The Growth & Increasing Cost of the Federal Prison System:
Drivers and Potential Solutions
Federal prison growth has multiple impacts.
Nancy La Vigne
Julie Samuels

The federal prison population has been growing dramatically; its current
population exceeds 218,000, 1 with projections of continued growth for the
foreseeable future. A wide array of actors—Members of Congress,
administration officials, a bipartisan cast of policy advocates, and
researchers—has concluded that this growth and its associated costs are
unsustainable. The basis for this conclusion varies:
Fiscal impact. Resources spent on the Bureau of Prisons (BOP) eclipse
other budget priorities.
Overcrowding risks. Overcrowded facilities can jeopardize the safety
of inmates and staff and limit opportunities for effective programming
that can reduce recidivism.

Fairness/equity concerns. High levels of incarceration may have
disproportionate impacts on certain subpopulations and communities.

This research was supported by
the Public Welfare Foundation.
The opinions, findings, and
conclusions or recommendations
expressed in this publication are
those of the authors and do not
necessarily reflect the views of
the Public Welfare Foundation,
the Urban Institute, its trustees,
or its funders. Permission is
granted for reproduction of this
document, with attribution to the
Urban Institute.

DECEMBER 2012

Inefficient resource allocation. Current research and recent
evidence-based policy changes implemented in states raise questions
about the cost-effectiveness of existing federal sentencing and
corrections policies.

The focus on this burgeoning population provides an opportunity to
explore the drivers of population growth and costs and to develop options
for stemming future growth that are consistent with public safety goals.

BOP projects continued growth.

BOP has experienced an almost tenfold increase in its population since
1980. In FY 2011, the BOP population increased by 7,541 inmates, and will
increase by an estimated 11,500 by the end of FY 2013. 2
Overall, BOP is operating at 39 percent above its rated capacity, with 55
percent crowding at high-security facilities and 51 percent at mediumsecurity facilities. 3 Since FY 2000, the inmate-to-staff ratio has increased
from about 4:1 to a projected 5:1 in FY 2013. 4 This degree of crowding
threatens the safety of both inmates and correctional officers, and it
undermines the ability to provide effective programming.

1

Prison is expensive.
Annual costs per inmate are $21,006 for minimum security, $25,378 for low security,
$26,247 for medium security, and $33,930 for high security. Average annual costs per
inmate housed in community corrections (residential reentry centers and home
confinement) for BOP are $25,838. 5 By contrast, the annual cost of supervision by probation
officers in the community is about $3,433 per offender. 6

Currently, more than half (56 percent) of the current federal inmate population is housed in
minimum- or low-security facilities. Almost 30 percent are housed in medium-security
facilities, and about 11 percent are housed in high-security facilities. 7

BOP growth creates opportunity costs.

The President’s FY 2013 budget request for BOP totals $6.9 billion, reflecting an increase of
$278 million (4.2 percent) from the FY 2012 enacted budget. These additional funds will
backfill currently open positions, enabling recently completed prisons to operate and, to a
limited degree, expand inmate programming. 8 However, these changes will not have any
substantial or sustainable impact on the overcrowding or inmate-to-staff ratio trends.

The BOP budget for FY 2013 accounts for over 25 percent of the DOJ budget. 9 As indicated in
figure 1, if present trends continue, the share of the DOJ budget consumed by BOP will grow
even further, approaching 30 percent in 2020. In these fiscally lean times, funding the
expanding BOP population crowds out other priorities, including federal investigators and
prosecutors and support for state and local governments. This situation is projected to
continue into the future. 10
Figure 1. BOP Budget as a Portion of Total DOJ Budget FY 2000-2012
(Projected through 2020)
$35.00
$30.00
Total DOJ
Discretionary
Budget (nonBOP)
BOP Enacted
Budget, Actual
Dollars

(Billions)

$25.00
$20.00
$15.00
$10.00
$5.00

29% of DOJ budget

$-

Source: Department of Justice Summary of Budget Authority by Appropriation FY 2000-2013. 11
2

The main drivers are front-end decisions about who goes to prison and
for how long.
About 90 percent of BOP inmates are sentenced offenders, mostly for federal crimes. 12 The
number and composition of offenders committed to federal prison result from the types of
cases investigated and charged in the federal system, the dispositions of those cases, and the
proportion of convicted offenders that receive a term of imprisonment. It is the combination
of volume of admissions and length of time served that drives the inmate population. The
length of stay is largely determined by the sentence imposed (informed by the relevant
statutory penalties and federal sentencing guidelines), 13 and most federal offenders
sentenced to prison serve at least 87.5 percent of their term of imprisonment, 14 generally
followed by a separate term of supervised release. Unlike parole, supervised release does not
replace a portion of the sentence of imprisonment but is in addition to the time spent in
prison.
Overview of sentenced offenders. As depicted in figure 2, from 2000 to 2010 the total
number offenders sentenced under the Sentencing Reform Act (felonies and class A
misdemeanors) increased from 59,846 to 83,946, or about 40 percent. 15
Figure 2. Offenders Sentenced for Felony and Class A Misdemeanors

90,000
80,000
70,000
60,000

Violent
Property offenses

50,000

Public-order

40,000

Weapon

30,000

Immigration
Drug

20,000
10,000
0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

Source: BJS Federal Justice Statistics Program (FJSP), http://bjs.ojp.usdoj.gov/fjsrc/, U.S.
Sentencing Commission data, as standardized by FJSP.

3

•

•
•

In FY 2010, about 90 percent of these sentenced offenders received a sentence of
imprisonment, with about 10 percent receiving probation. 16

The average sentence for all offenders with a term of imprisonment in FY 2010 was
54 months. Sentence lengths vary significantly by the type of offense, from an
average of 91 months for weapons offenders, 36 months for fraud, and 20 months for
immigration offenders. 17

Half the drug trafficking offenders sentenced in FY 2010 were in the lowest criminal
history category (Criminal History Category 1). 18

Drug trafficking offenders had an average sentence of 78 months. Figure 3 below
displays the average sentence by drug type.
Figure 3. Length of Imprisonment in Each Drug Type, Fiscal Year 2010
120

111
96.5

100
Average Prison Sentence (Months)

•

84.9
80

73.1

60

54.8

36.8

40

20

0
Powder
Cocaine
(n=5444)

Crack
Cocaine
(n=4742)

Heroin Marijuana
Meth
(n=1551) (n=5889) (n=4205)

Other
(n=1229)

Source: US Sentencing Commission, 2010 Sourcebook of Federal Sentencing Statistics
http://www.ussc.gov/Data_and_Statistics/Annual_Reports_and_Sourcebooks/2010/FigureJ.pdf

4

Drug offenders make up half of the BOP population.
For those imprisoned, the distribution of offenses varies across the admissions, release, and
stock (end-of-year) populations. As shown in figure 4, the mix of offenses for admissions and
releases are fairly similar to one another, with drugs and immigration each accounting for
about one-third of the cohort. However, drug offenders make up about half of the end-ofyear population. The length of sentences – particularly for drug offenders – is an important
determinant of the stock population and driver of population growth.
Figure 4. BOP 2010 Offense Distribution

100%

80%

Violent

60%

Property
Public-order
Weapon

40%

Immigration
Drug
20%

0%
Admissions

Population
(9/30/2010)

Releases

Source: BJS Federal Justice Statistics Program (FJSP), http://bjs.ojp.usdoj.gov/fjsrc/, BOP data,
as standardized by FJSP.

Our recent study of the growth in the BOP population from 1998 to 2010 confirmed that
time served in prison, particularly for drug offenses, was the largest determinant of the
growth in the population. 19 Changes in sentencing practices, prison release policies, or both
could directly address the time served, and thereby moderate prison population growth.

Supervision violations make up at least 15 percent of annual admissions.

Supervision violators include those on probation, supervised release, and parole.
Approximately one in seven BOP admissions was for a supervision violation in FY 2010; the
types of violations are not evident from the BOP data. 20 According to information from the
Administrative Office of the U.S. Courts, about 30 percent of post-conviction supervision
cases (15,561) closed with revocation. Technical violations accounted for 57 percent of the
revocations, minor violations for 6 percent, and major violations for 36 percent. 21
5

Front-end changes can most directly contain future growth.
Reducing sentence length, particularly for drug offenders, would be the most direct way to
slow the projected growth of the BOP population. Decreasing the number of offenders
committed to prison – both sentenced offenders and supervision violators – would also
reduce the long-term projections and cost for the system. BOP does not control either of
these drivers.

Back-end changes can help alleviate the pressure.

Although the main drivers of the BOP population are the number of offenders and sentence
length on the front end, sentence reductions on the back end can also ease crowding and
slow the population growth trend. While BOP plays a lead implementation role in most backend sentence reductions, current authorities are limited by statute and, in some cases,
budgetary constraints.

The federal system can learn from the states.

As with the federal system, states across the country have also experienced burgeoning
criminal justice populations and costs. Many have implemented policies to control the
growth and increase the effectiveness of spending to enhance public safety goals. These
policies include both legislative and administrative measures that change diversion
practices; revise sentencing laws; adjust good time and earned time provisions for
incarcerated offenders; improve community corrections to reduce the likelihood of
recidivism and the return of offenders to prison for technical violations; employ risk and
needs assessment tools more consistently across the criminal justice system; and improve
correctional and supervision practices to be more consistent with evidence-based practices.

While some aspects of the federal system differ from the states, many lessons can be learned
from the state experience. Chief among them is the need for the federal government to
enhance its community corrections capabilities and resources as it develops strategies to
contain its institutional population and accompanying costs.

Moving Forward

In developing strategies to address their prison populations, states typically analyzed
criminal justice trends to identify the factors driving the growth in the population and
convened stakeholders across the criminal justice system to discuss policy changes that
address those drivers. Similarly, for the federal system to address its prison population, an
important next step will be to develop policy options to inhibit the drivers of growth,
informed by a more detailed data analysis. Ultimately, controlling the growth of the BOP
population will require the cooperation and support of numerous players across all
branches of the federal system; as confirmed by a recent GAO report, 22 the Bureau of Prisons
cannot do this on its own. Congressional action will require both appropriators and
authorizers, with the House and Senate Judiciary Committees being central to implementing
solutions.

6

Notes
1

From Quick Facts about the Bureau of Prisons as of 10/27/12, http://www.bop.gov/news/quick.jsp
(hereafter “Quick Facts about BOP”)
2

U.S. Department of Justice Federal Prison System FY 2013 Performance Budget: Congressional
Submission (Washington, DC: 2012), http://www.justice.gov/jmd/2013justification/pdf/fy13-bop-sejustification.pdf (hereafter “BOP FY 2013 Congressional Budget Submission”).
3 BOP FY 2013 Congressional Budget Submission. Prison overcrowding is calculated using BOP’s baseline
of its rated capacity.
4 BOP FY 2013 Congressional Budget Submission.
5 According to the Government Accountability Office (GAO), BOP reimburses the residential reentry
centers half the overall per diem rate for each inmate in home detention, but did not have information on
the actual cost of the home detention (GAO, “Eligibility and Capacity Impact Use of Flexibilities to Reduce
Inmates’ Time in Prison,” GAO-12-320 [Washington, DC: 2012],
http://www.gao.gov/assets/590/588284.pdf).
6 Matthew Roland, memorandum to chief probation officers and chief pretrial services officers, April 2012.
7 Quick Facts about BOP.
8 U.S. Department of Justice, Federal Prison System (BOP) FY 2013 Budget and Performance Summary
(Washington, DC: 2012), http://www.justice.gov/jmd/2013summary/pdf/fy13-bop-bud-summary.pdf.
9 U.S. Department of Justice, Overview FY 2013 Budget and Performance Summary (Washington, DC:
2012), http://www.justice.gov/jmd/2013summary/pdf/fy13-bud-summary-requestperformance.pdf#summary.When pretrial detention is included, the total FY 2013 prisons and detention
budget request exceeds 30 percent of the DOJ budget request.
10 From statement of Matthew Axelrod, Associate Deputy Attorney General, U.S. Department of Justice,
before the U.S. Sentencing Commission, Hearing on Federal Sentencing Options after Booker, February 16,
2012, http://www.ussc.gov/Legislative_and_Public_Affairs/Public_Hearings_and_Meetings/2012021516/Testimony_16_Axelrod.pdf.
11 Budget figures were derived from yearly Department of Justice Summary of Budget Authority by
Appropriation tables. All figures are enacted amounts, although there is slight variation in the presentation
of these amounts (some are enacted with rescissions and/or transfers).
Projected budget calculations were made based on the equation: (T2-T1)/T1*100, where the first point in
time used was 1999 (percent change 1999-2000) and the last point used was 2013 (percent change 20122013). From this series, percent changes were averaged and applied as percent change for the following 10
years (2014-2020). Each subsequent year’s figure is based on the previous year’s figure and the percent
increase. For simplicity of presentation, only every fourth year is presented.
12 In addition to federally sentenced offenders (both new commitments and supervision violators), BOP
houses sentenced D.C. felony offenders (since 1997), and some pretrial or pre-sentencing offenders for the
U.S. Marshals Service and for Immigration and Customs Enforcement (Federal Bureau of Prisons, “Quick
Facts about the Bureau of Prisons,” http://www.bop.gov/about/facts.jsp).
13 The shift from mandatory to advisory sentencing guidelines and enactment of the Fair Sentencing Act are
examples of changes that can have a moderating effect on sentence lengths.
14 There are limited opportunities for some offenders to have their sentences reduced below 87.5 percent,
based on prison participation in residential drug treatment programming, and, in rare cases,
“compassionate release.”
15 From BJS Federal Justice Statistics Program (FJSP), http://bjs.ojp.usdoj.gov/fjsrc/,BOP data, as
standardized by FJSP. Note that the immigration cases captured by the U.S. Sentencing Commission do not
reflect the caseload of petty offenses that do not fall under the guidelines.

7

16

In total, 87.4 percent received prison only, 2.5 percent received prison/community split sentences, 7.3
percent received probation only, and 2.8 percent received probation and confinement (U.S. Sentencing
Commission, 2010 Sourcebook of Federal Sentencing Statistics [Washington, DC: 2011], table 12). From FY 2000
to FY 2010, the imprisonment rate reported by the U.S. Sentencing Commission increased from 85 to 90
percent.
17 From 2010 Sourcebook, table 14.
18 From 2010 Sourcebook, table 37.
19 Kamala Mallik-Kane, Barbara Parthasarathy, William Adams, “Examining Growth in the Federal Prison
Population, 1998 to 2010,” 2012, https://www.ncjrs.gov/pdffiles1/bjs/grants/239785.pdf.
20 From BJS Federal Justice Statistics Program (FJSP), http://bjs.ojp.usdoj.gov/fjsrc/, BOP data, as
standardized by FJSP.
21 Administrative Office of the United States Courts, 2011 Annual Report of the Director: Judicial Business
of the United States Courts (Washington, DC: 2012), table E-7A. Not all revoked offenders are sent back to
BOP.
22 GAO, “Growing Inmate Crowding Negatively Affects Inmates, Staff, and Infrastructure,” GAO-12-743
(Washington, DC: 2012).

8