...

RECIDIVISM OF FEDERAL
OFFENDERS RELEASED IN 2010

UNITED STATES SENTENCING COMMISSION
September 2021

United States Sentencing Commission
One Columbus Circle, N.E.
Washington, DC 20002
www.ussc.gov

Charles R. Breyer
Acting Chair
Patricia K. Cushwa
Ex Officio
Jonathan J. Wroblewski
Ex Officio

Kenneth P. Cohen
Staff Director
Glenn R. Schmitt
Director
Office of Research and Data

September 2021

		Ryan Cotter, Ph.D.
			Deputy Director
Office of Research and Data
		

Courtney Semisch, Ph.D.
Senior Research Associate
Office of Research and Data

		

David Rutter, M.P.A.
Research Associate
Office of Research and Data

United States Sentencing Commission

Recidivism of Federal Offenders Released in 2010

TABLE OF CONTENTS
1

9

19

INTRODUCTION
1		

Introduction

4

Key Findings

6		

Scope of Analysis

OFFENDER AND OFFENSE CHARACTERISTICS
10

Offender Characteristics

12

Original Federal Offense Types and Characteristics

14

Criminal History

15

Sentences Originally Imposed

RECIDIVISM FINDINGS
20

Overall

24

Age and Criminal History

31

Additional Characteristics

38

CONCLUSION

39

APPENDICES

53

ENDNOTES

i

INTRODUCTION

United States Sentencing Commission

INTRODUCTION
This report is the first in a series
continuing the United States Sentencing
Commission’s (the “Commission”) research
of the recidivism of federal offenders. It
provides an overview of the recidivism
of federal offenders released from
incarceration or sentenced to a term
of probation in 2010, combining data
regularly collected by the Commission with
data compiled from criminal history records
from the Federal Bureau of Investigation
(FBI). This report provides an overview
of recidivism for these offenders and
information on key offender and offense
characteristics related to recidivism. This
report also compares recidivism outcomes
for offenders released in 2010 to federal
offenders released in 2005.
Recidivism has been an ongoing
research priority for the Commission,
both in its development of the original
sentencing guidelines and in its continuing
duty to collect, analyze, and report
sentencing data.1 This study further
advances the Commission’s work in
this area by examining recidivism in the
wake of landmark changes in the federal
criminal justice system: the Supreme
Court’s January 12, 2005 decision in
United States v. Booker that rendered the
federal sentencing guidelines advisory and
expanded use of evidence-based practices
in federal supervision.2

2

In 2013, the Commission undertook
a comprehensive, multi-year study of the
recidivism of federal offenders. The first of
seven reports from that study, Recidivism
Among Federal Offenders: A Comprehensive
Overview,3 examined the recidivism of
25,431 federal offenders released from
prison or placed on probation in 2005. The
Commission found that almost one-half
(49.3%) of federal offenders released to
the community in 2005 were rearrested
over an eight-year period.4 In addition,
the report replicated the research of
others in demonstrating the relationship
between recidivism and an offender’s age
and criminal history.5 The Commission’s
analysis of recidivism rates by age and
criminal history demonstrated that the
highest rearrest rates occurred among the
youngest group of offenders and those with
the most serious criminal histories.6
The Commission issued six additional
publications in the series, all analyzing
the offenders released to the community
in 2005. Three of the reports provided
detailed analyses of recidivism among
specific offender groups: Recidivism Among
Federal Drug Trafficking Offenders, Recidivism
Among Federal Firearms Offenders, and
Recidivism Among Federal Violent Offenders.7
The remaining three reports provided
detailed analyses of specific recidivism
correlates: The Past Predicts the Future:
Criminal History and Recidivism of Federal

Recidivism of Federal Offenders Released in 2010

Offenders, The Effects of Aging on Recidivism
Among Federal Offenders, and Length of
Incarceration and Recidivism.8
This report updates the Commission’s
earlier studies by analyzing federal
offenders released into the community in
2010. The Commission combined offender
and offense information that it regularly
collects with recidivism data compiled in
partnership with the FBI to examine the
extent of recidivism, the types and timing
of rearrests, and offense and offender
characteristics of rearrested offenders.
The 32,135 offenders in this study
comprise the largest group of offenders
with an eight-year recidivism study period
examined by the Commission to date.
Notably, this analysis provides an
opportunity to examine recidivism in the
context of major changes in the federal
criminal justice system. The offenders in
this study were sentenced and released
in conjunction with two significant
transformations in federal sentencing and
supervision. The Supreme Court’s January
12, 2005 decision in Booker rendered the
guidelines advisory. 9 The overwhelming
majority (83.1%) of offenders in this study
were sentenced following that landmark
decision. Therefore, the majority of these
offenders were sentenced under the
advisory guideline system which provides
increased judicial discretion to impose

sentences outside of the guidelines. In
2009, the Administrative Office of the
United States Courts (AOUSC) began
implementing new evidence-based
practices in offender supervision.10 The
Federal Probation and Pretrial Services
Office (PPSO) adopted the Federal Post
Conviction Risk Assessment (PCRA), a
risk assessment tool that incorporates
criminogenic factors related to recidivism
to improve supervision outcomes.11
This study was not designed to directly
measure the effects of these sentencing
and supervision changes. Nevertheless,
by comparing the recidivism patterns
of offenders released before and after
these changes, it provides insight into
the possible impact of increased judicial
discretion and evidence-based supervision
programs on recidivism.

''

This analysis provides an opportunity
to examine recidivism in the context
of major changes in the federal
criminal justice system following the
Supreme Court’s decision in Booker
and increased use of evidence-based
practices in federal supervision.

3

United States Sentencing Commission

KEY FINDINGS

2
49.3%
REARRESTED

50.7%
NOT REARRESTED

1

The recidivism rate remained
unchanged for federal offenders
released in 2010 compared to
offenders released in 2005 despite
two intervening major developments
in the federal criminal justice system:
the Supreme Court’s decision in Booker
and increased use of evidence-based
practices in federal supervision.
• Over an eight-year follow-up
period, nearly one-half (49.3%) of
federal offenders released in 2010
were rearrested, the same rate for
offenders released in 2005 (49.3%).
• Other recidivism patterns also were
consistent for the two offender
cohorts.

4

For offenders who were
rearrested, the median time
to arrest was 19 months.
The largest proportion (18.2%) of
offenders were rearrested for the first
time during the first year following
release. In each subsequent year, fewer
offenders were rearrested for the
first time than in previous years. Most
offenders in the study were rearrested
prior to the end of supervision
terms. Of those offenders who were
sentenced to a term of supervision
and rearrested, 76.3 percent were
rearrested earlier than the expiration
of their originally imposed supervision
term.

Median Time to
Rearrest:
19 MONTHS

YEAR 8
49.3%

YEAR 1
18.2%

3

Assault was the most common
(20.7%) offense at rearrest. The
second most common offense
was drug trafficking (11.3%), followed
by: larceny (8.7%), probation, parole,
and supervision violations (8.1%), and
administration of justice offenses (7.5%).

Recidivism of Federal Offenders Released in 2010

4

Combined, violent offenses
comprised approximately onethird of rearrests; 31.4 percent
of offenders were rearrested for assault
(20.7%), robbery (4.5%), murder (2.3%),
other violent offense (2.3%), or sexual
assault (1.6%).

VIOLENT
REARREST
31.4%
NON-VIOLENT
REARREST
68.6%

6

Combined, the impact of CHC
and age on recidivism was even
stronger. During the eightyear follow-up period, 100 percent of
offenders who were younger than 21
at the time of release and in CHC IV, V,
and VI (the most serious CHCs) were
rearrested. In contrast, only 9.4 percent
of offenders in CHC I (the least serious
CHC) who were aged 60 and older at
release were rearrested.

Less
Younger

• Offenders in CHC I (the least serious
CHC) had the lowest rearrest rates
(30.2%) and offenders in CHC VI (the
most serious CHC) had the highest
rearrest rates (76.2%).
• In addition, nearly three-quarters
(72.5%) of offenders younger than
age 21 upon release were rearrested
during the study period compared
to 15.9 percent of offenders aged 60
and older.

More

100% rearrested

1

Age

5

Similar to findings in its previous
studies, the Commission found
age and Criminal History
Category (CHC) were strongly
associated with rearrests.

Criminal History

j
Older

□ -9.4%

7

Offenders sentenced for firearms
and robbery offenses had the
highest rearrest rates during the
eight-year follow-up period, with 70.6
percent and 63.2 percent, respectively.
In contrast, offenders sentenced for
fraud, theft, or embezzlement had the
lowest rearrest rate (35.5%).

5

United States Sentencing Commission

SCOPE OF ANALYSIS
Defining and Measuring
Recidivism
Recidivism “refers to a person’s relapse
into criminal behavior, often after the
person receives sanctions or undergoes
intervention for a previous crime.”12
Recidivism measures can provide policy
makers with information regarding the
relative threat to public safety posed
by various types of offenders, and the
effectiveness of public safety initiatives
in deterring crime and rehabilitating
offenders.13 Recidivism measures are used
by numerous public safety agencies to
measure program performance and inform
policy decisions on issues such as pretrial
detention, prisoner classification and
programming, and offender supervision in
the community.14
Two measures are foundational to
recidivism research, both of which can
impact the outcomes of recidivism analyses.
The first measure is the type of event used
to indicate a relapse into criminal behavior.
Recidivism typically is measured by
criminal acts that resulted in the rearrest,
reconviction, or reincarceration of an
offender.15 The Commission used rearrest
for this study for several reasons. Rearrest
is the most commonly used measure and
is the primary measure of recidivism used
by federal agencies in recent recidivism
studies.16 Federal agencies are using

6

rearrest as the primary measure because it
is a more reliable measure than reconviction
and reincarceration due to the incomplete
nature of disposition data.17 Criminal
records often fail to include reconvictions
and reincarcerations because jurisdictions
inconsistently report them. The records
the Commission used to compile the data
for this study reflect this inconsistency.
For example, records for 44.1 percent
of charges had no associated disposition
information. For these reasons, the
incomplete nature of disposition data used
to identify reconviction and reincarceration
events makes them unreliable measures of
recidivism.
It should be noted that using rearrest
does result in higher recidivism rates than
reconviction or reincarceration. Not only
are rearrests more consistently reported,
but also the evidentiary standard for an
arrest (probable cause) is less stringent than
the evidentiary standard for a conviction
and, therefore, incarceration (beyond a
reasonable doubt). Because not all arrests
result in conviction or incarceration,
rearrests can overstate recidivism.18 The
Commission’s rearrest measure also
includes arrests for alleged violations
(or revocations) of supervised release,
probation, or state parole, which also can
contribute to increased overall recidivism
rates. The Commission, however, excluded
rearrests for minor traffic offenses.

Recidivism of Federal Offenders Released in 2010

The second component of measuring
recidivism is the “follow-up period,”
the period of time over which events
are counted following release into the
community. After a starting event, in
this case, release from prison into the
community or placement on probation,
recidivism events are documented
through the end of the follow-up period.
The length of follow-up periods varies
across recidivism studies. Often, due to
limitations on available data, some studies
follow offenders for as little as six months.
Other studies follow offenders for several
years. Tracking offenders for a longer
duration provides a more accurate estimate
of recidivism or desistance from crime.19
The Commission used an eight-year followup period.

Methodology
This report provides a comprehensive
analysis of the recidivism of all federal
offenders who were released from federal
prison or sentenced to probation in
2010. The offenders in the study cohort
were identified in cooperation with the
Federal Bureau of Prisons (BOP) and the
AOUSC. The BOP provided identifying
information, release dates, and other
pertinent information for the Commission
to identify offenders released from
prison. The AOUSC provided identifying
information, revocation information, and
other pertinent information for offenders
sentenced to probation. The Commission
compiled the identifying information for
these offenders to obtain criminal records
in partnership with the FBI.

''

Federal agencies most commonly use rearrest as the primary recidivism measure
because it is a more reliable measure than reconviction and reincarceration
due to the incomplete nature of disposition data.

7

United States Sentencing Commission

The data used in this report
combines data regularly collected by the
Commission20 with data compiled as part
of a data sharing agreement with the FBI’s
Criminal Justice Information Services
Division.21 Through an agreement with
the FBI, the Commission collected and
processed criminal history records from all
state and federal agencies for the offenders
in the study.22 The Commission then
combined this criminal record data with
the data it collected about the offenders
when they were originally sentenced.

SUMMARY OF ANALYSIS

2010
YEAR OF RELEASE

32,135
OFFENDERS

8
YEAR FOLLOW-UP

The final study group of 32,135 offenders
satisfied the following criteria:
• United States citizens;
• Re-entered the community during 2010
after discharging their sentence of
incarceration or by commencing a term
of probation in 2010;
• Not reported dead, escaped, or
detained;23
• Have valid FBI numbers that could be
located in criminal history repositories
(in at least one state, the District of
Columbia, or federal records).

8

This report provides an analysis of
the overall recidivism rates during the
eight-year follow-up period for the
32,135 federal offenders identified for
this study, as well as key offense and
offender characteristics of this group.24 For
offenders who recidivated during the study
period, the analysis examines the elapsed
time from release to rearrest and the types
of offenses at rearrest. The analysis also
compares recidivism patterns to those of
federal offenders released in 2005.25

OFFENDER AND OFFENSE
CHARACTERISTICS

United States Sentencing Commission

OFFENDER AND
OFFENSE CHARACTERISTICS
Table 1. Offender Characteristics
Federal Offenders Released in 2010
Offender Characteristics
Race/Ethnicity
White

41.2%

Black

35.8%

Hispanic

18.1%

Other

4.9%

Gender
Male

82.7%

Female

17.3%

Education
Less than High School

34.9%

High School Graduate

38.4%

Some College

20.3%

College Graduate

6.4%

Age at Sentencing
Average

35 Years

Median

33 Years

Age at Release
Average

38 Years

Median

37 Years

Offender Characteristics
White offenders (41.2%) comprised
the largest group of offenders in the study
group, followed by Black (35.8%), Hispanic
(18.1%), and Other Races (4.9%) (Table
1).26 The offenders predominantly were
male (82.7%). Nearly two-thirds (65.1%)

10

of offenders were high school graduates,
including 6.4 percent who graduated
college. Approximately one-third (34.9%)
of offenders in the study did not complete
high school.

Recidivism of Federal Offenders Released in 2010

''

The offenders in this study group were similar to federal offenders released in 2005.

Figure 1. Age at Sentencing and Release
Federal Offenders Released in 2010

■

33.9%

Age at Sentencing

30.8%

Age at Release

35.0%

23.2%

4.0%

■

18.6%

23.1%
9.0%

1.5%

Younger than 21

21 - 29

30 - 39

At sentencing, the average age of
offenders in the study group was 35 years
(median 33 years),27 with offenders ranging
from 18 to 84 years of age. At the time of
release, the average age of offenders was
38 years (median 37 years). At release, the
largest proportion (35.0%) of offenders
were age 30 to 39 years old and only 5.3
percent were aged 60 or older (Figure 1).
The offenders in the study group were
similar to federal offenders released in
2005. Among offenders released in 2005,

40 -49

11.9%

50 - 59

3.7%

5.3%

60 and Older

the overwhelming majority (81.7%) were
male. White offenders (43.7%) comprised
the largest group of offenders, followed
by Black (33.9%), Hispanic (17.8%), and
Other Races (4.6%). The educational
attainment was as follows: 34.2 percent
did not complete high school, 36.9 percent
graduated high school, 21.4 percent
completed some college, and 7.5 percent
graduated college. The median age at
release was 36 years.28

11

United States Sentencing Commission

Original Federal Offense Types
and Characteristics
The largest proportion of offenders
in the study group, 43.1 percent, were
released following sentences for drug
trafficking offenses (Figure 2).29 The
remaining offenders were sentenced for
fraud, theft, or embezzlement (17.4%),
firearms (15.3%), immigration (4.4%),30
robbery (4.1%), or other types of offenses
(15.7%).31 A small proportion of offenders
were sentenced for a violent offense,
comprising 9.4 percent of the study group
(Figure 3).32

Figure 2. Federal Offense Type
Federal Offenders Released in 2010
DRUG TRAFFICKING

FRAUD/THEFT/EMBEZZLEMENT

FIREARMS

IMMIGRATION

ROBBERY

OTHER OFFENSES

-I
I

43.1%

17.4%

15.3%

4.4%

4.1%

15.7%

The largest proportion of offenders
released in 2005 were sentenced for drug
trafficking (41.6%). Fraud (13.6%) and
larceny (3.9%) were reported separately
for offenders released in 2005, combined,
they accounted for the third largest group
of offenders. The second largest group
was all other offenses (20.3%), followed
by firearms (12.8%), robbery (4.3%), and
immigration (3.5%).34

Offenders in this study committed
similar federal offenses as offenders
released in 2005. In the 2016 Recidivism
Overview Report, the Commission used
a slightly different offense classification
scheme for federal offenses.33
Nevertheless, the distribution of offense
types was similar for the two cohorts.

Figure 3. Violent Federal Offenses
Federal Offenders Released in 2010
43.1%

DRUG TRAFFICKING

FRAUD/THEFT/EMBEZZLEMENT

17.4%

VIOLENT
9.4%

FIREARMS
15.3%
A small proportion
of
offenders in the
study group
IMMIGRATION
4.4%
were originally sentenced
ROBBERY
4.1%
for a violent offense.
OTHER OFFENSES

12

15.7%

NON-VIOLENT
90.6%

Recidivism of Federal Offenders Released in 2010

Figure 4. Selected Sentencing Factors
Federal Offenders Released in 2010
0%

WEAPON INVOLVEMENT

11.7%

AGGRAVATING ROLE 5.0%

■

Applied

■

Not Applied

100%

88.3%

95.0%

MITIGATING ROLE

8.4%

91.6%

ACCEPTANCE OF
RESPONSIBILITY

92.2%

7.8%

Figure 4 provides information on
selected sentencing factors for offenders
in the study group. Weapon-involved
offenses were determined by the
application of a weapon-related specific
offense characteristic at sentencing or
a conviction under 18 U.S.C. § 924(c),35
or both.36 Sentence enhancements for
weapon-involved offenses applied to 11.7
percent of offenders in the study.
Chapter Three of the Guidelines Manual
provides adjustments pertaining to an
offender’s role in the offense. Relevant
to this study, Chapter Three provides
adjustments for offenders whose conduct
constitutes an aggravating role (§3B1.1)
or a mitigating role (§3B1.2) in the
offense. Section 3B1.1 provides a 2- to
4-level increase for offenders who acted
in an aggravating role in the offense.37
Conversely, §3B1.2 provides a 2- to 4-level
decrease for offenders who acted in a

mitigating role in the offense.38 Courts
applied these role adjustments for a small
proportion of offenders at the time of
sentencing; a slightly greater proportion
of offenders had offense level decreases
for mitigating role (8.4%) compared to
offenders with offense level increases for
aggravating role (5.0%).
Section 3E1.1 of the Guidelines Manual
provides for a 2- or 3-level decrease for an
offender who demonstrates acceptance
of responsibility for the offense.39 The
majority (92.2%) of offenders in the
study had an offense level decrease for
acceptance of responsibility at the time of
sentencing.

13

United States Sentencing Commission

Criminal History

Figure 5 shows the CHCs and underlying
criminal history points for the offenders in
the study. The largest proportion (45.3%)
of offenders were in CHC I, the lowest
Criminal History Category, with zero
or one criminal history point assigned.
The majority of CHC I offenders in the
study had zero points assigned under the
guidelines. Of the zero-point offenders,
nearly half (47.9%) were first-time
offenders who had no prior contact with
the criminal justice system; that is, they
did not have arrests or convictions prior
to the original offense of conviction.42
At the other end of the spectrum, 11.1
percent of offenders were in the most
serious Criminal History Category of VI.43
The number of criminal history points
underlying calculations of CHC VI ranged
from 13 to 56. The remaining offenders
were in CHC II (11.9%), CHC III (15.8%),
CHC IV (9.8%), and CHC V (6.1%).44

Chapter Four of the Guidelines Manual
provides for the calculation of a criminal
history score based primarily on the
type of sentence and length of any prior
sentence of imprisonment, among other
considerations.40 The guidelines provide
rules to determine the total number of
criminal history points applicable to an
offender’s prior convictions, which, in turn,
determine the offender’s Criminal History
Category (CHC) in the Sentencing Table.41
For example, three points are assigned
for each prior sentence of imprisonment
exceeding one year and one month. In
addition to points for prior sentences,
two additional points are added if the
defendant committed the federal offense
while under any criminal justice sentence,
such as probation. The total number of
criminal history points determine the
offender’s CHC, ranging from I to VI.

Figure 5. Criminal History Category and Underlying Criminal History Points
Federal Offenders Released in 2010
Criminal History Category

CHC I

CHC II

CHC III

CHC IV

CHC V

CHC VI

45.3%

11.9%

15.8%

9.8%

6.1%

11.1%

Criminal History Points

0

14

1

2

3

4

5

6

7

8

9 10 11 12

13 or more

Recidivism of Federal Offenders Released in 2010

Figure 6. Calendar Year of Federal Sentencing
Federal Offenders Released in 2010

83.1%

sentenced after
United States v. Booker
(January 12, 2005)

7,644

5,109 5,317
3,674
3,040
1,944
367 567
111 177 213 281

1,244
966 1,080

------------■
8

33

64

110

95

91

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

In addition to CHC, the guidelines
provide for enhanced penalties for some
repeat offenders; two of those provisions
are pertinent to this study. First, §4B1.1
(Career Offender) provides enhanced
penalties for offenders with an instant
conviction for a felony “crime of violence”
or a “controlled substance offense” (as
those terms are defined in §4B1.2) and who
have at least two prior felony convictions
for such offenses.45 Second, §4B1.4 (Armed
Career Criminal) provides enhanced
penalties for armed career criminals
pursuant to 18 U.S.C. § 924(e), the Armed
Career Criminal Act (ACCA). The ACCA
and, in turn, §4B1.4, provide increased
sentences for offenders who were
convicted under 18 U.S.C. § 922(g) and who
have at least three prior convictions for a
“violent felony” or “serious drug offense.”46
These enhancements applied to a very

small proportion of offenders in the study;
3.8 percent were originally sentenced
under the career offender or armed career
criminal provisions.47

Sentences Originally Imposed
The offenders in the study group
were originally sentenced between 1990
and 2010 (Figure 6). More than threequarters (76.2%) were sentenced prior
to the release year of 2010. Notably, the
overwhelming majority of offenders in
the study (83.1%) were sentenced after
January 12, 2005, the date of the Supreme
Court’s decision in Booker,48 that rendered
the sentencing guidelines advisory. In
addition, the offenders in the study were
released (or sentenced to probation) during
the PPSO’s implementation of updated
evidence-based supervision practices and
use of the PCRA.
15

United States Sentencing Commission

Figure 7. Type of Sentence Imposed
Federal Offenders Released in 2010

PROBATION + ALTERNATIVE
5.7%

7

PROBATION ONLY
9.9%

PRISON + ALTERNATIVE
3.6%

PRISON
ONLY
80.8%

Most offenders in the study group
originally were sentenced to a term of
imprisonment. As shown in Figure 7, a
prison-only sentence was imposed for
80.8 percent of offenders. Another 9.9
percent of offenders were sentenced
to probation only (i.e., where no type of
confinement was imposed). An additional
5.7 percent of offenders were sentenced
to terms of probation with some type
of alternative confinement, and 3.6
percent were sentenced to a combination
of imprisonment and alternative
confinement, such as a halfway house or
home confinement.49 Following the same
methodology used in the 2016 Recidivism
Overview Report, the Commission combined
these four sentencing categories into
two categories for analysis.50 The prison

16

category comprises the 84.4 percent of
offenders sentenced to prison only and
those sentenced to a combination of
imprisonment and alternative confinement;
the probation category comprises the
15.6 percent of offenders sentenced to
probation only and those sentenced to
probation with some type of alternative
confinement.
The average sentence imposed for
all offenders in the study group was 51
months.51 For offenders sentenced to
prison, the average imprisonment term was
nearly five years (59 months). The majority
(60.2%) of offenders sentenced to prison
were sentenced to terms ranging from
two years to less than ten years (Figure 8).
Offenders sentenced to prison terms of
ten years or longer comprised 14.2 percent

Recidivism of Federal Offenders Released in 2010

of imprisoned offenders.52 For offenders
sentenced to probation, the average
probation term imposed was slightly more
than three years (38 months). The largest
proportion (37.8%) of offenders sentenced
to probation were sentenced to a threeyear term.
Section 5D1.1(a) of the Guidelines
Manual provides that courts may impose
a term of supervised release following
a sentence of imprisonment for any
felony or misdemeanor and must do so if
required by statute.53 Nearly all (99.4%)
of the offenders sentenced to prison also
were sentenced to subsequent terms of
supervised release. The average term of
supervised release imposed was nearly
four years (46 months). As shown in Figure
8, courts most often imposed supervised

release terms of three to less than four
years, accounting for 52.3 percent of
offenders. A very small proportion of
offenders, 2.0 percent, were sentenced to
terms of supervised release of ten years or
more, or life. Nearly all (99.4%) offenders
in the study were sentenced to some type
of supervision (probation or supervised
release).

''

Nearly all offenders in the study group
were sentenced to some type of
supervision (probation or supervised
release).

Figure 8. Length of Imprisonment, Probation, and Supervised Release
Federal Offenders Released in 2010

LENGTH OF IMPRISONMENT

LENGTH OF PROBATION

LENGTH OF SUPERVISED RELEASE

AVERAGE 59 MONTHS

AVERAGE 38 MONTHS

AVERAGE 46 MONTHS

52.3%

37.8%
34.1%
27.2%

26.1%

-5.3% 4.6%

15.7%

19.1%
14.2%

11.2%

25.8%

- -•
4.7%

2.9% 7.7%

9.3%

1.6% 0.4%

17

United States Sentencing Commission

18

RECIDIVISM FINDINGS

United States Sentencing Commission

OVERALL FINDINGS

10

Table 2. Comparison of Overall Rearrest Findings
Federal Offenders Released in 2010 and 2005
Offenders Released
in 2010

Offenders Released
in 2005

49.3%

49.3%

Median Months to Rearrest

19

21

Median Number of Rearrests

2

2

Most Common Post Release
Offense

Assault
(20.7%)

Assault
(23.3%)

33

33

Percent Rearrested

Median Age at Release

Overall Recidivism Findings
The recidivism of federal offenders has
remained constant. Nearly half (49.3%) of
offenders released in 2010 were rearrested
within the eight-year follow-up period.54
This rate is identical to the rearrest rate
(49.3%) for federal offenders released in
2005.55 In addition, recidivism patterns
were the same for offenders released in
2010 and offenders released in 2005 (Table
2).
Among offenders in this study who
were rearrested, the median time to
arrest was 19 months. For one-half of
offenders who were rearrested, the first
rearrest occurred just over one and onehalf years following their initial release to
the community. The number of rearrests
during the follow-up period ranged from
one to 43 and the median number of
rearrests was two. Slightly less than onethird (30.7%) of offenders had a single
rearrest, but one-quarter (26.1%) were
rearrested five times or more. Assault was
20

the most serious and most common offense
at rearrest (20.7%).56 For offenders who
were rearrested, the median age at release
was 33 years (average 35 years).
As shown in Table 2, recidivism patterns
were unchanged from the Commission’s
findings in its 2016 Recidivism Overview
Report for federal offenders released in
2005. In that study, the Commission also
reported a median number of arrests of two
and a median age at release of 33 years.
The most common offense at rearrest also
was assault, and a slightly larger proportion
of offenders in that study had assault as the
most common type of offense (23.3%). The
median time to rearrest in that study was
slightly longer at 21 months.
Notably, these similar recidivism
patterns occurred in different federal
criminal justice environments. The
overwhelming majority of offenders
released in 2005 (77.8%) were sentenced
prior to Booker, under the mandatory
guideline system.57 In contrast, the

Recidivism of Federal Offenders Released in 2010

overwhelming majority of offenders
released in 2010 were sentenced postBooker, under the advisory guideline
system.58 In addition, the two groups of
offenders were released into different
supervisory systems. The offenders in
this study were released in 2010, when
the PPSO had increased its reliance on
evidence-based practices by implementing
the PCRA tool. While this study was
not specifically designed to assess the
impact of these changes on recidivism, it
is nevertheless notable that recidivism
patterns remained entirely unchanged in
their wake.

offenders released in 2005. As shown in
Figure 9, 18.2 percent of offenders in this
study were rearrested during the first year
following release into the community. In
the second year, an additional 10.4 percent
of offenders were rearrested for the first
time, and in the third year, an additional
6.8 percent were rearrested for the first
time. By the eighth year, only 1.8 percent
of offenders who were not previously
arrested recidivated for the first time.
Time to rearrest also is unchanged
from offenders released in 2005. In the
2016 Recidivism Overview Report, the
Commission reported that 16.6 percent
of offenders in the 2005 cohort were
rearrested during the first year of release,
and fewer offenders were rearrested for
the first time in each of the subsequent
years. Nearly identical to the current
findings, that study also showed that 10.5
percent of offenders were rearrested
for the first time in the second year, 6.6
percent were rearrested for the first time
in the third year, and only 1.8 percent of
offenders were rearrested for the first time
in the eighth year.59

Timing of Rearrest
The largest proportion of rearrests
occurred soon after release and steadily
decreased over time. Of those offenders
who were rearrested, the largest
proportion were rearrested for the first
time during the first year following release.
In each subsequent year, fewer offenders
were rearrested for the first time than in
previous years. This pattern also mirrors
that reported by the Commission for

Figure 9. Time to First Rearrest
Federal Offenders Released in 2010

Median Time to Rearrest:
19 months
35.4%

39.8%

43.1%

45.6%

47.5%

5

6

7

49.3%

28.6%
18.2%

0

1

2

3

4

8

Years After Release

21

United States Sentencing Commission

Rearrests and Federal Supervision
Status
To further explore the issue of timing
of rearrest, the Commission also examined
rearrests relative to federal supervision
status. As discussed above, nearly all
(99.4%) offenders in the study originally
were sentenced to a term of federal
supervision (either probation or supervised
release). The average supervision term
imposed was less than four years (44
months). Because almost all60 of the
supervision terms imposed were shorter
than the eight-year follow-up period, a
large proportion of offenders had the
opportunity to successfully complete their
originally imposed supervision terms prior
to the end of the study period. While the
data for this study included the length
of supervision terms imposed, it did not
include supervision status at the time of
rearrest. Therefore, for each rearrested
offender, the Commission compared the
length of supervision term imposed to the
elapsed time prior to rearrest to provide a
proxy of the offenders’ supervision status
at the time of rearrest.61
Overall, 49.2 percent of offenders
originally sentenced to any term of
supervision were rearrested during the
study period. Based on the length of terms
imposed, most offenders in the study
were rearrested prior to the end of those
terms. As shown in Figure 10, of those
offenders who were sentenced to a term of
supervision and rearrested, 76.3 percent
were rearrested earlier than the expiration
of their originally imposed supervision

22

Figure 10. Rearrests by Supervision Status
Federal Offenders Released in 2010 on Supervision
50.8%

OFFENDERS
NOT REARRESTED

49.2%

OFFENDERS REARRESTED

76.3%

PRIOR TO

END OF IMPOSED
SUPERVISION TERM

23.7%
AFTER

END OF IMPOSED
SUPERVISION TERM

•

I
term. The remaining 23.7 percent were
rearrested after the expiration of their
originally imposed supervision term.
Types of Rearrests
The Commission ranked offenses by
severity to analyze new offenses committed
by the offenders in the study. This severity
ranking was similar to the one in the
Commission’s 2016 Recidivism Overview
Report62 and presents new offenses in
order of seriousness. If an offender was
rearrested multiple times during the study
period or had multiple charges in an arrest,
the most serious offense according to this
ranking was reported as the type of offense
at rearrest.63

Recidivism of Federal Offenders Released in 2010

Figure 11. Most Serious Offense at Rearrest
Federal Offenders Released in 2010
VIOLENT OFFENSES
31.4%

MURDER
SEXUAL ASSAULT
ROBBERY
ASSAULT
OTHER VIOLENT
DRUG TRAFFICKING
BURGLARY
LARCENY
FRAUD
OTHER PROPERTY
DRUG POSSESSION
OTHER DRUG
WEAPON
OTHER SEX OFFENSES
DUI/DWI
IMMIGRATION

ADMINISTRATION OF JUSTICE
PROBATION/PAROLE/SUPERVISED RELEASE VIOL.
PUBLIC ORDER
OTHER

This analysis differs from the
Commission’s previous work in the
reporting of one rearrest category. To
provide an increased level of detail for
this analysis, the Commission expanded
the previously reported “public order”
category. The Commission examined the
offense types comprising the public order
category and determined that reporting
four separate categories provided a more
meaningful analysis. Therefore, the
previously existing public order category
is divided into administration of justice
offenses; probation, parole, and supervision
violations; other sex offenses; and public
order offenses.

2.3%

■

---1.6%

I

4.5%
20.7%

2.3%

■

■

11.3%

2.7%
8.7%

4.0%

■

2.4%
6.7%

4.8%

I

1.9%

I 0.5%

6.3%

0.8%

I
I

7.5%
8.1%

1.2%

1.7%

Using this method, the largest
proportion (20.7%) of offenders in this
study were rearrested for assault (Figure
11). The second most common offense
was drug trafficking (11.3%), followed
by larceny (8.7%), probation, parole,
and supervision violations (8.1%), and
administration of justice offenses (7.5%).
Other offense types each comprised less
than seven percent of rearrests, including
all remaining types of violent offenses.
Combined, violent offenses comprised
approximately one-third of rearrests; a
total of 31.4 percent of offenders were
rearrested for assault (20.7%), robbery
(4.5%), murder (2.3%), an other violent
offense (2.3%), or sexual assault (1.6%).64
23

United States Sentencing Commission

The predominant offenses at rearrest
for offenders released in 2010 are
consistent with the Commission’s previous
recidivism findings for offenders released
in 2005. The 2016 Recidivism Overview
Report also reported that assault was the
most common offense at rearrest; 23.3
percent of all federal offenders were
rearrested for assault, followed by public
order (15.5%),65 drug trafficking (11.5%),
and larceny (7.7%).66
Similarly, the Commission also
reported that violent offenses comprised
approximately one-third of rearrests, with
32.3 percent of offenders having been
rearrested for assault (23.3%), robbery
(4.7%), rape (1.9%), homicide (1.3%), or an
other violent offense (1.1%).

Age and Criminal History
Age and criminal history are
consistently strong predictors of
recidivism.67 This strong association is
demonstrated in the current study as
well as in the Commission’s prior study of
offenders released in 2005. For both the
2010 and 2005 cohorts, lower rearrest
rates were associated with older offenders
and higher rearrest rates were associated
with offenders with more extensive
criminal histories. The combined impact
of age and criminal history on recidivism
is more pronounced. This section of the
report focuses on these two primary
correlates of recidivism: age and criminal
history.

24

Age
Figure 12 shows the age distribution
of federal offenders released in 2010
and their corresponding recidivism rates.
As shown in the figure, more than half
of federal offenders were younger than
40 years of age at the time of release.
Figure 12 also demonstrates the inverse
relationship of age and recidivism. The
youngest offenders had the highest
recidivism rates, and those rates steadily
declined with increasing age. For example,
nearly three-quarters (72.5%) of offenders
younger than age 21 were rearrested
during the study period. Nearly twothirds (64.4%) of offenders aged 21 to 29
were rearrested and less than one-third
(30.8%) of offenders aged 50 to 59 were
rearrested. The lowest rearrest rates
were for offenders age 60 and older; 15.9
percent of offenders in that age group were
rearrested during the study period.
In addition to having higher rearrest
rates, younger offenders also recidivated
sooner after release compared to their
older counterparts. While the median time
to rearrest was 19 months for all offenders
in the study, as shown in Table 3, the
median time to rearrest for the youngest
group of offenders (younger than 21 years)
was 12 months. One-half of offenders in
that age group were rearrested before one
year had elapsed following their release,
and the other one-half of offenders were
rearrested after one year had elapsed
following their release. The median time
to rearrest increased for offenders in each
successive age group, nearly doubling to
almost two years (22 months) for the two
oldest groups of offenders.

Recidivism of Federal Offenders Released in 2010

Figure 12. Age at Release and Rearrest Rates
Federal Offenders Released in 2010

---

12,000

100.0%

Number of Offenders

10,000

Percent Rearrested

72.5%

80.0%

64.4%

8,000

60.0%

54.1%
6,000

11,256

42.7%
40.0%

4,000

7,440

7,425

2,000

30.8%

3,826

15.9%

20.0%

1,697
473

0

0.0%

Younger than 21

21 - 29

30 - 39

40 - 49

50 - 59

60 and Older

Age at Release

The demonstrated relationship
between age and recidivism is consistent
with previous Commission findings. In its
2017 Recidivism Age Report the Commission
demonstrated the overall relationship
between age and recidivism. In that study,
the Commission found that younger
offenders were more likely to be rearrested
than older offenders and were rearrested
faster than older offenders.68 For offenders
released in 2005, the rearrest rate was
highest (67.6%) among offenders younger

than 21 and declined in each subsequent
age group to 16.4 percent for offenders
aged 60 and older. In addition, offenders
who were younger than age 30 when
they were released had the shortest
median time to rearrest (17 months).
Conversely, offenders aged 60 and older
had the longest median time to rearrest (28
months).69 Furthermore, in that report the
Commission demonstrated that the agerecidivism relationship persisted across all
other factors in the analysis (e.g., criminal
history).70

Table 3. Time to Rearrest by Age Group
Federal Offenders Released in 2010

Criminal History Points and
Criminal History Category

Age at Release

Time to Rearrest
(median months)

Younger than 21

12

21 - 29 Years

16

30 - 39 Years

20

40 - 49 Years

21

50 - 59 Years

22

60 and Older

22

Criminal history also is strongly
correlated with recidivism. The
Commission examined the relationship
between rearrests and both the total
number of criminal history points and the
resulting CHC. This examination confirmed
that offenders with more extensive
criminal histories had higher rearrest rates.
25

United States Sentencing Commission

As shown in Figure 13, rearrest rates
increased with each CHC. The rearrest
rates ranged from less than one-third
(30.2%) of offenders in CHC I to nearly
one-half (49.4%) of offenders in CHC II,
and steadily increased to more than threequarters (76.2%) of offenders in CHC VI.

percentage points for offenders with one
criminal history point (42.3%); and then
increased steadily. With two exceptions
(8 points and 12 points) each additional
criminal history point was associated with a
higher rate of rearrest. For example, more
than half (57.9%) of offenders with four
criminal history points were rearrested,
more than two-thirds (69.2%) of offenders
with seven criminal history points were
rearrested, and 80.6 percent of offenders
with 13 or more criminal history points
were rearrested during the study period.

An offender’s criminal history points
largely determine the CHC.71 Therefore,
the relationship between recidivism
and criminal history is more evident
comparing rearrest rates and criminal
history points. As shown in Figure 14,
offenders with zero criminal history
points had the lowest rearrest rates.
Slightly more than one-quarter (26.8%) of
offenders with no criminal history points
assigned were rearrested during the study
period. Rearrests increased more than 15

These findings are consistent with
the Commission’s previous findings for
recidivism and criminal history for federal
offenders released in 2005. In its 2016
Recidivism Overview Report and its 2017
Recidivism Criminal History Report, the

Figure 13. Rearrest Rates by Criminal History Category
Federal Offenders Released in 2010
CRIMINAL HISTORY CATEGORY

75.4%

76.2%

V

VI

70.3%
61.9%
49.4%

30.2%

I

26

II

III

IV

Recidivism of Federal Offenders Released in 2010

Commission reported the lowest recidivism
rates among offenders with zero criminal
history points (30.2%), a sharp increase for
offenders with one criminal history point
(46.9%), and a generally steady increase
in recidivism with each additional criminal
history point.72 The 2016 Recidivism
Overview Report reported a recidivism rate
of 81.5 percent for offenders with ten or
more criminal history points and the 2017
Recidivism Criminal History Report reported
a recidivism rate of 85.7 percent for
offenders with 15 or more criminal history
points. In addition, both publications
reported that rearrest rates steadily
increased from a low of 33.8 percent for
offenders in CHC I to a high of 80.1 percent
for offenders in CHC VI.73

This study demonstrated notably
lower rearrest rates among offenders
in CHC I compared to offenders in the
higher CHCs. To further analyze the
relationship between criminal history and
recidivism, the Commission undertook a
closer examination of CHC I offenders. As
described above, offenders in CHC I are
assigned either zero or one criminal history
point. As a group, offenders with zero
criminal history points had a rearrest rate
of 26.8 percent. Among offenders with
zero criminal history points, approximately
half (47.9%) had no prior contact with the
criminal justice system. The remaining
half (52.1%) of offenders with zero points
had either prior arrests that did not result
in convictions or convictions that did
not qualify for criminal history points, or

Figure 14. Rearrest Rates by Criminal History Points
Federal Offenders Released in 2010
CRIMINAL HISTORY POINTS

80.6%

57.9%
48.9%

49.5%

2

3

61.8%

65.2%

69.2%

67.3%

7

8

71.3%

72.8%

9

10

75.2%

71.9%

42.3%

26.8%

0

1

4

5

6

11

12

13 or
More

27

United States Sentencing Commission

Figure 15. Rearrest Rates for Offenders with Limited Criminal History
Federal Offenders Released in 2010

MOST EXTENSIVE CRIMINAL

76

LIMITED CRIMINAL HISTORY
(CHC I)

62.8%

42.3%
32.7%
20.3%

Zero Points/
No Prior Contact

Zero Points/
Prior Contact

both. As shown in Figure 15, one-fifth
(20.3%) of zero-point offenders with no
prior contact with the criminal justice
system were rearrested during the study
period. In comparison, nearly one-third
(32.7%) of zero-point offenders with
prior contact were rearrested. However,
both groups of zero-point offenders were
rearrested at lower rates compared to
their CHC I counterparts with one criminal
history point, 42.3 percent of whom were
rearrested during the study period.
These findings also are similar to
the Commission’s findings in the 2017
Recidivism Criminal History Report. In that
publication, the Commission reported
that offenders with zero criminal history
points had a rearrest rate of 30.2 percent.
However, offenders who had no prior
contact with the criminal justice system
had a rearrest rate of 25.7 percent and
offenders with prior contact had a rearrest
rate of 37.4 percent.74
28

One Point

Career Offender/
Armed Career
Criminal

The Commission also analyzed career
offenders and armed career criminals
for both the 2010 and 2005 cohorts.
Recidivism rates for those offenders were
lower than for offenders who were in the
highest CHC, absent application of those
recidivist provisions. Only 3.8 percent of
offenders in this study were sentenced as
career offenders or armed career criminals.
However, almost two-thirds (62.8%) of
those offenders were rearrested during
the study period (Figure 16). These rates
were lower than the rearrest rates for
offenders in the most serious CHC of VI
(76.2%). This difference is attributable to
the provisions in §§4B1.1 and 4B1.4. Those
guideline provisions assign career offender
and armed career criminal status based on
a combination of the type of instant offense
and types of prior convictions.75 Because
those guidelines assign CHC based on
offender status in lieu of criminal history
points, the resulting CHCs often supersede
the otherwise applicable CHCs.76 For

CH

Recidivism of Federal Offenders Released in 2010

Figure 16. Rearrest Rates for Offenders with Most Extensive Criminal History
Federal Offenders Released in 2010
MOST EXTENSIVE CRIMINAL HISTORY

76.2%

D CRIMINAL HISTORY
(CHC I)

62.8%

42.3%
32.7%

Zero Points/
Prior Contact

One Point

Career Offender/
Armed Career
Criminal

example, more than half (51.0%) of career
offenders and armed career criminals had
higher CHCs than would have otherwise
applied based on criminal history points
alone.
Similarly, among offenders released
in 2005, 69.5 percent of offenders who
qualified as career offenders or armed
career criminals were rearrested,
compared to 80.1 percent of offenders in
CHC VI. The Commission also reported
similar findings in other studies. In its 2016
Report to the Congress: Career Offender
Sentencing Enhancements, the Commission
studied offenders who were released
between 2004 and 2006 who had been
sentenced as career offenders under
§4B1.1. Two-thirds (66.2%) of career
offenders were rearrested in the eight-year
study period.77 In its Armed Career Criminal
Report, the Commission studied offenders
who were released between 2009 and
2011 who had been sentenced as armed

CHC VI

career criminals pursuant to ACCA and
§4B1.4. More than half (59.0%) of armed
career criminals were rearrested in the
eight-year study period.78
The Combined Impact of Age and
Criminal History
Separately, age and criminal history
are consistent predictors of recidivism.
Considered together, they are even better
predictors of recidivism. Older offenders
were rearrested at lower rates compared
to younger offenders within each CHC,
and rearrests increased within each age
category across CHC. The Commission
reported a similar pattern for offenders
released in 2005 in its 2017 Recidivism Age
Report.79 In the current study, offenders
aged 60 and older at release and in CHC I
had the lowest rearrest rate, 9.4 percent
(Table 4). The rearrest rate was twice as
high (18.2%) for CHC II offenders in that
age group. Furthermore, the rearrest

29

United States Sentencing Commission

rate was five times as high (46.9%) for
CHC VI offenders in that age group; a rate
comparable to CHC I offenders aged 21
to 29 at release (48.5%). At the other end
of the spectrum, all offenders younger
than age 21 at release and in CHCs IV, V,
and VI were rearrested. Because CHC is
determined, in part, by the number and
length of prior sentences, only 19 offenders
were under the age of 21 and in CHC IV, V,
and VI. There were 347 offenders under
the age of 21 at release in CHC I, and twothirds (66.6%) of those offenders were
rearrested, a rate comparable to CHC V
offenders aged 40 to 49 (67.4%).

Similarly, in the 2005 cohort, offenders
aged 60 and older in CHC I had the lowest
rearrest rate, 11.3 percent. In comparison,
53.0 percent of CHC I offenders in the
youngest age group in that study, younger
than 30 years of age, were rearrested.
Considering CHC VI offenders in that
study, 37.7 percent of offenders aged 60
and older were rearrested, compared to
89.7 percent of offenders younger than 30
years of age.

''

Separately, age and criminal history
are consistent predictors of recidivism.
Considered together, they are even
better predictors of recidivism.

Table 4. Rearrest Rates by Age at Release and Criminal History Category
Federal Offenders Released in 2010
CRIMINAL HISTORY CATEGORY
(CHC)
AGE
Younger than 21
21 – 29
30 – 39
40 – 49
50 - 59
60 and Older

TOTAL

30

- - - --- - ---CHC I

CHC II

CHC III

CHC IV

CHC V

CHC VI

TOTAL

66.6%

84.1%

90.7%

100.0%

100.0%

100.0%

72.5%

48.5%

66.9%

79.1%

85.4%

86.4%

89.9%

64.4%

31.2%

49.6%

61.3%

71.4%

77.3%

81.1%

54.1%

22.3%

41.6%

50.9%

62.9%

67.4%

73.1%

42.7%

15.0%

30.8%

46.4%

48.6%

61.7%

65.3%

30.8%

9.4%

18.2%

28.1%

30.0%

48.6%

46.9%

15.9%

30.2%

49.4%

61.9%

70.3%

75.4%

76.2%

Recidivism of Federal Offenders Released in 2010

ADDITIONAL CHARACTERISTICS
Offender Characteristics
As shown in Table 5, male offenders had
substantially higher rearrest rates (52.3%)
compared to female offenders (35.1%)
during the study period. Black offenders
had the highest rearrest rates (58.2%),
followed by Other Races (50.2%), Hispanic
(46.5%), and White (42.9%) offenders.
These demographic differences in rearrest
rates, however, must be considered in light
of the association between criminal history
and recidivism discussed above. Black
offenders had the smallest proportion
(30.0%) of offenders in CHC I compared
to White (52.7%), Hispanic (55.3%), and
Other Races (58.2%). Also, Black offenders
had the largest proportion (16.5%) of
offenders in CHC VI compared to White
(9.1%), Hispanic (6.5%), and Other Races
(4.6%). Rearrest rates were comparable
for offenders in the highest CHC. For
offenders in CHC VI, the rearrest rates
were as follows: Black (75.0%), White
(78.3%), Hispanic (74.1%), and Other Races
(80.8%).
Rearrest rates decreased steadily for
each successive increase in educational
level. More than half of offenders who
did not complete high school (60.8%)
or graduated high school (50.2%) were
rearrested. In comparison, 37.2 percent
of offenders with some college and
19.6 percent of college graduates were
rearrested during the study period.

Table 5. Rearrest Rates by Offender Characteristics
Federal Offenders Released in 2010
Offender Characteristics

% Rearrested

Gender
Male

52.3%

Female

35.1%

Race/Ethnicity
White

42.9%

Black

58.2%

Hispanic

46.5%

Other

50.2%

Education
Less than High School

60.8%

High School Graduate

50.2%

Some College

37.2%

College Graduate

19.6%

Rearrest rates based on these offender
characteristics were the same for offenders
released in 2005. In the 2016 Recidivism
Overview Report, the Commission reported
that 52.2 percent of male offenders and
36.4 percent of female offenders were
rearrested during the study period. In
addition, eight years after release into
the community, Black offenders had been
arrested at the highest rates (59.1%),
followed by Other Races (49.4%), Hispanic
(49.1%), and White (41.7%) offenders.
Education levels were similarly associated
with different rates of recidivism.
Offenders who did not complete high
school had the highest recidivism rates
(60.4%), followed by high school graduates
(50.7%) and those with some college
(39.3%). College graduates had the lowest
rates (19.1%).80
31

United States Sentencing Commission

Figure 17. Rearrest Rates by Federal Offense Type
Federal Offenders Released in 2010

70.6%
63.2%

59.9%

56.7%

48.2%

48.0%
41.9%
35.5%

Drug
Trafficking

Firearms

Fraud/Theft/ Immigration
Embezzlement

Offense Characteristics
Rearrest rates varied substantially by
the type of offense for which offenders
were originally sentenced (Figure 17).
Offenders sentenced for a federal firearms
offense had the highest rearrest rate, 70.6
percent, followed by robbery (63.2%) and
immigration81 (56.7%). Fewer than half of
offenders sentenced for drug trafficking
(48.0%), fraud, theft, or embezzlement
(35.5%), and all other offenses (41.9%)

Offenders sentenced for a
federal firearms offense had
the highest rearrest rate.

Robbery

All Other

Violent
Offense

were rearrested.82 In addition, offenders
who were released following sentencing
for a violent offense were more likely to be
rearrested than non-violent offenders, 59.9
percent compared to 48.2 percent (Figure
18).
These findings reflect previous
Commission findings for federal offenders
released in 2005. In its 2016 Recidivism
Overview Report, the Commission reported
the highest rearrest rates for offenders
who were released following sentences
for firearms (68.3%) and robbery (67.3%)
offenses. In addition, the lowest rearrest
rates were for offenders sentenced for
fraud (34.2%), larceny (44.4%), and all
other (42.0%) offenses.83 In addition, in its
2017 Recidivism Criminal History Report the
Commission determined that 64.1 percent
of violent offenders were rearrested during
the study period.84

Non-Viole
Offense

Recidivism of Federal Offenders Released in 2010

Figure 18. Rearrest Rates by Violent Federal Offense Type
Federal Offenders Released in 2010

63.2%

59.9%

56.7%

48.2%
41.9%

%

eft/ Immigration
ment

Robbery

All Other

Violent
Offense

The Commission examined whether
offenders in the study were consistent
in their offending patterns. To make
this determination, the Commission
compared each offender’s original federal
offense type to the most serious offense
at rearrest. These comparisons did not
demonstrate consistent patterns. Similar
to the overall findings reported above,
assault was the most common type of
rearrest for offenders originally sentenced
for drug trafficking (19.9%), firearms
(26.6%), immigration (22.8%), and all other
types offenses (20.8%) (Table 6).
A pattern demonstrating consistency
in the most common type of rearrest
and federal offense appeared for only
two groups of offenders. For offenders
originally sentenced for fraud, theft, or
embezzlement, the most common type

Non-Violent
Offense

of rearrest was larceny (18.8%) and the
second most common type of rearrest was
assault (13.8%), but the third was fraud
(13.0%). In addition, the most common
type of rearrest was robbery (24.7%) for
offenders originally sentenced for robbery.

Table 6. Federal Offense Type by
Most Common Rearrest Type
Federal Offenders Released in 2010
Federal Offense Type

Most Common
Rearrest

Drug Trafficking

Assault
(19.9%)

Firearms

Assault
(26.6%)

Immigration

Assault
(22.8%)

All Other

Assault
(20.8%)

Fraud/Theft/Embezzlement

Larceny
(18.8%)

Robbery

Robbery
(24.7%)

33

United States Sentencing Commission

Figure 19. Prevalence of Violence (Instant Federal Offense and Rearrest)
Federal Offenders Released in 2010

0%

NON-VIOLENT FEDERAL OFFENSE

Violent
29.6%

VIOLENT FEDERAL OFFENSE

Violent
45.5%

Violent offenders were more likely to be
rearrested for a violent offense compared
to non-violent offenders; 45.5 percent
of offenders originally sentenced for a
violent offense also had a violent offense
as the most serious offense at rearrest.
In comparison, 29.6 percent of offenders
sentenced for non-violent offenses had a
violent offense as the most serious offense
at rearrest (Figure 19).

''

The Commission examined whether
offenders in the study were consistent
in their offending patterns and found
that violent offenders were more likely
to be rearrested for violent offenses
compared to non-violent offenders.

34

100%

REARREST TYPE

Non-Violent
70.4%

Non-Violent
54.5%

The Commission also assessed the
relationship between various sentencing
factors and recidivism. As shown in Figure
20, most sentencing factors were not
strongly associated with rearrest. More
than one-half (54.0%) of offenders whose
original federal offense involved a weapon
were rearrested, compared to less than
half (48.7%) of offenders whose offense did
not involve a weapon. In addition, rearrest
rates for offenders with and without
sentencing adjustments for mitigating
role and acceptance of responsibility were
comparable. In contrast, a notably smaller
proportion of offenders who received an
aggravating role adjustment (36.2%) were
rearrested compared to offenders who
did not receive the adjustment (49.9%).

Recidivism of Federal Offenders Released in 2010

Sentences Imposed

While the lower rearrest rate for arguably
more culpable offenders may seem
counterintuitive, two factors associated
with lower rearrest rates, age at release
and federal offense type, characterized
offenders who received the adjustment.
First, offenders who received a sentencing
adjustment for an aggravating role were,
on average, older at release (43 years)
than offenders who did not receive the
enhancement (38 years). Second, offenders
originally sentenced for offenses with
the lowest rearrest rates, fraud, theft, or
embezzlement, comprised nearly onethird (30.2%) of offenders who received
an aggravating role enhancement. That
proportion is almost twice the proportion
of fraud, theft, or embezzlement offenders
in the overall study group of 17.4 percent.

Imprisonment
More than half (53.0%) of offenders
sentenced to imprisonment were
rearrested (Figure 21), and there was
some variation in rearrest rates comparing
length of imprisonment terms. Offenders
originally sentenced to imprisonment
terms of less than six months had the
lowest rearrest rates, 41.7 percent. In
comparison, offenders who originally were
sentenced to imprisonment terms of two
to just under five years (55.0%) and five
to just under ten years (56.2%) had the
highest rearrest rates. The rearrest rate
for offenders originally sentenced to terms
of imprisonment of ten years or longer was
comparatively low at 50.3 percent.

Figure 20. Rearrest Rates by Selected Sentencing Factors
Federal Offenders Released in 2010
■

54.0%

Applied

■

Not Applied

49.9%

48.7%

49.8%
42.3%

49.5%

45.4%

36.2%

Weapon Involvement

Aggravating Role

Mitigating Role

Acceptance of
Responsibility

35

United States Sentencing Commission

The Commission’s findings in its
2020 Recidivism Incarceration Report
provide context for this outcome. That
study included two research designs that
matched offender comparison groups
to analyze the relationship between
length of incarceration and recidivism.
The Commission used various statistical
models and consistently found that
incarceration lengths of more than ten
years were associated with lower odds of
rearrest. Two sets of analyses in that study
demonstrated that offenders incarcerated
for more than ten years were between
30 percent and 45 percent less likely
to recidivate eight years after release
compared to groups of offenders sentenced
to shorter terms of imprisonment.85
The Commission found no statistically
significant relationship between recidivism
and incarceration lengths of 60 months or

less.86 In addition, when focusing on the
shortest period of incarceration studied
(12 to 24 months), the research designs
yielded varying results, neither of which
were statistically significant nor sufficiently
reliable to make evidence-based
conclusions.87
Probation
Fewer than one-third (29.3%) of
offenders sentenced to probation were
rearrested (Figure 22). Offenders
sentenced to a probation term of any
length had lower rearrest rates compared
to offenders sentenced to prison. The
largest proportion of offenders sentenced
to probation who were rearrested (31.5%)
were sentenced to terms of three years and
five years or more.

Figure 21. Rearrest Rates by Length of Imprisonment Terms
Federal Offenders Released in 2010

Imprisonment Rearrest Rate:
53.0%

55.0%
50.8%

50.0%

6 Months to
less than 1 Year

1 to less than
2 Years

56.2%
50.3%

41.7%

Less than
6 Months

36

2 to less than
5 Years

5 to less than
10 Years

10 Years
or More

Recidivism of Federal Offenders Released in 2010

Figure 22. Rearrest Rates by Length of Probation Terms
Federal Offenders Released in 2010

Probation Rearrest Rate:
29.3%

31.5%

26.5%

Less than
2 Years

31.5%

25.0%

2 to less than
3 Years

22.5%

3 to less than
4 Years

4 to less than
5 Years

5 Years
or More

Supervised Release
Nearly all (99.4%) offenders sentenced
to a term of imprisonment also were
sentenced to a term of supervised release,
and there was some variation in rearrest
rates comparing length of supervised
release terms (Figure 23). Offenders

sentenced to less than two years of
supervised release had the lowest rearrest
rates, 46.4 percent, compared to 56.3
percent for offenders originally sentenced
to three to just under four years of
supervised release.

Figure 23. Rearrest Rates by Length of Supervised Release Terms
Federal Offenders Released in 2010

Supervised Release
Rearrest Rate:
52.9%

56.3%

46.4%

47.5%

Less than
2 Years

2 to less than
3 Years

3 to less than
4 Years

53.4%

50.1%

49.7%

48.7%

4 to less than
5 Years

5 to less than
10 Years

10 Years
or more

Life

37

United States Sentencing Commission

CONCLUSION

This study examined the recidivism of
federal offenders released in 2010 during
an eight-year follow-up period. Almost
one-half (49.3%) of federal offenders were
rearrested during the study period. This
recidivism rate is identical to the rate
reported by the Commission for federal
offenders released in 2005. Notably, this
consistency ensued despite two substantial
changes in the federal sentencing
landscape: the change from a mandatory
to an advisory guideline system following
Booker, and the increasing reliance on
evidence-based practices in federal
supervision. Other recidivism patterns
also were the same for the two offender
cohorts.

For More
Information
Visit the Commission’s website
for additional resources on
recidivism.
www.ussc.gov

38

Consistent with its prior reports,88
the Commission found that the combined
factors of age and criminal history were
strongly associated with recidivism. All
offenders in the study aged 21 and younger
in CHC IV through VI were rearrested,
compared to 9.4 percent of offenders
aged 60 and older in CHC I. In addition,
offenders who originally were sentenced
for firearms offenses had the highest
rearrest rates of any offense type and
offenders originally sentenced for a violent
offense had higher rearrest rates compared
to non-violent offenders. The study
also examined the relationships of other
sentencing and offender characteristics
that also were consistent with prior studies.

Related Reports
Recidivism Series
2005 Study Cohort
Published 2016 - 2020

APPENDICES

United States Sentencing Commission

APPENDIX A
Methodology
The Commission entered into a data
sharing agreement with the FBI’s Criminal
Justice Information Services (CJIS) Division
and the Administrative Office of the
United States Courts (AOUSC) to provide
the Commission with secure electronic
access to criminal history records through
CJIS’s Interstate Identification Index
(III) and International Justice and Public
Safety Network (NLETS). Results received
using this system provide an individual’s
Criminal History Record Information
(CHRI) maintained by all U.S. states, the
District of Columbia, U.S. territories, and
federal agencies. Once the raw CHRI
was obtained, the Commission organized
and standardized the arrest and court
disposition information into an analytical
dataset. The resulting data contained CHRI
for 32,135 offenders with valid identifying
information and who were released in
2010.
Identifying the Study Cohort
The study cohort includes all federal
offenders who were U.S. citizens and
released from federal prison after serving
a sentence of imprisonment or placed on
probation in 2010. For offenders released
from prison, the Bureau of Prisons (BOP)
provided release dates and identifying
information for all offenders released
in 2010. The Commission identified
offenders placed on probation in 2010
40

and, with the assistance of the AOUSC,
identified and removed offenders who died
while on supervised release during the
recidivism follow-up period.
Processing the Criminal History
Record Information
The Commission entered into a data
sharing agreement with the FBI’s CJIS
Division and the AOUSC to acquire
electronic records of offender CHRI. The
AOUSC extracted offender CHRI through
its Access to Law Enforcement System
(ATLAS) which provides an interface to
III and NLETS. The III allows authorized
agencies to determine whether any
federal or state repository has CHRI on
an individual. Agencies can then securely
access specific state CHRI through NLETS.
As a result, ATLAS collects CHRI from all
state and federal agencies.
The ATLAS system returns the literal
text in the RAP sheets in the format
in which the original records appear:
dates of criminal justice system actions
(e.g., arrests); offense categories which
indicate the charges in the terminology
used by that agency (e.g., text strings or
numeric categories); subsequent action
tied to arrest charges (e.g., charges filed
by prosecutors, court findings of guilt,
etc.); and sentencing and corrections
information. All of these records are
subject to availability from the originating
source.

Recidivism of Federal Offenders Released in 2010

The ATLAS system also “parses”
records from RAP sheets received from
all 50 states, the District of Columbia, and
federal agencies. Parsing records involves
organizing key data elements into logical
components, for example: arrest, court, and
correctional events. Key data elements
include offender identifiers, dates of key
actions (e.g., arrests and convictions), the
criminal charges, and outcomes, such as
convictions and sentencing information,
when provided by the courts. The parsing
process collates the multi-state records
into a uniform structure, regardless of the
state, for all individuals with a valid FBI
number who were found in one or more
repositories across the country.
Standardizing the Criminal Records
After acquiring offender CHRI, the
Commission contracted with Integrity
One Partners (IOP) to consolidate records
for each offender and remove duplicative
or extraneous material.89 Following
this preliminary process, IOP utilized a
crosswalk created for the Commission’s
prior recidivism research90 to standardize
offense codes across states and federal
agencies. The crosswalk was updated
to standardize new offense codes not
mapped in the original crosswalk. The
crosswalk standardizes arrest and court
codes, regardless of originating sources,
into a common framework for analysis.
This step was needed because criminal

record repositories are primarily designed
to store records in ways that accurately
reflect the requirements of each state or
federal repository, such as the criminal
code for that jurisdiction. As a result,
any two repositories are likely to use
many unique text strings to indicate the
nature of the criminal charges and actions
taken in response to those charges. Thus,
standardizing the offense information was
necessary for cross-jurisdictional analysis.
Within each arrest cycle, arrest charges
were categorized using standardized
codes. A charge severity index was created
which incorporates both criminal law
classification (e.g., felony or misdemeanor)
and offense severity. Offenses were first
classified into standardized subcategories.
These subcategories were then further
grouped for analytical purposes into one
of 20 major crime categories in ranking
order by severity. For each offender, the
most severe major crime category was
identified in their arrest information. The
rearrest categories and their underlying
subcategories are provided in Table A.

41

United States Sentencing Commission

Table A. Rearrest Offense Categories and Charges

Murder

Murder of public officer
Murder
Attempted murder
Unspecified manslaughter/homicide
Nonnegligent manslaughter/homicide

Sexual Assault

Rape
Forcible sodomy
Fondling
Statutory rape
Luring minor by computer
Other sexual assault
Sexual assault unspecified

Robbery

Armed robbery
Robbery unspecified
Unarmed robbery

Assault

Aggravated/felony assault
Simple/misdemeanor assault
Assault unspecified
Assault of public officer
Intimidation
Hit and run driving with bodily injury
Intimidating a witness

Other Violent

Kidnapping
Blackmail/Extortion
Rioting
Child abuse
Other violent offense
Arson

Drug Trafficking

Trafficking cocaine/crack
Trafficking heroin
Trafficking marijuana
Trafficking methamphetamine
Trafficking other/unspecified controlled substance

Burglary

Burglary

42

Recidivism of Federal Offenders Released in 2010

Larceny

Motor vehicle theft
Grand/felony larceny
Petty/misdemeanor larceny
Larceny unspecified
Receiving stolen property
Trafficking stolen property
Unauthorized use of vehicle

Fraud

Fraud/forgery
Identity theft
Embezzlement
Bribery

Other Property

Destruction of property
Hit and run with property damage
Trespassing
Possession of burglary tools
Other property offense

Drug Possession

Possession of cocaine/crack
Possession of heroin
Possession of marijuana
Possession of methamphetamine
Possession of other/unspecified controlled substance

Other Drug

Unspecified cocaine/crack offense
Unspecified heroin offense
Unspecified marijuana offense
Unspecified methamphetamine offense
Unspecified other/unspecified drug offense

Weapon

Weapon offense

Other Sex Offense

Morals offense
Indecent exposure
Commercialized vice
Contributing to the delinquency of a minor

DUI/DWI

Driving while intoxicated/under the influence,
Substance unspecified
Driving while intoxicated/under the influence, alcohol
Driving while intoxicated/under the influence, drugs

Immigration

Immigration offense

43

United States Sentencing Commission

Administration of Justice Offenses

Escape from custody
Flight to avoid prosecution
Warrant
Contempt of court
Failure to appear
Violation of restraining order
Other court offense
Prison contraband offense
Sex offender registry offense
Obstruction of justice

Probation/Parole/
Supervised Release Violation

Parole violation
Unspecified probation/parole violation
Probation violation

Public Order Offenses

Family-related offense
Drunkenness/vagrancy/disorderly conduct
Invasion of privacy
Liquor law violation
Other public order offense
Curfew violation

Other/Unspecified Offenses

Vehicular manslaughter/homicide
Negligent (involuntary) manslaughter/homicide
Habitual offender
Runaway
Truancy
Ungovernability
Status liquor law violation
Miscellaneous status offense
Other offense
Unspecified inchoate offense
Military offense
Not applicable
Unspecified offense

44

Recidivism of Federal Offenders Released in 2010

APPENDIX B
Table B-1. Selected Offender Characteristics
Federal Offenders Released in 2010 and 2005
Release Year
Offender Characteristics

2010

2005

White

41.2%

43.7%

Black

35.8%

33.9%

Hispanic

18.1%

17.8%

4.9%

4.6%

Male

82.7%

81.7%

Female

17.3%

18.3%

Less than High School

34.9%

34.2%

High School Graduate

38.4%

36.9%

Some College

20.3%

21.4%

6.4%

7.5%

Average

35 Years

35 Years

Median

33 Years

33 Years

Average

38 Years

38 Years

Median

37 Years

36 Years

Race/Ethnicity

Other
Gender

Education

College Graduate
Age at Sentencing

Age at Release

45

United States Sentencing Commission

Table B-2. Selected Offense Characteristics and Sentencing Information
Federal Offenders Released in 2010 and 2005

Release Year

2010

2005

Drug Trafficking

43.1%

41.6%

Firearms

15.3%

12.8%

Fraud/Theft/Embezzlement

17.4%

13.6%

Immigration

4.4%

3.5%

Robbery

4.1%

4.3%

15.7%

24.2%

11.7%

9.8%

5.0%

6.4%

8.4%

10.6%

92.2%

90.7%

CHC I

45.3%

53.6%

CHC II

11.9%

12.2%

CHC III

15.8%

14.3%

CHC IV

9.8%

7.9%

CHC V

6.1%

4.4%

CHC VI

11.1%

7.6%

Prison

84.4%

81.2%

Probation

15.6%

18.8%

Average

51 Months

45 Months

Median

36 Months

30 Months

Average

59 Months

55 Months

Median

45 Months

37 Months

Offense Characteristics and Sentencing Information
Federal Offense Type*

All Other Offenses
Weapon Involvement
Enhancement Applied
Aggravating Role
Enhancement Applied
Mitigating Role
Reduction Applied
Acceptance of Responsibility
Reduction Applied
Criminal History Category

Sentence Type

Sentence Length

Imprisonment Length

* The Commission used slightly different offense classification schemes for federal offenses in the two analyses. The
offense type categories for this report were derived from the 30 Type of Crime categories the Commission reports
in its Sourcebook of Federal Sentencing Statistics. See U.S. Sent’g Comm’n, 2020 Sourcebook of Federal Sentencing
Statistics 210–14 (2021), https://www.ussc.gov/sites/default/files/pdf/research-and-publications/annual-reportsand-sourcebooks/2020/2020-Annual-Report-and-Sourcebook.pdf. The offense type categories reported in the
2016 Recidivism Overview Report were derived from the 33 Primary Offense categories the Commission previously
reported in its Sourcebook of Federal Sentencing Statistics. See U.S. Sent’g Comm’n, 2020 Sourcebook of Federal Sentencing
Statistics 167–70 (2018), https://www.ussc.gov/sites/default/files/pdf/research-and-publications/annual-reports-andsourcebooks/2017/2017SB_Full.pdf. To facilitate comparison, the offense types for offenders in the 2005 cohort were
updated in this table to conform to the classification scheme used for the 2010 cohort.

46

Recidivism of Federal Offenders Released in 2010

APPENDIX C
Figure C. Distribution of Other Offense Types at Original Sentencing
Federal Offenders Released in 2010
FRAUD/THEFT/EMBEZZLEMENT
17.4%

DRUG TRAFFICKING
43.1%
FIREARMS
15.3%

OTHER
OFFENSES
15.7%

IMMIGRATION
4.4%
ROBBERY
4.1%

Distribution of Other Offenses
Money Laundering

1.7%

Administration of Justice

1.6%

Child Pornography

1.5%

Tax

1.5%

Forgery/Counterfeit/Copyright

1.5%

Assault

1.0%

Sexual Abuse

0.9%

All Other

6.0%

47

United States Sentencing Commission

APPENDIX D
Table D-1. Selected Offender Characteristics
Federal Offenders Released in 2010 With No Rearrests
Offender Characteristics
Race/Ethnicity
White

46.4%

Black

29.6%

Hispanic

19.1%

Other

4.9%

Gender
Male

77.9%

Female

22.1%

Education
Less than High School

27.0%

High School Graduate

37.7%

Some College

25.2%

College Graduate

10.1%

Age at Sentencing
Average

39 Years

Median

37 Years

Age at Release

48

Average

41 Years

Median

40 Years

Recidivism of Federal Offenders Released in 2010

Table D-2. Selected Offense Characteristics and Sentencing Information
Federal Offenders Released in 2010 With No Rearrests
Offense Characteristics and Sentencing Information
Federal Offense Type
Drug Trafficking

44.2%

Firearms

8.9%

Fraud/Theft/Embezzlement

22.1%

Immigration

3.8%

Robbery

3.0%

All Other Offenses

18.0%

Criminal History Category
CHC I

62.4%

CHC II

11.8%

CHC III

11.9%

CHC IV

5.7%

CHC V

3.0%

CHC VI

5.2%

Sentence Type
Prison

78.3%

Probation

21.7%

49

United States Sentencing Commission

APPENDIX E
Table E. Rearrest Rates by Federal Offense Type
Federal Offenders Released in 2010
Offenders Rearrested

Total Offenders
Federal Offense Type
Drug Trafficking
Firearms
Fraud/Theft/Embezzlement
Robbery
Immigration

N

-~-=13,835
4,917

■
■

5,580
1,322
1,431

I
I

N
6,641
3,469

48.0
70.6

%

1,980
835
812

35.5
63.2
56.7

280

58.3

Assault
Child Pornography

321
494

223
204

69.5
41.3

Administration of Justice

510
368

189

37.1

Other

184

50.0

Sexual Abuse
Prison Offenses

283
160

159
133

56.2
83.1

Obscenity/Other Sex Offenses
Money Laundering

184
539

131
127

Drug Possession

84

Stalking/Harassing
Tax

165
103
483

Extortion/Racketeering

Forgery/Counterfeit/Copyright

480

63
61

151

43

Arson
Burglary/Trespass

71
46

34
32

Manslaughter

47

31

Murder
Bribery/Corruption

68
236

31
31

Commercialized Vice
Individual Rights

111
67

20
17

Environmental

93

17

Kidnapping
National Defense

20
21

11
5

Food and Drug
Antitrust

24
5

4
0

50

-•
•---

71.2
23.6
50.9
61.2
12.6
28.5
47.9
69.6
66.0
45.6
13.1
18.0
25.4
18.3
55.0
23.8
16.7
0.0

ENDNOTES

United States Sentencing Commission

ENDNOTES
1

28 U.S.C. § 995(a)(12)–(14).

2
543 U.S. 220 (2005) (striking the mandatory provision of 18 U.S.C. § 3553(a) and rendering the
guidelines advisory).
3
Kim Steven Hunt & Robert Dumville, U.S. Sent’g Comm’n, Recidivism Among Federal Offenders: A
Comprehensive Overview (2016), https://www.ussc.gov/sites/default/files/pdf/research-and-publications/
research-publications/2016/recidivism_overview.pdf [hereinafter 2016 Recidivism Overview Report].
4

Id. at 15.

5
U.S. Sent’g Comm’n, Supplementary Report on the Initial Sentencing Guidelines and Policy Statements
41–44 (1987), https://www.ussc.gov/sites/default/files/pdf/guidelines-manual/1987/manual-pdf/1987_
Supplementary_Report_Initial_Sentencing_Guidelines.pdf. See also William Rhodes, Christina Dyous, Ryan
Kling, Dana Hunt & Jeremy Luallen, ABT Assoc., Recidivism of Offenders on Federal Community Supervision
(2012), https://www.ojp.gov/pdffiles1/bjs/grants/241018.pdf [hereinafter Rhodes]; Alfred Blumstein, David P.
Farrington & Soumyo Moitra, Delinquency Careers: Innocents, Desisters, and Persisters, 6 Crime & Just. 187 (1985)
(describing two major cohort studies and the recidivism probability of youthful offenders increasing with
successive involvements with law enforcement); Paul Gendreau, Tracy Little & Claire Goggin, A Meta-Analysis of
the Predictors of Adult Offender Recidivism: What Works!, 34 Criminology 575 (1996) (explaining meta-analysis of
131 studies which found the strongest predictors of recidivism included criminogenic needs, criminal history,
social achievement, age/gender/race, and family factors) [hereinafter Gendreau].
6

See 2016 Recidivism Overview Report, supra note 3, at 5.

7
Louis Reedt, Kim Steven Hunt, James L. Parker, Melissa K. Reimer & Kevin T. Maass, U.S. Sent’g Comm’n,
Recidivism Among Federal Drug Trafficking Offenders (2017), https://www.ussc.gov/sites/default/files/
pdf/research-and-publications/research-publications/2017/20170221_Recidivism-Drugs.pdf; Matthew
J. Iaconetti, Tracey Kyckelhahn & Mari McGilton, U.S. Sent’g Comm’n, Recidivism Among Federal Firearms
Offenders (2019), https://www.ussc.gov/sites/default/files/pdf/research-and-publications/researchpublications/2019/20190627_Recidivism_Firearms.pdf [hereinafter 2019 Recidivism Firearms Report]; Kim
Steven Hunt, Matthew J. Iaconetti & Kevin T. Maass, U.S. Sent’g Comm’n, Recidivism Among Federal Violent
Offenders (2019), https://www.ussc.gov/sites/default/files/pdf/research-and-publications/‌researchpublications/2019/20190124_Recidivism_Violence.pdf [hereinafter 2019 Recidivism Violence Report].
8

Tracey Kyckelhahn & Trishia Cooper, U.S. Sent’g Comm’n, The Past Predicts the Future: Criminal History

and Recidivism of Federal Offenders (2017), https://www.ussc.gov/sites/default/files/pdf/research-and-

publications/research-publications/2017/20170309_Recidivism-CH.pdf [hereinafter 2017 Recidivism Criminal
History Report]; Kim Steven Hunt & Billy Easley, U.S. Sent’g Comm’n, The Effects of Aging on Recidivism Among
Federal Offenders (2017), https://www.ussc.gov/sites/default/files/pdf/research-and-publications/researchpublications/2017/20171207_Recidivism-Age.pdf [hereinafter 2017 Recidivism Age Report]; Ryan Cotter, U.S.
Sent’g Comm’n, Length of Incarceration and Recidivism (2020), https://www.ussc.gov/sites/default/files/pdf/
research-and-publications/research-publications/2020/20200429_Recidivism-SentLength.pdf [hereinafter
2020 Recidivism Incarceration Report].
9

See case cited supra note 2 and accompanying text.

10
Jonathan E. Hurtig & Lisa Marie Lenart, The Development of the Evidence-Based Practice Blue Print and
Where We are Now, 75 Fed. Prob. J. (2011), https://www.uscourts.gov/sites/default/files/75_2_6_0.pdf; Melissa

52

Recidivism of Federal Offenders Released in 2010
Alexander, Bradley Whitley & Christopher Bersch, Driving Evidence-Based Supervision to the Next Level: Utilizing
PCRA, “Drivers,” and Effective Supervision Techniques, 78 Fed. Prob. J. 2 (2014), https://www.uscourts.gov/sites/
default/files/probation_dec_2014_1219b.pdf.
11
Prob. & Pretrial Servs. Off., Admin. Off. of the U.S. Cts., An Overview of the Federal Post Conviction
Risk Assessment (2018), https://www.uscourts.gov/sites/default/files/overview_of_the_post_conviction_risk_
assessment_0.pdf.
12
Nat’l Inst. of Just., U.S. Dep’t of Just., Recidivism, (June 17, 2014), https://web.archive.org/
web/20160120175242/http://www.nij.gov/topics/corrections/recidivism/pages/welcome.aspx; see also
Michael D. Maltz, Recidivism 1, 54 (2001) [hereinafter Maltz].
13

See Maltz, supra note 12, at 7–20; see also Ryan King & Brian Elderboom, Urb. Inst., Improving Recidivism

as a Performance Measure (2014), https://www.bja.gov/Publications/UI-ImprovingRecidivism.pdf.

14
See, e.g., Christopher T. Lowenkamp, Marie VanNostrand & Alexander Holsinger, Investigating The Impact
Of Pretrial Detention On Sentencing Outcomes (2013), https://craftmediabucket.s3.amazonaws.com/uploads/
PDFs/LJAF_Report_state-sentencing_FNL.pdf.
15
See Maltz, supra note 12, at 61–64. See also Nat’l Inst. of Just., U.S. Dep’t of Just., Measuring Recidivism
(Feb. 20, 2008), https://web.archive.org/web/20160129195540/http://www.nij.gov/topics/corrections/
recidivism/pages/measuring.aspx.
16
See, e.g., Mariel Alper, Matthew R. Durose & Joshua Markman, Bureau of Just. Stat., U.S. Dep’t of Just.,
Update of Prisoner Recidivism: A 9-Year Follow-up Period (2005–2014) (2018), https://www.bjs.gov/content/
pub/pdf/18upr9yfup0514.pdf [hereinafter Alper]; Admin. Off. of the U.S. Cts., Just the Facts: Post-Conviction
Supervision and Recidivism, (Oct. 22, 2018), https://www.uscourts.gov/news/2018/10/22/just-facts-postconviction-supervision-and-recidivism#chart1; Rhodes, supra note 5.
17

See Maltz, supra note 12, at 55–60.

18

See Maltz, supra note 12, at 56–58.

19

See Alper, supra note 16.

20
The Commission collects and analyzes data on federal sentences to carry out its statutory
responsibilities. As authorized by Congress, the Commission’s research responsibilities include: (1) the
establishment of a research and development program to serve as a clearinghouse and information center
for the collection, preparation, and dissemination of information on federal sentencing practices, (2) the
publication of data concerning the sentencing process, (3) the systematic collection and dissemination of
information concerning sentences actually imposed and the relationship of such sentences to the sentencing
factors in 18 U.S.C. § 3553(a), and (4) the systematic collection and dissemination of information regarding
the effectiveness of sentences imposed. See 28 U.S.C. § 995(a)(12), (14)–(16). The Commission collects
information for every federal felony and Class A misdemeanor offense sentenced each year. Sentencing courts
are statutorily required to submit five sentencing documents to the Commission within 30 days of entry of
judgment in a criminal case, including: (1) the charging document, (2) the plea agreement, (3) the Presentence
Report, (4) the Judgment and Commitment order, and (5) the Statement of Reasons form. See 28 U.S.C. §
994(w)(1). For each case in its Individual Offender Datafile, the Commission routinely collects case identifiers,
sentencing data, demographic variables, statutory information, the complete range of court guideline
application decisions, and departure and variance information from these documents.
21
The data utilized in the course of conducting analyses in this report includes information obtained
pursuant to an interagency agreement with the FBI, which prohibits the Commission from releasing the
dataset.
22

Appendix A provides a detailed description of the data collection methodology.

53

United States Sentencing Commission

23
This includes any offenders released from BOP on detainer, which ordinarily indicates transfer of
custody to state court or to a state correctional facility following completion of their federal sentence.
24
Offenders were excluded from various analyses in this report due to missing information for the
variables required for those analyses.
25

Appendix B provides comparisons of offender and offense characteristics for the two study groups.

26
The 1,586 offenders comprising the Other Race category were American Indian/Alaskan Native
(52.0%), Asian or Pacific Islander (41.8%), or Multi-racial/Other (6.2%).
27
The median is the midpoint; half of the offenders were below the median and the remaining half of
offenders were above the median.
28

See 2016 Recidivism Overview Report, supra note 3, at 9.

29
The offense type categories were derived from the 30 Type of Crime categories the Commission
reports in its Sourcebook of Federal Sentencing Statistics. See U.S. Sent’g Comm’n, 2020 Sourcebook of Federal
Sentencing Statistics 210–14 (2021), https://www.ussc.gov/sites/default/files/pdf/research-and-publications/
annual-reports-and-sourcebooks/2020/2020-Annual-Report-and-Sourcebook.pdf.
30
U.S. citizens sentenced for immigration offenses primarily are sentenced under §2L1.1 (Smuggling,
Transporting, or Harboring an Unlawful Alien). See id. at 132.
31
Offense type categories were selected for analysis that comprised more than four percent of the total.
Information for the offense types comprising the “other” category is provided in Appendix C.
32
The Commission defined violent offenses based on guideline application. In cases where multiple
Chapter Two guidelines applied because the offender had multiple counts of conviction for different offenses,
this report identifies the offender as violent if any of the guidelines that applied was for a violent offense,
regardless of whether that guideline ultimately produced the highest offense level (i.e., the primary guideline).
See U.S. Sent’g Comm’n, Guidelines Manual, §1B1.5(a), comment. (n.1) (Nov. 2018) [hereinafter USSG].
33
The offense type categories reported in the 2016 Recidivism Overview Report were derived from the
33 Primary Offense categories the Commission previously reported in its Sourcebook of Federal Sentencing
Statistics. See U.S. Sent’g Comm’n, 2017 Sourcebook of Federal Sentencing Statistics 167–70 (2018), https://www.
ussc.gov/sites/default/files/pdf/research-and-publications/annual-reports-and-sourcebooks/2017/2017SB_
Full.pdf.
34

See 2016 Recidivism Overview Report, supra note 3, at 10.

35
Section 924(c) provides incremental, consecutive mandatory minimum penalties for the use, carrying,
or possession of a firearm during or in furtherance of a crime of violence or a drug trafficking crime. See 18
U.S.C. § 924(c).
36
Weapon involvement does not include: (1) cases in which a weapon is present in the offense, but the
offender was not convicted of 18 U.S.C. § 924(c) or did not receive a weapon-related sentencing enhancement,
(2) cases in which the specific enhancement can be applied for multiple reasons (e.g., the specific enhancement
can be applied if the offense involved either physical contact or if a dangerous weapon was possessed), or (3)
cases sentenced as weapon offenses under Chapter Two, Part K of the Guidelines Manual (unless they were
convicted of 18 U.S.C. § 924(c)).
37
USSG §3B1.1. Section 3B1.1 provides for a 4-level increase for an offender who was an organizer
or leader of a criminal activity that involved five or more participants or was otherwise extensive, a 3-level
increase for a manager or supervisor of criminal activity that involved five or more participants or was

54

Recidivism of Federal Offenders Released in 2010
otherwise extensive, and a 2-level increase for an organizer, leader, manager, or supervisor in any criminal
activity than otherwise described. Id.
38
USSG §3B1.2. Section 3B1.2 provides for a 4-level decrease for an offender who was a minimal
participant in any criminal activity, a 2-level decrease for any offender who was a minor participant in any
criminal activity, and a 3-level decrease for offenders falling between the two. Id.
39
USSG §3E1.1. Section 3E1.1 provides for a 2-level decrease if the offender clearly demonstrates
acceptance of responsibility for the offense. An additional 1-level decrease can apply if the offense level
determined prior to the 2-level decrease is level 16 or greater, and upon motion of the government stating
that the defendant has assisted authorities in the investigation or prosecution of his own misconduct by timely
notifying authorities of his intention to enter a plea of guilty. Id.
40
USSG Ch.4. Other considerations include the type of offense (certain minor offenses are excluded
from the criminal history score) and the length of time between the prior sentence and the instant federal
offense. The guidelines exclude certain prior convictions based on factors such as the type of offense (e.g., fish
and game violations), disposition (e.g., diversionary dispositions without a finding of guilt), or remoteness of the
conviction. See USSG §4A1.2.
41

USSG Ch.5, Pt.A (Sentencing Table).

42
Of the 11,338 offenders with zero criminal history points, the remaining 52.1% had prior contact
with the criminal justice system; that is, they had prior arrests that did not lead to convictions or they had
convictions that did not receive points due to the age of the conviction or the minor nature of the offense.
43
In some instances, the percentage of offenders comprising the criminal history points categories do
not sum to the total percentage of offenders comprising the associated CHC. This is because some guidelines
provide for a CHC that supersedes the CHC determined by criminal history points. See, e.g., USSG §4B1.1(b).
Additionally, in a small number of cases, the CHC is determined by court findings and differs from the calculated
criminal history points.
44
By comparison, the corresponding rates for offenders released in 2005 were as follows: CHC I
(53.6%), CHC II (12.2%), CHC III (14.3%), CHC IV (7.9%), CHC V (4.4%), and CHC VI (7.6%). 2016 Recidivism
Overview Report, supra note 3, at 10.
45

USSG §4B1.1(a).

46

See 18 U.S.C. § 924(e); USSG §4B1.4.

47
Of the 1,211 offenders sentenced under these provisions, 81.7% were sentenced as career offenders,
18.1% were sentenced as armed career criminals, and 0.2% were sentenced under both provisions.
48

See case cited supra note 2 and accompanying text.

49
The four types of sentences correspond to the four “Zones” (A–D) in the Sentencing Table of the
Guidelines Manual. See USSG Ch.5, Pt.A (Sentencing Table); see also USSG §§5B1.1, 5C1.1 (setting forth the
sentencing options for Zones A–D). Zone A authorizes probation only; Zone B authorizes probation with a
condition of confinement; Zone C authorizes a “split” sentence of imprisonment and community confinement
(e.g., home detention or a halfway house); and Zone D authorizes sentences of imprisonment only. See USSG
§§5B1.1, 5C1.1.
50
See 2016 Recidivism Overview Report, supra note 3, at 11. As in the 2016 Recidivism Overview Report,
the Commission analyzed the sentences imposed for offenders sentenced to alternatives for the current study.
Similar to the prior analysis, offenders in the current study who were sentenced to prison with some type of
alternative confinement had a median term of imprisonment of five months. In addition, nearly all (99.0%)
offenders sentenced to a Zone B sentence of probation with some type of alternative confinement were

55

United States Sentencing Commission
sentenced to home confinement or community confinement (e.g., a halfway house) rather than confinement in a
jail or prison.
51
The median sentence imposed was 36 months. Sentence imposed includes any time served amounts
and imprisonment under §5G1.3. Probation sentences are included as zero months. Any portion of a sentence
that is an alternative confinement as described in §5C1.1 is included. Life sentences and sentences exceeding
470 months are included in the calculation as 470 months.
52
Twelve offenders among those sentenced to ten years or longer were originally sentenced to
imprisonment terms of life. They are included in the release cohort because they were subsequently
resentenced to reduced terms. Sentences reported in this study are the sentences originally imposed.
The Commission did not collect resentencing information prior to fiscal year 2006; therefore, complete
resentencing information is not available for all offenders in the study.
53
USSG §5D1.1(a)–(c); 18 U.S.C. § 3583(a). Convictions for drug trafficking offenses and certain
kidnapping and sex offenses require a term of supervised release. See, e.g., 21 U.S.C. §§ 841, 846; 18 U.S.C.
§ 3583(k).
54

Selected data for offenders in the study who were not rearrested is provided in Appendix D.

55

2016 Recidivism Overview Report, supra note 3, at 15.

56

See infra note 63 and accompanying text.

57

2016 Recidivism Overview Report, supra note 3, at 11.

58
The rearrest rates for offenders released in 2010 were marginally different for offenders sentenced
before (53.5%) and after (48.5%) the Booker decision. Having been sentenced prior to January 12, 2005,
offenders sentenced pre-Booker naturally had longer sentences compared to offenders sentenced postBooker. Attributes related to their longer sentences also are associated with recidivism. For example, a smaller
proportion of pre-Booker offenders (19.0%) were in CHC I than post-Booker offenders (50.7%). In addition,
a larger proportion of pre-Booker offenders (17.9%) than post-Booker offenders (7.7%) were sentenced for a
violent offense. Conversely, a smaller proportion of pre-Booker offenders (1.7%) than post-Booker offenders
(20.6%) were sentenced for fraud, theft, or embezzlement.
59

2016 Recidivism Overview Report, supra note 3, at 16.

60
Courts originally imposed supervision terms of eight years or longer (including life) for only 2.7% of
offenders in the study.
61
This measurement is based on the supervision term imposed at the time of original sentencing and
does not account for any changes in supervision status following release. Such information was not available in
the data used for this study. Therefore, if a court terminated an offender’s supervision prior to the expiration
of the term initially imposed, that offender would still be considered under supervision for this analysis.
Alternatively, if a court extended an offender’s supervision beyond the term originally imposed, that offender
would be considered to have completed the supervision term for purposes of this analysis.
62

2016 Recidivism Overview Report, supra note 3, at 9, 31 n.24.

63
Accordingly, the data should not be interpreted to represent the overall frequency of the listed offense
among rearrests.
64
The “other violent” category includes violent offenses that do not fit into any of the specific violent
categories. It includes arson, blackmail/extortion, child abuse, kidnapping, rioting, and any other unspecified
violent offense.

56

Recidivism of Federal Offenders Released in 2010
65
The Commission separated rearrests for administration of justice offenses (e.g., failure to appear,
resisting arrest without violence, and tampering with evidence), probation, parole, and supervision
violations, and other sex offenses from public order offenses (e.g., disorderly conduct, vice crimes, and public
drunkenness) for this analysis, whereas the Commission previously combined those offense types into a single
“public order” category. See 2017 Recidivism Age Report, supra note 8, at 34 n.6. See also Courtney Semisch,
Kristen Sharpe & Alyssa Purdy, U.S. Sent’g Comm’n, Federal Armed Career Criminals: Prevalence, Patterns, and
Pathways 80 n.68 (2021), https://www.ussc.gov/sites/default/files/pdf/research-and-publications/researchpublications/2021/20210303_ACCA-Report.pdf [hereinafter Armed Career Criminal Report]. Combined, the
current findings for administration of justice, probation, parole, and supervision violations, other sex offenses,
and public order offenses generally are analogous to the 2016 finding for public order offenses. Reported
separately, or in combination, the Commission ranks these offense types low in its severity ranking. In addition,
the four separate offense categories reported in this study comprised the previously existing public order
category, making them largely comparable.
66

2016 Recidivism Overview Report, supra note 3, at 17.

67
See 2016 Recidivism Overview Report, supra note 3, at 18–19, 23. See also 2017 Recidivism Criminal
History Report, supra note 8; 2017 Recidivism Age Report, supra note 8.
68

2017 Recidivism Age Report, supra note 8, at 22–23.

69

Id. at 22.

70

Id. at 25.

71

See supra notes 39–40 and accompanying text.

72
2016 Recidivism Overview Report, supra note 3, at 18; 2017 Recidivism Criminal History Report, supra
note 8, at 7.
73
2016 Recidivism Overview Report, supra note 3, at 18–19; 2017 Recidivism Criminal History Report, supra
note 8, at 7–8.
74

2017 Recidivism Criminal History Report, supra note 8, at 9.

75

See supra notes 45–46 and accompanying text.

76

See USSG §§4B1.1(b), 4B1.4(c).

77
U.S. Sent’g Comm’n, Report to the Congress: Career Offender Sentencing Enhancements 39 (2016), www.
ussc.gov/sites/default/files/pdf/news/congressional-testimony-and-reports/criminal-history/201607_RtCCareer-Offenders.pdf.
78

Armed Career Criminal Report, supra note 65, at 43.

79

2017 Recidivism Age Report, supra note 8, at 25.

80

See 2016 Recidivism Overview Report, supra note 3, at 24.

81

See supra note 30.

82

Rearrest rates for the offense types comprising the “All Other” category are provided in Appendix E.

83
2016 Recidivism Overview Report, supra note 3, at 20. The Commission also reported on the recidivism
of firearms offenders in its 2019 Recidivism Firearms Report. Offenders in that study were designated as

57

United States Sentencing Commission
firearms offenders who: 1) were sentenced under §2K2.1, the primary sentencing guideline for unlawful
receipt, possession or transportation of firearms or ammunition, and prohibited transactions involving
firearms or ammunition, 2) sentenced pursuant to 18 U.S.C. § 924(e), the Armed Career Criminal Act (“ACCA”),
3) sentenced as career offenders pursuant to §4B1.1 who were also convicted of a federal firearms crime as
part of the instant offense, or 4) convicted under 18 U.S.C. § 924(c) who did not otherwise receive the ACCA
or career offender enhancement. All other offenders in the study were designated as non-firearms offenders.
Because of this specific definition, the firearms offenders in this study are not directly comparable to those in
the previous study. Nevertheless, 68.1% of firearms offenders in the previous study were rearrested. 2019
Recidivism Firearms Report, supra note 7, at 2, 17.
84
2017 Recidivism Criminal History Report, supra note 8, at 11–12. The Commission also reported on the
recidivism of violent offenders in its 2019 Recidivism Violence Report. Offenders in that study were designated
as violent who engaged in violent criminal conduct as part of their instant offense or had been arrested for a
violent offense in their past. In addition, offenders were designated as non-violent who neither engaged in a
violent instant offense nor been arrested for a violent crime in their past. Because of this specific definition, the
violent offenders in this study are not directly comparable to those in the previous study. Nevertheless, in the
previous study, 63.8% of violent offenders were rearrested, compared to 39.8% of non-violent offenders. 2019
Recidivism Violence Report, supra note 7, at 3–4.
85

2020 Recidivism Incarceration Report, supra note 8, at 4.

86

Id. at 4, 53 n.10.

87

Id. at 4.

88
See 2016 Recidivism Overview Report, supra note 3, at 18–19, 23. See also 2017 Recidivism Criminal
History Report, supra note 8; 2017 Recidivism Age Report, supra note 8.
89
Instances of arrest or sentencing that appeared to be duplicates of existing events were removed by
IOP. Minor traffic offenses (e.g., speeding) and arrest entries occurring outside of the eight-year follow-up
period were removed and, therefore, not used to ascertain recidivism.
90

58

See 2016 Recidivism Overview Report, supra note 3, Appendix B.

Recidivism of Federal Offenders Released in 2010

59

United States Sentencing Commission
www.ussc.gov

THURGOOD MARSHALL FEDERAL JUDICIARY BUILDING
ONE COLUMBUS CIRCLE N.E.
SUITE 2-500, SOUTH LOBBY
WASHINGTON, DC 20002-8002