection and Safety Act of 2006 (P.L. 109-248) requiring the BOP to
provide sex offender treatment to all inmates who are in need of and suitable for it, and with the
requirements in the Second Chance Act of 2007 (P.L. 110-199) that the BOP help prepare inmates
for re-entry.

(...continued)
receive a good time credit of up to 54 days per year to count toward serving the sentence. The amount of the credit is
subject to the determination of BOP. 18 U.S.C. §3624(b).
38
BOP’s FY2013 S&E Budget Justification, p. 26.
39
“Residential substance abuse treatment” is defined as a course of individual and group activities and treatment,
lasting at least six months, in residential treatment facilities set apart from the general prison population (which may
include pharmacotherapies, where appropriate) that may extend beyond the six-month period. 18 U.S.C.
§3621(e)(5)(A).
40
“Aftercare” is defined as placement, case management, and monitoring in a community-based substance abuse
treatment program when the prisoner leaves the custody of BOP. 18 U.S.C. §3621(e)(5)(C).
41
An “eligible prisoner” is defined as a prisoner who is determined by BOP to have a substance abuse problem and to
be willing to participate in a residential substance abuse treatment program. 18 U.S.C. §3621(e)(5)(B).
42
The following categories of inmates are not eligible for early release: (1) Immigration and Customs Enforcement
detainees; (2) pretrial inmates; (3) contractual boarders (for example, District of Columbia, state, or military inmates);
(4) inmates who have a prior felony or misdemeanor conviction for homicide, forcible rape, robbery, or aggravated
assault, or child sexual abuse offenses; (5) inmates who are not eligible for participation in a community-based program
as determined by the institution’s warden on the basis of his or her professional discretion; or (6) inmates whose current
offense is a felony. 28 C.F.R. §550.55.
43
BOP’s FY2013 S&E Budget Justification, p. 27.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

Figure 9. Per Capita Cost ($) of Rehabilitative Programs, FY2000-FY2012
900
800
700
600
500
400
300
200
100
0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
Fiscal Year
Substance abuse treatment

Education

Psychology services

The increasing cost of operating the federal prison system might be an issue for Congress as
policymakers seek to find ways to reduce discretionary spending. The appropriations committees
have already expressed concern that the continued growth in the federal prison population is not
sustainable.44 Unless there is a change in the upward trajectory in the number of inmates in the
federal prison system, Congress will face a decision regarding appropriation of additional funds
for the BOP. Assuming that growth in the budgetary requirements for the BOP exceeds the
growth in the allocation for the Administration of Justice budget function under the annual budget
resolution, policymakers might face a choice to reduce appropriations for other DOJ agencies or
DOJ grant programs in order to fund the federal prison system.

In the conference report for the Consolidated and Further Continuing Appropriations Act, 2012 (i.e., the “minibus,”
P.L. 112-55), the conferees expressed concern that “the current upward trend in the prison inmate population is
unsustainable and, if left unchecked, will eventually engulf the [Department of Justice's] budgetary resources.” U.S.
Congress, House of Representatives, Agriculture, Rural Development, Food and Drug Administration, and the Related
Agencies Programs for the Fiscal Year Ending September 30, 2012, and for Other Purposes, Conference Report to
Accompany H.R. 2112, 112th Cong., 1st sess., November 14, 2011, H.Rept. 112-284 (Washington, DC: GPO, 2011), p.
241.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

Prison Overcrowding
The growth of the federal prison population has resulted in the BOP incarcerating more inmates
than the federal prison system is rated to hold. Figure 10 shows that the level of overcrowding in
the federal prison system has changed over the years. Overcrowding reflects the difference
between how many inmates the federal prison system is “rated” to hold and how many inmates
the system actually holds.45 For example, if the overcrowding level is 35%, this means that the
number of inmates held in the federal system is 35% above the number of inmates the system is
rated to hold. Between FY1991 and FY1997, the federal prison system’s capacity nearly doubled
(a 95.2% increase) while the institutional population increased by 57.0% over this same time
period. The BOP was able to reduce prison overcrowding between FY1991 and FY1997 by
adding more bedspace, but it also changed the way it calculated rated capacity for each of its
facilities. Prior to FY1991, the BOP calculated its capacity based on single cell occupancy (i.e.,
one inmate per cell). But starting in FY1991, the BOP allowed for double bunking (i.e., two
inmates to one cell) in its facilities, which resulted in an increase in its rated capacity, which in
turn resulted in a decrease in overcrowding.
Data show that overcrowding in BOP facilities started to increase after FY1997, and it peaked in
FY2004 when overcrowding was at 41%. The BOP’s prison capacity expanded 30.7% between
FY1997 and FY2004 while the prison population grew by 50.9%. Overcrowding remained
around 35% between FY2005 and FY2010 after a steady growth between FY1997 and FY2004.
However, prison overcrowding increased to 39% by the end of FY2011, the highest level since
FY2004. Prison overcrowding decreased slightly in FY2012 to 38%, due to a decrease in the
institutional prison population (there were 378 fewer inmates held in BOP facilities in FY2012),
an increase in the number of beds (the BOP added 567 beds in FY2012), and greater use of
contract bedspace (there were 1,297 more inmates in contract facilities in FY2012 than there were
in FY2011).

Rated capacity, as calculated by the BOP, assumes some level of double bunking (i.e., two inmates to a cell) across
the federal prison system. The amount of double bunking allowed depends on the facility’s security level (i.e.,
minimum, low, medium, or high). The BOP calculates each facility’s rated capacity using the following formulas:
minimum and low security institutions at 100 percent double bunking; medium security institutions at 50 percent
double bunking and; high security institutions at 25 percent double bunking. For example, if a high security facility had
500 cells, the facility’s rated capacity would be 625 inmates. The rated capacity is intended to reflect the number of
prisoners that the institution can house safely and securely with adequate access to services and rehabilitative programs.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

Figure 10. Overcrowding in the Federal Prison System, FY1981-FY2012
80%
70%
60%
50%
40%
30%
20%
10%

2011

2009

2007

2005

2003

2001

1999

1997

1995

1993

1991

1989

1987

1985

1983

1981

Fiscal Year
Source: FY1981-FY2012 crowding levels were provided by the U.S. Department of Justice, Bureau of Prisons.
Notes: Overcrowding rates for FY1981-FY1990 are not directly comparable to crowding rates after FY1990
because the BOP changed the way that it calculated its capacity.

Data from the BOP indicate that while federal prisons have been operating at 30% or more over
rated capacity for more than a decade, the overcrowding problem is worse in high- and mediumsecurity male facilities (see Table 2). Medium-security male facilities were operating at well over
50% of rated capacity for the early part of the previous decade. Overcrowding in mediumsecurity facilities was brought down in the latter part of the previous decade because the BOP
opened several additional facilities, but medium security overcrowding levels have been slowly
increasing since FY2006. For the earlier part of the previous decade, overcrowding in high
security male facilities, in general, was not as bad as it was in medium security facilities.
However, since FY2006, high security facilities have been more crowded than medium and low
security facilities and above the average overcrowding within the federal prison system in
general. Since FY2006, overcrowding in high security male facilities has been near or above
50%.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

Table 2. Overcrowding in All BOP facilities and Low-, Medium-, and High-Security
Male Facilities, FY1995-FY2012
Fiscal Year

All Facilities

Male Low

Male
Medium

Male High

1995

25%

33%

52%

40%

1996

24%

31%

42%

65%

1997

22%

23%

37%

52%

1998

26%

27%

48%

56%

1999

31%

37%

51%

2000

32%

42%

50%

54%

2001

32%

38%

58%

42%

2002

33%

39%

58%

41%

2003

39%

59%

58%

2004

41%

45%

62%

49%

2005

34%

43%

42%

35%

2006

36%

41%

37%

53%

2007

37%

35%

42%

53%

2008

36%

35%

44%

50%

2009

37%

40%

47%

49%

2010

37%

43%

53%

2011

39%

37%

51%

55%

2012

38%

40%

47%

51%

Source: U.S. Department of Justice, Bureau of Prisons.
Notes: All BOP facilities include minimum security male facilities and secure female facilities, in addition to low,
medium, and high security male facilities.

The BOP has attempted to address prison overcrowding by placing a growing proportion of
federal inmates in contract facilities. While this has undoubtedly helped control prison
overcrowding to some degree, it is also drawing resources away from BOP’s other operations.
Figure 11 shows how growth in funding for the Contract Confinement46 decision unit compares
to growth in the Inmate Care and Programs47 and Institutional Security and Administration48
46
This decision unit provides for the costs associated with the confinement of federal inmates in contract facilities,
which include private prisons, residential re-entry centers, state and local facilities, and home confinement. It provides
funding for the management and oversight of contract confinement functions. The decision unit also provided funding
for the National Institute of Corrections.
47
This decision unit covers the cost of inmate food, medical supplies, institutional and release clothing, welfare
services, transportation, gratuities, staff salaries, and operational costs of functions directly related to providing inmate
care. It provides funding for inmate programs, including education and vocational training, psychological services,
religious programs, and drug treatment. All of the drug treatment programs discussed above are funded from this
decision unit. The decision unit also covers costs associated with regional and central office administration and support
related to providing inmate care.
48
This decision unit covers costs associated with the maintenance of facilities and institution security. It funds
institution maintenance, motor pool operations, powerhouse operations, institution security and other administrative
(continued...)

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

decision units and the S&E account overall. Figure 11 shows that in most fiscal years since
FY1999, growth in funding for contract confinement exceeded that of the S&E account overall.
Increased funding for contract confinement has come at the cost of slower growth for the Inmate
Care and Programs and Institutional Security and Administration decision units, especially the
latter, which in most fiscal years grew at a rate below either the S&E account or the Inmate Care
and Programs decision unit, or both.
Figure 11. Growth in Funding for Contract Confinement, Inmate Care and
Programs, Institutional Security and Administration, and
Overall Salaries and Expenses
60.0%
50.0%

Percent change

40.0%
30.0%
20.0%
10.0%
0.0%
-10.0%

Salaries and Expenses

Inmate Care and Programs

Institutional Security and Administration

Contract Confinement

2011-2012

2010-2011

2009-2010

2008-2009

2007-2008

2006-2007

2005-2006

2004-2005

2003-2004

2002-2003

2001-2002

2000-2001

1999-2000

-20.0%

Source: Presentation of data provided by the U.S. Department of Justice, Bureau of Prisons.
Notes: Figure does not include funding for the Management and Administration decision unit.

Continued increases in prison overcrowding could be an issue for Congress given concerns about
a potential link between prison overcrowding and increases in assaults and other inmate
misconduct. Research on the link between prison overcrowding and inmate misconduct has been
inconsistent, with some research showing a positive association between the two, other research
(...continued)
functions. The decision unit also covers costs associated with regional and central office administrative and
management support functions such as research and evaluation, systems support, financial management, budget
functions, safety, and legal counsel.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

showing no relationship, and some research even suggesting that there is a negative relationship
between overcrowding and misconduct (i.e., as overcrowding increases, misconduct decreases).49
A group of researchers conducted a meta-analysis that sought to synthesize the results of research
on the overcrowding-misconduct link. Their research, which was based on 16 studies that
provided 120 estimates of the correlation between overcrowding and inmate misconduct,
concluded that, overall, overcrowding did not substantially influence inmate misconduct.50
Based on its own research, in which it collected data from 73 all-male low-, medium-, and highsecurity federal prisons from July 1996 to December 2004, the BOP concluded that there is a
significant positive relationship between overcrowding and misconduct. The analysis conducted
by the BOP included statistical methods to control for stable traits within each prison and to test
the effect of other variables that prior research indicated were related to inmate misconduct. The
BOP estimated that for every one percentage point increase in a prison’s overcrowding (measured
as the ratio of the number of inmates to the prison’s rated capacity), the prison’s annual serious
assault rate increased by 4.1 assaults per 5,000 inmates.51 To put this figure in perspective, in
2009 (the most recent year data are available), BOP data indicate that the annual serious assault
rate was 16 assaults per 5,000 inmates.52
Why do the results of the BOP’s analysis disagree with the results of the meta-analysis described
above? One possible explanation is the relatively long period over which the BOP collected data
for its study. Research has shown that the prevalence of misconduct increased with the length of
the study.53 As such, the results of the BOP’s analysis might suggest a link between overcrowding
and misconduct because it included more data points than the studies included in the meta49

Travis W. Franklin, Courtney A. Franklin, and Travis C. Pratt, “Examining the Empirical Relationship Between
Prison Crowding and Inmate Misconduct: A Meta-analysis of Conflicting Research Results,” Journal of Criminal
Justice, vol. 34, no. 4 (July-August 2006), p. 401, hereinafter “Examining the Empirical Relationship Between Prison
Crowding and Inmate Misconduct.”
50
It has been argued that this body of research does not lend itself to meta-analytic techniques because studies vary in
units of analysis, the definition of prison overcrowding, and attempts to control for other variables that could affect the
overcrowding-misconduct link. In addition, meta-analytic techniques would give the same weight to both rigorous and
cursory studies of the link between overcrowding and misconduct. The researchers incorporated these concerns into
their research design. Their results suggested that the effect sizes calculated using studies of differing methodological
rigor were not dissimilar enough to affect the results of the study. They also found that different definitions of prison
overcrowding did not lead to different results. The researchers address the issue of inconsistency in the use of control
variables across studies by using data from the study to simply test whether there was a bivariate correlation between
prison overcrowding and inmate misconduct, thereby negating the influence of other control variables in the outcome.
While this is the most simplistic test of a relationship between two variables, and it does not exclude any intervening
variables that might explain the relationship between overcrowding and misconduct, it stands to reason that introducing
any control variables into the analysis would only dissipate the strength of any relationship between overcrowding and
misconduct. Because the researchers concluded that there is not a relationship to begin with, it appears unlikely that
additional variables would change their conclusion. Ibid., pp. 407-408. Gerald G. Gaes, Prison Crowding Research
Reexamined, U.S. Department of Justice, Bureau of Prisons, Washington, DC, January 1994, p. 65,
http://www.bop.gov/news/research_projects/published_reports/cond_envir/oreprvariance.pdf.
51
U.S. Department of Justice, Bureau of Prisons, The Effects of Changing Crowding and Staffing Levels in Federal
Prisons on Inmate Violence Rates, Executive Summary, October 2005, hereinafter “The Effects of Changing Crowding
and Staffing Levels in Federal Prisons on Inmate Violence Rates.”
52
Based on data downloaded from U.S. Department of Justice, Bureau of Prisons, Assault Graph Spreadsheets by Year,
http://www.bop.gov/news/research_projects/assaults/assault_spreadsheets/assaults_graph_spreadsheets_by_year.jsp.
The annual assault rate was calculated using the total number of serious assaults against both inmates and correctional
officers and the average annual prison population over each of the 12 months in the year.
53
John Wooldredge, Timothy Griffin, and Travis Pratt, “Considering Hierarchical Models for Research on Inmate
Behavior: Predicting Misconduct with Multilevel Data,” Justice Quarterly, vol. 18, no. 1 (March 2001), p. 212.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

analysis. Also, it has been argued that because of differences between federal and state prisons,
the results of research that tests the link between overcrowding and misconduct in the federal
prison system might not be directly comparable to similar analyses using state-level data.54 If this
is the case, the results of the meta-analysis may not be directly comparable to the results of the
BOP’s study because most of the studies included in the meta-analysis used state-level data.
A Government Accountability Office (GAO) report on prison crowding highlights some of the
problems overcrowding can cause in BOP facilities and how those problems might contribute to
inmate misconduct. The GAO reported that in order to manage the growing inmate population,
the BOP has had to triple or quadruple bunk some inmates and in other instances the BOP has had
to convert common areas, such as a television room, into temporary housing space, which can
result in inmates with a higher propensity for violence spending more time with other inmates.55
In addition, due to prison crowding, inmates may experience crowded bathroom facilities,
reduced shower times, shortened meal times, longer waits for food service, and limited
recreational activities.56 The increasing number of inmates housed in BOP facilities might
decrease the availability of program opportunities, resulting in inmate idleness and waiting lists
for rehabilitative programs like education, vocation training, substance abuse treatment, and faithbased reentry programs.57
The reduction in rehabilitation opportunities can affect the BOP’s ability to manage the prison
population. As mentioned above, inmates who successfully complete a residential substance
abuse treatment program can have up to one year taken off of their sentence. However, the BOP
reported long wait lists for admission to a residential substance abuse treatment program, which
limited the BOP’s ability to admit inmates early enough to allow them to earn the maximum
reduction in their sentences.58 Also, under current law, in order for inmates to earn their full
allotment of good time credit per year, one of the conditions is that the inmate is making
satisfactory progress on completing a GED (assuming the inmate does not have a GED or a high
school diploma).59 Overcrowding can also decrease the number of meaningful work opportunities
available to inmates. Within any given prison there are only so many jobs related to operating and
maintaining the prison for inmates to participate in, and with recent changes to how executive
branch agencies procure goods produced by the Federal Prison Industries (FPI), there are fewer
opportunities for an inmate to work in a FPI factory.60

54
Benjamin Steiner and John Wooldredge, “Rethinking the Link Between Institutional Crowding and Inmate
Misconduct,” The Prison Journal, vol. 89, no. 2 (June 2009), pp. 227-228.
55
U.S. Government Accountability Office, Bureau of Prisons: Growing Inmate Crowding Negatively Affects Inmates,
Staff, and Infrastructure, GAO-12-742, September 2012, p. 18, http://www.gao.gov/assets/650/648123.pdf, hereinafter
“GAO prison crowding report.”
56
Ibid., p. 19.
57
Ibid., pp. 19-20.
58
Ibid., pp. 20-21.
59
18 U.S.C. §3624(b)(1).
60
For more information on changes in how executive branch agencies procure goods produced by the FPI and the
number of inmates working in FPI factories, see CRS Report RL32380, Federal Prison Industries: Overview and
Legislative History, by Nathan James.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

Inmate-to-Staff Ratio
Another issue related to the growth of the federal prison population is increasing inmate-to-staff
ratios. Table 3 shows both the inmate-to-staff ratio and the inmate-to-correctional officer ratio for
the federal prison system since FY2000.61 The inmate-to-staff ratio has held steady
(approximately five inmates to every BOP staff member) since FY2005, but this is higher than the
inmate-to-staff ratio in FY2000 (approximately four inmates to every BOP staff member). The
inmate-to-correctional officer ratio has been approximately twice the inmate-to-staff ratio. The
ratio of inmates to correctional officers was approximately 10 to 1 in FY2012, roughly the same
as it was in FY2000. However, the inmate-to-correctional officer ratio was nearly 11 to 1 in
FY2004 and FY2005. The inmate-to-correctional officer ratio has remained fairly steady even
though the BOP has increased the number of correctional officers by 2% or more in 9 of the past
13 fiscal years. To put these figures in perspective, the Bureau of Justice Statistics reports that in
2005, the inmate-to-staff ratio for all state correctional agencies was 3.3 to 1 and the ratio of
inmates to correctional officers was 4.9 to 1.62 The inmate-to-staff and inmate-to-correctional
officer ratios for the five largest state correctional systems (in terms of inmate population) in 2005
were, respectively,
•

California: 3.6 to 1 and 6.1 to 1,

•

Texas: 4.3 to 1 and 5.9 to 1,

•

Florida: 3.8 to 1 and 4.9 to 1,

•

New York: 2 to 1 and 3 to 1, and

•

Georgia: 3.6 to 1 and 5.4 to 1.63

“Staff” includes all employees of a facility whereas “correctional officers” only include employees whose primary
duties are to supervise inmates.
62
James J. Stephen, Census of State and Federal Correctional Facilities, 2005, U.S. Department of Justice, Office of
Justice Programs, Bureau of Justice Statistics, NCJ 222182, Washington , DC, October 2008, p. 5,
http://bjs.ojp.usdoj.gov/content/pub/pdf/csfcf05.pdf.
63
Ibid., p. 22.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

Table 3. Inmate-to-Staff and Inmate-to-Correctional Officer Ratios
for the Federal Prison System
Number of Inmates per...
Fiscal Year

All BOP Staff

Correctional
Officers

2000

4.1

9.9

2001

4.1

9.7

2002

4.3

10.1

2003

4.5

10.5

2004

4.7

10.8

2005

4.9

10.9

2006

4.9

10.6

2007

4.9

10.6

2008

4.9

10.4

2009

4.9

10.6

2010

4.8

9.8

2011

4.9

10.2

2012

4.8

10.0

Source: Based on a CRS analysis of data provided by the U.S. Department of Justice, Bureau of Prisons.
Notes: Ratios were calculated using figures on the number of inmates held in BOP facilities at the end of the
fiscal year.

While it might appear somewhat surprising that the BOP would have an inmate-to-correctional
officer ratio that was twice the ratio of inmates to staff—especially since the data presented above
indicate that state correctional agencies have inmate-to-correctional officer ratios that are much
closer to inmate-to-staff ratios—the BOP trains all of its staff as correctional officers, which
means that the bureau does not need as many correctional officers to maintain the same level of
security as other correctional agencies. For example, in many state prisons it would not be
uncommon for a correctional officer to be stationed in a classroom while an instructor was
teaching a class. However, since all staff in BOP facilities have been trained as correctional
officers, the BOP does not place a correctional officer in classrooms during instructional periods.
A review of BOP staffing levels by the GAO suggests that federal prison facilities might be
understaffed. The GAO cites a DOJ study from August 2010 that concluded that nearly all BOP
facilities had fewer correctional officers on staff than needed.64 The GAO noted that with the
exception of hiring staff when a new facility opens, the number of staff positions has generally
not increased as the prison population has increased, which is reflected by the slow, but relatively
steady, increase in the inmate-to-staff ratio (the GAO reported that the inmate-to-staff ratio in
FY1997 was 3.6:1).65 The GAO reported that staffing in BOP facilities is, on average, less than
90% of authorized levels. The fact that the inmate-to-staff ratio has increased while the inmate-to64
65

GAO prison crowding report, p. 23.
Ibid., p. 22.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

correctional officer ratio has remained fairly steady might be the result of the way that wardens
make staffing decisions. The BOP funds staffing levels to the extent possible after the costs of
caring for the inmate population (e.g., food, clothing, and medical care) are met. A warden might
choose to use his or her allotted funding to fill more correctional officers positions while leaving
more support staff positions unfilled. The overall inmate-to-staff ratio, which includes staff in the
BOP’s central office and regional offices, might mask staffing issues experienced by individual
facilities. The GAO reported that the BOP calculates a ratio of inmates to institutional staff. From
FY2006 to FY2011, the inmate to total institutional staff ratio for all facilities and for all male
facilities was approximately 5.2:1. Also, the overall inmate-to-staff and inmate-to-correctional
officer ratios do not reflect the fact that those ratios can vary based on the type of institution, the
time of day, and the day of the week.66 For example, the inmate-to-correctional officer ratio might
be higher in low-security facilities compared to medium- and high-security facilities (though,
arguably, this should be expected because lower security inmates should require less supervision).
In addition, the inmate-to-correctional officer ratio might be lower during the night shift than
during the day shifts.
The GAO also reported that staffing levels might be affected by some identified recruitment
challenges the BOP faces. For example, some BOP officials reported that they have had problems
with finding enough qualified candidates.67 Furthermore, officials have reported problems with
hiring professional staff (e.g., psychologists or medical staff) because BOP salaries were less than
those paid for similar work in the surrounding community.68

Ibid. pp. 23-24.
Ibid., p. 22.
68
Ibid.
67

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

Figure 12. Change in BOP Staff, Correctional Officers,
and Institutional Inmate Population
10,000

8,000

6,000

4,000

2,000

2011-2012

2010-2011

2009-2010

2008-2009

2007-2008

2006-2007

2005-2006

2004-2005

2003-2004

2002-2003

2001-2002

2000-2001

-2,000

Fiscal Year
Staff

Correctional Officers

Inmates

Source: Based on a CRS analysis of data provided by the U.S. Department of Justice, Bureau of Prisons.
Notes: Changes in the inmate population is only for inmates incarcerated in BOP facilities, not the total inmate
population which would include inmates held in private facilities.

The data also show that in order to bring down either the inmate-to-staff ratio or the inmate-tocorrectional officer ratio, there would have to be a significant increase in the BOP’s S&E
appropriation relative to the growth in the size of the federal prison population.69
Inmate-to-staff ratios might be an issue for Congress because it could mean that BOP facilities are
less secure. In the research conducted by the BOP that evaluated causes of inmate misconduct
(discussed above), the BOP estimated that a one-inmate increase in a prison’s inmate-to-staff ratio
increased the prison’s annual serious assault rate by 4.5 assaults per 5,000 inmates.70 In addition,
more inmates per staff member could mean less access to rehabilitative programming because
higher inmate-to-staff ratios could mean that BOP would not have the resources to meet the
increasing demand for education, substance abuse treatment, and other rehabilitative programs.
Through the Second Chance Act of 2007 (P.L. 110-199), Congress has required the BOP to help
69

For more information on how the BOP’s appropriations are split between decision units and what funding under
those decision units is used for, see CRS Report R42486, The Bureau of Prisons (BOP): Operations and Budget, by
Nathan James.
70
The Effects of Changing Crowding and Staffing Levels in Federal Prisons on Inmate Violence Rates.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

prepare inmates for re-entry, and the BOP has identified rehabilitative programming as a key
component of its re-entry strategy for federal inmates.71 Therefore, if the BOP does not have the
resources to hire staff at a rate commensurate with the rate of increase in the federal prison
population, the BOP might not be able to fully prepare inmates for their transition back to the
community.

Prison Construction and Maintenance
Funds for new prison construction and expansion and the modernization and repair of existing
facilities come from BOP’s Buildings and Facilities (B&F) account. Appropriations for the B&F
account have fluctuated over the past 32 fiscal years. Congress made a substantial investment in
prison construction and expansion from FY1989-FY1992, with an appropriation of nearly $3
billion for the B&F account. There was another spike in appropriations for the B&F account
between FY1999 and FY2004, when Congress appropriated approximately $3.4 billion. A
comparison of the history of appropriations for the B&F account and historical overcrowding
levels in BOP facilities shows that there is a lag between Congress appropriating funding for
additional bedspace and a reduction in prison overcrowding. This is because it can take several
years for the BOP to identify a location for a new prison, award the contracts for construction of
the facility, complete construction, and bring the prison “online” by hiring new employees to staff
the facility.
Data provided by the BOP indicate that it is becoming more expensive to expand federal prison
capacity. The data presented in Figure 13 represent the total cost of each prison the BOP has built
since 1994 divided by the rated capacity of the facility; this provides an indication of how much it
has cost the BOP to expand its rated capacity by one inmate over the past 18 years. The data show
that, in general, the per inmate cost of expanding the system’s rated capacity increased.

BOP’s FY2013 S&E Budget Justification, p. 33.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

Figure 13. The Cost of Expanding Rated Prison Capacity by One Inmate
250,000

Cost of per inmate rated capacity

200,000

150,000

100,000

50,000

Mendota FCI

Butner FCI

McDowell FCI

Tucson USP

Bennettsville FCI

Herlong FCI

Victorville FCI

Yazoo City FCI

Williamsburg FCI

Gilmer FCI

Big Sandy USP

Atwater USP

Lee USP

Victorville FCI

Philadelphia FDC

Brooklyn FDC

Edgefield FCI

Beaumont FCC

Forrest City FCI

Taft FCI

Beckley FCI

Cumberland FCI

Florence FCC

Greenvile FCI

Source: Based on a CRS analysis of data provided by the U.S. Department of Justice, Bureau of Prisons.
Notes: Each bar in Figure 13 represents the total cost of building a new prison divided by the number of beds
in the completed prison. The data are presented in chronological order, with the first bar on the left (Greenville
FCI) representing a facility completed in 1994 and the last bar on the right (Aliceville FCI; not labeled)
representing a facility completed in 2011.

The growing prison population and related overcrowding is contributing to the deterioration of
the BOP’s facilities. As discussed above, the B&F account, in addition to providing funding for
prison construction, also provides funding for the maintenance of the BOP’s prison facilities. The
BOP currently operates 117 prison facilities, and approximately one-third of those facilities are
more than 50 years old.72 Moreover, failure to perform adequate maintenance on existing
facilities can result in larger capital investment in future years as prisons and utility systems
deteriorate and, according to the BOP, can cause direct and/or indirect security problems.73
According to the BOP, some prisons are experiencing “extensive wear and tear, as well as
premature deterioration” because the facilities are holding more inmates than they were originally
designed to hold.74 The BOP reports that it has a backlog of 154 modernization and repair (M&R)

72
U.S. Department of Justice, Bureau of Prisons, FY2013 Performance Budget, Congressional Submission, Buildings
and Facilities, p. 5, http://www.justice.gov/jmd/2013justification/pdf/fy13-bop-bf-justification.pdf.
73
Ibid.
74
Ibid., p. 23.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

projects with an approximate cost of $349 million for FY2012,75 which is up from a backlog of
134 projects with an approximate cost of $302 million in FY2011.76 The amount required to work
through the current backlog of M&R projects is more than what Congress has appropriated for
M&R projects going back to FY1999 (see Table A-2). The estimate only reflects “major” projects
(repairs costing $300,000 or more) that have been approved by the BOP’s administration for
funding when appropriations are available. As such, the $349 million estimate probably does not
capture the full extent of all needed repairs.
Appropriations for the B&F account have enabled the BOP to expand bedspace at a rate where it
can manage overcrowding but not reduce it. The federal prison population is increasing at a rate
where new bedspace is accounted for as soon as it is available. However, recent reductions in
funding for the New Construction decision unit under the B&F account mean that the BOP will
lack the funding to begin any new prison construction in the near future, which could result in
increased overcrowding in the federal prison system (see Table A-2).
The current state of expansion in prison capacity means that the BOP cannot close older prisons.
It is likely that more money will have to be spent to properly maintain a prison the longer it is in
operation, especially when it is housing more inmates than it was rated to hold. In addition to
having lower maintenance costs, newer prisons also have the advantage of updated designs and
technology upgrades that allow for fewer correctional officers to safely monitor more inmates.

Select Policy Options
The analyses presented above show that the growth in the federal prison population over the past
three decades has resulted in an increasingly expensive federal prison system that is overcrowded
and aging and where facilities might not be staffed at an optimal level. Congress could choose to
address the mounting number of federal inmates either in the context of existing correctional
policies or by changing the current policies. Specific policy options under these two courses are
discussed in more detail below.

Continuing or Expanding Current Correctional Policies
Under the umbrella of continuing existing policies, Congress could consider addressing issues
related to the burgeoning federal prison population by (1) expanding the capacity of the federal
prison system, (2) continuing to invest in rehabilitative programming, (3) placing more inmates in
private correctional facilities, or some combination of the three.

Expanding the Capacity of the Federal Prison System
Arguably one of the most straightforward approaches for managing the steadily increasing
number of federal inmates is to expand the capacity of the federal prison system. Congress could
choose to mitigate some issues related to federal prison population growth by appropriating more
funding so the BOP could expand prison capacity to alleviate overcrowding, update and properly
75

Ibid., p. 27.
U.S. Department of Justice, Bureau of Prisons, FY2012 Performance Budget, Congressional Submission, Buildings
and Facilities, p. 25, http://www.justice.gov/jmd/2012justification/pdf/fy12-bop-bf-justification.pdf.
76

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

maintain existing facilities, and hire additional staff. While a large-scale expansion of the federal
prison system might help reduce overcrowding, it takes several years for a prison to be built and
be ready to accept inmates, so if Congress chooses to appropriate funding for an expansion of the
BOP’s infrastructure, it could be several years before overcrowding is reduced. Even if the federal
prison population stabilized at the current level, the BOP would have to add over 50,000
additional beds in order to eliminate overcrowding. The BOP projects the federal prison
population to increase to nearly 250,000 inmates by FY2018.77 These projections are based on
what has happened in the past, and as discussed above, the federal prison population has been
increasing for more than 30 years. Should Congress choose to invest in a wide-scale expansion of
prison capacity, and the prison population decreases in the future, the surplus bedspace could
allow the BOP to close some of its older facilities, which, in general, require more maintenance
and need higher inmate-to-staff ratios to safely operate.
Critics contend that expanding the capacity of the federal prison system does not address the
continued growth of the federal prison population. Also, this policy option would not resolve the
issue of the rising cost of the federal prison system; in fact, it could exacerbate it. However,
alternatives that would reduce the federal prison population would most likely involve
prosecuting fewer people in federal courts, providing ways for inmates to be released before they
served a significant proportion of their sentences, putting more inmates into diversionary
programs, or placing more offenders on some form of community supervision. If Congress does
not wish to take any of these steps, a large-scale expansion of the federal prison system might be
the sole way to manage the effects of an increasing prison population. Some may argue that in
order to protect public safety Congress should appropriate the funding necessary to expand the
federal prison system rather than adopt policy changes that would reduce the prison population
through early releases, alternatives to incarceration, or fewer prosecutions.

Investing in Rehabilitative Programs
A review of the literature on rehabilitative programs (e.g., academic and vocational education,
cognitive-behavioral programs, and both community- and prison-based drug treatment) suggests
that there are enough scientifically sound evaluations to conclude that they are effective at
reducing recidivism, which could potentially help stem growth in the federal prison population in
the future.78 The BOP offers a variety of rehabilitation programs such as academic and vocational
education, work programs through the Federal Prison Industries (FPI), substance abuse treatment,
and cognitive-behavioral programs that focus on promoting pro-social behavior.79 One possible
option for reducing the federal prison population would be to ensure that the BOP has the
resources it needs to provide rehabilitative services to inmates.
At a time when some policymakers are considering reducing discretionary funding for federal
agencies, there might be some effort to restrain the growth of the BOP’s appropriations, including
for rehabilitative services. The BOP has to administer the federal prison system within the funds
appropriated for it by Congress. As shown in Figure 9, the per capita cost of providing
77

U.S. Department of Justice, Bureau of Prisons, FY2012 Performance Budget, Congressional Submission, Salaries
and Expenses, p. 2, http://www.justice.gov/jmd/2012justification/pdf/fy12-bop-se-justification.pdf.
78
Doris Layton MacKenzie, What Works in Corrections: Reducing the Criminal Activities of Offenders and
Delinquents (New York: Cambridge University Press, 2006), pp. 331-333, hereinafter, “What Works in Corrections.”
79
For more information on the BOP’s rehabilitative programs see CRS Report R41525, Federal Prison Inmates:
Rehabilitative Needs and Program Participation, by Nathan James.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

rehabilitative programming for inmates has been increasing. If the BOP does not have sufficient
resources, it might not be able to provide rehabilitative programming to all inmates who need it.
The federal prison population has continued to grow, and unless there is a change in the trend
over the past 30 years, it would appear likely that there will be a growing need for and cost of
rehabilitative programming in the federal prison system.
It could be argued that in order to reduce the growing cost of operating the federal prison system,
the BOP should reduce funding for rehabilitative programming and invest solely in providing for
the subsistence of inmates and maintaining a level of staffing that is adequate to ensure that
federal prisons are secure. However, reducing programming opportunities might result in more
inmate idleness, which might result in more inmate misconduct. Moreover, as noted above, the
BOP is authorized to reduce an inmate’s sentence by up to one year for successfully completing a
residential substance abuse treatment program; therefore, reducing programming opportunities
could hamper one of the few avenues the BOP has for releasing inmates early. It is also possible
that BOP might be able to realize some long-term cost savings by successfully rehabilitating
inmates. For example, research by the Washington State Institute for Public Policy (WSIPP)
suggests that effective rehabilitation programs can result in cost savings.80
As policymakers consider the appropriate level of funding for the BOP in light of concerns about
the federal deficit and potential freezes or reductions in non-defense discretionary spending, they
could consider whether it is prudent to increase resources for the BOP’s rehabilitative programs in
the near-term in order to realize potential long-term benefits. As outlined above, the BOP’s
appropriations have increased along with the federal prison population. Yet, current funding may
be inadequate for the BOP to provide more rehabilitative opportunities to federal inmates. The
size of the effect that decreased recidivism amongst federal offenders would have on the BOP’s
budget would depend on how many new inmates the BOP incarcerates. If new commitments
exceed the number of inmates released who do not return to prison then the demand for prisons,
personnel, and inmate programs and services would continue to grow, although possibly at a
slower rate. If the number of new commitments is less than the number of inmates released who
do not return to prison then the demand for prisons, personnel, and inmate programs and services
would decrease. However, even if the growth of the federal prison population slows, the demand
for increased BOP appropriations may continue.

Placing More Inmates in Private Prisons
The BOP has placed an increasing share of federal inmates in contract facilities as a way of
managing the growth in the federal prison population. Congress might also consider whether
more federal inmates should be housed in private facilities as a means of reducing crowding in
federal prisons and potentially reducing the cost of operating the federal prison system. The
number of inmates under the BOP’s jurisdiction held in contract facilities has steadily increased
since the early 1980s, and the BOP expects the trend to continue into the later part of this decade.
However, the growth in the number of inmates held in contract facilities is mostly the result of
more inmates being placed in Residential Reentry Centers (RRCs) at the end of their sentences.
Most BOP inmates held in private correctional facilities are low-level, non-citizen offenders. The
debate about whether to house inmates in privately operated correctional facilities has been
80
Steve Aos, Marna Miller, and Elizabeth Drake, Evidence-based Public Policy Options to Reduce Future Prison
Construction, Criminal Justice Costs, and Crime Rates, Washington State Institute for Public Policy, Olympia, WA,
October 2006, http://www.wsipp.wa.gov/rptfiles/06-10-1201.pdf.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

framed by two overarching questions: (1) can private facilities incarcerate inmates at a lower cost
and (2) can private facilities provide services that are equal or superior to the services provided in
public institutions?
The BOP attempted to answer these questions, at the behest of Congress, by operating the Taft
Correctional Institution (TCI) as a private facility as a part of a privatization demonstration
project.81 The BOP awarded a contract to the Geo Group (formerly Wackenhut Corrections
Corporation) to operate the facility from 1997 to 2007.82 The BOP, through the National Institute
of Justice, funded an evaluation of TCI and three similar BOP facilities,83 which was conducted
by Abt Associates, Inc. In addition, the BOP’s Office of Research, in conjunction with the Center
for Naval Analyses (CNA) Corporation, conducted its own evaluation of TCI and the similar
facilities. Both the Abt and BOP evaluations found that TCI was cheaper to operate on a per diem
basis than the three comparable facilities, but the two evaluations offer different conclusions as to
how much was saved by operating TCI as a private institution. The Abt analysis concluded that
the average per diem cost of incarceration for the three BOP-operated facilities in FY1999 was
18.9% greater than the per diem cost of incarceration for TCI; in FY2000 it was 20.0% greater; in
FY2001 it was 17.5% greater; and in FY2002 it was 14.8% greater.84 In comparison, the BOP
analysis concluded that the average per diem cost of incarceration for the three BOP-operated
facilities in FY1999 was 4.0% greater than the per diem cost of incarceration for TCI; in FY2000
it was 5.4% greater; in FY2001 it was 0.3% greater; and in FY2002 it was 2.2% greater.85 The
two primary reasons for the different conclusions are economies of scale86 realized by TCI and
differences in how per diem rates were calculated.87 TCI had on average 300 more inmates each
year than the three BOP-operated prisons, which means that TCI was able to take advantage of
economies of scale that decreased average costs. In the BOP analysis, the researchers adjusted for
these economies of scale by estimating what expenditures would have been for the BOP facilities
if they had prison populations similar to TCI. In addition, the Abt analysis assumed that the BOP
would not provide many resources to support TCI’s operations, resulting in a large amount of
savings from reduced indirect overhead costs. The BOP analysis assumed that the BOP would
continue to incur some overhead expenses related to overseeing TCI. As such, the BOP included a
10-12% overhead rate in its analysis.
Research that reviewed the results of state and local efforts to privatize correctional systems
generally found that it is questionable whether privatization can deliver lower costs and whether

The conference report (H.Rept. 104-863) for the Omnibus Consolidated Appropriations Act, 1997 (P.L. 104-208)
incorporates, by reference, language from the Senate report (S.Rept. 104-353) to accompany the Senate committeereported version of H.R. 3814 (104th Congress), that requires the BOP to undertake “a 5-year prison privatization
demonstration project” involving the facility that the BOP built in Taft, CA.
82
The Taft Correctional Institution is still operating as a private facility. After the contract with the Geo Group expired
in 2007, the contract was recompeted and it was awarded to Management and Training Corporation.
83
The three similar facilities included in the evaluation were FCI Yazoo City, FCI Elkton, and FCI Forrest City.
84
Gerry Gaes, “Cost, Performance Studies Look at Prison Privatization,” NIJ Journal , no. 259 (March 2008), p. 33.
85
Ibid.
86
“Economies of scale” generally refers to the increase in efficiency of production that accompanies expanded
production. In economic terms, this means that the average cost of the good produced decreases and production
increases because fixed costs are shared over an increased number of goods. In terms of the BOP evaluation of TCI,
“economies of scale” would refer to the decreased per prisoner costs resulting from spreading the prison’s operating
costs over a greater number of inmates.
87
Ibid., p. 34.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

services provided by private prisons are comparable to services provided by public prisons.88 One
of the first studies to quantitatively summarize the results of several evaluations of prison
privatization efforts found that regardless of whether the prison was privately or publicly
operated, the economies of scale, the prison’s age, and the prison’s security level were the most
significant determinants of the daily per diem cost.89 The researchers concluded that “[a]lthough
specific privatization policy alternatives may result in modest cost savings…relinquishing the
responsibility of managing prisons to the private sphere is unlikely to alleviate much of the
financial burden on state correctional budgets.”90 Their conclusions are echoed by a review of the
literature on privatization. In this analysis, the researchers concluded “that prison privatization
provides neither a clear advantage nor disadvantage compared with publicly managed prisons.
Neither cost savings nor improvements in quality of confinement are guaranteed through
privatization.”91 However, even though both studies limited their analyses to the most
methodologically sound evaluations, these evaluations are still limited to the same issues
described above, namely, what costs are considered when the evaluators calculated whether
privatization could lower correctional costs. As discussed above, these assumptions can have a
considerable effect on the results of the evaluation.
Placing more inmates in private facilities could help alleviate overcrowding in federal prisons
without the need to invest in a large-scale expansion of federal prison bedspace. Expanding
capacity through contracting for additional bedspace rather than building new prisons could give
Congress the flexibility to reduce capacity if the federal prison population decreased in the future.
However, research suggests that moving federal prisoners into private prisons might not help to
control the rising costs of the federal prison system. Also, medium and high security facilities are
the most crowded, and the BOP is less inclined to place medium and high security inmates in
private facilities. Congress might also consider whether it wants to place a greater portion of the
federal prison population in the custody of private operators when the BOP has less direct
oversight over the day-to-day operations of private facilities.

Changing Existing Correctional and Sentencing Policies to Reduce
the Prison Population
Policymakers might also consider whether they want to revise some of the changes that have been
made to federal criminal justice policy over the past three decades. A confluence of these changes
has resulted in an increasing number of offenders being sent to federal prisons. Should Congress
decide to change federal criminal justice policy to try to reduce the number of inmates held in
federal prisons, policymakers might start by considering which offenders are incarcerated and the
length of their sentences.
88
Travis C. Pratt and Jeff Maahs, “Are Private Prisons More Cost-effective Than Public Prisons? A Meta-analysis of
Evaluation Research Studies,” Crime and Delinquency, vol. 45, no. 3 (July 1999), pp. 358-371; Dina Perrone and
Travis C. Pratt, “Comparing the Quality of Confinement and Cost-effectiveness of Public Versus Private Prisons: What
We Know, Why We Do Not Know More, and Where to Go From Here,” The Prison Journal, vol. 83, no. 3 (September
2003), pp. 301-322; Brad W. Lundahl, Chelsea Kunz, and Cyndi Brownell, et al., “A Meta-analysis of Cost and Quality
of Confinement Indicators,” Research on Social Work Practice, vol. 19, no. 4 (July 2009), pp. 383-394.
89
Travis C. Pratt and Jeff Maahs, “Are Private Prisons More Cost-effective Than Public Prisons? A Meta-analysis of
Evaluation Research Studies,” Crime and Delinquency, vol. 45, no. 3 (July 1999), p. 367.
90
Ibid., pp. 367-368.
91
Brad W. Lundahl, Chelsea Kunz, and Cyndi Brownell, et al., “A Meta-analysis of Cost and Quality of Confinement
Indicators,” Research on Social Work Practice, vol. 19, no. 4 (July 2009), p. 392.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

Changes to Mandatory Minimum Penalties
As discussed earlier, the USSC concluded that, in part, mandatory minimum penalties have
contributed to the growing federal prison population. It might be argued that some or all
mandatory minimum penalties should be repealed as a way to manage the growth of the federal
prison population. Allowing defendants to be sentenced using the federal sentencing guidelines
could allow for more individualized sentencing, thereby allowing the court to mete out
punishment using an array of variables that reflect a more nuanced analysis of a defendant’s
culpability. Opponents of widespread use of mandatory minimum penalties contend that they are
a blunt instrument with which to determine a proper sentence. The USSC reported that “certain
mandatory minimum provisions apply too broadly, are set too high, or both, to warrant the
prescribed minimum penalty for the full range of offenders who could be prosecuted under the
particular criminal statute.”92 Also, to the extent that mandatory minimum penalties have
contributed to sentence inflation as a result of the USSC incorporating them into the federal
sentencing guidelines, repealing some mandatory minimum penalties might reduce the amount of
time inmates serve in federal prison.
However, proponents of the continued use of mandatory minimum penalties contend that after the
Supreme Court’s ruling in United States v. Booker93 and its progeny (e.g., Gall v. United States94
and Kimbrough v. United States95), which rendered the sentencing guidelines effectively advisory,
Congress has a responsibility to set minimum penalties for some offenses as a way to limit
judicial discretion, thereby preventing unwanted sentencing disparities. It has been argued that
mandatory minimum penalties promote uniformity and fairness for defendants, transparent and
predictable outcomes, and a higher level of truth and integrity in sentencing.96 Also, should
Congress choose to repeal some or all mandatory minimum penalties, policymakers would
relinquish their ability to control the amount of time inmates serve for certain offenses.
Even if Congress chooses not to repeal any mandatory minimum sentences, policymakers could
review current mandatory minimum penalties to ensure that they are (1) not excessively severe,
(2) narrowly tailored to apply only to those offenders who warrant such punishment, and (3)
applied consistently.97

Alternatives to Incarceration
During the 1980s many states instituted a series of alternatives to incarceration as a way to
respond to an increasing number of convicted offenders and wide-scale prison overcrowding.98
Prior to this, sentencing options were limited to incarceration or probation. However, there was
growing sentiment that some crimes were too severe to be punished by placing the offender on
92

Mandatory Minimum Penalties in the Federal Criminal Justice System, p. 345.
543 U.S. 220 (2005).
94
552 U.S. 38 (2007).
95
552 U.S. 85 (2007).
96
Erik Luna and Paul G. Cassell, “Mandatory Minimalism,” Cardozo Law Review, vol. 32, no. 1 (September 2010), p.
11.
97
Mandatory Minimum Penalties in the Federal Criminal Justice System, p. 368.
98
What Works in Corrections, p. 304.
93

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

probation, but those same crimes were not severe enough to warrant incarceration. Therefore,
states started to develop a series of alternative sentences that fell somewhere between probation
and incarceration. These alternatives included house arrest, electronic monitoring, intensive
supervision, boot camps, split sentences, day reporting centers, fines, and community service.99
The programs provide graduated sanctions that might be more appropriate than either probation
or incarceration, and provided a higher level of offender restraint and accountability than
traditional probation. Some also provide higher levels of treatment or services for problems such
as substance abuse, low education levels, and unemployment.
A majority of the evaluations of intensive supervision and electronic monitoring programs found
that there was no significant difference in recidivism rates between offenders sentenced to
alternatives to incarceration and offenders in control groups.100 This means that increasing
surveillance and control of offenders’ activities does not decrease their criminal activities.
Ironically, while these programs were created as a means of reducing the number of incarcerated
individuals, the increased surveillance might increase the probability that violations of the terms
of probation will be detected, which could increase the number of inmates as probationers are
often incarcerated for technical violations. One shortcoming of the research is that since most
intensive supervision programs increase the probability of detection, there is no way to tell if the
underlying level of criminality changed between the treatment and control groups, i.e., the
increased probability of detection might mean that offenders in the control group are simply more
likely to be caught when they commit crimes, even though offenders in the control group commit
crime at the same, or even higher, rate. Also, the research tended to focus on whether the
restraining aspects of the program could reduce recidivism. Some evaluations found that inmates
who received treatment while participating in an intensive supervision program were less likely to
be arrested.101

Placing More Inmates on Probation
Congress could consider whether there are alternative ways to properly manage offenders
convicted of committing relatively minor crimes without sending them to prison. Data from BJS
show that in FY2010 over half of inmates entering federal prison were sentenced to three years or
less. Given the relatively short sentences these inmates received, it is likely that they were
sentenced for relatively minor offenses. One policy option Congress could consider is amending
penalties for some offenses to allow more defendants to be placed on probation rather than being
sentenced to a period of incarceration. However, the Booker decision that rendered the federal
sentencing guidelines advisory might influence any debate Congress would have over who would
be placed on probation. The sentencing guidelines placed substantial restrictions on when courts
could sentence defendants to probation. Under §5B1.1 of the sentencing guidelines, defendants
can only be placed on probation if their sentence under the guidelines is equal to or less than 15
months. Nonetheless, after the Supreme Court’s ruling in Booker, federal judges are not required
to impose a sentence within the range calculated under the sentencing guidelines. Therefore,
judges can impose probation for offenders unless (1) the defendant has been convicted of a class
A or B felony,102 (2) probation is statutorily precluded as a sentencing option, or (3) the defendant
99

Ibid.
Ibid., p. 306.
101
Ibid., p. 318.
102
A class A felony is an offense where the maximum term of imprisonment authorized is life imprisonment or death.
A class B felony is an offense where the maximum term of imprisonment authorized is 25 years or more. 18 U.S.C.
(continued...)
100

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

is sentenced to a term of imprisonment for the same or different offense that is not a petty
offense.103
Data from the Administrative Office of the U.S. Courts suggest that in the post-Booker era courts
have not chosen to sentence outside the guideline range to place more offenders on probation.
Since the mid-1970s, a dwindling proportion of defendants sentenced for federal offenses were
placed on probation (see Figure 14). The decrease in the proportion of sentenced defendants
placed on probation coincides with the implementation of the federal sentencing guidelines. The
data also show that the proportion of defendants that were placed on probation has further
decreased since the Booker decision. Approximately 14% of defendants in FY2004 received
probation; by FY2011, about 10% of defendants were placed on probation.
Data show that the risk of recidivism for probationers is the highest in the first year after being
placed on probation.104 It has been argued that surveillance and services should be front-loaded
(i.e., more intensive at the beginning of a term of probation) to try to mitigate recidivism and
other negative consequences that might occur during the first year that an offender is serving on
probation.
A common argument from advocates of decreasing the use of incarceration is that it is cheaper to
supervise an offender in the community than it is to incarcerate that individual. The
Administrative Office of the U.S. Courts reports that the average annual cost of probation
supervision was $3,938 per probationer.105 In comparison, the average annual cost of
incarceration for FY2010 was $25,627 per inmate. However, some of the lower cost of probation
relative to incarceration might be the result of fewer and lower-risk offenders being placed on
probation. It is possible that the annual cost of probation would increase if Congress expanded the
number of people placed on probation and implemented some of the changes discussed below.

(...continued)
§3559(a).
103
18 U.S.C. §3561(a).
104
Ibid., p. 519.
105
Administrative Office of the U.S. Courts, “Newly Available: Costs of Incarceration and Supervision in FY 2010,”
press release, June 23, 2011, http://www.uscourts.gov/news/newsview/11-06-23/
Newly_Available_Costs_of_Incarceration_and_Supervision_in_FY_2010.aspx.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

Figure 14. Proportion of Sentenced Defendants in Federal Courts
Placed on Probation
60.0%

50.0%

40.0%

30.0%

20.0%

10.0%

2010

2008

2006

2004

2002

2000

1998

1996

1994

1992

1990

1988

1986

1984

1982

1980

1978

1976

1974

1972

1970

0.0%

Fiscal Year
Source: FY1970-FY1996 data taken from Sourcebook of Criminal Justice Statistics, 1996, Table 5.27. Data for
FY1997-FY2011 taken from Judicial Business of the U.S. Courts for each respective fiscal year.

Should Congress choose to expand probation as a sentencing option for more offenses, research
suggests that probation programs that use a validated risk assessment tool to sort offenders into
high- and low-risk groups and focus resources and supervision on higher-risk offenders might be
more effective at reducing recidivism.106 Research also suggests that probation programs that
offer a mix of evidence-based treatment that is delivered to offenders who are the most likely to
benefit from it along with surveillance are more effective at reducing recidivism than
surveillance-only probation.107 As one expert noted, “‘[t]reatment’ alone is not enough, nor is
‘surveillance’ by itself adequate. Programs that can increase offender-to-officer contact and
[emphasis original] provide treatment have reduced recidivism.”108 Researchers have found that
participants in probation programs that subject probationers with substance abuse issues to
frequent random drug testing and that require probationers who violate the terms of their
probation to serve intermediate sanctions, such as a short stay in jail, are less likely to recidivate
106

Joan Petersilia, “Community Corrections: Probation, Parole, and Prisoner Reentry,” in Crime and Public Policy, ed.
James Q. Wilson and Joan Petersilia (New York: Oxford University Press, 2011), pp. 521-522, hereinafter,
“Community Corrections: Probation, Parole, and Prisoner Reentry.”
107
Ibid., p. 522.
108
Ibid.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

than those who were on regular probation. 109 Another option Congress might consider is allowing
probationers who strictly adhere to their conditions of probation to be released early. Research
has shown that an earned discharge strategy can reduce recidivism.110

Expanding the Use of Residential Reentry Centers
Congress could also consider extending the BOP’s authority to place inmates with short sentences
who are deemed to be low security risks directly into Residential Reentry Centers (RRCs, i.e.,
halfway houses). A New York Times (Times) investigation of halfway houses in New Jersey might
raise some questions amongst policymakers about whether placing some federal inmates in RRCs
rather than federal prisons is a viable policy option. Like the BOP, the New Jersey Department of
Corrections contracts with private halfway houses to give inmates the opportunity to reestablish
themselves in society before being released from custody. The Times investigation found that
approximately 5,100 inmates have escaped since 2005 and that some inmates have committed
new offenses after they escaped.111 However, many escaped inmates were captured or returned
within hours or days.112 The Times also uncovered instances of lax security because counselors
were either poorly trained, outnumbered, or feared for their safety; inmate-on-inmate violence;
and questionably delivered rehabilitative services.113
The Department of Justice’s Office of the Inspector General (OIG) completed an audit of six
RRCs114 that the BOP had contracts with to determine whether the RRCs operations were
conducted in compliance with the BOP’s requirements and whether the BOP effectively
administers and monitors RRC contracts.115 The OIG’s audit included a review of the files of 177
inmates who were transferred to one of the six RRCs between FY2008 and FY2010. The OIG
concluded that the RRCs adequately met most Statement of Work (SOW)116 requirements;
however, the six RRCs did not fully comply with conditions related to substance abuse testing,
escapes, and authorized inmate absences. Specifically, the OIG found that 30% of the inmates
with histories of substance abuse did not have documentation showing that they were given
required drug tests.117 The OIG found that from FY2008 to FY2010, approximately 3% of the
109

Kevin McEvoy, “HOPE: A Swift and Certain Process for Probationers,” NIJ Journal, no. 269 (March 2012), p. 17.
Community Corrections: Probation, Parole, and Prisoner Reentry, p. 524.
111
Sam Dolnick, “As Escapees Stream Out, a Penal Business Thrives,” New York Times, June 17, 2012, p. A1.
112
Ibid., p. A16.
113
Sam Dolnick, “Poorly Staffed, a Halfway House in New Jersey is Mired in Chaos,” New York Times, June 18, 2012;
Sam Dolnick, “At Penal Unit, A Volatile Mix Fuels a Murder,” New York Times, June 19, 2012.
114
The RRCs were located in Denver, CO; Leavenworth, KS; El Paso, TX; Boston, MA; Washington, DC; and Kansas
City, MO.
115
U.S. Department of Justice, Office of the Inspector General, Audit of the Federal Bureau of Prisons’ Contracting for
and Management of Residential Reentry Centers, Audit Report 12-20, Washington , DC, March 2012, p. 5,
http://www.justice.gov/oig/reports/2012/a1220.pdf.
116
The Statement of Work (SOW) for RRCs is the primary document that outlines all of the BOP’s requirements for
contractors operating RRCs. The SOW requires RRCs to develop individualized program plans for each inmate that
focus on, when applicable, reestablishing relationships with family, obtaining and maintaining employment, obtaining
drug and alcohol abuse treatment, and finding housing. At the same time each RRC must: (1) be able to locate and
verify the whereabouts of inmates at all times; (2) establish an surveillance program to deter and detect the illegal
introduction of drugs and alcohol into the facility; (3) effectively discipline inmates; and (4) prepare and maintain
required documentation. Ibid., p. 3.
117
RRCs are required to randomly test at least 5% of all inmates for drugs and alcohol monthly, with a minimum of
one inmate tested per month. In addition, an inmate with a known history of drug abuse, or who is suspected of illegal
(continued...)
110

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

inmates transferred to the six RRCs included in the audit escaped.118 In addition, the OIG found
that 92 of the 177 inmates included in the audit had 434 authorized absences where they returned
to the RRC more than one hour late, and no disciplinary action was taken or documented for 65
of the 71 inmates who returned more than one hour late with no reason for their tardiness.119 The
OIG also concluded that the BOP’s monitoring procedures were sufficient to identify most RRC
deficiencies related to compliance with the SOW and that corrective actions were implemented.
The Times investigation of halfway houses in New Jersey and the OIG’s audit of RRCs might
raise some potential issues for Congress. The Times report suggests that several of the problems
experienced in the halfway houses that were the subject of its investigation resulted from the New
Jersey Department of Corrections and local sheriffs’ departments using halfway houses as a
means of reducing prison and jail overcrowding, which resulted in inmates with violent histories
and/or who were convicted for violent offenses being placed in halfway houses. These inmates
were then supervised by employees with little training, who were not correctional officers and
who, in some instances, feared the inmates because they were substantially outnumbered. This
suggests that if Congress wanted to use RRCs as a way of reducing overcrowding in federal
prisons that placement in RRCs should be limited to low-level, non-violent offenders. The Times
article includes accounts from staff who reported fearing for their safety while patrolling the
halfway houses at night because of lax security and high inmate-to-staff ratios. This might mean
that should RRCs be used as a way to reduce the number of inmates held in federal prisons, the
BOP will need to ensure that RRCs have properly trained and adequate staff and that the RRCs
have satisfactory security measures in place. The findings from the OIG audit suggest that the
BOP might need to increase its oversight of the RRCs it contracts with. This could mean that the
BOP would need additional staff and an increase in its travel budget so BOP staff could make
more frequent visits to RRCs.
If policymakers were concerned about whether RRCs are a valid alternative to placing some
offenders in federal prison, Congress could choose to provide funding for a program that would
allow the federal government to contract with local jails to provide short-term bedspace. One
possible example is the Cooperative Agreement Program (CAP) whereby the U.S. Marshals
Service (USMS) provided capital investment funding to local jails in exchange for guaranteed
(...continued)
drug use, must be tested no less than four times per month. RRCs are also required to give inmates a breathalyzer test
every time they return from an unsupervised activity. Ibid., pp. 7-8.
118
The OIG notes that there are three types of escapes: regular, technical new, and technical old. “Regular” escapes
most closely fit the definition of what most would consider an escape. A regular escape is when an inmate fails to
remain in RRC custody by (1) not reporting to the facility for admission at the scheduled time; (2) not remaining at the
approved place of employment, training, or treatment during the hours specified by the terms of the employment,
training, or treatment program; (3) not returning to the facility at the time prescribed; (4) not returning from an
authorized furlough or pass at the time and place stipulated; (5) not abiding by conditions of employment, or curfew
conditions of home confinement; or (6) leaving the facility without the permission of RRC staff. “Technical new”
escapes occur when an inmate fails to remain in the RRC’s custody by being arrested for a new charge. “Technical old”
escapes occur when an inmate fails to remain in the RRC’s custody by being arrested for an outstanding warrant while
residing at a RRC. Of the escapes reported in the OIG’s report, 75.3% were regular escapes, 12.7% were technical new
escapes, and 11.9% were technical old escapes. Ibid. p. 15.
119
RRCs can only authorize an inmate to leave for approved activities, including job searches, employment, religious
services, and visitations with family and friends. Other than for employment or programming activities, such as drug
abuse counseling, an inmate must generally be at the RRC facility from 9:00 p.m. to 6:00 a.m., unless the director has
granted an exemption. RRCs are allowed to grant passes or furloughs to release inmates overnight or to travel distances
of more than 100 miles. Ibid., pp. 18-19.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

bedspace for federal detainees.120 While the CAP was limited to securing bedspace for people in
the custody of the USMS (i.e., people who have not yet been convicted of a crime), it is possible
that the program could be expanded to allow the federal government to expand local jail capacity
in order to secure bedspace for some lower-level federal inmates who are serving short sentences.
It is likely that jails would be more secure than RRCs. In addition, jails are staffed by correctional
officers, who might be better prepared to supervise federal inmates.
Allowing the BOP to confine more low-level inmates in RRCs would mean fewer inmates would
be placed into already overcrowded facilities while still receiving a punishment for criminal
behavior and supervision of their actions for a given period of time. However, FY2011 data
published by the BOP show that per capita expenditures for RRCs ($26,163) were higher than the
per capita cost of confining a prisoner in a minimum ($18,849), low ($23,780), or medium
($23,780) prison, so placing more low-level inmates in RRCs might not generate a substantial
amount of savings.121
Congress could also consider whether to require courts to place certain offenders in RRCs for
violating the terms of their supervised release rather than returning them to prison. As mentioned,
the BOP might not save a significant amount of money by placing a greater number of inmates in
RRCs, but by placing more of these short-term inmates in RRCs the BOP would have additional
bedspace. In addition, the BOP would not have to invest time and money into re-processing the
offender through the prison system.122 This is not to suggest that all inmates who have their
supervised release revoked would be suitable for RRC placement. Indeed, inmates who are
arrested and/or convicted for serious offenses would most likely need to be placed in a secure
facility. However, offenders who have their supervised release revoked for technical violations
(e.g., repeatedly failing drug tests) might be suitable for placement in a less secure environment
that still allows for monitoring of their actions.
All of the alternatives to incarceration discussed above place the offender in the community,
which means there is some level of risk that the offender could commit new offenses, because
even though the offender would be supervised, the level of supervision would most likely provide
a lower level of control over the individual’s actions than would be provided by correctional
officers in a secure environment.

Early Release Measures
One possible way to reduce the growth of the federal prison population would be to expand the
early release measures for federal inmates. There are several options Congress could consider if
policymakers wanted to expand early release options for federal inmates, including (1) reinstating
parole, (2) expanding good time credits, and (3) expanding the conditions under which courts
could reduce sentences pursuant to 18 U.S.C. §3582(c)(1)(A).

120

Funding for this program was discontinued after FY2004.
See http://www.bop.gov/foia/fy11_per_capita_costs.pdf.
122
For more information on how inmates are processed through the federal prison system, see CRS Report R42486,
The Bureau of Prisons (BOP): Operations and Budget, by Nathan James.
121

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

Reinstating Parole
One option Congress might consider is whether to reinstate parole in the federal system. As
discussed above, inmates sentenced for an offense in a federal court committed after November 1,
1987 are not eligible to be released on parole. Parole is one way correctional authorities can
release inmates who are deemed to be at a low-risk for recidivism and place them in community
supervision for the remainder of their sentences.
Should Congress consider reinstating parole for federal inmates, there are several salient issues
that policymakers might think about. First, how would a parole system work within the current
determinate sentencing structure used in federal courts? Traditionally, discretionary parole has
been combined with an indeterminate sentencing structure (i.e., a system whereby the court could
impose a sentence for a crime within a range prescribed in law). Indeterminate sentences allow
the court to tailor sentences to each defendant, but this gave rise to concerns about whether some
sentences were arbitrary and unfair. For example, two defendants who were convicted for similar
crimes might receive different sentences depending on which judge happened to be presiding
over their case. When combined with a parole board’s discretion over when, if ever, someone
would be granted parole, two defendants who were convicted of similar crimes could end up
serving significantly different amounts of time in prison.
Congress sought to limit the discretion of the federal judiciary and the executive branch when it
eliminated parole and replaced indeterminate sentencing with the sentencing guidelines. Parole
might not be irreconcilable with a determinate sentencing structure. Courts could continue to use
sentencing guidelines as a guidepost for determining a defendant’s sentence and each inmate
could then be eligible for parole after serving a certain portion of his or her sentence. However,
should Congress allow federal inmates to be eligible for parole, it would grant the executive
branch, through the U.S. Parole Commission (hereinafter, “commission”), some measure of
control over determining how much time an inmate serves in prison. Congress might choose to
limit some of the commission’s discretion by setting a higher threshold for determining what
portion of an inmate’s sentence must be served before he or she is eligible to be placed on
parole.123
Should Congress choose to reinstate parole for federal inmates, another key question would be
whether eligibility would be made retroactive to inmates who were sentenced for federal crimes
after November 1, 1987. As discussed above, approximately 3% of inmates currently incarcerated
in federal facilities are still eligible for parole, which means that at the end of FY2011 there were
nearly 204,600 inmates in federal prison who were not eligible for parole. Making eligibility for
parole retroactive could potentially reduce the federal prison population in a shorter amount of
time than it would if only newly convicted inmates were eligible for parole consideration. Data
from the BOP indicate that nearly three-quarters of inmates have served at least 25% of their
sentence, meaning that if Congress reinstated the old parole eligibility rules, a majority of federal

123

Federal inmates who are eligible for parole (i.e., inmates sentenced before November 1, 1987) can be released after
serving one-third of their sentences (if sentenced to a term of incarceration greater than one year) or after 10 years if
sentenced to life or a term or incarceration over 30 years. However, the sentencing court could designate a minimum
term of imprisonment the defendant would have to serve before being eligible for parole. The minimum term of
imprisonment designated by the court could be less, but not more, than one-third of the sentence imposed. The
sentencing court could also fix the maximum sentence to be served, at which point the inmate could be released on
parole. 18 U.S.C. §§4205(a) and 4205(b), as it was in effect before being repealed by section 218(a) of P.L. 98-473.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

inmates would be eligible for parole consideration.124 It would appear likely that the commission
and the U.S. Probation and Pretrial Services office would need increased resources in order to
properly manage what would likely be a significant increase in their caseloads.
There might be some concern about whether allowing federal inmates to be released on parole
would pose a threat to public safety. Concerns about recidivism are not unfounded. Research
published by the Bureau of Justice Statistics found that over two-thirds (67.5%) of inmates
released in 1994 were rearrested within three years and nearly half (46.9%) were convicted for a
new crime.125 While the data are dated, this remains one of the most comprehensive studies on
recidivism. Concerns about offenders committing new crimes while on parole have led some
jurisdictions to implement intensive supervision programs where parolees are subject to more
rigorous conditions of release and more frequent contacts with a parole officer. While intensive
supervision programs might in theory reduce the likelihood that parolees commit new offenses
while in the community, the body of research on intensive supervision programs suggests that
these programs do not reduce recidivism.126 Depending on how recidivism is defined, intensive
supervision programs may actually increase “recidivism” because they are more likely to detect
technical violations of the conditions of release.127 This can create a paradox for policymakers:
parole might be considered as a means of reducing the prison population, but it might actually
increase the number of inmates in prisons as more return to prison for violating the conditions of
parole. Should Congress choose to reinstate parole, policymakers might consider evidence-based
measures so that parole helps as many inmates successfully transition back into the community as
possible. The options Congress could consider are similar to those outlined above for successful
probation programs, namely
•

using a validated risk assessment tool to sort parolees into high- and low-risk
groups;

•

ensuring that parolees with a demonstrated need for rehabilitative programming
have access to evidence-based, appropriately delivered programs;

•

requiring parolees who violate their conditions of release to serve intermediate
sanctions rather than returning them to prison; and

•

allowing parolees who strictly adhere to the conditions of their parole to be
released early.

Also, like probationers, data indicate that parolees are at the highest risk for recidivism during
their first year of parole.128 This suggests that in order to decrease the risk of recidivism services
should be more intensive during the parolee’s first year on release. Some research suggests that

124

Data provided by the U.S. Department of Justice, Bureau of Prisons.
Patrick A. Langan and David J. Levin, United States Department of Justice, Bureau of Justice Statistics, Recidivism
of Prisoners Released in 1994, Report NCJ193427, June 2002. For a summary of this study and other studies on
recidivism, see CRS Report RL34287, Offender Reentry: Correctional Statistics, Reintegration into the Community,
and Recidivism, by Nathan James.
126
Doris Layton MacKenzie, What Works in Corrections: Reducing the Criminal Activities of Offenders and
Delinquents (New York: Cambridge University Press, 2006), p. 310.
127
Ibid.
128
Community Corrections: Probation, Parole, and Prisoner Reentry, p. 524.
125

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

intensive supervision programs can reduce recidivism when they are combined with treatment
and rehabilitative programming.129

Expanding Good Time Credits
Another potential policy option Congress could consider as a means to slow the growth of, or
possibly reduce, the federal prison population is to expand the BOP’s authority to grant good time
credit to inmates. As outlined above, Congress abolished parole for federal inmates in the 1980s,
which means that inmates cannot be released before serving their entire sentence, minus any good
time credit, even if the inmate’s risk of recidivism is low. Under current law, the BOP can grant
up to 54 days of good time credit per year to inmates serving a sentence of more than one year,
assuming the inmate has demonstrated “exemplary compliance with institutional disciplinary
regulations” and is making satisfactory progress on completing a GED (assuming the inmate does
not have a GED or a high school diploma).130
In addition to the amount of good time credit an inmate can earn, the BOP is allowed to reduce a
non-violent inmate’s sentence by up to one year if the inmate participates in residential substance
abuse treatment.131 It has been argued that teaming good time credit with a program that places
inmates with objectively assessed needs and risks in evidence-based programs to address those
needs and risks can reduce recidivism and cut prison costs.132 Congress could consider allowing
the BOP to award good time credit for inmates who have a need for and successfully complete
rehabilitative programs other than residential drug abuse treatment. However, expanding good
time credit for participation in rehabilitative programming would likely require Congress, at least
in the short term, to expand funding for rehabilitative programs and inmate skills and needs
assessments. While expanding current good time credit policies might help reduce prison
overcrowding, there might be some concern that the BOP would effectively be reducing inmates’
sentences without the sentencing court’s approval. Additional good time credit would also allow
inmates to be released before serving a significant (85%) portion of their sentence, a key rationale
for why parole was eliminated in the first place. In addition, some may feel that regardless of an
inmate’s efforts to rehabilitate himself or herself or the risk he or she would pose to society when
released, the inmate was sent to prison as a punishment for a crime, hence the inmate should
serve his or her full sentence.

Sentence Reduction
In addition to allowing the BOP to grant more good time credit to inmates, Congress could also
consider whether to amend the conditions under which courts can reduce an inmate’s sentence.
Under current law (18 U.S.C. §3582(c)(1)(A)), the BOP can petition the court to reduce an
inmate’s sentence if the court finds that “extraordinary and compelling reasons warrant such a
reduction”; or the inmate is at least 70 years of age, has served at least 30 years of his or her
sentence, and a determination has been made by the Director of the BOP that the inmate is not a
129
Doris Layton MacKenzie, What Works in Corrections: Reducing the Criminal Activities of Offenders and
Delinquents (New York: Cambridge University Press, 2006), p. 318.
130
18 U.S.C. §3624(b)(1).
131
18 U.S.C. §3621(e)(2)(B).
132
Dora Schriro, “Is Good Time a Good Idea? A Practitioner’s Perspective,” The Federal Sentencing Reporter, vol. 21,
no. 3 (February 2009), p. 181.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

danger to the safety of any other person or the community. Congress required the USSC, when
issuing a policy statement regarding sentence modification under 18 U.S.C. §3582(c)(1)(A), to
“describe what should be considered extraordinary and compelling reasons for sentence
reduction, including the criteria to be applied and a list of specific examples.”133 Under §1B1.13
of the U.S. Sentencing Guidelines, the USSC deemed the following circumstances to be
“extraordinary and compelling reasons” for a sentence reduction:
•

The inmate is suffering from a terminal illness.

•

The inmate is suffering from a permanent physical or medical condition, or is
experiencing deteriorating physical or mental health because of the aging
process, that substantially diminishes the ability of the defendant to provide selfcare within the environment of a correctional facility and for which conventional
treatment promises no substantial improvement.

•

The death or incapacitation of the inmate’s only family member capable of caring
for the inmate’s minor child or minor children.

•

As determined by the Director of the BOP, there exists in the inmate’s case an
extraordinary and compelling reason other than, or in combination with, the
reasons described above.

Pursuant to 28 U.S.C. §944(t), rehabilitation of an inmate is not, by itself, an extraordinary and
compelling reason for granting a sentence reduction. If the court grants a sentence reduction
under 18 U.S.C. §3582(c)(1)(A), the court is also allowed to impose a term of probation or
supervised release, with or without conditions, for a period up to the amount of time that was
remaining on the inmate’s sentence.
One of the critiques of this program is that it relies on the BOP to petition the court for a review
of an inmate’s sentence. One commentator argues that the BOP narrowly interprets when inmates
should be allowed to apply for a sentence reduction, effectively limiting applications to cases
where the inmate is terminally ill and near death.134 The regulations governing the program do not
state that consideration for a sentence reduction under this program will be limited to cases where
the inmate is terminally ill, but it does state that consideration for a sentence reduction under 18
U.S.C. §3582(c)(1)(A) is limited to “particularly extraordinary or compelling circumstances
which could not reasonably have been foreseen by the court at the time of sentencing.”135
Congress could consider modifications to the requirements for sentence reduction under 18
U.S.C. §3582(c)(1)(A) to allow more inmates to have their sentences reduced. For example,
Congress could consider allowing courts to consider rehabilitation—either as an extraordinary
and compelling reason on its own, or in consort with other reasons—when making determinations
about sentence reductions. Expanding the authority of courts to grant a sentence reduction could
allow inmates deemed to be a low threat to public safety to be placed in the community earlier,
thereby freeing up bedspace in federal prisons.

133

28 U.S.C. §994(t).
Stephen R. Sady, “Second Look Resentencing under 18 U.S.C. §3582(c) as an Example of Bureau of Prisons
Policies That Result in Overincarceration,” Federal Sentencing Reporter, vol. 21, no. 3 (February 2009), p. 167.
135
28 C.F.R. §571.60.
134

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

An inmate granted a sentence reduction could still be required to serve a term of supervised
release, which would allow federal probation officers to monitor the inmate after he or she is
released, a possible advantage over allowing inmates to be released early by increasing good
conduct time. However, it is likely that the judicial branch would require additional resources in
order to process more applications for sentence reductions under the program and properly
monitor inmates whose sentences were reduced but who were placed on supervised release. Also,
there might be a question as to whether this would turn the courts into de facto parole boards.
Congress eliminated parole in the federal system, in part, over concerns that inmates were
incarcerated for less than an appropriate amount of time and disparities in decisions over who
received parole. Under this possible system, inmates could be released before serving a majority
of their sentences, but Congress could address this concern by not allowing inmates to be eligible
for a sentence reduction until they have served a certain portion of their entire sentence.
A potentially more difficult issue for Congress to address is how judges would make decisions if
granted broader authority to reduce sentences under the program. It is possible that an inmate’s
opportunity to receive a sentence reduction would depend on which judge ruled on the inmate’s
petition. This concern mirrors some of the concerns that existed about how much sway parole
boards held over who was granted parole.
Congress could also consider amending the requirements under 18 U.S.C. §3582(c)(1)(A) so that
inmates could be released before turning 70. Research indicates that most offenders “age-out” of
crime; that is, the older offenders get, the less likely they are to commit new crimes.136 It appears
likely that more elderly inmates could safely be released from confinement and placed on home
confinement for the remainder of their sentences.137 However, the data show that granting early
release to elderly offenders would only have a minimal effect on prison overcrowding. In
FY2010, approximately 14% were over the age of 50 and about 4% of inmates were over the age
of 60.138 In addition, while elderly inmates might pose a reduced threat to public safety, there is
136

Lindsey Devers, Desistance and Developmental Life Course Theories: Research Summary, U.S. Department of
Justice, Office of Justice Programs, Bureau of Justice Assistance, Washington, DC, November 9, 2011, p. 7,
http://www.ojp.usdoj.gov/BJA/pdf/DesistanceResearchSummary.pdf.
137
Under section 231(g) of the Second Chance Act of 2007 (P.L. 110-199), the BOP was directed to conduct a pilot
program in FY2009 and FY2010 whereby eligible inmates would be placed on home confinement for the remainder of
their sentences. Inmates eligible to participate in the pilot program were 65 or older; non-violent or non-sex offenders;
not serving a life sentence; severed the greater of 10 years or 75% of their sentences; did not have a history of escape or
escape attempts; and were determined to not be at risk for recidivism. The Government Accountability Office (GAO)
reported that of the 855 inmates who applied for the pilot program, 71 (8.3%) were determined by the BOP to have met
the criteria for the program and were eventually placed on home confinement. The GAO noted that as of June 2012,
none of the inmates placed on home confinement had recidivated or violated the terms of release. However, the BOP
reported that it did not save any money by placing elderly inmates on home confinement; in fact, the BOP reported that
it cost approximately $540,000 more to place the inmates on home confinement. The GAO contends that the BOP’s
conclusions might not be a reliable indicator of the potential cost of the program should it be continued or expanded.
First, while the BOP knows what it paid RRCs to monitor inmates placed on home confinement, the BOP does not
know the exact cost of home confinement. The BOP negotiates with RRCs to provide supervision of inmates placed on
home confinement. RRCs are paid a per diem rate to house an inmate and they are paid 50% of the per diem rate to
supervise an inmate placed on home confinement. However, the BOP does not require RRC contractors to separate the
cost of home confinement services and RRCs bedspace, so the BOP does not actually know the cost of home
confinement. Second, some of the costs of the pilot program would have been incurred regardless because the BOP is
currently authorized to place all of the inmates in the program on home confinement for up to six months. Government
Accountability Office, Federal Bureau of Prisons: Methods for Estimating Incarceration and Community Corrections
Costs and Results of the Elderly Offender Pilot, GAO-12-807R, Washington, DC, July 27, 2012, http://www.gao.gov/
assets/600/593089.pdf.
138
Data from the U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics, Federal Criminal
(continued...)

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

likely to be some sentiment that any offender, regardless of age and safety risk, should serve his
or her entire sentence.

Modifying the “Safety Valve” Provision
There are other potential amendments to the criminal code Congress could consider if
policymakers wanted to potentially reduce the size of the federal prison population. For example,
Congress could consider expanding the “safety valve” provision under 18 U.S.C. §3553(f). The
safety valve provision allows judges to impose a sentence without regard to the mandatory
minimum sentences for certain drug offenses139 if the following conditions are met:
•

The defendant does not have more than one criminal history point, as determined
under the sentencing guidelines.

•

The defendant did not use violence or credible threats of violence or possess a
firearm or other dangerous weapon (or induce another participant to do so) in
connection with the offense.

•

The offense did not result in death or serious bodily injury to any person.

•

The defendant was not an organizer, leader, manager, or supervisor of others in
the offense, as determined by the sentencing guidelines, and was not engaged in a
continuing criminal enterprise.

•

No later than the time of the sentencing hearing, the defendant has truthfully
provided to the government all information and evidence the defendant has
concerning the offense or offenses that were part of the same course of conduct
or of a common scheme or plan, but the fact that the defendant has no relevant or
useful or other information to provide or that the government is already aware of
the information shall not preclude a determination by the court that the defendant
has not complied with the requirements of the provision.

Currently, the safety valve provision cannot be applied to defendants facing a mandatory
minimum sentence for an offense that is not drug-related. The safety valve provision was enacted
after Congress became concerned that the mandatory minimum sentencing provisions could result
in equally severe penalties for both the more and the less culpable offenders.140 Congress could
consider expanding the provision so that it would apply to defendants facing mandatory minimum
sentences for offenses other than drug crimes. This option would allow Congress to retain
mandatory minimum penalties that can still be applied to more serious offenders while allowing
judges to sentence less serious offenders to shorter periods of incarceration.
(...continued)
Case Processing Statistics, http://www.bjs.gov/fjsrc/.
139
These offenses are trafficking in various controlled substances (21 U.S.C. §841), possession of certain controlled
substances (21 U.S.C. §844), attempt or conspiracy to violate controlled substance provisions carrying mandatory
minimum sentences (21 U.S.C. §846), smuggling controlled substances in violation of 21 U.S.C. §§952, 953, 955, 957,
or 959 (21 U.S.C. §960), and attempt or conspiracy to violate the controlled substance import/export provisions (21
U.S.C. §963).
140
CRS Report R41326, Federal Mandatory Minimum Sentences: The Safety Valve and Substantial Assistance
Exceptions, by Charles Doyle. See also U.S. Congress, House Committee on the Judiciary, Mandatory Minimum
Sentencing Reform Act of 1994, to accompany H.R. 3979, 103rd Cong., 2nd sess., March 24, 1994, H.Rept. 103-460
(Washington: GPO, 1994), p. 2.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

One idea put forth is to amend current law so that judges could apply the safety valve in instances
where the recommended sentencing guideline range is below the mandatory minimum penalty
and where the defendant’s offense did not result in death or serious bodily injury to anyone and
the defendant has provided the government with all information and evidence available to the
defendant.141 Under the proposal, the defendant could not be sentenced to less than the minimum
of the sentencing range calculated under the sentencing guidelines. Many of the conditions placed
on the current safety valve provision would remain (e.g., not using violence or possessing a
weapon and not being an organizer or leader in the offense), but rather than being a bar from
being eligible for the safety valve, they would be factors for the court to consider when deciding
whether to sentence a defendant below the mandatory minimum penalty. Judges would be
required to state for the record why they chose to impose a sentence below the mandatory
minimum penalty, and those decisions would be subject to appellate review. However, as noted
above, the USSC has incorporated many mandatory minimum sentences into the sentencing
guidelines. Therefore, in some instances the guideline sentence might be equal to or exceed the
mandatory minimum penalty, which would render the proposed safety valve provision moot. One
possible solution to this conundrum would be to allow the USSC to give due consideration to
mandatory minimum penalties when formulating the sentencing guidelines, but not requiring the
USSC to make the guidelines consistent with mandatory minimum penalties.142

Repealing Federal Criminal Statutes for Some Offenses
One of the highlighted reasons for the growth in the federal prison population was the
“federalization” of offenses that were traditionally under the sole jurisdiction of state authorities.
Policymakers could consider revising the U.S. Code so that federal law enforcement focuses on
crimes where states do not have jurisdiction over the offenses or where the federal government is
best suited to investigate and prosecute the offenders (e.g., the offense involves multiple
individuals acting together to commit crimes across several states). Some crimes will always be
federal offenses. For example, the federal government will always be responsible for prosecuting
individuals who commit immigration-related offenses because immigration laws are solely the
jurisdiction of the federal government. However, over the years the federal government has
become more involved in investigating, prosecuting, and incarcerating people who commit drug
offenses and offenses where a convicted felon is found to be in possession of a firearm. In many
instances, states have criminal penalties for individuals who commit these types of crimes. For
example, in his testimony before the House Subcommittee on Crime, Terrorism, and Homeland
Security at a hearing on the unintended consequences of mandatory minimum penalties, Eric
Sterling, the President of the Criminal Justice Policy Foundation argued
[f]ederal drug cases should focus exclusively on the international organizations that use their
profits from the manufacture and distribution of cocaine, opium and heroin,
methamphetamine, and cannabis to finance assassinations, terrorism, wholesale corruption
and bribery, organized crime generally, and the destabilization of our allies…Every state in
the U.S. has a great capacity to investigate, prosecute and punish the high-level local drug
traffickers that operate within their jurisdiction. State and local police and prosecutors
outnumber federal agents and prosecutors. State prisons far exceed the capacity of federal
prisons…Almost none of the crack [cocaine] dealers that proliferate in countless U.S.
141

Paul G. Cassell and Erik Luna, “Sense and Sensibility in Mandatory Minimum Sentencing,” Federal Sentencing
Reporter, vol. 23, no. 3 (February 2011), p. 222.
142
Ibid., p. 225.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

neighborhoods warrant federal prosecution. There are neighborhood criminals and their
crimes are state crimes. If a state’s law does not adequately punish a crack [cocaine] dealer,
that is the state’s problem. Inadequate state laws do not warrant wasting very scarce,
powerful federal resources even on serious neighborhood criminals [emphasis original].143

Scaling back the scope of the federal criminal code could help reduce the size of the federal
prison population in the future by reducing the number of people prosecuted and sentenced to
incarceration in federal courts. However, this policy option could increase the burden on state
criminal justice systems since they would be responsible for prosecuting and incarcerating
offenders who are no longer tried in federal courts. By year-end 2011, according to the BJS, 24
state correctional systems were at or above their highest capacity, and another 10 state
correctional systems were between 90% and 99% of their highest capacity.144 Since nearly threequarters of states have prison systems that are operating at 90% of capacity or higher, it would
appear that if the federal government chooses not to prosecute some offenses, thereby leaving
states with the responsibility to do so, it would require states to either expand their prison
capacities or possibly decline to prosecute some offenses. Also, it is possible that an expansion of
state correctional systems could have a significant effect on state finances. The Vera Institute of
Justice reported that state correctional spending has nearly quadrupled over the past two decades,
which makes it the fasting growing budget item after Medicaid.145 The National Association of
State Budget Officers reported that in FY2011, correctional expenditures accounted for 7.4% of
all state general fund expenditures, with a range of 24.3% (Michigan) to 2.9% (Minnesota).146
Since states typically cannot run a budget deficit, any expansion in correctional expenditures
would have to be paid for with cuts to other state services, increased taxes, or issuing bonds.

Conclusion
There are almost 195,000 more inmates incarcerated in federal prisons currently than there were
in FY1980, a nearly 790% increase in the federal prison population. The growth in the federal
prison population is the result of several changes to the federal criminal justice system, including
expanding the use of mandatory minimum penalties; the federal government taking jurisdiction in
more criminal cases; and eliminating parole for federal inmates.
The analyses presented in this report show that the BOP faces several challenges resulting from
the increasing number of inmates in the federal prison system. First, it is increasingly expensive
to operate the federal prison system. Second, the federal prison system is becoming more
overcrowded, especially in high- and medium-security male prisons. Third, the cost of caring for
more inmates has left the BOP in a position where it might not be able to hire an adequate number
143
Statement of Eric E. Sterling, President of the Criminal Justice Policy Foundation, U.S. Congress, House Committee
on the Judiciary, Subcommittee on Crime, Terrorism, and Homeland Security, Mandatory Minimums and Unintended
Consequences, Hearing on H.R. 2934, H.R. 834 and H.R. 1466, 111th Cong., 1st sess., July 14, 2009, H.Hrg. 111-48
(Washington: GPO, 2010), pp. 114-115.
144
Prisoners in 2011, p. 31.
145
Christian Henrichson and Ruth Delaney, The Price of Prisons: What Incarceration Costs Taxpayers, Vera Institute
of Justice, New York, NT, January 2012, p. 2, http://www.vera.org/download?file=3542/
Price%2520of%2520Prisons_updated%2520version_072512.pdf.
146
National Association of State Budget Officers, 2010 State Expenditure Report: Examining Fiscal 2009-2011 State
Spending, Washington, DC, 2011, p. 56, http://www.nasbo.org/sites/default/files/
2010%20State%20Expenditure%20Report.pdf.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

of staff. Finally, the steady influx of inmates combined with dwindling levels of funding for
prison construction and maintenance has placed a strain on the federal prison system’s
infrastructure.
If Congress moves forward with a debate about prison and sentencing reform, policymakers
might also consider whether incarcerating more offenders would continue to generate a
significant deterrent effect. Research suggests that while incarceration did contribute to lower
violent crime rates in the 1990s, there are declining marginal returns associated with everincreasing levels of incarceration.147 The diminishing level of return resulting from higher levels
of incarceration might be explained by the fact that higher levels of incarceration are likely to
include more offenders who are either at the end of their criminal careers or who were at a low
risk of committing crimes at a high rate (so-called “career criminals”).148 Another possible reason
for diminishing marginal returns might be that more of the individuals incarcerated over the past
three decades have been incarcerated for crimes where there is a high level of replacement.149 For
example, if a serial rapist is incarcerated, not only is there the potential to prevent future sexual
assaults that would have been committed by the offender, but it is also probable that no one else
will take that offender’s place. However, if a drug dealer is incarcerated, it is possible that
someone will step in to take that person’s place; therefore, no further crimes may be averted by
incarcerating the individual.
The unprecedented increase in the federal prison population over the past three decades was not
the result of a singular policy change; likewise, it would appear that, should Congress choose to
address prison population growth, it would require a multi-faceted approach to reduce the number
of federal inmates in any substantive manner. Congress might consider options—such as
expanding the capacity of the federal prison system, continued investment in rehabilitative
programs, and placing inmates in private prisons—that would either continue or expand current
correctional policies. However, Congress might also consider changing or reversing some of the
policies that have been put into place over the years which contributed to the increasing number
of federal prison inmates. Some of these options include placing some inmates in alternatives to
incarceration, such as probation, or expanding early release options by allowing inmates to earn
more good time credit or allowing inmates to be placed on parole once again. Congress could
consider reducing the amount of time inmates are incarcerated in federal prisons by limiting the
number of crimes subject to mandatory minimum penalties or reducing the length of the
mandatory minimum sentence. Finally, policymakers could consider allowing states to investigate
and prosecute offenses that have become subject to federal jurisdiction over the past three
decades.

147
Anne Morrison Piehl and Bert Useem, “Prisons,” in Crime and Public Policy, ed. Joan Petersilia and James Q.
Wilson, 2nd ed. (New York: Oxford University Press, 2011), p. 542.
148
Doris Layton MacKenzie, “Reducing the Criminal Activities of Known Offenders and Delinquents: Crime
Prevention in the Courts and Corrections,” in Evidence-based Crime Prevention, ed. Lawrence W. Sherman, David P.
Farrington, Brandon C. Welsh and Doris Layton MacKenzie (New York: Routledge, 2002), p. 337.
149
Bert Useem and Anne Morrison Piehl, Prison State: The Challenge of Mass Incarceration (New York: Cambridge
University Press, 2008), p. 74.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

Appendix. Select BOP Data
Table A-1. Appropriations for the BOP by Account; Number of Inmates Under the
BOP’s Jurisdiction; and the Number and Capacity of and Overcrowding in
BOP Facilities
Appropriations amounts are in thousands of dollars
Prison Population

Appropriations
Fiscal
Year
1980

Salaries
and
Expenses

Buildings
and
Facilities

$323,884

$5,960

Total
$329,844

Institution

Contract

24,268

372

Prison Facilities

Total
24,640

Capacity

Overcrowding

—

—
11%a

Number

1981

351,759

10,020

361,779

26,195

118

26,313

23,648a

1982

378,016

56,481

434,497

28,133

2,398

30,531

24,072a

17%a

1983

412,133

66,667

478,800

30,214

3,002

33,216

23,936a

26%a

1984

464,850

47,711

512,561

32,317

3,478

35,795

24,874a

30%a

1985

536,932

86,043

622,975

36,001

4,329

40,330

25,532a

41%a

1986

561,480

44,082

605,562

41,506

4,549

46,055

27,785a

49%a

1987

656,941

219,249

876,190

44,194

5,184

49,378

27,854a

59%a

1988

772,013

297,076

1,069,089

44,142

6,371

50,513

28,143a

57%a

1989

962,016

612,914

1,574,930

51,153

6,609

57,762

31,727a

61%a

1990

1,138,778

1,511,953

2,650,731

58,021

6,915

64,936

34,239a

69%a

1991

1,363,645

374,358

1,738,003

64,131

7,377

71,508

42,531

51%

1992

1,649,121

462,090

2,111,211

70,670

9,008

79,678

48,527

46%

1993

1,793,470

339,225

2,132,695

79,799

8,766

88,565

57,610

39%

1994

1,962,605

269,543

2,232,148

85,850

9,312

95,162

64,751

33%

1995

2,319,722

276,301

2,596,023

90,159

10,799

100,958

72,039

25%

1996

2,546,893b

334,728

2,881,621

94,695

10,748

105,443

76,442

24%

1997

2,748,427c

435,200

3,183,627

101,091

11,198

112,289

83,022

22%

1998

2,847,777d

255,133

3,102,910

108,207

14,109

122,316

86,051

26%

1999

2,888,853e

410,997

3,299,850

117,295

16,394

133,689

89,581

31%

2000

3,111,073f

556,780

3,667,853

125,560

19,565

145,125

94,927

32%

2001

3,469,739

833,822

4,303,561

130,327

26,245

156,572

100

98,425

32%

2002

3,805,118

807,808

4,612,926

137,527

25,909

163,436

102

103,262

33%

2003

4,044,788

396,632

4,441,420

146,212

26,287

172,499

103

105,193

39%

2004

4,414,313

393,515

4,807,828

152,518

27,377

179,895

109

108,537

41%

2005

4,571,385

205,076

4,776,461

159,501

27,893

187,394

116

118,652

34%

2006

4,830,160

99,961

4,930,121

162,514

30,070

192,584

114g

119,510

36%

2007

5,012,433

432,425

5,444,858

167,323

32,697

200,020

114

122,189

37%

2008

5,346,740

372,720

5,719,460

165,964

35,704

201,668

114

122,366

36%

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

Prison Population

Appropriations
Salaries
and
Expenses

Buildings
and
Facilities

2009

5,600,792

575,807

2010

6,106,231

2011
2012

Fiscal
Year

Total

Prison Facilities

Total

Number

Overcrowding

Institution

Contract

Capacity

6,176,599

172,423

36,336

208,759

115

125,778

37%

99,155

6,205,386

173,289

36,938

210,227

116

126,713

37%

6,282,410

98,957

6,381,367

177,934

39,834

217,768

117

127,795

39%

6,551,281

90,000

6,641,281

177,556

41,131

218,687

118

128,359

38%

Source: U.S. Department of Justice, Bureau of Prisons.
Notes: The BOP did not provide capacity and overcrowding data for FY1980. Appropriation amounts in Table
A-1 include all supplemental and reprogrammed appropriations and any rescissions of enacted budget authority,
but they do not include rescissions of unobligated balances. From FY1980 to FY1995, funding for the National
Institute of Corrections (NIC) was included in a separate account. Since FY1996, funding for the NIC has been
included in the S&E account. Funding for the NIC for FY1980-FY1995 was added to the S&E account to make
funding for the S&E account comparable across fiscal years.
a.

Capacity and overcrowding were calculated based on single cell occupancy

Includes $13.5 million appropriated from the Violent Crime Reduction Trust Fund.

Includes $25.2 million appropriated from the Violent Crime Reduction Trust Fund.

Includes $26.1 million appropriated from the Violent Crime Reduction Trust Fund.

Includes $26.5 million appropriated from the Violent Crime Reduction Trust Fund.

Includes $22.5 million appropriated from the Violent Crime Reduction Trust Fund.

In FY2006, the BOP closed four stand-alone prison camps and activated two new prisons.

Congressional Research Service

The Federal Prison Population Buildup: Overview, Policy Changes, Issues, and Options

Table A-2. Appropriations for the BOP, by Decision Unit, FY1999-FY2012
Appropriations in thousands of dollars
Buildings and Facilities

Salaries and Expenses
Inmate Care
and
Programs

Institution
Security and
Administration

Contract
Confinement

Management
and
Administration

1999

$1,090,148

$1,401,349

$255,062

$142,249

$322,963

$88,034

2000

1,123,856

1,494,809

344,773

147,635

441,003

115,777

2001

1,208,480

1,601,518

511,579

148,162

710,816

123,006

2002

1,311,335

1,722,567

620,145

151,071

675,040

132,768

2003

1,454,481

1,868,538

567,365

154,404

262,956

133,676

2004

1,624,308

2,050,417

574,473

165,115

177,620

164,000

2005

1,682,656

2,094,917

627,135

166,677

25,372

179,704

2006

1,740,011

2,227,223

683,031

179,895

48,115

51,846

2007

1,774,048

2,297,055

757,851

183,479

368,875

63,550

2008

1,942,077

2,409,112

814,529

181,022

302,720

70,000

2009

2,070,002

2,495,196

840,933

194,661

465,180

110,627

2010

2,215,992

2,708,651

981,112

200,476

25,386

73,769

2011

2,294,174

2,783,664

996,772

207,800

25,335

73,622

2012

2,421,272

2,880,290

1,040,213

209,506

23,035

66,695

Fiscal
Year

New
Construction

Modernization
and Repair

Source: U.S. Department of Justice, Bureau of Prisons.
Notes: Amounts in Table A-2 include all supplemental and reprogrammed appropriations and any rescissions
of enacted budget authority, but they do not include rescissions of unobligated balances.

Author Contact Information
Nathan James
Analyst in Crime Policy
njames@crs.loc.gov, 7-0264

Congressional Research Service