Public Safety,
Public Spending
Forecasting America’s Prison Population 2007–2011

About the Public Safety
Performance Project
n operating project of The Pew Charitable
Trusts, the Public Safety Performance Project
seeks to help states advance fiscally sound, datadriven policies and practices in sentencing and corrections
that protect public safety, hold offenders accountable and
control corrections costs. The project helps states
diagnose the factors driving prison growth and provides
policy audits to identify options for reform, drawing on
solid research, promising approaches and best practices in
other states. The initiative also helps state officials,
practitioners and others share state-of-the-art knowledge
and ideas through policy forums, public opinion surveys,
multi-state meetings, national, regional and state-level
convenings, and online information about what works.

Justice, and former bureau chief of the Bureau of
Research and Data Analysis for the Florida Department
of Corrections.
• Richard Berk, professor of criminology and statistics,
University of Pennsylvania, and former Distinguished
Professor of Statistics and Sociology at UCLA.
• Gerald Gaes, visiting scientist at the National Institute
of Justice, criminal justice consultant and former
director of research for the Federal Bureau of Prisons.

The project works closely with the Pew Center on the
States (PCS), a division of Pew. By conducting nonpartisan
research and analysis, educating the public and federal and
state policy makers, bringing together diverse stakeholders,
and encouraging pragmatic, consensus-based solutions,
PCS identifies and advances effective public policy
approaches to critical issues facing states.

Substantial contributions to the report also were made by
the Vera Institute of Justice and the Council of State
Governments Justice Center, partners of the Public Safety
Performance Project. Staff of both organizations reviewed
drafts of the report and offered excellent comments and
insights that were instrumental to its completion.

A

About this Report
This report was prepared for the Public Safety
Performance Project by the JFA Institute, a well-respected,
Washington-based, nonprofit consulting firm. JFA is led
by James Austin, Wendy Naro and Tony Fabelo, three
nationally renowned researchers with deep expertise in
state criminal justice policy and statistics. JFA conducts
prison population forecasts under contract with a number
of states, and several other states use JFA’s software to
make their projections.
The report was reviewed by three independent specialists
in prison population forecasting:
• William Bales, associate professor, Florida State
University, College of Criminology and Criminal

While these experts have screened the report for
methodology and accuracy, neither they nor their current
or former organizations necessarily endorse its findings
or conclusions.

We also would like to thank the 50 state correctional
agencies and the federal Bureau of Prisons, which
provided much of the data used to create the national
forecast and other parts of this report.

Contact Information
For more information, please visit www.pewpublicsafety.org
or contact Project Director Adam Gelb at
agelb@pewtrusts.org or (404) 848-0186.
The Pew Charitable Trusts is driven by the power of knowledge
to solve today’s most challenging problems. Pew applies a
rigorous, analytical approach to improve public policy, inform the
public and stimulate civic life. We partner with a diverse range of
donors, public and private organizations and concerned citizens
who share our commitment to fact-based solutions and goal-driven
investments to improve society.

Public Safety Performance Project

i

Executive Summary

fter a 700-percent increase in
the U.S. prison population between
1970 and 2005, you’d think the
nation would finally have run out of
lawbreakers to put behind bars.

A

But according to Public Safety, Public Spending:
Forecasting America’s Prison Population 20072011, a first-of-its-kind projection, state and
federal prisons will swell by more than
192,000 inmates over the next five years.

The national price tag is staggering. The
projected 192,023 new prisoners—leave aside
the current population of more than 1.5
million inmates—could cost as much as $27.5
billion: potentially a cumulative $15 billion in
new operating costs and $12.5 billion in new
construction costs by 2011. Every additional
dollar spent on prisons, of course, is one
dollar less that can go to preparing for the
next Hurricane Katrina, educating young
people, providing health care to the elderly,

This 13-percent jump triples the projected

or repairing roads and bridges.

growth of the general U.S. population, and
will raise the prison census to a total of more
than 1.7 million people. Imprisonment levels
are expected to keep rising in all but four
states, reaching a national rate of 562 per

The number of women prisoners is projected

you put them all together in one place, the

to grow by 16 percent by 2011, while the

incarcerated population in just five years will
outnumber the residents of Atlanta,

male population will increase 12 percent. In
some states this disparity is particularly
striking. Nevada, for example, is projecting a
36-percent increase in female prisoners over
the next half-decade.

National Prison Population, 1980-2011
2,000,000
1,800,000
1,600,000
1,400,000

Projected

1,200,000
1,000,000
Actual

800,000
600,000
400,000
200,000
1983

outnumber women behind bars, but women
are playing a dubious kind of catch-up here.

100,000, or one of every 178 Americans. If

Baltimore and Denver combined.

0
1980

Don’t picture this parade of prisoners as an
exclusively male group. Nationwide, men

1986

1989

1992

1995

1998

Sources: Bureau of Justice Statistics (historical) and JFA Institute

ii Public Safety, Public Spending

2001

2004

2007

2011

Gender differences aren’t the only area in
which trends vary widely among states and
regions. Although national prison populations
aren’t currently growing at the same furious
pace as they were a few years back, in some
states and regions growth rates remain in
crisis mode. Prison populations in the West,
Midwest and South are expected to increase

by double-digit percentages between 2006
and 2011, led by the West with a projected
growth rate of 18 percent. The Northeast,
with its slow population growth and steady
crime rates, will see slower but still costly
growth of 7 percent during the same period.

National Prison Incarceration Rate, 1980-2011
600
500
Projected
400
Actual
300

A few other trends add to the image of states’
prisons and budgets stretched at the seams:
Over the next five years, the average
inmate will be more likely to be female or
elderly—both groups that have special
needs and higher costs.
In some states, corrections officials, already

200
100
0

•

•

1980

1983

1986

1989

1992

1995

1998

2001

2004

2007

2011

Sources: Bureau of Justice Statistics (historical) and JFA Institute

State Highlights

having difficulty hiring and keeping guards
on the job, are becoming more and more

•

concerned about finding and retaining
qualified personnel to staff new prisons.
In some states, especially in the West,

This report provides forecasts for prison
populations and incarceration rates for all 50
states. Among its findings:
By 2011, without changes in sentencing or

•

release policies, Alaska, Arizona, Idaho,
Montana and Vermont can expect to see
one new prisoner for every three currently

Midwest and South, methamphetamine
cases have become significant contributors

•

to prison growth.
In the past few years, many states have
enacted enhanced penalties for sex crimes.
The impact of most of these laws on prison
populations and state budgets will be felt
beyond the five-year window of this report.

•

in the system.
Similarly, barring reforms, there will be one
new prisoner for every four now in prison

•

in Colorado, Washington, Wyoming,
Nevada, Utah and South Dakota.
Incarceration rates are expected to spike in
Arizona and Nevada, from 590 and 540
prisoners per 100,000 residents,
respectively, to 747 and 640. Particularly
worrisome is the growth in the population
of young males, the group at highest risk of
criminal activity. Both states have recently
increased their prison population forecasts
because of the combined impact of

10 Highest-Growth States
(by percent increase)
Montana
Arizona
Alaska
Idaho
Vermont
Colorado
Washington
Wyoming
Nevada
Utah

41%
35%
34%
34%
33%
31%
28%
27%
27%
25%

•

demographics and policies that increase
prison terms.
Louisiana, which has the highest
incarceration rate among states, with 835
prisoners per 100,000 residents, expects
that figure to hit 859 by 2011.

• Florida is anticipated to cross the 100,000prisoner threshold within the next five

Public Safety Performance Project

iii

•

•

years, the only state other than Texas and
California to do so.
None of the states is projecting an actual
decrease in its number of prisoners between
2006 and 2011. The report projects no
growth in Connecticut, Delaware and New
York.
The Midwest’s prison population continues
to rise primarily because of increases in new
prison admissions and parole violations.

•

Iowa’s prison population is expected to
increase at a slower rate than other
Midwest states.
Though the Northeast boasts the lowest
incarceration rates, it has the highest costs
per prisoner, led by Rhode Island ($44,860),
Massachusetts ($43,026) and New York
($42,202). The lowest costs are generally in
the South, led by Louisiana ($13,009),
Alabama ($13,019) and South Carolina
($13,170).

Driving Forces
Predicting the future is a risky business, of
course. In Charles Dickens’ Christmas Carol,
Scrooge asks the last ghost that appears to
him, “Are these the shadows of the things
that Will be? Or are they shadows of things
that May be, only?”
In the world of criminal justice policy, as much
as in Dickens’ famed tale, nothing is inevitable.
The size and attributes of a state’s prison
population are linked to an array of factors.
Population growth and crime rates can be the
fuel for this fast-moving train, but the throttle
is in the hands of state leaders who make
related policy choices. Some of these decisions
are made on the basis of careful analysis of
facts and history. Others are predicated on
anecdote and the impact a single, particularly

iv Public Safety, Public Spending

heinous crime can have on the public’s views
about the appropriate punishment for that
type of offense and incarceration in general.
The size of a state’s prison system is
determined by two simple factors: how many
people come in and how long they stay. Yet
both variables are the products of a dizzying
array of influences, from policy-level decisions
and the discretion that judges, prosecutors
and corrections officials exercise in individual
cases, to the larger forces at work in society.
During the past three decades, a number of
changes in states’ sentencing and corrections
policies have been particularly significant.
These include movement from indeterminate
to determinate sentencing; abolition of parole
and adoption of truth-in-sentencing
requirements; lower parole grant rates; passage
of “three-strikes” laws; and establishment of
sentencing guidelines. While the impact of
reforms varies in each state, the states report
that these policy decisions are among the major
drivers of their prison populations.

Implications for Public
Safety and Public Policy
It’s a tempting leap of logic to assume that
the more people behind bars, the less crime
there will be. But despite public expectations
to the contrary, there is no clear cause and
effect. In fact, the question of the effect that
imprisonment has on crime rates cannot be
solved with simple arithmetic. It requires
something more like a social policy calculus.
The central questions are ones of
effectiveness and cost. Total national
spending on corrections has jumped to more
than $60 billion from just $9 billion in 1980,

and yet recidivism rates have barely changed.
More than half of released prisoners are back
behind bars within three years. If states want
the best results from their correctional
systems over the next five years—both in
terms of public safety and public spending—
how should they approach the significant
prison population growth that is anticipated?

Methodology Overview
Forecasting prison populations has grown more sophisticated since
the days of estimating using time series or trend analysis, which
showed what had already happened but failed to make accurate
projections of future patterns. Today’s more advanced models are
designed to mimic the flow of the correctional system based on
probabilities of prison admissions and inmate lengths of stay.

That question is the chief challenge states are
facing. They are not fated to such high rates
of prison growth by factors out of their
control. The policy choices they make—the
sentencing and release laws, programs and
practices they enact and fund—are principal
determinates of the size, effectiveness and
cost of their corrections systems.
The key is for policy makers to base their
decisions on a clear understanding of the
costs and benefits of incarceration—and of
data-driven, evidence-based alternatives that
can preserve public safety while saving muchneeded tax dollars. To begin the process of
looking at costs and benefits, state policy
makers need to know whether, and at what
rate, their correctional system is likely to
grow, and how their system’s growth rate
compares to that of other states. By providing
this comparative data, this forecast can assist
states in their efforts to develop cost-effective
options that reduce corrections expenditures
while protecting public safety.
Those last two words—public safety—are of
particular consequence. No policy maker is
likely to (or should) pursue a path that saves
prison money if it runs a substantial risk of
increasing recidivism or crime rates. On the
other hand, an option that can lead to better
public safety outcomes while saving money is
the picture that goes alongside the dictionary
definition of win-win.

This national prison projection report was generated from data from
the states themselves. The federal Bureau of Prisons and 42 states
(including the 36 states that use advanced simulation methods)
provided their official forecasts to form the basis of this report. Those
jurisdictions accounted for 92 percent of the national prison population
as of 2005. The remaining eight states were unable to provide
projections, so researchers calculated estimates using the states’ own
most recent monthly population counts and available admission and
release data. Those estimates—for Alaska, Connecticut, Delaware,
Maine, New York, Utah, Wisconsin and Wyoming—are not official
forecasts.
Researchers also contacted each state to obtain the most current costs
per prisoner. The cost figures included administrative support, program
services and facility maintenance. If a state contracts with a private
prison company, researchers attempted to incorporate those into the
annual cost figure.
It’s important to note that an increase or decrease in a state prison
population will not yield a direct change in operating costs. Some
states whose prison populations grow by only a small amount will
experience only marginal cost increases, such as the costs of medical
care and food; they will likely not need to hire additional staff or build
new cells. Other states may pass a tipping point and proceed with
constructing new prisons and taking on new staff.
It’s possible, too, that the projected population may involve
disproportionately lower-custody inmates or that a state may employ
alternative, lower-cost housing methods and divert some offenders into
community punishments. These scenarios would result in an
overestimate of future costs if the estimate is made using an average
cost per inmate.
Capital costs for corrections are more difficult to project than operating
costs. Prison beds cost about $65,000 to construct, but total
construction cost figures exclude renovation and conversion of
existing bed space.
For these reasons, the report does not provide cost estimates for each
individual state.

Public Safety Performance Project

v

Table of Contents

Introduction .....................................................................................................................1
Forecasting Correctional Populations ........................................................................3
Micro-simulation Models ..........................................................................................................6
Accuracy of the Projection Models...........................................................................................7
National Prison Population Projection Estimates ...................................................9
Growth of Women Prisoners Will Continue to Outpace Males............................................10
Age of Inmates (and the Cost of Their Medical Care) is Expected to Rise .........................11
Corrections Workforce Recruitment and Retention is a Growing Concern .........................11
Rise in Methamphetamine-related Cases ...............................................................................11
Impact of Enhanced Sex Offender Sentences Will Be Felt Beyond Five Years .....................12
Regional and State Trends..........................................................................................13
Northeastern Region ...............................................................................................................13
Midwestern Region .................................................................................................................14
Southern Region .....................................................................................................................15
Western Region .......................................................................................................................17
Estimating Current and Future Prison Costs .........................................................18
Methodological Issues.............................................................................................................19
Current Operational Costs .....................................................................................................20
Estimates of Future Operational Costs ..................................................................................21
Capital Costs...........................................................................................................................22
The Relationship Between Incarceration and Crime Rates.................................23
Public Safety, Public Spending:
The Challenge Ahead for State Policy Makers........................................................25
Appendix.........................................................................................................................27

Public Safety Performance Project

vii

Tables and Figures
Table 1:
Figure 1:
Figure 2:
Figure 3:
Figure 4:

Adult Correctional Populations, 1980-2005..........................................................2
Schematic Flow of Prison Population Components..............................................4
Crime and Incarceration Rates by State, 2004 .....................................................5
National At-Risk Population: Males Between 18-34.............................................6
Accurate Projections: West Virginia, 2004-2006 ..................................................8

Figure 5:

Projections Responding to Change: Nevada, 2005-2006 .....................................8

Figure 6:

Projected National Prison Population and Incarceration Rate, 2006-2011 ........10

Figure 7:

Projected Change in Regional Incarceration Rates, 2006-2011 .........................10

Table 2:

Ohio 10-Year Prison Population Projections, 2007-2016 .....................................14

Figure 8:

Projected Year-End Resident Population by Region, 2006-2011 ........................15

Table 3:

Nevada 10-Year Prison Population Projections, 2007-2016.................................16

Table 4:

Arizona 10-Year Prison Population Projections, 2007-2016 ................................16

Table 5:

Costs Per State Prisoner, 1984-2005 ...................................................................21

Figure 9:

National Crime and Imprisonment Trends, 1931-2005 .....................................23

Appendix
Table
Table
Table
Table

A-1:
A-2:
A-3:
A-4:

Key State Data, 2005 ...........................................................................................27
State, Regional and National Residential Populations, 2005-2011.....................28
State Prison Populations by Region, 2006-2011 .................................................29
State Prison Populations by Growth Rate, 2006-2011 .......................................30

Table A-5: State Incarceration Rates by Region, 2006-2011 ................................................31
Table A-6: State Incarceration Rates by Growth Rate, 2006-2011.......................................32
Table A-7: Annual Operating Costs per Inmate...................................................................33
Table A-8: Sources of State Prison Population Projections...................................................34
Table A-9: Sources of State Inmate Costs .............................................................................36

viii Public Safety, Public Spending

Introduction

his report estimates the
future size and cost of the state and
federal prison systems. It examines
the reasons for the projected growth and, since
prison expansion is generally intended to
reduce crime, it outlines what we currently
know about the relationship between
incarceration and crime rates. Finally, the

T

report highlights the efforts of some states to

world in incarceration rates, well above Russia
and Cuba, which have the next highest rates
of 607 and 487 per 100,000. Western
European countries have incarceration rates
that range from 78 to 145 per 100,000.3
Probation and parole populations have
skyrocketed alongside the rapid growth in the
state and federal prison systems. Since 1980,

control corrections spending while protecting
public safety and holding offenders
accountable for their actions.

the total correctional population has grown

The past three decades have witnessed an

adults are on probation, making that the
largest component of the correctional system;

historic increase in the nation’s penal system
at all levels. In 1970, the state and federal
prison population was less than 190,000. The
latest report by the U.S. Department of Justice
puts the 2005 population at nearly 1.5
million. Further, almost 750,000 people are
incarcerated in local jails, resulting in a total
incarcerated population of almost 2.2 million,
or 737 per 100,000 U.S. population.1 Put
differently, for every 1,000 U.S. residents,
seven are incarcerated either in jail or prison
on any given day. Each year, over 600,000
people are admitted to state and federal
prisons. A much larger number (over 10
million) go to local jails. There are another
4.3 million ex-convicts living in the U.S.2
The U.S. imprisons significantly more people
than any other nation. China ranks second,
imprisoning 1.5 million of its much larger
citizen population. The U.S. also leads the

from 1.8 million to over 7 million people
(Table 1). While the prison population has
grown at the fastest rate, more than 4 million

it too has nearly tripled since 1980.
While noteworthy in their own right, national
trends tend to mask significant state-level
variation. This is the case both for
incarceration (covering jails and prisons)4 and
the population under community supervision
(including parole and probation). For
example, while the national prison
incarceration rate in 2005 was 491 per
100,000 residents, Louisiana had the highest
prison incarceration rate (797 per 100,000)

At year-end
2005, there were
almost 2.2 million
people—one in
every 136 U.S.
residents—
in U.S. jails
and prisons.

followed by fellow Southern states Texas
(691), Mississippi (660) and Oklahoma (652).
Maine had the lowest incarceration rate (144),
followed by Minnesota (180), Rhode Island
(189) and New Hampshire (192).5
While it is generally true that Southern states
have high incarceration rates while

Public Safety Performance Project

1

national correctional system. Currently, each
state bears responsibility for forecasting its
own population. A national forecast such as
this will have several important uses.

TABLE 1

Adult Correctional Populations, 1980–2005
Population
Probation
Jail
Prison
Parole
Total Adults
Under Corrections

1980
1,118,097
183,988
319,598
220,438

2005
4,162,536
747,529
1,461,132
784,408

% Change
272%
306%
357%
255%

1,842,100

7,155,605

288%

Adult Population
% of Adults Under
Corrections

162.8 Million

222.3 million

36%

1.1%

3.2%

Sources: U.S. Department of Justice, Office of Justice Programs. Prisoners in 2005, Bureau
of Justice Statistics Bulletin, by Paige M. Harrison and Allen J. Beck (Washington, D.C.:
November 2006), NCJ 215092; U.S. Department of Justice, Office of Justice Programs.
Prison and Jail Inmates at Midyear 2005, Bureau of Justice Statistics Bulletin, by Paige M.
Harrison and Allen J. Beck (Washington, D.C.: May 2006), NCJ 213133 and U.S. Department
of Justice, Office of Justice Programs. Probation and Parole in the US 2005, Bureau of
Justice Statistics Bulletin, by Lauren E. Glaze and Thomas P. Bonozar (Washington, D.C.:
November 2006), NCJ 215091

Northeastern states have low rates, there is
considerable variation even among states
from the same region or sharing similar
crime rates. For example, North and South
Dakota had low but very different
incarceration rates in 2005: 208 per 100,000
for North Dakota versus double that—443—
for South Dakota. In the South, North
Carolina’s incarceration rate is 360 while
South Carolina’s is 525.6 As discussed later,
these pronounced differences in incarceration
rates often reflect different sentencing laws and
correctional policies that have been adopted by
policy makers. In other words, the size and
attributes of a state’s prison population are
heavily determined by policy choices.
In light of that, it would be valuable for policy
makers and the public to understand the likely
future outcomes in states that have adopted
varying policies. While the U.S. Department
of Justice provides accurate and
comprehensive historical data on the size and
attributes of the various correctional
populations, there is no organization or agency
that provides estimates of the future size of the

2 Public Safety, Public Spending

First, state policy makers need to know how
much their correctional system is likely to
grow, if at all, so that they at least can ensure
that sufficient funds are available to support
growth. This is especially true for the jail and
prison systems that must maintain standards of
care for their prisoners. Second, because
differences in population increases often reflect
differences in criminal justice policies,
understanding such policy differences and their
impact on prison populations and costs can
help policy makers better evaluate whether
they should pursue reforms. Third, given the
large and increasing amount of taxpayer funds
being devoted to prison systems, policy makers
want to ensure that their investments in public
safety are generating their intended results. If
other states are slowing the growth of their
prison populations while achieving better
public safety outcomes, such as lower
recidivism rates or lower crime rates, policy
makers want to know that.
Finally, the costs of constructing and operating
jail and prison systems are an ongoing concern
for policy makers. Between 1982 and 2003,
national spending on criminal justice increased
from $36 billion to $186 billion. Over $61
billion of that total is allocated to local, state,
and federal corrections.7 Indeed, corrections
spending—which consists primarily of budgets
for jails and prisons—grew by more than 570
percent during that period, faster than any
other aspect of the criminal justice system.
Given the phenomenal period of growth in
correctional populations and its associated
costs to the taxpayer, public officials are
becoming increasingly concerned about what
the costs will look like in the future.

Forecasting Correctional
Populations
stimating the future size of any
correctional system is part science and
part judgment. Criminal justice policy
is a dynamic phenomenon and is difficult to
predict with a high degree of certainty.
During the past three decades, we have
witnessed a wide array of policy shifts in
sentencing, including some states abolishing

E

The basic formula is:
Prison admissions x length of stay (LOS) =
Average Daily Population (ADP)8

parole, moving from indeterminate to

This simplistic formula becomes far more
complex when one begins to understand the
myriad factors that can influence admissions
and the LOS. Relatively minor changes in
admissions or LOS can have an enormous

determinate sentencing, establishing
sentencing guidelines, and adopting truth-insentencing and “three-strikes” laws. Many of

in a prison system is 30 months, an increase
of three months in the LOS would increase

these changes were intended to remove repeat
offenders from the streets. But as the cost of
corrections has skyrocketed, so has interest in
finding cost-effective options that could
reduce expenditures without jeopardizing
public safety.

impact on the ADP. For example, if the LOS

the ADP by 10 percent. Changes in the LOS
can be achieved by modifying sentence
lengths, awarding or rescinding good time
credits, changing parole eligibility dates, and
paroling (or not paroling) offenders at either
their initial parole date or

Between probation, parole,
jail and prison, the U.S.
correctional population
exceeds 7 million people.
One in every 32 U.S. adult
residents is currently under
correctional supervision.

Identifying these options requires sound

at a subsequent parole
hearing.

research, comprehensive analysis and reliable
forecasting techniques to better inform
policy makers and the public about the
consequences of current and proposed
policies. Estimating the future prison
population is the beginning of this enterprise,
not the end. Decision makers need to
understand why prison populations are
growing and how future changes will affect
the system.

Figure 1 illustrates the
various internal and
external factors that
influence ADP and
therefore influence a
forecast of the future ADPs.
External factors reflect the
interplay of demographic, socio-economic and
crime trends that produce arrests, and

In the simplest terms, prison populations
(and all correctional populations) are the
result of two factors: the number of people
admitted to prison and how long they stay.

offenders’ initial entry into the criminal justice
process. Criminologists have long noted that
certain segments of the population have higher
rates or chances of becoming involved in
crime, being arrested and being incarcerated.

Public Safety Performance Project

3

FIGURE 1

Schematic Flow of Prison Population Components

Demographics—at risk population

Crime

Arrests

New
Charge

Convictions

New
Charge

Probation
Prison

Technical
Violator

Technical
Violator

Parole/Community Supervision

Release to Community

4 Public Safety, Public Spending

FIGURE 2

Crime and Incarceration Rates by State, 2005

Incarceration Rate per 100,000 residents

800

LA

700

TX
MS

OK

600

AL

500

ID
KYVA

SD
400

MO GA
SC AZ
FL
AR NV
CO
TN
AK
WY
OH MD
IN
MT
NC OR
IL
KS
NM HI
MI
DECA

CTWI
NY NJPA
WV IA
VT
NE
MA
NH ND
RI
MN
ME

300
200

WA

UT

100
0
0

1000

2000

3000

4000

5000

6000

Crime Rate per 100,000 residents

Source: FBI Uniform Crime Report and BJS

This is known as the “at-risk” population,
which generally consists of younger males.
The high crime rate ages are 15-25, while the
high adult incarceration rate is between the
ages of 18 and 35. When the at-risk
population is expected to increase in a
jurisdiction, one can also expect some
additional pressure on criminal justice
resources, all things being equal.
Figure 2 shows the association between crime
rates (which are produced in part by
demographic and socio-economic trends) and
incarceration rates. The figure plots the crime
and incarceration rates for each state, showing
that states with low crime rates tend to have
lower incarceration rates. The spread of states
up and to the right on the graph shows that
states with higher crime rates tend to have high
incarceration rates. The last section of this
report summarizes what is known about the
relationship between crime and incarceration.

It is unfortunate but true that AfricanAmericans and Hispanics have significantly
higher arrest and incarceration rates than
whites. One must also factor in the extent to
which these racial and ethnic groups within
these age ranges are also projected to
increase. As shown in Figure 3, the number
of at-risk African-American and Hispanic
males has been increasing over the past few
years. States that are projected to have a
larger at-risk population over the next decade
also are likely to experience continued
pressures on criminal justice and correctional
resources based on demographic growth.
Internal factors reflect the various decision
points within the criminal justice system that
cumulatively determine prison admissions
and LOS. These decisions begin with police
and end with correctional officials who,
within the context of the court-imposed
sentences, have the authority to release,

Public Safety Performance Project

5

criminal justice systems often vest considerable
discretion in their public leaders who construct
these policies and procedures. A complete
understanding of these complex influences is
essential to the accuracy of planning and
forecasting a prison or jail population.

FIGURE 3

National At-Risk Population:
Males Between 18-34
25 m

20 m

White

8m

Micro-simulation Models
7m

Hispanic

6m
5m

Black

4m
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

Source: U.S. Census Bureau

Traditionally, prison populations were
estimated using time series or trends analysis.
This was easy to do since the historic counts
were readily available and it required little
—
skill to use such methods. These methods
were very inaccurate, however, especially in

recommit, give and restore a wide array of
good time credits, and offer supervision and
services that may reduce recidivism.9

an environment where policy is very

For example, one of the most difficult

or future criminal justice policies and

numbers to estimate is the number of prison
admissions for the next five years. As

sentencing legislation.

suggested by Figure 1, people come to prison

To better account for such a complex and
dynamic system, a new generation of microsimulation models has been developed to help
decision makers estimate the effects of current
policies and the likely consequences of specific
policy proposals. These micro-simulation

for three basic reasons: (1) they have been
directly sentenced by the courts to a prison
term (new court commitments); (2) they have
failed to complete their term of probation and
are now being sentenced to prison for a
violation of the conditions of their release or
new crime; or, (3) they have failed their term
of parole (or post-release supervision) and are
being returned to prison for a violation of the
conditions of their release or new crime.
Almost two-thirds of the estimated 600,000plus people who are admitted to prison are
those who have failed to complete probation
or parole. A projection model thus should
have a “feedback loop” that captures the
expected rate of probation and parole failures.
The impact of recently enacted sentencing
laws, judicial decisions and other criminal
justice policy choices also must be considered
in a population forecast. These complex factors
also vary from state to state. State and local

6 Public Safety, Public Spending

dynamic. Time series models can show only
what has already occurred; they cannot
estimate future populations based on current

models are designed to mimic the flow of (1)
the current prisoner population, and (2) the
expected new admissions over the projection
horizon based on these internal factors. Based
on stochastic entity simulation methods, the
models mimic the actual flow of the
correctional system based on current and future
probabilities of being admitted to prison under a
particular legal status, with a certain sentence
for a certain crime, and being released at a
certain time based on probabilities of receiving
good time and being released on parole.
Similarly, each person released to probation or
parole has a certain probability of being
revoked for a new crime or technical violation
and being returned to prison for a certain
period of time before being re-released. All of

these “probabilities” are based on the current
behavior of the decision makers.

Accuracy of the
Projection Models
A recurring question about any projection
model is its accuracy. In one sense this is the
wrong question to ask, since a forecast of any
correctional system is predicated upon the
assumptions of future criminal justice policy.
Because such policies are constantly in flux,
the projection must be modified as lawmakers
adopt new policies and correctional officials
adjust their administrative procedures. For

Time series or regression models are not able
to employ such techniques and thus are less
able to demonstrate their accuracy. Moreover,
because they are based on historical patterns
that do not account for contemporary policies
or laws, they often either over- or
underestimate short-term developments.
Figures 4 and 5 highlight recent accuracy
analyses for West Virginia and Nevada, both
of which employ simulation models. West
Virginia reflects a fairly stable policy
environment, so the 2004 projection has been
quite accurate for the past two years.

example, if a parole board implements new

Conversely, the Nevada estimate issued in

parole guidelines that serve to increase the

March 2005 began to display an

rate of parole for low-risk prisoners from 35
percent to 50 percent, the projection model’s

underestimate in fall 2005. This was caused
by a significant and unexpected surge in new

parole grant rates must be similarly adjusted

court commitments, largely from the Las

and thus show a lower forecast. If the
legislature adopts a longer sentencing range

Vegas metropolitan area. The model’s new
court intake estimates were then adjusted

for drug dealers that is not retroactive to the

with the assumption that new admissions
would continue to grow at the 2006 rather
than the 2005 rate. As shown in the graph,

current prisoner population, the new
admission stream must be altered and will
show a higher projection.
Despite the nuances of the dynamic policy
arena, the models must demonstrate that they
would be accurate if policies remain constant.
The micro-simulation models are especially
adept in this regard if they are designed to
model both the current and future correctional
populations. For the first 12 to 18 months of a
projection, the current parole and prison
populations have a large influence on the
forecast since it takes that long for large
numbers of that population to exit. Further, the
micro-simulation models are loaded with the
most current data to reflect current practices
and are then “started” several months in the
past to see if they are mimicking actual
monthly counts of admissions, releases and
populations. Only when this test has been
successful is the forecast deemed “accurate.”

this single change in the new admission
assumption increased the 10-year forecast by
over 900 prisoners.

Public Safety Performance Project

7

5000

Two of the most significant examples of
overestimates occurred in Virginia after it
adopted truth-in-sentencing laws and in
California after it adopted its “three-strikes”
mandatory sentencing laws. The Virginia
error resulted in a massive over-construction
plan to build prison beds that were not
needed. In subsequent years Virginia was able

4800

to cancel some of its construction plans and

4600

recoup some of its losses by renting out the

FIGURE 4

Accurate Projections: West Virginia, 2004-2006
5800
5600

Inmates

5400
Actual

5200

Projected

surplus prison beds at a profit to states that
had crowded systems.

4400
4200
Jan
04

Apr
04

Jul
04

Oct
04

Jan
05

Apr
05

Jul
05

Oct
05

Jan
06

Apr
06

Jul
06

In California, the original estimate was that the
Source: JFA Institute

“three-strikes” legislation would more than
double the inmate population from 121,000
prisoners in 1994 to over 245,000 in 1999. It
turned out that the prison population rose to

FIGURE 5

Projections Responding to Change:
Nevada, 2005-2006

160,000. The estimate was off by a staggering

13500
July 2006 projection
13000

Inmates

12500

Actual population

12000
11500
April 2005 projection

11000

bargain a large number of cases to lesser

10500
10000
Jan
05

Mar
05

May
05

Jul
05

Sep
05

85,000 inmates. The primary source of the
error was an assumption that all criminal cases
that fit the criteria for either a second- or thirdstrike sentence would be so prosecuted. In
reality, prosecutors used the law to plea

Nov
05

Jan
06

Mar
06

May
06

Jul
06

Sep
06

Source: JFA Institute

The level of accuracy raises the issue of
under- and overestimates. It is fair to say that
correctional officials are more fearful of an
underestimate, which may lead to crowding
and perhaps a more dangerous prison
environment. Overestimates typically pose
little operational problem to prison officials
who may welcome a surplus of vacant prison
beds or at least a reduction in existing
crowding. However, overestimates are viewed
with disdain by some state fiscal analysts,
who may feel (rightly or wrongly) that the
projections were manipulated by the prison
agency to secure extra, unneeded funding.

8 Public Safety, Public Spending

charges. And in several major counties,
including San Francisco and Alameda
(Oakland), prosecutors rarely applied the law.10
The lesson for “projectionists” is that they
must anticipate adjustments that practitioners
will make to new policies that strain their
agencies’ capacities or their local community
standards. For instance, it can’t be assumed
that mandatory sentencing laws will be strictly
followed by prosecutors or the courts. For this
reason it is useful to discount the estimated
effects of such laws.

National Prison Population
Projection Estimates
o make an estimate of the
U.S. prison population, the researchers
for this report contacted each of the 50
states and the federal Bureau of Prisons (BOP)
and requested their current official population
projections. Where available, projections by
gender were also requested.

T

The BOP and 42 states provided at least a fiveyear prison population forecast. These
reporting jurisdictions accounted for 92 percent
of the national prison population as of 2005.
For the remaining eight states, researchers
made estimates based on current population
trends and extrapolated for five years.11
Figures 6 and 7 provide the national and
regional estimates based on the data received
from the states and the BOP and the
estimates for states with no official projection.
Detailed tables for each state are shown in
the appendix. The national and state
estimates reveal the following major trends:
1. The nation’s state and federal prison
population will reach 1,722,477 by 2011—
an increase of approximately 192,000 over
a five-year period.
2. This rate of growth—about 38,400 more
inmates per year—is markedly higher than
the growth rate of the past three years.
3. The prison incarceration rate will continue
to grow, from 491 per 100,000 U.S.
residents in 2005 to 511 per 100,000 in
2006, then to 562 per 100,000 in 2011.

4. The Western region will have the largest
prison population increase (18 percent)
while the Northeast will experience the
smallest growth (7 percent).
5. There is considerable variation among the
states. Montana, Alaska, Arizona, Idaho,
Vermont and Colorado all are poised to
grow by more than 30 percent under
current criminal justice policies.
Conversely, Connecticut, Delaware, New
York and Maryland are expected to have
little if any growth.
6. Four states—Florida, California, Arizona
and Texas—and the federal prison system
will account for more than 87,000
additional prisoners, or about 45 percent of
the total prison population increase.
In reviewing these trends
and discussing them with

By 2011, America’s prison
population is projected to
increase by 192,000 to over
1.7 million inmates.
One in every 178 U.S.
residents will live in prison.

the states, researchers
learned that a wide array of
factors were influencing
these estimates. For a
number of Southern and
Western states, demographic
growth, particularly for the
at-risk population, was a
major concern. This was especially true in
Arizona, Nevada and Texas, all of which have
recently increased their prison population
estimates because of increases in prison
admissions for new court sentences or
probation revocations. However, incarceration

rates in all three states will grow, meaning that

Public Safety Performance Project

9

FIGURE 6:

Projected National Prison Population
and Incarceration Rate, 2006-2011
US Prison Population
in millions
1.80

Inmates per
100,000 residents
600
562

1.75

440

A region-by-region
summary of the
estimates and factors
that underpin the
estimated growth
follows. But before
proceeding to these
regional variations, a
number of other

420

policy-related issues

400

merit discussion.

580

553
545

Incarceration Rate

1.70

560

534
1,722,477

521

520

1,654,668
Prison Population

1.60

500

1,614,808
1.55
1.50

540

1,686,495

511

1.65

480

1,568,822

460

1,530,454

1.45
1.40
2006

2007

2008

2009

2010

2011

inability to reduce
recidivism rates—all
contributed to the
higher projections.

These issues emerged
during researchers’

interviews with state
correctional officials and planners who are
directly involved in the states’ forecasts.

Source: JFA Institute

FIGURE 7

Growth of Women Prisoners Will
Continue to Outpace Males

Projected Change in Regional
Incarceration Rates, 2006-2011

12%

The female prisoner population, while well
below the size of the male prisoner
population, has been growing at a faster rate

10%

for many years. The Bureau of Justice

16%
14%

8%
6%
4%
2%
0%
Northeast

Midwest

South

West

Source: JFA Institute

the greater prison admissions or longer LOS,
or both, are causing the prisons to grow faster
than the general population. In these and other
states, state officials reported that the
cumulative effects of lengthy mandatory prison
terms adopted in the 1980s and 1990s, reduced
parole grant rates, and high numbers of parole
and probation violators—coupled with an

10 Public Safety, Public Spending

Statistics (BJS), part of the U.S. Department
of Justice, notes in its most recent prison
population report that the female population
has grown by 57 percent since 1995,
compared to a 34-percent increase for males.12
For this forecast, 25 states, covering only
about one-third of the national prison
population, were able to provide their
projections by gender. In these 25 states,
females are expected to grow at a faster rate
(16 percent) than males (12 percent).
Researchers’ interviews with other state
correctional officials suggest that higher
female growth rates are likely to continue in
the other states as well.

Disaggregating in this manner is desirable
because women have unique security and
programmatic needs that may not be met if
the size of the female population is not
properly estimated. For example, women are
typically housed in much lower-security-level
facilities than men and require a lower staffto-inmate ratio. The construction of female
facilities is increasingly designed to meet the
unique custody and service needs of women.
Also, because the female prison population
has risen faster for the past decade, failure to
perform separate forecasts by gender could
distort growth estimates for women
prisoners.
In addition, females generally pose a
significantly lower risk to public safety than
males. BJS studies of female recidivism rates
have consistently shown that women have a
lower recidivism rate than males and are far
less likely to commit a violent or sex crime
upon release.13 The disproportionate increases
in the female prison population, then, are
somewhat ironic.

Corrections Workforce
Recruitment and Retention
is a Growing Concern
As their prison populations increase, states
need to find qualified applicants for
correctional officer positions and other prison
jobs. Many of the state officials contacted for
this report expressed concern that even if
they can secure the necessary funding to
build and operate an expanded prison
system, it will be increasingly difficult to find
qualified workers to fill these positions.
These officials already face a high turnover
rate and a growing number of “baby
boomer” employees now nearing retirement.
A number of Southern states (especially
Louisiana, Mississippi and Alabama) are
hoping to increase salary levels to attract and
retain qualified staff to work in prisons that
are often located in economically depressed
rural areas. Such increased salaries will carry
an obvious fiscal burden for state
governments.

Methamphetamine-related
Cases are on the Rise

Age of Inmates (and the
Cost of Their Medical Care)
is Expected to Rise

Many states are seeing significant growth

BJS reports that the average age of prisoners
being released to parole has increased from
31 to 34 between 1990 and 1999.14 There are

example, meth-related admissions more than
tripled, from 977 inmates in fiscal years 1999
and 2000 to 3,579 in fiscal years 2004 and

no more recent national data, and states were

2005. With meth offenders currently serving
an average of 5.5 years in prison, officials
estimate that the cumulative cost of housing
these inmates alone will exceed $340
million.15

not able to provide prisoner age projections
for this report, but policy experts and state
officials are concerned that the aging trend
will accelerate largely because of the longer
prison terms being served under various
sentencing and release laws and policies. This
presents a major fiscal concern for states,
because as the prison population ages, the
medical costs of the corrections system are
expected to rise accordingly.

in prison admissions related to
methamphetamine addiction. In Georgia, for

The rise of meth cases is not readily reflected
in the current forecast, but correctional
officials have become increasingly concerned
that larger proportions of the probation and
parole populations are using the drug and
thereby increasing the likelihood of probation

Public Safety Performance Project

11

and parole revocations. To control the
problem and its impact on prisons, many
correctional officials are calling for more
community-based treatment beds and wider
adoption of evidence-based practices for
treating meth abusers.

Impact of Enhanced Sex
Offender Sentences Will Be
Felt Beyond Five Years
Many states have recently passed sentencing
laws for sex offenders that require a lengthier
period of incarceration and/or a lengthier and
more intense period of parole supervision.
One example is California, which under the
recently passed Proposition 83 requires sex
offenders to be tracked electronically for life.
This law will no doubt increase the number
of parolees returned to prison for technical
violations. In Kansas, a law enacted in 2006
will result in approximately 150 persons
convicted of child sex crimes being sentenced
to prison for terms approximately 16 years
longer than under earlier sentencing
practices.

12 Public Safety, Public Spending

The current five-year state projections do not
reflect the long-term effects of such laws. The
laws typically are not retroactive, and because
many of these offenders already spend longer
than five years behind bars, the impact of the
longer sentences will not be felt on
populations and budgets for some time
beyond the next five years. Over the next
two decades, however, one can expect the
number of prisoners convicted of sex crimes
to expand rapidly.

Regional and State Trends

Northeastern Region
The Northeast historically has the lowest
incarceration rates, which will continue to be
true well into the next decade. Led by New
York, Massachusetts, New Jersey and
Connecticut, these states are estimating little if
any growth. Part of the explanation for this
trend is demographic, as this region is
estimated to grow slowly. Crime rates also are
relatively low. The stability of incarceration
rates results from more than demography and
crime rates, however; states also have adopted
new policies that have controlled prison
population growth. In both Massachusetts
and New Jersey, for example, parole grant
rates have increased while state leaders have
resisted calls to increase sentencing lengths.
Connecticut may provide one of the most
striking and successful examples of policy
intervention. Using data-driven analyses,
Connecticut policy makers identified that
parole and probation violators were driving
much of the prison growth. They passed
legislation in 2004 that set a goal of reducing
parole and probation revocations by 20
percent, and hired 96 new probation officers,
reducing caseloads from approximately 160
cases per officer in January 2004 to
approximately 100 cases per officer in June
2005.
As part of a “justice reinvestment” strategy,
Connecticut redirected $13 million of the
expected savings from those reforms into

recidivism reduction initiatives. They funded
two programs targeting violators, and required
the development of a comprehensive re-entry
plan, with focus on the specific neighborhoods
to which most prisoners were returning.
Within two years following the development
and adoption of this strategy, Connecticut
went from having one of the fastest-growing
prison populations in the nation to
experiencing a decline steeper than almost
any other state. Crime rates in Connecticut
also dropped during this period, faster than
they were falling in the nation overall.
Another big story in the
Northeast has been New
York, where the prison
population has declined
from a peak of 72,889 in
1999 to its current level of
about 63,000. Virtually all
of this historic decline has
resulted from dramatic

Change in five-year projected
state prison populations varies
radically, from no growth in
New York, Delaware and
Connecticut to 41 percent
growth in Montana.

reductions both in serious
crime and in the number of felony arrests,
much of which can be linked to the wellknown reforms within the New York City
police department.16 Indeed, admissions to
state prison from New York City fell from
20,580 in 1993 to 8,490 in 2005. While the
state has not issued a formal prison
population forecast, the most recent trends
show no reason to expect significant increases
over the next five years.

Public Safety Performance Project

13

extending parole terms, especially for sex
offenders. Although the Department of
Corrections has expanded the programmatic
opportunities available to inmates, and linked
participation to additional good-conduct
credits, these efforts have not offset the
impact of sentencing initiatives enacted in
Illinois during the late 1990s.
Ohio had been experiencing declining prison
populations since 1999 as a result of a
sentencing reform initiative. Now the state is
experiencing increases because of higherthan-expected prison admissions. A surge in

Midwestern Region

admissions of white females from a number

The prison population of the Midwest
continues to grow, primarily as a result of
increases in prison admissions from both new
court admissions and parole violations. In
some states the long-term effects of truth-insentencing laws that were enacted more than
a decade ago are now affecting lengths of
stay. In Illinois, for example, prison
admissions have increased every year, with
the system thus setting new highs annually.
Parole violation rates are at a record high,
and policy makers have enacted several laws

of rural counties has been especially

TABLE 2

Ohio 10-Year Prison Population Projections,
2007–2016
Date
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
% Change

Male
43,965
45,485
47,563
49,354
50,889
52,625
53,832
55,384
56,941
58,184
59,756
36%

Female
3,554
3,726
3,985
4,249
4,416
4,598
4,699
4,802
4,914
5,088
5,214
47%

Note: 2006 figure is the actual population as of 10/2/06.

14 Public Safety, Public Spending

Total
47,519
49,211
51,548
53,603
55,305
57,223
58,531
60,186
61,855
63,272
64,970
37%

dramatic. Based on these developments, Ohio
estimates it will add over 17,000 inmates to its
prison population over the next 10 years, a
37-percent increase. The female population
will grow at an even faster rate of 47 percent.
Kansas is another Midwestern state that has
changed its direction. Between 2003 and 2006,
the prison population remained fairly stable.
With the passage of new child sex offender
legislation and increases in the number of
offenders being imprisoned for violating
probation, the state’s latest forecast shows that
the prison population will increase from
approximately 9,000 to 11,231 by 2016. These
projections would be even higher were it not
for recent legislative actions and correctional
policy changes that will hold technical parole
violators accountable with graduated sanctions
prior to returning them to prison.
Iowa provides an interesting example of a
state in which the prison population is
projected to grow, but at a slower rate than
other Midwestern states. There have been
fewer new court commitments for the state in
recent years, although that has been
somewhat offset by higher rates of probation

and community
supervision (parole)
admissions. To
control its prison
population, Iowa also
relaxed its truth-insentencing laws,
dropping its
requirement of time

FIGURE 8

Projected Year-End Resident Population
by Region, 2006-2011
320 m
Total
290 m
120 m
South

110 m

served from 85
percent to 70
percent, and
increased the number
of paroles. As a
result of these
changes, Iowa’s
growth rate is
projected to be low

100 m

for the next five
years. The long-term
estimates are higher,

40 m

80 m

West

70 m

Midwest
60 m
Northeast

50 m

2006

2007

2008

2009

2010

Source: JFA Institute

other decisions the state has made, such as
abolishing or restricting parole for certain

which has one of the largest state prison
populations, is estimated to grow by an
additional 13,656 prisoners over the next five

crimes and increasing sentences for sex

years. Florida, another large state, will

offenders. Iowa estimates its prison
population will rise from 8,737 in 2005 to
11,240 in 2015. As in Ohio, the female

incarcerate more than 100,000 people by
2011. At the same time, Maryland and

however, because of the long-term effects of

population is projected to grow faster than
the male population.

Southern Region
The Southern states traditionally have had
the highest rates of incarceration, and that
will continue to be the case. Figure 8 shows
the projected populations of the four regions,
with the South having the greatest projected
growth. Yet the forecast shows Southern
states moving in different directions over the
next five years.
Some Southern states, such as Texas, Florida,
Georgia, South Carolina and West Virginia,
are projecting significant increases. Texas,

2011

Delaware have stable population trends.
These states have been very active in
adopting a variety of
reforms designed to control
By
prison population growth.

2011, the imprisonment
rate of the South will
exceed that of the
Northeast by 85 percent.

Texas’s prison system will
continue to grow in part
because of simple
demographics: the state is
expected to grow by more than 2.3 million

residents over the next five years, for a total
population of over 25 million. However, its
incarceration rate is projected to grow as well,
the result primarily of low parole grant rates
and a high number of probation revocations.
Texas policy makers have begun to evaluate

Public Safety Performance Project

15

TABLE 3

Nevada 10-Year Prison Population Projections,
2007–2016
Year

Male
Population
2005
11,075
July 2006
11,662
2006
12,081
2007
12,496
2008
12,984
2009
13,727
2010
14,378
2011
15,188
2012
15,935
2013
16,727
2014
17,515
2015
18,243
2016
19,066
Numeric Change 2006 – 2016
6,985
Percent Change 2006 – 2016
57.8%

Female
Total
Population Population
1,008
12,083
1,134
12,796
1,158
13,239
1,236
13,732
1,305
14,289
1,402
15,129
1,484
15,862
1,576
16,764
1,657
17,592
1,755
18,482
1,849
19,364
1,957
20,200
2,057
21,123
899
7,884
77.6%
59.6%

Note: Numbers represent end of calendar-year figures (with the exception of the July 2006
figure, which represents the July 31, 2006, population). Year 2005 and July 2006 rows show
actual population figures.

changes on both fronts to help slow the
anticipated growth. The state parole board is
analyzing its compliance with parole
guidelines and may change its decisionmaking criteria. And in their 2007 session,
Texas lawmakers are expected to consider
major policy initiatives to reform probation,
increase intermediate sanctions and expand
treatment capacity in the correctional system.
In Louisiana, partly in response to the
devastation caused by hurricanes Katrina and
Rita, the state legislature passed several bills
designed to reduce the length of incarceration
modestly by granting more good time to
prisoners who complete treatment programs
and have satisfactory work conduct records.
The state has also enacted a law limiting to
90 days the amount of time a probation or
parole technical violator can serve in prison
for a first revocation. Louisiana also is
launching a number of reforms to expedite

TABLE 4

Arizona 10-Year Prison Population Projections,
2007–2016
Year

Male
Population
2005
30,626
July 2006
31,837
32,415
2006
2007
34,814
2008
36,958
2009
39,672
2010
42,182
2011
43,933
2012
45,834
2013
47,243
2014
48,650
2015
49,841
2016
51,008
Numeric Change 2006 – 2016 19,171
Percent Change 2006 – 2016 60.2%

Female
Total
Population Population
2,909
33,535
3,062
34,899
35,965
3,228
3,375
38,189
3,687
40,645
3,942
43,614
4,210
46,392
4,388
48,381
4,557
50,391
4,812
52,055
4,980
53,630
5,054
54,895
5,216
56,224
2,154
21,325
70.3%
61.1%

Note: Numbers represent end of calendar year figures (with the exception of the July 2006
figure, which represents the July 31, 2006, population). Year 2005 and July 2006 rows show
actual population figures.

16 Public Safety, Public Spending

parole hearings. Its prison population is
expected to rise by 4 percent over the next
five years.
Maryland and West Virginia have adopted
new parole guidelines that increase parole
grant rates for low-risk prisoners. The
Georgia parole board also relaxed its selfimposed rule that required certain offenders
to serve 90 percent of sentence, allowing
some inmates in that group to be considered
for earlier release. Maryland also enacted
new parole hearing procedures to ensure that
prisoners who are being granted parole are
actually released when they become eligible.
In addition, the Maryland parole board
adopted narrower length-of-stay ranges to
reduce how long some offenders are
incarcerated before being paroled.

Western Region
Virtually all of the Western states, with the
exception of California and Oregon, will
increase their prison populations by 20
percent or more. While Montana will have
the greatest percentage increase, Arizona,
California and Colorado will see the greatest
growth in absolute numbers in the West.

probation for repeat
Arizona’s prison population
offenders, with a few
is projected to increase by
specified exceptions. This
more than 60 percent over
collection of varied trends
and developments could
the next decade.
make Arizona a leader in
prison growth. As shown in Table 4, the 10year forecast shows the state’s prison
population increasing to 56,224.

This region’s estimated growth is in part the
result of demographics. For example, while
the U.S. population is expected to grow by
approximately 4.5 percent in the next five
years, the Western region will increase by 6.4
percent. Arizona and Nevada’s populations
are expected to increase by a dramatic 13
percent and 14 percent, respectively.
Nevada, which has a mostly discretionary
release system, has significantly increased its
10-year forecast, as the state experienced
larger-than-expected admissions from the Las
Vegas metropolitan area. Despite efforts to
counteract this surge through a higher parole
grant rate, Nevada is now poised to house
one of the fastest-growing prisoner
populations in the nation. Its prison
population is projected to increase from
about 13,200 in 2006 to over 21,000 by 2016
(see Table 3). As in other states, the female

California also is a determinate state with no
discretionary parole. However, it actually
lowered its fall 2006 population projection
from its spring 2006 estimate because of
lower-than-expected growth in new court
commitments. The long-term estimate is for
continued growth, because of both population
increases in the at-risk age cohort and the
cumulative effects of the state’s two- and
three-strikes legislation. The state also returns
an extremely high number of released inmates
to prison, especially for violations of their
terms of supervision. These ominous trends
have led Governor Arnold Schwarzenegger to
propose reducing or eliminating formal parole
supervision for low-risk offenders and
establishing a sentencing commission.

population is expected to increase at a faster
rate than the male population.
Arizona is a determinate sentencing state with
an 85-percent truth-in-sentencing law for all
prisoners, giving it little short-term flexibility
to moderate inmates’ length of sentence and
temper its growth. The recently passed
Proposition 301 negates the mandatory
probation provision in the criminal code for
first- and second-offense drug possession for
methamphetamine offenses. Further, the
Maricopa County Attorney’s Office recently
announced a new policy to disallow pleas to

Public Safety Performance Project

17

Estimating Current
and Future Prison Costs
n addition to forecasting the
national prison population over the next
five years, this report estimates the
additional fiscal costs of the expanding state
and federal prison systems. Based on
calculations described in detail below,
researchers estimate that prison operating
costs will increase by at least $2.5 billion per

I

year to as much as $5 billion per year by
2011. The price of building new prison beds
could reach $12.5 billion. In sum, the

The U.S. may need an
additional $27.5 billion over
the next five years to
accommodate projected prison
expansion and operations.

estimated 192,000 new
prisoners could cost as much
as $27.5 billion over the next
five years.

The cost of a prison system
is traditionally separated
into two broad categories:
operational and capital. Operational costs
reflect the day-to-day expenses of operating a
correctional facility, including the central
office and support services surrounding that
facility. While the largest component of
operational costs is personnel (salary and
fringe benefits), this category also reflects
items such as utilities, food, office and
medical supplies, communication services,
transportation, program services and a
variety of contracted support services such as
electrical, building maintenance and
information technology.
Capital costs are generally limited to one-time
purchases of land, construction of new

18 Public Safety, Public Spending

buildings, renovation of existing structures
and equipment. Unlike operating costs,
capital expenditures can fluctuate
dramatically from year to year depending on
decisions to build or not build new facilities.
For example, it may require at least five years
to open a new prison once a state has
decided to build one. A significant amount of
time is needed to identify an appropriate site,
develop the necessary architectural plans,
prepare the site for construction and secure
the necessary building permits from state and
local authorities. The actual construction can
often take two years with the normal delays
incident to any construction schedule. The
costs associated with a construction project
can show up during the fiscal year in which
the funds were authorized or be recorded as
costs are incurred.
Since 1984, the U.S. Department of Justice
has conducted periodic cost analyses for
each state and the District of Columbia; the
most recent study was published in 2004 and
used 2001 figures.17 At that time, it was
reported that state correctional agencies
spent $29.5 billion on correctional facilities,
with $28.4 billion spent on operating
expenses and $1.1 billion on capital costs.
(Approximately two-thirds of the operating
costs were linked to salaries and fringe
benefits.) With 1,252,743 prisoners in
custody in 2001, the average (mean) annual
cost per prisoner was $22,650.

Looking back, a comparison suggests that the
costs per prisoner stabilized between 1996
and 2001. The 1996 cost analysis found that
the average cost per inmate had steadily
increased from $16,300 in 1984 to $18,400 in
1990 and $20,100 in 1996, using constant
1996 dollars. In its more recent report, BJS
noted that when adjusted for inflation, the

1. Regional and State Variation in Costs. As noted
above, there is considerable variation across
the regions and even among the states within
a region. If one region or only certain states
from certain regions are experiencing the bulk
of the increases, the cost estimates must
account for these regional and state variations.

1996 cost per prisoner in 2001 dollars was

Just as incarceration rates themselves vary
widely by state, the 2001 BJS report found

2. Marginal Versus “Fully Loaded” Operational
Costs. An increase or decrease in a state prison
population will not yield a direct,
proportionate increase or decrease in
operating costs.18 This is because some states
whose prison populations may grow by only

considerable variation among state operating

a small amount likely will absorb that growth

costs. The most expensive prison systems

in existing facilities and with current staff.

tended to be in the Northeast region ($33,037
per prisoner per year) and the least expensive

They would experience only marginal cost
increases for medical care and daily costs

were in the Southern region ($16,479). The

such as water, food, electricity and gas.

$22,515, which was only slightly below the
actual 2001 figure of $22,650.

least expensive states were Alabama ($8,128),
Mississippi ($12,795), Missouri ($12,867),
Louisiana ($12,951) and Texas ($13,808)—the
same states that tended to have the highest
incarceration rates. The most expensive states
were Maine ($44,379), Rhode Island
($38,503), Massachusetts ($37,718),
Minnesota ($36,836), New York ($36,835),
Alaska ($36,730), and Oregon ($36,060).
While wages and benefits account for much
of the variation among the states, the other
key factor is the inmate-to-staff ratio. The
BJS report showed that Maine had the lowest
inmate per staff ratio (1.7 inmates per
employee), while Alabama had the highest
(6.8). Lower numbers of correctional officers
per inmates can reduce costs but also raise
risks to the safety of staff and inmates.

Methodological Issues
A number of methodological challenges make
estimating future prison costs problematic.
Several approaches are available, but each
must be sensitive to the following issues.

3. Tipping Point Effects. Related to the marginal
cost issue is the possibility that a very small
increase in a state’s prison population could
trigger a major increase in costs per prisoner.
This could result if in the past an agency has
been using controlled crowding measures to
control costs. However, at some “tipping” point

Prison beds each
cost between
$25,000 and
$100,000 to
build, depending
on inmate
security level.

a modest increase in the prison population may
result in a decision to construct and staff one
or more new prisons. This in turn would
significantly increase the cost-per-inmate figure.
4. Differences in Cost-containment Approaches
Adopted by the States. States use very different
approaches to reduce or control their costs
for a growing prison system. Some contract
with private prisons or local jails, while
others simply start reducing programs and
converting program space to housing units.
Because each state will approach its growth
situation differently, it would be useful to
identify those approaches and make the
appropriate adjustments.

Public Safety Performance Project

19

5. Average Costs. Related to the points above, an
estimate that uses an average cost per inmate
may well overestimate true future costs if the
state applies alternative housing methods or
changes sentencing or release laws and
practices, or if the projected population will
include a disproportionate number of lowercost inmates. There are significant differences
in the cost of housing minimum-, mediumand maximum-custody prisoners, males and
females, healthy and sick, young and old, etc.
Unless these differences are accounted for, the
estimated costs may be inaccurate.

private prison company, researchers made
every attempt to ensure those costs were
incorporated in the annual cost figure.
Each state’s cost-per-inmate rate was compared
with the BJS 2001 figure. If there was a
significant difference, researchers contacted the
state to discuss the matter and then made a
determination of the most accurate cost-peryear figure. If a state did not reply to the
request, researchers used the 2001 figures and
then adjusted them for inflation, using the
estimates provided by the federal Bureau of
Labor Statistics.19 Nonetheless, there were

Any estimate of future costs should take into

some major differences between the 2001 and

account, or at least acknowledge, that the

FY 2005-06 numbers that have not yet been

future average cost per inmate may vary
based on these and other factors, and
therefore is difficult to estimate.

accounted for.

Under ideal circumstances, another
comprehensive survey would be completed to
duplicate the detailed state-by-state census the
BJS conducted in 2001. Unfortunately, a
study of this nature was beyond the resources
of this research effort. Also, the purpose of
this report was solely to estimate the possible
cost to the state and federal government for
corrections in 2011. It’s simply not feasible to
claim that such costs certainly will occur,
because states could adopt a variety of yetunknown cost-saving strategies. However,
because it is clear that costs will increase by
some amount as a result of the large projected
growth in the prison population, it is useful to
estimate what those costs could be.

Current Operational Costs
Researchers for this report contacted each
state to obtain its most current cost per
inmate. These cost rates included
administrative support, program services
(public and contracted), and facility
maintenance. Where a state contracts with a

20 Public Safety, Public Spending

Using this approach, the current average
annual operational cost per prisoner for the
states was $23,876. For the federal Bureau of
Prisons (BOP) the cost was $23,429. Table 5
summarizes the trends in costs per state
prisoner from 1984 to FY2005-06 based on
the BJS data and this report’s survey of the
50 states, controlling for inflation. Whether
one uses 1996 or 2005 inflation-adjusted
figures, the pattern is the same: a steady
increase from 1984 to 1996 followed by
relative stability through 2001 and then a
decline in FY 2005-06.
The decline in FY2005-06 operating costs
could be the result of several factors. First, for
each of the other years, cost data were
obtained from BJS, and BJS data may be
different from the information researchers
received from the states for this report. But if
the decline is real, then it may be attributed to
other factors. For example, the largest
increases in the nation’s prison population
have occurred in the West and South, where
costs per prisoner tend to be lower, driving
down the overall average costs. There have

been efforts to make
corrections more
efficient through
procurement reforms
and privatization of a
variety of services.
And there is the
possibility that

TABLE 5

Costs Per State Prisoner, 1984–2005
Cost per
inmate

1984

1990

1996

2001

FY2005–
2006

1996 Dollars
2005 Dollars

$16,300
$20,289

$18,400
$22,903

$20,100
$25,019

$20,065
$23,941

$19,181
$23,876

crowding more
prisoners in existing facilities has reduced the
average cost per prisoner.

Estimates of Future
Operational Costs
Researchers made two estimates of future

Table A-7 in the appendix shows the 2001 and
FY 2005-06 costs per state, both with and
without adjustments for inflation. As with the
previous BJS reports, the 2005 data show
major differences among the states and the
regions. The Northeast continues to have the
highest costs per prisoner, led by Rhode Island
($44,860), Massachusetts ($43,026) and New
York ($42,202). The lowest rates are largely in
the South, led by Louisiana ($13,009),
Alabama ($13,019), South Carolina ($13,170)
and Mississippi ($13,428).
The table also shows that some states have
significantly increased or lowered their costs
per prisoner, even when adjusted for inflation.
States that have lowered their costs include
Oregon, Maine, Minnesota, Michigan, Ohio,
South Carolina and Hawaii. States where the
rates have increased significantly are
California, Alabama, Rhode Island, New York,
Alaska and Massachusetts.
In some of the states—New York,
Massachusetts, Maryland and Rhode Island—
the cost-per-inmate rate has increased but
there has been an associated decline or
leveling off in the prison population. So their
prison populations have dropped or stabilized
but the “fixed costs” of operating their prison
systems continue to increase.

operational costs. Under the first, researchers
multiplied the current (FY2005-06) costs per
prisoner obtained from each state and the BOP
by the projected 2011 prisoner populations.
Through this method, they took into account
the significant variation in costs per prisoner by
state. However, this estimate does not control
for marginal costs, tipping effects or innovative
methods for controlling costs in the face of
population increases. This is the typical
method used by the states in making fiscal
impact statements on pending legislation or
administrative reforms. There typically is no
effort to account for marginal costs or to assess
what the actual cost increases have been in the
past for each inmate increase in the prison
population. One should assume that, as a
result, such estimates by the states are too high.

Federal and state
governments are
projected to need
as much as
$15 billion
in additional
operational funds
over the next
five years.

Using this approach, the state and federal
operational budgets, which totaled just under
$35 billion in 2005, would increase by an
estimated $5 billion a year to almost $40
billion annually by 2011 in constant dollars.
In cumulative terms, this 14-percent increase
means the states and federal government
would spend a cumulative $15 billion in
operating costs over just the next five years
to accommodate the projected growth.
Under the second (and more conservative)
method for estimating future operating costs,

Public Safety Performance Project

21

researchers calculated the actual cost changes
between 2001 and FY2005-06 in relation to
the change in the prisoner population. This
method assumes no mechanical incremental
increase in the operational budgets for each
additional prisoner added to the daily
population. The BJS report and the state
survey conducted for this report show that

need for additional bed capacity. There also
are many ways by which states fund prison
construction costs that may not fully surface
during the projection period. For example, if
prison construction is being funded through a
30-year bond, the “true” cost of the new beds
will be far above the actual construction costs
because of debt service on the bond.

while the prisoner population increased from
2005, the total operating budgets for the states
only increased from $28,374,273 in 2001 to an
estimated $30,802,574 in FY2005-06.

Nonetheless, it is important to make some
estimate of the number of new beds each
state and the BOP would need to construct
based on their projections and the
construction costs associated with this bed

Assuming the BJS 2001 and the state-reported

demand. In general, the states reported

FY2005-06 cost comparisons are valid for

construction costs that ranged from $25,000

most states, the marginal annual cost for
housing each additional prisoner was $13,797
(not adjusted for inflation).20 This is 57
percent below the $23,876 figure cited earlier.
If one applies the $13,797 rate to the projected

for a minimum-security bed to more than
$100,000 for a maximum-security cell.

1,345,217 in 2001 to 1,480,223 by the end of

Because there are no “average” estimates,

192,000 increase in prisoners, the projected

researchers believe the best approximation to
use is a midpoint of $65,000 in capital costs
per bed. This figure reflects what most would

additional costs to state prison budgets by
2011 would be $2.5 billon annually in

consider the costs of a “typical” mediumsecurity bed, which covers the largest custody

constant dollars, rather than the $5 billion
cited earlier. That would accumulate to an
additional $7.5 billion in prison operations
spending over the next five years.

level of most prison systems.

Capital Costs
Estimating how much money the states and
the federal government are likely to spend on
prison construction over the next five years is
a tenuous undertaking. As described above,
some of the projected inmate growth may be
averted by changes in sentencing or release
policies. Even if growth is not averted, states
may choose to accommodate new inmates in
existing facilities by double- or triple-celling
inmates, converting program space into
dormitories or other means. On the other
hand, in some states construction costs may
be related to the need to replace aging and
dysfunctional facilities, not any projected

22 Public Safety, Public Spending

Applying the $65,000 estimate of
construction costs to the projected need for
192,000 additional prison beds, the total
construction costs would be approximately
$12.5 billion in 2006 dollars. This estimate
may be conservative, as it excludes
renovation or conversion of existing prison
bed space and assumes no financing costs.
For example, California’s Governor
Schwarzenegger recently requested a total of
$10.9 billion in mostly bond financing to
construct a combination of 78,000 jail, prison,
and juvenile correctional beds—an average of
approximately $135,000 per bed. Similarly,
Colorado has announced that it will need to
build a number of 1,000-bed prisons at a cost
of $87,000 per bed.

The Relationship
Between Incarceration
and Crime Rates
iven the projected increase
in the prison population and
associated costs, it is useful to
review the potential impact of further
incarceration on public safety. There has been
much political and academic debate on the
relationship between the use of incarceration
and crime rates.21 The common expectation is

G

that crime rates will decline as the number of
people in prison increases, and that crime will
increase if incarceration rates fall. The logic of

To demonstrate that incarceration causes
crime to go up or down, one must show (1)
that there was an increase or decrease in
incarceration before the crime rate changed
(temporal assumption); (2) that a statistical
relationship existed between crime and
imprisonment rates after the change in
incarceration rate (empirical association); and
(3) that there are no other factors that could
explain the change in the crime rate (nonspurious assumption). For example, crime rates
may have declined as a result of a decline in

this argument is that the crime rate falls when
we incapacitate people who are committing
crimes and deter those who might otherwise
become involved in criminal activities absent

the baby boomers, more effective community
policing, or many other factors.

the threat of imprisonment. A different
contention is that other social and economic
factors, such as poverty rates and education

Figure 9 shows that the increase in national
crime rates beginning in 1964 was not

the number of welfare recipients, the aging of

levels, have a greater
FIGURE 9

7000

500

6000
5000

400
4000

300

Crime Rate
3000

200
Incarceration Rate
100

2000
1000

0

0
1931
1933
1935
1937
1939
1941
1943
1945
1947
1949
1951
1953
1955
1957
1959
1961
1963
1965
1967
1969
1971
1973
1975
1977
1979
1981
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005

Incarceration Rate

600

Crime Rate

(Sentenced Inmates in State and Federal Institutions,
per 100,000 residents)

National Crime and Imprisonment Trends, 1931-2005
(Sentenced Inmates in State and Federal Institutions,
per 100,000 residents)

impact on crime than
imprisonment rates.
The general
consensus among
criminologists is that
crime rates are the
product of a complex
set of factors,
including but not
limited to
imprisonment.

Source: FBI Uniform Crime Reports, BJS

Public Safety Performance Project

23

predated by a drop in the incarceration rate.
Whatever caused the crime rate to increase
from 1965 to 1974, therefore, was not a
change in imprisonment, which remained
stable until 1975 and then started increasing
after crime rates had stabilized.22 Some
experts posit that while the increase in crime
rates that began in 1965 may have been

A recent review conducted by the Vera
Institute of Justice of all the major studies of
the relationship between incarceration and
crime shows disparate findings, with different
estimates of whether the relationship exists,
what the relationship may be, and even
whether incarceration rates at some point
may actually increase crime.25 The Vera

caused by other socio-economic and

review found that “the most sophisticated

demographic factors, the significant increase

analyses generally agree that increased

in the use of imprisonment has helped lower
crime rates. Meanwhile, after 1975, policy
makers passed many laws that increased the
probability of being sentenced to prison

incarceration rates have some effect on
reducing crime,” accounting for perhaps 25
percent of the drop in crime during the
1990s. But “analysts are nearly unanimous in
their conclusion that continued growth in
incarceration will prevent considerably fewer,
if any, crimes than past increases did and will
cost taxpayers substantially more to

rather than to jail or probation, and
dramatically increased the length of those
prison sentences.
A decade ago, James Q. Wilson suggested
that the U.S. had reached a tipping point of
“diminishing returns” from our investment in
prisons.23 According to Wilson, judges have
always been tough on violent offenders and
have incarcerated them for relatively long
sentences. However, as states expanded
incarceration, they dipped “deeper into the
bucket of persons eligible for prison,
dredging up offenders with shorter and
shorter criminal records.”24 Increasing the
proportion of convicted criminals sent to
prison, like lengthening time served beyond
some point, has produced diminishing
marginal returns in crime reductions. This
does not mean an absence of returns—just
that the benefit to public safety of each
additional prisoner consistently decreases.

24 Public Safety, Public Spending

achieve.”26

Public Safety,
Public Spending:
The Challenge Ahead
for State Policy Makers
t’s hard to place a value on the
peace of mind and sense of justice that a
victim, his or her loved ones, and society
as a whole, receive when an assailant is locked
away behind bars. This powerful and rightful
response, perhaps more than any other factor,

I

system to produce the best possible outcomes
at the best price. This means less crime and
fewer victims, lower recidivism rates, and

drives states to build more and more prisons.

Driven by hard data, the projections in this

But Americans expect the corrections system
to do more than just punish.27 They expect it
to protect public safety and reduce crime—by
deterring would-be criminals, by separating

more resources for investments like education,
health and economic development.

report clearly outline for state policy makers
the increases in their prison populations, and
worsening cost crises some are facing. States
will ignore these facts at their own peril. If
nothing changes, taxpayers will spend as much

the most dangerous people from society, and

as $27.5 billion more on prisons over the next

by helping lower-risk offenders and inmates
returning to society become productive, crime-

five years, and the jury is still out as to
whether that investment will yield
commensurate results in crime control.

free citizens.

And the effect of corrections spending on

This report does raise red flags, but it also can
be used to help diagnose problems rooted
within state corrections systems. Throughout
the report, and in accompanying state profiles,

other state priorities is particularly strong as
almost all of it comes from the states’ own
coffers, with minimal reliance on federal aid.
By contrast, the majority of health care
funding in many states comes from the federal
government, primarily through Medicaid.

Pew’s Public Safety Performance Project
showcases states that are examining the
performance of their sentencing and
corrections systems and using that data to
determine the steps necessary to improve their
outcomes.

The high cost and high stakes of corrections
rightly puts a premium on performance.
Taxpayers, victims of crime, prosecutors,
police, judges–everyone wants the corrections

The profiles highlight states that have
broadened their approaches to criminal justice,
making prisons one item on a larger menu of
options for dealing with the wide spectrum of

States pay a high price for these services.
Prisons are the fourth-largest state budget item
behind health, education and transportation.

Public Safety Performance Project

25

criminal behavior. They have employed new,
cost effective strategies for managing their
prison populations, such as establishing
sentencing guidelines, improving parole release
practices, and holding probation and parole
violators accountable with graduated
sanctions. They have developed new
programs proven effective at reducing
recidivism, such as drug courts, day reporting
centers and comprehensive re-entry programs.
And states have deployed new technologies,
such as instant-result drug tests and risk
assessments that help judges and corrections
professionals match offenders with the right
levels and types of supervision and services.
The increases in prison populations and costs
predicted in this report are worrisome, but
they are not inevitable. These projections and
the Public Safety Performance Project’s
profiles on innovative states should serve as
tools for policy makers and others, who can
use the data and lessons learned across the
country to boost the performance of the
corrections systems in their own states.

26 Public Safety, Public Spending

Appendix
TABLE A-1 Key State Data
State/Region
U.S. total*
Federala*
State*
Northeast*
Connecticutb
Maine
Massachusettsc*
New Hampshire
New Jerseyd*
New York
Pennsylvania
Rhode Islande*
Vermontb
Midwest*
Illinoisf
Indiana
Iowag*
Kansasf
Michigan
Minnesotak*
Missourif
Nebraska
North Dakota
Ohiof
South Dakota
Wisconsin
South*
Alabama
Arkansas
Delawareb
Floridaj*
Georgiah*
Kentucky
Louisiana
Maryland
Mississippi
North Carolinai*
Oklahomaf
South Carolina
Tennesseef
Texasm*
Virginia
West Virginia
West*
Alaskab
Arizona
California
Coloradof
Hawaiib
Idaho
Montanan*
Nevada
New Mexico
Oregon
Utah
Washington
Wyoming

Prison Population
2005
1,480,223
187,394
1,292,829
164,074
13,121
1,905
10,385
2,520
26,746
62,743
42,345
2,767
1,542
252,438
44,919
24,416
8,737
9,068
49,546
8,874
30,803
4,330
1,327
45,854
3,454
21,110
580,860
27,003
13,383
3,972
86,563
51,904
19,215
36,083
22,143
19,335
36,620
23,245
22,464
26,369
151,925
35,344
5,292
295,457
2,781
31,411
168,982
21,456
4,422
6,818
2,625
11,644
6,292
13,390
6,269
17,320
2,047

Incarceration
Rate 2005
500
56
435
298
373
144
239
192
313
326
340
189
247
383
351
388
294
330
489
180
529
245
208
400
443
380
539
591
479
467
499
572
459
797
394
660
360
652
525
440
691
464
291
431
414
521
466
457
340
472
373
474
323
365
252
273
400

Projection
Methodology
–
Simulation
–
–
None
Simulation
Simulation
Time Series
Flow
None
Simulation
Simulation
None
–
Simulation
Simulation
Simulation
Simulation
Simulation
Simulation
Time Series
Simulation
Simulation
Simulation
Simulation
None
–
–
Simulation
Simulation
Simulation
Simulation
Simulation
Simulation
Simulation
Simulation
Simulation
Simulation
Simulation
Simulation
Simulation
Simulation
Simulation
–
None
Time Series
Simulation
Time Series
Simulation
None
None
Simulation
Simulation
Simulation
None
Simulation
None

*

a

b

c

d

e

f
g

h

i

j

k

m
n

State population differs from
BJS report to mirror projections
populations
Federal prison population
provided by the Bureau of
Prisons.
Prisons and jails form one
integrated system. Data
include total jail and prison
population.
Prison population on December
27, 2005 from 'Quarterly Report
on the Status of Prison
Overcrowding, Fourth Quarter
2005', Massachusetts
Department of Correction,
January 2006
(http://www.mass.gov/Eeops/d
ocs/doc/research_reports/4th_0
5_overcrowding.pdf).
New Jersey 2005 prison total
taken from New Jersey
Department of Corrections
Offender Characteristics
Report, Policy Analysis &
Planning and represent
population on January 9, 2006
(http://www.nj.gov/corrections/
offender_statistics/2006/Whole
Doc_Off_Char2006.PDF).
Prison population provided by
Rhode Island Department of
Corrections.
Includes some inmates
sentenced to 1 year or less.
Iowa prison population
extrapolated using fiscal year
end counts.
Prison population provided by
Georgia Department of
Corrections.
North Carolina prison
population data obtained from
NCDOC web page database.
Prison population from 'Trends
in Florida Prison Admissions
and Populations December
2005', Florida Department of
Corrections
(www.dc.state.fl.us/pub/pop/20
05/pop.pdf)
Prison population on January 1,
2006 from 'An Outcome
Evaluation of the Challenge
Incarceration Program',
Minnesota Department of
Corrections, October 2006
(http://www.corr.state.mn.us/p
ublications/documents/CIPEval
uationReport10-06.pdf)
Prison population from Texas
Department of Criminal Justice.
Institutional prisoners only.

Source: Bureau of Justice Statistics, JFA Institute

Public Safety Performance Project

27

TABLE A-2 State, Regional and National Residential Populations,

2005-2011
State/Region

U.S. total
Northeast
Connecticut
Maine
Massachusetts
New Hampshire
New Jersey
New York
Pennsylvania
Rhode Island
Vermont
Midwest
Illinois
Indiana
Iowa
Kansas
Michigan
Minnesota
Missouri
Nebraska
North Dakota
Ohio
South Dakota
Wisconsin
South
Alabama
Arkansas
Delaware
Florida
Georgia
Kentucky
Louisiana
Maryland
Mississippi
North Carolina
Oklahoma
South Carolina
Tennessee
Texas
Virginia
West Virginia
West
Alaska
Arizona
California
Colorado
Hawaii
Idaho
Montana
Nevada
New Mexico
Oregon
Utah
Washington
Wyoming

Estimated
End of
FY 2005
295,859,883
54,641,895
3,510,297
1,321,505
6,398,743
1,309,940
8,717,925
19,254,630
12,429,616
1,076,189
623,050
65,971,974
12,763,371
6,271,973
2,966,334
2,744,687
10,120,860
5,132,799
5,800,310
1,758,787
636,677
11,464,042
775,933
5,536,201
106,954,892
4,557,808
2,779,154
843,524
17,789,864
9,072,576
4,173,405
4,523,628
5,600,388
2,921,088
8,683,242
3,547,884
4,255,083
5,962,959
22,859,968
7,567,465
1,816,856
68,291,122
663,661
5,939,292
36,132,147
4,665,177
1,275,194
1,429,096
935,670
2,414,807
1,928,384
3,641,056
2,469,585
6,287,759
509,294

Projected
End of
Year 2006
299,020,242
55,125,296
3,527,755
1,330,601
6,561,571
1,336,023
8,832,766
19,325,562
12,477,239
1,096,344
637,437
66,440,637
12,769,657
6,294,276
2,986,331
2,768,324
10,277,845
5,247,934
5,813,035
1,752,320
636,036
11,510,978
776,480
5,607,424
108,364,091
4,548,208
2,807,016
851,327
18,015,259
9,126,400
4,195,783
4,559,786
5,692,070
2,933,689
8,893,893
3,542,715
4,302,577
6,044,730
23,336,489
7,690,340
1,823,813
69,090,218
670,332
6,091,570
36,653,225
4,682,434
1,297,844
1,439,585
943,862
2,450,937
1,927,339
3,651,845
2,470,439
6,299,318
511,492

Projected
End of
Year 2007
301,714,686
55,326,451
3,542,998
1,338,420
6,588,248
1,350,148
8,888,287
19,364,721
12,509,376
1,102,511
641,745
66,721,923
12,814,117
6,323,155
2,993,970
2,779,287
10,322,902
5,296,992
5,844,610
1,757,362
636,323
11,531,425
779,477
5,642,306
109,700,000
4,562,068
2,826,758
860,938
18,359,934
9,259,442
4,216,491
4,575,884
5,752,927
2,945,086
9,022,205
3,556,830
4,344,306
6,097,782
23,711,224
7,781,912
1,826,218
69,966,312
676,799
6,243,949
37,059,690
4,725,181
1,311,156
1,461,544
951,064
2,518,059
1,943,387
3,690,289
2,505,704
6,365,366
514,128

Projected
End of
Year 2008
304,398,677
55,517,038
3,557,405
1,346,036
6,613,573
1,364,301
8,941,682
19,399,297
12,540,294
1,108,389
646,063
66,995,549
12,856,636
6,351,281
3,000,886
2,789,995
10,366,379
5,346,312
5,875,874
1,762,181
636,532
11,550,391
782,351
5,676,735
111,039,357
4,575,841
2,846,270
870,411
18,711,584
9,391,842
4,236,461
4,591,185
5,813,760
2,955,983
9,151,193
3,570,816
4,385,599
6,150,954
24,086,241
7,873,339
1,827,882
70,846,733
683,524
6,399,280
37,464,007
4,767,794
1,323,543
1,483,723
958,172
2,586,289
1,958,693
3,729,815
2,541,227
6,434,081
516,588

Source: U.S. Census Bureau
Note: End of year number estimated by calculating the mid-point population for mid-year estimates/projections.

28 Public Safety, Public Spending

Projected
End of
Year 2009
307,071,996
55,697,475
3,570,942
1,353,460
6,637,676
1,378,473
8,993,050
19,429,579
12,569,978
1,113,952
650,367
67,260,924
12,897,157
6,378,648
3,007,035
2,800,366
10,408,204
5,395,836
5,906,750
1,766,765
636,618
11,567,842
785,077
5,710,628
112,381,872
4,589,514
2,865,500
879,730
19,070,439
9,523,469
4,255,690
4,605,658
5,874,573
2,966,363
9,280,841
3,584,641
4,426,424
6,204,210
24,461,352
7,964,633
1,828,840
71,731,725
690,537
6,557,534
37,866,419
4,810,307
1,335,077
1,506,077
965,145
2,655,610
1,973,191
3,770,424
2,577,036
6,505,545
518,827

Projected
End of
Year 2010
309,740,694
55,869,287
3,583,757
1,360,672
6,660,827
1,392,698
9,042,819
19,456,238
12,598,434
1,119,205
654,640
67,517,883
12,935,988
6,405,269
3,012,367
2,810,442
10,448,077
5,445,509
5,937,243
1,771,126
636,578
11,583,777
787,646
5,743,865
113,729,952
4,603,075
2,884,503
888,882
19,437,214
9,654,210
4,274,242
4,619,402
5,935,371
2,976,225
9,411,179
3,598,415
4,466,747
6,257,637
24,837,867
8,055,867
1,829,120
72,623,572
697,745
6,719,039
38,268,509
4,852,893
1,346,002
1,528,573
971,931
2,726,022
1,986,937
3,812,060
2,613,198
6,579,809
520,859

Projected
End of
Year 2011
312,417,989
56,034,887
3,596,091
1,367,596
6,683,355
1,406,971
9,091,630
19,480,389
12,625,803
1,124,186
658,868
67,767,236
12,973,810
6,431,257
3,016,907
2,820,284
10,485,819
5,495,282
5,967,445
1,775,300
636,442
11,598,234
790,056
5,776,403
115,088,601
4,616,554
2,903,384
897,843
19,813,082
9,784,054
4,292,249
4,632,560
5,996,219
2,985,630
9,542,453
3,612,293
4,506,628
6,311,407
25,218,315
8,147,172
1,828,762
73,527,265
705,148
6,884,382
38,673,873
4,895,793
1,356,313
1,551,126
978,498
2,797,632
2,000,105
3,854,789
2,649,851
6,657,015
522,743

% Change
2006-2011
4.5%
1.7%
1.9%
2.8%
1.9%
5.3%
2.9%
0.8%
1.2%
2.5%
3.4%
2.0%
1.6%
2.2%
1.0%
1.9%
2.0%
4.7%
2.7%
1.3%
0.1%
0.8%
1.7%
3.0%
6.2%
1.5%
3.4%
5.5%
10.0%
7.2%
2.3%
1.6%
5.3%
1.8%
7.3%
2.0%
4.7%
4.4%
8.1%
5.9%
0.3%
6.4%
5.2%
13.0%
5.5%
4.6%
4.5%
7.7%
3.7%
14.1%
3.8%
5.6%
7.3%
5.7%
2.2%

TABLE A-3 State Prison Populations by Region, 2006-2011
State/Region

U.S. total
Federal1
State
Northeast
Connecticut
Maine4
Massachusetts6
New Hampshire
New Jersey
New York
Pennsylvania
Rhode Island
Vermont
Midwest
Illinois3
Indiana
Iowa3
Kansas
Michigan2
Minnesota
Missouri
Nebraska
North Dakota
Ohio3
South Dakota
Wisconsin4
South
Alabama
Arkansas
Delaware
Florida
Georgia
Kentucky
Louisiana
Maryland3
Mississippi
North Carolina3
Oklahoma3
South Carolina
Tennessee
Texas
Virginia
West Virginia
West
Alaska4
Arizona
California
Colorado2
Hawaii
Idaho5
Montana5
Nevada
New Mexico
Oregon2
Utah4
Washington2
Wyoming4

Estimated
End of
Year 2006
1,530,454
192,584
1,337,870
168,176
14,000
1,978
10,670
2,620
27,309
63,000
44,096
2,853
1,650
256,613
45,687
25,061
8,857
8,924
49,974
8,899
30,135
4,706
1,384
47,519
3,442
22,025
603,876
28,430
13,737
3,972
89,815
53,685
21,459
38,094
23,156
22,812
38,257
25,089
24,070
26,186
152,671
37,198
5,246
309,205
2,951
35,965
173,100
22,624
4,105
7,206
2,812
13,239
7,006
13,411
6,552
18,088
2,147

Projected
End of
Year 2007
1,568,822
200,696
1,368,126
170,838
14,000
2,053
10,780
2,699
28,051
63,000
45,596
2,901
1,758
261,076
46,273
25,249
9,282
9,185
50,743
9,115
29,824
4,953
1,420
49,211
3,594
22,227
615,562
28,789
14,264
3,972
92,569
55,051
21,650
38,488
23,220
23,288
38,865
26,175
24,819
26,590
154,766
37,686
5,370
320,651
3,130
38,189
177,573
23,927
4,281
7,669
3,017
13,732
7,431
13,600
6,848
19,000
2,254

Projected
End of
Year 2008
1,614,808
206,982
1,407,826
173,076
14,000
2,131
10,910
2,780
28,369
63,000
47,096
2,924
1,866
267,174
46,967
26,179
9,659
9,383
51,857
9,385
29,512
5,052
1,458
51,548
3,745
22,429
635,968
28,966
14,790
3,972
96,568
56,310
23,690
38,738
23,270
23,746
39,394
31,992
25,568
26,965
158,090
38,330
5,579
331,608
3,321
40,645
180,979
25,357
4,457
8,141
3,233
14,289
7,795
13,924
7,157
19,945
2,366

Projected
End of
Year 2009
1,654,668
212,283
1,442,385
175,349
14,000
2,212
11,040
2,863
28,704
63,000
48,596
2,960
1,974
274,877
47,708
27,058
9,898
9,505
53,044
9,609
31,216
5,182
1,499
53,603
3,904
22,651
649,085
29,298
15,246
3,972
100,482
57,463
24,525
38,951
23,320
24,005
40,059
32,633
26,317
27,273
160,555
39,304
5,682
343,075
3,523
43,614
183,955
26,894
4,633
8,625
3,464
15,129
8,044
14,294
7,480
20,937
2,483

Projected
End of
Year 2010
1,686,495
217,385
1,469,110
177,585
14,000
2,296
11,180
2,949
29,100
63,000
50,096
2,882
2,082
281,289
48,539
28,154
10,071
9,821
54,441
9,701
31,577
5,243
1,535
55,305
4,069
22,833
656,408
29,739
15,703
3,972
103,158
58,509
25,455
39,241
23,370
24,367
40,860
28,058
27,066
27,388
163,331
40,383
5,808
353,829
3,737
46,392
186,565
28,261
4,809
9,125
3,712
15,862
8,244
14,719
7,818
21,978
2,607

Projected
End of
Year 2011
1,722,477
221,882
1,500,595
180,154
14,000
2,383
11,310
3,037
29,586
63,000
51,596
3,052
2,190
287,622
49,497
28,728
10,284
10,074
55,687
10,063
31,937
5,273
1,580
57,223
4,241
23,035
669,072
30,461
16,057
3,972
106,042
59,449
26,209
39,491
23,420
24,673
41,676
28,345
27,815
27,582
166,327
41,476
6,077
363,748
3,964
48,381
188,772
29,685
4,985
9,654
3,977
16,764
8,477
15,110
8,171
23,071
2,737

% Change
2006-2011
13%
15%
12%
7%
0%
21%
6%
16%
8%
0%
17%
7%
33%
12%
8%
15%
16%
13%
11%
13%
6%
12%
14%
20%
23%
5%
11%
7%
17%
0%
18%
11%
22%
4%
1%
8%
9%
13%
16%
5%
9%
12%
16%
18%
34%
35%
9%
31%
21%
34%
41%
27%
21%
13%
25%
28%
27%

Source: JFA Institute
1
Source: Bureau of Prisons
2
State provided projections short of 2011. Similar growth rates were applied to complete.
3
State provided projections on a FY basis. December figures were extrapolated in these states.
4
Average annual change from 2001-2005 applied yearly to generate forecast.
5
Both FY adjusted and short of 2011.
6
Massachusetts represents both civil and criminal inmates.

Public Safety Performance Project

29

TABLE A-4 Prison Populations by Growth Rate, 2006-2011
State/Region

U.S. total
Federal1
State
Montana5
Arizona
Alaska4
Idaho5
Vermont
Colorado2
Washington2
Wyoming4
Nevada
Utah4
South Dakota
Kentucky
Hawaii
New Mexico
Maine4
Ohio3
Florida
Pennsylvania
Arkansas
Iowa3
New Hampshire
West Virginia
South Carolina
Indiana
North Dakota
Minnesota
Oklahoma3
Kansas
Oregon2
Nebraska
Virginia
Michigan2
Georgia
California
Texas
North Carolina3
Illinois3
New Jersey
Mississippi
Alabama
Rhode Island
Massachusetts6
Missouri
Tennessee
Wisconsin4
Louisiana
Maryland3
Connecticut
New York
Delaware

Estimated
End of
Year 2006
1,530,454
192,584
1,337,870
2,812
35,965
2,951
7,206
1,650
22,624
18,088
2,147
13,239
6,552
3,442
21,459
4,105
7,006
1,978
47,519
89,815
44,096
13,737
8,857
2,620
5,246
24,070
25,061
1,384
8,899
25,089
8,924
13,411
4,706
37,198
49,974
53,685
173,100
152,671
38,257
45,687
27,309
22,812
28,430
2,853
10,670
30,135
26,186
22,025
38,094
23,156
14,000
63,000
3,972

Projected
End of
Year 2007
1,568,822
200,696
1,368,126
3,017
38,189
3,130
7,669
1,758
23,927
19,000
2,254
13,732
6,848
3,594
21,650
4,281
7,431
2,053
49,211
92,569
45,596
14,264
9,282
2,699
5,370
24,819
25,249
1,420
9,115
26,175
9,185
13,600
4,953
37,686
50,743
55,051
177,573
154,766
38,865
46,273
28,051
23,288
28,789
2,901
10,780
29,824
26,590
22,227
38,488
23,220
14,000
63,000
3,972

Projected
End of
Year 2008
1,614,808
206,982
1,407,826
3,233
40,645
3,321
8,141
1,866
25,357
19,945
2,366
14,289
7,157
3,745
23,690
4,457
7,795
2,131
51,548
96,568
47,096
14,790
9,659
2,780
5,579
25,568
26,179
1,458
9,385
31,992
9,383
13,924
5,052
38,330
51,857
56,310
180,979
158,090
39,394
46,967
28,369
23,746
28,966
2,924
10,910
29,512
26,965
22,429
38,738
23,270
14,000
63,000
3,972

Source: JFA Institute
1
Source: U.S. Bureau of Prisons
2
State provided projections short of 2011. Similar growth rates were applied to complete.
3
State provided projections on a FY basis. December figures were extrapolated in these states.
4
Average annual change from 2001-2005 applied yearly to generate forecast
5
Both FY adjusted and short of 2011
6
Massachusetts represents both civil and criminal inmates.

30 Public Safety, Public Spending

Projected
End of
Year 2009
1,654,668
212,283
1,442,385
3,464
43,614
3,523
8,625
1,974
26,894
20,937
2,483
15,129
7,480
3,904
24,525
4,633
8,044
2,212
53,603
100,482
48,596
15,246
9,898
2,863
5,682
26,317
27,058
1,499
9,609
32,633
9,505
14,294
5,182
39,304
53,044
57,463
183,955
160,555
40,059
47,708
28,704
24,005
29,298
2,960
11,040
31,216
27,273
22,651
38,951
23,320
14,000
63,000
3,972

Projected
End of
Year 2010
1,686,495
217,385
1,469,110
3,712
46,392
3,737
9,125
2,082
28,261
21,978
2,607
15,862
7,818
4,069
25,455
4,809
8,244
2,296
55,305
103,158
50,096
15,703
10,071
2,949
5,808
27,066
28,154
1,535
9,701
28,058
9,821
14,719
5,243
40,383
54,441
58,509
186,565
163,331
40,860
48,539
29,100
24,367
29,739
2,882
11,180
31,577
27,388
22,833
39,241
23,370
14,000
63,000
3,972

Projected
End of
Year 2011
1,722,477
221,882
1,500,595
3,977
48,381
3,964
9,654
2,190
29,685
23,071
2,737
16,764
8,171
4,241
26,209
4,985
8,477
2,383
57,223
106,042
51,596
16,057
10,284
3,037
6,077
27,815
28,728
1,580
10,063
28,345
10,074
15,110
5,273
41,476
55,687
59,449
188,772
166,327
41,676
49,497
29,586
24,673
30,461
3,052
11,310
31,937
27,582
23,035
39,491
23,420
14,000
63,000
3,972

% Change
2006-2011
13%
15%
12%
41%
35%
34%
34%
33%
31%
28%
27%
27%
25%
23%
22%
21%
21%
21%
20%
18%
17%
17%
16%
16%
16%
16%
15%
14%
13%
13%
13%
13%
12%
12%
11%
11%
9%
9%
9%
8%
8%
8%
7%
7%
6%
6%
5%
5%
4%
1%
0%
0%
0%

TABLE A-5 State Incarceration Rates by Region, 2006-2011
State/Region

U.S. total
Federal
State
Northeast
Connecticut
Maine
Massachusettsc
New Hampshire
New Jersey
New York
Pennsylvania
Rhode Island
Vermont
Midwest
Illinois
Indiana
Iowa
Kansas
Michigan
Minnesota
Missouri
Nebraska
North Dakota
Ohio
South Dakota
Wisconsin
South
Alabama
Arkansas
Delaware
Florida
Georgia
Kentucky
Louisiana
Maryland
Mississippi
North Carolina
Oklahoma
South Carolina
Tennessee
Texas
Virginia
West Virginia
West
Alaska
Arizona
California
Colorado
Hawaii
Idaho
Montana
Nevada
New Mexico
Oregon
Utah
Washington
Wyoming

Projected
End of
Year 2006
511
64
447
305
397
149
163
196
309
326
353
260
259
386
358
398
297
322
486
170
518
269
218
413
443
393
557
625
489
467
499
588
511
835
407
778
430
708
559
433
654
484
288
448
440
590
472
483
316
501
298
540
364
367
265
287
420

Projected
End of
Year 2007
521
67
453
309
396
154
164
201
317
326
365
264
275
392
362
400
310
331
493
173
512
282
223
427
462
395
565
632
506
464
509
599
515
843
406
792
434
737
574
438
658
487
294
461
465
619
482
509
328
529
318
553
384
370
275
300
439

Projected
End of
Year 2008
534
68
462
313
395
159
166
206
319
325
376
265
291
400
367
414
323
338
502
177
505
287
229
447
480
398
580
635
523
461
526
608
562
847
404
806
437
899
589
442
667
493
305
474
491
651
488
537
340
557
340
567
401
377
286
313
460

Projected
End of
Year 2009
545
69
470
316
394
165
167
211
322
325
388
268
307
411
372
427
330
341
513
181
533
294
236
464
500
400
588
641
537
459
542
616
580
850
403
814
441
916
603
445
672
502
311
487
518
690
494
567
352
586
363
593
412
385
296
327
482

Projected
End of
Year 2010
553
70
474
320
394
171
169
216
325
325
399
260
322
420
378
443
336
352
525
181
537
298
241
479
520
402
591
650
552
456
551
623
601
855
402
824
446
786
617
445
678
513
318
499
547
725
498
593
363
615
387
613
421
395
308
342
505

Projected
End of
Year 2011
562
71
480
324
393
177
171
221
330
324
411
275
338
428
384
451
342
360
536
187
542
299
248
495
541
405
599
665
562
454
561
629
617
859
401
833
452
792
631
446
685
524
332
510
577
747
501
620
375
646
414
640
431
403
319
357
529

% Change
2006-2011
10%
10%
7%
6%
-1%
19%
5%
13%
7%
0%
16%
6%
31%
11%
7%
13%
15%
12%
10%
10%
5%
11%
14%
20%
22%
3%
7%
6%
15%
-3%
13%
7%
21%
3%
-1%
7%
5%
12%
13%
3%
5%
8%
16%
14%
31%
27%
6%
28%
19%
29%
39%
18%
19%
10%
20%
24%
26%

Source: JFA Institute
Note: The forecasted incarceration rates are calculated using adjusted Census projections and state prison population forecasts.

Public Safety Performance Project

31

TABLE A-6 State Incarceration Rates by Growth Rate, 2006-2011
State/Region

U.S. total
Federal
State
Montana
Alaska
Vermont
Idaho
Colorado
Arizona
Wyoming
Washington
South Dakota
Kentucky
Utah
Ohio
Maine
New Mexico
Hawaii
Nevada
Pennsylvania
West Virginia
Iowa
Arkansas
North Dakota
Indiana
New Hampshire
South Carolina
Florida
Oklahoma
Kansas
Nebraska
Minnesota
Michigan
Oregon
Virginia
Illinois
Mississippi
Georgia
New Jersey
Alabama
California
Rhode Island
North Carolina
Massachusettsc
Texas
Missouri
Tennessee
Wisconsin
Louisiana
New York
Connecticut
Maryland
Delaware

Projected
End of
Year 2006
511
64
447
298
440
259
501
483
590
420
287
443
511
265
413
149
364
316
540
353
288
297
489
218
398
196
559
499
708
322
269
170
486
367
484
358
778
588
309
625
472
260
430
163
654
518
433
393
835
326
397
407
467

Projected
End of
Year 2007
521
67
453
318
465
275
529
509
619
439
300
462
515
275
427
154
384
328
553
365
294
310
506
223
400
201
574
509
737
331
282
173
493
370
487
362
792
599
317
632
482
264
434
164
658
512
438
395
843
326
396
406
464

Projected
End of
Year 2008
534
68
462
340
491
291
557
537
651
460
313
480
562
286
447
159
401
340
567
376
305
323
523
229
414
206
589
526
899
338
287
177
502
377
493
367
806
608
319
635
488
265
437
166
667
505
442
398
847
325
395
404
461

Projected
End of
Year 2009
545
69
470
363
518
307
586
567
690
482
327
500
580
296
464
165
412
352
593
388
311
330
537
236
427
211
603
542
916
341
294
181
513
385
502
372
814
616
322
641
494
268
441
167
672
533
445
400
850
325
394
403
459

Source:JFA Institute
Note: The forecasted incarceration rates are calculated using adjusted Census projections and state prison population forecasts.

32 Public Safety, Public Spending

Projected
End of
Year 2010
553
70
474
387
547
322
615
593
725
505
342
520
601
308
479
171
421
363
613
399
318
336
552
241
443
216
617
551
786
352
298
181
525
395
513
378
824
623
325
650
498
260
446
169
678
537
445
402
855
325
394
402
456

Projected
End of
Year 2011
562
71
480
414
577
338
646
620
747
529
357
541
617
319
495
177
431
375
640
411
332
342
562
248
451
221
631
561
792
360
299
187
536
403
524
384
833
629
330
665
501
275
452
171
685
542
446
405
859
324
393
401
454

% Change
2006-2011
10%
10%
7%
39%
31%
31%
29%
28%
27%
26%
24%
22%
21%
20%
20%
19%
19%
19%
18%
16%
16%
15%
15%
14%
13%
13%
13%
13%
12%
12%
11%
10%
10%
10%
8%
7%
7%
7%
7%
6%
6%
6%
5%
5%
5%
5%
3%
3%
3%
0%
-1%
-1%
-3%

TABLE A-7 Annual Operating Costs per Inmate
State/Region

Federal
State
Northeast
Connecticut
Maine
Massachusetts
New Hampshire
New Jersey
New York
Pennsylvania
Rhode Island
Vermont
Midwest
Illinois
Indiana
Iowa
Kansas
Michigan
Minnesota
Missouri
Nebraska
North Dakota
Ohio
South Dakota
Wisconsin
South
Alabama
Arkansas
Delaware
Florida
Georgia
Kentucky
Louisiana
Maryland
Mississippi
North Carolina
Oklahoma
South Carolina
Tennessee
Texas
Virginia
West Virginia
West
Alaska
Arizona
California
Colorado
Hawaii
Idaho
Montana
Nevada
New Mexico
Oregon
Utah
Washington
Wyoming

Annual Operating Cost
Per Inmate 2001

Annual 2001 Costs
Adjusted to 2005 Dollars

Annual Operating Cost
Per Inmate 2005

$22,632
$22,650
$33,037
$26,856
$44,379
$37,718
$25,949
$27,347
$36,835
$31,900
$38,503
$25,178
$24,779
$21,844
$21,841
$22,997
$21,381
$32,525
$36,836
$12,867
$25,321
$22,425
$26,295
$13,853
$28,622
$16,479
$8,128
$15,619
$22,802
$20,190
$19,860
$17,818
$12,951
$26,398
$12,795
$26,984
$16,309
$16,762
$18,206
$13,808
$22,942
$14,817
$25,231
$36,730
$22,476
$25,053
$25,408
$21,637
$16,319
$21,898
$17,572
$28,035
$36,060
$24,574
$30,168
$28,845

24,010
23,941
35,048
28,467
47,042
39,981
27,506
28,988
39,045
33,814
40,813
26,689
26,228
23,155
23,151
24,377
22,664
34,477
39,046
13,639
26,840
23,771
27,873
14,684
30,339
18,476
8,616
16,556
24,170
21,401
21,052
18,887
13,728
27,982
13,563
28,603
17,288
17,768
19,298
14,636
24,319
15,706
26,720
38,934
23,825
26,556
26,932
22,935
17,298
23,212
18,626
29,717
38,224
26,048
31,978
30,576

23,429
23,876
35,584
29,527
35,012
43,026
28,143
28,000
42,202
31,029
44,860
28,846
23,296
21,622
21,531
23,383
21,944
28,743
29,260
14,183
25,079
25,692
23,011
14,157
28,932
17,991
13,019
17,608
24,500
22,211
17,017
18,170
13,009
30,244
13,428
24,986
16,986
13,170
20,940
14,622
21,248
16,976
29,608
42,082
19,795
34,150
26,248
18,370
16,115
25,710
17,676
26,971
24,665
23,000
29,005
33,048

Change from Inflation
Adjusted 2001 Costs to
Actual 2005 Costs
-581
-65
536
1,060
-12,030
3,045
637
-988
3,157
-2,785
4,047
2,157
-2,932
-1,533
-1,620
-994
-720
-5,734
-9,786
544
-1,761
1,921
-4,862
-527
-1,407
-485
4,403
1,052
330
810
-4,035
-717
-719
2,262
-135
-3,617
-302
-4,598
1,642
-14
-3,071
1,270
2,888
3,148
-4,030
7,594
-684
-4,565
-1,183
2,498
-950
-2,746
-13,559
-3,048
-2,973
2,472

Source: Bureau of Justice Statistics and JFA Institute
Note: Inflation assumed at 1.5% per year.

Public Safety Performance Project

33

TABLE A-8 Sources of State Prison Population Projections
State

Projections Source

Alabama

Addressing the Crisis: Charting the Course for Reform, Alabama Sentencing Commission 2006, p. 62

Alaska

N/A

Arizona

JFA

Arkansas

Arkansas Department of Correction, Sentencing Commission, and Department of Community Correction Ten
Year Adult Secure Population Projection 2006-2016, produced for the Arkansas Sentencing Commission by JFA
Associates, LLC, authors: Roger Ocker & Wendy Ware, July 2006

California

California Department of Corrections and Rehabilitation website
(http://www.cya.ca.gov/ReportsResearch/OffenderInfoServices/Projections/F06pub.pdf)

Colorado

Colorado Division of Criminal Justice December 2005 Prison Projections & Legislative Council Staff December
2005 Prison Population Projections

Connecticut

N/A

Delaware

N/A

Florida

Detailed Monthly Forecast: October 12, 2006, Florida Criminal Justice Estimating Conference
(http://edr.state.fl.us/conferences/criminaljustice/ES10122006.pdf)

Georgia

Georgia Department of Corrections

Hawaii

10-Year Corrections Master Plan Update, pg. 2-9, December 2003, Carter Goble Associates, Inc.

Idaho

Idaho Offender Population Forecast FY 2007 through 2010, August 30, 2006, State of Idaho Department of
Correction (http://www.corr.state.id.us/facts/monthly_stats/FY2007Forecast.pdf)

Illinois

Illinois Department of Corrections

Indiana

Indiana Department of Correction

Iowa

Iowa Prison Population Forecast, Iowa Department of Human Rights Division of Criminal and Juvenile Justice
Planning, November 2006

Kansas

2006 Corrections Briefing Report, Kansas Department of Corrections
(http://www.dc.state.ks.us/briefrep/2006BriefRep.pdf)

Kentucky

Kentucky Department of Corrections

Louisiana

Louisiana Department of Public Safety and Corrections

Maine

N/A

Maryland

Maryland Department of Public Safety and Correctional Services

Massachusetts

Massachusetts Department of Correction

Michigan

Report to the Legislature Pursuant to P.A. 154 of 2005 Section 401, Prison Population Projection Report January
2006, MDOC Office of Research & Planning

Minnesota

Minnesota Prison Population Projections Fiscal Year 2006 Report, p.9, Minnesota Department of Corrections
(http://www.corr.state.mn.us/publications/documents/ProjectionsReport-FY06_000.pdf)

Mississippi

Mississippi Department of Corrections Ten Year Adult Secure Population Projection: 2004-2015, produced for
the Mississippi Department of Corrections by JFA Associates, LLC, authors: Gillian Thompson & Wendy Ware,
November 2006

Missouri

Missouri Department of Corrections

Montana

Montana Department of Corrections webpage
(http://www.cor.mt.gov/resources/reports/PopulationForecast.pdf)

Nebraska

Nebraska Department of Correctional Services Ten Year Adult Secure Population Projection 2007-2017,
produced for the Nebraska Department of Correctional Services by JFA Associates, LLC, authors: Roger Ocker
& Wendy Ware, July 2006
continued next page

34 Public Safety, Public Spending

TABLE A-8 Sources of State Prison Population Projections (continued)
State

Projections Source

Nevada

Nevada Department of Corrections Ten Year Adult Secure Population Projection, produced for the Nevada
Department of Administration, Budget Division by JFA Associates, LLC, authors: Gillian Thompson & Wendy
Ware, November 2006

New Hampshire

New Hampshire Department of Corrections

New Jersey

New Jersey Department of Corrections, Office of Policy & Planning

New Mexico

New Mexico Corrections Department Ten Year Adult Secure Population Projection, Revision C, FY 2007-2016,
produced under contract for the New Mexico Corrections Department by JFA Associates, LLC, authors: Roger
Ocker & Wendy Ware, June 2006

New York

N/A

North Carolina

North Carolina Sentencing & Policy Advisory Commission FY 2005-2015 Population Projections, prepared in
conjunction with Department of Correction's Office of Research and Planning, January 2006

North Dakota

Study of the Facilities and Operations of the North Dakota Department of Corrections, Vol. II: Population
Projections and Capacity Needs Analysis, June 15, 2002, Security Response Technologies, Inc.

Ohio

Ohio Prison Population Projections and Intake Estimates, Bureau of Research, Office of Policy and Offender
Reentry, Ohio Department of Rehabilitation and Correction, author: Brian Martin, February 2006

Oklahoma

Oklahoma Criminal Justice Resource Center, April 2006 Oklahoma Prison Population Projection
(http://www.ocjrc.net/pubFiles/InmatePopulation/OklahomaPrisonPopulationProjection_2006.pdf)

Oregon

Oregon Corrections Population Forecast October 2006, Vol. XII No. 2, Office of Economic Analysis, Department
of Administrative Services

Pennsylvania
Rhode Island

Pennsylvania Population Projection Committee Report Update, September 2005
Rhode Island Department of Corrections Adult Prison Population Forecast FY 2006, produced for the Rhode
Island Department of Corrections by JFA Associates, LLC, authors: Roger Ocker & Wendy Ware.

South Carolina

South Carolina Department of Corrections

South Dakota

South Dakota Department of Corrections

Tennessee

The Tennessee Department of Correction Fiscal Year 2005-2006 Annual Report Coordinated and Published by
the Policy, Planning, and Research Division, authors: Linda M. Nutt, Cynthia Taylor, Sara Conte
(http://www.state.tn.us/correction/pdf/0506anlrpt.pdf)

Texas

Adult Incarceration Projected Population, Texas Legislative Budget Board, January 2007

Utah

N/A

Vermont

Vermont Department of Corrections

Virginia

Virginia Secretary of Public Safety & Policy Advisory Commission

Washington

Washington State Department of Corrections

West Virginia

West Virginia Correctional Population Forecast: 2004-2014: A Study of the State's Prison Population, December
2006, Criminal Justice Statistical Analysis Center, authors: Theresa K. Lester & Stephen M. Haas

Wisconsin

N/A

Wyoming

N/A

Source: JFA Institute

Public Safety Performance Project

35

TABLE A-9 Sources of State Inmate Costs
State

Cost Source

Alabama

http://www.doc.state.al.us/docs/AnnualRpts/2005AnnualReport.pdf

Alaska

http://www.gov.state.ak.us/omb/06_OMB/budget/DOC/dept20.pdf

Arizona

http://www.azcorrections.gov/adc/reports/CAG/CAGJun05.pdf

Arkansas

Arkansas Department of Corrections

California

http://www.cya.ca.gov/divisionsboards/aoap/factfiguresarchive/factsfigures3rdq2005.html

Colorado

Colorado Department of Corrections Statistical Report Fiscal Year 2004, Office of Planning & Analysis,
Kristi L. Rosten

Connecticut

http://www.ct.gov/doc/cwp/view.asp?a=1492&q=265472

Delaware

http://www.state.de.us/correct/pdfs/BudgetInformationFY05.pdf & Delaware DOC

Florida

Florida Department of Corrections; ttp://www.dc.state.fl.us/upu/annual/0405/budget.html

Georgia

http://www.dcor.state.ga.us/pdf/FY05AnnualReportPart2.pdf

Hawaii

http://www.hawaii.gov/psd/documents/reports/PSD_AnnualReport_2004.pdf;
http://www.ojp.usdoj.gov/bjs/pub/pdf/p04.pdf

Idaho

http://www.corr.state.id.us/facts/fact_sheets/QuickFactsJuly2006.pdf

Illinois

http://www.idoc.state.il.us/subsections/reports/department_data/Department%20Data%202005.pdf

Indiana

http://www.in.gov/indcorrection/facts.htm

Iowa

http://www.doc.state.ia.us/Documents/QuickFacts.pdf

Kansas

2007 Corrections Briefing Report', Kansas Department of Corrections
(http://www.dc.state.ks.us/briefrep/2006BriefRep.pdf)

Kentucky

Kentucky Department of Corrections

Louisiana

Louisiana Department of Public Safety and Corrections

Maine

Regional Average

Maryland

Maryland Department of Public Safety and Correctional Services

Massachusetts

http://www.mass.gov

Michigan

http://www.michigan.gov/documents/2004_Annual_Report_147719_7.pdf

Minnesota

http://www.doc.state.mn.us/aboutdoc/stats/documents/NotableStatistics7-06_000.pdf

Mississippi

http://www.mdoc.state.ms.us/Research%20and%20Statistics/OffenderCostPerday/
Cost%20Per%20Inmate%20Day%20FY%202005.pdf

Missouri

http://www.doc.mo.gov/pdf/AR%202005.pdf

Montana

http://www.cor.mt.gov/Facts/FAQ.asp; http://www.cor.mt.gov/Resources/Reports/PopulationForecast.pdf

Nebraska

http://www.corrections.state.ne.us/administration/statistics/reportdocs/05annualreport.pdf

Nevada

http://www.doc.nv.gov/stats/annual/fy2005.pdf

New Hampshire

http://www.nh.gov/nhdoc/divisions/publicinformation/documents/annual2005.pdf

New Jersey

http://www.state.nj.us/corrections/freqntlyasked.html

New Mexico

Regional Average

New York

New York Department of Correctional Services

North Carolina

http://www.doc.state.nc.us/dop/cost/

North Dakota

http://www.state.nd.us/docr/docr/BiennialReport03-05.pdf

Ohio

http://www.drc.state.oh.us/web/Reports/FactSheet/July%202005.pdf

Oklahoma

http://www.doc.state.ok.us/newsroom/facts/06-01%20Facts%20at%20a%20Glance.pdf

Oregon

http://www.oregon.gov/DOC/PUBAFF/docs/pdf/quickfacts.pdf

Pennsylvania

http://www.cor.state.pa.us/stats/lib/stats/2006budgetpresentation.pdf

Rhode Island

http://www.doc.ri.gov/administration/Cost%20Per%20Offender%20-%202006.pdf
continued next page

36 Public Safety, Public Spending

TABLE A-9 Sources of State Inmate Costs (continued)
State

Cost Source

South Carolina

http://www.doc.sc.gov/FAQs/FAQs.html

South Dakota

http://www.state.sd.us/corrections/miscellaneous_stats.htm

Tennessee

http://www.state.tn.us/correction/faq.html

Texas

Texas Department of Criminal Justice

Utah

http://corrections.utah.gov/faq.html

Vermont

http://www.doc.state.vt.us/pageflip/pageflip.pl/picture?book=FF2006&seqno=1196

Virginia

http://www.vadoc.state.va.us/about/facts/financial/2005/05percapita.pdf

Washington

http://www.doc.wa.gov/BudgetAndResearch/ResearchData/DOCStatisticalBrochureNov06P282.pdf

West Virginia

West Virginia Division of Corrections

Wisconsin

http://www.wi-doc.com/index_adult.htm

Wyoming

Regional Average

Source: JFA Institute

Public Safety Performance Project

37

Endnotes

1

2

U.S. Department of Justice, Office of Justice Programs. Prisoners in 2005, Bureau
of Justice Statistics Bulletin, by Paige M. Harrison and Allen J. Beck (Washington,
D.C.: November 2006), NCJ 215092 and U.S. Department of Justice, Office of
Justice Programs. Prison and Jail Inmates at Midyear 2005, Bureau of Justice Statistics
Bulletin, by Paige M. Harrison and Allen J. Beck (Washington, D.C.: May
2006), NCJ 213133.
U.S. Department of Justice, Office of Justice Programs. Prevalence of Imprisonment
in the U.S. Population, 1974-2001, by Thomas P. Bonczar (Washington, D.C.:
August 2003), NCJ 197976.

3

King’s College, London, International Centre for Prison Studies. Prison Brief—
Highest to Lowest Rates. Online. Available:
http://www.kcl.ac.uk/depsta/rel/icps/world brief. Accessed: 2006.

4

A common error is to lump together the terms “jail” and “prison.” In general,
jails are operated by county government and are reserved for persons who are
awaiting trial or who have been sentenced to a term of less than one year.
Prisons are operated by state agencies and typically house persons with felony
sentences of one year or more.

5

U.S. Department of Justice, Office of Justice Programs. Prisoners in 2005, Bureau
of Justice Statistics Bulletin, by Paige M. Harrison and Allen J. Beck (Washington,
D.C.: November 2006), NCJ 215092

6

Ibid.

7

http://www.ojp.usdoj.gov/bjs/glance/tables/exptyptab.htm

8

The formula actually requires one to specify the LOS in years to produce an
annualized ADP. So if the LOS is not in years but days, one must divide the
sum by 365 days to produce an LOS in years.

9

The amount of discretion correctional authorities have to release prisoners
varies according to each state’s sentencing structure. The majority of states
have indeterminate sentencing systems, which offer the greatest amount of discretion since they allow parole boards to release inmates once they have served
their minimum sentence. States with determinate sentencing structures provide
some level of discretion to release prisoners based on good-time and special
program credits.

10

Austin, James, John Clark, Patricia Hardyman, and D. Alan Henry. 1999. “The
Impact of ‘Three Strikes and You’re Out’,” Punishment and Society, Vol 1(2):
131-162.

11

For six of the eight states, the average annual percent change was calculated
from 2001 to 2005 and applied each year to future years. The other two states,
Delaware and New York, have been showing declines over this time frame.
Researchers contacted both states to determine if the downward trends might
continue. Based on these contacts and a review of recent prison population
trend data, this report assumes no growth over the next five years. It should be
emphasized that for these eight non-reporting states the estimates used in this
report are not official forecasts.

12

13

U.S. Department of Justice, Office of Justice Programs. Prisoners in 2005, Bureau
of Justice Statistics Bulletin, by Paige M. Harrison and Allen J. Beck (Washington,
D.C.: November 2006), NCJ 215092.
U.S. Department of Justice, Bureau of Justice Statistics. Recidivism of Prisoners
Released in 1994, (Washington, D.C.: Bureau of Justice Statistics, June 2002),
NCJ 193427.

38 Public Safety, Public Spending

14

http://www.ojp.usdoj.gov/bjs/reentry/characteristics.htm

15

Data from Georgia Department of Corrections, Georgia Board of Pardons and
Paroles, and Georgia Criminal Justice Coordinating Council.

16

For a detailed presentation of the New York experience, see Michael Jacobson,
Downsizing Prisons: How to Reduce Crime and End Mass Incarceration (New York: New
York University Press, 2005), Chapter 4.

17

Stephen, James J. (June 2004). State Prison Expenditures 2001. Washington, D.C.:
U.S. Department of Justice, Bureau of Justice Statistics.

18

The one exception to this rule is where a state is contracting out to private
prisons or local jails and where the contract allows for the cost to the private or
local facility to vary directly to the number of inmates it is housing. For
example, in Louisiana, local jails bill the state for each state inmate it houses on
each day at a cost of $22 per day.

19

U.S. Department of Labor, Bureau of Labor Statistics. http://data.bls.gov/cgibin/cpicalc.pl.

20

There were some states where the differences between 2001 and FY2005-06
were so large that researchers decided to exclude them based on face validity
concerns. Also excluded were states that showed significant declines in their
costs between 2001 and FY 2005-06.

21

For a review of the more recent studies on the link between incarceration rates
and crime rates, see the following reports: Michael Jacobson, Downsizing Prisons:
How to Reduce Crime and End Mass Incarceration (New York: New York University
Press, 2005), Chapter 4. Michael Lynch, “Beating a Dead Horse: Is There Any
Basic Empirical Evidence of the Deterrent Effect of Imprisonment,” Crime, Law
and Social Change vol. 31, no. 4 (1999) p. 361. Tomislav V. Kocandizic and
Lynne M. Vieraitis, “The Effect of County-Level Prison Population Growth on
Crime Rates,” Criminology & Public Policy, vol, 5, no. 2 (May 2006), p. 234.
Raymond V. Leidka, et al, “The Crime-Control Effect of Incarceration: Does
Scale Matter?” Criminology & Public Policy, vol. 5, no. 2 (May 2006), pp. 245-276.
William Spelman, “The Limited Importance of Prison Expansion,” in The
Crime Drop in America, ed. Alfred Blumstein, Revised Edition, (New York:
Cambridge University Press, 2006), pp. 97-129.

22

For all historical incarceration rates presented in the figures in this section, the
source is the Sourcebook of Criminal Justice Statistics Online,
http://www.albany.edu/sourcebook/wk1/t6292004.wk1. The U.S. crime rate and
state crime rates presented in this section are those compiled and reported by
the National Disaster Center, at
http://www.disastercenter.com/crime/uscrime.htm

23

James Q. Wilson, “Crime and Public Policy” in James Q. Wilson and Joan
Petersilia, Crime ICS Press, Oakland, California 1995, p. 489-507.

24

Ibid, p. 501.

25

Don Stemen, “Reconsidering Incarceration: New Directions for Reducing
Crime,” Vera Institute of Justice, New York, New York, January 2007.

26

Ibid, p. i.

27

See, for instance, “The National Center for State Courts Sentencing Attitudes
Survey,” July 2006. Findings from a poll of 1,502 randomly selected adults
included that 76 percent of Americans “would rather see their tax dollars
support programs that try to prevent crime by helping offenders find jobs and
get treatment than be used to build more prisons.”

1025 F Street NW, 9th Floor
Washington, DC 20004-1409
phone 202.552.2000 | fax 202.552.2299