Pennsylvania
Department of Corrections

Recidivism Report
2013

Tom Corbett

John E. Wetzel

Governor

Secretary

Recidivism Report
PRS Overview
The Bureau of Planning, Research and Statistics (PRS)
directs the maintenance of historical data records for
the Department and conducts ongoing planning,
research, and evaluation activities designed to
provide key decision-makers with quantitative data
analysis for use in formulating and evaluating
departmental policies and practices. Specific functions
performed by the Office include planning, program
evaluation, corrections research, data analysis, statistical
repository, policy development and analysis.

Kristofer Bret Bucklen, Director
Planning, Research and Statistics
STAFF

2 0 1 3
Report Preparation
Authors.......................................Nicolette Bell
Kristofer Bret Bucklen
Kiminori Nakamura
Joseph Tomkiel
Angelo Santore
Lorraine Russell
Robert Orth

Nicolette Bell……………………….Chief of Research &
Evaluation
Robert Flaherty .....................Chief of Data Analysis
& Management Support

Technical Support.......................Planning, Research
and Statistics Staff

Lois Good...............................Community Corrections
Analyst
Runglux Kuba.........................Information Coordinator
Dean Lategan.........................Population and
Projections Analyst
Stacey O’Neill.........................Security Data Analyst
Lorraine Russell......................Research and Data
Analyst
Angelo Santore.......................Reentry/Program
Measures Analyst
Joseph Tomkiel......................Research Manager
Jessica Campbell………………….Data Analyst

TECHNICAL ADVISOR
Kiminori Nakamura……………….Professor of Criminology
& Criminal Justice,
University of Maryland

For additional information contact
The Bureau of Planning, Research and Statistics at
(717) 728-4051 Fax (717) 728-4180

COMMONWEALTH OF PENNSYLVANIA
DEPARTMENT OF CORRECTIONS
February 8, 2013

I am pleased to present the Pennsylvania Department of Corrections’ 2013 Recidivism Report,
which we believe to be a landmark state recidivism study. This groundbreaking and comprehensive
study represents the keystone of the Corbett Corrections Reform initiative, establishing a “new
normal” in our criminal justice system by focusing on reducing crime. This report was produced by
staff from the department’s Bureau of Planning, Research, and Statistics. They are to be commended
for their work on this comprehensive report. The scope of this report is impressive, and sets the bar
high for future analysis of state recidivism rates.
The report presents a mixed picture of recidivism rates in Pennsylvania. While on the one hand
reincarceration rates are going down, rearrest rates have been flat or slightly rising. For the most part,
recidivism rates have remained virtually unchanged over at least the past decade in Pennsylvania.
While this is disappointing, it also presents an opportunity. Over the past year, under the leadership of
Governor Corbett, fundamental transformations to Pennsylvania’s criminal justice system have been
enacted into law as a part of the administration’s Justice Reinvestment Initiative (JRI). In the Corbett
Corrections Reform initiative, population and cost, although both remain essential measurements, will
not be the sole numbers. The “new normal” is to expect and require quantifiable results. Citizens of
the Commonwealth should have every expectation of a corrections system that actually helps people
correct themselves; one that is based on research, not on anecdotal stories and innuendo. Changes
resulting from JRI are expected to significantly improve public safety, reduce recidivism, and lower
correctional costs for the citizens of the Commonwealth in the years to come. I view this report as the
first step towards measuring our progress in reaching these goals. Make no mistake; crime reduction
will always be the benchmark for performance measurement when we talk about recidivism reduction
efforts. As such, this report is our baseline for going forward.
The details of this report are worth exploring. Some truly innovative measures of recidivism
are provided, such as the fraction of total arrests in Pennsylvania that are attributable to ex-offenders
released from state prison, an analysis of the degree to which ex-offenders specialize in certain crime
types when they reoffend, and an analysis of recidivism rates by geographic location. A section is also
included which provides estimates of the potential cost savings for various recidivism reduction
scenarios.
Continued...
OFFICE OF THE SECRETARY
1920 Technology Parkway | Mechanicsburg, PA 17050 | 717.728.4109 | Fax 717.728.4178 | www.cor.state.pa.us

A special section of this report also examines recidivism rates for our Community Corrections
Center (CCC) system. This section is really an update to a previous analysis of the CCC system
provided in a study conducted by Dr. Edward Latessa at the University of Cincinnati in 2009. The
findings here are largely consistent with Dr. Latessa’s previous findings. We know from this updated
analysis that we have a lot of work to do to improve outcomes in our CCC system. Fortunately, many
of the legislative changes accomplished through JRI are specifically targeted towards improving the
CCC system. Again, this report sets the baseline for going forward, as we focus our CCC system
around performance-based recidivism reduction outcomes.
At the Pennsylvania Department of Corrections we believe that one of the most fundamental
methods for accomplishing our goals of less crime, less prison population, and less taxpayer costs, is
to utilize timely, accurate, and reliable data to guide policy. A scientific, data-driven approach offers
similar benefits to the field of corrections as it does to other fields of practice such as medicine, for
improving lives and saving money. I believe we also have an obligation to provide data and
evaluation in a public and transparent manner. This report reflects such an approach.
The report also benefited tremendously from our partnership with Dr. Kiminori Nakamura, a
professor in the Criminology & Criminal Justice department at the University of Maryland. Dr.
Nakamura was a co-author on this report, and also served as a technical advisor. We have been
working with Dr. Nakamura over the past year, under a researcher-practitioner partnership grant
through the National Institute of Justice. Under this grant, Dr. Nakamura is on loan from his
university on a part-time basis, as an “embedded criminologist” in our department. He serves as a
partner and a general scientific advisor, not just with this study but with all of our research efforts. I
thank him for his role in this report.
We trust that you find this report useful and informative. We also hope that this report will
generate some significant discussions surrounding the implications of its findings for recidivism
reduction policy.
Lastly, I want to thank the entire staff at the Pennsylvania Department of Corrections, for their
ongoing work and dedication towards improving the safety of the citizens of the Commonwealth of
Pennsylvania.
Sincerely,

John E. Wetzel
Secretary of Corrections
JEW/KBB/dls

2013 Recidivism in Pennsylvania
Table of Contents
Introduction……………………………………………………………………………………………………………………..……..
Section 1: Recidivism Rate Trends………………………………………………………………………………………

1
5

3-Year Recidivism Rates…………………………………………………………………………………………………………..

5

2000-2010 Recidivism Rates…………………………………………………………………………………………………….

6

Time to Reincarceration ………………………………………………………………………………………………………….
5-Year Recidivism Rates…………………………………………………………………………………………………………..

9
10

20-Year Long View of Reincarceration Rates……………………………………………………………………………

11

Section 2: Recidivism Rates by Geographic Areas………………………………………………………...

12

County……………………………………………………………………………………………………………..........................

12

Metropolitan Area…………………………………………………………………………………………………………………..

15

Section 3: Recidivism Rates by Demographics…………………………………………………………..……

17

Gender…………………………………………………………………………………………………..………………………………..

17

Race/Ethnicity………………………………………………………………………………………………………………………….

17

Age Groups………………………………………………………………………………………………………………………………

18

Prior Arrests/Incarcerations…………………………………………………………………………………………………….

19

Section 4: Recidivism Rates by Crime Types.…………………………………………………………………....

20

Commitment Crime Types……………………………………………………………………………………………………….

21

Rearrest Crime Types………………………………………………………………………………………….…………………..

22

Section 5: Recidivism Crime Type Specialization...…………………………………………………….…..

23

Section 6: Recidivism Rates by Type of Release……………..…………………..……………………….….

24

Section 7: Recidivism as a Fraction of Total Arrests in Pennsylvania………………….

25

Section 8: Cost of Recidivism………………………………………………………………………………………….……

26

Section 9: Community Corrections Recidivism…….……………………………………………………….

27

Parole Release Type…………………………………………………………………………………………………………………

27

Modeled Recidivism…………………………………………………………………………………………………………………

30

Parole to Center Recidivism Compared to Street……………………………………………………………………..

32

Recidivism by Vendor………………………………………………………………………………………………………………

34

Appendix A—Technical Definitions of Recidivism/Data Sources……………………..……..
Appendix B—End Notes……………………………………………………………………………………………………..…..

37
38

2013 Recidivism in Pennsylvania
Table of Contents
Tables
Table 1: 2000-2010 Rearrest Rates…………………………………………………………………………………………. 6
Table 2: 2000-2010 Reincarceration Rates………………………………….………………………………………….. 7
Table 3: 2000-2010 Overall Recidivism Rates………………….………………………………………………………. 8
Table 4: 5-Year Recidivism Rates………………….………………………………………………….……………………… 10
Table 5: 20-Year Long View of Reincarceration………………….……………………………………………………. 11
Table 6: Top 10 Counties with Highest Rearrest Rates………………….…………………………………………. 12
Table 7: Top 10 Counties with Highest Reincarceration Rates………………….……………………………… 13
Table 8: Top 10 Counties with Highest Overall Recidivism Rate………………….…………………………… 14
Table 9: Metropolitan Areas with Highest 3-Year Rearrest Rates…………….……………………………… 15
Table 10: Metropolitan Areas with Highest 3-Year Reincarceration Rates……………….……………… 15
Table 11: Metropolitan Areas with Highest 3-Year Overall Recidivism Rates……….………………….. 16
Table 12: 3-Year Recidivism Rates by Commitment Crime Type for 2008 Releases…………………. 21
Table 13: 3-Year Recidivism Rates by Aggregate Crime Category for 2008 Releases……………….. 21
Table 14: Breakdown of 3-Year Rearrests by Rearrest Crime Type for 2008 Releases……………… 22
Table 15: 3-Year Rearrest Crime Type Specialization (2008 Releases)……………………………………... 23
Table 16: 3-Year Recidivism Rates by Type of Release for 2008 Releases………………………………… 24
Table 17: 3-Year Recidivism Rates by Type of Parole Release for 2008 Releases…………………….. 24
Table 18: Breakdown of 3-Year Reincarceration by Type of Return for 2008 Parole Releases…. 24
Table 19: 2010 Pennsylvania Arrests Attributable to Released Inmates.……….………………………… 25
Table 20: 2010 Pennsylvania Arrest Rates (per 100.000) for Released Inmates and General
Population……………………..……………………………………………………………………………………….. 25
Table 21: Cost Savings by Reduction in 1-Year Reincarceration Rate………………………………………. 26
Table 22: Cost Savings by Reduction in PA DOC Admissions of Previously Released
Inmates…………………………………………………………………………………………………………………… 26
Table 23: Rearrest Rates by Parole Release Type…………………………………………………………………….. 27
Table 24: Reincarceration Rates by Parole Release Type…………………………………………………………. 28
Table 25: Overall Recidivism Rates by Parole Release Type…………………………………………………….. 29
Table 26: Modeled Overall Recidivism Rates by Parole Release Type……………………………………… 30
Table 27: Modeled Overall Recidivism Rates by Six Month Survival Time……………………………….. 31
Table 28: Rearrest Rates by Vendor….…………………………………………………………………………………….. 34
Table 29: Reincarceration Rates by Vendor…………………………………………………………………………….. 35
Table 30: Overall Recidivism Rates by Vendor…………………….…………………………………………………… 36

2013 Recidivism in Pennsylvania
Table of Contents
Figures
Figure 1: Pennsylvania’s Recidivism Flow………………………………………………………………………………… 4
Figure 2: 2000-2008 3-Year Recidivism Rates …………………………………………………………………………. 5
Figure 3: 2000-2010 Rearrest Rates..………………………………………………………………………………….…… 6
Figure 4: 2000-2010 Reincarceration Rates…………………………………………………………………………….. 7
Figure 5: 2000-2010 Overall Recidivism Rates…………………………………………………………………………. 8
Figure 6: 3-Year Reincarceration Rates by Time to Reincarceration Event (2008 Releases)…….. 9
Figure 7: 5-Year Recidivism Rates (2006 Releases)………………………………………………………………….. 10
Figure 8: 20 Year Long View of Reincarceration Rates…………………………………………………………….. 11
Figure 9: 3-Year Rearrest Rates by county in Pennsylvania……………………………………………………… 12
Figure 10: 3-Year Reincarceration Rates by County in Pennsylvania……………………………………….. 13
Figure 11: 3-year Overall Recidivism Rates by County in Pennsylvania……………………………………. 14
Figure 12: 3-Year Recidivism Rates by Gender………………………………………………………………………… 17
Figure 13: 3-Year Recidivism Rates by Race/Ethnicity……………………….………………………………….... 17
Figure 14: 3-Year Rearrest Rates by Age Group………………………………………………………………………. 18
Figure 15: 3-Year Reincarceration Rates by Age Group..………...…………………………………..………….. 18
Figure 16: 3-Year Overall Recidivism Rates by Age Group...…………………………………………..………… 18
Figure 17: 3-Year Overall Recidivism rates by Prior Arrests…………………………………………………….. 19
Figure 18: 3-Year Overall Recidivism Rates by Prior Incarcerations…………………………………………. 19
Figure 19: 3-Year Rearrest Rates as a Percent of Total Rearrests…………………………………………….. 22
Figure 20: Percent of Admissions in PA DOC Attributable to Recidivists………………………………….. 26
Figure 21: 1-Year Rearrest Rates by Parole Release Type………………………………………………………… 27
Figure 22: 1-Year Reincarceration Rates by Parole Release Type…………………………………………….. 28
Figure 23: 1-Year Overall Recidivism Rates by Parole Release Type (2005-2011 Releases)………. 29
Figure 24: 1-Year Overall Recidivism Rate Percentage Point Difference Compared to Parole to
Street………………………………………………………………………………………………………………..…… 32
Figure 25: 3-Year Rearrest Rate Percentage Point Difference Compared to Parole to the
Street (2008-2009 Releases)…………………………………………………………………………………… 34
Figure 26: 3-Year Reincarceration Rate Percentage Point Difference Compared to Parole to
the Street (2008-2009 Releases)………….…………………………………………………………………. 35
Figure 27: 3-Year Overall Recidivism Rate Percentage Point Difference Compared to Parole to
the Street (2008-2009 Releases)…………………………………………………………………………….. 36

2013 PA Recidivism Report

Recidivism in Pennsylvania
Introduction

One in 200 adult Pennsylvanians is currently
incarcerated in a Pennsylvania State Correctional
Institution. Ninety percent of the inmates currently
in a Pennsylvania state prison will eventually be
released. According to findings in this report, a
large proportion of those released will return to
some sort of offending behavior. This report
presents recidivism statistics for offenders released
from the custody of the Pennsylvania Department
of Corrections. Recidivism is measured by three
different methods in this report: rearrest,
reincarceration, and overall recidivism (see box
below for a description of each measure).

Highlights:









Recidivism Defined:



Rearrest is measured as the first instance of
arrest after inmates are released from state
prison.



Reincarceration is measured as the first
instance of returning to state prison after
inmates are released from state prison.
Overall Recidivism is measured as the first
instance of any type of rearrest or
reincarceration after inmates are released
from state prison.

Pennsylvania Department of Corrections



Approximately 6 in 10 released inmates
recidivate (are rearrested or reincarcerated)
within three years of release from prison.
Overall recidivism rates have been stable over
the last ten years.
Rearrest rates have been slowly increasing over
the last ten years.
Reincarceration rates peaked around 2005 and
began to decline in the most recent years.
Despite a drop starting in 2005, reincarceration
rates were slightly higher in the most recent
years than they were in 1990.
Offenders returning to urban areas are more
likely to be rearrested, however those returning
to rural areas are more likely to be
reincarcerated.
Dauphin County reports the highest overall
recidivism rates.
Released inmates do not appear to heavily
specialize in the same crime type when they
reoffend. The most specialized type of recidivist
is the property offender. The least specialized
type of recidivist is the violent offender.
Released inmates are more likely to be
reincarcerated (mostly for technical parole
violations) than rearrested during the first 18
months after release from prison, and thereafter
are significantly more likely to be rearrested.

1

2013 PA Recidivism Report
Highlights (Continued):
 More than half of those who return to prison
within three years after release will do so within
the first year of release. The first year is by far the
most risky period for recidivism.
 Younger released inmates are more likely to
recidivate than older inmates. A released inmate
who is under 21 at the time of release from prison
is more than twice as likely to recidivate within
three years than a released inmate who is over age
50 at the time of release from prison.
 Those with prior prison stays are more likely to
recidivate than those who have never been in state
prison. A released inmate who has already served
one or more times in a state prison has around a
25 percentage point higher recidivism rate than
one who is released from state prison for the first
time.
 Those with more prior arrests are more likely to
recidivate than those with fewer prior arrests. A
released inmate who has 10 or more prior arrests
is greater than 6 times more likely to recidivate
than a released inmate who has no prior arrest
history other than the arrest for the current
incarceration.
 Property offenders are significantly more likely to
recidivate than other types of offenders.
 DUI, rape, and arson offenders have the lowest
recidivism rates. While the 3-year overall
recidivism rate for all offenders is 59.9%, the
overall rate for DUI is 38.4%, for rape is 49.3%, and
for arson is 46.3%. The highest overall recidivism
rates are for stolen property (79.6%), burglary
(72.5%), and kidnapping (73.2%).

Pennsylvania Department of Corrections

















Nearly three-fourths of the rearrest offenses
committed by released inmates within three years
after their release from prison are for less serious
(Part II) offenses. Half (51%) are for a drug or
property offense. Only 17% of all rearrests are for
violent offenses (1.3% for murder).
Approximately 10% of all arrests in Pennsylvania
during 2010 were arrests involving released
inmates who had previously (in the last 10 years)
served time in state prison.
Per capita arrest rates for violent crimes are 14
times higher among released inmates than among
the general public.
Inmates who are released under parole supervision
are more likely to be reincarcerated, however, less
likely to be rearrested for a new offense than their
counterparts who complete their maximum
sentence (max outs).
Nearly two-thirds of all reincarcerations within
three years of release from prison are for technical
parole violations.
Those released inmates who are paroled after
failing parole at least once in the past have a
recidivism rate of about 12 percentage points
higher than those who are released onto parole for
the first time.
PA DOC can save approximately $44.7 million
annually by reducing its 1-year reincarceration rate
by 10 percentage points.
PA DOC can save approximately $16.5 million
annually by reducing admissions to state prison
who are recidivists by 10 percentage points.

2

2013 PA Recidivism Report
Highlights (Continued):
 Overall recidivism rates for released inmates who
transition through a Community Corrections
Center (CCC) have generally declined since 2005.
 In most recent years, the rearrest rates for
released offenders who are paroled to a Center are
lower than for those who are paroled directly
home (“to the street”), whereas reincarceration
rates and overall recidivism rates are higher for
those who are paroled to a Center compared to
those who are paroled directly home (“to the
street”).
 After accounting for other important differences
which may affect whether a released inmate is
paroled to a Center versus paroled directly home,
those paroled to a Center still demonstrate a
higher overall recidivism rate than those paroled
directly home (65.7% vs. 61.2% respectively, for
the most recent 3-year overall recidivism rates).
 Among those released offenders who survived at
least six months in the community without
recidivating, those who spent their first 3 to 6
months in a Center had a significantly lower 1-year
overall recidivism rate than those who were
paroled directly home (15% vs. 18%).

Pennsylvania Department of Corrections

3

2013 PA Recidivism Report
Figure 1: Pennsylvania’s Recidivism Flow

Release Types:
Parole: Inmates released from state prison to
serve the rest of their sentence on parole.
Max Out: Inmates released from state prison
after serving their maximum sentence.
Reincarceration Types:
Technical Parole Violation (TPV): A TPV occurs
when a parolee violates a condition of his/her
parole that is not necessarily an illegal act (i.e.,
entering a bar or not reporting to an agent).
Convicted Parole Violation (CPV): A CPV occurs
when a parolee violates a condition of parole
that is also against the law (i.e., using drugs).
New Court Commitment: A new court
commitment occurs when a released inmate is
arrested, convicted in court, and is sentenced to
prison for a new criminal charge.

Pennsylvania Department of Corrections

Figure 1 depicts a typical recidivism flow for
Pennsylvania’s state correctional system. PA DOC can
release inmates through two mechanisms: parole and
max out. Released inmates can return to PA DOC
through a technical parole violation (TPV), a convicted
parole violation (CPV), or as a new court commitment
(see box on the left for the explanations of different
release and reincarceration types).
Those who are paroled can return to prison
through a TPV, a CPV, or a new court commitment. A
parolee can be rearrested without being
reincarcerated, and conversely can be reincarcerated
without being rearrested.
Those who are released from prison by
maxing out their sentence can only return to prison
after they are arrested for a new crime, convicted,
and sentenced to prison through a court. Note that a
released inmate who is rearrested is not always
reincarcerated. But if reincarceration in state prison
is the given sentence for the arrest, the recidivist will
then be reincarcerated with PA DOC and will be
paroled or max out again after serving new time.
4

2013 PA Recidivism Report
SECTION 1: Recidivism Rate Trends
Figure 2: 2000-2008 3-Year Recidivism Rates
Overall Recidivism Rate

Reincarceration Rate

Rearrest Rate

70%
65%

Recidivism Rates

60%
55%
50%
45%

40%
35%
2000

2001

2002

2003

2004

2005

2006

2007

2008

Year of Release

Figure 2 shows a comparison of 3-year
recidivism rates for inmates released between 2000
and 2008. Those released from prison who were
reincarcerated or rearrested within three years of
their release date were included in these measures.
The 3-year reincarceration rate peaked at 49.3% in
2005 and declined to 43.0% in 2008. The 3-year
rearrest rates have been consistently higher than the
reincarceration rates. The 3-year rearrest rate has
grown from 47.2% in 2000 to 50.7% in 2008.
The 3-year overall recidivism rate has
remained relatively stable over the eight years
shown. In the latest year (2008), 70.6% of the overall
recidivism measure consisted of rearrest events,
while reincarceration events accounted for the other
29.4%.
Pennsylvania Department of Corrections

5

2013 PA Recidivism Report
From 2000 to 2008, the rearrest rates for
released inmates in Pennsylvania grew slightly.
However, according to Table 1, in 2010, the 6-month
and 1-year rearrest rates declined (12.3% and 23.7%,
respectively). The 2008 3-year rearrest rate was
50.7%. The 6-month rearrest rate peaked in 2009
(14.5%), the 1-year rearrest rate peaked in
2007/2008 (25.9%), and the 3-year rearrest rate
peaked in 2005 (51.1%).
Figure 3 depicts the 6-month, 1-year, and
3-year rearrest rates for inmates released from
Pennsylvania state prisons from 2000 to 2010. The
3-year rearrest rate has been more than double the
1-year rate in most years.

Table 1: 2000 - 2010 Rearrest Rates
Year of
Release

Rearrest Rates
6-Month

1-Year

3-Year

2000

12.4%

23.0%

47.2%

2001

12.9%

23.8%

47.6%

2002

12.7%

23.3%

48.2%

2003

12.0%

23.0%

48.4%

2004

12.7%

23.6%

49.1%

2005

13.8%

25.1%

51.1%

2006

13.2%

25.1%

50.6%

2007

13.9%

25.9%

50.4%

2008

14.2%

25.9%

50.7%

2009

14.5%

25.4%

N/A

2010

12.3%

23.7%

N/A

Figure 3: 2000-2010 Rearrest Rates
6-Month

1-Year

3-Year

60%

Rearrest Rates

50%

40%
30%
20%
10%

0%
2000

2001

Pennsylvania Department of Corrections

2002

2003

2004
2005
2006
Year of Release

2007

2008

2009

2010

6

2013 PA Recidivism Report
Table 2 shows the reincarceration rates of
Pennsylvania inmates released between 2000 and
2010. The reincarceration rates rose during the first
half of the decade and declined slightly in the second
half, although, the 6-month (12.0%) and 1-year
(22.5%) reincarceration rates in 2010 increased
slightly. The 2008 3-year reincarceration rate was
43.0%, the lowest in the previous eight years. Given
that the 3-year reincarceration rates have generally
tracked the 6-month and 1-year reincarceration rates,
it is likely that the 3-year reincarceration rate may
increase for those released in 2009 and 2010.
The 6-month, 1-year, and 3-year
reincarceration rates are depicted in Figure 4. The
reincarceration rates usually doubled from six months
to one year. After one year, the reincarceration rates
seemed to slow down, given that the 3-year
reincarceration rates typically are not quite double
the 1-year rates of the same year.

Table 2: 2000—2010 Reincarceration Rates
Reincarceration Rates

Year of
Release

6-Month

1-Year

3-Year

2000

12.5%

24.0%

45.9%

2001

13.9%

25.8%

46.3%

2002

13.0%

24.9%

45.4%

2003

13.7%

26.1%

47.1%

2004

13.9%

27.2%

48.0%

2005

16.3%

29.2%

49.3%

2006

14.6%

26.3%

46.0%

2007

12.5%

23.4%

43.9%

2008

11.2%

22.0%

43.0%

2009

10.7%

20.1%

N/A

2010

12.0%

22.5%

N/A

Figure 4: 2000-2010 Reincarceration Rates
6-Month

1-Year

3-Year

60%

Reincarceration Rates

50%
40%
30%
20%
10%
0%
2000

2001

Pennsylvania Department of Corrections

2002

2003

2004
2005
2006
Year of Release

2007

2008

2009

2010

7

2013 PA Recidivism Report
According to Figure 5, the overall recidivism
rates for inmates released from state prison in
Pennsylvania between 2000 and 2010 appear
strikingly steady. In 2010, the 6-month overall
recidivism rate declined slightly (20.0%) while the 1year overall recidivism rate was also slightly down at
35.0%. The 2008 3-year overall recidivism rate was
62.0%. The 6-month overall recidivism rate peaked
in 2001 (22.6%), the 1-year overall recidivism rate
peaked in 2001 and again in 2005 (38.6%), and the 3year overall recidivism rate peaked in 2005 (64.4%).
See Table 3 for the full breakdown of the overall
recidivism rates.
Over the ten-year span, approximately 64% of
the first recidivism events have been a rearrest while
only 36% have been a reincarceration.

Table 3: 2000 - 2010 Overall Recidivism Rates
Overall Recidivism Rates

Year of
Release

6-Month

1-Year

3-Year

2000

21.4%

37.2%

63.4%

2001

22.6%

38.6%

63.1%

2002

21.7%

37.7%

63.1%

2003

21.0%

37.6%

63.0%

2004

20.7%

37.5%

63.8%

2005

22.2%

38.6%

64.4%

2006

20.2%

36.4%

62.7%

2007

19.9%

35.8%

62.2%

2008

20.6%

37.0%

62.0%

2009

20.8%

35.5%

N/A

2010

20.0%

35.0%

N/A

Figure 5: 2000-2010 Overall Recidivism Rates
6-Month

1-Year

3-Year

70%

Overall Recidivism Rates

60%
50%
40%
30%
20%
10%
0%
2000

2001

Pennsylvania Department of Corrections

2002

2003

2004
2005
2006
Year of Release

2007

2008

2009

2010

8

2013 PA Recidivism Report
Figure 6: 3-Year Reincarceration Rates by Time To Reincarceration (2008 Releases)
1,600

1,400

Number of Reincarcerated Inmates

1,200

25%
return
within
5 months

1,000

800

50% return
within 10
months

600

400

75% return within 19 months
200

0
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
Months to Reincarceration

Figure 6 displays the number and proportion of
recidivism events among those who are
reincarcerated within 3 years from release. The
overall declining curve suggests that those who
return to prison tend to do so relatively soon after
their release. According to Figure 6, over half of the
inmates released in 2008 who were reincarcerated
within three years were reincarcerated within 12
months of their release. In fact, more than 1,000
inmates were reincarcerated per month during
each month, through month 12 after release. Three
quarters of the inmates released in 2008 who were
reincarcerated within three years were returned to
prison in approximately 19 months.

Pennsylvania Department of Corrections

9

2013 PA Recidivism Report
Figure 7: 5-Year Recidivism Rates in Pennsylvania (2006 Releases)
Reincarceration Rate

Rearrest Rate

Overall Recidivism Rate

70%

Recidivism Rates

60%
50%
40%
30%

20%
10%

5 Year

Q3

Q2

Q1

4 Year

Q3

Q2

Q1

3 Year

Q3

Q2

Q1

2 Year

Q3

Q2

Q1

1 Year

Q3

Q2

Q1

0%

Time Since Release
(Quarterly increments)

Figure 7 shows the cumulative recidivism
rates for inmates released in 2006, over a five year
period of time since release. The reincarceration
rates are slightly higher than the rearrest rates in the
first year and a half after release. At the second year
mark, the rearrest rates surpass and remain higher
than the reincarceration rates.
Table 4: 5-Year Recidivism Rates
Reincarceration
Rate

Rearrest
Rate

Overall
Recidivism
Rate

1 Year

26.3%

25.1%

36.2%

2 Year

39.2%

40.2%

53.1%

3 Year

46.0%

50.7%

62.7%

4 Year

50.0%

57.7%

68.6%

5 Year

52.8%

60.7%

71.1%

Pennsylvania Department of Corrections

According to Table 4, after the first year
period, the reincarceration rate is 26.3%, the rearrest
rate is 25.1% and the overall recidivism rate is 36.2%
for the inmates released in 2006. After three years,
the reincarceration rate is 46.0%, the rearrest rate is
50.7%, and the overall recidivism rate is 62.7%.
Slightly more than half of those who recidivated
(rearrested or reincarcerated) within three years
actually recidivated within the first year. This shows
the slowing rate of recidivism as time since release
elapses. Finally, the 5-year reincarceration rate is
52.8%, rearrest rate is 60.7%, and the overall
recidivism rate is 71.1%. The 5-year recidivism rates
increased from the 3-year rates by only a small
increment, indicating a further slow-down of
recidivism rates as the time since release grows
longer. This slow down can be seen in Figure 7 as
the slopes of the recidivism lines increasingly flatten
over time.

10

2013 PA Recidivism Report
Figure 8: 20-Year Long View of Reincarceration Rates

60%

1-Year

3-Year

Reincarceration Rates

50%

40%

30%

20%

10%

0%
1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

2010

Year of Release

Taking a longer view, from 1990 to 2010,
reincarceration rates have remained fairly stable in
Pennsylvania, ranging from 20% to 29% for inmates
reincarcerated within one year, and 41% to 50% for
those reincarcerated within three years of their
release from state prison (see Figure 8).1 Both 1year and 3-year rates had a peak in 1994 and
trough in 1996. After another peak in 2005,
reincarceration rates began to decline from 2005 to
2009, reaching a low in 2009 with a 1-year rate of
20.1%. However, in 2010, the 1-year rate increased
by almost 10%, suggesting that an upward trend in
reincarceration rates may be occuring, given that
the 3-year rates appear to follow the trends of the
1-year rates historically.

Pennsylvania Department of Corrections

Table 5: 20-Year Long View of Reincarceration
Inmates Reincarcerated
Year of Inmates
Release Released

1-Year

3-Year

Number

Rate

Number

Rate

1990

6,702

1,461

21.8%

2,788

41.6%

1992

8,057

2,023

25.1%

3,766

46.7%

1994

8,523

2,360

27.7%

4,306

50.5%

1996

7,049

1,493

21.2%

2,939

41.7%

1998

8,927

2,048

22.9%

3,807

42.6%

2000

10,934

2,628

24.0%

5,015

45.9%

2002

11,030

2,744

24.9%

5,012

45.4%

2004

13,913

3,780

27.2%

6,680

48.0%

2006

13,762

3,625

26.3%

6,328

46.0%

2008

13,814

3,042

22.0%

5,944

43.0%

2010

16,764

3,767

22.5%

N/A

N/A

11

2013 PA Recidivism Report
SECTION 2: Recidivism Rates By Geographic Areas
Table 6 shows the ten counties with the
highest 3-year rearrest rates. The county designation
represents where the released inmate was originally
convicted before commitment to state prison.
Between 2006 and 2008, the average statewide 3year rearrest rate was 50.7%. The counties with the
larger populations such as Philadelphia, Allegheny,
Dauphin, Delaware, and Montgomery have some of
the highest rearrest rates, and drive up the 3-year
rearrest rate for Pennsylvania as a whole. In fact, the
median 3-year rearrest rate for Pennsylvania counties
was only 43%. The overall median rearrest rate can
be used as a benchmark to compare counties in
Pennsylvania (see Figure 9 for the 3-year rearrest
rates for all 67 Pennsylvania counties).

Table 6: Top 10 Counties with Highest
Rearrest Rates
3–Year Rearrests

County

2006-2008
Releases

Number

Rate

Philadelphia

10,394

6,249

60.1%

Carbon

61

36

59.0%

Dauphin

1,739

1,005

57.8%

Blair

349

196

56.2%

Montgomery

1,211

648

53.5%

Allegheny

2,826

1,482

52.4%

Delaware

1,363

701

51.4%

Perry

67

34

50.7%

York

1,297

641

49.4%

Beaver

276

135

48.9%

Figure 9: 3-Year Rearrest Rates by County in Pennsylvania

Pennsylvania Department of Corrections

12

2013 PA Recidivism Report
Table 7 shows the ten counties with the
highest 3-year reincarceration rates. Similar to Table
6, the county designation represents the county
where the released inmate was originally convicted
before commitment to state prison. The average
statewide 3-year reincarceration rate in Pennsylvania
between 2006 and 2008 was 43%. The median
reincarceration rate for all counties was 41%. In
contrast to the rearrest rates, which tended to show
higher rates for more populous counties, the
counties with the highest reincarceration rates are
mostly rural and relatively less populous. Figure 10
shows the 3-year reincarceration rates of all 67
Pennsylvania counties.

Table 7: Top 10 Counties with Highest
Reincarceration Rates
County

2006-2008
Releases

3–Year Reincarcerations
Number

Rate

Montour

30

16

53.3%

Clinton

79

42

53.2%

Lackawanna

809

421

52.0%

Bedford

80

41

51.3%

Lycoming

578

281

48.6%

Union

103

50

48.5%

Huntingdon

52

25

48.1%

Dauphin

1,748

827

47.3%

Franklin

450

210

46.7%

Lehigh

958

444

46.3%

Figure 10: 3-Year Reincarceration Rates by County in Pennsylvania

Pennsylvania Department of Corrections

13

2013 PA Recidivism Report
Table 8 shows the ten counties with the
highest 3-year overall recidivism rates. The
statewide average overall recidivism rate for
Pennsylvania between 2006 and 2008 is 62%, while
the median overall recidivism rate of Pennsylvania’s
67 counties is 54%. This discrepancy between the
statewide recidivism rates and the median county
rate suggests that more populous counties, such as
Dauphin, Philadelphia, and Allegheny tend to have
higher overall recidivism rates which drive up the
statewide rate. Figure 11 shows the 3-year overall
recidivism rates for all 67 counties.

Table 8: Top 10 Counties with Highest Overall
Recidivism Rate
County

2006-2008
Releases

3–Year Overall Recidivism
Number

Rate

Dauphin

1,739

1,171

67.3%

Philadelphia

10,394

6,811

65.5%

Allegheny

2,826

1,748

61.9%

Montgomery

1,211

747

61.7%

Blair

349

215

61.6%

Cambria

205

125

61.0%

Lycoming

607

369

60.8%

Lackawanna

896

543

60.6%

York

1,297

780

60.1%

Huntingdon

55

33

60.0%

Figure 11: 3-Year Overall Recidivism Rates by County in Pennsylvania

Pennsylvania Department of Corrections

14

2013 PA Recidivism Report
Table 9 shows the 3-year rearrest rates by
Pennsylvania metropolitan area2 for inmates released
in 2006 to 2008. Consistent with the rearrest rates by
county, the Philadelphia metropolitan area had the
highest 3-year rearrest rate of the 2006-2008
released inmates. The Harrisburg-Carlisle
metropolitan area rate was second. Rounding out the
top five metropolitan areas with the highest rearrest
rates are Altoona, York-Hanover, and Pittsburgh. The
top five metropolitan areas contain large
Pennsylvania cities.

Carlisle, York-Hanover, and Allentown are also
included in the five highest metropolitan areas
according to their 3-year reincarceration rates. As
shown on the previous map of incarceration rates by
county, these less populous metropolitan areas tend
to have higher reincarceration rates.

Table 10 shows that Williamsport had the
highest 3-year reincarceration rate for the 2006-2008
released inmates. Scranton-Wilkes-Barre, Harrisburg-

Table 9: 3-Year Rearrest Rates by
Metropolitan Areas

Metropolitan Area

2006-2008
Releases

Table 10: 3-Year Reincarceration Rates by
Metropolitan Areas

3-Year Rearrests
Metropolitan Area
Number

Rate

Philadelphia

14,398

8,248

57.3%

Harrisburg-Carlisle

2,059

1,159

Altoona

349

York-Hanover

2006-2008
Releases

3-Year
Reincarcerations
Number

Rate

56.3%

Williamsport
Scranton-Wilkes Barre

578
1,517

281
721

48.6%
47.5%

196

56.2%

Harrisburg-Carlisle

2070

945

45.7%

1,297

641

49.4%

York-Hanover

1,278

577

45.1%

Pittsburgh

4,916

2,408

49.0%

Allentown

1,755

776

44.2%

Williamsport

607

295

48.6%

Johnstown

194

82

42.3%

Allentown

1,806

852

47.2%

Lebanon

400

169

42.3%

Lancaster

856

389

45.4%

Pittsburgh

4808

2026

42.1%

Scranton-Wilkes Barre

1,658

737

44.5%

Altoona

339

141

41.6%

Johnstown

205

88

42.9%

Philadelphia

14084

5791

41.1%

Reading

1,667

701

42.1%

Reading

1,629

669

41.1%

Erie

1,424

573

40.2%

Erie

1,357

546

40.2%

Lebanon

419

168

40.1%

Lancaster

868

335

38.6%

State College

158

60

38.0%

State College

159

49

30.8%

Pennsylvania Department of Corrections

15

2013 PA Recidivism Report
Table 11: 3-Year Overall Recidivism Rate by
Metropolitan Area

Metropolitan Area

2006-2008
Releases

3-Year Overall
Recidivism
Number

Rate

Harrisburg-Carlisle

2,059

1,344

65.3%

Philadelphia

14,398

9,082

63.1%

Altoona

349

215

61.6%

Johnstown

205

125

61.0%

Williamsport

607

369

60.8%

York-Hanover

1,297

780

60.1%

Pittsburgh

4,916

2,912

59.2%

Scranton-Wilkes Barre

1,658

962

58.0%

Allentown

1,806

1,037

57.4%

Lancaster

856

457

53.4%

Reading

1,667

865

51.9%

Lebanon

419

212

50.6%

Erie

1,424

715

50.2%

State College

158

66

41.8%

Table 11 shows the 3-year overall recidivism rates of the 2006-2008 releases ranked by
metropolitan areas. The Harrisburg-Carlisle metropolitan area had the highest average 3-year
overall recidivism rate based on inmates released between 2006 and 2008, followed by
Philadelphia, Altoona, Johnstown, and Williamsport.

Pennsylvania Department of Corrections

16

2013 PA Recidivism Report
SECTION 3: Recidivism Rates By Demographics
Figure 12: 3-Year Recidivism Rates by
Gender
Female

Figure 13: 3-Year Recidivism Rates by
Race/Ethnicity
Black

Male

Hispanic

66.8%

63.2%
46.9%

51.8%
36.8%

Rearrests

28.4%

44.2%

Reincarcerations Overall Recidivism

Figure 12 shows 3-year recidivism rates by gender,
suggesting that men are at a higher risk of being both
rearrested and reincarcerated within three years of
their release from Pennsylvania state prison when
compared to women.

Pennsylvania Department of Corrections

White

57.4% 57.8%

56.9%
46.5% 44.7%

Rearrests

45.5%

42.0% 40.7%

Reincarcerations

Overall Recidivism

Figure 13 shows 3-year recidivism rates broken down
by race/ethnicity3, suggesting that Blacks report the
highest rates of rearrest rates and overall recidivism,
followed by Hispanics, and non-Hispanic Whites4. On
the other hand, reincarceration rates by race are
much more similar.

17

2013 PA Recidivism Report
Figure 14: 3-Year Rearrest Rates By Age

Group

63.7%

56.9%

51.4%

Figure 16: 3-Year Overall Recidivism Rates
by Age Group

77.9%

45.2%

67.3%

63.4%

38.8%

57.8%
37.3%

Under 21

21-29

30-39

40-49

Above 50

Figure 15: 3-Year Reincarceration Rates by
Age Group

Under 21

21-29

21-29

30-39

40-49

Above 50

The 3-year reincarceration rates of inmates
released in 2008 show a similar declining
reincarceration rate pattern with age, according to
Figure 15.
The 3-year overall recidivism rates by age
group follow the same declining pattern as with the
rearrest and reincarceration rates, according to
Figure 16.

63.1%
49.0%

Under 21

44.6%

30-39

37.4%

40-49

28.0%

Above 50

These age group findings suggest that age has
a strong negative correlation with recidivism. In
other words, the older an inmate is at the time of his/
her release, the less likely he/she is to recidivate.

Figure 14 shows the 3-year rearrest rates by
age at time of release, suggesting that younger age
groups5 are at the highest risk for recidivating. A 21
year old released inmate’s risk of being rearrested is
almost 25 percentage points higher than an over 50
year old inmate.

Pennsylvania Department of Corrections

18

2013 PA Recidivism Report
Figure 17: 3-Year Overall Recidivism Rates
by Prior Arrests

Figure 18: 3-Year Overall Recidivism Rates
by Prior Incarcerations

79.2%
64.4%

82.8%

83.1%

84.5%

1

2

3+

58.0%

41.9%

12.4%
0

1-4

5-9

10 +

NOTE: The number of priors does not include the current arrest.

0

NOTE: The number of priors does not include the current incarceration.

Prior criminal history appears to also be highly
associated with whether an inmate will continue to
commit crimes after being released from state prison.
Figures 17 and 18 show the overall recidivism rates of
inmates released in 2008 by the number of prior
arrests or incarcerations6, respectively. As depicted,
the general trend is that the risk of recidivating
increases with higher numbers of priors.

Figure 18 depicts a large jump in the 3-year
overall recidivism rate between inmates released
from Pennsylvania state prison for the first time (zero
prior incarcerations) and those released inmates who
had been incarcerated before (more than one prior
incarcerations). After an inmate is released from
Pennsylvania state prison with at least one prior, he/
she is more than 80% likely to be rearrested or
According to Figure 17, the risk of recidivating reincarcerated within three years of release.
within three years, by either rearrests or
reincarcerations, increases as the number of prior
arrests increases.

Pennsylvania Department of Corrections

19

2013 PA Recidivism Report
SECTION 4: Recidivism Rates By Crime Types
Table 12 depicts the 3-year recidivism rates
for inmates released in 2008, by the type of crime
committed that led to their original incarceration in a
Pennsylvania state prison. It is important to note that
inmates who recidivated were not necessarily
rearrested or reincarcerated for the same crime as
the original commitment crime.

crime is 39.6%, which is 4.5 percentage points below
the 3-year reincarceration rate for Part I crime. The
Part II 3-year overall recidivism rate is 58.1%, which is
4.5 percentage points lower than the overall
recidivism rate for Part I crimes. The Part II offenses
that had higher 3-year rearrest rates were: Other
Assault, Stolen Property, Forgery, Drug Offenses,
Weapons, and Prison Breach. The Part II offenses
The 3-year rearrest, reincarceration, and
that had higher 3-year reincarceration rates were:
overall recidivism rates for Part I crime are 48.6%,
Stolen Property, Forgery, Drug Offenses, Weapons,
44.1%, and 62.6%, respectively.
Prison Breach, and Part II Other. The Part II offenses
The Part I offenses with higher 3-year rearrest that had higher 3-year overall recidivism rates were:
and reincarceration rates were: Robbery, Aggravated Other Assault, Fraud, Stolen Property, Forgery, Other
Assault, Burglary, and Theft/Larceny. The Part I
Sexual Offenses, Weapons, Prison Breach, and
offenses that had higher 3-year overall recidivism
Kidnapping.
rates were: Robbery, Burglary, and Theft/Larceny.
Table 13 depicts the 3-year recidivism rates by
The 3-year rearrest rate for Part II crime is
aggregate crime categories for inmates released in
48.5%, very close to the 3-year rearrest rate for Part I 2008. Property crimes had the highest 3-year
crime. The 3-year reincarceration rate for Part II
recidivism rates for all three measures of recidivism.

Pennsylvania Department of Corrections

20

2013 PA Recidivism Report
Table 12: 3-Year Recidivism Rates by Commitment Crime Type for 2008 Releases
3-Year Rearrests
Offense Category
Part I
Murder/
Manslaughter
Forcible Rape
Robbery
Aggravated Assault
Burglary
Theft/Larceny
Arson
Total: Part I
Part II
Other Assault
Fraud
Stolen Property
Forgery
Statutory Rape
Other Sexual Offenses
Drug Offenses
Weapons
DUI
Prison Breach
Kidnapping
Part II Other
Total: Part II
Grand Total

3-Year Reincarcerations

3-Year Overall Recidivism

Number

Rate

Number

Rate

Number

Rate

144

33.0%

145

33.3%

227

52.1%

78
881
567
504
526
17
2,717

25.8%
52.8%
48.8%
52.6%
53.7%
21.3%
48.6%

71
806
516
457
449
21
2,465

23.5%
48.4%
44.4%
47.7%
45.9%
26.3%
44.1%

149
1050
700
695
639
37
3,497

49.3%
63.0%
60.2%
72.5%
65.3%
46.3%
62.6%

103
20
148
100
5
120
2,143
279
184
126
16
670
3,914
6,631

51.8%
47.6%
63.0%
49.5%
41.7%
31.8%
50.6%
60.0%
27.6%
62.4%
39.0%
48.4%
48.5%
48.6%

59
15
116
85
3
99
1,695
206
169
103
16
625
3,191
5,656

29.6%
35.7%
49.4%
42.1%
25.0%
26.3%
40.0%
44.3%
25.4%
51.0%
39.0%
45.2%
39.6%
41.4%

123
38
187
131
6
227
2,427
333
256
144
30
783
4,685
8,182

61.8%
90.5%
79.6%
64.9%
50.0%
60.2%
57.3%
71.6%
38.4%
71.3%
73.2%
56.6%
58.1%
59.9%

NOTE: The total 3-year reincarceration, rearrest, and overall recidivism rates do not match the 3-year rates presented at the beginning of the
report due to missing offense category data. Also, rearrest totals are missing 30 of the original incarceration offenses.

Table 13: 3-Year Recidivism Rates by Aggregate Crime
Category for 2008 Releases
Crime Category

Rearrest
Rate

Reincarceration
Rate

Overall Recidivism
Rate

Violent
Property
Drugs
Public Order/Other

45.6%
52.7%
50.6%
46.3%

40.9%
45.8%
40.0%
40.6%

59.9%
69.2%
57.3%
55.8%

Pennsylvania Department of Corrections

21

2013 PA Recidivism Report
Table 14: Breakdown of 3-Year Rearrests by
Rearrest Crime Type for 2008 Releases
Offense Category
Part I
Murder/
Manslaughter
Forcible Rape
Robbery
Aggravated
Assault
Burglary
Theft/Larceny
Arson
Total: Part I
Part II
Other Assault
Fraud
Stolen Property
Forgery
Statutory Rape
Other Sexual
Offenses
Drug Offenses
Weapons
DUI
Prison Breach
Kidnapping
Part II Other
Total: Part II
Grand Total

Rearrests

% of Total

84

1.3%

40
281

0.6%
4.2%

287

4.3%

278
804
5
1,779

4.2%
12.1%
0.1%
26.7%

230
107
290
8
0

3.5%
1.6%
4.4%
0.1%
0.0%

165

2.5%

1,931
299
585
166
6
1,095
4,882
6,661

29.0%
4.5%
8.8%
2.5%
0.1%
16.4%
73.3%
100.0%

inmates were rearrested for within three years of
their 2008 release from a Pennsylvania state prison.
Four out of 10 of the Part II rearrests were drug
offenses. Other significant Part II offenses that
released inmates were rearrested for DUIs and a
variety of other minor offenses (i.e., “Part II Other”).

Figure 19: 3-Year Rearrest Rates as a
Percent of Total Rearrests

Public
Order/
Other
32%

Violent
17%

Property
22%

Drugs
29%

Figure 19 depicts the percentage breakdown
of rearrest into aggregate crime categories: violent,
property, drugs, and public order/other8. The highest
percentage of rearrests occurred for Public Order/
Table 14 displays the crime type of the most
serious rearrest charge for inmates released in 2008 Other (32.3%), followed by Drug offenses (29.0%),
who were rearrested within three years. Part I crimes Property crimes (22.4%), and Violent crimes (16.3%).
accounted for 26.7% of the rearrests within 3 years of As mentioned previously, the types of crime that a
released inmate was rearrested for is not necessarily
release. Almost half of the crimes that released
inmates were rearrested for were in the theft/larceny the same type of crime that he/she was originally
category. Other common Part I crimes that released incarcerated for.
inmates were arrested for were: Aggravated Assault,
Burglary, and Robbery. Part II crimes accounted for
the other 73.3% of the crimes for which released
Pennsylvania Department of Corrections

22

2013 PA Recidivism Report
SECTION 5: Recidivism Rates By Crime Type
Specialization
Table 15: 3-Year Rearrest by Commitment and Rearrest Crime Types (2008 Releases)
Crime Type for Original
Commitment

Rearrest Crime Type
Violent

Property

Drugs

Public Order/
Other

No Rearrest

Violent
Property

13.1%
7.1%

9.0%
24.7%

10.4%
9.0%

12.7%
11.9%

54.8%
47.3%

Drugs
Public Order/Other

7.3%
7.7%

8.0%
11.2%

22.4%
12.3%

12.8%
16.8%

49.4%
52.1%

In this report, crime type specialization is
defined as the propensity for released inmates to be
rearrested for a crime type that is the same as the
crime type for the original commitment. Table 15
displays the combination of commitment crime types
(the rows) and the percentage of different rearrest
crime types (including the possibility of no rearrest
within three years). This allows us to examine what
proportion of those who were initially committed for
each of the four crime categories were rearrested for
the same crime category, or for a different category.
The values in the diagonals of the table (highlighted in
yellow) represent the proportion recidivating for the
same crime type as their commitment offense (i.e.,
specialists). The values in the off-diagonals represent
the proportion committing different crime types than
their commitment offense (i.e., non-specialists).
According to Table 15, some degree of
specialization seems to exist among the inmates
released in 2008. The tendency of specialization is
particularly stronger for property and drug crimes.
Released inmates who were originally incarcerated
for property crimes returned to property crimes at a

Pennsylvania Department of Corrections

24.7% rearrest rate, while rearrests for violent (7.1%),
drugs (9.0%), and public order/other (11.9%) crimes
were at lower rates. Released inmates who were
incarcerated for drug crimes returned to drug crimes
at a 22.4% rate, while violent (7.3%), property (8.0%),
and public order/other (12.8%) crimes were at
significantly lower rates.
Specialization is less evident in violent and
public order crimes. Those who were originally
incarcerated for violent crimes were rearrested for a
violent crime 13.1% of the time, a slightly higher rate
than the rates for public order/other (12.7%), drugs
(10.4%), and property (9.0%). Finally, inmates
originally incarcerated for public order/other crimes
returned to public order/other rearrests at 16.8%,
which is higher than the rates for drugs (12.3%),
property (11.2%), and violent (7.7%). Overall, this
specialization pattern of property and drug offenders
tend to have higher propensity to repeat similar
crimes is consistent with what has been found in
national recidivism studies. In general, released
inmates tend to generalize rather than specialize in
their recidivism crime types.

23

2013 PA Recidivism Report
SECTION 6: Recidivism Rates By Type of Release
Table 16: 3-Year Recidivism Rates by Type of
Release for 2008 Releases
Type of Release Reincarceration Rate Rearrest Rate
Parole
Max Out

50.5%
20.4%

47.1%
62.0%

Table 17: 3-Year Recidivism Rates by Type of
Parole Release for 2008 Releases
Type of Parole Reincarceration Rate

Rearrest Rate

Initial Parole

46.9%

43.7%

Reparole

58.6%

55.1%

Table 18: 3-Year Breakdown of Reincarceration
by Type of Return for 2008 Parole Releases
Type of Return

% of Total Returns

Technical Parole Violator

61.5%

Convicted Parole Violator

33.4%

New Court Commitment

5.0%

NOTE: Does not include 2008 Releases who maxed out their
sentences. All Max Out releases should return as new court commits
in the event that they return to prison.

Pennsylvania Department of Corrections

For an in-depth description of the recidivism
flow in the Pennsylvania state correctional system,
refer to Figure 1 on page 6. According to Table 16,
50.5% of the inmates released on parole in 2008
were reincarcerated within three years, while only
20.4% of the released inmates who maxed out in
2008 were reincarcerated within three years. The
higher reincarceration rate for paroled inmates is
likely due to violating the conditions of their parole,
since prisoners who max-out are not subject to such
conditions. Of the inmates released in 2008 who
were paroled, 47.1% were rearrested within three
years while 62.0% of those who maxed out were
rearrested within three years.
According to Table 17, 46.9% of the inmates
paroled in 2008 for the first time (initial parole) were
reincarcerated within three years of their release,
while 59.8% of the inmates paroled in 2008 for the
second or more time (reparole) were reincarcerated
within three years of their release. Of those paroled
in 2008, 43.7% paroled for the first time were
rearrested within three years, while 55.1% of those
paroled for the second or more time were
rearrested within three years.
Of the parolees who were reincarcerated
within three years of their 2008 release date, 61.5%
were returned as TPVs (see Table 18). Another
33.4% of reincarcerated parolees were returned as
CPVs. The remaining 5.0% were reincarcerated
through the court system as a new court
commitment.

24

2013 PA Recidivism Report
SECTION 7: Recidivism as a Fraction of Total Arrests
Table 19: 2010 Pennsylvania Crime Types by Released Offenders
Violent

Property

Drugs

TOTAL

Arrests of Released Inmates in 2010

2,506

4,661

5,087

12,254

Total Arrests in 20109

20,275

48,739

51,443

120,457

% of Arrests Attributable to Released Inmates

12.4%

9.6%

9.9%

10.2%

Table 19 depicts the most serious crimes per
arrest by released inmates in Pennsylvania as a percentage
of the total Part I10 arrests in Pennsylvania in 2010.
Inmates released from a Pennsylvania state prison
between 2000 and 2010 were included in this analysis.

Conversely, the violent crime arrest rate for the general
population in Pennsylvania was 205 per 100,000
individuals in 2010. These rates indicate that released
inmates were 14 times more likely to be arrested for a
violent crime in Pennsylvania in 2010 than individuals in
the general population. Following this logic, inmates
In Table 19, the 2010 crimes committed by
released from a Pennsylvania prison between 2000 and
released inmates are broken down into Violent (12.4%),
2010 were 11 times more likely to be arrested for property
Property (9.6%), and Drugs (9.9%) categories. These three
and drug crimes in 2010. Overall, inmates released from a
crime categories were used to produce an average of
Pennsylvania state prison between 2000 and 2010 were 12
10.2%, the best estimate for the total serious crime in a
times more likely to be arrested for a crime in 2010 than
year attributable to released inmates in Pennsylvania. The
the general population.
serious crimes included in the Violent category were
murder, manslaughter, forcible rape, robbery, and
Overall comparisons are misleading though.
aggravated assault. The serious crimes included in the
Inmates released in 2000 were far less likely to be arrested
Property category were burglary, larceny/theft, motor
in 2010 than inmates released in 2009. In fact, inmates
vehicle theft, and arson. All drug offenses were included
released in 2000 were only three times more likely to be
in the Drugs category.
arrested in 2010 than the general population. Conversely,
inmates released in 2009 were 17 times more likely to be
Table 20 shows the arrest rates of inmates
arrested in 2010 than the general population (18 times
released from a Pennsylvania state prison between 2000
more likely for violent crimes, 16 times for property
and 2010 as a ratio of the arrest rates for the general
crimes, and 17 times for drug crimes). This suggests that
civilian population at risk for arrest in the time frame. For
recidivism is mostly attributable to recently released
example, the violent crime arrest rate for released inmates
inmates, and the longer that released inmates remain
in 2010 was 2,905 per 100,000 released inmates.
arrest free, the less likely that they are to be rearrested.
Table 20: 2010 Pennsylvania Crime Types by Released Offenders As a Ratio of General
Population11

Arrest Rate for Released Inmates in 2010
Arrest Rate for General Population in 2010
Ratio (Released Inmate/General Public)
Pennsylvania Department of Corrections

Violent

Property

Drugs

TOTAL

2,905

5,403

5,896

14,203

205

492

519

1,216

14-to-1

11-to-1

11-to-1

12-to-1
25

2013 PA Recidivism Report
SECTION 8: Cost of Recidivism
Table 21: Cost Savings by Reduction in
1-Year Reincarceration Rate
1-Year
Reincarceration Rate

Annual Bed
Days

Annual Cost
Savings
(in millions)

Reduced by 1
Percentage Points

48,768

$0.8

Reduced by 5
Percentage Points

234,930

$15.0

Reduced by 10
Percentage Points

475,035

$44.7

Figure 20: Percent of Admissions In PA DOC
Attributable to Recidivists

First Time
Admissions
50.4%

Repeat
Admissions
49.6%

Table 22: Cost Savings by Reduction in
Admissions of Previously Released Inmates
Admissions of Released
Inmates

Annual Bed
Days

Annual Cost
Savings
(in millions)

Reduced by 1
Percentage Points

25,024

$0.4

Reduced by 5
Percentage Points

126,626

$3.9

Reduced by 10
Percentage Points

257,573

$16.5

Pennsylvania Department of Corrections

Table 21 shows the estimated annual cost
savings by reducing the 1-year reincarceration rate by
one, five, and 10 percentage points. The cost savings
were calculated by taking the released inmates who
were reincarcerated in 2010, from the 2009 and 2010
releases, and reducing their numbers to attain a
reincarceration rate of one, five, and 10 percentage
points lower. Based on a 10 percentage point
reduction in the 1-yr recidivism rate, the PA DOC would
save approximately 475,035 bed-days, or
approximately $44.7 million annual cost savings.
Further, a second calculation was performed to
estimate the annual cost savings that the PA DOC could
achieve by reducing the number of admissions of
inmates who had been previously released from a
Pennsylvania state prison (i.e., repeat offenders, or
recidivists). This is another useful way of looking at
population reduction and cost savings from recidivism
reduction. As depicted in Figure 20, approximately
49.6% of the total annual state prison admissions in
2010 were offenders who had previously served time
in a Pennsylvania state prison. Just slightly more than
half (50.4%) of the admissions in 2010 were first time
inmates.
Recidivists who are admitted to state prison
take up approximately 1.3 million bed-days in a given
year, at a cost of $121.2 million per year. If the
percentage of DOC admissions who had at least one
prior state prison admission was reduced by 10
percentage points (39.6% of admissions rather than
49.6%), this reduction in annual recidivist admissions
would result in an annual bed-day reduction of
approximately 257,573 beds, or an annual cost savings
of $16.5 million.

26

2013 PA Recidivism Report
Section 9: Community Corrections Recidivism
Table 23: Rearrest Rates by Parole Release Type
6-Month Rearrests
Release Year
2005-06 Releases
2008-09 Releases
2010-11 Releases

1-Year Rearrests

3-Year Rearrests

Parole to
Street
12.0%
12.2%

Parole to
Center
11.7%
10.0%

Parole to
Street
23.2%
23.4%

Parole to
Center
25.4%
21.8%

Parole to
Street
49.2%
48.1%

Parole to
Center
52.5%
47.1%

11.8%

8.9%

21.1%

17.7%

N/A

N/A

Figure 21: 1-Year Rearrest Rates by Parole Release Type
30%
25%
20%

15%
10%
5%
0%
2005-06 Releases

2008-09 Releases

Parole to Street

Community Corrections Centers (CCCs), also
known as halfway houses, provide a transitional
process by allowing residents monitored contact
with jobs and reentry services. The CCCs house
inmates granted parole by the Pennsylvania Board
of Probation and Parole. The PA DOC also
contracts with private vendors (CCFs) to provide
specialized treatment and transitional supervision
services, many in the area of substance abuse
programming.

Pennsylvania Department of Corrections

2010-11 Releases

Parole to Center

According to Figure 21, the 1-year rearrest
rates of releases who were paroled directly home (i.e.,
“to the street”) were lower than for those paroled to a
Community Corrections Center (CCC) in 2005 and
2006. From 2008 to 2011, the 1-year rearrest rates
were higher for those paroled to the street.
Table 23 shows that the 1-year rearrest rates
of those paroled to a CCC have declined over time.
The 1-year rearrest rate of 2005-06 releases paroled to
a CCC was 25.4%, while the 1-year rearrest rate was
17.7% for the 2010-11 releases. This trend did not
hold for those paroled to the street, whose 1-year
rearrest rates held steady and then declined for the
2010-11 releases.
27

2013 PA Recidivism Report
Table 24: Reincarceration Rates by Parole Release Type

Release Year

6-Month
Reincarcerations

1-Year
Reincarcerations

3-Year
Reincarcerations

2005-06 Releases

Parole to
Street
11.8%

Parole to
Center
18.1%

Parole to
Street
26.3%

Parole to
Center
36.0%

Parole to
Street
47.5%

Parole to
Center
58.7%

2008-09 Releases
2010-11 Releases

9.3%
9.8%

16.1%
19.3%

22.1%
22.5%

32.0%
33.5%

44.0%
N/A

53.3%
N/A

Figure 22: 1-Year Reincarceration Rates by Parole Release Type
40%
35%
30%
25%
20%
15%
10%
5%
0%
2005-06 Releases

2008-09 Releases

Parole to Street

According to Figure 22, the 1-year
reincarceration rates of releases from 2005 to 2011
for those who were paroled to the street were
consistently lower than for hose paroled to a CCC.
Also, the 1-year reincarceration rates seemed
to be declining over time, despite a slight increase for
the most recent releases. The 1-year reincarceration
rate of 2005-06 releases who were paroled to a CCC
was 36.0%, whereas the 1-year reincarceration rate
dropped to 33.5% for the 2010-11 releases to a CCC.
Mirroring this trend, the 1-year reincarceration rate
of 2005-06 releases paroled to the street was 26.3%,
whereas the 1-year rate dropped to 22.5% for the

Pennsylvania Department of Corrections

2010-11 Releases

Parole to Center

2010-11 releases. Table 24 shows the 6-month and 3year reincarceration rates for the same release years.
In each case, the reincarceration rates are higher for
those paroled to a CCC than for those paroled to the
street.

28

2013 PA Recidivism Report
Table 25: Overall Recidivism Rates By Parole Release Type
6-Month Overall
1-Year Overall
3-Year Overall
Recidivism
Recidivism
Recidivism
Release Year
Parole to Parole to Parole to
Parole to
Parole to
Parole to
Street
Center
Street
Center
Street
Center
2005-06 Releases

18.6%

22.8%

35.5%

42.6%

61.5%

68.6%

2008-09 Releases

17.4%

22.1%

33.8%

41.6%

59.7%

66.7%

2010-11 Releases

18.1%

24.0%

32.7%

40.5%

N/A

N/A

Figure 23: 1-Year Overall Recidivism Rates By Parole Release Type
50%

45%
40%
35%
30%
25%

2005-06 Releases

2008-09 Releases

Parole to Street

According to Figure 23, the 1-year overall
recidivism rates of releases from 2005 to 2011 who
were paroled to the street were consistently lower
than for those who were paroled to a CCC.

2010-11 Releases

Parole to Center

rates for the same release groups. In each case, the
overall recidivism rates have been higher for those
paroled to a CCC than for those paroled to the street.

Also, the overall recidivism rates seem to be
decline over time. The 1-year overall recidivism rate
for 2005-06 releases to a CCC was 42.6%. For 201011 releases to a CCC, the 1-year overall recidivism
rate decreased to 40.5%. Mirroring this trend, the 1year overall recidivism rate for 2005-06 releases to
the street was 35.5%, but for 2010-11 releases to the
street the 1-year rate dropped to 32.7%. Table 25
also shows the 6-month and 3-year overall recidivism

Pennsylvania Department of Corrections

29

2013 PA Recidivism Report
The descriptive comparison of recidivism
rates by parole release type in the previous pages is
informative, but the observed differences in the
recidivism rates may not represent statistically
significant differences and may be due to chance
variation or the influence of factors that vary
between those who are paroled to the street and
those who are paroled to a center which are not yet
accounted for. Table 26 shows the overall recidivism
rates by parole release type while controlling for
various important predictors of recidivism such as
age, race, prior criminal history, and risk score

(LSI-R)12. The differences in modeled recidivism rates
by parole release type essentially mirror the
descriptive differences in Table 25. Across the
various release years (2005-2006, 2008-2009, 20102011), the recidivism rates of those who are paroled
to a center are about 5 percentage points higher
than the rates of those who are paroled to the
street, despite the differences being narrower than
the descriptive differences in Table 25 as a result of
statistically accounting for the other factors
mentioned above (e.g., age, race, prior criminal
history, etc.).

Table 26: Modeled Overall Recidivism Rates by Parole Release Type
6-Month
1-Year
3-Year
Overall Recidivism
Overall Recidivism
Overall Recidivism
Release Year
Parole to
Parole to
Parole to
Parole to
Parole to
Parole to
Street
Center
Street
Center
Street
Center
2005-06 Releases

17.0%

20.2%

34.1%

39.5%

63.2%

67.9%

2008-09 Releases

16.4%

19.7%

33.1%

38.8%

61.2%

65.7%

2010-11 Releases

17.6%

22.6%

32.3%

38.1%

N/A

N/A

Pittsburgh CCC
Pennsylvania Department of Corrections

30

2013 PA Recidivism Report
The higher recidivism rates of those who are
paroled to a center do not necessarily indicate that
the parolee’s chance of recidivating increases as a
result of being sent to a center. It could indicate that
close monitoring provided by the centers (and to
some degree Parole staff) help detect violating
behaviors of parolees (criminal or otherwise) that
would remain undetected if parolees did not live in
centers. If this is true and centers essentially better
detect violating behaviors and remove high-risk
parolees from centers through arrests and
reincarcerations, then we might expect that those
parolees who are discharged from centers without
recidivism have lower recidivism rates. Also, those
who are successfully discharged from a center may
benefit from the programs and treatments they
receive while at the center. In order to examine this
possibility further, we compared the recidivism rates
of those who were discharged from a center and
stayed recidivism-free for at least six months after
their release from prison with those who were
paroled to the street and stayed recidivism-free for
at least 6 months.
The results in Table 27 show that among
those who remained recidivism-free for at least six
months, there was no statistically significant
difference in overall recidivism rates between
parolees who were assigned to a center and
discharged successfully and parolees who were
paroled to the street, both at one year after their
release from prison (19.0% vs. 18.0% respectively)
and three years after their release from prison
(53.0% vs. 52.0% respectively).13 We also looked at
whether the length of stay at a center matters to the
recidivism rates of parolees who were discharged
from a center and stayed recidivism-free for at least
six months. Again, the recidivism rates of those who

Pennsylvania Department of Corrections

were assigned to a center were statistically no higher
than the rates of those who were paroled to the
street, but those who stayed at a center for three to
six months actually had statistically lower recidivism
rates than those paroled to the street. The fact that
a longer stay at a center is associated with lower
recidivism rates than the rates of those paroled to
the street is consistent with the possibility that
centers efficiently detect and help sanction violations
and remove high-risk parolees so that those who are
successfully discharged from a center consist of
relatively low-risk parolees. Regardless of the
explanation, we were able to substantiate in this
analysis at least one comparison where those who
were paroled to a center had a lower recidivism rate
than those who were paroled directly to the street.

Table 27: Modeled Overall Recidivism Rates by
Six Month Survival Time
Parole Type

Overall Recidivism Rates
1-Year

3-Year

Parole To Center

19.0%

53.0%

< 1 Month

17.0%

60.0%

1 to <3 Months

19.0%

54.0%

3 to <6 Months

15.0%*

50.0%

18.0%

52.0%

Parole To Street

NOTE: Parole To Center 3 to <6 Months 1-Year Overall Recidivism
rate is significantly different from Parole To Street at p < .05

31

2013 PA Recidivism Report

SHARON CCC
RENEWAL2
KEENAN HOUSE/TT
CONEWAGO HARRISBURG
PHILADELPHIA CCC4
ATKINS HOUSE
MINSEC YORK STREET
KEYSTONE CORRECTIONAL
JOSEPH E. COLEMAN CENTER
HAZLETON TREATMENT CTR
MINSEC OXFORD
LIBERTY MANAGEMENT
KINTOCK-ERIE AVENUE
GATEWAY-SHEFFIELD
ADAPPT TREATMENT SERVICE
ALLE-KISKI PAVILION
CAPITOL PAVILION
JOHNSTOWN CCC
SCRANTON CCC
ERIE CCC
RENEWAL, INC.
PENN PAVILION
MINSEC OF SCRANTON
YOUTHBUILD/CRISPUS ATKS
GAUDENZIA-SIENA HOUSE
WERNERSVILLE CCC30
MINSEC BROAD STREET
GATEWAY-ALIQUIPPA
GATEWAY-BRADDOCK
SCRANTON CATH SOC SVCS
HARRISBURG CCC
ALLENTOWN CCC
GATEWAY-ERIE
ADCM
PHILADELPHIA CCC2

Figure 24: 1-Year Overall Recidivism Rate Percentage point Difference Compared to
Parole To Street

35
30
25
20
15
10
5

-5
-10
-15
-20

HANNAH HOUSE
CONEWAGO-WERNERSVILLE
PITTSBURGH CCC
SELF HELP MOVEMENT
YORK CCC
TRANSITIONAL LIVING CTR
GAUDENZIA-COMMON GROUND
DIAG & REHAB CENTER
GAUDENZIA-ERIE
WERNERSVILLE CCC27
MINSEC CHESTER

0

Figure 24 shows the overall recidivism rates
for all the individual Community Corrections Centers
(CCCs) and contracted facilities (CCFs) with more
than 10 parolees, in comparison to the recidivism
rate of those who are paroled to the street. By
setting the recidivism rate of the “parole to the
street” group at zero, the recidivism rates for the
centers are shown as the percentage points higher
or lower than the recidivism rates of parole to the
street, ordered from lowest to highest. Reflecting
Pennsylvania Department of Corrections

1-Year Overall
Recidivism Rate for
Parole to Street

the overall patterns in Table 25, only about a
quarter of the centers have lower recidivism rates
than those paroled to the street, and the majority of
centers have much higher recidivism rates than
those paroled to the street.

32

2013 PA Recidivism Report
The next three tables and figures (tables 28-30,
figures 25-27) show the recidivism rates for some of
the major contractors of community corrections
facilities in Pennsylvania, along with the recidivism rate
of state-run community corrections centers. The
recidivism rates are displayed by the type of recidivism
measure (rearrest, reincarceration, overall recidivism),
by the release year (2005-2006, 2008-2009, 20102011), and by the length of follow-up period (6 months,
1 year, 2 years). Aside from several contractors and the
state-run centers showing lower rearrest rates than
those parole to the street across different release years
and follow-up times, the contract facilities and the
state-run centers almost always show higher overall
recidivism rates.
There are several ways to display comparisons
between contractors and state-run centers in terms of
recidivism rates. One way is to look at the rank order
of contractors and state-run centers by recidivism rates
across different recidivism measures. For the 3-year
follow-up, Gateway and Minsec facilities tend to have
the highest recidivism rates for rearrest,
reincarceration, and overall recidivism based on the
2008-09 releases, as shown in figures 25-27.

Interestingly, CEC is one of the contractors with the
highest 3-year rearrest rates, but had the lowest
reincarceration rate among contractors and state-run
centers, although still higher than those who were
paroled to the street. Firetree and Renewal
consistently demonstrated fairly low recidivism rates
across recidivism measures, according to figures 25-27.
Another way to evaluate comparisons between
contractors and state-run centers in terms of recidivism
is to look at the relative change of recidivism rates over
time (across release years). For the 6-month and 1year overall recidivism rates, Gateway demonstrated
the largest increase in recidivism over time, whereas
Renewal demonstrated the largest decrease in
recidivism over time. Firetree also demonstrated a
large increase in overall recidivism over time, at least
for the 1-year rate. Kintock showed highly fluctuating
rates, with a large drop from 2005-06 to 2008-09, but
then an increase from 2008-09 to 2010-11. Yet
another way to assess comparisons in recidivism rates
is to examine rates across the three follow-up periods
(6-months, 1-year, and 3-year). Gateway and the state
run centers are both again among the top highest
overall recidivism rates across the three different
follow-up periods.

Johnstown CCC
Pennsylvania Department of Corrections

33

2013 PA Recidivism Report
Table 28: Rearrest Rates By Vendor
2005-06 Release Cohort

2008-09 Release Cohort

2010-11 Release Cohort

Vendor
(# of Centers)

6-Month

1-Year

3-Year

6-Month

1-Year

3-Year

6-Month

1-Year

3-Year

CEC (4)

14.8%

31.5%

54.9%

12.4%

21.6%

51.6%

9.0%

19.1%

N/A

Firetree (4)

6.5%

15.2%

47.8%

9.8%

19.7%

39.3%

9.2%

16.9%

N/A

Gaudenzia (9)

6.7%

13.5%

50.0%

9.2%

20.2%

46.8%

6.6%

11.6%

N/A

Gateway (4)

7.1%

19.0%

38.1%

9.1%

25.8%

51.5%

10.3%

21.8%

N/A

Kintock (2)

14.9%

31.0%

63.2%

9.6%

22.8%

47.1%

13.3%

26.7%

N/A

Minsec (7)

15.2%

30.3%

59.3%

10.8%

22.9%

50.7%

6.4%

12.3%

N/A

Renewal (2)

3.4%

24.1%

48.3%

3.6%

16.4%

29.1%

2.8%

9.7%

N/A

Parole To Street

12.0%

23.2%

49.2%

12.2%

23.4%

48.1%

11.8%

21.1%

N/A

Parole To CCC

9.2%

24.6%

48.6%

10.4%

20.4%

42.5%

13.2%

23.9%

N/A

Parole To CCF

12.1%

25.5%

53.2%

9.8%

22.1%

48.1%

8.3%

16.8%

N/A

CEC (4)

5

Gateway (4)

Minsec (7)

Figure 25: 3-Year Rearrest Rate Percentage Point Difference Compared to Parole to the
Street (2008-2009 Releases)

Firetree (4)

-10

-20

Kintock (2)

3-Year Rearrest rate for
Parole to Street

Renewal (2)

-15

Parole To CCC

-5

Gaudenzia (9)

0

Pennsylvania Department of Corrections

34

2013 PA Recidivism Report
Table 29: Reincarceration Rates by Vendor
2005-06 Release Cohort

2008-09 Release Cohort

2010-11 Release Cohort

Vendor
(# of Centers)

6-Month

1-Year

3-Year

6-Month

1-Year

3-Year

6-Month

1-Year

3-Year

CEC (4)

17.3%

40.7%

61.7%

16.0%

33.2%

51.6%

21.1%

36.0%

N/A

Firetree (4)

15.2%

27.2%

56.5%

19.7%

24.6%

55.7%

13.8%

38.5%

N/A

Gaudenzia (9)

17.3%

33.7%

52.9%

12.8%

37.6%

56.0%

17.2%

26.8%

N/A

Gateway (4)

19.0%

35.7%

54.8%

18.2%

36.4%

56.1%

27.6%

54.0%

N/A

Kintock (2)

21.8%

47.1%

69.0%

14.0%

27.2%

52.9%

16.7%

36.7%

N/A

Minsec (7)

17.2%

35.2%

57.2%

17.9%

33.2%

56.5%

22.8%

34.2%

N/A

Renewal (2)

27.6%

48.3%

72.4%

12.7%

30.9%

54.5%

16.7%

34.7%

N/A

Parole To Street

11.8%

26.3%

47.5%

9.3%

22.1%

44.0%

9.8%

22.5%

N/A

Parole To CCC

26.1%

40.1%

62.0%

20.4%

34.6%

53.3%

17.6%

30.2%

N/A

Parole To CCF

16.7%

35.3%

58.1%

15.1%

31.4%

53.2%

19.5%

34.0%

N/A

8

Minsec (7)

Gateway (4)

Gaudenzia (9)

Firetree (4)

Parole To CCC

10

CEC (4)

12

Kintock (2)

14

Renewal (2)

Figure 26: 3-Year Reincarceration Rate Percentage Point Difference Compared to Parole to
the Street (2008-2009 Releases)

6
4
2
0

3-Year Reincarceration rate for
Parole to Street

Pennsylvania Department of Corrections

35

2013 PA Recidivism Report
Table 30: Overall Recidivism Rates By Vendor
2005-06 Releases

2008-09 Releases

2010-11 Releases

Vendor
(# of Centers)

6-Month

1-Year

3-Year

6-Month

1-Year

3-Year

6-Month

1-Year

3-Year

CEC (4)

22.2%

45.1%

71.0%

24.0%

41.2%

67.2%

24.4%

42.1%

N/A

Firetree (4)

18.5%

31.5%

64.1%

26.2%

39.3%

60.7%

18.5%

43.1%

N/A

Gaudenzia (9)

18.3%

36.5%

65.4%

18.3%

43.1%

65.1%

22.2%

33.3%

N/A

Gateway (4)

26.2%

45.2%

64.3%

21.2%

43.9%

71.2%

32.2%

59.8%

N/A

Kintock (2)

25.3%

48.3%

78.2%

19.1%

37.5%

65.4%

21.7%

41.7%

N/A

Minsec (7)

22.8%

42.1%

69.7%

23.8%

41.3%

70.0%

24.7%

38.8%

N/A

Renewal (2)

31.0%

62.1%

75.9%

14.5%

41.8%

63.6%

18.1%

40.3%

N/A

Parole To Street

18.6%

35.5%

61.5%

17.4%

33.8%

59.7%

18.1%

32.7%

N/A

Parole To CCC

30.3%

47.9%

70.4%

27.9%

45.4%

67.5%

27.3%

42.0%

N/A

Parole To CCF

21.5%

41.6%

68.3%

20.8%

40.7%

66.5%

23.5%

40.3%

N/A

3

Firetree (4)

CEC (4)

Kintock (2)

6

Renewal (2)

9

Gaudenzia (9)

Minsec (7)

12

Parole To CCC

15

Gateway (4)

Figure 27: 3-Year Overall Recidivism Rate Percentage Point Difference Compared to Parole
to the Street (2008-2009 Releases)

0

3-Year Overall Recidivism
rate for Parole to Street

Pennsylvania Department of Corrections

36

2013 PA Recidivism Report
Appendix A—Technical Definition of Recidivism/Data Sources
Definition of Recidivism
The PA DOC identifies a recidivist as an inmate who, after release from prison, commits a new offense or violates parole,
resulting in an arrest, an incarceration, or both. It is important to note that this report only captures recidivism events that
occurred in Pennsylvania, and does not include recidivism events that may have occurred in another state. The recidivism
rate for rearrests, reincarcerations, and overall recidivism is calculated using:
𝑅𝑒𝑐𝑖𝑑𝑖𝑣𝑖𝑠𝑚 𝑅𝑎𝑡𝑒 (𝑡, 𝑦) =

# of released inmates who recidivated within time period t
# of total releases in calendar year y

where t is length of recidivism follow-up time and y is the release year.
The PA DOC has generally defined its benchmark recidivism follow-up period as three years after prison release. This followup period is generally recognized as an optimal follow-up period for capturing recidivism as a stable and reliable measure. In
addition to three-year rates, this report also examines six-month and one-year rates, as well as at least one comparison of
five-year rates.
In order to provide maximum insight into recidivism of inmates released from the PA DOC, data on arrests have been
collected in addition to standard reincarceration data. Arrest data was used to calculate rearrest rates for released inmates.
Many recidivism studies use multiple measures of recidivism, including rearrest and reincarceration rates.
Recidivism rates for Community Corrections Centers (CCCs) and Contract Facilities (CCFs) were only calculated for those who
were paroled from prison to a Center. This report did not examine recidivism rates for Center residents who were in a
Center for a technical parole violation (e.g., “halfway back” cases and TPV Center cases). Recidivism rates for pre-release
offenders in Centers were not included either. To maximize comparability between those paroled to a Center and those
paroled “to the street”, this report further only examined the sub-set of parole release cases who received a “parole to an
approved home plan” Parole Board action, some who transitioned through a Center (i.e., the “Parole to Center” group) and
others who were paroled directly home (i.e, the “Parole to Street” group). We think this is an important methodological
improvement over previous attempts to evaluate recidivism rates for Pennsylvania’s CCCs and CCFs.
Data Sources: Releases and Reincarceration Data
Reincarceration data for this report was extracted from PA DOC internal databases by the Bureau of Planning, Research and
Statistics. The data used represents released inmates by release year. Demographic information (e.g., age, sex, race) and
commitment data (e.g., primary offense type) was collected from release records. Only inmates released permanently were
included- that is, the releases included all inmates whose incarceration sentence had been satisfied. This includes some
inmates whose sentence involves a period of post-prison supervision.
Data Sources: Rearrest Data
The Pennsylvania State Police (PSP) provided arrest data for this report. The PSP receives arrest reports from local police
agencies within the state. Since arrest reports from local agencies are not mandated by law, this data may underreport actual
arrests of released inmates. Computerized criminal history files drawn from this statewide database were used to provide
arrest data to the PA DOC.

Pennsylvania Department of Corrections

37

2013 PA Recidivism Report
Appendix B—End Notes
1. Rearrest and Overall Recidivism rates were not available for the 20-year time period
2. Metropolitan Areas as defined by the PA Department of Labor (www.paworkstats.state.pa.us).
Allentown : Carbon, Lehigh, Northampton
Altoona: Blair
Erie: Erie
Harrisburg-Carlisle: Cumberland, Dauphin, Perry
Johnstown: Cambria
Lancaster: Lancaster
Lebanon: Lebanon
Philadelphia: Philadelphia, Delaware, Chester, Bucks, Montgomery
Pittsburgh: Allegheny, Armstrong, Beaver, Butler, Fayette, Washington, Westmoreland
Reading: Berks
Scranton-Wilkes-Barre: Lackawanna, Luzerne, Wyoming
State College: Centre
Williamsport: Lycoming
York-Hanover: York
3. Race/ethnicity categories are measured as mutually exclusive, according to the inmate’s response upon entry into state
prison.
4. Other race/ethnicity categories are not used in this report because they make up less than 1% of the releases in any
given year.
5. Age groups are determined based on equal sizes of the inmates released in 2008.
6. The number of prior arrests and incarcerations were determined based on equal groupings of the inmates released in
2008.
7. Risk score based on the LSI-R assessment given upon entry into state prison. The LSI-R™ assessment is a quantitative
survey of offender attributes and offender situations relevant for assessing criminal risk of re-offending, and making
decisions about levels of supervision and treatment. The instrument’s applications include assisting in the allocation of
resources, helping to make probation and placement decisions, making appropriate security level classifications, and
assessing treatment progress. The 54 LSI–R items include relevant factors for making decisions about risk level and
treatment.
8. Breakdown of Broad Crime Categories:
 Violent—Murder/Manslaughter, Forcible Rape, Robbery, Aggravated Assault, Other Assault, Statutory Rape,
Other Sexual Offenses, Kidnapping
 Property—Burglary, Theft/Larceny, Arson, Fraud, Stolen Property, Forgery
 Drugs—Drug Offenses
 Public Order/Other—Weapons, DUI, Prison Breach, Part II Other
9. Arrests according to 2010 Pennsylvania State Uniform Crime Report (PA State Police, 2012).
10. Part I crimes were only included in this analysis because some Part II crime, such as simple assaults, may not be fully
reported to the Pennsylvania State Police.
11. Rates in Table 20 are per 100,000 population in Pennsylvania.
12. The complete set of controlled predictors consists of age at release, race, marital status, count of prior institutional
misconducts, count of prior incarcerations, LSI-R score, violent commitment offense indicator committing county, sex
offender indicator, status of completing prescribed institutional treatment, and time served in prison. The controlled
predictors are set at their mean values.
13. The follow-up time of 1 year and 3 years includes the 6 months of recidivism-free time assumed for this analysis.

Pennsylvania Department of Corrections

38