September 2012

States Report Reductions in Recidivism

I

n many jurisdictions, state and local government
recidivism. As policymakers are under tremendous
-

Reductions in Statewide
Recidivism Rates for 2005
and 2007 Prison Releases

among people released from prison would save money

someone released from prison or jail will reoffend.
-

to reduce recidivism: focusing resources on individuals

Percentage
change in
recidivism rate*
for 2005 and
2007 releases

community supervision policies and practices; and
cate declines in statewide recidivism rates for adults
improvements.
report, nor is it an evaluation of any state’s recidivism

Kansas

-15%

289

Michigan

-18%

862

-9%

235

-11%

1,278

-8%

138

-11%

1,212

-6%

141

Mississippi
Ohio
Oregon
Texas
Vermont

or practice.

1. Marshall Clement, Matthew Schwarzfeld, and Michael Thompson,
The National Summit on Justice Reinvestment and Public Safety: Addressing
Recidivism, Crime, and Corrections Spending (New York: Council of State
Governments Justice Center, 2011).

number fewer
returned
to Prison
for the 2007
release grouP **

*Percentage change in recidivism rate is calculated by dividing
the percentage-point change by the initial recidivism rate, which
yields the percentage by which the recidivism rate changed.

-

nearest tenth of a percent, see Figure 1 in the Appendix.

-

individuals who returned to prison in the latest data-year from
the number that would have returned had the state not reduced
its recidivism rate. The number is calculated based on a single
release cohort, but if the number of people released and recidivism
rates were to remain constant, the number would also represent
an annualized average. See Figure 2 in the Appendix for further
illustration.
2. Results from a national survey on recidivism may be found in the
2011 report State of Recidivism: The Revolving Door of America’s Prisons, in
which the Pew Center on the States compares the three-year recidivism
rates for 1999 and 2004 prison releases. The report shows that, in
addition to the states highlighted in this brief, many other states have
also achieved recidivism reductions.

prison would have been even higher
reduced its recidivism rate.

Reductions and Strategies
Kansas3
-

Three-year
recidivism rate
for 2005 prison
releases

Three-year
recidivism rate
for 2007 prison
releases

Decline in
recidivism rate

39 percent

33 percent

15 percent

ing and workforce development services.
cies to design local strategies to reduce revocations; it
cessful completion of educational, vocational, and treatment programs.
-

-

-

“

One of my wardens constantly asks his staff, right down to the line staff, ‘What can we
do to reduce recidivism?’ This gets them thinking that reentry is an important part of
what they do…that they can do something to improve the likelihood that the people
who leave their custody are successful when they return home.”
Ray Roberts, Secretary of Kansas Department of Corrections

3. The Kansas Department of Corrections (KDOC), through the
recidivism as “the number of inmates released from the DOC during
a given calendar year who returned to a Kansas prison within 36
months after release, divided by the number of inmates released during
that calendar year.” Neither the KDOC nor ASCA includes rearrests in
recidivism calculations. Recidivism-related information is gathered

2

States Report Reductions in Recidivism

by analyzing data that is tracked for each released individual. Kansas
measures recidivism based at three junctures after release—at 12
months, 24 months, and 36 months. Separate rates are calculated
for those persons returning to KDOC with new sentences and those
returning with no new sentences. Kansas calculates its recidivism rate
based on the total number of releases.

Michigan4

Mississippi7

Three-year
recidivism rate
for 2005 prison
releases

Three-year
recidivism rate
for 2007 prison
releases

40 percent

33 percent

Decline in
recidivism rate

Three-year
recidivism rate
for 2005 prison
releases

Three-year
recidivism rate
for 2007 prison
releases

Decline in
recidivism rate

18 percent

31 percent

28 percent

9 percent

-

-

ate system overcrowding, reserving critical resources
-

-

services.
ments in recidivism rates, in part due to legislation
-

facilities and minimum-security camps.

improve its recidivism rate.

“

We know that the majority of those incarcerated will be rejoining society and their
successful reentry is as critical to public safety as a sentence served.… Effective prisoner
reentry is an important component of smart justice. Michigan’s prisoner reentry
program has been a major contributor to lower recidivism rates for the state.”
Governor Rick Snyder (R, MI)

4.
return to prison within three years of release for either a new prison
sentence or for a technical violation of parole conditions. Recidivism
analysis is based on follow-up data for three years after each individual’s
release, determining whether the offender returned to prison as a
Parole Violator Technical (PVT), Parole Violator New Sentence during the
parole term (PVNS), or with a new prison sentence after the expiration
of the parole term. Michigan calculates its recidivism rate based on the
total number of releases on parole to Michigan counties (i.e., excluding
paroles to other states or paroles to other jurisdictions’ custody).
5.
agency conducted an analysis of multiple release-year cohorts that
participated in the in-reach phase of the program. This internal analysis
examined outcomes against baseline expectations for the 1998

release-year cohort, which was the year before the Michigan Prisoner
Reentry program began.
6. See Figure 1 of the Appendix.
7.
state calculates recidivism by tracking the return to inmate status
for individuals who are placed on parole, Earned Release Supervision,
house arrest, or probation, or who are released unconditionally from
inmate status every calendar year. The rate does not distinguish
between individuals on community supervision who are returned
to inmate status for technical violations of the terms of their release
or those who return to inmate status for committing a new offense.
Mississippi calculates its recidivism rates based on the total number of
releases.

States Report Reductions in Recidivism

3

Ohio8

Oregon10

Three-year
recidivism rate
for 2005 prison
releases

Three-year
recidivism rate
for 2007 prison
releases

38 percent

34 percent

Decline in
recidivism rate

Three-year
recidivism rate
for 2005 prison
releases

Three-year
recidivism rate
for 2007 prison
releases

Decline in
recidivism rate

11 percent

30 percent

27 percent

8 percent

prevention, treatment, and intervention programs
drawing on evaluation outcomes from a series of stud-

ing assessments and effective interventions, planning
case supervision, and improving communication skills.

provide incentive funding for community corrections

criminogenic risk assessments.
-

and resources.
Over a longer period, Oregon’s recidivism rate fell from

strategies for reducing crime and recidivism, includplaced under supervision upon release from prison,
and improving reentry services for individuals returning from incarceration.

8.

in the recidivism rate. Recidivism for individuals on probation is

based on returns to incarceration for a new crime, prison sanction, or
technical violation of the conditions of parole supervision. Although an

recidivism rate by tracking release and admission cohorts within a
cohorts include those who were released to parole supervision or

year. Ohio calculates its recidivism rates based on the total number of
releases.
9. See Figure 1 of the Appendix.
10.
follow-up period, including individuals who are convicted of a new
felony but are not reincarcerated. Technical violations are not counted

4

States Report Reductions in Recidivism

cohorts exclude individuals released following a return to incarceration
for a revocation. Although an individual may be sentenced more than
once to probation supervision, each new probation sentence is counted
separately. Oregon calculates its recidivism rates based on the total
number of releases.
11. See Figure 1 of the Appendix.

Texas12

tem of graduated sanctions for parolees, allowing for
-

Three-year
recidivism rate
for 2005 prison
releases

Three-year
recidivism rate
for 2007 prison
releases

Decline in
recidivism rate

27 percent

24 percent

11 percent

to prison.

programs and alternatives to incarceration, including

Comparing Recidivism Rates
This brief focuses on comparing the change in an individual state’s recidivism rate from one period to another,

as opposed to comparing that rate to another state’s recidivism rate, or to the rate of recidivism nationally.14
methodology for calculating recidivism. For example, some state measurements of recidivism account only
for reincarceration, while others include reconvictions that do not result in a prison or jail sentence. In Texas,
for instance, parolees who are temporarily placed in an Intermediate Sanction Facility as an alternative to
rate.
In addition, the composition of each state’s prison population is distinct. Incarcerated populations can
vary by risk level. For example, a state that sentences to prison large numbers of people who are at low risk
of reoffending will logically have a lower recidivism rate than a state that uses its prison facilities for people

responsible for prison and jail operations and there is no county jail system.
Because of these and other factors, comparing recidivism rates from state to state or comparing a state
recidivism rate with the national average is discouraged. In addition, national recidivism data should be used
improvement or investment.

12.
previous criminal involvement. The state calculates separate recidivism
rates based on rearrest and reincarceration for its state prison, state
jail, treatment facility, and sanction facility populations. Individuals
who violate the conditions of their parole and are sanctioned to an
Intermediate Sanction Facility are not counted as recidivists; they
revocation to prison. The recidivism rate included in this report is for

reincarceration of releases from state prisons only, and it is calculated
by tracking reincarceration within a three-year period after release.
Texas calculates the recidivism rate for state prison reincarcerations
based on the total number of releases.
13. See Figure 1 of the Appendix.
14. The statistics provided in this report are focused on prison releases
only.

States Report Reductions in Recidivism

5

Vermont15

gramming, including increased capacity of commu-

Three-year
recidivism rate
for 2005 prison
releases

Three-year
recidivism rate
for 2007 prison
releases

Decline in
recidivism rate

44 percent

41 percent

6 percent

treatment option.

-

viding intensive community supervision.

States Everywhere Committed to Reducing Recidivism
In December 2011, the Council of State Governments Justice Center co-organized a national forum on reentry
and recidivism that was attended by leaders from all 50 states, in partnership with the Association of State
Correctional Administrators, the U.S. Department of Justice’s Bureau of Justice Assistance, the Public Welfare
Foundation, and the Pew Center on the States. Corrections directors, reentry coordinators, legislators, and
judges worked together at this historic event to develop concrete strategies to reduce recidivism in their states.
In the weeks after the event, leaders of the state departments of corrections in 43 of 50 states completed a
continued assistance in this area.
Nearly all of these states currently measure recidivism, with 80 percent producing annual recidivism reports
that show year-to-year trends.
More than 80 percent of the 43 states have developed or are currently developing a plan to reduce recidivism.
Leaders in 29 states have either already set a recidivism-reduction target or anticipate setting a target in the
near future.

reduction targets and connect with other states to share information about their successes and challenges in
reducing recidivism.

15.

incarceration and who return to prison within three years of release
for a conviction of a new offense or a violation of supervision resulting
in an incarceration sentence of at least 90 days. Although individuals

recidivism as a conviction for an offense committed after release from
incarceration. The state calculates its recidivism rate by tracking
individuals who are released after a sentence of more than one year of

6

States Report Reductions in Recidivism

16. See Figure 1 of the Appendix.

APPENDIX

Figure 1: Statewide Recidivism Rates for 2000 – 2008 Releases17
To provide a broader view, below is recidivism data gathered for 2000 to 2008 release cohorts.
2000

2001

2002

2003

2004

2005

2006

2007

2008

n/a

n/a

n/a

n/a

n/a

38.6%

34.2%

32.9%

33.7%

Michigan

43.5%

41.9%

39.3%

39.1%

40.8%

40.5%

36.4%

33.2%

31.5%

Mississippi

28.6%

29.7%

28.6%

30.3%

n/a

30.6%

29.4%

27.9%

27.7%

Ohio

39.0%

39.0%

38.8%

39.5%

38.9%

38.4%

36.4%

34.0%

31.2%

Oregon

30.9%

30.6%

29.0%

31.1%

31.5%

29.8%

28.2%

27.5%

27.7%

Texas

31.2%

28.2%

28.5%

28.2%

28.0%

27.2%

26.0%

24.3%

n/a

Vermont

45.5%

43.9%

46.2%

42.4%

43.2%

43.7%

45.0%

40.9%

n/a

Kansas

Figures in bold are used elsewhere in this document.

Figure 2: Reductions in Recidivism and Numbers Returned to Prison
Percentage-Point
change between
2005 and 2007
releases

Percentage change
in recidivism rate
for 2005 and 2007
releases*

Kansas

-5.7%

-14.8%

Michigan

-7.3%

-18%

Mississippi

-2.7%

-8.8%

Ohio

-4.4%

Oregon

number
released in 2007

number fewer
returned to
Prison for the 2007
release grouP**

5,063

289

11,805

862

8,608

235

-11.5%

29,042

1,278

-2.3%

-7.7%

5,987

138

Texas

-2.9%

-10.7%

41,808

1,212

Vermont

-2.8%

-6.4%

5,025

141

Figures in bold are used elsewhere in this document.
*Percentage change in recidivism rate is calculated by dividing the percentage-point change by the initial recidivism rate, which
yields the percentage by which the recidivism rate changed.
that would have returned had the state not reduced its recidivism rate. The number is calculated based on a single release cohort, but
if the number of people released and recidivism rates were to remain constant, the number would also represent an annualized
average.

17. The states discussed in the following tables use a tracking period of
three years after release from incarceration. For example, the recidivism
rate for 2000 releases was calculated using data from 2000 through

2003 and the rate for 2008 releases was calculated using data from
2008 through 2011.

States Report Reductions in Recidivism

7

Sources
Much of the data on statewide recidivism included in this report was provided by state departments of
corrections. Additional data came from State of Recidivism: The Revolving Door of America’s Prisons (Washington, DC:
The Pew Center on the States, 2011) and “Reforming a System: An Inside Perspective on How Ohio Achieved a
Record-Low Recidivism Rate” by Gary Mohr, Director of the Ohio Department of Rehabilitation and Correction
(National Reentry Resource Center Newsletter, March 12, 2012). The states featured in this report noted that they had
received support in developing and implementing recidivism-reduction strategies from various organizations,
including the Bureau of Justice Assistance, the Pew Center on the States, the Council of State Governments Justice
Center, and the National Institute of Corrections.

The Council of State Governments (CSG) Justice Center is
at the local, state, and federal levels from all branches of
government. The CSG Justice Center provides practical,
nonpartisan advice and consensus-driven, evidence-based
strategies to increase public safety and strengthen communities.
The CSG Justice Center’s Justice Reinvestment Initiative to address corrections spending and public safety is a
partnership with the Public Safety Performance Project of the Pew Center on the States and the U.S. Department of
Justice’s Bureau of Justice Assistance. These efforts have provided data-driven analyses and policy options to state
leaders in 16 states.
For more information, visit www.justicecenter.csg.org.

This project was supported by Grant No. 2010-MU-BX-K084 awarded by the
Bureau of Justice Assistance. The Bureau of Justice Assistance is a component

Bureau of Justice Assistance
U.S. Department of Justice

opinions in this document are those of the author and do not represent the
Assistance, please visit www.bja.gov.

The Pew Center on the States is a division of The Pew Charitable Trusts

approach to improve public policy, inform the public and stimulate civic
life. Launched in 2006, The Public Safety Performance Project helps states
and corrections that protect public safety, hold offenders accountable, and control corrections costs. For more
information, visit www.pewcenteronthestates.org.

8

States Report Reductions in Recidivism