one in

31

THE LONG REACH OF

AMERICAN CORRECTIONS

MA R C H 2 0 0 9

ABOUT THIS REPORT
The Pew Charitable Trusts applies the power of knowledge to solve today’s most challenging problems. The
Pew Center on the States identifies and advances effective policy approaches to critical issues facing states.
In 2006, the Pew Center on the States launched the Public Safety Performance Project (PSPP) to help
states advance fiscally sound, data-driven policies and practices in sentencing and corrections that
protect public safety, hold offenders accountable, and control corrections costs.
PEW CENTER ON THE STATES
Susan Urahn, managing director
Adam Gelb, director, PSPP
Richard Jerome, manager, PSPP
Jake Horowitz, senior associate, PSPP
Joe Gavrilovich, administrative assistant, PSPP
Data Consultants (see Methodology Notes): American Probation and Parole Association (Carl Wicklund,
executive director; Matthew DeMichele, senior research associate) and JFA Associates (Jim Austin,
president; Keith Hardison, consultant)
ACKNOWLEDGMENTS
We would like to thank consultant Jenifer Warren, principal report author; William Saylor and Anthony
Iwaszko of the Federal Bureau of Prisons; the Federal Probation and Pretrial Services System and
Administrative Office of U.S. Courts; Dennis Schrantz of the Michigan Department of Corrections and
Jeriel Heard and Andre Simenauer of the Wayne County Sheriff’s office; the survey respondents from
prison, probation and parole agencies in all 50 states; our survey partners including the Crime and
Justice Institute, the Council of State Governments Justice Center, the National Center for State Courts
and the Vera Institute of Justice; Eric Cadora and Charles Schwartz of the Justice Mapping Center Inc.;
Mike Heffner, Lucy Pope and Denise Kooper of 202design; David Draine, associate, research; and the
communications team for communications support and creative direction of this report (Janet Lane,
project director, communications; Andrew McDonald, senior officer; Carla Uriona, manager; Jessica
Riordan, senior associate; Alyson Freedman, administrative associate).
For additional information on the Pew Center on the States and its Public Safety Performance Project,
please visit www.pewcenteronthestates.org.
For fact sheets about the size and cost of the correctional populations in each of the states, see the
online version of this report at www.pewcenteronthestates.org/publicsafety.
Suggested citation: Pew Center on the States, One in 31: The Long Reach of American Corrections
(Washington, DC: The Pew Charitable Trusts, March 2009).
©2009 The Pew Charitable Trusts
901 E Street NW, 10th Floor
Washington, DC 20004

2005 Market Street, Suite 1700
Philadelphia, PA 19103

Table of Contents
Executive Summary....................................................................................................................................................1
America’s Surging Correctional Population...............................................................................................................................4
Community Corrections: Population Growth Exceeds Prisons .............................................................................4
Sidebar: The Role of Probation and Parole...............................................................................................................5
Who’s Under Supervision?..........................................................................................................................................................6
Case Study: Geographic Concentration in Michigan..................................................................................................6
Sidebar: An Even Wider Net?.........................................................................................................................................10
Prisons: 32% of the Growth, 88% of the Cost ........................................................................................................................11
Community Corrections Far Less Expensive ................................................................................................................12
Probation and Parole Stretched Thin ...............................................................................................................................12
Sidebar: Policy Choices Put Bluegrass State in a Bind ....................................................................................15
More Prison Spending Brings Lower Public Safety Returns ..........................................................................................17
The Myth of the “Average Prisoner”....................................................................................................................................17
The Declining Impact of Incarceration on Crime.......................................................................................................18
Three Strikes for Incarceration...............................................................................................................................................19
Prisons Reconsidered..................................................................................................................................................................20
Community Corrections: A Strategy for Safety and Savings.........................................................................................22
Sort Offenders by Risk to Public Safety ............................................................................................................................23
Base Intervention Programs on Science..........................................................................................................................23
Harness Technology ...................................................................................................................................................................24
Impose Swift and Certain Sanctions for Violations....................................................................................................26
Create Incentives for Success ................................................................................................................................................26
Sidebar: “HOPE” for Improvement in Hawaii.........................................................................................................27
Measure Progress..........................................................................................................................................................................29
A Rare Moment in Time .....................................................................................................................................................................31
Figures 7 Million and Counting...................................................................................................................................................4
Less Than One-Third Behind Bars..............................................................................................................................5
1 in 31: Doing the Math .................................................................................................................................................5
Who’s Under Correctional Control? .........................................................................................................................7
Wide Variance in Correctional Control...................................................................................................................7
Correctional Control in Michigan .............................................................................................................................8
Prisons Dominate Spending .....................................................................................................................................11
Explosive Growth in Prison Spending .................................................................................................................12
State Daily Costs Per Offender .................................................................................................................................13
Daily Cost Details: Colorado ......................................................................................................................................13
NY Cuts Crime and Incarceration...........................................................................................................................21
Kansas Population Trend .............................................................................................................................................29
Endnotes.....................................................................................................................................................................................................33
Methodology Notes .............................................................................................................................................................................36
Jurisdictional Notes ..............................................................................................................................................................................38
Appendix Table A-1: National Correctional Populations, 1982-2007........................................................................40
Table A-2: State and National Correctional Spending................................................................................41
Table A-3: State Correctional Populations, Year End 2007 .......................................................................42
Table A-4: Adult Incarceration Rates (Jail and Prison) ................................................................................43
Table A-5: Adult Community Supervision Rates (Probation and Parole) .........................................44
Table A-6: Adult Correctional Control Rates (Jail, Prison, Probation and Parole).........................45

One in 31: The Long Reach of American Corrections

Executive Summary
States face the worst fiscal crisis in a generation.
Shrinking budgets are forcing governors
and legislators to examine all areas of public
spending for possible savings, even those that
have been off limits.
Corrections is a prime target for cuts. Last year it
was the fastest expanding major segment of state
budgets, and over the past two decades, its
growth as a share of state expenditures has been
second only to Medicaid. State corrections costs
now top $50 billion annually and consume one in
every 15 discretionary dollars.
The remarkable rise in corrections spending wasn’t
fate or even the natural consequence of spikes in
crime. It was the result of state policy choices that
sent more people to prison and kept them there

“The fact that so many Americans,
including hundreds of thousands
who are a threat to no one, are
incarcerated means that something
is wrong with our criminal justice
system and the way we deal with
both dangerous criminals and
those whose behavior we simply
don’t like.”
David Keene, President, American Conservative Union
Personal communication
February 5, 2009

longer. The sentencing and release laws passed in
the 1980s and 1990s put so many more people
behind bars that last year the incarcerated

population reached 2.3 million and, for the first
time, one in 100 adults was in prison or jail.
The escalation of the prison population has been
astonishing, but it hasn’t been the largest area of
growth in the criminal justice system. That would
be probation and parole—the sentenced
offenders who are not behind bars.
With far less notice, the number of people on
probation or parole has skyrocketed to more than
5 million, up from 1.6 million just 25 years ago.
This means that 1 in 45 adults in the United
States is now under criminal justice supervision in
the community, and that combined with those in
prison and jail, a stunning 1 in every 31 adults, or
3.2 percent, is under some form of correctional
control. The rates are drastically elevated for men
(1 in 18) and blacks (1 in 11) and are even higher
in some high-crime inner-city neighborhoods.

Community Corrections:
Big Promise, Little Support
Probation and parole, the dominant
community corrections programs, have had
larger population growth than prisons but far
smaller budget growth. Looking at a handful
of states that were able to provide long-term
spending figures, eight times as many new
dollars went to prisons as went to probation
and parole. And while fewer than one out of
three offenders is behind bars, almost nine out
of 10 corrections dollars are spent on prisons.
Incarceration understandably costs more. Prisons
must house, feed and provide medical care to the

One in 31: The Long Reach of American Corrections

1

EXECUTIVE SUMMARY

“…Focus must be placed on
locking up the most dangerous
people instead of diverting time
and money to incarcerate the
wrong people.”
U.S. Sen. Jim Webb (D-Va.)
http://webb.senate.gov

most dangerous offenders. But the price gap is
nevertheless staggering: on average, the daily
cost of supervising a probationer in fiscal 2008
was $3.42; the average daily cost of a prison
inmate, $78.95, is more than 20 times as high.
Community corrections agencies have been
further strained by a host of added
responsibilities. On top of crushing caseloads,
new laws, such as statutes mandating lifetime
supervision of some offenders, and expanded
roles like sophisticated cyber-crime detection,
have created new obligations for departments
already stretched thin. The expanded duties are
a partial recognition of the role that community
corrections plays in protecting public safety, but
they have come without sufficient investments in
staff, equipment and other support.
Despite the meager funding and ballooning
workload, there have been significant advances
in community supervision. Sophisticated risk
assessment tools now help determine which
offenders require the most supervision and
what sort of monitoring and services they need.
Global positioning systems, rapid-result drug
tests and other technology can track offenders’
whereabouts and behavior. Offender supervision,
treatment and re-entry programs are
incorporating solid research on how to cut
recidivism. Performance incentives are

2

increasingly available for both offenders and
agencies, and managers are doing a better job
tracking new arrests, collection of victim
restitution and other key outcomes.
Taken together and implemented well, these
approaches can produce double-digit reductions
in recidivism and save states money along the
way. If policy makers want these results, though,
they will have to invest in the overburdened
system of community corrections.

Opportunity in Crisis
After an extraordinary, quarter-century expansion
of American prisons, one unmistakable policy
truth has emerged: We cannot build our way to
public safety.
Serious, chronic and violent offenders belong
behind bars, for a long time, and the expense of
locking them up is justified many times over. But
for hundreds of thousands of lower-level inmates,
incarceration costs taxpayers far more than it
saves in prevented crime. And new national and
state research shows that we are well past the
point of diminishing returns, where more
imprisonment will prevent less and less crime.
With the costs of imprisonment rising and the
benefits falling, our ability to keep communities
safe depends more than ever upon our ability to
better manage the 5 million offenders on
probation and parole.
The current budget crisis presents states with an
important, perhaps unprecedented opportunity
to do so. Rather than trying to weather the
economic storm with short-term cost saving
measures, policy leaders should see this as a
chance to retool their sentencing and corrections

Public Safety Performance Project | Pew Center on the States

EXECUTIVE SUMMARY

“We have to fundamentally rethink
prisons.”
Newt Gingrich
American Enterprise Institute forum
March 27, 2008

systems. If we had stronger community
corrections, we wouldn’t need to lock up so
many people at such a great cost. By redirecting
a portion of the dollars currently spent on
imprisoning the lowest-risk inmates, we could
significantly increase the intensity and quality of
supervision and services directed at the same
type of offenders in the community.
This is not a call to slash funding for prison
operations. Though efficiencies undoubtedly
can be wrung from prisons1—like any other
government agency—they must be safe and
secure and adequately staffed and equipped.
Savings significant enough to truly bolster
community supervision can come only from
reductions of the inmate population large
enough to warrant the closure of entire
cellblocks or institutions.

This reinvestment strategy wouldn’t put a stop
to all new crimes. But it would significantly cut
recidivism—both for offenders coming out of
prison and those diverted from prison in the
first place—and do it at a fraction of the cost of
a prison bed.
A number of states are seizing the moment,
rethinking old policies and reallocating some
correctional dollars. Texas and Kansas are off to a
strong start, providing community corrections
agencies with more resources and authority, but
also giving them incentives and holding them
accountable for results. States such as Arizona,
Michigan, Pennsylvania and Vermont are now
following with innovations of their own.
The bipartisan leadership in these states and the
advances in correctional practice deserve more
than a passing glance, especially in a fiscal crisis
that demands more than ever that taxpayer
dollars be wisely spent. Armed with the
conviction that our current crime and punishment
policies are not delivering satisfactory results,
policy makers have a chance to both balance
their budgets and deliver better public safety.

One in 31: The Long Reach of American Corrections

3

America’s Surging
Correctional Population
Last year, the Pew Center on the States
reported that for the first time, more than
1 in every 100 adults in the United States was
confined behind bars. That sobering news came
as a shock for many Americans and sparked
discussions about incarceration and its fiscal
and social costs in the media, at universities,
in statehouses, and around dinner tables.
For policy makers, the 1 in 100 milestone was a
reminder that state policy choices have driven
the rise in prison populations. The explosive
prison growth of the past 30 years didn’t happen
by accident, and it wasn’t driven primarily by
crime rates or broad social and economic forces
beyond the reach of state government. It was the
direct result of sentencing, release and other
correctional policies that determine who goes
to prison and how long they stay.2

supervision grew by a staggering 3,535,660 to
a total of 5.1 million. Though the percentage
increase of those under community supervision
was not as large as the growth of those in
custody, the absolute number of probationers
and parolees grew by more than twice as much.
In 1982, 72 percent of offenders were managed in
the community, with about 28 percent behind
bars. At the end of 2007, the most recent year for
which figures are available, 31 percent were locked
up and 69 percent were on probation or parole. So
over the past quarter century, the nation has put
1.6 million more people behind bars, yet prisons
and jails still hold roughly the same proportion of
offenders and criminal suspects. (See Appendix A-1
for complete figures.)

7 MILLION AND COUNTING
Community Corrections:
Population Growth
Exceeds Prisons
While the 1 in 100 statistic has seeped into the
national consciousness, many Americans remain
unaware that a much larger number of offenders
are not behind bars at all but receive their
punishment in the community. The raw numbers
illustrate this story in dramatic fashion. During
the past quarter-century, the number of prison
and jail inmates has grown by 274 percent. The
additional 1,680,661 inmates brought the total
population in custody to 2.3 million. During the
same period, the number under community

4

Led by probation, the correctional population has
tripled in 25 years.
8,000,000
7,000,000
6,000,000
PROBATION
4,293,163

5,000,000
4,000,000

PAROLE
824,365
PRISON
1,512,576

3,000,000
2,000,000
1,000,000

JAIL
780,581

0

82

87

92

97

02

07

SOURCE: Bureau of Justice Statistics Correctional Surveys available at
http://www.ojp.usdoj.gov/bjs/glance/tables/corr2tab.htm.
NOTE: Due to offenders with dual status, the sum of these four correctional categories
slightly overstates the total correctional population.

Public Safety Performance Project | Pew Center on the States

D

A M E R I C A’ S S U R G I N G C O R R E C T I O N A L P O P U L AT I O N

LESS THAN ONE-THIRD BEHIND BARS
Despite a 274
percent increase
in incarceration,
the vast majority
of offenders under
correctional
control remain
in 0the community.

SOURCE: Calculation based on
data from the Bureau of Justice
Statistics Correctional Surveys
available at
http://www.ojp.usdoj.gov/bjs/
glance/tables/corr2tab.htm.

PROBATION
and PAROLE
1982

72%

1987

75%

1992

73%

1997

70%

2002

70%

2007

69%

PRISON
and JAIL

[For details of each state’s correctional population
and expenditures, see the state fact sheets included
in the online version of this report.]

28%

Who’s Under Supervision?
25%
27%
30%

30%
31%

Adding up all probationers and parolees,
prisoners and jail inmates, you’ll find America
now has more than 7.3 million adults under some
form of correctional control. That whopping
figure is more than the populations of Chicago,
Philadelphia, San Diego and Dallas put together,
and larger than the populations of 38 states and
the District of Columbia.3 During Ronald Reagan’s
first term as president, 1 in every 77 adults was
under the control of the correctional system in
the United States. Now, 25 years later, it is 1 in 31,
or 3.2 percent of all adults.4

Looking at the numbers through the lenses of
race and gender reveals stark differences. Black
adults are four times as likely as whites and nearly
2.5 times as likely as Hispanics to be under
correctional control. One in 11 black adults—
9.2 percent—was under correctional supervision
at year end 2007. And although the number of
female offenders continues to grow, men of all
races are under correctional control at a rate five
times that of women.5
Geography adds another revealing facet to the
picture. In Georgia, 1 in 13 adults is under the
correctional system’s authority, but in New
Hampshire, the figure is just 1 in 88. While
Southern states maintain the nation’s highest
incarceration rates, the addition of probationers
and parolees to the mixture casts a spotlight on
states that supervise massive numbers of
people in the community. The 10 states with
the largest number of people in the corrections
system include those with reputations for
toughness, like Texas and Louisiana, but also
Idaho, Ohio and Massachusetts. Similarly,

1 IN 31: DOING THE MATH
ONE
PRISON POPULATION
JAIL POPULATION
PROBATION POPULATION
PAROLE POPULATION
CORRECTIONAL POPULATION

TWO
1,512,576
780,581
4,293,163
824,365
7,328,200

THREE
229,030,637

ADULT POPULATION

ONE IN EVERY 31 U.S. ADULTS
IS UNDER CORRECTIONAL CONTROL

CORRECTIONAL POPULATION

7,328,200

SOURCE: Calculation based on data from the Bureau of Justice Statistics “Prisoners at Yearend 2007” as well as “Probation and Parole at Yearend 2007” available
at http://www.ojp.usdoj.gov/bjs and the U.S. Census State Population Estimates.
NOTE: Probation, parole, jail and prison populations do not sum to total due to offenders with dual status. Prison and jail populations differ from past reports due to method of counting
prisoners held in jail.

One in 31: The Long Reach of American Corrections

5

A M E R I C A’ S S U R G I N G C O R R E C T I O N A L P O P U L AT I O N

THE ROLE OF PROBATION AND PAROLE
The public’s perception of corrections most commonly centers on prisons and jails—buildings with bars,
locked cells and uniformed guards. But far more offenders pay for their crimes through community
sanctions, including drug courts, home detention and electronic monitoring, residential facilities with
treatment, and day reporting centers.
The centerpiece of community corrections is probation and parole. Offenders placed on probation—
derived from the Latin word probatum, for “the act of proving”—are typically lower level offenders who
are allowed to remain in the community provided they exhibit good behavior and meet other conditions
while supervised by a probation officer. With origins in this country dating to the mid-19th century,
probation is ordered by a judge and served under threat of more serious sanctions. If a probationer
violates conditions governing his or her community release, a judge may impose additional rules or
require a term in custody.
Parolees, by contrast, are offenders who have spent time in prison and are released to complete the
remainder of their sentence under supervision in the community. Intended in part to smooth a prisoner’s
transition back to society, parole, which became prevalent at the turn of the 19th century, is sometimes
ordered by appointed parole boards, which also craft conditions governing a parolee’s release. More
often, the date of parole release reflects an offender’s original sentence, perhaps shortened by credits
for a clean disciplinary record or completion of in-prison programs. In the community, parolees are
supervised by a parole officer and subject to similar rules as those on probation. If parolees violate the
rules of their release, they too face penalties including re-incarceration.
The most recent numbers, from year end 2007, show that nearly 4.3 million adults are on probation
in the United States, with almost half of them having been convicted of felonies. Property and drug
charges accounted for more than 50 percent of probationers, followed by driving while intoxicated
and other criminal traffic violations (18 percent), violent crimes (17 percent) and other offenses (13
percent).6 Parolees, meanwhile, are substantially fewer, with about 824,000 reported at the end of
2007. They also are more likely than probationers to have been convicted of a violent crime (26
percent) or a drug offense (37 percent).7

the 10 states with the lowest correctional
control rates include rural and northeastern
states like Iowa and Maine, but also states with
large urban populations, such as New York,
and with long sentences for violent offenders
like Virginia.

6

Case Study: Geographic
Concentration in Michigan
But even these statewide averages hide extreme
geographic concentrations. Michigan, with a
correctional control rate of 1 in 27—not far from
the national average—provides a useful illustration.
Mapping just the 122,165 jail and prison inmates,

Public Safety Performance Project | Pew Center on the States

A M E R I C A’ S S U R G I N G C O R R E C T I O N A L P O P U L AT I O N

WHO’S UNDER CORRECTIONAL CONTROL?
Correctional control rates vary drastically across demographic lines.

TOTAL 1 IN 31

WHITE 1 IN 45

WOMEN 1 IN 89

HISPANIC 1 IN 27

MEN 1 IN 18

BLACK 1 IN 11

SOURCE: Calculation for year end 2007 based on data from the Bureau of Justice Statistics “Prisons and Jails at Midyear” series as well as “Probation and Parole at Yearend” series available
at http://www.ojp.usdoj.gov/bjs and the U.S. Census State Population Estimates.

WIDE VARIANCE IN CORRECTIONAL CONTROL
Share of adults under correctional control, year end 2007.

WA
1 in 30
OR
1 in 33

NV
1 in 48
CA
1 in 36

NH
1 in 88
MT
1 in 44
ID
1 in 18

ND
1 in 63

WY
1 in 38

WI
1 in 39

SD
1 in 40
NE
1 in 44

UT
1 in 64

AZ
1 in 33

CO
1 in 29

NM
1 in 35

KS
1 in 53

OK
1 in 42

TX
1 in 22

AK
1 in 36

VT
1 in 46

MN
1 in 26

IA
1 in 54

MO
1 in 36

NY
1 in 53
MI
1 in 27

OH
IN
IL
1 in 26 1 in 25
1 in 38
WV
1 in 68

VA
1 in 46

KY 1 in 35

NC
1 in 38

TN 1 in 40
AR
1 in 29
LA
1 in 26

PA
1 in 28

MA
1 in 24
RI
1 in 26
CT
1 in 33
NJ
1 in 35
DE
1 in 26
DC
MD
1 in 27 1 in 21

SC
1 in 38
MS
1 in 38

AL
1 in 32

GA
1 in 13

FL
1 in 31
HI
1 in 32

ME
1 in 81

Highest fifth
Second highest
Middle fifth
Second lowest
Lowest fifth

SOURCE: Calculation includes offenders in state and federal jail, prison and community supervision and is based on data from the U.S. Census State Population Estimates, the Bureau of
Justice Statistics Correctional Surveys available at http://www.ojp.usdoj.gov/bjs/glance/tables/corr2tab.htm, the U.S. Bureau of Prisons, the Administrative Office of U.S. Courts and the
Pew Public Safety Performance Project.

One in 31: The Long Reach of American Corrections

7

A M E R I C A’ S S U R G I N G C O R R E C T I O N A L P O P U L AT I O N

CORRECTIONAL CONTROL IN MICHIGAN
Nationally, 1 in 31 adults is under some form of correctional control. But they are not evenly spread
across or within states. To illustrate the concentration of correctional populations and costs, the Pew
Center on the States partnered with the Justice Mapping Center to map the home address of all
adults in jail and prison or under parole or felony probation supervision in one state. Michigan was
selected because its correctional indicators are near the national averages and data were available.
Excluding misdemeanor and other lower-level probationers, these mapped populations account for
122,165 of the state’s 278,808 adults who are behind bars or supervised in the community.

8

Public Safety Performance Project | Pew Center on the States

A M E R I C A’ S S U R G I N G C O R R E C T I O N A L P O P U L AT I O N

GEOGRAPHIES OF CORRECTIONAL CONTROL

Wayne County
Detroit
East Side
Brewer Park

ANNUAL
CORRECTIONS
COST
(IN MILLIONS)
$546.9
$393.0

COMMUNITY
SUPERVISED
POPULATION
22,624
13,390

TOTAL
CORRECTIONAL
POPULATION
37,267
24,272

PERCENT
OF
ADULTS
2.6
4.1

CORRECTIONAL
CONTROL RATE
1 in 38
1 in 25

1,269

1,646

2,915

4.5

1 in 22

$46.1

78

104

182

6.3

1 in 16

$2.9

INCARCERATED
POPULATION
14,643
10,882

NOTES: Based on data from the Michigan Department of Corrections, the Wayne County Sheriff’s Office, the Federal Bureau of Prisons, the Administrative Office of U.S. Courts, the U.S.
Census and the Pew Center on the States. The difference between Michigan’s true correctional control rate of 1 in 27 and the rate of 1 in 61 reflected in these maps is due to the exclusion of
approximately 157,000 misdemeanant probationer residences.

One in 31: The Long Reach of American Corrections

9

A M E R I C A’ S S U R G I N G C O R R E C T I O N A L P O P U L AT I O N

AN EVEN WIDER NET?
Probation, parole, jail and prison population data, as used in this report, are provided voluntarily by state
agencies and account for the vast majority of supervised offenders in the United States. There are,
however, many others involved in our fragmented correctional system for whom there are no reliable
state-by-state data.
In particular, recent research points to a hidden population supervised pre-trial, by drug courts or
alternative sentencing units, and other specialized programs. The National Criminal Justice Treatment
Practices Survey of 20058 sampled agencies representing 72 counties and estimated that nationwide as
many as one million offenders were under correctional supervision but not included in conventional
probation and parole counts.
Additionally, there are well over 100,000 offenders in prisons of the U.S. territories, Immigration and
Customs Enforcement facilities and juvenile residential placements.9 These offenders also are not
typically included in incarceration rate calculations.

parolees and felony probationers—excluding
nearly 157,000 non-felon probationers—reduces
the figures considerably, to only 1 in 61 adults
across the Wolverine State.10 In Wayne County
(the state’s most populous county), however, the
figure is 1 in 38 and in Detroit it is 1 in 25. Further
investigation reveals that in the East Side, 1 in 22
adults are under correctional authority. And in
the blocks around Brewer Park, the number is a
startling 1 in 16 and would be even higher with
a count of non-felon probationers.
These disturbing patterns are repeated in most
major metropolitan areas of the United States.
While people must be held to account for their
crimes, a number of researchers have shown the
dire consequences of such a high geographic

10

concentration of people in the corrections
system. Because offenders from highincarceration areas also fulfill the roles of family
members, neighbors, economic consumers and
producers, removing them from the community
can result in a wide range of costly side effects,
from family disruption and neighborhood
destabilization to depressed wages and even
increased AIDS infection rates.11
Policy makers must consider these implications,
along with the budgetary cost of the corrections
system itself, in crafting fiscally responsible crime
control strategies. This includes deployment of
probation and parole officers to the hardest-hit
neighborhoods, where they, like community
police officers, can be more effective.

Public Safety Performance Project | Pew Center on the States

Prisons: 32% of the Growth,
88% of the Cost
State correctional budgets spiked along with
their offender populations in recent years. In
FY2008, states are estimated to have spent more
than $47 billion of general funds on corrections,
a 20-year jump of 303 percent.12 Add in another
$4 billion in state special funds and bonds, and
about $900 million from the federal government,
and total state spending for corrections is
estimated to top $52 billion.13 (See Appendix A-2
for additional cost information.)
This growth rate outpaced budget increases for
nearly all other essential government services
tracked over the same period, from elementary
and secondary education (205 percent) to
transportation (82 percent), higher education (125
percent) and public assistance (9 percent). Only
Medicaid spending grew faster than spending on
corrections, increasing 492 percent in the last two
decades.14 As a share of total state general fund
spending, corrections has grown from 5.2 percent
in 1988 to 6.9 percent today.15 For all levels of
government, total corrections spending has
reached an estimated $68 billion, an increase of
330 percent since 1986.16
To get a better picture of how states have
invested their corrections dollars, the Pew Center
on the States and several partners recently
completed the first national survey of corrections
spending by function in the past seven years.17
Thirty-four states, accounting for 58 percent of
total state correctional populations,18 made
complete data available while the others did not.

The largest beneficiaries of those mushrooming
budgets, by far, have been prisons. For the most
part, probation, parole and other programs that
manage offenders outside prisons and jails have
scrambled for funds needed to keep pace with
expanding caseloads of offenders with
increasingly complex and demanding problems.
In FY2008, these 34 states spent $18.65 billion
on prisons but just $2.53 billion on probation
and parole, a ratio of more than seven to one.
Viewed over time, the spending gap looks just
as substantial. For eight geographically diverse
states19 that were able to provide data for the
past 25 years, 88 percent of the increase in
corrections spending was directed toward
prisons, which now consume nearly nine out
of every ten state corrections dollars.

PRISONS DOMINATE SPENDING
Across 34 states, nearly 9 of 10 correctional dollars went
to prisons in FY2008.

12%

AMOUNT TO
PROBATION
AND PAROLE
$2.53 billion
AMOUNT TO
PRISONS
$18.65 billion

88%
TOTAL CORRECTIONS
SPENDING
$21.18 billion
SOURCES: Spending figures were collected from AR, AL, AK, CO, DE, GA, ID, IA, KY, LA, ME, MD,
MI, MN, MS, MO, MT, NC, ND, NE, NH, NM, NY, OK, OR, PA, RI, SC, SD, TN, TX, VT, VA and WY.

One in 31: The Long Reach of American Corrections

11

PRISONS: 32% OF THE GROW TH, 88% OF THE COST

EXPLOSIVE GROWTH IN PRISON SPENDING
Across 8 states, 88 percent of additional corrections
spending since FY1983 has gone to prisons.

$5,672.74
million

by guards than it is to monitor him or her
in the community. Prisons and jails also are
buildings that need to be cooled, heated
and lighted, equipped with security, and
continually cleaned and maintained.
The difference in cost between institutional and
community corrections, however, is huge. While
there is wide variance among states, in 2008
prisons cost our 34 surveyed states an average
of about $79 per inmate per day—or almost
$29,000 per year. In contrast, the average daily
costs for managing an offender in the community
in these states ranged from $3.42 per day for
probationers to $7.47 per day for parolees or
about $1,250 to $2,750 a year, respectively.

$788.80
million

$930.06
million

$136.48
million
1983

2008

PROBATION
AND PAROLE

1983

2008

PRISON

SOURCES: Only eight states could provide 25-year spending histories (AL, GA, LA, MO, MT,
NY, OR and WY).

So while the incarcerated population has
added only half as many offenders as
community supervision over the last quarter
century, if the survey states are representative
of the nation, prisons have received almost
90 percent of the new funding.

Community Corrections
Far Less Expensive
Society should expect to pay more to punish its
most serious and violent offenders by removing
them from our communities. Not surprisingly,
then, it’s more expensive to house and feed an
offender in a facility watched around-the-clock

12

Another reason community corrections costs
less is that offenders are often required to pay a
substantial share of the tab. In Colorado, for
instance, probationers under the authority of the
state pay a $50 per month supervision fee, and
some drug and sex offenders pay a surcharge on
top of that. Ninety-four percent of the funding for
treatment services provided by the court is covered
by these probationer fees as are 5.5 percent of all
staffing costs.20 Additionally, offenders ineligible for
probation but diverted from prison to residential
community corrections beds paid $11.75 million
toward their own housing, meals and treatment,
nearly $900,000 in child support, and over $1.2
million in state taxes and over $3 million in federal
taxes in fiscal year 2007.21 See chart, next page, for
Colorado spending details.

Probation and Parole
Stretched Thin
Managing offenders in the community, when
done well, produces appreciable costs savings and
public safety outcomes. However, the funding

Public Safety Performance Project | Pew Center on the States

PRISONS: 32% OF THE GROW TH, 88% OF THE COST

STATE DAILY COSTS PER OFFENDER
1 day in prison costs more than 10 days on parole or 22 days on probation.
LOW
AVERAGE
HIGH

$1.38
$3.42
$7.89

LOW
AVERAGE
HIGH

$1.22
$4.00

LOW
AVERAGE
HIGH

PROBATION AGENCIES

PROBATION AND PAROLE AGENCIES
$9.76

$3.51
$7.47

PAROLE AGENCIES
$13.28

PRISON SYSTEMS

$35.69

LOW
AVERAGE
HIGH

$78.95
$130.16

SOURCES: Spending figures were collected from AR, AL, AK, CO, DE, GA, ID, IA, KY, LA, ME, MD, MI, MN, MS, MO, MT, NC, ND, NE, NH, NM, NY, OK, OR, PA, RI, SC, SD, TN, TX, VT, VA and WY.
NOTE: Caution should be used in making interstate comparisons since a wide variety of factors beyond agency performance or efficiency can account for daily cost differences. Some states
have separate probation and parole agencies while others have combined them.

struggle has stretched probation and parole
staffing woefully thin, leading to inflated caseloads
with a high ratio of offenders to officers. The
average probation officer now has about 100
offenders on his or her caseload; parole tends to
be slightly lower, at about 60 offenders per
officer.22 Agencies often put higher risk and high
stakes cases on priority caseloads.

That’s a logical compromise but one which leaves
many other offenders without supervision or services
adequate to prevent a relapse into destructive
behavior, including committing new crimes.
The low priority of probation and parole has
forced officers in some regions to do without
important and sometimes basic tools of the

DAILY COST DETAILS: COLORADO
Per offender per day costs vary substantially both between and within supervision categories.

PROBATION

$3.07 Regular
$8.97 Intensive
$61.86 Minimum Security
$74.80 Medium Security
$91.90 Maximum Security

PRISON

PAROLE

$9.32 Regular
$22.79 Intensive

SOURCE: Colorado Commission on Criminal and Juvenile Justice, “2008 Annual Report” and Colorado Department of Corrections. All figures fiscal year 2007.

One in 31: The Long Reach of American Corrections

13

PRISONS: 32% OF THE GROW TH, 88% OF THE COST
trade. In Cook County (Chicago), Illinois, for
example, probation officers don’t have personal
computers to help them perform everyday case
work, exchange information with other agencies
or investigate criminal histories.23 By contrast,
many parole officers in California have handheld
PDAs, a convenience that allows them to access
files and accomplish other tasks from the field.24
Beyond often lacking the basic resources and
technology, community corrections agencies
have been assigned a widening array of
responsibilities, often without the funds to carry
them out. Over the past decade or so, for
example, an explosion of well-intentioned laws
governing the supervision of sex offenders has
created a multitude of new duties, or expanded
existing ones, for probation and parole
departments. The new responsibilities include
conducting DNA testing, mental health screening
and risk assessments for sex offenders, as well as
continual registration checks of their address and
work status. These are vital public safety tasks,
but they are too seldom backed up with the
resources to conduct them and are further
watering down supervision across the board.
To make matters worse, the economic situation this
year is forcing states to consider cutting back on
what limited resources community corrections
agencies do have. In Sacramento County, California,
76 probation officer positions—9 percent of the

14

“Currently, we spend next to nothing
on community-based corrections.
We get what we pay for.”
Prof. John J. DiIulio
The Wall Street Journal
March 12, 1999

total force—are on the chopping block. In
Washington State, half of all taxpayer-supported
drug treatment beds are slated for elimination. And
in Florida, the two streams converge as lawmakers
consider cutting both $3 million in drug treatment
slots and 66 probation officer positions. Across the
nation, tight budgets are jeopardizing the basics of
community supervision: caseloads, services and
day-to-day resources.
Without adequate resources and authority,
community supervision agencies are hardpressed to fulfill their traditional case
management workloads, let alone adequately
handle their new responsibilities. The huge
increase in corrections spending has favored
prisons over probation and parole by nearly nine
to one. Supervising 1 in 45 adults and holding
them accountable to victims will require that
funding gap to narrow. The sheer scale of
community supervision obliges policy makers to
recognize the major role of probation and parole
agencies in helping states protect public safety
and control public spending.

Public Safety Performance Project | Pew Center on the States

PRISONS: 32% OF THE GROW TH, 88% OF THE COST

C ASE STUDY: POLIC Y CHOICES PUT BLUEGRASS STATE IN A BIND
A variety of factors influence the size and cost of the corrections system. Crime and a rising resident
population of a state certainly play a role, but studies show correctional policies and practices that
determine who is sent to prison and how long they stay—from sentencing laws such as “three strikes”
to the extensive use of prison to punish supervision rule violators—are more decisive factors.25
Kentucky’s experience provides a vivid, if not uncommon, illustration of the consequences of the
corrections policy choices that state leaders make.

The Cause
Kentucky’s prison population has surged over the past eight years, jumping by 50 percent to more than
22,000 inmates.26 With the fastest growing prison system in the country, the Bluegrass State could have
an incarcerated population of 31,000 within the
coming decade.27 The growth has been propelled
largely by a series of tough-on-crime measures that
began in 1974 with passage of the first version of
the state’s “persistent felon law.” The original law
required three strikes to trigger “persistent felon”
status, but that was cut to two in 1976, and several
other measures in the 1990’s elevated
misdemeanors to felonies, reclassified offenses as
higher level felonies and enhanced the penalties

“Nobody’s willing to change the
laws because everybody wants
zero tolerance on everything.
But there’s something [that’s]
going to have to give.”
Terry Carl, Kenton County (KY) jailer
Lexington Herald Leader
January 13, 2008

for a variety of crimes.28 Another law extends
sentences by not counting time served on parole toward completion of the sentence if a parolee was
revoked back to prison. These and other measures have created a desperate need for more prison space
and sent the state’s corrections budget rocketing upward.29
If Kentucky’s rapid prison growth had led to dramatic decreases in crime, it could be justified on public
safety grounds. But it hasn’t. From 1987 to 2007, the state’s imprisonment rate grew nearly 250 percent,
from well below the national average (147 per 100,000 residents compared with a U.S. rate of 228) to
slightly above the national average (512 versus 506 per 100,000).30 During that time, Kentucky’s violent
crime rate fell 13 percent, but the national violent crime rate fell 23 percent. Kentucky’s property crime
rate fell 14 percent, also badly trailing the national property crime rate, which fell 34 percent.31

The Consequences
With state prisons jam-packed, Kentucky has been forced to pay county jails to house overflow offenders.
County officials appreciate the income—and, in fact, have come to rely heavily upon it in recent years.
But now their lock-ups, too, have become severely overcrowded, some so much that inmates are
This Case Study continues on page 16

One in 31: The Long Reach of American Corrections

15

PRISONS: 32% OF THE GROW TH, 88% OF THE COST

Continued from page 15

sleeping on the floor.32 Moreover, the county jails are designed for short, pre-trial stays, and are less wellequipped than prisons to provide the drug treatment, mental health care and other services designed to
reduce the risk of recidivism. Kentucky’s jails now hold about 20,000 people—putting them 10 percent
beyond their intended capacity—but a recent count showed only 400 slots in treatment programs.33 The
prison system, meanwhile, holds about 15,000 inmates34 and has about 1,000 treatment beds.35 One jail
treatment supervisor, in Kenton County, summed up the dire need for program slots in vivid terms: “I
have a waiting list as long as the New York phone book.”36
A good share of responsibility for this predicament lies with the state’s chronic underinvestment in
community corrections. Between fiscal years 2003 and 2008, the state increased annual corrections
spending by $100 million.37 Ninety percent of this additional spending, however, was channeled to
prisons, with only 1 in 10 new dollars going to support probation and parole. The evidence in Kentucky
suggests that saving pennies on community supervision costs taxpayers dollars in prison expenses: 3,101
of the 17,700 Kentuckians on parole in 2007—1 in every 6—were returned to prison for committing a new
crime or breaking the technical rules of release.38 On the street, these parolees cost taxpayers under $10
per day, but behind bars they cost over $50 per day. A stronger community supervision system could have
prevented many of these parolees from returning to prison and at a fraction of the cost.
These troubles would be difficult enough to manage in good economic times. But Kentucky, like other
states, faces a fiscal crunch. The state is looking at a projected revenue shortfall of $1.3 billion over the
next 18 months and is bracing for cuts to police and other important government services.39 In late 2008,
Kentucky’s counties gave the Commonwealth something else to worry about, suing to force the state to
cover the cost of housing inmates in county jails before their trial and sentencing.40 Now, on top of
everything else, the projected tab for the corrections budget this fiscal year is $521 million, nearly five
times the amount spent 20 years ago.41

16

Public Safety Performance Project | Pew Center on the States

More Prison Spending Brings
Lower Public Safety Returns
States are facing their worst fiscal crisis in years.
With revenues down and public needs rising,
policy makers are confronting wrenching budget
decisions. Reluctant to raise taxes—at any time
but especially when their constituents are
financially stressed—lawmakers across the
country are locked in bitter battles over where to
cut spending and by how much.

“It’s not about being tough on crime
or soft on crime. We are facing a
huge economic challenge here.
Are we doing the right thing?”
William Wren, NH Commissioner of Corrections
Concord Monitor
January 25, 2009

All told, analysts forecast a $312 billion hole in
state budgets over the next two years.42 This fiscal
year alone, 42 states and the District of Columbia
are grappling with a combined $46 billion
deficit.43 In response, officials are scavenging for
dollars wherever they can, cutting back on
everything from government basics—such as
how often the grass gets mowed outside the
state capitol—to education and services for
persons with disabilities and mental illness.

Against this grim backdrop, prison spending is
deservedly receiving new scrutiny. The central
questions: What has our massive investment
bought us? How can we curb and reallocate
corrections spending in ways that protect public
safety and produce better results for taxpayers,
offenders and society at large?

Tennessee Gov. Phil Bredesen has told state
department chiefs to prepare for a budget deficit
that could hit $1 billion and has warned that cuts to
higher education and health care are on the table.
Virginia, meanwhile, already has reduced spending
by $2 billion and is preparing for another round of
cuts, including a possible $400 million reduction in
Medicaid. Out West in Washington, state lawmakers
are scrambling to plug a $500 million hole in the
current budget and cope with a projected deficit of
more than $5 billion for the next two-year budget
cycle. “It will be heartbreaking. We don’t have any
money. We simply don’t have any money,” said Rep.
Maralyn Chase, a Democrat from Shoreline.44

The Myth of the
“Average Prisoner”

These are challenging questions, but they do
have answers.

There is little debate that imprisonment has
protected communities from many of the most
violent and menacing criminals, and that some
offenders should be locked up purely for the sake
of punishment. But in casting a wider net for
criminals, prisons have snagged many smaller fish.
A growing body of research is showing the limits
of incarceration as a sanction for these lower level
and less frequent lawbreakers, both in terms of its
cost-effectiveness and its impact on crime.

One in 31: The Long Reach of American Corrections

17

M O R E P R I S O N S P E N D I N G B R I N G S LO W E R P U B L I C S A F E T Y R E T U R N S
To understand this, it’s important to remember
that all offenders aren’t the same. They present
different threats to public safety, and thus their
incarceration pays vastly different dividends.
Criminologists long ago demonstrated that
imprisonment of the average offender serves to
avert many crimes that would otherwise carry
considerable public cost. But more recent and
refined research reveals that measuring the
impacts of the average prisoner hides as much as
it reveals because offenders—and their crimerelated impacts—vary so dramatically.

“What we’ve done with the laws
we passed over the last 20 years
is thrown our net out there too
widely and picked up too many
little fish. We filled our prisons
with non-violent, first-time
offenders, and with no noticeable
increase in public safety.”
State Sen. Stewart Greenleaf (R-PA)
NCSL Roundtable
September 26, 2008

One such study, published by the Manhattan
Institute, ranked all male inmates entering the
Arizona prison system in terms of the harm they
created in the year before incarceration. Those at
the 80th percentile of harm, the research showed,
created almost $220,000 in social costs. But those
at the 50th percentile—the median—inflicted
$25,500 in social costs, while those at the 20th
percentile were responsible for just $3,950 in
social costs.45 The authors concluded that for
Arizona and the two other states they analyzed
(New York and New Mexico), incarceration for half
of all entering prisoners would cost taxpayers

18

more than it was worth, in terms of crimes
avoided.46

The Declining Impact of
Incarceration on Crime
Aside from evidence that incarceration doesn’t
“pay” for all current prisoners, there are separate
reasons to question its value as a broadly applied
correctional tool for the future. One is what
economists call the law of diminishing returns.
Here, diminishing returns means that the larger
the group of offenders scooped up by prisons,
the lower the payoff for states in terms of crime
reduction.47 It certainly pays to remove the most
prolific offenders from the streets. But once they
are locked up, more incarceration grabs the
second and third and tenth tier offenders who
are less likely to commit as many crimes. So
gradually, the crime-prevention payoff declines.
Diminishing doesn’t necessarily mean no returns
at all, but it does mean that each additional
prison cell provides less public safety benefit.
Many states appear to have reached a “tipping
point” where additional incarceration will have
little if any effect on crime. Washington State, for
example, found that the number of crimes
committed by its average prisoner dropped from
62 in 1980 to 37 in 1990 and 18 by 2001.48 Back in
1980, state researchers found, each prison bed
represented a positive benefit-to-cost ratio. But
during the 1990s and the first part of this decade,
prison expansion captured less and less harmful
offenders, leading to a dilution of impact.49 Put
simply, after 20 years, locking up more drug and
property offenders in Washington began to cost
more than it was worth.

Public Safety Performance Project | Pew Center on the States

M O R E P R I S O N S P E N D I N G B R I N G S LO W E R P U B L I C S A F E T Y R E T U R N S
Researchers have conducted similar analyses in
other states, such as North Carolina50 and
Oregon,51 and reached conclusions of similarly
diminishing returns. Indeed, in Washington, from
1980 to 2001, the benefit-to-cost ratio for drug
offenders plummeted from $9.22 to $0.37. That is,
for every one dollar invested in new prison beds
for drug offenders, state taxpayers get only 37
cents in averted crime.52 An updated analysis
from 2006 found that incarceration of offenders
convicted of violent offenses remained a positive
net benefit, while property and drug offenders
offered negative returns.53
More recently, scholars have explored the tipping
point concept in incarceration on a 50-state basis.
A 2006 study suggests that, after exceeding a
threshold in the range of 325 to 430 inmates per

“Bed for bed, prisons become less
effective as they fill up.”
William Spelman, University of Texas
“The Limited Importance of Prison Expansion”
2000

100,000 residents, incarceration fails to reduce
crime—and may even increase it.54 Imprisonment
was more useful, the authors argue, when state
incarceration rates hovered around 111 per
100,000 in the 1970s, or around 207 per 100,000
in the 1980s, than when they accelerated to 397
per 100,000 in the 1990s.55 Today, of course, the
national rate of imprisonment is significantly
higher—506 per 100,000.56

Three Strikes for Incarceration
The potency of incarceration is further
diminished by three other forces, researchers
have found. The first, sometimes referred to as the
“replacement effect,” applies largely to crimes that
occur as part of a market, such as fencing stolen
property or, most notably, drug transactions.
Once incarcerated, drug dealers tend to be
quickly replaced by new dealers and, as during
the crack epidemic, the new recruits can be
younger and more prone to violence than their
predecessors.57 Thus while drug dealers no doubt
deserve punishment, most leading researchers,
and many law enforcement officials, now agree
that incarcerating the foot soldiers in drug gangs,
not to mention drug users, has a negligible
impact on crime.58 Moreover, by creating job
openings in drug-dealing organizations, it draws
more people into criminal lifestyles and may in
certain cases exacerbate crime.59
Secondly, statistics have long shown that crime is
an occupation of the young, so imprisoning
offenders beyond the age at which they would
have likely given up their criminal ways brings little
benefit—but big expenses. As James Q. Wilson, the
noted political scientist at Pepperdine University,
has written, “Some thugs may mug and murder
until the day they die, but they are the exception.
Age slows us all down, mugger and victim alike.”60
The graying of the nation’s prisons suggests that
policy makers have not paid much heed to this
well-established criminological fact. Rather, many
have embraced longer sentences through
broadly defined “three strikes” statutes and parole
policies that are hiking up the average age of
inmates—and the costs to states of treating their
more serious medical conditions.

One in 31: The Long Reach of American Corrections

19

M O R E P R I S O N S P E N D I N G B R I N G S LO W E R P U B L I C S A F E T Y R E T U R N S
Consider California. Between 1980 and 2007, the
average age of California inmates increased from
27 to 37. In 2008, the Golden State’s prisons
held more than 22,000 offenders over age 50,
representing about 13 percent of all adult
inmates. That’s more than twice the proportion
of over-50 convicts in California prisons just a
decade earlier.61
Finally, research has shed important new light
on the impact of one of incarceration’s most
fundamental selling points: deterrence. Today,
it is widely agreed that deterrence is more a
function of a sanction’s certainty and swiftness
than its severity. This means that the 36th month
of a 3-year prison term costs taxpayers just
as much as the first month, but its value as a
deterrent is far less. Unfortunately, the corrections
system has put more and more of its eggs into
the severity basket, spending billions to extend
prison terms—for property and drug offenders as
well as violent and sex offenders—but doing
little to raise the chances that criminals and
supervision violators are caught and brought
quickly to justice.

“Very large increases in the prison
population can produce only modest
reductions in crime rates.”
James Q. Wilson
The Public Interest
Fall 1994

1990s.63 More recently, however, another expert,
Bruce Western of Harvard University, estimated
that only 10 percent of that decade’s decline in
crime was due to increased incarceration.64
The disparities underscore the fact that estimates
by researchers in this field vary wildly and are
highly sensitive to statistical techniques and
modeling assumptions.65 Whatever level of crime
reduction was achieved is worth applauding.
What cannot be overlooked, however, is that
even the statistical models most generous to
prisons find that most of the crime drop was
attributable to forces other than incarceration.
These include a strengthening economy, aging
drug epidemics and changes in law enforcement,
including the expansion of police forces and the
adoption of new policing strategies.66

Prisons Reconsidered
This is not to say that prisons haven’t reduced
crime. One widely respected expert, William
Spelman of the University of Texas, concluded
that prison growth over the 25-year period
ending in 1997 reduced the violent crime rate by
roughly 35 percent.62 Imprisonment, he asserted
further, was responsible for about one-quarter of
the significant drop in violent crime during the

20

The questionable value of prisons as a deterrent,
combined with other factors that reduce
incarceration’s effectiveness and overshadowed
by a constellation of factors that drive and
suppress crime rates, add up to an often
overlooked truth: states can carefully reduce
incarceration and still protect—and even
improve—public safety.

Public Safety Performance Project | Pew Center on the States

M O R E P R I S O N S P E N D I N G B R I N G S LO W E R P U B L I C S A F E T Y R E T U R N S

NY CUTS CRIME AND INCARCERATION
Between 1997 and 2007, New York State bucked the
national trend in prison growth while leading all states
in the violent crime decline.

+14%

VIOLENT CRIME
RATE

U.S.
New York

U.S.

New York

–15%
–24%
INCARCERATION
RATE

–40%

SOURCE: The Bureau of Justice Statistics “Prisons at Yearend” series and the Federal Bureau
of Investigation Uniform Crime Reports.

New York has demonstrated this point in
dramatic terms. Between 1997 and 2007, New
York experienced both the greatest decrease in
violent crime and, simultaneously, the greatest
decrease in prison population and incarceration
rate of any state in the country. During that
decade, the national prison population grew by
more than 350,000 inmates, a 28 percent jump
that corresponded to a 14 percent increase in
the national incarceration rate. Over the same
time period, New York’s prison population
declined by almost 6,500 inmates, a 9.4 percent
dip that amounted to a 15 percent drop in the
incarceration rate.67 To the surprise of many at
the time, New York’s violent crime rate fell a
remarkable 40 percent during the decade, while
the national violent crime rate dropped by a
much smaller measure, 24 percent. In terms of
crime and prison contraction, New York led all
regions of the country and every individual state.68

One in 31: The Long Reach of American Corrections

21

Community Corrections:
A Strategy for Safety and Savings
Building more prisons is not a cost-effective
path to greater public safety. But even if states
wanted to add new cells, they will be hardpressed over the next few budget cycles to find
the money to build them.

“We are never going to build our
way out of there being crime. We
don’t want to put that many people
in jail, and we can’t afford to.”
Chief District Court Judge Joseph Turner, Guildford County, NC
Greensboro News and Record
January 25, 2009

Policy makers must confront the reality that, for
the foreseeable future, roughly seven out of every
ten offenders will continue to serve all or part of
their sentences in the community. Ensuring public
safety and balancing a budget, then, require
states to strengthen badly neglected community
corrections systems, so they can become credible
options for more of the lowest risk offenders who
otherwise would be in prison. This means states
must take a harder look at which offenders should
be locked up and which can be managed
effectively in the community. It means they must
give community corrections agencies the tools
and incentives they need to do their jobs
effectively and hold them accountable for
implementing the supervision strategies that
reflect the wisdom gathered through a quartercentury of research on recidivism reduction.

22

Some states, such as Kansas, Texas and Arizona,
are already well underway. To help spread news
of their good work and share other useful
approaches, the Pew Center on the States in 2008
brought together leading policy makers,
correctional practitioners and researchers to
identify ways to help corrections agencies adopt
the most effective research-based practices. From
those discussions came the “Policy Framework to
Strengthen Community Corrections.” The
framework includes measures that provide
incentives for offenders to stay crime- and drugfree and fiscal incentives for agencies to improve
their success rates—both strategies that can
create new resources for community corrections
agencies without requiring new appropriations.
A detailed menu of policy options, including
suggested language for legislation, executive
orders or court rules, is available at
www.pewcenteronthestates.org/publicsafety.69

“Every time we keep a released
inmate from re-offending, we keep
an innocent person from becoming a
victim, and we save taxpayer dollars.”
CO Gov. Bill Ritter
State of the State Address
January 10, 2008.

Since its inception, the guiding philosophy of
community corrections has bounced back and
forth between law enforcement and social work.
The hallmark of the new approaches is that they

Public Safety Performance Project | Pew Center on the States

C O M M U N I T Y C O R R E C T I O N S : A S T R AT E G Y F O R S A F E T Y A N D S AV I N G S
create a blend of the two strategies that focuses
on a primary mission—preventing crime—and
that is far more potent than either punishment or
treatment by themselves. The key components of
this 21st century corrections system are detailed
in the pages that follow.

A FRAMEWORK FOR LESS CRIME AT LOWER COST
1 Sort Offenders By Risk to Public Safety
2 Base Intervention Programs on Science
3 Harness Technology
4 Impose Swift and Certain Sanctions
5 Create Incentives for Success
6 Measure Progress

1

Sort Offenders by Risk
to Public Safety

A pivotal starting point for community corrections
is the ability to sort offenders by risk—that is, to
accurately separate those who are more likely to
cause great harm from those who may cause
relatively little harm. For decades, that sensitive and
crucial task was left to the educated hunches of
prosecutors, judges and probation or parole officers.
Fortunately, a new generation of risk assessment
tools can now help officials more accurately predict
not only how likely a person is to commit a new
offense but also whether that offense will be a
violent one.70

change, such as an offender’s age at the time of
first arrest, as well as “dynamic” factors, changeable
characteristics such as an offender’s living situation
or current drug use. The risk score then can be
compared with other offenders and used to guide
decisions about whether a particular offender
should go to prison, what level of community
supervision is the best fit, and which interventions
will target the attitudes and behaviors that drive
that specific offender’s criminal activity.
Virginia uses a risk assessment instrument for
felony theft, fraud and drug offenders who
would otherwise be sent to prison under the
state’s sentencing guidelines. Defendants whose
assessment scores are low, based on elements
of the crimes and individual characteristics, are
steered away from prison. In 2008, more than
1,400 of these offenders were sentenced to
community corrections in lieu of prison.71 A
separate assessment for certain sex offenders is
used to find the highest risk cases and double
or triple their terms behind bars.72
Despite significant advances in risk assessment,
the science is still evolving and will always
amount to risk management, not risk elimination.
Such evaluations are not fool-proof, reflecting
instead the best estimate of what a given
person will do. But simple logic dictates that
aside from locking all offenders behind bars
forever, it is impossible to guarantee they will
remain crime-free.
2

While risk tools vary in terms of what they evaluate,
and how much they cost to administer, they
generally rely on a checklist of factors that allow
clinicians to establish a risk score for individual
offenders. These include “static” factors that don’t

Base Intervention
Programs on Science

Along with establishing a sophisticated system
for sorting offenders, states must ensure their
community corrections options are rooted in
today’s robust body of research. While states may

One in 31: The Long Reach of American Corrections

23

C O M M U N I T Y C O R R E C T I O N S : A S T R AT E G Y F O R S A F E T Y A N D S AV I N G S
take different paths toward this goal, they should
always develop and implement policies based on
the best available science. Evidence-based
programs should identify desired outcomes for
offenders and include a means for measuring
progress. Moderate and high risk offenders
should have an individual case plan based on
their risk assessment, and they should be
assigned to programs targeting their unique
behaviors and needs.
Supervision agencies should concentrate their
resources on higher-risk people, times and places.
Risk assessment instruments can help identify the
individuals who need higher intensity supervision
and services. Greater attention also should be
paid to offenders who have just been released,
the times when research shows they are most
likely to fail. Citing a study of over 240,000
offenders released from prison in 13 states, the
National Research Council reported that the
probability of arrest is twice as high in the first
month of supervision as in the 15th month.73
Finally, as the maps of Michigan attest,
supervision and services should be located in the
neighborhoods where offenders live. Too often,
monitoring and resources are located far from
these high-stakes neighborhoods, impeding both
control and rehabilitation.
Agencies striving for better performance are
delivering front-loaded resources to their riskiest
cases in the neighborhoods where the offenders
live. When rooted in these and other evidencebased principles, community corrections
programs can deliver encouraging results.
The implementation of evidence-based practices
results in an average decrease in crime of between

24

10 percent and 20 percent, whereas programs that
are not evidence-based tend to see no decrease
and even a slight increase in crime.74 Interventions
that follow all evidence-based practices can
achieve recidivism reductions of 30 percent.75

“The [evidence based practices] law
is intended to focus our funding
on services that work and get the
greatest return on our investment.”
OR Gov. Ted Kulongoski
governor.oregon.gov
November 29, 2007

In one widely cited 2006 review of more than
550 program evaluations, the Washington State
Institute for Public Policy found that a moderateto-aggressive investment in evidence-based
programs would save state taxpayers $2 billion,
avert prison construction and reduce the crime
rate.76 Some states were already believers, like
Oregon. In 2003, Oregon’s legislature required
that by the 2005 biennium, one-quarter of all
program funding for youth and adult offenders
go to interventions that were evidence-based.
By the 2007 biennium, half of those dollars
were to be spent on evidence-based programs,
and by the 2009 biennium, lawmakers
directed that 75 percent of funding be used
for interventions that are evidence-based.
3

Harness Technology

One supervision technique that is playing an
increasingly important role in many community
corrections programs wasn’t even around 20
years ago—electronic monitoring. Although
conceived as a correctional strategy in the 1960s,
electronic monitoring of offenders did not
become a reality until the 1980s.77

Public Safety Performance Project | Pew Center on the States

C O M M U N I T Y C O R R E C T I O N S : A S T R AT E G Y F O R S A F E T Y A N D S AV I N G S
With dramatic advances in technology, affordable
electronic monitoring today allows officials to
conduct “active” monitoring, in which an offender
wears a transmitter, usually in the form of an ankle
bracelet, that sends a continuous location tracking
signal to a monitoring center. In recent years, such
monitoring has evolved to include the use of
Global Positioning Satellite technology—first
developed by the Department of Defense in the
1970s—to give supervision agents increasingly
detailed information about an offender’s
whereabouts.78 In certain cases, for instance, a
supervisor may be alerted if an offender violates
his parole or probation by going to a location
where he is prohibited by his supervision
conditions. While an alert may not prevent a crime,
the knowledge that law enforcement has such
tracking ability can be a deterrent. “We can’t be on
their doorstep 24/7, but GPS is a way for us to
monitor location and compliance of someone in
the community,” said Chief U.S. Probation Officer
Ken Young. “We can, with reasonable certainty,
know where someone is or has been.”79
Florida is among those states that have used
electronic monitoring extensively and with
positive results. In the early 1980s, Florida
launched a home confinement program for drug,
property and other offenders dubbed
“community control.” Later that decade, the state
began using radio frequency tracking of certain
offenders in the program, and by the 1990s,
Florida had added GPS monitoring to its list of
options for those on community control.
A study of more than 75,000 offenders who
passed through the program between 1998 and
2002 found that, after controlling for offender risk,
those assigned to either form of electronic
monitoring were significantly less likely to

reoffend or abscond.80 On the minus side,
electronic monitoring’s overall record in reducing
recidivism is mixed, and it places significant
new demands on supervisory agents.
Nevertheless, the tool is becoming more
commonly used as an alternative sanction for
some offenders and as an adjunct to traditional
community supervision practices for others.81
Technology also is changing the way in which
offenders are monitored for drug and alcohol
use. Some agencies subject offenders to random
tests for alcohol through breathalyzer equipment
in their homes. “Ignition interlocks” installed in a
vehicle prevent a person from starting the
engine if alcohol is detected in his system. The
driver must blow in the device and pass a breath
test before the vehicle will start. Other agencies
equip offenders with ankle bracelets that can
detect the offender’s blood alcohol level as
ethanol vapor migrates through the skin.
Technology offers policy makers a spectrum of
options that are more intense than traditional
face-to-face community supervision yet far
cheaper than incarceration. Tracking devices and
sensors allow probation and parole officers to
monitor offenders’ whereabouts and behavior in
ways that could hardly have been imagined
when the prison boom began. But if states are
going to make full use of these advances, they
must back the technology with adequate
resources and policies to respond when
offenders are caught breaking the rules.

Impose Swift and Certain
Sanctions for Violations
4

In building stronger community corrections
systems, states should be mindful that
punishment imposed on offenders who break

One in 31: The Long Reach of American Corrections

25

C O M M U N I T Y C O R R E C T I O N S : A S T R AT E G Y F O R S A F E T Y A N D S AV I N G S
the rules of their supervision must be swift,
certain and proportionate. If applied in that way,
sanctions can stop misbehavior early in the
game, thereby reducing the odds that parolees
and probationers will commit more serious
violations and land in an expensive prison cell.
But making swift, certain and proportionate
a reality is a challenge in many of today’s
underfunded, understaffed probation and parole
agencies. Officers struggle with high caseloads,
a lack of suitable community sanctions, and
cumbersome administrative hurdles as they
try to hold violators accountable. As a result,
they often delay pursuing violations before a
court or parole board until an offender has
committed a significant number of
transgressions, at which point revocation to
prison becomes the likely penalty.
To remedy this problem, probation and parole
agencies need an array of graduated sanctions,
as well as clear authority to impose them.
A typical continuum ranges from community
service programs on one end to more restrictive
options such as day reporting centers and
even secure residential treatment facilities
on the other. To maximize the certainty and
swiftness of the sanctions, states should
provide parole and probation agencies the
authority to move offenders up and down
the ladder of sanction programs—even
including short stays in jail—without first
requiring a time-consuming trip back to court.
Georgia has taken this very step, through
a successful program called Probation Options
Management. It allows chief probation
officers or hearing officers within the Georgia
Department of Corrections to impose

26

administrative sanctions on violators in
certain circumstances. An evaluation of the
program shows it reduced by 70 percent
the average number of days offenders spent
in jail awaiting court disposition of their
violations cases,82 saving local jails $1.1 million.83
The program also drastically reduced the
amount of time probation officers spent
waiting in courthouses for violations cases to
be heard, thus freeing up hours that could be
spent on actual supervision of offenders.84
5

Create Incentives for Success

An effective community corrections framework
needs three other staples: incentives for offenders
to change their behavior, a payoff for agencies that
succeed and a system for measuring their results.
The first of these boils down to a fundamental
principle of psychology: When it comes to
motivating people to change their behavior,
carrots work better than sticks.85 The prevailing
philosophy of many community supervision
agencies is the opposite—to try to catch
offenders doing something wrong. But many
agencies, led by drug courts, are now learning
how to use the carrot of positive reinforcement
to keep offenders on the straight and narrow.
What kind of carrots? A variety of approaches
are now afoot, from graduation ceremonies to
gift certificates from local businesses and
removal of restrictions such as curfews. Some
states are starting to push even further, telling
probationers and parolees that they can earn
time off their sentences if they comply with all
of their terms of supervision.
Carrots can work for correctional managers, too.
If community corrections agencies succeed in

Public Safety Performance Project | Pew Center on the States

C O M M U N I T Y C O R R E C T I O N S : A S T R AT E G Y F O R S A F E T Y A N D S AV I N G S

“HOPE” FOR IMPROVEMENT IN HAWAII
With prisons overflowing and correctional budgets straining state finances, probation—allowing lowerrisk offenders to undergo community supervision provided they meet certain conditions—is playing an
increasingly vital role in our nation’s criminal justice system. But all too often, the practice of probation
yields disappointing results.

The Challenge
Probation officers are faced with overwhelming caseloads, outdated technology and cumbersome court
processes for sanctioning violators. As a result, they often are unable to detect when their charges
break the rules or respond with meaningful penalties when they do. Some probationers, convinced that
slip-ups won’t bring immediate consequences, rack up pages of violations for failed drug tests, missed
appointments and other transgressions. Then, at some arbitrary point when they are eventually
brought back to court for a violation hearing, many offenders receive society’s most expensive
punishment—a stay in prison. This approach defies what research and common sense tell us about
effective deterrence and behavior change: punishment is far more effective if it is swift, certain and
proportionate than delayed, unpredictable and severe.
Such was the case in Hawaii until 2004, when Circuit Court Judge Steven Alm decided to create Hawaii’s
Opportunity Probation with Enforcement, or HOPE. The Oahu program involves close partnerships with
prosecutors and defense counsel, police, wardens, and treatment providers, and it is delivering
encouraging results.
HOPE notifies probationers that the old rules remain in place but will now be enforced. That means failures
to comply with frequent but random drug tests, office visits and treatment requirements are met with
immediate sanctions, typically a few days in jail, time that is served over the weekend for probationers with
legitimate jobs. Those who cannot abstain from drugs are placed in residential treatment.

The Results
Preliminary results of a randomized controlled trial found that HOPE participants were less than half as
likely to test positive for drugs (11 percent versus 26 percent) or miss appointments (5 percent versus 12
percent). Early results from a matched comparison group study were even more promising. Arrest rates
for HOPE probationers were three times lower than for the comparison group, and they experienced
significantly lower revocation rates as well (9 percent versus 31 percent).86
This is deterrence in action: a credible threat, combined with resources for those who want to change,
averts both the offending behavior and the need for and cost of punishment.
“Our offenders know that if they use drugs today, they will go to jail tomorrow,” Judge Alm says. “That
means something.”

One in 31: The Long Reach of American Corrections

27

C O M M U N I T Y C O R R E C T I O N S : A S T R AT E G Y F O R S A F E T Y A N D S AV I N G S
thinning the throng of offenders sent back to
prison for new crimes or rule violations, states reap
savings by avoiding prison costs. Those savings
should, in turn, be shared with the successful
community supervision agencies, which can
use these funds to expand their success. This
redirection of dollars can allow states to strengthen
their overall community corrections product
without the need to appropriate new funds.
Among the states that have embraced performance
incentives, Arizona is a recent standout. As often
happens, Arizona’s initiative was sparked by a high
crime rate and a prison population explosion that
was draining taxpayer dollars. From 1997 to 2007,
the state inmate count grew 60 percent, from 23,484
to 37,746, leading to a doubling of the corrections
budget.87 Projections forecast another 50 percent
jump in the prison population by 2017, at an
estimated cost to state taxpayers of $2-3 billion.88
Despite the prison growth, the state still had the
highest crime rate in the nation. State Rep. Bill
Konopnicki, a Republican from Safford who pushed
for reforms along with Republican State Sen. John
Huppenthal of Chandler, painted a gloomy picture
of the prospects: “If we decide to do nothing, we are
in effect committing an additional one billion dollars
in state tax dollars to grow our prison system.”89
Instead, Arizona last year adopted the Safe
Communities Act (SB 1476), a sweeping bill that
creates performance incentives for both
offenders and the county-based probation
supervision system. One part of the law gives
probationers an incentive to pay court-ordered
restitution, complete community service
assignments and comply with their other
conditions of supervision. For every month that
an offender complies with the terms of
supervision, the legislation authorizes the courts

28

to reduce the length of probation by up to 20
days. Slip-ups result in a loss of the earned time.
Under a second part of the bill, signed in June by
then-Gov. Janet Napolitano, counties that reduce
recidivism are awarded 40 percent of the money
the state saves by not having to house repeat
offenders and probation rule violators in its
prisons. The refund is then used by counties to
improve victims’ services and expand access to
drug treatment and other recidivism-reducing
programs. Projections show that if counties
reduce probation revocations by 10 percent, the
state could save nearly $10 million, with 40
percent of that amount returned to the local level.
Faced with a spiking prison population and high
rates of failure by community-based offenders,
Kansas is another state that has taken performance
incentive funding to heart. After recognizing that
about two-thirds of all prison admissions were
probation and parole rules breakers, and that more
than half of the violators needed substance abuse
or mental health treatment, Kansas took action.
Under SB 14, passed in 2007, the state provides
$4 million annually in performance-based grants

“We were faced with spending
millions of dollars on new prisons
to house the expanding population.
Instead, we developed bipartisan
legislation that resulted in treatment
programs for nonviolent drug
offenders and innovative and
collaborative release efforts for inmates
returning to their communities.”
KS Gov. Kathleen Sebelius
State of the State Address
January 12, 2009

Public Safety Performance Project | Pew Center on the States

C O M M U N I T Y C O R R E C T I O N S : A S T R AT E G Y F O R S A F E T Y A N D S AV I N G S

KANSAS PRISON POPULATION TREND
Senate Bill 14 passed in 2007 and was projected to
avert the need for more prisons for 10 years.
12000
11,231
2006 PROJECTION
11000
10000
9251

ACTUAL

Will the gains hold? Budget woes and other
forces are putting them to the test. Recently,
Kansas legislators adopted sentence
enhancements that are fueling projections for
a 10 percent growth in the prison population
over the next decade. At the same time, budget
cuts threaten the very reforms and incentives
that served to reduce the inmate population
pressure and put Kansas on stable footing.

9000
SENATE BILL 14 APPLIED
8000
7455
7000

96 97 98 99 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17

SOURCE: Kansas Sentencing Commission
NOTE: For more information about the Kansas reforms, visit the Kansas page at
www.pewpublicsafety.org

to community corrections programs that increase
probationer and parolee success rates by 20
percent. The grant money goes hand-in-hand
with efforts to train supervision staff in evidencebased practices for effectively managing offenders
in the community.90
Only a couple of years have passed, but Kansas
is already reporting noticeable results. Overall,
the state’s prison population dropped 3.6 percent
between midyear 2007 and year end 2008.
A primary contributor to this drop is a 7
percent reduction in FY2008 of the number
of probationers sent to prison for condition
violations (the top source of prison admissions
in FY2007). Prison admissions of parolees for rule
violations (down 2.2 percent since FY2003) and
new crimes (down 47 percent since FY2003),
as well as parole absconding rates (under 4
percent of the entire caseload), are at or near
all-time lows.91

6

Measure Progress

Incentives, evidence-based programs and offender
sorting all should produce better results—less
crime, fewer victims, and more room in state
budgets for other pressing priorities. But even the
best designed systems must be held accountable
through a method for measuring progress. Just as
law enforcement has shifted from simply counting
arrests to measuring and accepting responsibility
for reducing crime, corrections also needs to
evaluate outcomes of its work.
An admirable standard for public safety
performance measurement was set in the mid1990s by the New York City Police Department’s
Compstat program. Short for “compare statistics,”
Compstat involves the continuous evaluation of
agency performance through live, ongoing
audits. Information on crimes, arrests and other
critical measures is distributed to managers
department-wide, then reviewed in weekly
sessions where unit commanders are called
before their leaders to explain crime trends as
well as their strategic and tactical responses. This
combination of real-time data and transparent,
immediate feedback created incentives to adopt
practices that better protect the public. The
ultimate payoff: Compstat and better crime
analysis helped New York City reduce crime.92

One in 31: The Long Reach of American Corrections

29

C O M M U N I T Y C O R R E C T I O N S : A S T R AT E G Y F O R S A F E T Y A N D S AV I N G S
Momentum is building to adapt Compstat’s core
principles—accurate and timely intelligence;
deployment of resources where they are most
needed; effective tactics; and relentless follow-up
and assessment—to the community corrections
field.93 The overall goal is to lower recidivism rates
among probationers and parolees, but other key
performance measures include employment,
substance abuse and payment of victim
restitution rates. Another yardstick would track
whether supervised offenders are successfully
discharged at the end of their supervision term.

30

Agencies in several states, including Maryland,
New York and Georgia, have adopted Compstatlike systems and are beginning to show
promising outcomes. The rate at which offenders
successfully complete their parole terms in
Georgia, for example, has risen by four
percentage points under the new approach. It
may not sound like much, but each percentage
point is estimated to save the state $6 million to
$7 million in reduced incarceration costs.94

Public Safety Performance Project | Pew Center on the States

A Rare Moment in Time
The revelation last year that 1 in 100 adults is
behind bars led to action in some states as
political leaders took a fresh look at sentencing
and correctional policies with an eye toward
better balancing public safety, offender
accountability and the realities of tight budgets.
Today, economic crisis is again changing the
game. States are in dire fiscal shape, slashing
programs and services in ways that will exact a
considerable human toll. To balance their budgets,
many will have to slow prison growth or even
shutter entire institutions. Community corrections
programs, already strained from years of neglect,
will be asked, once again, to do more with less.
But tight budgets can inspire better policy
making and a heightened vigilance to ensure
every tax dollar delivers maximum value for the
public. Such is the case today with respect to the
massive, expensive and underperforming
correctional system in America.
Research and experience have led practitioners,
analysts and policy makers to develop a set
of sentencing and correctional principles
that meet that challenge. With adequate
resources and authority, courts and community
corrections professionals can determine
which offenders should be in prison and for
how long. With new supervision strategies and
technologies, the lower-risk offenders can
be managed safely and held accountable in
the community, at lower cost and with better
results than incarceration achieves.

“We won’t get true public safety and
protection for crime victims until we
invest in community corrections –
because most offenders are not behind
bars, but living as our neighbors.”
Anne Seymour, National Crime Victim Advocate
Personal Communication
2009

These efforts need to be strengthened, not scaled
back. Cutting them may appear to save a few
dollars, but it won’t. It will fuel the cycle of more
crime, more victims, more arrests, more
prosecutions and still more imprisonment.
Better performance in community corrections can
cut crime and avert the need not only for new
prisons but even for some we already have. And
the accrued savings, if used to reinforce probation
and parole, support early-intervention strategies,
or shore up the high-stakes neighborhoods where
prisoners come from and return to, can generate
even further reductions in crime and incarceration.
Some states are putting research into action
and carefully modernizing their correctional
playbooks. Others should follow suit.
Meaningful progress will take time, and will
require focus and determination from state
leaders. But doing nothing is unacceptable.
Continuing down the same path is an affront to
taxpayers who rightly expect government to learn
from its failings and build upon its success.

One in 31: The Long Reach of American Corrections

31

Endnotes
1

See Ten Steps Corrections Directors Can Take to Strengthen
Performance, The Pew Charitable Trusts (Washington D.C.: May
2008); and Protecting Public Safety During a Budget Crisis:
Managing Corrections, National Governors Association, Social,
Economic and Workforce Development Division (Washington
D.C.: Dec. 2008).
2 Evidence on the causes of prison growth from 1980 until 1996
points almost entirely to sentencing and corrections policies.
According to a seminal article on the issue, the rise in
incarceration of non-drug offenders, who constitute the vast
majority of the prison population, was due entirely to a higher
probability of incarceration after an arrest and an increase in
time-served after commitment to a penal institution. Cited in
Alfred Blumstein and Allen J. Beck, “Population Growth in U.S.
Prisons, 1980-1996,” Prisons (1999).
3 In descending order of population, the states are WA, AZ, MA,
IN, TN, MO, MD, WI, MN, CO, AZ, SC, LA, KY, OR, OK, CT, IA, MS,
AR, KS, UT, NV, NM, WV, NE, ID, ME, NH, HI, RI, MT, DE, SD, AK,
ND, VT, DC, WY.
4 Glaze, Lauren E. and Thomas P. Bonczar, 2007 Statistical Tables,
Probation and Parole in the United States (Washington, D.C.:
Bureau of Justice Statistics, Dec. 2008). Available online at
http://www.ojp.usdoj.gov/bjs/pub/pdf/ppus07st.pdf.
5 Available data do not permit breakdown of rates by race and
gender.
6 Glaze, Lauren E. and Thomas P. Bonczar, 2007 Statistical Tables,
Probation and Parole in the United States (Washington, D.C.:
Bureau of Justice Statistics, Dec. 2008). Available online at
http://www.ojp.usdoj.gov/bjs/pub/pdf/ppus07st.pdf.
7 Ibid.
8 Taxman, Faye S., Douglas W. Young, Brian Wiersema, Anne
Rhodes, and Suzanne Mitchell, “The National Criminal Justice
Treatment Practices Survey: Multilevel Survey Methods and
Procedures,” Journal of Substance Abuse Treatment, 32 (3), (April
2007).
9 West, Heather C., and William J. Sabol, Prisoners in 2007, U.S.
Department of Justice, Bureau of Justice Statistics
(Washington, D.C.: Dec. 2008). Available online at
http://www.ojp.usdoj.gov/bjs/pub/pdf/p07.pdf.
10 One in 27 Michiganders are under correctional control but, for
mapping purposes, The Pew Center on the States and partner,
the Justice Mapping Center, were only able to secure
residential information for 122,165 of the 278,808 supervised
offenders in the state. In particular, due to the state’s structure
for misdemeanant probationers, we were unable to gather
geographic information for a large portion of the state’s overall
probationer population. Misdemeanor offenses are those that
typically result either in fines, community supervision, or shortterm stays in local or county jails. Felony offenses, by contrast,
can result in prison stays exceeding one year, in addition to all
the sanctions open to misdemeanants.
11 On neighborhood destabilization, see Todd R. Clear and Dina R.
Rose, “When Neighbors Go to Jail: Impact on Attitudes About
Formal and Informal Social Control.” Summary of a presentation
at the U.S. Department of Justice, National Institute of Justice,
July 1999. On decreased wage growth, see Bruce Western,

12

13
14
15
16

17

18

19
20

21

22

23
24
25

Punishment and Inequality in America (New York: Russell Sage
Foundation, 2006). On incarceration and AIDS infection rates,
see Rucker Johnson and Steven Raphael, “The effects of male
incarceration dynamics on AIDS infection rates among AfricanAmerican women and men,” Journal of Law and Economics
[forthcoming]; available online at http://gsppi.berkeley.edu/
faculty/sraphael/aids-paper-july-2007.pdf.
Analysis based on National Association of State Budget
Officers, “State Expenditure Report” series. Available online at
http://www.nasbo.org/publications.php. Budget figures
throughout this report are not adjusted for inflation.
Ibid.
Ibid.
Ibid.
Bureau of Justice Statistics, “Direct expenditures by criminal
justice function, 1982-2006.” Available online at
http://ojp.usdoj.gov/bjs/glance/tables/exptyptab.htm.
While organizations such as the National Association of State
Budget Officers and the National Conference of State
Legislators annually conduct surveys of state expenditures,
they do not distinguish between spending on prisons,
probation and parole when measuring corrections
expenditures. The most recent known similar survey was
conducted by the Criminal Justice Institute in 2002. The lead
partner for this fiscal year 2008 survey was the American
Probation and Parole Association, and other partners included
the Crime and Justice Institute, the Council of State
Governments Justice Center, the National Center of State
Courts and the Vera Institute of Justice.
They are AR, AL, AK, CO, DE, GA, ID, IA, KY, LA, ME, MD, MI, MN,
MS, MO, MT, NE, NC, ND, NH, NM, NY, OK, OR, PA, RI, SC, SD, TN,
TX, VT, VA and WY.
They are LA, WY, AL, MO, MT, GA, OR and NY.
Original data from Pew Center on the States’ survey
administered by the American Probation and Parole
Association.
Colorado Division of Criminal Justice, “Colorado Community
Corrections: Annual Report FY 2007.” Available online at
http://dcj.state.co.us/occ/pdf/Annual%20Report%200607%20FINAL.pdf.
Current national probation caseload estimates are not
available. The most recent national publication to address the
issue was The Corrections Yearbook (Middletown, CT: The
Criminal Justice Institute, 2002). According to the yearbook, in
2001, average regular probation caseloads in 11 reporting
states were 127 offenders per officer. About 15 percent of
offenders were on specialized caseloads, which had offenderto-officer ratios of 29:1 for intensive supervision and 40:1 for
specialized supervision. Parole caseloads were derived from
original data collected for this report.
Wicklund, Carl, American Probation and Parole Association
executive director. Interview by phone, Jan. 8, 2009.
Ibid.
One in 100: Behind Bars in America 2008, Public Safety
Performance Project, The Pew Charitable Trusts (Washington,
D.C.: Feb. 2008).

One in 31: The Long Reach of American Corrections

33

E N D N OT E S
26 West, Heather C., and William J. Sabol, Prisoners in 2007, U.S.
Department of Justice, Bureau of Justice Statistics
(Washington, D.C.: Dec. 2008). Kentucky’s state prison system
has jurisdiction over 22,000 inmates but custody of only
15,000. The remainder, as described, is housed in county jail
facilities.
27 Lawson, Robert, “PFO Law Reform: A Crucial First Step Toward
Sentencing Sanity in Kentucky,” Kentucky Law Journal, 97 (1),
(2008-2009).
28 Ibid.
29 Ibid.
30 Analysis based on Bureau of Justice Statistics, “Prisons at
Yearend” series, available online at
http://www.ojp.usdoj.gov/bjs/prisons.htm. These rates are
calculated with adults and minors as “residents” and thus are
not comparable with other rates in this paper, which are adult
resident rates.
31 Based on analysis of the Federal Bureau of Investigation’s
Uniform Crime Reports (UCR), data available online at
http://bjsdata.ojp.usdoj.gov/dataonline/Search/Crime/State/St
ateCrime.cfm.
32 Cheves, John, “With prisons full, many county jails are
overflowing with felons,” The Lexington Herald Leader (Jan. 13,
2008).
33 Ibid.
34 Kentucky’s state prison system has jurisdiction over 22,000
inmates but custody of only 15,000. The remainder, as
described, is housed in county jail facilities.
35 Cheves, John, “With prisons full, many county jails are
overflowing with felons,” The Lexington Herald Leader (Jan. 13,
2008).
36 Ibid.
37 Original data from Pew Center on the States’ survey
administered by the American Probation and Parole
Association.
38 Glaze, Lauren E. and Thomas P. Bonczar, 2007 Statistical Tables,
Probation and Parole in the United States (Washington, D.C.:
Bureau of Justice Statistics, Dec. 2008). Available online at
http://www.ojp.usdoj.gov/bjs/pub/pdf/ppus07st.pdf. The
number of parolees cited here differs from elsewhere in this
report because this figure includes offenders on parole at any
time during the year rather than at year end.
39 Associated Press, “Kentucky Legal Experts Agree on Prison
Report” (Nov. 25, 2008).
40 Ellis, Ronnie, “Counties Sue State Over Inmate Costs,” McCreary
County Record (Nov. 10, 2008).
41 Analysis based on National Association of State Budget
Officers, “State Expenditure Report” series. Available online at
http://www.nasbo.org/publications.php.
42 Government Accountability Office, Update of State and Local
Government Fiscal Pressures, Letter to U.S. Senate Finance
Committee (January 26, 2009). Available online at
www.gao.gov/new.items/d09320r.pdf.
43 Center on Budget and Policy Priorities, “States in Trouble Due
to Economic Downturn,” Policy Points (Jan. 29, 2009). Available
online at http://www.cbpp.org/policy-points10-20-08.htm.
44 Cornfield, Jerry, “Legislature’s $6 billion problem,” HeraldNet
(Jan. 11, 2009).
45 Piehl, Anne M., Bert Useem and John J. DiIulio, Jr., Right-Sizing
Justice: A Cost-Benefit Analysis of Imprisonment in Three States
(New York: Manhattan Institute, Sept. 1999).

34

46 Ibid. Note that these were entering inmates, not current
inmates. Assuming that more harmful offenders serve longer
terms, we would expect the average current inmate to have
higher social costs than the average entering inmate. Also,
because these costs are based on reported criminal activity,
they are both reliant on offender accuracy and inclusive only
of crimes averted through incapacitation. As such, they do not
capture deterrent, or any other, effects.
47 Piquero, Alex and Alfred Blumstein, “Does Incapacitation
Reduce Crime,” Journal of Quantitative Criminology, 23, pp.
267-285 (2007); Wilson, James Q., “Prisons in a Free Society,”
The Public Interest (Fall, 1994); DiIulio, John J., Jr., “Two Million
Prisoners Are Enough,” The Wall Street Journal (March 12,
1999); Stemen, Don, Reconsidering Incarceration: New
Directions for Reducing Crime (New York: Vera Institute of
Justice, Jan. 2007); Spelman, William, “The Limited Importance
of Prison Expansion,” The Crime Drop in America (Cambridge,
MA: Cambridge University Press, 2000); Liedka, Raymond V.,
Anne M. Piehl and Bert Useem, “The Crime-Control Effect of
Incarceration: Does Scale Matter?” Criminology and Public
Policy, 5 (2), (2006).
48 Washington State Institute for Public Policy, The Criminal Justice
System in Washington State: Incarceration Rates, Taxpayer Costs,
Crime Rates, and Prison Economics (Olympia, WA: Jan. 2003).
49 Ibid.
50 Yearwood, Douglas L., Richard Hayes, Justin Davis and James
Klopovic, “Comparing Incarceration and Community
Corrections Costs in North Carolina,” Journal of Community
Corrections (Summer 2008).
51 State of Oregon Criminal Justice Commission, “Report to the
Legislature” (Jan. 2007). Available online at
http://www.oregon.gov/CJC/docs/2007cjcreport.pdf.
52 Washington State Institute for Public Policy, The Criminal
Justice System in Washington State: Incarceration Rates, Taxpayer
Costs, Crime Rates, and Prison Economics (Olympia, WA: Jan.
2003).
53 Aos, Steve, “Evidence-Based Policy Options to Reduce Crime,
Criminal Justice Costs and Prison Construction,” Symposium on
Alternatives to Incarceration (Washington, D.C.: United States
Sentencing Commission, July 14-15, 2008).
54 Liedka, Raymond V., Anne M. Piehl and Bert Useem, “The
Crime-Control Effect of Incarceration: Does Scale Matter?”
Criminology and Public Policy, 5 (2), (2006).
55 Ibid.
56 West, Heather C., and William J. Sabol, Prisoners in 2007, U.S.
Department of Justice, Bureau of Justice Statistics
(Washington, D.C.: Dec. 2008). This rate represents the number
of prisoners in America divided by the total residents of the
nation. This report and One in 100: Behind Bars in America 2008,
used the total number of prison and jail inmates divided by
the adult population.
57 See, for example, Blumstein, Alfred, The Impact of Incarceration
on Crime: Two National Experts Weigh In, Public Safety
Performance Project, The Pew Charitable Trusts (Washington,
D.C.: April 2008).
58 Piehl, Anne M. and John J. DiIulio, Jr., ‘Does Prison Pay?’ Revisited
(Washington, D.C.: Brookings Institution, Jan. 1995).
59 Liedka, Raymond V., Anne M. Piehl and Bert Useem, “The
Crime-Control Effect of Incarceration: Does Scale Matter?”
Criminology and Public Policy, 5 (2), (2006); Wilson, James Q.,
“Prisons in a Free Society,” The Public Interest (Fall, 1994).

Public Safety Performance Project | Pew Center on the States

E N D N OT E S
60 Wilson, James Q., “Prisons in a Free Society,” The Public Interest
(Fall, 1994).
61 Thompson, Don, “Aging Inmates Adding to State’s Prison
Strain, Costing More,“ Associated Press (July 7, 2008).
62 Spelman, William, “The Limited Importance of Prison
Expansion,” The Crime Drop in America (Cambridge, MA:
Cambridge University Press, 2000).
63 Ibid.
64 Western, Bruce, Punishment and Inequality in America (New
York: Russell Sage Foundation, 2006).
65 Spelman, William, “Specifying the Relationship Between Crime
and Prisons,” Journal of Quantitative Criminology, 24, pp. 149178 (2008); Piquero, Alex and Alfred Blumstein, “Does
Incapacitation Reduce Crime,” Journal of Quantitative
Criminology, 23, pp. 267-285 (2007); Stemen, Don,
Reconsidering Incarceration: New Directions for Reducing Crime
(New York: Vera Institute of Justice, Jan. 2007).
66 Western, Bruce, Punishment and Inequality in America (New
York: Russell Sage Foundation, 2006); Spelman, William, “The
Limited Importance of Prison Expansion,” The Crime Drop in
America (Cambridge, MA: Cambridge University Press, 2000).
67 Analysis of data from the Bureau of Justice Statistics, “Prisoners
at Yearend” series.
68 Analysis of data from the Federal Bureau of Investigation,
Uniform Crime Reports.
69 The options are supported by a separate set of guiding
principles for community corrections developed by leaders in
the field. These leaders were convened by The Urban Institute
in June 2008 with funding from the National Institute of
Corrections, the Bureau of Justice Assistance and the JEHT
Foundation. For a summary of these principles, see Putting
Public Safety First, available online at
http://www.pewcenteronthestates.org/report_detail.aspx?id=
46538.
70 Douglas, Kevin and Jennifer Skeem, “Violence Risk Assessment:
Getting Specific About Being Dynamic,” Psychology, Public
Policy and Law, 11 (3), pp. 347-383 (2005); Hilton, N. Zoe, Grant
T. Harris and Marnie E. Rice, “Sixty-six years of research on the
clinical versus actuarial prediction of violence,” The Counseling
Psychologist, 34 (400), (2006).
71 Virginia Criminal Sentencing Commission, 2008 Annual Report.
72 Ibid.
73 National Research Council, Parole, Desistance from Crime, and
Community Integration (Washington, D.C.: The National
Academy Press, 2007).
74 See, for example, Andrews, D.A. and Craig Dowden, “The RiskNeed-Responsivity Model of Assessment in Human Service
and Prevention and Corrections Crime Prevention
Jurisprudence,” Canadian Journal of Criminology and Criminal
Justice, 49 (4), (2007); Aos, Steve, Marna Miller and Elizabeth
Drake, Evidence-Based Public Policy Options to Reduce Future
Prison Construction, Criminal Justice Costs, and Crime Rates
(Olympia, WA: Washington State Institute for Public Policy, Oct.
2006); Lipsey, Mark W. and Francis T. Cullen, “The Effectiveness
of Correctional Rehabilitation: A Review of Systematic
Reviews,” Annual Review of Law and Social Science, 3 (2007).
75 Andrews, D.A., et al, “Does correctional treatment work?: A
clinically-relevant and psychologically-informed meta-analysis,”
Criminology, 28, pp. 269-404 (1990).

76 Aos, Steve, Marna Miller and Elizabeth Drake, Evidence-Based
Public Policy Options to Reduce Future Prison Construction,
Criminal Justice Costs, and Crime Rates (Olympia, WA:
Washington State Institute for Public Policy, Oct. 2006).
77 Bottos, Shauna, An Overview of Electronic Monitoring in
Corrections: The Issues and Implications (Ottawa, ON:
Correctional Service of Canada, April 2007).
78 “GPS: Your Supervisory Officer is Watching,” The Third Branch:
Newsletter of the Federal Courts, 39 (4), (April 2007).
79 Ibid.
80 Padgett, Kathy G., William D. Bales and Thomas G. Blomberg,
“Under Surveillance: An Empirical Test of the Effectiveness and
Consequences of Electronic Monitoring,” Criminology and
Public Policy, 5 (1), (Feb. 2006).
81 Ibid.
82 Speir, John, et al, An Evaluation of Georgia’s Probation Options
Management Act (Atlanta: Applied Research Services, Oct. 24,
2007).
83 Grant, Johnny, Georgia State Senator. Remarks before the
National Conference of State Legislatures Fall Forum, Dec. 11,
2008.
84 Speir, John, et al, An Evaluation of Georgia’s Probation Options
Management Act (Atlanta: Applied Research Services, Oct. 24,
2007).
85 Latessa, Edward, Francis Cullen and Paul Gendreau, “Beyond
Professional Quackery: Professionalism and the Possibility of
Effective Treatment,” Federal Probation, 66 (2), pp. 43-49 (2002);
Crime and Justice Institute, Implementing Evidence-Based
Practice in Community Corrections: The Principles of Effective
Intervention (Boston: 2004).
86 Hawken, Angela and Mark Kleiman, “Research Brief: Evaluation
of H.O.P.E. Probation” (July 2008). Available online at
http://www.pewcenteronthestates.org/uploadedFiles/HOPE_R
esearch_Brief.pdf.
87 Analysis based on Bureau of Justice Statistics, “Prisons at
Yearend” series, available online at
http://www.ojp.usdoj.gov/bjs/prisons.htm; and, Council of
State Governments Justice Center, Recent and Projected Growth
of the Arizona Prison Population (Feb. 2007), available online at
http://justicereinvestment.org/states/arizona/pubmaps-az.
88 Council of State Governments Justice Center, Recent and
Projected Growth of the Arizona Prison Population (Feb. 2007).
89 The Council of State Governments Justice Center, “Arizona’s
Prison Population Projected to Grow Twice as Fast as General
Resident Population, Independent Study Finds.” Press Release,
Feb. 6, 2007. Available online at http://justicecenter.csg.org/
downloads/AZ+Pew+Press+Release+PDF.pdf.
90 Council of State Governments Justice Center, Justice
Reinvestment State Brief: Kansas (Oct. 2007).
91 Kansas Department of Corrections, “Statistical Profile: FY 2008
Offender Population” (Oct. 2008). Available online
athttp://www.dc.state.ks.us/publications/Statistical%20Profile%
20FY%202008%20Offender%20Population.pdf.
92 See You Get What You Measure: CompStat for Community
Corrections, Public Safety Performance Project, The Pew
Charitable Trusts (Washington, D.C.: Nov. 2007). Available
online at http://www.pewcenteronthestates.org/
report_detail.aspx?id=32106.
93 Ibid.
94 Ibid.

One in 31: The Long Reach of American Corrections

35

Methodology Notes
Overview: This report analyzed prison, jail, parole and
probation populations individually and as a share of the total
adult population both on a national and a state level. Trends
over time in these corrections populations and as a share of
the adult population were expressed in a 25-year span of year
end figures, beginning with year end 1982 and ending at year
end 2007. These year end data were derived through a variety
of methods explained below.
National Corrections Populations: Data from the Bureau of
Justice Statistics (BJS) Correctional Surveys (available at
www.ojp.usdoj.gov/bjs/ glance/tables/corr2tab.htm) were
used for all national-level correctional population figures.
These include national prison, jail, probation and parole
population figures. Where national corrections populations are
expressed as a rate, such as “1 in 31,” they have been combined
with adult resident population data from the U.S. Census.
Adult Population: U.S. adult resident population figures were
derived nationally and for each state from midyear data
prepared by the U.S. Census Bureau, State Population
Estimates, going back to midyear 1981. These midyear census
figures were then averaged to create year end figures which
were used for all calculations throughout this report. The rate
of growth for midyears 2006 to 2007 was applied to midyear
2007 figures to derive projected midyear 2008 figures; these
were then averaged with the midyear 2007 figures to estimate
year end figures for 2007.
State Corrections Populations: A variety of sources were
compiled to generate the different components of the total
corrections population: prisoners, jail inmates, probationers
and parolees. The different data sources for each component
are described below. Additionally, the methods used to
compensate and adjust for missing data are discussed.
Prison Inmates: State prison figures used throughout the
report include BJS year end state prison counts as well as year
end counts and estimates of federal inmates by state of
reported residence from the Federal Bureau of Prisons (BOP).
The prison inmate figures in this report exclude state prisoners
held in local jails; they have been counted as part of the jail
population as described in the section on jail inmates.

36

year end counts used in this report for the years 1982 through
1998. BOP does not have home-state addresses for all inmates.
As a percentage of the total BOP population, the number with
a reported home state residence hovers around 83 percent
from 1999 to 2003, and climbs steadily to 88 percent from 2004
to 2007. To conservatively estimate the total number of BOP
inmates that came from all states for the years 1982 to 1998,
the 1999 to 2003 “83 percent” average was applied to the total
BOP population reported for each of those years. Then, each
state’s 1999 to 2003 average share of BOP inmates was applied
to the estimated “83 percent” count for each year in the 1982 to
1998 period. This gave a rough estimate of the number of
federal prisoners from each state for the years 1982 to 1998
which allowed the state level prison population estimates to
better reflect the actual prison population.
Jail Inmates: State jail figures are based largely on BJS surveys
of jail inmates conducted in February 1978 and at midyears
1983, 1988, 1993, 1999 and 2005. Since statewide jail counts
are not available in the intervening years, a straight-line
estimation was applied to obtain jail populations for each of
those years, and to adjust all figures to year end counts. This
method provided the jail population estimates from 1982 to
2006. To this data set was added year end 2007 state jail
counts derived from survey data wherever available and
estimates where states wouldn’t or couldn’t respond to the
survey questions.
While BJS was able to provide an estimated national year end
2007 jail population, individual figures for each state were not
available. The year end 2007 state jail counts used in this
report include a combination of counts reported to JFA
Associates by 22 states and Washington D.C. and estimations
for 23 states. Five states with unified jail and prison systems
(CT, VT, RI, DE and HI) were assigned year end 2007 jail
populations of zero, consistent with BJS’s reporting in all
previous years when these states’ jail inmates were counted in
their reported prison populations. Though often considered a
unified system state, Alaska has a small local jail population
and was therefore included in the estimation process.

State prison counts as reported to BJS are conventionally
calculated using only those inmates held under state
jurisdiction or custody. Nationally, this method excludes nearly
200,000 inmates who are held in the federal prison system and
leads to state imprisonment figures that, purely due to statefederal jurisdictional boundaries, are lower than corresponding
national figures. In order to provide a more complete account
of prisoners by state, federal inmates were added back into
each state’s BJS-reported counts. Overall, this adjustment
allows, for example, BOP prisoners from Wisconsin to be
counted as Wisconsin prisoners. This was done using BOP data.

JFA Associates surveyed all 50 states and DC and received jail
population counts from 23 of them. All told, JFA reported jail
populations for CA, DC, FL, IN, KY, LA, ME, MD, MA, MI, MN, NJ,
NM, NY, OH, PA, SC, TN, TX, VA, WA, WV and WI totaling nearly
547,000 inmates, representing 70 percent of BJS’s reported
2007 national jail population. For the remaining 23 states, a
year end 2007 jail population was estimated by applying the
rate of growth experienced by the 22 respondent states
between midyear 2005 and year end 2007 to the midyear
2005 population of the 23 estimated states. The rate, 1.7
percent, appears to be a conservative growth estimate, as BJS
recently reported a 1.9 percent rate of growth for jails
nationally from midyear 2006 to midyear 2007.1

The BOP provided year end counts of federal inmates by state
of reported residence for the years 1999 through 2007. These
counts were used in this report, and were also used to estimate

To avoid double-counting of prisoners, a count or estimate of
prisoners held in local jails was subtracted from each state’s
jurisdictional prisoner count. BJS provided these data from

Public Safety Performance Project | Pew Center on the States

M E T H O D O LO G Y N OT E S
1999 to 2007; these counts were used in this report, and were
also used to estimate year end counts used in this report for
the remaining years, 1982 to 1998. The average percent of
inmates who would have been double-counted (by state)
during the years from 1999 to 2005 was applied to the total
jail and prison count of each state for all years before 1999. For
11 states, the average was zero and most states were relatively
consistent. The modeling is based on counts from 1999 to
2005 because this is the period for which BJS state-by-state
data, or imputed jail populations are available.
Probationers and Parolees: State probation and parole figures
include BJS year end counts as well as counts from the
Administrative Office of the U.S. Courts (AOC). As with the U.S.
Census data, the AOC data were midyear counts, which were
averaged to create year end estimates for the purposes of this
report. Similar to the how federal prisoners are conventionally
absent from state prison counts, state probation and parole
statistics typically ignore offenders in the states under
supervision in the community by federal authorities. The AOC
provided counts by state of community supervision offenders
under federal jurisdiction for the entire 25-year period from
1982 to 2007. For counting purposes, the federal definitions
“judge probation,” “magistrate judge probation” and “pretrial
diversion” were combined to form a single federal probation
category. A single federal parole category was constructed
out of the various federal forms of post-prison supervision—
“term of supervised release,” “parole,” “mandatory release,”
“military parole” and “special parole.” These federal probation
and parole categories were added to each state’s BJS-reported
probation and parole counts, respectively, to obtain total
statewide probation and parole counts.
Complicating issues: There were numerous data challenges
that make it difficult to provide an accurate state level count.
These include problems with double counting, shifting
definitions and missing data on the community corrections
population. These issues and the steps taken to deal with
them are presented below.
Double-counting: Offenders involved with more than one
criminal justice agency could be double-counted and
artificially inflate most measures of correctional control. For
example, an offender on probation or parole might be
imprisoned or jailed but not removed from the probation or
parole rolls. BJS has used increasingly sophisticated measures
to avoid double-counting. In its most recent such release of
data, BJS adjusted for possible overlap between probation,
parole, prison and jail counts. Its adjusted total correctional
population amounts to a 1.11% reduction in the sum of the
separate probation, parole, prison and jail counts. Pew could
not perform such adjustments for the 50-states with available
data. This could lead to overestimates, a risk that could affect
states like Georgia that have many agencies, some privatized,
handling large corrections populations. In Georgia’s case,
there are concerns that some individuals on probation in
multiple jurisdictions might be counted separately for each
jurisdiction and that private agencies report counts of cases
under supervision rather than individual offenders.
Change in counting definition: In 1998, BJS revised the
probation survey used in its population counts to include more
reporting agencies. In states like Georgia and Idaho, this revision
expanded counts to include court-based populations, often of

misdemeanants. One consequence is that these additional
reporting agencies increase the risk of double-counting,
discussed above. Unfortunately, due to the limited data on the
specifics of these additional groups included in the updated
probation and parole statistics, it is impossible to determine the
degree to which the additional reporting agencies are
correcting previous underestimates of the correctional
population and the degree to which the additional agencies
are contributing to overestimates of that population.
Share of correctional population in institutional versus
community settings: Double-counting of offenders with
multiple criminal justice statuses and the change in counting
rules would tend artificially to inflate the share of the
correctional population that is under supervision in the
community. However, as noted in the sidebar, “An Even Wider
Net?,” there may be a large number of offenders in pre-trial
supervision programs, drug courts or other court-based
alternative sentencing units, and other specialized programs
who are not picked up in conventional probation or parole
counts. These populations have likely increased over time,
especially due to the proliferation of drug courts. These
various counting issues offset each other to some unknown
degree. A more precise estimate of the community
supervision population, and therefore its share of the total
correctional population, will be identified only when more
extensive and detailed surveys are designed and conducted.
Spending Figures: To collect current and past prison,
probation and parole spending figures from all states, the Pew
Center on the States partnered with the American Probation
and Parole Association (APPA). APPA designed a survey and
coordinated data collection with partner organizations
including the Crime and Justice Institute, the Council of State
Governments Justice Center, the National Governors
Association and the Vera Institute of Justice. Forty-five states
completed at least a portion of the survey and 34 provided
data on probation, parole and prison expenditures for both
FY08 and at least one baseline fiscal year. Analysis of these
data, performed both by APPA and Pew, forms the basis of the
fiscal analyses included in this report and the state fact sheets.
The APPA survey asked for total fiscal year expenditures as
well as per diem costs for administering probation, parole and
prison supervision and services. While respondents were
asked to exclude capital costs, they were requested to include
costs for personnel, operations, treatment and an undefined
“other” category. Respondents were asked for this information
both in FY2008 and an historical baseline as far back as
FY1983 (or in five-year increments from that point forward).
For states that were unable to complete the APPA survey,
statistics were gathered from the National Association of State
Budget Officers (NASBO) State Expenditure Reports. These
reports contain an impressive scope of state spending
information, going back more than 20 years. The corrections
spending figures they contain, however, do not distinguish
between corrections expenditures including probation, parole
and prison.
Throughout the report, spending figures have not been
adjusted for inflation.
1

Sabol, William and Todd D. Minton. June 2008. Jail Inmates at Midyear 2007.
Washington, DC: Bureau of Justice Statistics, U.S. Department of Justice.

One in 31: The Long Reach of American Corrections

37

Jurisdictional Notes
Within the 50 states and the District of Columbia there are
hundreds of prison, probation and parole agencies (in
addition to many more jails and community corrections
agencies) operating with different population and budget
counting rules. The following notes are provided to explain
some of these differences and to account for many of the
idiosyncrasies in the reported data. The notes are based on
reports collected by the Bureau of Justice Statistics (BJS) and
the American Probation and Parole Association, as well as
direct contacts with state officials, but they are not a
complete description of all counting issues.
Alabama: Probation and parole spending figures include all
probationers under the Department of Corrections, but do
not include some offenders sentenced to community
supervision. Rental costs are included in operating costs.
Treatment services were not provided internally in 1983.
Most prison treatment services are now provided by the
DOC internally. Mental health treatment is contracted out to
private companies. These costs are grouped under
professional services that include health care services, leased
bed contracts, and all treatment programs.

Georgia: The budget total for 1983 did not provide a specific
subcategory total for parole supervision or other agency
functions. A close approximation of the supervision portion
of the budget was calculated by the respondent by
determining parole supervision’s share of the 2008 budget
(69 percent) and applying it to the 1983 total. Figures for
parole in 2008 include funds for GPS monitoring, not
included in the 1983 budget. Georgia’s probation population
appears to be inflated both by a number of local ordinance
violators under the jurisdiction of the state courts and by
counts by private probation providers that reflect probation
cases rather than probationers. This means that some
probationers with multiple convictions may be counted more
than once. The population count also may include a number
of people whose probation terms have ended but for whom
there are outstanding warrants.

Alaska: Though the state’s prison system manages most jail
inmates, BJS reports that there are several dozen jail inmates
in local jails throughout the state. A state jail population was
therefore estimated for Alaska (see Methodology Notes).

Hawaii: For 1998, the total budget expenditure for probation
includes payroll costs but does not include fringe benefits.
Hawaii maintains a unified state jail and prison system and,
per reporting to the BJS, has in this report a single figure for
its incarcerated population.

Arkansas: In 2008, the Department of Community Corrections
operated community corrections beds and probation and
parole. As opposed to later years, some probation and parole
costs were included in prison costs in 1984.

Illinois: Illinois does not have a reported parole population
in the BJS parole survey of 2006. For this report, this void was
filled by a straight-line average of the state’s 2005 and 2007
figures.

Colorado: Per diem costs for Colorado prisons are weighted
to include both state and privately managed facilities.
Probation figures do not cover all expenditures excluding,
for example, grants for pilot programs and victim services. In
all reported years, costs cover only those probationers in the
state courts as well as from the District Court of Denver, but
exclude probationers from the City or County of Denver and
the approximately 20,000 probationers in Colorado
supervised by private agencies funded through offender
supervision fees. Probation figures were adjusted by survey
respondent to account for differences in funds received from
the collection of drug offender assessment fees, which were
collected in 2008 but not 2003.

Louisiana: Probation and parole expenditure figures include
offender fees.

District of Columbia: Tracking correctional populations in the
District is complicated by the transfer to federal custody of all
District prisoners as a result of the 1997 Revitalization Act. For
this report, District probationers, parolees and jail inmates
were counted as described in the methodology section. The
District’s prisoner count consists of BJS-reported figures for
the period 1982-2000, of BOP-reported Superior Court
sentenced prisoners for the period 2002-2007, and of an
average of the BJS 2000 figure and the BOP 2002 figure for the
year 2001. Because the 2002-2007 BOP figures would have
overlapped with the BOP data on prisoners by reported home
state address (see Methodology Notes), this latter category of

38

inmates was excluded from DC’s prisoner calculation. Also
excluded from the District’s counts are Federal District Court
sentenced prisoners in the BOP and an anomalous figure
reported to the BJS of prisoners held in local jails in 2000.

Maine: Maine does not have a reported parole population in
the BJS parole survey from 1985-1990. For this report, this void
was filled by a straight-line average of the state’s 1984 and 1991
figures.
Maryland: Treatment programs such as the Substance
Abusing Offender Program and the Urinalysis and Treatment
Program did not exist in 1988 but are reflected in the 2008
expenditure figures. Prison treatment costs include medical
services, which in 1988 were approximately $15.3 million
across the Division of Corrections. In 2008, these costs were
contracted out and totaled approximately $107.2 million in
expenditures, all falling under contractual services.
Michigan: Maps of Michigan’s correctional population
were prepared by the Justice Mapping Center, Inc.
(www.justicemapping.org). Geographic data on standing
populations of the state’s prisoners (as of May 20, 2008),
parolees (as of May 28, 2008) and probationers (as of July 15,
2008) was provided by the Michigan Department of
Corrections; on the state’s county jail inmates (average daily
populations for 2007) by the JPIS report from the Michigan
Department of Corrections and by the Wayne County
Sheriff’s office; on the state’s federal prisoners by the Bureau

Public Safety Performance Project | Pew Center on the States

J U R I S D I C T I O N A L N OT E S
of Prisons (as of December 21, 2008); and on the state’s
community supervised, federal custody offenders by the
Eastern and Western Districts of Michigan, Administrative
Office of U.S. Courts (as of January 7, 2009). The 2008 prison
costs include juveniles adjudicated as an adult or youthful
trainee, which was not the case in 1998. Michigan’s figure for
corrections’ share of general fund spending (22 percent in
FY2008) is not comparable with similar figures from other
states, because in 1994, Michigan separated its K-12
education system into a different fund.
Minnesota: Prison costs only include prisons operated by
the Department of Corrections and contracted facilities, and
exclude private prison costs in Minnesota. Probation and
parole figures were provided by the respondent from the
Department of Corrections that is in charge of probation
and parole supervision for 55 of Minnesota’s 87 counties.
This respondent was able to provide budget subsidy totals
provided by the state legislature through the Minnesota
Community Corrections Act. This figure excludes local
funding of probation and parole but does capture a large
portion of probation and parole spending in Minnesota, and
it is consistent across reported time periods.
Missouri: Prison costs include juveniles sentenced as adults.
All personnel costs exclude fringe benefits which are paid
separately for all state employees. In 2007 and 2008, the State
Office of Administration assumed control of budgets for
maintenance functions and information systems from other
state agencies. Missouri officials made adjustments to the
2008 per offender costs to account for this difference.
Probation and parole costs in 2008 include two additional
community release centers and six community supervision
centers (totaling $13,035,480).
Montana: Probation and parole spending in 1983 included
37 prerelease beds and juvenile aftercare, both of which were
removed from the budget by 2008. Personnel costs in 1983
were included in operating costs, but by 2008 they became a
separate line item included in the overall budget. In 1983, the
alcohol and drug abuse treatment programs had separate
budgets from probation and parole, all of which fell under
the Department of Institutions. In 2008, probation and parole
budgets included all community corrections alcohol and
drug programs. The state’s survey respondent reported that
there were broad changes in budgeting as certain costs were
added to probation and parole spending and others were
moved to other agencies including, for example, the removal
of probation and parole costs for juvenile supervision.
New York: All personnel figures exclude fringe benefits such as
health insurance and retirement benefits. These costs are
handled in a different fund. Local assistance funds are included
for parole and probation figures in 2008, but only probation in
1983. Prison budgets include $300 million in capital costs for
2008 and $75.3 million in capital costs in 1983.
North Dakota: Probation and parole budgets for both
reported time periods are included in a Field Services
category which includes the five divisions of administration,
victim services, interstate compact, security and supervision,
and treatment.
Ohio: Prison costs exclude non-expense items (e.g.,
transfers) and capital costs. 1983 figures include two-thirds
of central office costs. On July 1, 2007, Ohio implemented a

new accounting system, the Ohio Administrative
Knowledge System (OAKS), which brought about some
changes to accounting categories, but the state’s survey
respondent indicated that this shift should not affect survey
responses. Reported parole figures are from the Parole and
Community Services division which is the parent agency for
Adult Parole Authority. These figures include some
probation costs for mostly rural portions of the state.
Oklahoma: Oklahoma’s 2007 probation figure is missing
from the annual BJS report. For this report, this void was
filled by applying the 2005-2006 rate of growth (which was
negative) to the year end 2006 figure.
Oregon: Probation and parole budgets in 1983 included
misdemeanors that are not included in 2008 probation and
parole figures. In 1983, probation and parole offices were
operated by the state, and in 2008 all but two jurisdictions
were operated by counties through state funding received
as an intergovernmental block grant. This change gives the
counties more flexibility in allocating the funds. Probation
and parole costs were separated by the survey respondent.
Pennsylvania: According to state sentencing laws, inmates
with maximum sentences of less than two years are subject
to the courts’ paroling authority and are typically supervised
by county adult probation departments. Data for these
jurisdictions are included in the state’s figures. The state’s
survey respondent indicated that, on average, approximately
15-19 percent of the supervised population is comprised of
these special probation referrals from the courts.
Pennsylvania probation figures are for the supervision of
probationers by county adult probation departments.
Rhode Island: The total adult prison spending amount does
not include administration costs such as finance, human
resources, and information technology charges. Earlier
probation and parole budget figures do not include
expenditures for electronic monitoring. Rhode Island
maintains a unified state jail and prison system and, per
reporting to the BJS, has in this report a single figure for its
incarcerated population.
Texas: All personnel figures exclude employee benefits,
which are budgeted through other state agencies. All prison
figures exclude inmates held in private prisons.
Vermont: In 1994, the prison budget did not allocate central
administration and management costs to facilities, and all
treatment costs were centrally administered. In 2008, such
prison costs are included. The state’s survey respondent
noted that Vermont moved in 1999 toward private prison
facilities, and these costs are not included. Probation and
parole costs were separated by the state’s survey
respondent. Vermont maintains a unified state jail and prison
system and, per reporting to the BJS, has in this report a
single figure for its incarcerated population.
Wyoming: The 1983 probation and parole cost figures
include juveniles placed under supervision by the court.
Probation and parole treatment costs for 2008, but not 1983,
include substance abuse assessments, cognitive behavioral
programming, and supportive services associated with drug
courts. The prison costs for 2008 have increased due to
private sector charges for medical and mental health services.
Wyoming total correctional cost figures were reported by the
state Legislative Service Office, January 2009.

One in 31: The Long Reach of American Corrections

39

TA B L E A - 1

National Correctional Populations, 1982-2007
2007
2006
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994
1993
1992
1991
1990
1989
1988
1987
1986
1985
1984
1983
1982

Rate of
Correctional Control:
1 in X

Total Correctional
Population

Probation

Parole

Jail

Prison

31
31
32
32
32
32
32
33
33
33
35
36
37
38
39
40
41
43
45
49
52
55
59
65
70
77

7,328,200
7,211,400
7,051,900
6,995,100
6,924,500
6,758,800
6,581,700
6,445,100
6,340,800
6,134,200
5,734,900
5,490,700
5,342,900
5,148,000
4,948,300
4,765,400
4,537,900
4,350,300
4,057,800
3,715,800
3,461,400
3,241,100
3,013,100
2,690,700
2,476,800
2,194,400

4,293,163
4,237,023
4,166,757
4,143,792
4,120,012
4,024,067
3,931,731
3,826,209
3,779,922
3,670,441
3,296,513
3,164,996
3,077,861
2,981,022
2,903,061
2,811,611
2,728,472
2,670,234
2,522,125
2,356,483
2,247,158
2,114,621
1,968,712
1,740,948
1,582,947
1,357,264

824,365
798,202
780,616
771,852
769,925
750,934
732,333
723,898
714,457
696,385
694,787
679,733
679,421
690,371
676,100
658,601
590,442
531,407
456,803
407,977
355,505
325,638
300,203
266,992
246,440
224,604

780,581
766,010
747,529
713,990
691,301
665,475
631,240
621,149
605,943
592,462
567,079
518,492
507,044
486,474
459,804
444,584
426,479
405,320
395,553
343,569
295,873
274,444
256,615
234,500
223,551
209,582

1,512,576
1,492,973
1,448,344
1,421,345
1,390,279
1,367,547
1,330,007
1,316,333
1,287,172
1,224,469
1,176,564
1,127,528
1,078,542
990,147
909,381
850,566
792,535
743,382
683,367
607,766
562,814
526,436
487,593
448,264
423,898
402,914

See methodology and state notes sections for definitions and exceptions.
Sources include the Bureau of Justice Statistics (correctional populations) and the Pew Center on the States (1 in X figures, based upon analysis of data from the U.S. Census State
Population Estimates and Bureau of Justice Statistics).
Total correctional population counts are not equal to the sum of probation, parole, jail and prison counts due to offenders with dual status.

40

Public Safety Performance Project | Pew Center on the States

TA B L E A - 2

State and National Correctional Spending
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan 1
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming 2

FY2008 Total General
Fund Corrections Spending,
in millions

FY2008 Corrections as a
Percent of State General
Fund Spending

$420
$240
$951
$348
$9,657
$625
$699
$200
n/a
$2,819
$1,100
$228
$207
$1,363
$669
$353
$341
$521
$625
$153
$1,192
$1,250
$2,178
$460
$266
$575
$169
$179
$253
$101
$1,581
$277
$2,871
$1,254
$65
$1,794
$491
$763
$1,836
$185
$487
$81
$675
$2,958
$330
$116
$1,254
$917
$181
$1,076
$103

2.5%
4.7%
9.5%
8.0%
9.3%
8.6%
4.3%
6.1%
n/a
10.0%
5.9%
4.3%
7.3%
6.1%
5.3%
6.0%
5.6%
5.5%
6.4%
4.9%
8.2%
4.6%
22.0%
2.6%
6.4%
6.8%
8.6%
5.1%
7.9%
6.8%
4.8%
4.6%
5.4%
6.2%
5.4%
7.3%
7.0%
10.6%
6.7%
5.5%
6.6%
7.1%
5.5%
6.8%
5.7%
9.4%
7.6%
6.3%
4.7%
8.0%
5.7%

2008
2007
2006
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994
1993
1992
1991
1990
1989
1988

State General Fund
Corrections Spending,
in millions

Corrections as a Percent
of State General Fund
Spending

$47,335
$43,904
$40,078
$38,239
$35,744
$35,285
$34,364
$33,571
$32,195
$29,733
$27,021
$25,440
$24,847
$23,251
$20,062
$17,435
$16,504
$15,890
$14,453
$12,887
$11,744

6.9%
6.7%
6.7%
6.9%
7.0%
7.2%
6.9%
6.9%
7.1%
7.1%
6.9%
6.8%
6.9%
6.7%
6.2%
5.7%
5.6%
5.7%
5.5%
5.3%
5.2%

All cost figures from the National Association of State Budget Officers, State Expenditure Reports.
FY2008 figures are estimates.
1
On Michigan corrections' share of general fund spending, see state notes.
2
Wyoming cost figures reported by State Legislative Service Office, January 2009.

One in 31: The Long Reach of American Corrections

41

TA B L E A - 3

State Correctional Populations, Year End 2007
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming

Rate of
Correctional
Control: 1 in X

Total
Correctional
Population

Probation

Parole

Jail

Prison

Federal Prisoners and
CommunitySupervised Offenders

32
36
33
29
36
29
33
26
21
31
13
32
18
38
26
54
53
35
26
81
27
24
27
26
38
36
44
44
48
88
35
35
53
38
63
25
42
33
28
26
38
40
40
22
64
46
46
30
68
39
38

108,843
14,005
144,221
73,193
755,256
128,186
82,655
25,082
22,892
462,435
562,763
31,620
63,231
252,776
181,459
42,294
39,275
91,993
122,207
12,852
156,776
206,241
278,808
152,319
56,208
125,613
16,997
30,195
40,172
11,628
191,473
42,197
282,215
181,435
7,885
351,879
65,720
89,589
346,268
31,250
88,352
15,211
117,428
797,254
29,023
10,622
129,681
165,725
21,065
110,642
10,631

51,745
6,416
76,830
31,676
353,969
77,635
57,493
16,696
6,485
274,079
435,361
19,426
48,663
142,790
126,562
22,776
16,131
42,510
39,006
7,853
98,470
175,419
182,706
127,797
21,623
56,240
9,106
18,910
13,461
4,650
126,390
20,774
119,963
111,446
4,468
254,898
26,038
43,732
176,987
26,137
42,721
5,870
56,179
434,309
10,829
7,059
51,954
118,885
7,890
53,230
5,358

7,790
1,544
6,807
19,388
123,764
11,086
2,177
535
5,569
4,654
23,111
2,110
3,114
33,354
10,362
3,546
4,842
12,741
24,085
32
13,856
3,209
21,131
4,744
2,015
19,849
966
800
3,653
1,653
15,043
3,527
53,669
3,311
342
17,575
2,349
22,658
78,107
462
2,433
2,812
10,496
101,748
3,597
936
6,850
13,017
1,830
16,986
706

15,401
66
15,743
6,229
82,662
13,871
0
0
2,900
64,547
45,732
0
3,852
20,408
15,540
3,699
7,022
18,337
33,627
1,838
13,632
13,394
18,100
8,085
11,617
10,639
2,304
3,151
7,231
1,757
19,627
8,345
28,400
17,464
960
20,560
9,748
6,661
35,347
0
13,137
1,456
23,590
67,885
6,854
0
27,583
12,137
3,628
13,931
1,577

27,816
5,167
37,700
13,307
171,500
22,666
20,924
7,276
6,606
97,072
49,337
5,978
6,744
45,215
25,130
8,732
8,696
14,545
20,461
2,222
23,282
11,300
50,190
8,950
17,479
29,857
2,940
4,505
13,245
2,891
25,359
6,350
62,602
37,970
1,368
50,731
23,957
13,925
45,969
4,018
23,862
3,256
19,248
159,016
5,223
2,145
32,972
17,410
4,907
23,028
2,028

6,091
812
7,141
2,593
23,361
2,928
2,061
575
1,332
22,083
9,222
4,106
858
11,009
3,865
3,541
2,584
3,860
5,028
907
7,536
2,919
6,678
2,743
3,474
9,028
1,681
2,829
2,582
677
5,054
3,201
17,581
11,244
747
8,115
3,628
2,613
9,858
633
6,199
1,817
7,915
34,296
2,520
482
10,322
4,276
2,810
3,467
962

See methodology and state notes for definitions and exceptions.
Sources include the Bureau of Justice Statistics (probation, parole and prison populations, December 31, 2007), the Federal Bureau of Prisons, the Administrative Office of U.S. Courts
and the Pew Center on the States (jail populations and 1 in X figures, based upon analysis of data from the U.S. Census State Population Estimates and Bureau of Justice Statistics).
Total correctional population figures may exceed total correctional population due to offenders with dual status.

42

Public Safety Performance Project | Pew Center on the States

TA B L E A - 4

Adult Incarceration Rates (Jail and Prison)
2007

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

1982

Rank

1 in X

Percent of
Adults

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51

50
55
69
70
71
75
76
82
83
83
88
88
89
89
90
92
94
97
97
98
100
102
102
103
104
105
108
109
110
111
111
115
118
120
121
132
133
136
140
140
143
148
154
155
179
187
190
204
204
211
226

2.00%
1.81%
1.44%
1.42%
1.41%
1.33%
1.32%
1.22%
1.21%
1.21%
1.14%
1.14%
1.13%
1.13%
1.11%
1.08%
1.06%
1.03%
1.03%
1.02%
1.00%
0.98%
0.98%
0.97%
0.96%
0.95%
0.92%
0.92%
0.91%
0.90%
0.90%
0.87%
0.85%
0.84%
0.82%
0.76%
0.75%
0.74%
0.72%
0.71%
0.70%
0.68%
0.65%
0.64%
0.56%
0.53%
0.53%
0.49%
0.49%
0.47%
0.44%

1 in X

Percent of
Adults

74
205
247
169
215
208
275
186
190
226
209
224
270
171
298
391
330
394
308
272
415
309
243
191
401
283
448
437
211
327
420
314
457
386
446
303
348
486
408
564
424
294
533
312
817
662
572
740
587
726
488

1.35%
0.49%
0.41%
0.59%
0.47%
0.48%
0.36%
0.54%
0.53%
0.44%
0.48%
0.45%
0.37%
0.58%
0.34%
0.26%
0.30%
0.25%
0.32%
0.37%
0.24%
0.32%
0.41%
0.52%
0.25%
0.35%
0.22%
0.23%
0.47%
0.31%
0.24%
0.32%
0.22%
0.26%
0.22%
0.33%
0.29%
0.21%
0.24%
0.18%
0.24%
0.34%
0.19%
0.32%
0.12%
0.15%
0.17%
0.14%
0.17%
0.14%
0.20%

Growth in Incarceration
Rate, 1982-2007
48%
272%
256%
141%
203%
176%
263%
127%
131%
173%
139%
154%
205%
93%
232%
324%
252%
307%
217%
176%
314%
204%
137%
86%
285%
169%
314%
300%
93%
195%
280%
173%
287%
223%
267%
130%
162%
258%
192%
303%
197%
99%
247%
101%
357%
254%
200%
264%
188%
243%
116%

Calculations based on data from the U.S. Census State Population Estimates, the Bureau of Justice Statistics Correctional Surveys, the U.S. Bureau of Prisons and the Pew Public
Safety Performance Project. See methodology notes for details.

One in 31: The Long Reach of American Corrections

43

TA B L E A - 5

Adult Community Supervision Rates
(Probation and Parole)
2007
Rank
Georgia
Idaho
Massachusetts
Minnesota
Rhode Island
Ohio
Texas
Indiana
District of Columbia
Michigan
Maryland
Washington
Pennsylvania
Delaware
Arkansas
Colorado
Oregon
Connecticut
Hawaii
New Jersey
Louisiana
Florida
Illinois
Arizona
California
Missouri
Alabama
Kentucky
New Mexico
North Carolina
Vermont
Wisconsin
Alaska
Wyoming
Nebraska
South Dakota
Tennessee
Montana
South Carolina
New York
Iowa
Mississippi
Oklahoma
Virginia
Kansas
North Dakota
Nevada
Utah
Maine
West Virginia
New Hampshire

1 in X

1982
Percent of
Adults

1 in X

Percent of
Adults

Growth in Supervision
Rate, 1982-2007

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16

15
21
28
30
31
32
32
35
35
37
37
37
37
38
41
41

6.50%
4.71%
3.58%
3.37%
3.26%
3.16%
3.14%
2.89%
2.82%
2.70%
2.69%
2.68%
2.68%
2.63%
2.43%
2.42%

48
186
163
114
104
183
52
158
43
179
53
49
129
82
238
137

2.09%
0.54%
0.61%
0.88%
0.96%
0.55%
1.92%
0.63%
2.34%
0.56%
1.90%
2.04%
0.78%
1.22%
0.42%
0.73%

212%
775%
483%
284%
238%
479%
64%
356%
20%
384%
42%
32%
245%
117%
476%
232%

17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51

43
44
45
46
48
50
54
55
56
56
57
57
58
58
60
60
61
63
64
64
68
70
71
82
83
86
92
94
96
98
106
120
126
131
155

2.32%
2.25%
2.23%
2.17%
2.07%
2.00%
1.86%
1.83%
1.79%
1.78%
1.76%
1.74%
1.73%
1.71%
1.67%
1.66%
1.63%
1.59%
1.56%
1.56%
1.47%
1.44%
1.40%
1.22%
1.20%
1.17%
1.09%
1.06%
1.04%
1.02%
0.95%
0.83%
0.79%
0.76%
0.64%

97
68
112
111
129
124
109
121
96
142
158
141
170
84
83
149
151
208
109
192
228
172
110
135
175
183
135
180
112
327
107
100
253
376
246

1.03%
1.47%
0.89%
0.90%
0.78%
0.81%
0.92%
0.82%
1.04%
0.70%
0.63%
0.71%
0.59%
1.20%
1.20%
0.67%
0.66%
0.48%
0.91%
0.52%
0.44%
0.58%
0.91%
0.74%
0.57%
0.55%
0.74%
0.55%
0.89%
0.31%
0.94%
1.00%
0.40%
0.27%
0.41%

125%
53%
149%
141%
166%
148%
102%
122%
71%
153%
178%
146%
194%
43%
39%
146%
147%
231%
71%
200%
236%
148%
54%
64%
110%
113%
47%
91%
17%
235%
1%
-17%
100%
186%
58%

Calculations based on data from the U.S. Census State Population Estimates, the Bureau of Justice Statistics Correctional Surveys, the Administrative Office of U.S. Courts and the
Pew Public Safety Performance Project. See methodology notes for details.
Population changes between 1982 and 2007 result both from changes in the true supervised populations and changes in survey instruments. In particular, the Bureau of Justice
Statistics amended the annual probation survey to include probationers under local jurisdiction (i.e. not under state jurisdiction). This change in definition may, for some states,
result in an artificially inflated growth figure.

44

Public Safety Performance Project | Pew Center on the States

TA B L E A - 6

Adult Correctional Control Rates
(Jail, Prison, Probation and Parole)
2007

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

1982

Growth in Control
Rate, 1982-2007

Rank

1 in X

Percent of
Adults

1 in X

Percent of
Adults

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16

13
18
21
22
24
25
26
26
26
26
26
27
27
28
29
29

7.92%
5.71%
4.82%
4.56%
4.10%
4.03%
3.89%
3.85%
3.80%
3.79%
3.77%
3.67%
3.65%
3.58%
3.46%
3.41%

37
128
27
42
127
116
79
98
106
90
59
41
110
99
102
134

2.68%
0.78%
3.69%
2.38%
0.79%
0.86%
1.27%
1.02%
0.94%
1.11%
1.69%
2.42%
0.91%
1.01%
0.98%
0.74%

196%
633%
31%
91%
420%
366%
207%
278%
304%
241%
123%
51%
301%
253%
251%
358%

17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51

30
31
32
32
33
33
33
35
35
35
36
36
36
38
38
38
38
38
39
40
40
42
44
44
46
46
48
53
53
54
63
64
68
81
88

3.33%
3.22%
3.15%
3.09%
3.08%
3.07%
3.03%
2.88%
2.85%
2.83%
2.81%
2.77%
2.76%
2.65%
2.62%
2.61%
2.61%
2.61%
2.57%
2.52%
2.49%
2.41%
2.28%
2.26%
2.19%
2.16%
2.07%
1.89%
1.88%
1.85%
1.58%
1.57%
1.48%
1.23%
1.14%

42
74
90
90
74
59
79
87
108
104
97
90
69
128
60
70
105
83
111
130
124
90
125
87
108
73
66
93
87
132
234
83
226
167
184

2.36%
1.34%
1.12%
1.11%
1.36%
1.69%
1.27%
1.14%
0.92%
0.97%
1.03%
1.11%
1.46%
0.78%
1.67%
1.44%
0.95%
1.21%
0.90%
0.77%
0.81%
1.11%
0.80%
1.15%
0.92%
1.37%
1.52%
1.08%
1.15%
0.76%
0.43%
1.21%
0.44%
0.60%
0.54%

41%
140%
182%
177%
126%
82%
140%
152%
208%
193%
173%
150%
90%
239%
57%
82%
174%
116%
185%
228%
208%
117%
186%
97%
137%
58%
36%
75%
63%
144%
270%
30%
233%
106%
109%

Calculations based on data from the U.S. Census State Population Estimates, the Bureau of Justice Statistics Correctional Surveys, the U.S. Bureau of Prisons, the Administrative
Office of U.S. Courts and the Pew Public Safety Performance Project. See methodology notes for details.
Population changes between 1982 and 2007 result both from changes in the true supervised populations and changes in survey instruments. In particular, the Bureau of Justice
Statistics amended the annual probation survey to include probationers under local jurisdiction (i.e. not under state jurisdiction). This change in definition may, for some states,
result in an artificially inflated growth figure.

One in 31: The Long Reach of American Corrections

45

9 0 1 E S T R E E T , N W , 1 0 TH F L O O R • W A S H I N G T O N , D C 2 0 0 0 4
W W W. P E W C E N T E R O N T H E S TAT E S . O R G