Office of the Inspector General
U.S. Department of Justice
Office
of the Inspector General

U.S. Department of Justice

Review of the Department’s 

Implementation of Prosecution and 

Sentencing Reform Principles under 

the Smart on Crime Initiative 


Evaluation and Inspections Division 17-04

June 2017

EXECUTIVE SUMMARY 

Introduction
In August 2013, the U.S. Department of Justice (Department) and then
Attorney General Eric H. Holder, Jr., announced the Smart on Crime initiative,
which highlighted five principles to reform the federal criminal justice system.
Smart on Crime encouraged federal prosecutors to focus on the most serious cases
that implicate clear, substantial federal interests. In the first principle, the
Department required, for the first time, the development of district-specific
prosecution guidelines for determining when federal prosecutions should be
brought, with the intent of focusing resources on fewer but the most significant
cases. The second principle of Smart on Crime announced a change in Department
charging policies so that certain defendants who prosecutors determined had
committed low-level, non-violent drug offenses, and who had no ties to large-scale
organizations, gangs, or cartels, generally would not be charged with offenses that
imposed a mandatory minimum prison sentence.
The Office of the Inspector General (OIG) initiated this review to evaluate the
Department’s implementation of the first two principles of Smart on Crime, as well
as the impact of those changes to federal charging policies and practices. We
assessed the 94 U.S. Attorney’s Office districts’ implementation and the impact of
the Smart on Crime policy on not charging drug quantities implicating mandatory
minimum sentences in circumstances where the defendants were low-level,
non-violent offenders with limited criminal histories. We also assessed the
implementation and impact of the policy that required prosecutors to consider
certain factors before filing a recidivist enhancement that would increase the
sentence of a drug defendant with a felony record pursuant to 21 U.S.C. § 851.
On May 10, 2017, the Attorney General issued a new charging and
sentencing policy to all federal prosecutors that effectively rescinds the specific
charging policies and practices outlined by Smart on Crime.1 We did not review this
new policy as part of this review, which examined the implementation of the
prosecution and sentencing reform principles under the Smart on Crime initiative.
Nevertheless, we believe that the lessons learned from the Department’s
implementation of the Smart on Crime initiative, and the challenges faced in

The Department’s new charging policy states that it is a “core principle” that all federal
prosecutors should charge and pursue “the most serious readily provable offense… By definition, the
most serious offenses are those that carry the most substantial guidelines sentence, including
mandatory minimum sentences.” Prosecutors may make exceptions from this policy, based upon
careful consideration, with supervisory approval, for reasons that must be documented in the file.
Additionally, the new policy provides that prosecutors in most cases should recommend sentences
within the advisory guidelines range, again absent supervisory approval for reasons documented in
the file. The memorandum indicated that the Deputy Attorney General had been directed to oversee
implementation of the new policy and to issue any clarification and guidance he deemed appropriate
“for its just and consistent application.” See Memorandum to all Federal Prosecutors, from Attorney
General, Jefferson B. Sessions III, Department Charging and Sentencing Policy, May 10, 2017.
1

i

assessing its impact, can be of assistance to the Department when seeking to
implement its new and any future charging policies and practices.
Results in Brief
We found that the Department made progress implementing the first two
Smart on Crime principles, but we also identified several shortcomings in its efforts,
including some failures to update national and local policies and guidelines and a
lack of communication with local law enforcement partners regarding changes to
these polices and guidelines in some instances.
We found that, while the Department issued policy memoranda and guidance
to reflect its Smart on Crime policies, the U.S. Attorneys’ Manual (USAM), a primary
guidance document for federal prosecutors, was not revised until January 2017,
more than 3 years after Smart on Crime was launched, even though Department
officials established a deadline of the end of 2014 to do so. Further, we determined
that 74 of 94 districts had developed or updated their local policies to reflect the
Smart on Crime policy changes regarding mandatory minimum charging decisions.
Of the remaining 20 districts, some provided incomplete information to the OIG as
to whether they had updated their prosecution guidelines or policy memoranda to
reflect the Smart on Crime policy changes regarding mandatory minimum charging
decisions in drug cases; in others, the district policies provided appeared to be
inconsistent with the Smart on Crime policies in whole or in part; and some told us
that they relied on the Holder memoranda for direction but did not develop or
update any of their district policies or guidance documents to reflect the Smart on
Crime policy changes.
We also found that 70 of 94 districts had incorporated Smart on Crime
recidivist enhancement policy changes into their districts’ prosecution guidelines or
policy memoranda. However, of the remaining 24 districts, 20 provided information
to the OIG with respect to recidivist enhancements that appeared to be inconsistent
with the 2013 Holder memoranda in whole or in part, or reported to the OIG that
they followed the Holder memorandum but did not specifically revise their district
policies to reflect Smart on Crime policy changes. The four remaining districts
provided information that did not reflect the Smart on Crime policy changes on
filing recidivist enhancements. Finally, we found that 10 districts failed to update
their policies to reflect Smart on Crime policy changes with regard to both
mandatory minimum charging decisions and recidivist enhancements.
While most districts updated their prosecution guidelines, it is not clear
whether all districts worked in conjunction with their law enforcement partners to
develop or update existing district-specific prosecution guidelines for determining
when federal prosecutions should have been brought and in what priority areas,
consistent with the first principle of Smart on Crime. In our interviews, 3 of 14 U.S.
Attorneys did not recall having these discussions. In addition, 19 of 25 Drug
Enforcement Administration and Federal Bureau of Investigation Special Agents in
Charge either could not recall having these discussions or were not aware of
whether they had occurred. In response to our survey of U.S. Attorney’s Offices,
Criminal Chiefs from nine districts responded that their district did not work with
ii

law enforcement partners to develop or revise district guidelines and Criminal
Chiefs from four districts responded that they were not sure whether this had
occurred.
We further found that the Department’s ability to measure the impact of the
first two Smart on Crime principles is limited because it does not consistently collect
data on charging decisions. For example, while the Legal Information Office
Network System (LIONS), the U.S. Attorneys’ Offices’ case management system,
allows federal prosecutors generally to track information about their cases, data
fields relevant to Smart on Crime were not always present or updated.
Due to these limitations, the Department has relied on U.S. Sentencing
Commission (USSC) data to assess the impact of the first two Smart on Crime
principles. However, using USSC data to measure the impact of Smart on Crime’s
charging policies is challenging because the USSC collects data from courts on
sentencing decisions by judges and does not receive data from prosecutors about
their charging decisions. In that regard, the USSC data does not allow assessments
regarding charges that prosecutors could have brought but chose not to bring.
Nevertheless, based on our own analysis of USSC sentencing data over the
period from 2010 through 2015, we found that sentencing outcomes in drug cases
had shifted in a manner that was consistent with the first two principles of Smart on
Crime. This was reflected by significantly fewer mandatory minimum sentences
being imposed in drug cases nationwide, as well as a decrease in mandatory
minimum sentences for those defendants who might otherwise have received such
a sentence in the absence of the 2013 Holder memoranda. For example, the rate
of federal drug offenders sentenced without a mandatory minimum rose from
40 percent in 2012 to 54 percent in 2015.
In addition, we found that the percentage of federal drug offenders with two
criminal history points who were not subject to a mandatory minimum sentence
increased from 44 percent in FY 2012 to 64 percent in FY 2015. Generally, these
offenders were eligible for relief under the Smart on Crime policy but were not
eligible for “safety valve” relief (a provision that allows judges to sentence
offenders below a mandatory minimum) because they had more than one criminal
history point.
With regard to the use of recidivist enhancements, USSC estimates and OIG
survey results indicated that such enhancements have become less common since
Smart on Crime. For example, USSC data sampling reflected that prosecutors filed
recidivist enhancements in about 20.6 percent of eligible cases in FY 2012, but did
so in only about 17.6 percent of eligible cases in FY 2014. Moreover, this decline
occurred despite a rise in the percentage of defendants eligible for such an
enhancement.
We also found that some regions in the country diverged from these overall
national trends. For example, while drug convictions decreased nationally by
19 percent, the decrease was far larger in the Southwest Border region. Further,
the West, Pacific Northwest, and Hawaii and Island Territories regions actually

iii

showed increases in the number of drug convictions. As a result, we determined
that national trends should not be interpreted in such a way as to conclude that
Smart on Crime had a uniform impact across all the nation’s districts.
In order for the Department to be able to more accurately assess the
implementation and impact of its charging policies and practices, whether related to
Smart on Crime or otherwise, we believe the Department needs to collect relevant
and timely charging data. Similarly, it should ensure that the USAM accurately
reflects Department charging policies and that all districts have local prosecution
guidelines in place that reflect those policies.
Recommendations
In this report, we make three recommendations to ensure that all federal
prosecutors have clear and consistent guidance and fully understand all Department
charging policies, and to enable the Department to more accurately measure the
effectiveness of its charging policy decisions.

iv

TABLE OF CONTENTS 

INTRODUCTION ............................................................................................ 1 

Background ........................................................................................ 1




The USAM and the Smart on Crime Charging Principles ............................. 4 

The Department’s Efforts to Assess the Impact of the Smart on

Crime Charging Policies ........................................................................ 6 

Previous Work Examining Aspects of Smart on Crime......................................... 7 

Scope and Methodology of the OIG Review....................................................... 8 

RESULTS OF THE REVIEW .............................................................................. 9 

While the Department Made Progress Implementing Its Smart on

Crime Policies, We Found Shortcomings, Including Failures to 

Update National and Local Policies and a Lack of Communication 

with Local Law Enforcement Partners, which Could Have Limited 

the Potential Effectiveness of the Initiative .............................................. 9 

The Department’s Ability to Measure the Impact of Smart on Crime

or Other Charging Policies Has Been Limited because Data on

Charging Decisions Has Not Been Consistently Collected.......................... 18 

USSC Data Reflects that There Have Been Shifts in Drug Sentencing

Consistent with Smart on Crime’s First Two Principles, but Some

Regions and Districts Have Diverged from National Trends ...................... 20 

CONCLUSION AND RECOMMENDATIONS ........................................................ 34 

Conclusion ........................................................................................ 34




Recommendations ............................................................................. 35




APPENDIX 1: METHODOLOGY OF THE OIG REVIEW........................................ 36 

Standards ......................................................................................... 36 

Data Analysis .................................................................................... 36




Survey ........................................................................................... 38




Document Analysis ............................................................................ 38




Interviews ........................................................................................ 39




APPENDIX 2: SUMMARY OF SURVEY RESULTS ............................................... 40 


v

APPENDIX 3:	 HOLDER MEMORANDUM, DEPARTMENT POLICY ON

CHARGING AND SENTENCING, MAY 19, 2010 ............................ 47 

APPENDIX 4:	 HOLDER MEMORANDUM, FEDERAL PROSECUTION 

PRIORITIES, AUGUST 12, 2013 ................................................ 50 

APPENDIX 5:	 HOLDER MEMORANDUM, DEPARTMENT POLICY ON

CHARGING MANDATORY MINIMUM SENTENCES AND

RECIDIVIST ENHANCEMENTS, AUGUST 12, 2013 ........................ 53 

APPENDIX 6:	 HOLDER MEMORANDUM, GUIDANCE REGARDING § 851 

ENHANCEMENTS IN PLEA NEGOTIATIONS, 

SEPTEMBER 24, 2014 ............................................................. 56 

APPENDIX 7:	 U.S. SENTENCING COMMISSION FEDERAL SENTENCING 

GUIDELINES MANUAL, § 5C1.2., “SAFETY VALVE” ...................... 57 

APPENDIX 8: 	 THE DEPARTMENT'S RESPONSE TO THE DRAFT REPORT .............. 59 

APPENDIX 9: 	 OIG ANALYSIS OF THE DEPARTMENT'S RESPONSE ..................... 61 


vi

INTRODUCTION

Background
On August 12, 2013, then Attorney General Eric H. Holder, Jr., gave a speech
to the American Bar Association’s Annual Convention that outlined the U.S.
Department of Justice’s (Department) “Smart on Crime” initiative.2 Smart on Crime
resulted from a review of the criminal justice system to “identify reforms that would
ensure federal laws are enforced more fairly and — in an era of reduced budgets —
more efficiently.”
The five announced Smart on Crime principles were: (1) prioritize
prosecutions to focus on the most serious cases; (2) reform sentencing to eliminate
unfair disparities and reduce overburdened prisons; (3) pursue alternatives to
incarceration for low-level, non-violent crimes; (4) improve reentry to curb repeat
offenses and re-victimization; and (5) “surge” resources to prevent violence and
protect the most vulnerable populations.
The cost of maintaining the federal prison system imposes a heavy burden on
taxpayers and, as the Department has acknowledged, results in less funding for the
Department’s other critical law enforcement and national security missions. For
fiscal year (FY) 2016, the Federal Bureau of Prisons’ (BOP) budget was $7.5 billion
and accounted for 26 percent of the Department’s discretionary budget. Moreover,
at the end of FY 2016, the BOP operated at 16 percent over capacity.3 As the
Department has previously stated, overcrowding presents critical safety challenges
for both BOP staff and inmates and has a negative impact on the ability of the BOP
to promptly provide inmate treatment and training programs that promote effective
reentry and reduce recidivism.
In implementing Smart on Crime, the Department stated that it was
designed, in part, to address these budget realities and to chart a course that
controls prison overcrowding and spending while ensuring sufficient resources for
policing and prosecution, effective prisoner reentry, prevention and intervention
programs, and adequate drug treatment.
The first Smart on Crime principle changed Department policies to focus
more directly on what were identified as the most serious cases that implicated
clear, substantial federal interests which, according to the Department, included
protecting Americans from national security threats, violent crime, and financial
fraud, as well as protecting the most vulnerable members of society. Also, for what
the Department described as the first time, district-specific prosecution guidelines

U.S. Department of Justice (DOJ), Smart on Crime: Reforming the Criminal Justice System
for the 21st Century (August 2013).
2

Although that rate is down from the 30 percent over capacity rate as of the end of FY 2014,
the Department projects that BOP institutions will remain overcrowded through FY 2017 and beyond.
3

1


were required to determine when federal prosecutions should be brought, based on
priorities that “will often depend on local criminal threats and needs.”4
The second Smart on Crime principle sought to “reform sentencing to
eliminate unfair disparities and reduce overburdened prisons” by revising Department
charging policies so that certain defendants who have committed low-level,
non-violent drug offenses and who have no ties to large-scale organizations, gangs,
or cartels, would no longer be charged with offenses that trigger the imposition of
mandatory minimum sentences.5 According to a memorandum issued by Attorney
General Holder on the same day he announced the Smart on Crime initiative,
prosecutors were instructed to decline to charge the quantity of drugs necessary to
trigger a mandatory minimum sentence under Title 21 of the U.S. Code for
defendants who met each of the following criteria (the “Holder factors”):
	 The defendant’s relevant conduct does not involve the use of violence, the
credible threat of violence, the possession of a weapon, the trafficking of
drugs to or with minors, or the death or serious bodily injury of any person;
	 The defendant is not an organizer, leader, manager or supervisor of others
within a criminal organization;
	 The defendant does not have significant ties to large-scale drug trafficking
organizations, gangs, or cartels; and
	 The defendant does not have a significant criminal history. A significant
criminal history will normally be evidenced by three or more criminal history
points but may involve fewer or greater depending on the nature of any prior
convictions.
The Holder memorandum also provided that prosecutors should decline to file an
information pursuant to 21 U.S.C. § 851 that would increase a defendant’s
sentence based on prior felony drug convictions unless the defendant “[was]
involved in conduct that makes the case appropriate for severe sanctions,” and it
provided a similar but not identical list of factors that prosecutors should have
considered in making this decision.6

4

2013).

DOJ, Smart on Crime: Reforming the Criminal Justice System for the 21st Century (August

The second principle of Smart on Crime also stated that the Attorney General planned to
work with Congress to pass legislation that would reform mandatory minimum laws. We did not
review this aspect of Smart on Crime as part of this review.
5

We use “Holder memorandum” to refer to any of the following, by year: Eric Holder, Jr.,
Attorney General, memorandum to All Federal Prosecutors, Department Policy on Charging and
Sentencing, May 19, 2010; memorandum to Heads of Department of Justice Components and United
States Attorneys, Federal Prosecution Priorities, August 12, 2013; memorandum to United States
Attorneys and Assistant Attorney General for the Criminal Division, Department Policy on Charging
Mandatory Minimum Sentences and Recidivist Enhancements in Certain Drug Cases, August 12, 2013;
(Cont’d)
6

2


In February 2015, Attorney General Holder stated that “preliminary” data
showed that Smart on Crime worked “exactly as intended” and had “a real and
measurable impact on the decisions made by federal prosecutors from coast to
coast.”7 Attorney General Holder stated that data showed federal prosecutors were
more selective in bringing certain drug prosecutions; the most serious drug crimes
were attracting the highest scrutiny; and the Department experienced a “historic”
decline in the number of mandatory minimum sentences sought by federal
prosecutors.
In March 2016, the Department reported that U.S. Sentencing Commission
(USSC) data for FY 2015 showed that charging decisions made by federal
prosecutors were focused on more serious drug cases and federal prosecutors
brought fewer indictments carrying a mandatory minimum.8 Also, according to BOP
data, as of September 2016, the federal prison population had decreased by more
than 12 percent, or over 26,000 inmates, from its peak in 2013.
Although the Office of the Inspector General (OIG) has conducted reviews
related to other aspects of Smart on Crime, including our recent review on the use
of pretrial diversion and diversion-based court programs discussed below, the OIG
had not examined whether Smart on Crime’s directives had the intended impact on
federal charging practices. For this review, we focused on the first two principles of
Smart on Crime. Specifically, we assessed the 94 U.S. Attorney’s Office (USAO)
districts’ implementation of the initiative’s first principle that required the
establishment of district-specific policies for prioritizing prosecutions and the second
principle that gave prosecutors discretion to not charge drug quantities implicating
mandatory minimum sentences in circumstances where the defendants were lowlevel, non-violent offenders with limited criminal histories, as well as discretion
regarding whether to file recidivist enhancements pursuant to 21 U.S.C. § 851 that
increases sentences of defendants with prior felony drug convictions.9
As the OIG was finalizing this report, we were informed that the Department
was preparing to release a new charging and sentencing policy. This policy was
and memorandum to Department of Justice Attorneys, Guidance Regarding § 851 Enhancements in
Plea Negotiations, September, 24, 2014. See Appendices 3–6, respectively, for the memoranda.
7 DOJ Press Release, “Attorney General Holder Delivers Remarks at the National Press Club,”
www.justice.gov/opa/pr/attorney-general-holder-delivers-remarks-national-press-club, February 17,
2015 (accessed June 14, 2017).

DOJ Press Release, “New Smart on Crime Data Reveals Federal Prosecutors Are Focused on
More Significant Drug Cases and Fewer Mandatory Minimums for Drug Defendants,”
https://www.justice.gov/opa/pr/new-smart-crime-data-reveals-federal-prosecutors-are-focused-moresignificant-drug-cases-and, March 21, 2016 (accessed June 14, 2017).
8

9 2013 Holder memorandum on Charging Mandatory Minimum Sentences and Recidivist
Enhancements (see Appendix 5).

21 U.S.C. § 851 requires prosecutors to file one or more notices of prior felony drug
convictions in order to trigger enhanced sentences for drug offenses, for example, from a mandatory
sentence depending on drug type and quantity of 5 to 10 years, from 10 to 20 years, or from 20 years
to life.

3


released on May 10, 2017, and effectively rescinded the charging and sentencing
policies established in the Smart on Crime initiative.10 We did not review the
Department’s new policy as part of this review. Nevertheless, we believe that the
lessons learned from the Department’s implementation of the Smart on Crime
initiative, and the challenges faced in assessing its impact, can be of assistance to
the Department when seeking to implement its new and any future charging
policies and practices.
In this section, we describe the guidance in the U.S. Attorneys’ Manual
(USAM), which federal prosecutors look to in the discharge of their duties, and
Attorney General Holder’s August 12, 2013, memorandum to federal prosecutors
regarding charging certain drug cases (the 2013 Holder memorandum). We also
discuss the Department’s efforts to assess the impact of Smart on Crime,
particularly focusing on how these reforms have affected the sentences of
defendants charged with certain drug offenses since the reforms were launched.
Finally, we discuss our previous work related to the Smart on Crime reforms.
The USAM and the Smart on Crime Charging Principles
For nearly 3 decades, the USAM’s “Principles of Federal Prosecution” have
guided federal prosecutors in the discharge of their duties and helped to ensure
federal cases are prosecuted according to consistent standards.11 Over the years,
the USAM and its Principles of Federal Prosecution have been updated and refined
to reflect changes in the law and Department policies.12
Originally promulgated by Attorney General Benjamin R. Civiletti on July 28,
1980, the Principles of Federal Prosecution direct that federal prosecutors “should
charge, or should recommend that the grand jury charge, the most serious offense
that is consistent with the nature of the defendant’s conduct, and that is likely to
result in a sustainable conviction.”13 In a 1993 memorandum, Attorney General
Janet Reno adopted this language but supplemented it with a requirement that
federal prosecutors’ charging decisions should be based on “an individualized
assessment of the extent to which particular charges fit the specific circumstances
of the case, as consistent with the criminal code, and maximize the impact of
10 The Department’s new charging policy states that it is a “core principle” that all federal
prosecutors should charge and pursue “the most serious readily provable offense… By definition, the
most serious offenses are those that carry the most substantial guidelines sentence, including
mandatory minimum sentences.” Prosecutors may make exceptions from this policy, based upon
careful consideration, with supervisory approval, for reasons that must be documented in the file.
Additionally, the new policy provides that prosecutors in most cases should recommend sentences
within the advisory guidelines range, again absent supervisory approval for reasons documented in
the file. The memorandum indicated that the Deputy Attorney General had been directed to oversee
implementation of the new policy and to issue any clarification and guidance he deemed appropriate
“for its just and consistent application.” See Jefferson B. Sessions III, Attorney General,
memorandum to All Federal Prosecutors, Department Charging and Sentencing Policy, May 10, 2017.
11

USAM, 9-27.001, et seq.

12

USAM, 9-27.300.

13

USAM, 9-27.001 and 9-27.300.

4


federal resources on crime.”14 In a September 2003 memorandum, Attorney
General John Ashcroft eliminated the requirement of an “individualized assessment”
and instructed federal prosecutors that they “must charge and pursue the most
serious, readily provable offense or offenses that are supported by the facts of the
case.”15
On May 19, 2010, Attorney General Holder issued a memorandum to all
federal prosecutors regarding the Department’s policy on charging and sentencing
decisions. The 2010 Holder memorandum expressly superseded the 2003 Ashcroft
memorandum, scaled back the mandatory language, and stated that a federal
prosecutor “should ordinarily charge” the most serious offense that is consistent
with the nature of the defendant’s conduct and that is likely to result in a
sustainable conviction. Further, the 2010 Holder memorandum states that this
determination must always be made in the context of “an individualized assessment
of the extent to which particular charges fit the specific circumstances of the case,
are consistent with the purpose of the Federal Criminal Code, and maximize the
impact of federal resources on crime.”16
As referenced above, in August 2013, Attorney General Holder announced
reforms associated with Smart on Crime and concurrently released two memoranda
to U.S. Attorneys (USA) and Department components to outline specific policy
changes.17 The first memorandum addressed the initiative’s first principle
regarding prioritizing prosecutions, and required “all USAs, in conjunction with their
law enforcement partners, to develop — or update existing — district specific
guidelines for determining when federal prosecutions should be brought and in what
priority areas.”18 The second memorandum addressed reforms to charging and
sentencing practices, and revised the Department’s charging policies regarding drug
quantities that trigger mandatory minimum sentences for certain non-violent, lowlevel drug offenders and provided guidance to federal prosecutors limiting the filing
of recidivist enhancements in some drug cases pursuant to 21 U.S.C. § 851.

14 Janet Reno, Attorney General, memorandum to Holders of U.S. Attorneys’ Manual, Title 9,
Reno Bluesheet on Charging and Plea Decisions, October 12, 1993.
15 John Ashcroft, Attorney General, memorandum to All Federal Prosecutors, Memo regarding
Policy on Charging of Criminal Defendants, September 22, 2003, 2.
16

2010 Holder memorandum, 3 (see Appendix 3).

2013 Holder memorandum on Federal Prosecution Priorities and 2013 Holder memorandum
on Mandatory Minimum Sentences and Recidivist Enhancements (see Appendices 4 and 5,
respectively).
17

18 District-specific prosecution guidelines determine when federal prosecutions should be
brought and in what priority areas by defining what cases serve a district’s substantial federal
interests. District guidelines are informed by a number of factors, including both national and local
law enforcement priorities.

5


The Department’s Efforts to Assess the Impact of the Smart on Crime
Charging Policies
In order to assess the impact of the Smart on Crime policies, the Department
developed a tracking document in 2013 that included the Smart on Crime goals; a
process to measure the initiative’s effects; a timeline for implementation of the
initiative; and a series of milestones (completed, in-process, and future). As part
of its plan for measuring changes in how defendants were charged, the Department
selected metrics to track over the 2 years following the initiative’s implementation,
including the number of defendants sentenced to mandatory minimums under
21 U.S.C.§ 841.
The tracking document also measured the use of the “safety valve” and the
frequency with which low-level, non-violent drug offenders received mandatory
minimum sentences.19 Further, the Department determined the number of
sentences enhanced pursuant to 21 U.S.C. § 851. This included an analysis of
conviction data on a quarterly basis from FY 2010 to the present, with the end of
FY 2013 serving as a baseline measure. In addition to generally requiring USAOs to
track data on charging decisions, the Department’s tracking document called for
USAOs to update their district-specific prosecution guidelines by September 2014,
and for the Department to update the USAM to be consistent with the Smart on
Crime policies by the end of 2014.
Since the Department has not historically collected charging or conviction
data, it relied on data collected by the USSC in order to measure the impact of
Smart on Crime.20 The USSC is an independent agency located in the judicial
branch of government, which was created by the Sentencing Reform Act of 1984 to,
among other things, reduce sentencing disparities and promote transparency and
proportionality in sentencing through its promulgation of federal sentencing
guidelines. The USSC collects, analyzes, and distributes a broad array of
information on federal sentencing practices, including data on the number of
defendants who were sentenced to 5- or 10-year mandatory minimum sentences,
and 21 U.S.C. § 851 recidivist enhancements according to drug type, but it does
not track the charging decisions prosecutors make that underlie those judicial
sentencing decisions.

19 The “safety valve” refers to 18 U.S.C. § 3553(f), which provides that the court shall impose
a sentence without regard to an otherwise applicable statutory mandatory minimum sentence when a
defendant meets certain conditions enumerated in the statute and truthfully provides to the
government all relevant information and evidence the defendant may have. The other conditions
establishing eligibility to be sentenced under the safety valve are similar to, but somewhat more
limited than, the Holder factors. Separate from the safety valve, the court also may impose a
sentence below an otherwise applicable mandatory minimum sentence based upon a government
motion establishing that the defendant has provided “substantial assistance in the investigation or
prosecution of another person.” U.S.S.G. 5K1.1 and 18 U.S.C. 3553(e).
20

DOJ, Smart on Crime Briefing Metrics “tracking” document, September 2013, 4–5.

6


Previous Work Examining Aspects of Smart on Crime
OIG, Audit of the Department’s Use of Pretrial Diversion and Diversion-Based Court
Programs as Alternatives to Incarceration, Audit Report 16-19 (July 2016)
Smart on Crime reforms outlined a range of options for federal prosecutors to
consider to ensure just punishments for low-level, non-violent offenders, including
the increased use of alternatives to incarceration such as pretrial diversion and
diversion-based court programs. In the OIG’s July 2016 report, we found that the
availability and use of these programs varied substantially across federal judicial
districts, that the Department had not evaluated the effectiveness of the USAOs’
use of pretrial diversion or their participation in diversion-based court programs,
and that the USAOs did not maintain sufficient reliable data to enable the OIG to
comprehensively evaluate the effectiveness of these programs.21 We also found
that the pretrial diversion information that the Executive Office for United States
Attorneys (EOUSA) captured may have been underreported or inconsistently
reported, and we determined not only that the number of successful pretrial
diversion program participants varied greatly among the USAOs, but also that the
USAOs’ participation in diversion-based court programs was limited.
OIG, Review of the Impact of an Aging Inmate Population on the Federal Bureau of
Prisons, Evaluation and Inspections Report 15-05 (May 2015)
Pursuant to the second principle of Smart on Crime, the BOP expanded its
compassionate release criteria for aging inmates. However, the Department
significantly limited the number of inmates eligible for this expanded compassionate
release policy by imposing several requirements, including that inmates be at least
age 65 to be eligible and have served at least 10 years of their sentence. The OIG
found in its May 2015 review that only two inmates had been released under this
new provision in the year since it was adopted and that, according to institution
staff, it is difficult for aging inmates to meet all of the eligibility requirements of the
BOP’s new provisions. The OIG’s analysis showed that if the BOP reexamined these
eligibility requirements in a manner consistent with Smart on Crime goals, its
compassionate release program could result in significant cost savings for the BOP,
as well as assist in managing the inmate population.22

In 2016, the U.S. Government Accountability Office (GAO) also found that the Department
does not reliably track the use of some alternatives to incarceration, including the use of pretrial
diversion. The GAO concluded that, by revising the Department’s system to track the different types
of pretrial diversion programs and issuing guidance as to when staff are to enter their use into its
database, the Department would have more reliable and complete data. See GAO, Federal Prison
System: Justice Has Used Alternatives to Incarceration, But Could Better Measure Program
Outcomes, GAO-16-516 (June 2016).
21

In 2013, the OIG found that the BOP’s compassionate release program had been poorly
managed and implemented inconsistently, likely resulting in eligible inmates not being considered for
release and in terminally ill inmates dying before their requests were decided. See DOJ OIG, The
Federal Bureau of Prisons’ Compassionate Release Program, Evaluation and Inspections
Report I-2013-006 (April 2013).
22

7


U.S. Government Accountability Office, Federal Prison System: Justice Could Better
Measure Progress Addressing Incarceration Challenges, GAO-15-454 (June 2015)
The U.S. Government Accountability Office (GAO) reviewed the Department’s
16 Smart on Crime indicators and found that, while they were well-linked to the
effort’s overall goals, in many cases the indicators lacked other key elements of
successful performance measurement systems, such as clarity and context.23 For
example, the GAO found that 7 of the 16 indicators were confusing or did not
represent the information the indicator name implied and that 13 of the
16 indicators lacked the needed contextual information to appropriately interpret
their results. The GAO recognized that measuring performance can be a challenge,
especially for prosecutorial agencies such as the Department, but noted that
improved data collection and clearly defined goals and progress measures can
assist agencies in developing effective performance measurement systems. The
GAO noted that the Department would have been better positioned to more
effectively measure its Smart on Crime efforts by relying on such options.
Scope and Methodology of the OIG Review
This review examined how the Department established policies following its
Smart on Crime announcement to prioritize prosecutions to focus on the most
serious cases and whether charging practices changed in a manner consistent with
Smart on Crime principles and implementing memoranda. Our fieldwork occurred
from January 2016 through July 2016 and consisted of document reviews, data
analysis, and interviews. We analyzed USSC data from FY 2010 through FY 2015
and how the Department collects and analyzes data related to Smart on Crime. We
also analyzed USAO staffing data from the same period that might be relevant to
overall case numbers and charging practices.
We prepared and conducted a comprehensive survey of Criminal Chiefs,
supervisory attorneys, and line attorneys in all 94 USAO districts, addressing their
district policies, practices, and their views regarding Smart on Crime.24
Additionally, we requested and reviewed information and documentation from each
district on its local prosecution guidelines and policies regarding charging
mandatory minimums and recidivist enhancements. We also interviewed staff from
the Office of the Deputy Attorney General, the Office of Policy and Legislation, the
Office of Legal Policy, EOUSA, and the USSC. Additionally, we interviewed USAs
and Criminal Chiefs in selected districts, as well as Special Agents in Charge at
Federal Bureau of Investigation and Drug Enforcement Administration field offices.
See Appendix 1 for more information about the OIG’s methodology.

23

See page 18 for further discussion regarding the 16 Smart on Crime indicators.

24 Our survey contained 32 multiple-choice and open-ended questions regarding how the
districts’ prosecution priorities and charging policies have been affected by Smart on Crime and the
issuance of the 2013 Holder memoranda. See Appendix 1 for more information and Appendices 4 and
5 for the applicable memoranda.

8


RESULTS OF THE REVIEW 

While the Department Made Progress Implementing Its Smart on Crime
Policies, We Found Shortcomings, Including Failures to Update National
and Local Policies and a Lack of Communication with Local Law
Enforcement Partners, which Could Have Limited the Potential
Effectiveness of the Initiative
We found that overall the U.S. Attorney’s Office (USAO) districts took steps to
implement the Smart on Crime policy changes, such as generally updating their local
policies and charging practices. However, the Department failed to update the U.S.
Attorneys’ Manual (USAM) to reflect the Smart on Crime policy changes, which could
have resulted in federal prosecutors receiving inconsistent guidance when charging
certain drug cases. Although most districts developed or updated their policies to
reflect the Smart on Crime reforms related to charging mandatory minimums and/or
recidivist enhancements, some districts did not develop or update their policies as
directed, while others developed policies that are in whole or in part inconsistent with
Smart on Crime, particularly regarding recidivist enhancements. Finally, it is not
clear that the USAOs in some districts consulted their law enforcement partners when
updating their local prosecution guidelines as required by Smart on Crime.
The Department Failed to Update the USAM to Reflect Smart on Crime Policy
Changes, which Could Have Resulted in Federal Prosecutors Receiving Inconsistent
Guidance When Charging Certain Drug Cases
The USAM’s Principles of Federal Prosecution are, among other things,
designed to “assist in structuring the decision-making process of attorneys for the
government,” “facilitate the task of training new attorneys in the proper discharge of
their duties,” and “contribute to more effective management of the government’s
limited prosecutorial resources by promoting greater consistency among the
prosecutorial activities of all United States Attorney’s offices.”25 As discussed
elsewhere in this report, the Department developed a tracking document with
milestones for the completion of tasks related to the implementation of Smart on
Crime. One of these milestones required the completion of USAM revisions by the
end of 2014 to ensure federal prosecutors throughout the country had consistent
guidance when charging drug cases. We found that the Department did not update
the USAM until January 2017, more than 3 years after Smart on Crime was launched.
Prior to January 2017, the USAM contained language directing prosecutors to
charge “the most serious offense that is consistent with the nature of the
defendant’s conduct, and that is likely to result in a sustainable conviction,” and did
not discuss any of the specific directives contained in Smart on Crime policies.26
25

USAM, 9-27.001.

26 We also found that the USAM did not include or reference additional guidance to federal
prosecutors for cases that were charged before the issuance of the 2013 Holder memoranda (see
(Cont’d)

9


While USABook, the internal informational portal for USAOs maintained by the
Executive Office for United States Attorneys (EOUSA), did have a page on the
Smart on Crime initiative with links to the Smart on Crime policy and supporting
memoranda, we believe that the failure of the Department to update the USAM
could have been confusing for federal prosecutors given its importance in guiding
prosecutorial discretion and because the Smart on Crime policies required them to
consider several factors (the “Holder factors”) in certain drug cases when charging
low-level, non-violent drug offenders with limited criminal histories.
Also prior to January 2017, we found that the USAM did not include the
Holder factors that federal prosecutors must consider in cases involving the
applicability of Title 21 mandatory minimum sentences based on drug type and
quantity, or the similar but somewhat distinct factors they were to consider when
deciding whether to file recidivist enhancements.27 In fact, as to the latter, we
found that the USAM appeared to have retained a presumption in favor of the filing
of 21 U.S.C. § 851 recidivist enhancements that was inconsistent with the Smart on
Crime initiative and the Holder memorandum. Specifically, prior to January 2017,
Section 9-27.200(B) provided in pertinent part as follows:
Every prosecutor should regard the filing of an information under
21 U.S.C. § 851 concerning prior convictions as equivalent to the filing
of charges. Just as a prosecutor must file a readily provable charge,
he or she must file an information under 21 U.S.C. § 851 regarding
prior convictions that are readily provable and that are known to the
prosecutor prior to the beginning of trial or entry of plea. The only
exceptions to this requirement are where: (1) the failure to file or the
dismissal of such pleadings would not affect the applicable guideline
range from which the sentence may be imposed; or (2) in the context
of a negotiated plea, the United States Attorney, the Chief Assistant
United States Attorney, the senior supervisory Criminal Assistant
United States Attorney or within the Department of Justice, a Section
Chief or Office Director has approved the negotiated agreement. The
reasons for such an agreement must be set forth in writing. Such a
reason might include, for example, that the United States Attorney’s
office is particularly overburdened, the case would be time-consuming
Appendices 4 and 5). See also Eric Holder, Jr., Attorney General, memorandum to All United States
Attorneys and the Assistant Attorney General for the Criminal Division, Retroactive Application of
Department Policy on Charging Mandatory Minimum Sentences and Recidivist Enhancements in
Certain Drug Cases, August 29, 2013.
We use “Holder memorandum” to refer to any of the following, by year: Eric Holder, Jr.,
Attorney General, memorandum to All Federal Prosecutors, Department Policy on Charging and
Sentencing, May 19, 2010; memorandum to Heads of Department of Justice Components and United
States Attorneys, Federal Prosecution Priorities, August 12, 2013; memorandum to United States
Attorneys and Assistant Attorney General for the Criminal Division, Department Policy on Charging
Mandatory Minimum Sentences and Recidivist Enhancements in Certain Drug Cases, August 12, 2013;
and memorandum to Department of Justice Attorneys, Guidance regarding § 851 Enhancements in
Plea Negotiations, September, 24, 2014. See Appendices 3–6, respectively, for the memoranda.
27

See Appendix 5 for the applicable 2013 Holder memorandum.

10


to try, and proceeding to trial would significantly reduce the total
number of cases disposed of by the office. The permissible
agreements within this context include: (1) not filing an
enhancement; (2) filing an enhancement which does not allege all
relevant prior convictions, thereby only partially enhancing a
defendant's potential sentence; and (3) dismissing a previously filed
enhancement.
The Chief of Staff to then Deputy Attorney General Sally Q. Yates told us that
federal prosecutors need to follow the USAM, and she acknowledged that the USAM
should have been updated to include the Smart on Crime charging policies and that
it was the Office of the Deputy Attorney General’s responsibility to finalize USAM
revisions. Nonetheless, the Deputy Attorney General’s Chief of Staff said that the
lack of USAM revisions should not have prevented Assistant U.S. Attorneys (AUSA)
from following the Smart on Crime policies because “all AUSAs have a copy of these
policies and they are on-line and the policies make clear that they supersede the
things in the USAM, which is a compilation of what gets rolled out over time. So,
technically, these policies supersede it.”
Nevertheless, since AUSAs are required to follow the USAM, which is widely
viewed as a seminal guidance document for federal prosecutors across the country,
we believe that the Department’s delay in updating the USAM could have created a
risk that defendants could be charged and sentenced in a manner that was
inconsistent with the Smart on Crime policy, with defendants serving potentially
disparate sentences as a result. This risk could have logically increased if a district
failed to update its local prosecution guidelines to reflect the Smart on Crime policy
changes.
Smart on Crime also called for the USAM to be updated to reflect the
requirement that U.S. Attorneys (USA) develop district-specific prosecution
guidelines to assist AUSAs in determining when federal prosecutions should have
been brought because “each U.S. Attorney is in the best position to articulate the
priorities that make sense for that area.”28 However, prior to January 2017, we
found that the USAM was not updated to reflect this Smart on Crime requirement
either. We discuss the efforts the districts made to develop or update their districtspecific prosecution guidelines below.
Most Districts Updated or Revised Their Charging Policies Related to Mandatory
Minimum Sentences in Certain Drug Cases
According to the Holder memorandum, “prosecutors should continue to
ascertain whether a defendant is eligible for any statutory mandatory minimum
statute or enhancement.”29 However, in cases involving the applicability of Title 21
mandatory minimum sentences based on drug type and quantity, prosecutors were
See DOJ, Smart on Crime: Reforming the Criminal Justice System for the 21st Century
(August 2013).
28

29

See Appendix 5 for the applicable 2013 Holder memorandum.

11


instructed to decline to charge the quantity necessary to trigger a mandatory
minimum sentence if the defendant met each of the four specified criteria described
above.30
The OIG requested prosecution guidelines, policy memoranda, checklists, and
worksheets related to charging practices in drug cases from each of the 94 USAOs
to determine whether they incorporated the directives contained in the Holder
memoranda.
We found that 74 of 94 districts’ (79 percent) local charging policies reflected
the policy changes outlined in the Holder memoranda for assessing cases prior to
indictment, or for filing enhancements. In most cases, districts simply added the
Holder memoranda language to their local drug prosecution guidelines, developed a
worksheet or checklist to ensure AUSAs considered the Holder factors before
making charging decisions, or issued a separate memorandum to AUSAs regarding
this policy change.31
We further found that in 8 of 94 districts (9 percent), while the districts
reported to us that they relied on the Holder memoranda for direction, they had not
developed or updated any district policies or other guidance documents to reflect
the Smart on Crime policy changes. In response to our survey, two Criminal Chiefs
from one district said that they noticed fewer drug cases where mandatory
minimums were charged for low-level drug defendants. However, two Supervisors
and an AUSA from this same district said that they continued to charge and
approve charging mandatory minimums for such defendants.
Also, while most survey respondents from one district said that they were
less likely to charge or approve mandatory minimums for low-level drug
defendants, the Criminal Chief and a Supervisor in this district said that there are
more drug cases where such defendants are currently charged with mandatory
minimums. Even though this district may be charging drug cases with more
mandatory minimum sentences because it is focusing on the most serious
defendants, one AUSA from this district said that the initiative “implies that we
prosecute low-level drug dealers as a general practice — that assumption is
insulting.”
In addition, in 4 of the 94 districts (4 percent), the documentation provided
to the OIG appeared to be inconsistent with the Holder memoranda in whole or in
part. Generally, these districts sent an email or memorandum to their AUSAs that
required them to assess their cases in light of the Smart on Crime policy, but the
districts did not update their existing prosecution guidelines. For example, the
30

These criteria are hereinafter referred to as the “Holder factors.”

We did not distinguish among these districts based on the method by which they
communicated the Holder factors. We do believe that, even where districts have adopted worksheets
or checklists, or issued policy memoranda, they also should have revised their district charging policies
so that the relevant and most current policy is consistent with district practice and is available to all
AUSAs and law enforcement agencies with whom they work in a single, readily accessible location.
31

12


current Criminal Division Manual for one of the districts did not reflect the Smart on
Crime policy changes regarding mandatory minimums, yet this district’s prosecution
memorandum template, which was updated in January 2015, required an analysis
of the Holder factors prior to the indictment of a case.32
In addition, one district issued a policy on August 15, 2014, that appeared to
us to effectively abrogate the Holder memoranda policy changes for at least some
defendants because the district included two additional factors to be considered
when determining whether to charge a mandatory minimum sentence. However, in
a follow-up interview and subsequent email, the Criminal Chief for this district
indicated that the language used to describe the additional factors was “in-artfully
drawn,” and that the language served in practice to exclude charging mandatory
minimums and filing 21 U.S.C. § 851 enhancements against minor offenders such
as a drug mule.33
Finally, 8 of 94 districts (9 percent) did not initially provide complete
information to the OIG to indicate that they had updated their local policies to
reflect the Smart on Crime policy changes.34 We found that, generally, these
districts’ policies predated Smart on Crime, made no mention of the Holder
memoranda, or contained guidance related only to their internal policies regarding
recidivist enhancements. In response to our survey, Criminal Chiefs, Supervisors,
and line attorneys for these eight districts said that their districts were less likely
under Smart on Crime to charge or approve charging low-level defendants with
drug quantities that trigger mandatory minimum sentences. Line attorneys
responded that they incorporated an analysis of the Holder factors into their
prosecution memoranda and templates.
However, the prosecution memorandum template that we reviewed from one
of the districts did not include any language to indicate that such an analysis was
required or occurred on a regular basis. In fact, the prosecution guidelines
provided by this district regarding charging, plea agreements, and sentencing was
dated September 2010, predating the 2013 Holder memoranda. Although this
district provided a “mandatory minimum checklist” dated December 29, 2014, that
included the Holder factors, its Criminal Chief indicated in response to our

32 As a result of the OIG’s inquiry regarding this district’s prosecution guidelines, this district
removed language from its prosecution guidelines that still allowed AUSAs to file 21 U.S.C. 851
enhancements and use them as leverage in plea negotiations or as a trial penalty.

Of course, many such defendants might have been eligible for a sentence below an
otherwise applicable mandatory minimum based either upon the safety valve or a government motion
based upon substantial assistance. It is beyond the scope of this review to determine whether, in
practice, the implementation of the Smart on Crime principles significantly changed sentencing
outcomes for such individuals.
33

34 Throughout the OIG’s work on this report, we followed up with those districts that failed to
respond to our requests for information to attempt to obtain the most complete information for our
analysis. See Appendix 1 for more information on our methodology.

13


September 2016 follow-up request that the district had not formally updated its
prosecution guidelines, as required.
Most Districts Updated or Revised Their Charging Policies Related to Recidivist
Enhancements
The Holder memorandum stated that “prosecutors should decline to file an
information pursuant to 21 U.S.C. § 851 unless the defendant is involved in conduct
that makes the case appropriate for severe sanctions.” It further provided that,
“when determining whether an enhancement is appropriate, prosecutors should
have considered a list of factors similar, but not identical to the Holder factors for
deciding whether to charge mandatory minimum-inducing drug quantities.35
In addition, the Department policy as set forth in the September 2014
Holder memorandum prohibited federal prosecutors from using recidivist
enhancements “for the sole or predominant purpose of inducing a defendant
to plead guilty” and encouraged them to make the § 851 filing decision at the
time the case is charged or soon after.36
We found that 70 of 94 districts (74 percent) incorporated the Smart on
Crime recidivist enhancement policy changes into their district prosecution
guidelines or policy memoranda. In most cases, the districts updated their
prosecution guidelines, policy memoranda, templates, or worksheets to reflect the
Smart on Crime policy. However, we found that 14 districts’ prosecution guidelines
regarding recidivist enhancements appeared to be inconsistent with the Holder
memoranda in whole or in part, and 10 others either provided documentation to the
OIG that did not reflect Smart on Crime policy changes with respect to 21 U.S.C.
§ 851 recidivist enhancements or told us that they did not update their policies, but
that they nevertheless relied on the 2013 Holder memoranda.
Specifically, some of the policies we reviewed allowed prosecutors to use
recidivist enhancements as a trial penalty, thereby inducing a defendant to plead
guilty. For example, one district’s guidance stated that “in order to trigger
enhancement provisions concerning multiple drug convictions or dangerous drug
offenders, statutes require the Government to file an ‘Enhancement Information’
prior to the entry of a plea of guilty or the selection of a jury.” While “dangerous
drug offenders” are the type of defendants discussed in the Holder factors that
would be subject to “severe sanctions,” under the Smart on Crime guidance, the
§ 851 filing decision should have occurred at the time the case was charged or soon
after, and not simply prior to the entry of a guilty plea or before jury selection,
which, while compliant with the statute, could have opened the door to potential
misuse of the enhancement. Similarly, another district’s prosecution guidelines
allowed prosecutors to file enhancements “before trial or guilty plea, and also

35

See Appendix 5 for the applicable 2013 Holder memorandum.

36

See Appendix 6 for the 2014 Holder memorandum.

14


before Jencks material was provided to the defense. Timely guilty pleas may
warrant agreement not to file an 851.”37
In addition, we found it more problematic that two districts developed additional
factors for an AUSA to consider when making enhancement decisions that may have
been contrary to the spirit and intent of the Smart on Crime policy regarding recidivist
enhancements. In one of the districts, a memorandum on 21 U.S.C. § 851
enhancements appropriately discussed the Holder factors, but this district’s § 851
approval form contained 15 other factors, two of which required AUSAs to consider a
possible guilty plea when making § 851 recidivist enhancement determinations.
Four of 94 districts (4 percent) provided documentation to the OIG that did not
reflect Smart on Crime policy changes with respect to 21 U.S.C. § 851 recidivist
enhancements. The USAs of two districts told us in interviews that they were being
much more selective when using recidivist enhancements, yet the documentation
provided did not reflect the policy changes. One of the USAs told us that in the past
this district used these enhancements as a bargaining tool; however, the district no
longer uses them to negotiate plea agreements, “since that is now against the rules.”
The other USA told us that the district “never used it as a bargaining tool, but used it
to charge all eligible defendants with the enhancements.”38
The responses to our survey for the remaining two of these four districts
indicated that AUSAs in these districts either reviewed the Holder memoranda at
intake or when drafting charges, referred to the specific Holder factors that are
incorporated in their districts’ prosecution guidelines, or included the basis for filing
a 21 U.S.C. § 851 enhancement in pre-indictment prosecution memoranda.39
Finally, the six remaining districts reported to the OIG that they followed the
Holder memoranda, but they did not provide any supporting documents to show
that the guidance regarding 21 U.S.C. § 851 recidivist enhancements was
communicated in their districts. Responses to our survey from these districts were
split, indicating that they either updated their prosecution priorities or that their

37 “Jencks” or “Jencks material” refers to evidence used during the course of a federal criminal
prosecution in the United States that usually consists of documents relied upon by government witnesses
who testify at trial. The material is described as inculpatory, favoring the government prosecution of a
criminal defendant. “Jencks” is typically provided to the defense in advance of a witness’s testimony at trial.
38 In a follow-up discussion with the Deputy Criminal Chief for this district, we were told that
this district rarely files recidivist enhancements and that criminal AUSAs were trained on the
requirements for filing these enhancements in an October 2014 training session.
39 We reached out to two of the districts for clarification regarding these issues. The Criminal
Chief for one of the districts said AUSAs draft an email to include the basis for filing a 21 U.S.C. § 851
enhancement. Also, the USA and the First Assistant U.S. Attorney reviewed the district’s prosecution
guidelines when the Holder memorandum was issued and believed that it satisfied the Holder
memorandum, but they planned to update their local policies to include specific guidance regarding
recidivist enhancements. The Deputy Criminal Chief for the other district told us in an email that “the
unwritten policy is that the factors in the Narcotics Section Charging Policy [which discusses the
Holder factors] were used in determining whether to file 851 enhancements.”

15


district already followed policies similar to Smart on Crime policy when making §
851 enhancement determinations.
Most Districts Required Supervisory Approval for Drug Cases in Which a Mandatory
Minimum Is Charged or a Recidivist Enhancement Is Filed
After setting forth the factors to be considered in charging both mandatory
minimum triggering drug quantities and recidivist enhancements, the 2013 Holder
memorandum “reminded” prosecutors “that all charging decisions must be reviewed
by a supervisory attorney to ensure adherence to the Principles of Federal
Prosecution” and the policies set forth in the 2010 and 2012 memoranda. We
found that most districts required supervisory approval before a case could be
indicted with a drug quantity that will trigger a mandatory minimum sentence
(84 percent of districts) and before an AUSA could file a 21 U.S.C. § 851 recidivist
enhancement (89 percent of districts). Supervisory approval is generally required
from the AUSA’s direct supervisor, the Branch Chief, Criminal Chief and/or Deputy
Criminal Chief, a Senior Litigation Counsel, or, in some cases, the First Assistant
U.S. Attorney or the USA.
However, in some districts, while supervisory approval was required, the
documents we were provided did not specify this requirement. When we requested
clarification on this issue, district officials told us that all indictments triggering
mandatory minimums or filing recidivist enhancements required supervisory
approval. Some of the districts we spoke with provided additional worksheets,
indictment approval forms, or checklists to illustrate the requirement for
supervisory approval. Other districts told us that they would update their policies in
the future to ensure this requirement is documented in their district policies. As
noted above, we believe that the Department should ensure that supervisory
approval is required, communicated, and documented in district policies to reflect
the current Department policy.
While Most USAOs Updated Their District Guidelines, It is Not Clear Whether All
USAOs Consulted with Their Law Enforcement Partners as Required by the Smart
on Crime Initiative
To ensure that federal law enforcement priorities align with the prosecution
goals of all districts, Smart on Crime policy required that USAs develop or update
existing district-specific guidelines, “in conjunction with their law enforcement
partners.”40 In response to our survey of all 94 districts, 51 of 64 districts’ Criminal
Chiefs (80 percent) who responded to this question indicated that they worked in
conjunction with their law enforcement partners as required by Smart on Crime
policy. However, the remaining 13 districts (20 percent) responded that they did
not work with their law enforcement partners to develop or revise district guidelines
(9 districts) or that they were not sure (4 districts). Also, 3 of the 14 USAs

40

See Appendix 4 for the applicable 2013 Holder memorandum.

16


(21 percent) that we interviewed said they could not recall working with law
enforcement to develop or update their prosecution guidelines.41
Further, five of 25 Drug Enforcement Administration (DEA) and Federal
Bureau of Investigation (FBI) Special Agents in Charge (SAC) that we interviewed
(20 percent) said that they could not recall having worked with the USAs in their
district to develop or revise district guidelines that align with Smart on Crime. We
note that an additional 14 of the 25 SACs (56 percent) had started in their positions
after the Department implemented Smart on Crime and were not aware of whether
the USAs in their district had worked with the previous SAC to develop or revise
district guidelines. While it is possible that discussions between USAs and SACs
regarding districts’ prosecution priorities may have occurred on an ongoing basis,
we found that three of the SACs we spoke with had not become familiar with Smart
on Crime until they prepared for our interview. For example, an FBI SAC we
interviewed stated
that he did not learn
Example of Strong Collaboration – Southern District of Florida
about Smart on
Based on our review of the 94 USAO districts’ policies, we found
Crime until the week
that
the
Southern District of Florida was an example of a district that
before our interview.
worked particularly collaboratively with its state, local, and federal law
enforcement partners in updating its district-specific prosecution
guidelines and training its staff on Smart on Crime policy changes.
This USAO’s leadership consulted with its law enforcement partners
and determined that the district’s prosecution guidelines, while broad,
were consistent with the Holder memoranda and allowed the district
to handle new priorities as they arose. The district also provided
information to the OIG noting its efforts to collaborate with federal law
enforcement partners, including the DEA; the FBI; the Bureau of
Alcohol, Tobacco, Firearms and Explosives; the U.S. Secret Service;
and others by providing training at various management conferences
and interagency working groups regarding the development of
district-specific prosecution guidelines consistent with Smart on
Crime.

By failing to
work with their
federal law
enforcement
partners to develop
or update districtspecific prosecution
guidelines, or to
keep these partners
informed of
significant changes
in Department
policy, districts were not only out of compliance with the Holder memoranda, but
they could have risked creating a disconnect between investigative agencies and
the prosecutors who handle their cases, which may have led to the use of limited
federal law enforcement resources on investigations that ultimately may not have
resulted in expected federal prosecution. Conversely, by working with law
enforcement partners to develop or update district-specific prosecution guidelines
and by keeping them informed of any major changes in Department policy, districts
can maximize resources and focus on the most serious investigations (see the text
box above). While the charging and sentencing practices set forth in the Smart on
Crime initiative have been effectively rescinded by the Attorney General’s May 10,
2017, memorandum, we believe that it remains important for the Department to
ensure appropriate coordination between federal prosecutors and their law
Three USAs that we interviewed were appointed after the Department had implemented
Smart on Crime and speculated that the previous USA in their district may have had these discussions
with their district law enforcement partners.
41

17


enforcement partners in order to target prosecution efforts at the most serious
“local criminal threats and needs.”
The Department’s Ability to Measure the Impact of Smart on Crime or
Other Charging Policies Has Been Limited because Data on Charging
Decisions Has Not Been Consistently Collected
The first two principles of Smart on Crime directed federal prosecutors’ priorities
and charging decisions. However, senior Department officials told us that the EOUSA
case management system, the Legal Information Office Network System (LIONS), would
not be useful for tracking Smart on Crime’s impact on charging decisions, particularly
regarding what would otherwise be safety valve cases and recidivist enhancements.42
Department officials said that LIONS is a user-driven system and that its core
functions involve tracking information such as when a case was opened, when it was
charged, what statute was charged, and what sentence the defendant received.43
While, for example, LIONS allows federal prosecutors to track information on recidivist
enhancements that are charged, federal prosecutors are not required to track this data.
A June 2015 U.S. Government Accountability Office (GAO) report found that
the Department had not established an effective performance measurement system
for Smart on Crime.44 In April 2014, the Department had identified 16 categories of
data, or “key indicators,” to assess the effectiveness of Smart on Crime. Of those
key indicators, 10 involved data collected by the U.S. Sentencing Commission
(USSC) relevant to Smart on Crime principles on prosecutorial priorities and
charging policies, such as the number of drug defendants who received mandatory
minimum sentences, the number of drug defendants who possessed a weapon, and
the number who received a recidivist enhancement.45

As discussed above, the safety valve allows a judge to sentence drug offenders below the mandatory
minimum term if certain conditions, similar to but somewhat more limited than the Holder factors, are met.
42

LIONS is the database used by the 94 USAOs to track case information. According to the
LIONS user manual, the database’s functions allow offices to meet notification requirements for
victims and witnesses, create caseload calendars, monitor workloads, make case assignments, and
respond to data inquiries. Also, EOUSA uses LIONS data to respond to statistical information requests
from the Office of Management and Budget, Congress, and the public. Further, LIONS data provides
the figures for the Attorney General’s Annual Report and the United States Attorneys’ Annual
Statistical Report and is used to formulate budget estimates, justify budget requests, and allocate
resources, including personnel, among the various districts.
43

GAO, Federal Prison System: Justice Could Better Measure Progress Addressing
Incarceration Challenges, GAO-15-454 (June 2015).
44

Specifically, the ten indicators were: “(1) Number of defendants subject to (sentenced to)
mandatory minimum sentence under 21 U.S.C. § 841; (2) (Number of defendants sentenced to) no
mandatory minimum; (3) (Number of defendants sentenced to) 5-year mandatory minimum; (4) (Number of
defendants sentenced to) 10-year mandatory minimum; (5) No mandatory minimum plus safety valve
(number of drug defendants not charged with an offense carrying a drug mandatory minimum that also
qualified for the safety valve); (6) Mandatory minimum plus safety valve (number of drug defendants charged
with an offense carrying a drug mandatory minimum that also qualified for the safety valve);
(Cont’d)
45

18


The Department, however, did not publicly specify how it would interpret these
indicators to assess whether Smart on Crime was successfully implemented. The GAO
recommended that the Department establish target directions for the indicators to
better reflect elements of effective performance measurement systems, but the
Department did not concur with this recommendation, arguing that prosecutors must
charge cases in a manner appropriate to the individual circumstances of cases rather
than with the goal of advancing national numerical targets.
Using these same indicators, however, in March 2016, the Department
reported that its assessment of trends in USSC data showed that Smart on Crime
was having the desired impact.46 Specifically, the Department reported, based on
these sentencing trends, that federal prosecutors:
	 Took on fewer and more serious drug cases: The number of drug offenders
convicted decreased nationwide, while the percentage of drug offenders with
a weapon or an aggravating role increased.
	 Referred more safety valve cases to state prosecutors: The percentage of
drug offenders who qualified for the safety valve (first-time, non-violent drug
offenders whose cases did not involve guns) decreased.47
	 Charged fewer mandatory minimums: The percentage of offenders
sentenced to mandatory minimum terms decreased even while prosecutors
pursued more serious cases.
	 Experienced no negative impact on defendant cooperation with the government:
The rates of drug offenders pleading guilty and cooperating with the government
to make cases against others stayed constant. Federal prosecutors’ ability to
obtain defendants’ cooperation did not decline as a result of the reduced number
of mandatory minimums charged and recidivist enhancements filed.
While we agree with the Department that certain USSC data points can be
used to draw inferences as to how Smart of Crime is affecting sentencing
proceedings, we do not believe that the USSC data is a substitute for the
Department collecting data that it is uniquely positioned to gather and that would
allow it to undertake a more consequential and meaningful analysis of the impact of
Smart on Crime or other charging policies. There are significant limitations on the
extent to which USSC data can inform the Department and the public as to
(7) Number of drug defendants with weapon involvement; (8) Number of drug defendants with an aggravating
role adjustment; (9) Number of drug defendants with a mitigating role adjustment; and (10) Number of 851enhanced charges (number of 851-enhanced mandatory minimum sentences).” The remaining six indicators
addressed other Smart on Crime principles such as reentry programs and drug diversion courts.
46 DOJ Press Release, “New Smart on Crime Data Reveals Federal Prosecutors Are Focused on
More Significant Drug Cases and Fewer Mandatory Minimums for Drug Defendants,” March 21, 2016.
47 We did not examine federal referrals of cases to state prosecutors or state sentencing data,
which would help to confirm whether low-level drug offenders were referred more often to state
prosecutors. Further, we are unaware of any such analysis by the Department. Absent such information,
we are unable to assess the Department’s claim that a decline in the number of federal safety valve cases
means that federal prosecutors referred more safety valve eligible cases to state prosecutors.

19


(a) whether prosecutorial charging decisions are being made consistent with Smart
on Crime or other charging policies and (b) the impact of Smart on Crime or other
charging policies on sentencing outcomes. Perhaps the most significant limitation is
that the USSC collects data from judges on the sentencing decisions that they have
made but does not receive data from prosecutors about their charging decisions.
Thus, while the USSC data informs the public on how judges sentenced
defendants based on charges that were actually brought by prosecutors and that
ultimately resulted in a conviction, the data does not allow for any assessment as to
charges that (a) could have been but were not brought by prosecutors or (b) did not
result in a sentencing proceeding (i.e., where a defendant was acquitted on all counts
or the charges were dismissed). Similarly, the USSC does not have data on what a
convicted defendant’s minimum sentence would have been had prosecutors sought a
mandatory minimum charge or recidivist enhancement at sentencing. Accordingly, we
believe that the Department’s limited collection of relevant and reliable data on
charging decisions presents challenges for its and the public’s further assessment of
the implementation and effectiveness of Smart on Crime or other charging policies.
USSC Data Reflects that There Have Been Shifts in Drug Sentencing
Consistent with Smart on Crime’s First Two Principles, but Some Regions
and Districts Have Diverged from National Trends
According to the Department’s statements at the time of the Smart on Crime
announcement, the first two principles of Smart on Crime were designed to focus
scarce federal law enforcement resources on the most serious cases that clearly
align with federal interests and to reserve the most severe penalties in drug cases
for serious, high-level, or violent offenders. Our analysis of USSC and other data
shows that drug sentences have shifted, in some areas substantially, in a manner
consistent with these first two Smart on Crime principles. Specifically, we found
that during the implementation of Smart on Crime, there were fewer mandatory
minimum sentences imposed in drug cases, more drug offenders received weapon
or aggravating role enhancements at sentencing, and there were fewer sentencings
of low-level drug offenders federally. Additionally, while there was limited data
available for analysis, we also found indications of a reduction in the number of
recidivist enhancements sought by prosecutors at sentencing.
Another goal of the first two Smart on Crime principles was to reduce the
burden on federal prisons that were over capacity. We noted that the federal
prison population began to decline in 2014 for the first time in over 30 years. While
this decline coincided with the announcement of Smart on Crime, it also coincided
with the USSC’s adoption of its Drugs Minus Two guidelines amendment, which
lowered the base offense level by two levels for most drug offenses, and its decision
to make that guideline amendment retroactive, both of which became effective on
November 1, 2014.48 As evidenced by the data developed by the USSC regarding
The Federal Sentencing Guidelines assign each offense a base offense level, ranging from
1 to 43, primarily determined by the seriousness of a particular offense. More serious types of crime
(Cont’d)
48

20


the impact of its guideline decisions, the decisions not only had an immediate
impact on thousands of pending federal sentencings involving drug offenses, but
also impacted the release date for thousands of inmates convicted of drug
sentences that had already been imposed.49 By contrast, the Department did not
have sufficient data regarding prosecutorial charging decisions to allow us to
analyze whether the Department’s policy changes had an immediate effect on the
federal prison population or whether that effect would be seen over a period of
time. Accordingly, we did not undertake such an evaluation.
USSC Data Indicates that the Implementation of Smart on Crime’s First Two
Principles Had an Impact at the National Level
The OIG used USSC data and the Department’s list of Smart on Crime key
indicators to conduct an analysis that covered several of the data categories
mentioned in the Department’s March 2016 press release. We found similar
indications that Smart on Crime may have had the nationwide effects desired by
the Department when it announced the initiative in 2013.50
Fewer Mandatory Minimum Convictions in Drug Cases
As part of Smart on Crime, the Department directed federal prosecutors to
decline to charge the quantity of drugs necessary to trigger a mandatory minimum
sentence in cases when certain factors were present.51 We found that there has
been a significant increase in the percentage of drug offenders not sentenced to
mandatory minimum sentences nationally since Smart on Crime was announced.
Specifically, the percentage of federal drug offenders not sentenced to a mandatory
minimum term rose from 39.7 percent in FY 2012 to 49.9 percent in FY 2014. By
FY 2015, over half of all federal drug offenders, 54.2 percent, were sentenced with

have higher base offense levels, which are then increased or decreased by specific offense
characteristics that may be typical of the offense. Other adjustments discussed below, such as those
made for aggravating or mitigating roles, can also affect the final adjusted offense level for a
particular defendant. The total adjusted offense level for the offense or offenses of conviction and the
defendant’s criminal history category are used to arrive at the sentencing guideline range, which the
sentencing court considers in imposing sentences, subject to any applicable mandatory minimum
sentence as discussed herein, or to any statutory maximum that may apply.
See § 2D1.1 of the Federal Sentencing Guidelines for the complete Drug Quantity Table.
The amendment to the Federal Sentencing Guidelines became effective on November 1, 2014.
49

DOJ Press Release, “New Smart on Crime Data Reveals Federal Prosecutors Are Focused on
More Significant Drug Cases and Fewer Mandatory Minimums for Drug Defendants.” USSC data
provided to the OIG yielded slightly different figures in our analysis than the Department cited in its
press release, but these variations were minor. The variations were caused by differences in
methodology between the USSC’s publication of data in its sourcebook, which was also the
methodology that the Bureau of Justice Statistics used to provide data to the OIG, and the data the
USSC provided to the Department. These methodological differences included the exclusion of cases
missing guideline information and offenders sentenced for simple possession, counting mandatory
minimums of less than 5 years as if they were not mandatory minimums, and the inclusion of
mandatory minimums over 10 years with the count of 10-year mandatory minimums.
50

51

See Appendix 5 for the applicable 2013 Holder memorandum.

21


no mandatory minimum, according to USSC data.52 Conversely, USSC data shows
that the percentage of 5-year mandatory minimum cases decreased, from
28.0 percent of federal drug offenders’ sentences in FY 2012 to 20.8 percent by
FY 2015. Similarly, the percentage of federal drug offenders sentenced with a
10-year mandatory minimum or greater, typically based on larger quantities of
drugs, fell from 32.4 in FY 2012 to 25.0 in FY 2015.
In order to further evaluate whether Smart on Crime had an effect on
prosecutors’ charging decisions, we examined data on the rates of defendants with
mandatory minimum sentences who received relief under the safety valve provision
and the rates of defendants with two criminal history points at higher base offense
levels who received a mandatory minimum sentence. Additionally, we analyzed
prosecutors’ survey responses regarding their charging practices.


Mandatory Minimum Convictions and Safety Valve Relief

The safety valve is a statutory exception for drug offenders who meet certain
conditions similar to, but more limited than, the Holder factors. Under the safety
valve provision, even if a defendant is charged with and convicted of a mandatory
minimum offense, a sentencing judge is not required to impose a mandatory
minimum sentence if the defendant meets the safety valve conditions. The most
significant difference between the safety valve conditions and the Holder factors is
that, under the safety valve, defendants can have no more than one criminal
history point, while the Smart on Crime policy applied to drug offenders who do not
have “a significant criminal history,” which will “normally be evidenced by three or
more criminal history points,” but may involve more or fewer depending on the
nature of the prior convictions.
Although there is overlap between the safety valve and the Holder factors, some
defendants who appeared to have met the Holder factors’ criteria might still have
received safety valve relief at sentencing rather than benefiting from Smart on Crime
at the charging stage of a case. For example, when a defendant is determined to have
one criminal history point, a prosecutor could determine based upon the defendant’s
prior conviction record that the nature of the defendant’s prior convictions constituted
a significant criminal history, and therefore, the prosecutor decided to charge the
defendant with an offense carrying a mandatory minimum sentence, while the judge

We note that the most common types of primary drug offenses shifted over this time period
and that each drug requires different quantities to trigger a mandatory minimum sentence. For
example, the percentage of federal drug convictions involving methamphetamine and heroin rose
every year from FY 2010 to FY 2015, while the percentage involving cocaine fell. We observed that
methamphetamine offenders experienced the largest drop in the rate of 10-year mandatory minimum
sentences after Smart on Crime, with 60.5 percent of methamphetamine offenders receiving 10-year
mandatory minimums in FY 2012 but only 35.4 percent receiving 10-year mandatory minimums in
FY 2015. Additionally, from FY 2012 to FY 2015, the percentage of methamphetamine offenders who
were sentenced with no mandatory minimum of any length rose from 16.7 to 45.5 percent. We were
unable to determine what effect, if any, this shifting of the types of drugs being prosecuted had on the
Department’s charging decisions separate and apart from the impact of Smart on Crime.
52

22


might nevertheless determine that the defendant’s criminal history is not that
significant and grant the defendant relief under the safety valve provision.
We found that there has been a substantial decrease in the percentage of
convictions where the defendant received the safety valve at sentencing since
Smart on Crime was announced. Given the data cited previously showed a
decrease in the percentage of drug offenders being convicted of mandatory
minimum offenses, this decrease is to be expected given that there are fewer
mandatory minimum offenders who need the benefit of the safety valve. We also
found a substantial increase since the implementation of Smart on Crime in the
percentage of offenders who were not eligible to receive safety valve relief but who
nevertheless were not convicted of a mandatory minimum offense, a category that
would include defendants who met the Holder factors and were charged in a
manner consistent with Smart on Crime principles.53
Nevertheless, the percentage of drug offenders who received the safety valve
in FY 2015 was still significant, at just over 13 percent. We had expected this
amount to be lower because most of these defendants would also have met the
criteria of Smart on Crime and therefore would have been within the universe of
defendants that the Holder memorandum determined should not have been
charged with a mandatory minimum offense by the prosecutor. However, given the
absence of Department charging data, we were unable to assess the reasons why
these defendants were charged with mandatory minimum offenses. See Table 1 for
the percentage of drug offenders who received mandatory minimums and who
qualified for the safety valve during the years for which the OIG received USSC
data, FY 2010 through FY 2015.
Table 1 

Percentage of Drug Offenders Receiving Mandatory Minimums or the 

Safety Valve, FY 2010 – FY 2015


% No Mandatory Minimum
Conviction/Would Not Have Been
Safety Valve Eligible
% Mandatory Minimum
Conviction/Not Eligible for Safety
Valve
% No Mandatory Minimum
Conviction/Would Have Been
Safety Valve Eligible
% Mandatory Minimum
Conviction/Received Safety Valve

FY
2010

FY
2011

FY
2012

FY
2013

FY
2014

FY
2015

21.5%

23.4%

24.6%

27.2%

32.3%

34.9%

41.5%

38.6%

36.9%

38.1%

34.4%

32.9%

13.3%

15.0%

14.7%

10.2%

17.2%

19.1%

23.7%

23.0%

23.8%

24.5%

16.1%

13.1%

Source: OIG analysis of USSC data

The safety valve does not repeal or eliminate mandatory minimum sentences. See
Appendix 7 of this report and 18 U.S.C. § 3553(f).
53

23


Table 1 reflects that charging decisions shifted in FY 2014 and 2015, after
Smart on Crime. Specifically, in FY 2010, prior to Smart on Crime, the largest
category of drug offenders (41.5 percent) were those sentenced with mandatory
minimums who did not receive safety valve relief, followed by about half as many
offenders who obtained relief from mandatory minimums through the safety valve
(23.7 percent). With these two categories combined, almost two-thirds of all
federal drug cases sentenced in that year resulted in the imposition of mandatory
minimum sentences or would have done so absent relief under the safety valve.
However, by FY 2015, safety valve relief had become less essential in helping
offenders avoid mandatory minimums. As a result, safety valve relief from
mandatory minimums dropped over 10 percentage points, to only 13.1 percent of
FY 2015 drug offenders, while defendants given mandatory minimum sentences fell
to 32.9 percent, meaning that with these two categories combined, less than half of
all federal drug cases resulted in mandatory minimum sentences or would have
done so without application of the safety valve. Similarly, offenders who would not
have been eligible for the safety valve and who nevertheless were not charged with
an offense triggering a mandatory minimum made up the largest category of drug
convictions (34.9 percent).
	 Mandatory Minimum Rates for Offenders with Two Criminal
History Points
In addition, we analyzed the mandatory minimum sentence rates of drug
offenders who were sentenced with two criminal history points and at base offense
levels 24 or higher.54 As stated previously, these offenders would not have been
eligible for relief under the safety valve provision due to their criminal history but
could have fallen within the Smart on Crime criteria. Therefore, these drug
offenders potentially stood to benefit the most from Smart on Crime policies
because they are a category of drug offenders who previously might have received
a mandatory minimum sentence but were able to avoid this under the initiative.
We found that a higher percentage of these offenders were sentenced without
mandatory minimums in FY 2014 and FY 2015 compared to the previous fiscal
years. See Figure 1 below for the percentage of offenders with two criminal history
points who were sentenced with mandatory minimums.55

Drug offenders with two criminal history points comprised 1,125 offenders sentenced with a
primary drug offense on average each fiscal year.
54

This analysis did not exclude offenders who received an adjustment in their sentence for
providing substantial assistance since the rate of these offenders remained constant throughout our
scope.
55

24


Figure 1
Percentage of Offenders with Two Criminal History Points and Base
Offense Level 24 or Higher Receiving Mandatory Minimums
FY 2010 – FY 2015
60.0%
55.0%
50.0%

49.1%

48.5%

47.8%
44.9%

45.0%
40.0%
35.0%

34.5%
32.3%

39.1%
35.6%

30.0%

35.7%

34.8%

25.0%

27.6%

20.0%

23.4%

15.0%
10.0%
FY10

FY11

FY12

FY13

FY14

FY15

Mandatory Minimum Sentence Imposed ‐ All Base Offense Levels
Mandatory Minimum Sentence Imposed ‐ Base Offense Level 24 or Higher
Source: OIG analysis of USSC data

	 Survey Data on Charging Drug Quantities That Trigger
Mandatory Minimums
The OIG’s survey data also suggests that charging practices changed since
Smart on Crime. Nearly half of the respondents to the OIG’s survey (49.4 percent,
or 404 of 817 AUSAs) stated that the Smart on Crime criteria affected their
charging decisions with regard to low-level, non-violent drug offenders without
significant criminal histories.56 Specifically, these prosecutors and supervisors
responded that they were now less likely to recommend or approve charging drug
quantities that would trigger mandatory minimum sentences for low-level drug
defendants. A further 19.7 percent of the survey respondents (161 of 817 AUSAs)
stated that it was already their practice prior to Smart on Crime not to charge lowlevel, non-violent defendants with mandatory minimum-triggering drug quantities.

The 817 respondents were those AUSAs who responded substantively to the question
regarding mandatory minimums. This number excludes those respondents who skipped the question
or marked “not applicable” as an answer. See Appendix 2 for a summary of the OIG’s survey results.
56

25


The Prioritization of Federal Drug Prosecutions
As noted above, the Department had previously cited several metrics which
showed that its Smart on Crime policy changes met their intended outcomes and
that the Department prosecuted fewer but more serious cases to better prioritize
resources. These metrics included a drop in the overall number of federal drug
convictions, a rise in the percentage of offenders with a weapon or aggravating role
as an indication that more serious offenders were being prosecuted, and a drop in
the percentage of offenders who were eligible for the safety valve as an indication
that fewer low-level offenders were being prosecuted.
Our analysis found these trends as well. However, the metrics the
Department used and the available sentencing data limited our ability to draw
definitive conclusions regarding the influence the Smart on Crime charging policies
had because these trends could be influenced by a number of other factors and do
not include information on the reasons for charging decisions. We discuss our
analysis of the metrics and their limitations below.


The Number of Federal Drug Convictions

We found that the overall number of federal drug convictions decreased by
18.8 percent between FY 2012 and FY 2015, but this decrease occurred within the
context of shifting trends in federal criminal cases as whole, making it difficult to
identify the causes for the decrease. For example, during this period, the number
of all federal criminal convictions decreased by 15.6 percent and drug cases as a
percentage of all criminal cases actually increased. According to the USSC, in
FY 2010, defendants with a drug crime as the primary offense accounted for
28.9 percent of federal convictions, rising to 30.2 percent in FY 2012 and then to
31.8 percent by FY 2015. In comparison, the next largest offense category,
immigration, made up 34.4 percent of federal convictions in FY 2010 but decreased
to 32.2 percent in FY 2012 and then to 29.3 percent by FY 2015. See Figure 2
below for the number of federal drug offenders during the years for which the OIG
received USSC data, FY 2010 through FY 2015.57

Department and USSC officials we interviewed also suggested that declines in USAO
attorney staffing levels resulting from a Department hiring freeze, as well as the 2013 budget
sequestration, may have had an influence on trends in the number of federal drug convictions over
our period. We found that a third of districts were understaffed by 10 percent or more on average
from FY 2012 through 2015. However, the OIG’s analysis of USAO staffing data did not find a
statistically significant correlation between understaffing in the districts and changes in the number of
drug cases.
57

26


Figure 2

Number of Federal Drug Convictions and the Percentage of Drug 

Convictions to All Federal Criminal Convictions 

FY 2010 – FY 2015 

100%


25,000
20,000

24,360
 25,264
 25,357


80%
22,881
 21,893


15,000

20,598


40%

10,000
5,000

60%

28.9%


29.1%


30.2%


31.2%


31.7%


31.8%


FY10

FY11

FY12

FY13

FY14

FY15

0

20%

Percent of All Convictions

Number of Convcitions

30,000

0%
Number of Drug Convictions
Drug Convictions as a Percentage of All Criminal Convictions

Source: OIG analysis of USSC data



More Serious Drug Offenders Prosecuted

We found that the percentage of drug defendants who possessed a weapon
rose from 15 percent in FY 2012 to 16.2 in FY 2014, and to 17.2 percent in
FY 2015. The percentage of defendants with an aggravating role in the offense,
such as leadership of a drug conspiracy, under the sentencing guidelines also
increased, from 6.6 percent of drug offenders in FY 2012 to 7.1 percent in 2014
and 7.7 percent in 2015.58 See Figure 3 below for the percentage of drug offenders
with a weapon or aggravating role from FY 2010 through FY 2015.59

In determining a defendant’s offense level, adjustments may be made for the defendant
having an “aggravating” or “mitigating” role. The aggravating role adjustment relates to whether the
defendant is an organizer, manager, or leader of the criminal activity at issue. If the defendant has
such a role, the adjusted offense level can be increased by up to four offense levels.
58

59 The percentage of defendants who received a mitigating role adjustment of two to four
offense levels downward, for playing a minor or minimal role in a drug conspiracy, showed a less
distinct trend, remaining close to 18 percent from FY 2010 through FY 2013, rising to 19.1 percent in
FY 2014, and then falling to 16.7 percent in FY 2015.

27


Figure 3

Percentage of Drug Offenders with a Weapon or an 

Aggravating Role, FY 2010 – FY 2015

20%
15%

16.4%

17.2%

16.2%

16.2%

6.6%

7.0%

7.1%

7.7%

FY12

FY13

FY14

FY15

15.6%

15.0%

6.0%

6.0%

FY10

FY11

10%
5%
0%
% Weapon

% Aggravating Role

Source: OIG analysis of USSC data

Another indication that federal prosecutors may have been focusing on more
serious drug cases is that they were prosecuting drug offenders with larger drug
quantities. The base offense level is the starting point for determining the
seriousness of an offense, with larger drug quantities having higher base offense
levels. While the average base offense level for drug offenders remained relatively
similar in each fiscal year, we found that the percentage of drug offenders with a
base offense level of 24 or higher had increased. Specifically, from FY 2011 to
FY 2012, 73 to 74 percent of drug offenders had a base offense level of 24 or
higher, as compared to FY 2013 to FY 2015, where 77 to 79 percent of these drug
offenders had a base offense level of 24 or higher.60


Fewer Low-Level Drug Offenders Prosecuted

Our analysis also showed that offenders who were eligible for the safety
valve (first-time, non-violent drug offenders whose cases did not involve guns) fell
from 38.5 percent of those federally sentenced in FY 2012 to 32.3 percent by
FY 2015. In addition, the percentage of drug offenders with a limited criminal
history (those with three or fewer criminal history points in Criminal History
Categories I or II) saw a modest decline.61 Defendants with such lesser criminal
60 As noted above, in November 2014, the USSC enacted the Drugs Minus Two guidelines
amendment, which decreased the overall base offense levels for most drug offenses by two offense
levels. Therefore, the increase in the number of drug offenders who were sentenced at offense level
24 or higher also seems to indicate that they had even higher drug quantities, but benefited from the
overall two-level reduction promulgated under the amendment.

According to the Federal Sentencing Guidelines, defendants are assigned criminal history
points based upon prior criminal sentences, and then placed in one of six Criminal History Categories.
(Cont’d)
61

28


histories went from 65.1 percent of all federal drug offenders in FY 2012 to
61 percent in FY 2015. See Figure 4 for the percentage of drug offenders with a
Criminal History Category of I or II or who were eligible for the safety valve from
FY 2010 through FY 2015. The decline in these percentages may indicate that
federal prosecutors were charging relatively fewer low-level drug defendants while
Smart on Crime was in effect.
Figure 4

Percentage of Drug Offenders with a Criminal History 

Category of I or II or Who Were Eligible for the Safety Valve 

FY 2010 – FY 2015 

70%
60%

63.2%

65.0%

65.1%

37.0%

37.9%

38.5%

FY10

FY11

62.1%

62.0%

61.0%

34.7%

33.3%

32.3%

FY13

FY14

FY15

50%
40%
30%
20%
10%
0%
FY12

% Criminal History I or II

% Safety Valve

Source: OIG analysis of USSC data

The Number of Recidivist Enhancements Filed Appears to Have Declined since the
Implementation of Smart on Crime
The 2013 Holder memorandum directed federal prosecutors to decline to file
recidivist enhancements unless the defendant’s conduct made the case appropriate
for severe sanctions.62 The Department provided to the OIG a summary of a USSC
project conducted to estimate the number of recidivist enhancements filed under

Criminal History Category I is the least serious category and includes many first-time defendants.
Criminal History Category VI is the most serious category and includes offenders with extensive criminal
records. Defendants with two or three criminal history points are in Criminal History Category II.
62 21 U.S.C. § 851 requires prosecutors to file an information setting forth any prior felony
drug conviction that will be used to increase a defendant’s sentence and provides procedures by which
a defendant can challenge any such conviction, which, if applied by the court, can potentially enlarge a
mandatory minimum sentence from 5 to 10 years, from 10 to 20 years, or from 20 years to life. We
were unable to fully assess this key aspect of Smart on Crime’s second principle because, as
previously discussed, the Department collects limited data on recidivist enhancements. See
Appendix 5 for the applicable Holder memorandum and the factors discussed therein.

29


21 U.S.C. § 851.63 Our review of the USSC project summary found that the
Department’s use of recidivist enhancements since the implementation of Smart on
Crime appeared to have declined.
According to the USSC, in FY 2012, federal prosecutors filed recidivist
enhancements in approximately 20.6 percent of eligible cases, but in only
17.6 percent of eligible cases in FY 2014.64 Also, this decline in the rate of recidivist
enhancements occurred despite a rise in the percentage of eligible defendants. An
estimated 27.9 percent of drug offenders (7,078 of 25,357 defendants) were
eligible for a recidivist enhancement in FY 2012, while 31.8 percent (6,972 of
21,893) were eligible in FY 2014.
In addition, of the recidivist enhancements that federal prosecutors filed,
many were withdrawn. In FY 2012, prosecutors withdrew 17.8 percent of filed
recidivist enhancements prior to sentencing. The percentage increased to 26.1 in
FY 2014, resulting in a total of approximately 1,200 recidivist enhancements
applied at sentencing in FY 2012 and 905 in FY 2014. Finally, the percentage of
federal drug defendants who actually received a recidivist enhancement in FY 2012
was estimated at 3.4 percent (approximately 850). The percentage decreased to
2.9 percent (approximately 625) in FY 2014.
Our survey results supported the USSC’s findings based on its sampling
project. Approximately 51.6 percent of our survey respondents (428 of
829 AUSAs) indicated that since Smart on Crime they became less likely to file
recidivist enhancements when a defendant is eligible. An additional 20.6 percent of
survey respondents (171 of 829 AUSAs) stated that it was already their practice
prior to Smart on Crime not to file these enhancements in most cases when the
defendant was eligible.65
Some Regions Diverged from National Trends in Number of Drug Cases and
Mandatory Minimums
While the total number of federal drug convictions decreased nationwide
following the implementation of Smart on Crime, we found noticeable variations in the
number of drug convictions and mandatory minimums in different regions of the
country. In some regions, the number of drug convictions increased, while other

At the time of our review, the USSC’s project provided the only available data on the rates
of recidivist enhancement filings for the fiscal years immediately before and after Smart on Crime
(FY 2012 and FY 2014).
63

The USSC stated that it determined eligibility by identifying those cases in its sample in
which the offender had previously been convicted of a drug trafficking, drug possession, or other type
of drug offense and in which the sentence imposed was 13 months or longer. Offenders meeting
these criteria were deemed to have been convicted of a prior felony drug offense and thus met the
requirement for application of an 851 enhancement.
64

The 829 respondents were those AUSAs who responded substantively to the question
regarding recidivist enhancements. This number excludes those respondents who skipped the question or
marked “not applicable” as an answer. See Appendix 2 for a summary of the OIG’s survey results.
65

30


regions saw significantly larger declines, proportionally. See Figure 5 for the percent
changes in the number of drug convictions by region from FY 2012 through FY 2015.
Figure 5
Percent Changes in the Number of Drug Convictions by Region
FY 2012 – FY 2015
30%
20%
10%
0%
‐10%
‐20%
‐30%
‐40%

% Change FY 12‐15

Southwest
Border
‐34.2%

Florida

California
(non‐
border
districts)

East

South

Midwest

‐31.3%

‐19.9%

‐16.8%

‐13.4%

‐12.4%

Hawaii and
Pacific
Island
Northwest
Territories
3.0%

15.3%

West
22.0%

Source: OIG analysis of USSC data

As Figure 5 shows, the West region, which includes districts such as Idaho,
Wyoming, and the non-border districts of Texas (the Northern and Eastern Districts
of Texas), taken together, saw a 22 percent increase in the number of drug
convictions from FY 2012 to FY 2015. The Pacific Northwest region (Oregon,
Washington, and Alaska), as well as the region consisting of Hawaii and the Island
Territories (Puerto Rico, Guam, the Northern Mariana Islands, and the Virgin
Islands), also experienced an increase in the number of drug convictions following
the implementation of Smart on Crime.
We found that many states and several territories in these regions had a
relatively high proportion of methamphetamine offenders, which made up
50 percent or more of all drug sentences. For example, in Guam and the Northern
Mariana Islands, 100 percent of drug offenders sentenced in FY 2015 were
convicted of methamphetamine offenses, while in Hawaii the proportion was
94 percent. In Montana, Wyoming, Utah, and Idaho — all in the West region — the
proportions were 92, 74, 72, and 71 percent, respectively. Across all 94 districts,
there was a statistically significant positive correlation between the percentage of
methamphetamine offenders in FY 2015 and the percentage increase in the number
of drug cases from FY 2012 to FY 2015. According to the USSC, the number of
federal methamphetamine offenders rose across the country by 41 percent from
FY 2011 to FY 2015, and these offenders also made up the highest percentage of
drug offenders in 27 states in FY 2015, mostly in the South, Midwest, and Western
regions.

31


We also found that districts in the Southwest Border region (Arizona, New
Mexico, Southern California, Southern Texas, and Western Texas) collectively saw a
34.2 percent decrease in the number of drug convictions from FY 2012 to 2015. By
comparison, all other regions combined experienced an 11.3 percent decrease in
the number of drug convictions. Other regions with large declines in the number of
drug convictions were Florida, the East (from Maine to the District of Columbia),
and California (excluding the border). The Midwest and the South saw more
moderate decreases.66
Weapons and Aggravating Roles at the District Level
As discussed above, the Department identified cases in which a weapon was
involved or the offender had an aggravating role as indicative of more serious drug
offenders. The percentage of drug cases in which a weapon was involved increased
nationally from 16.4 percent to 17.2 percent from FY 2010 to FY 2015, and the
percentage of defendants with an aggravating role increased from 6.0 percent to
7.7 percent during the same period. At the district level, however, these two
indicators varied and showed a less clear trend than the aggregated data.
Specifically, in 40 of 94 districts the percentage of offenders with a weapon
and the percentage of defendants with an aggravating role trended in opposite
directions from FY 2012 to FY 2015. In 18 districts, the percentage with a weapon
rose while the percentage with an aggravating role fell. In 22 districts, the
percentage with a weapon fell and the percentage with an aggravating role rose. In
15 districts, both the percentage with a weapon and the percentage with an
aggravating role trended downward. In the remaining 39 districts, both indicators
increased during the period of our review. Nevertheless, the fact that these
indicators so often displayed contradictory trends at the district level suggests that
local crime conditions may play a significant role in these numbers and that caution
should be exercised in drawing a conclusion from the national trend that
prosecutors have charged more serious offenders since Smart on Crime.
The Impact of the Southwest Border Region on National Trends
We found that the Southwest Border region faced a number of unique
circumstances that made its districts outliers from national trends identified through
our analysis of USSC data. A senior Department official told us that due to its
location, this region handles more drug cases that involve border crossings of
foreign nationals (often low-level, non-violent drug offenders) as compared to other
regions. Because many of these border region offenders are foreign nationals, they
are almost always prosecuted federally. These border drug offenders also often
have little or no criminal history in the United States.
During FYs 2012 through 2015, the Southwest Border region’s five districts
were responsible for a range of 26 to 33 percent of all drug convictions annually.
This region significantly contributed to the 19 percent decrease in drug convictions
We discuss the unique circumstances of the Southwest Border region further in “The
Impact of the Southwest Border Region on National Trends” below.
66

32


nationally from FY 2012 to FY 2015. We found that if we removed the Southwest
Border region from our analysis the decrease in drug convictions nationally would
be only 11 percent.
Also, this region consistently had the highest percentage of offenders who
were eligible for safety valve relief and a high percentage of offenders with three or
fewer criminal history points compared to other regions. From FY 2012 through
2015, more than half of drug offenders each year in the Southwest Border region
were eligible for safety valve relief, approximately twice the percentage in all other
regions combined. In addition, this region had the highest proportion of marijuana
offenders compared to other regions, ranging from 59 percent of its drug offenders
in FY 2012 to 42 percent in FY 2015. Finally, as Table 2 shows, all regions of the
country saw increased percentages of drug offenders sentenced without a
mandatory minimum from FY 2012 through FY 2015, but the Southwest Border
Region consistently had the highest percentage over this time period.
Table 2 

Percentages of Drug Offenders Sentenced with

No Mandatory Minimum, by Region 

FY 2012 – FY 2015

Region
Southwest
Border

FY 2012

FY 2014

FY 2015

47.0%

59.8%

65.4%

South

43.4%

55.3%

60.8%

East
Pacific
Northwest
Midwest

42.9%

52.6%

52.5%

36.6%

49.7%

52.0%

32.3%

42.6%

51.6%

West

30.7%

41.3%

43.8%

California

28.0%

38.4%

41.1%

40.7%

37.0%

22.1%

26.2%

Florida
27.0%
Hawaii and
Island
21.0%
Territories
Source: OIG analysis of USSC data

Given that the Southwest Border region was a major driver of many of the
national trends, and in light of the significant divergence reflected in Table 2,
national trends should not be interpreted in such a way as to conclude that Smart
on Crime had a uniform impact across all of the nation’s districts.

33


CONCLUSION AND RECOMMENDATIONS 

Conclusion
The OIG found overall that the Department took a number of actions to
implement the first two principles of Smart on Crime, but we also identified several
shortcomings in its efforts. While the Department issued policy memoranda and
guidance for the application of Smart on Crime policy changes, the U.S. Attorneys’
Manual, used by federal prosecutors to make charging decisions consistent with
Department policy, was not revised to reflect Smart on Crime policies until January
2017, more than 3 years after Smart on Crime was launched.
We found that most districts had updated local prosecution guidelines to
reflect local priorities as required by Smart on Crime and had adopted policies and
procedures to implement the initiative’s provisions regarding the use of mandatory
minimums and recidivist enhancements. The latter was done, in most cases, by
updating prosecution guidelines, policy memoranda, templates, or worksheets to
reflect the policy changes. We believe that the Department should ensure that all
districts update their policies to ensure that they are consistent with current
Department charging policies.
We also found that most districts required supervisory approval before a case
could be indicted with a drug quantity that would trigger a mandatory minimum
sentence and before an Assistant U.S. Attorney could file a 21 U.S.C. § 851
recidivist enhancement. However, in some districts, the supervisory approval
process was not documented in the policies we were provided. We believe that the
Department should ensure that supervisory approval is required, communicated,
and clearly documented in each district. Further, while most U.S. Attorney’s Offices
(USAO) had updated their guidelines, it is not clear whether all districts had
consulted with their law enforcement partners, as required by Smart on Crime.
This obviously needs to occur on an ongoing basis.
The Department’s ability to measure the effectiveness of Smart on Crime or
other charging policies is limited because it does not consistently collect data on
charging decisions. The U.S. Attorneys’ case management system does not keep
track of all the requisite information, and what information it collects is not always
updated. Due to these limitations, the Department has relied on U.S. Sentencing
Commission (USSC) data to assess the effectiveness of Smart on Crime. However,
we believe that the Department cannot rely solely on after-the-fact sentencing
information that does not contain all the requisite data, and that USAOs should
collect and report charging data so that the Department can accurately measure
the implementation and impact of its charging polices.
Nevertheless, based on our own analysis of USSC data, we found that the
Department’s Smart on Crime plan had an impact on charging decisions, as
reflected in shifts in a number of different metrics that we believe to be consistent
with the initiative’s first two principles and the responses we received from our
survey to all of the USAOs. We also found that some regions diverged from
national trends. For example, while drug convictions decreased nationally by
34


19 percent, we found that the Hawaii and Island Territories Region, the Pacific
Northwest Region, and the West Region had an increase in the number of drug
convictions.
Recommendations
In order to ensure that all federal prosecutors have clear and consistent
guidance regarding Department charging policies, we recommend that the
Department:
1.	

Ensure that the U.S. Attorneys’ Manual accurately reflects Department
charging policies.

2.	

Ensure that all U.S. Attorney’s Offices consult with their law enforcement
partners to make sure that their policies are current and consistent with local
prosecution priorities.

In order for the Department to accurately measure the impact of its charging
policies, we recommend that the Department and the Executive Office for United
States Attorneys:
3.	

Require all U.S. Attorney’s Offices to collect charging data that will enable the
Department to determine whether its charging and sentencing policies are
being effectively implemented.

35


APPENDIX 1 

METHODOLOGY OF THE OIG REVIEW
In this review, the OIG examined how the Department prioritized
prosecutions to focus on the most serious cases, as well as the implementation of
the Smart on Crime charging reforms that were intended to eliminate what the
former Administration characterized as unfair disparities in the criminal justice
system. The review covered the period from FY 2012 through FY 2015. Our
fieldwork, performed from January 2016 through July 2016, included data analysis,
a survey of U.S. Attorney’s Office (USAO) district Criminal Chiefs and Assistant U.S.
Attorneys (AUSA), document analysis, and interviews. It did not assess the new
charging and sentencing policy announced by the Attorney General by
memorandum dated May 10, 2017. The following sections provide additional
information about our methodology.
Standards
The OIG conducted this review in accordance with the Council of the
Inspectors General on Integrity and Efficiency’s Quality Standards for Inspection
and Evaluation (January 2012).
Data Analysis
The OIG requested U.S. Sentencing Commission (USSC) data related to
Smart on Crime, broken down by USAO district. In accordance with USSC and
Department policy, the OIG requested this data via the Criminal Division, which
functions as the Department’s official representative to the USSC. However,
breaking down the USSC data by USAO district was a project that the Criminal
Division said it was not in a position to undertake.
The OIG then requested and timely received this same district-level USSC
data from the Bureau of Justice Statistics (BJS), also broken down by fiscal year
from FY 2010 through FY 2015. The data the BJS provided used methodology
identical to the USSC’s methodology for producing its online interactive sourcebook
data tables and was based on raw data that the USSC publishes annually in the
form of large and highly complex data files that are made available online for
researchers and the general public. Specifically, the BJS provided to the OIG
district-level breakdowns by fiscal year of the information contained in Drug Data
Tables 33 and 37 – 45 of the USSC’s online interactive sourcebook, covering:


Primary Drug Type of Offenders Sentenced Under Drug Guidelines;



Criminal History Category of Drug Offenders in Each Drug Type;



Plea and Trial Rates of Drug Offenders in Each Drug Type;



Weapon Involvement of Drug Offenders in Each Drug Type;



Role Adjustment of Drug Offenders in Each Drug Type;



Acceptance of Responsibility of Drug Offenders in Each Drug Type;
36


	 Drug Offenders Receiving Mandatory Minimums in Each Drug Type;
	 Drug Offenders Receiving Safety Valve and Mandatory Minimums in Each
Drug Type;
	 Drug Offenders Receiving Safety Valve in Each Drug Type; and
	 Sentences Relative to the Guideline Range for Drug Offenders in Each Drug
Type (12 categories).
The OIG used these district-level data tables as the basis for our own further
analysis of national, regional, and district-level trends. Of particular note in the
USSC and BJS methodologies for analyzing the data was that (1) simple possession
cases were included, (2) mandatory minimums of less than 5 years were included
with the “no mandatory minimum” sentences, and (3) mandatory minimums of
greater than 10 years were included with the 10-year mandatory minimum
sentences. In some tables, including the table on the primary drug type of
offenders sentenced under drug guidelines, the data excluded cases missing
complete guideline information.
To perform a regional analysis of the data, the OIG assigned the districts in
various states to approximate regions of the country as follows:
	 California (non-border): California Central, Eastern, and Northern 

Districts;

	 East: Connecticut, Delaware, District of Columbia, Maine, Maryland,
Massachusetts, New Hampshire, New Jersey, New York, Pennsylvania, Rhode
Island, and Vermont;
	 Florida: Middle, Northern, and Southern Districts of Florida;
	 Hawaii and Island Territories: Guam, Hawaii, the Northern Mariana
Islands, Puerto Rico, and the Virgin Islands;
	 Midwest: Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri,
Nebraska, North Dakota, Ohio, Oklahoma, South Dakota, and Wisconsin;
	 Pacific Northwest: Alaska, Oregon, and Washington;
	 South: Alabama, Arkansas, Georgia, Kentucky, Louisiana, Mississippi, North
Carolina, South Carolina, Tennessee, Virginia, and West Virginia;
	 Southwest Border: Arizona, California Southern, New Mexico, Texas
Southern, and Texas Western; and
	 West: Colorado, Idaho, Montana, Nevada, Texas Eastern, Texas Northern,
Utah, and Wyoming.
Additionally, the OIG used the USSC’s publicly available raw data files to
perform an analysis on federal drug offenders with two criminal history points to
37


determine trends in mandatory minimums for these offenders, as well as an
analysis of trends in base offense levels, generally following a methodology similar
to the BJS’s methodology for reproducing the USSC’s online interactive sourcebook
data tables.
Finally, the OIG requested and received from the Executive Office for United
States Attorneys (EOUSA) data on the number of AUSA positions allocated, vacant
positions, and AUSAs on board each fiscal year from 2010 through 2015. We used
this EOUSA staffing data to examine national and district-level USAO staffing trends
and to further analyze the USSC data for trends according to district sizes, as
determined by the number of AUSA positions, as well as the degree of AUSA
understaffing in districts and regions, though we were unable to draw any
correlations between district staffing and charging practices. Additionally, the OIG
requested and received from EOUSA data on the total numbers of sentences for
drug offenses as well as total numbers of sentences for other broad categories of
offenses, including immigration, white collar, violent crime, and all criminal
offenses, for FY 2013 through FY 2015. The OIG analyzed this data to examine
whether significant changes occurred in the proportions of these broad categories of
offenses over the period.
Survey
The OIG surveyed Criminal Chiefs, supervisory attorneys, and line attorneys
in all 94 USAO districts on their practices and views with regard to Smart on Crime.
The survey contained 32 multiple choice and open-ended questions regarding how
each district’s prosecution priorities and charging policies were affected by Smart
on Crime and the issuance of the 2013 Holder memoranda. See Appendix 2 for a
summary of the survey questions and results and Appendices 4 and 5 for the 2013
Holder memoranda.
Document Analysis
We requested and reviewed documentation related to Smart on Crime from
the Office of the Deputy Attorney General, the Criminal Division, EOUSA, the
Federal Bureau of Prisons, the Federal Bureau of Investigation (FBI), and the Drug
Enforcement Administration (DEA). The documents we reviewed included
memoranda, policies, reports, meeting agendas, presentation materials, budget
documents, strategic objective review documents prepared for the Office of
Management and Budget, the results of an earlier EOUSA survey of the districts,
case management system user guides, and emails. The OIG also received copies of
emails summarizing the USSC’s research on recidivist enhancements via the
Criminal Division.
Additionally, we requested and reviewed information and documentation
from each district on its local prosecution guidelines and policies regarding charging
mandatory minimums and recidivist enhancements. On February 29, 2016, the
OIG made an initial request to Criminal Chiefs of the 94 USAOs for “any internal
prosecution guidelines, prosecution thresholds, or other related policies that
respond to the Attorney General’s directive, found in the attached August 12, 2013,

38


memorandum entitled Federal Prosecution Priorities,” that all U.S. Attorneys
“develop — or update existing — district-specific guidelines for determining when
federal prosecutions should be brought and in what priority areas.” The OIG also
requested “any internal office policies or checklists that relate to (1) your office’s
charging practices with respect to mandatory minimum sentences for drug
offenses, and (2) your office’s filing or utilization of recidivist enhancements under
21 USC 851.” In order to obtain a response from all the districts, we made
additional requests on March 23 and June 24, 2016.
Interviews
We interviewed staff from the Office of the Deputy Attorney General, the
Office of Policy and Legislation, the Office of Legal Policy, EOUSA, the USSC, and
the BJS. We also interviewed 14 U.S. Attorneys and 3 Criminal Chiefs in the
District of Columbia, Maryland, and the Eastern District of Virginia, as well as
30 Special Agents at field offices and headquarters of the FBI and the DEA. The
OIG selected these interviewees based on a preliminary data analysis that showed
particularly high or low rates of mandatory minimum sentences.

39


APPENDIX 2
SUMMARY OF SURVEY RESULTS
Overall Number of Respondents
Positions

Number of Respondents

Line Attorney/Line Assistant

1,064 (68.6%)

Supervisor

293 (18.9%)

Criminal Chief

79 (5.1%)

Other

114 (7.4%)

Grand Total

1,550

Are you familiar with the Smart on Crime initiative and issuance of the Attorney General’s
related August 2013 memorandum?
Position
No
Not sure
Yes
Grand Total
23
40
Line Attorney/Line Assistant
1,001 (94.1%)
1,064
(2.2%)
(3.8%)
Supervisor

-

-

293 (100%)

293

Criminal Chief

-

-

79 (100%)

79

23

40

1,373

1,436

Grand Total

Did you receive any training regarding the principles underlying the Smart on Crime
initiative?
Regularly Handle Drug or Violent Crime Cases
Positions
Criminal Chief
Received training
Not sure
Line Attorney/Line
Assistant
Received training
Did not receive training
Supervisor
Received training
Did not receive training
Grand Total

Yes

Not Sure

76 (96.2%)

No

Grand Total

-

3 (3.8%)

79

75 (98.7%)

-

3 (100%)

78

1 (1.3%)

-

-

1

740 (69.5%)

7 (0.7%)

317 (29.8%)

1,064

664 (89.7%)

7 (100%)

249 (78.5%)

920

76 (10.3%)

-

68 (21.5%)

144

219 (74.7%)

-

74 (25.3%)

293

210 (95.9%)

-

64 (86.5%)

274

9 (4.1)

-

10 (13.5%)

19

1,035

7

394

1,436

40


The August 2013 Smart on Crime memorandum directed USAOs to develop or update
existing district-specific guidelines or action plans to focus on the most serious cases. Has
your district done so?
Criminal
Line
Response
Supervisor
Grand Total
Chief
Assistant
(No response)
2 (0.3)
2 (0.9%)
4
No, neither formally nor informally
Not sure

-

15 (2.0%)

1 (0.5%)

16

3 (3.9%)

195 (26.4%)

16 (7.3%)

214

-

4 (0.5%)

1 (0.5%)

5

9 (11.8%)

108 (14.6%)

37 (16.9%)

154

64 (84.2%)

416 (56.2%)

162 (73.9%)

642

76

740

219

1,035

Not yet, but development of a
formal policy is in progress.
Yes, informally, but no written
policy document has been issued or
distributed.
Yes, my district has developed or
updated formal written guidelines.
Grand Total

Has your district worked with its law enforcement partners in developing or revising the
district’s guidelines in light of Smart on Crime?
Response
Criminal Chief
Supervisor
Grand Total
Yes
Not sure
No
(No response)
Grand Total

58 (76.3%)

142 (64.8%)

200

6 (7.9%)

56 (25.6%)

62

11 (14.5%)

17 (7.8%)

28

1 (1.3%)

4 (1.8%)

5

76

219

295

How has the implementation of your district’s guidelines since Smart on Crime affected
your office’s prosecution priorities?
Criminal
Line
Supervisor
Grand Total
Response
Chief
Assistant
Our district’s prosecution priorities
have changed since implementation
of the new guidelines.

27 (35.5%)

273 (36.9%)

82 (37.4%)

382

41 (53.9%)

223 (30.1%)

105 (47.9%)

369

2 (2.6%)

192 (25.9%)

9 (4.1%)

203

4 (5.3%)

29 (3.9%)

13 (5.9%)

46

2 (2.6%)

15 (2.0%)

4 (1.8%)

21

(No response)

-

1 (0.1%)

2 (0.9%)

3

Not applicable (e.g., our district has
not issued new guidelines)

-

7 (0.9%)

4 (1.8%)

11

76

740

219

1,035

The new guidelines have not
affected our office’s prosecution
priorities — we were already
following similar principles prior to
implementation.
Not sure
Other (please specify in the space
below)
The new guidelines have not
affected prosecution priorities — we
were not following these principles
beforehand and still are not.

Grand Total

41


Approximately how many of your attorneys’ cases include defendants who are low-level,
non-violent drug offenders who do not have significant criminal histories as defined in the
Smart on Crime initiative (fewer than three criminal history points)?
Response
Criminal Chief
Line Assistant
Supervisor
Grand Total
All or almost all
12 (1.6%)
2 (0.9%)
14
Almost none

37 (48.7%)

324 (43.8%)

125 (57.1%)

486

Half or more

3 (3.9%)

62 (8.4%)

9 (4.1%)

74

Less than half

17 (22.4%)

200 (27.0%)

35 (16.0%)

252

None

15 (19.7%)

123 (16.6%)

44 (20.1%)

182

4 (5.3%)

19 (2.6%)

4 (1.8%)

27

76

740

219

1,035

Not sure
Grand Total

Have the new Smart on Crime criteria affected your charging decisions with regard to such
defendants?
Criminal
Line
Supervisor
Grand Total
Response
Chief
Assistant
(No response/did not receive
16 (21.1%)
127 (17.2%)
45 (20.5%)
188
question)
No, I handle few cases involving
such defendants.

4 (5.3%)

102 (13.8%)

27 (12.3%)

133

14 (18.4%)

113 (15.3%)

34 (15.5%)

161

No, in the past I have generally
recommended charging drug
quantities that trigger mandatory
minimums whenever available and
continue to do so.

-

33 (4.5%)

2 (0.9%)

35

No, my attorneys handle few cases
involving such defendants.

1 (1.3%)

-

4 (1.8%)

5

Not applicable

-

28 (3.8%)

2 (0.9%)

30

Not sure

-

30 (4.1%)

3 (1.4%)

33

7 (9.2%)

31 (4.2%)

8 (3.7%)

46

34 (44.7%)

276 (37.3%)

94 (42.9%)

404

76

740

219

1,035

No, I was already generally
approving/recommending not to
charge mandatory minimumtriggering drug quantities against
such defendants and continue to do
so.

Other (please specify in the space
below)
Yes, I am now less likely to
approve/recommend charging drug
quantities that trigger mandatory
minimums against such defendants.
Grand Total

42


Have the new Smart on Crime criteria affected your decisions to refer such defendants to
state courts for prosecution rather than charging them federally?
Criminal
Line
Supervisor
Grand Total
Responses
Chief
Assistant
(No response/did not receive
15 (19.7%)
125 (16.9%)
45 (20.5%)
185
question)
No, I continue to charge such
7 (9.2%)
121 (16.4%)
18 (8.2%)
146
defendants federally.
No, my attorneys/I handle few
cases involving such defendants.

8 (10.5%)

144 (19.5%)

37 (16.9%)

189

23 (30.3%)

103 (13.9%)

59 (26.9%)

185

Not applicable

1 (1.3%)

45 (6.1%)

3 (1.4%)

49

Not sure

1 (1.3%)

50 (6.8%)

2 (0.9%)

53

6 (7.9%)

32 (4.3%)

12 (5.5%)

50

15 (19.7%)

120 (16.2%)

43 (19.6%)

178

76

740

219

1,035

No, I was already generally sending
such defendants to state courts.

Other (please specify in the space
below)
Yes, I am now more likely to refer
such defendants for state
prosecution.
Grand Total

Please choose the analysis that best describes how your attorneys determine whether a
defendant meets the Smart on Crime criteria for declining to charge drug quantities that
would trigger mandatory minimums. (Please check all that apply.)67
Criminal
Line
Supervisor
Grand Total
Responses
Chief
Assistant
Attorneys are required to review
the Smart on Crime memorandum
20 (15.7%)
156 (14.7%)
50 (15.2%)
226
at intake or when drafting charges.
Attorneys refer to the specific
Smart on Crime criteria that are
24 (18.9%)
234 (22.1%)
84 (25.5%)
342
incorporated in our district
prosecution guidelines.
Attorneys fill out a worksheet,
checklist, or charge sheet to
25 (19.7%)
176 (16.6%)
57 (17.3%)
258
document the basis for charging
quantities that trigger mandatory
minimums.
Attorneys set forth the basis for
charging quantities that trigger
mandatory minimums on a pre50 (39.4%)
373 (35.2%)
119 (36.2%)
542
indictment prosecution
memorandum.
Other
8 (6.3%)
28 (2.6%)
8 (2.4%)
44
Attorneys in our district do not
20 (1.9%)
2 (0.6%)
22
undertake this kind of analysis.
Not sure
42 (3.9%)
3 (0.9%)
45
Not applicable
Grand Total

-

31 (2.9%)

6 (1.8%)

37

127

1,060

329

1,516

For questions with “check all that apply,” the grand total will be the number of responses,
not the number of people who responded.
67

43

Since the implementation of Smart on Crime, are there more, fewer, or about the same
number of cases in your office in which prosecutors decline to charge drug quantities that
would trigger mandatory minimums?
Criminal
Supervisor
Grand Total
Responses
Chief
About the same number of cases

14 (18.4%)

37 (16.9%)

51

Fewer cases
More cases with mandatory minimum-triggering
drug quantities NOT charged
Not applicable
Not sure
(No response/did not receive question)

13 (17.1%)

17 (7.8%)

30

31 (40.8%)

102 (46.6%)

133

3 (3.9%)
15 (19.7%)

2 (0.9%)
16 (7.3%)
45 (20.5%)

2
19
60

76

219

295

Grand Total

At what point do your attorneys typically file a 21 U.S.C. § 851 information?
Criminal
Line
Supervisor
Responses
Chief
Assistant
(No response/did not receive
7 (9.2%)
356 (48.1%)
55 (25.1%)
question)
At indictment or soon after
35 (46.1%)
168 (22.7%)
97 (44.3%)
During plea negotiations

Grand Total
418
300

-

7 (0.9%)

-

7

3 (3.9%)

34 (4.6%)

6 (2.7%)

43

27 (35.5%)

162 (21.9%)

56 (25.6%)

245

Not applicable

1 (1.3%)

1 (0.1)

-

2

Not sure

1 (1.3%)

7 (0.9%)

2 (0.9%)

10

Other (please specify)

2 (2.6%)

5 (0.7%)

3 (1.4%)

10

76

740

219

1,035

Immediately prior to trial
It depends on the circumstances of
the case/investigation at hand.

Grand Total

Is this a change from your attorneys’ practice prior to the implementation of Smart on
Crime?
Criminal
Line
Supervisor
Grand Total
Responses
Chief
Assistant
(No response/did not receive
7 (9.2%)
358 (48.4%)
56 (25.6%)
421
question)
No, I followed the same practice
26 (34.2%)
238 (32.2%)
87 (39.7%)
351
prior to Smart on Crime.
Yes, I used to file a 21 U.S.C. § 851
18 (23.7%)
56 (7.6%)
32 (14.6%)
106
information later in the process.
Not applicable
3 (3.9%)
40 (5.4%)
6 (2.7%)
49
Not sure
Yes, I used to file a 21 U.S.C. § 851
information earlier in the process.
No, my attorneys followed the same
practice prior to Smart on Crime.
Yes, my attorneys used to file a
21 U.S.C. § 851 information later
in the process.
Yes, my attorneys used to file a
21 U.S.C. § 851 information earlier
in the process.
Grand Total

4 (5.3%)

30 (4.1%)

11 (5.0%)

45

5 (6.6%)

18 (2.4%)

6 (2.7%)

29

8 (10.5%)

-

13 (5.9%)

21

3 (3.9%)

-

7 (3.2%)

10

2 (2.6%)

-

1 (0.5%)

3

76

740

219

1,035

44


Have the new Smart on Crime criteria related to 21 U.S.C. § 851 recidivist enhancements
affected your charging decisions?
Criminal
Line
Supervisor
Grand Total
Responses
Chief
Assistant
Yes, I am now less likely to
file/approve filing a 21 U.S.C. § 851
48 (63.2%)
267 (36.1%)
113 (51.6%)
428
recidivist enhancement in cases
where the defendant is eligible.
No, charging decisions regarding
21 U.S.C. § 851 recidivist
enhancements have not changed —
1 (1.3%)
48 (6.5%)
13 (5.9%)
62
absent extraordinary or special
circumstances, my attorneys file
them if the defendant is eligible.
No, charging decisions regarding
21 U.S.C. § 851 recidivist
7 (9.2%)
enhancements have not changed —
121 (16.4)
43 (19.6%)
171
I (or my attorneys) rarely file them
when the defendant is eligible.
No, my district typically has few
defendants eligible for 21 U.S.C.
3 (3.9%)
8 (1.1%)
6 (2.7%)
17
§ 851 enhancements.
Other
13 (17.1%)
41 (5.5%)
19 (8.7%)
73
Not sure
2 (2.6%)
72 (9.7%)
4 (1.8%)
78
Not applicable
1 (1.3%)
39 (5.3%)
2 (0.9%)
42
(No response/did not receive
1 (1.3%)
144 (19.5%)
19 (8.7%)
164
question)
Grand Total
76
740
219
1,035
Please choose the analysis that best describes how your attorneys determine whether a
defendant meets the Smart on Crime criteria for § 851 recidivist enhancements. (Please
check all that apply.)
Criminal
Line
Supervisor
Grand Total
Responses
Chief
Assistant
Attorneys are required to review
the Smart on Crime memorandum
19 (12.7%)
157 (15.2%)
61 (16.3%)
237
at intake or when drafting charges.
Attorneys refer to the specific
Smart on Crime criteria that are
27 (18.0%)
216 (20.9%)
86 (22.9%)
329
incorporated in our district
prosecution guidelines.
Attorneys fill out a worksheet,
checklist, or charge sheet to
35 (23.3%)
175 (16.9%)
66 (17.6)
276
document the basis for filing
21 U.S.C. § 851 enhancements.
Attorneys set forth the basis for
filing a 21 U.S.C. § 851
55 (36.7%)
353 (34.1%)
132 (35.2%)
540
enhancement on a pre-indictment
prosecution memorandum.
Other (please specify in the space
14 (9.3%)
40 (3.9%)
20 (5.3%)
74
below)
Attorneys in our district do not
17 (1.6%)
3 (0.8%)
20
undertake this kind of analysis.
Not sure
53 (5.1%)
3 (0.8%)
56
Not applicable
24 (2.3%)
4 (1.1%)
28
Grand Total
150
1,034
375
1,560

45


Since Smart on Crime, are there more, fewer, or about the same number of cases in your
office where prosecutors decline to file 21 U.S.C. § 851 informations?
Row Labels
Criminal Chief
Supervisor
Grand Total
More cases

36 (47.4%)

94 (42.9%)

130

About the same number

15 (19.7%)

52 (23.7%)

67

Fewer cases

17 (22.4%)

25 (11.4%)

42

6 (7.9%)

27 (12.3%)

33

-

1 (0.5%)

1

2 (2.6%)

20 (9.1%)

22

76

219

295

Not sure
Not applicable
(Did not answer)
Grand Total

Does your district have a supervisory review process in place to ensure that prosecutors are
considering Smart on Crime principles when making charging decisions? (Please check all
that apply.)
Criminal
Line
Supervisor
Grand Total
Responses
Chief
Assistant
Yes, supervisors must sign off on
prosecutors’ written explanations
for charging mandatory
61 (46.2%)
minimum-triggering drug
502 (43.4%)
171 (45.8%)
734
quantities to ensure that they are
consistent with the principles of
Smart on Crime.
Yes, supervisors must sign off on
prosecutors’ written explanations
for charging 21 U.S.C. § 851
59 (44.7%)
recidivist enhancements to
403 (34.9%)
153 (41.0%)
615
ensure that they are consistent
with the principles of Smart on
Crime.
Yes, our district uses another
method of supervisory review to
10 (7.6%)
ensure compliance with Smart on
122 (10.6%)
43 (11.5%)
175
Crime (please specify in the
space below).
No, there is no supervisory
review process in place to ensure
2 (1.5%)
25 (2.2%)
1 (0.3%)
28
compliance with the principles of
Smart on Crime.
Not sure
104 (8.9%)
5 (1.3%)
109
Grand Total

132

46


1,156

373

1,661

APPENDIX 3

HOLDER MEMORANDUM, DEPARTMENT POLICY ON CHARGING
AND SENTENCING, MAY 19, 2010

Q)fficl' 11f till' ~thlft\l'l1 (!)l'tll'nd
IDns l!iulltlll1 . ID. <i. ~O~3t1

May 19.2010

MEMORANDUM TO ALL FEDERAI.I'ROSECUTORS
From:

A .,..Eric II. Holder, Jr.

~ Anomey General

Subject

Depanmem I'olic y on Charg ing and Sentencing

The reasoned cxcreise of prosccutorial di scretion is esscmial to the fair, clTeetive. and
even· hand ed administration of the federal crimina l laws. lki;isions about ,~hcther 10 init iate
charges. what charges and enhancements to pursue. when to accept a negotiated pIca. and how 10
advocate m scnteneing. are among the most fundamental duties of federal pro'SCeutors. For
nearly three dcrades. the Principles of Federdl I'rosecution. as reflected in Title 9 of the U.S.
Al1orneys' Manual, Chapter 27. have guidt-d f~-deral proseculors in the discharge oflhese duties
in panieular and in thei r responsibility to sec k jusliee in the enforcement of Ihe federal criminal
laws in ge neral. The purposc of this memOF,lndum is to reamnn the guidance provided by thosc
Principles.
]'crsons who commit similar crimes and have similar culpability should, to the eS len!
possible, be Ireated similarly _ Unwarranted dispari ties may result from disregard for this
fundamental principle. They can aloo result, hov.'Cvcr, from a failure to analyze careful ly and
distinguish thc specific faelS and circumstances of each particular case. Indeed, equal justice
depends on individualizcdjusticc. and sman law enforeement demands it. Acco rdingly.
decisions regarding charging. pica agrcemenlS, and advocacy al sentencing must be made on the
merits of each case, taking into account an individuali~.cd assessment of the defendant's conduct
and criminal history and the circumstances rclating to commission of the offense (includin g the
impaci of thc crimc on victims). lhe needs of the communities we serve. and f~-deral rcsourc(.>s
and priori ties. Prosecutors must always be mindful of our duty to ensun.- Ihat th(.>se decisions an.­
madc without unwarranted consider~tion of such faetor1i as raec. gender. cthn icity, or sex ual
orientation.
Charl! ln~ DeclSlons Charging decisions should be informed by reason and by Ihe
gcneral purposes of criminal law enforcement: punishment. public ""fety. detcrrence, and
rchabilitation. Thtse decisions shou ld also renecl thc priorities of the Dcpanment and of each
district. C harges should ord inaril y be brought iflherc is probable cause to believe that a person
has comrnillcd a federal offense and there is suffic ient admissiblc evidcnce to obtain and sustain
p conviction, unless"TJ(I substantial fedcral interest" would be served, the person is subject to

47


2
"effective prosecution" elsewhere. or there is "an adequate non-criminal alternative to
prosecution" [U SAM 9-27.200 el seq.].
Moreover, in accordance wilh long-standing principle, a federal prosecutor should
ordinarily charge "the most serious offense that is consistent with the nature of the defendant's
conduct, and that is li kely to result in a sustainable conviction" [USAM 9-27.300]. This
detetmination, however, must always be made in the contex t of " al1 ind ividualized assessment of
the extent to wh ich particular charges fit the specific circumstances urlhe case, are consistent
with the purpose of the Federal crimin<.ll code, and maximize the impact of Federal resources on
crime" IUSAM 9-27.100J. In all cases, the charges should fairly represent the defendant's
crim ina! conduct, and due consideration should be given to the defendant's substantial assistance
in an investigation or prosecution. As a general matter, the decision whether to seek a statutory
sentencing enhancement should be guided by these same principles.
All charging decisions must be reviewed by a supervisory attorney. All but the most
routine indictments should be accompanied by a prosccution memorandum that idcntifies thc
chargi ng options supported by the evidence and the law and explai ns the charging decision
therein. Each office shall promulgate written guidance describing its internal indictment review
1
process.
Plea Agreements: Plea agreements should re nect the totality of a defendant's conduct.
Thest: agreements are govt:med by the same fundamcntal principle a~ charging decisions:
prosecutors should seek a plea to the most serious off ense that is consistent with the nature of thc
defendant's conduct and likely to result in a sustainable conviction, infornled by an
individualized assessment of the specific facts and circumstances of each particular case.
Charges should not be filed simpl y to exert leverage to induce a plea, nor should charges be
abandoned to arrive at a plea bargairi that docs not re neel the seriousness of the defendant's
conduct. All pica agreements should be consistent with the Prillciples of Federal Prosecution
and must be reviewed by a supervisory attorney. Each office shall promulgate written guid<lllce
regarding the standard elements required in ils plea agreements, including the waivers ora
dcfendant's rights.
.
Advocacy at Sentencmg: As the Supreme Court has recognized, Congress has identified
lhe [actors for courts to consider when imposing sentences pursuant to 18 U.S.C. §3553.
Consistent with the statute and with the advisory sentencing guidelines as the
touchstone, prosecutors should seek sentences that rcflect lhe seriousness of the offense, promote
respect for the law, provide just punishment, afford deterrence, protect the public, and offer
defendants an opportunity for effective rehabili tation. [n the typical case, the appropriate
balance among these purposes will continue to be renecteu by the applicable guidelines range,
and prosecutors should generall y continue to advocate for a sentence w ithin that range. 111e
advisory guidelines remai n important in further ing the gO:11 of national uniformity throughout the
.federal system. But consistent with the Principles of Federal Prosecution and given the advisory
I This memorandum has no impaet on the guidance provided in the September 22. 2003 memorandum and
elsewhere regarding "fasl-1raek" programs. Inlhose districts where an approved "fasH rack" program has becn
established, ~h'lfgmg decisions and disposition of charges mu51 comply with the Ikpllrtment's requirements for that
program.

48


J
nature of the guidelines, advocacy at sentencing- like charging decisions and pica agrcements­
mu~l

also follow from an individualized assessment of the facts and circumstances of each
particular case. All prosccutorial requests fo r departures or variances- \lpward or downward­

must be based upon specific and artlculablc factors, and require supervisory approval. Each
otticc shall provide trai ning fo r effective advocacy al sentencing.
With respect to charging decisions, plea agreemen ts, and advocacy at sen tencing, the
mechanisms established for obtain ing supervisory approval should be designed to ensure, as
much as possi ble, adherence to the Principles o f Federal Prosec ution and the guidance provided
by this memorandu m, as well as district-wide consistency. Supervisory attorneys selected to
review exercises of discretion should be skilled, ex perienced, and thoroughly tiltn iliar with
Department and district-spe(.;i fi c policies, priorities, and practices . All guidance described above
must be shared wi th the Executive Office for U.S. Attorneys upon promulgation.
This memorand um supersedes previous Department guidance on charging and sentencing
including the September 22, 2003 memorandum issued by Attorney General John Ashcroft
("Department Po licy Concerning Charging Criminal Offenses, Disposition of C harges, and
Sentencing"), the July 2, 2004 memorandum issued by Deputy Attorney General James Corney
(" Department Legal Positions and Policies in Light of Blakely v. Washingwn"), and the January
28,2005 memorandum issued by Deputy Attorney General James Corney ("Department Policies
and Procedures Concerning Sentenc ing").

49


APPENDIX 4

HOLDER MEMORANDUM, FEDERAL PROSECUTION PRIORITIES,
AUGUST 12, 2013 


QOOjce of t~e %lttomep ~eneta{
Wa~bi ltgtolt , lD .~.

20530

August 12, 2013

MEMORANDUM TO HEADS OF DEPARTMENT OF JUSTICE COMPONENTS
AND UNITED STATES ATTORNEYS
FROM:

THE ATTORNEY GENERAL

SUBJECT:

Federal Prosecution Priorities

C-

We must always endeavor to ensure we use our limited resources in a manner that is
consistent with our responsibility to effectively enforce the criminal law. With the curreni
strains on Department of Justice resources, we must maximize efrorts to prosecute the right
criminal cases consistent with our mission. OUT reaction to these budget challenges should not
be driven by statistics alone. Rather, we must set quality, evidence· based priorities for the types
of cases we bring with an eye toward promoting public safety, deterrence, and fairness. This
necessarily will mean focusing our resources on fewer but the most significant cases.

In order to initiate charges against a defendant, prosecutors must detennine not only that
the person's conduct constitutes a federal offense and admissible evidence is sufficient to obtain
and sustain a conviction. but also that the prosecution serves a substantial federal interest, the
person is not subject to effective prosecution elsewhere. and there is no adequate non~criminal
alternative to prosecution. See Attorney General Ho lder Memorandum (May 19, 2010): USAM
9-27.220.
Of these criteria that must be satisfied before bringing a case, it is ofprirnary importance
to assess whether a prosecution serves a substantial federal interest. For this reason. 1 am asking
all United States Attorneys, in conjunction with their law enforcement partners. to develop - or
update existing - district-specific guidelines for detennining when federal prosecutions should
be brought and in what priority areas.
Your local guidelines should define what cases serve a district's substantial federal
interests and should be informed by a number of factors, including both national and local law
enforcement priorities. Nationally. the Department's priorities are (1) protecting Americans
from national security threats; (2) protecting Americans from violent crime; (3) protecting
Americans from financial fraud; and (4) protecting the most vulnerable members of our society.
Locally, a particular district's priorities will often depend on local criminal threats and needs,
and each United States Attorney is in the best position to clearly articulate his or her priorities.
The United States Attorney' s Manual already contains specific guidance to prosecutors
on determining whether a prosecution would serve a substantial federal interest. See USAM 927.230. While designed to help prosecutors analyze individual cases, the USAM also wiU

50


Memorandum to Heads of Department of Justice Components
and United States Attorneys

Page 2

inform United States Attorneys in shaping their district-specific criteria. In addition, the below
list of factors relevant to assessing the federal interest in different types of criminal cases should
supplement the USAM 1 s excellent guidance and further inform your efforts.
Finally, it is important that federal law enforcement agency priorities in each district align
with our prosecution goals. For this reason, your evaluations should be conducted jointly with
federal law enforcement leaders and in consultation with state, local, and tribal law enforcement
as appropriate.
I am confident that these thoughtful and reasoned assessments will better advance our
shared law enforcement mission.
Factors to Consider when Developing District Investigative and Prosecution Priorities

Explicit Federal Priorities
•

Does the matter fall within the Department's four law enforcement priorities?

•

lfthe offense charged does not fall within the Attorney General 's stated priorities,
does the defendant's ancillary or underlying conduct nevertheless implicate the
Department's stated priorities?

•

Is the matter part of a program or priority for which Congress, the Department of
Justice, or another federal agency has provided specific funding?

•

Would accepting the case potentially lead to another case implicating more
significant Department or Government-wide priorities?

Primary or Exclusive Jurisdiction
•

Did the criminal conduct occur on a federal enclave, such as a national park or a
military base where the U.S. Attorney' s office bas exc1usivejurisdiction?

•

Does the primary criminal conduct to be addressed involve statutory schemes, such as
immigration laws, that are enforced exclusively by the federal government?

•

Does the crime have international, interstate, or multi-district impact?

•

Is a federal agency or employee a victim oftbe crime?

Effective Alternative!,·
•

Is the defendant currently serving a state or federal sentence of imprisonment for
other criminal conduct, and if so, is the defendant unlikely to be released soon?

51


Memorandum to Heads of Department of Justice Components
and United States Attorneys

Page 3

•

Does the state or local government have a statute and sentencing scheme that
adequately addresses the criminal conduct?

•

If the criminal conduct occurred in Indian Country, will tribal statutes and available
sentences adequately address the criminal conduct?

•

Is the matter a "complex" case requiring resources or evidence collection capabilities
beyond the scope of local abilities?

•

Would taking the case federally be a particularly effective mechanism for removing a
violent person or repeat offender from society?

•

When a case involves a matter that, pursuant to the above cri teri~ would be more
appropriately brought by the state, does the case nevertheless address an
extraordinary public safety or public health concern to which the U.S. Attorney's
Office is particularly able to respond? If so, is the case part of a carefully-designed
local program or initiative coordinated between the United States Attorney and local
law enforcement?

52


APPENDIX 5

HOLDER MEMORANDUM, DEPARTMENT POLICY ON CHARGING 

MANDATORY MINIMUM SENTENCES AND RECIDIVIST 

ENHANCEMENTS, AUGUST 12, 2013 


(i)fficc

,'f tllc l\ttllrllCI;! ijirnrnd

ltJas llington.l!I. (!i. 2115:lI'
August 12,20 13

».lr-

MEMORANDUM TO TIlE UNITED STATES ATTORNEYS AND
ASSISTANT ATTORNEY GENERAL FOR THE CRIMINAL DIVISION
FROM:
SUBJECT:

TI-IE ATTORNEY GENERAL

C

~

Deportment Policy on Charging Mandatory Minimum SentenceS

and Recidivist Enhancements in Certain Drug Cases
In Alleyne v. United Stales, 133 S.C. 2151 (2013). the Supreme Court held that any fact that
increases the statutory mandatory minimum sentence is an element afthe crime that must be
submitted to the jury and found beyond a reasonable doubt This means that for a defendant to be
subject to a mandatory minimum sentence, prosecutors must ensure thai the charging document
includes those elements orthe crime that trigger the statutory minimum penalty.
The Supreme Court's decision in Alleyne heightens the role a prosecutor plays in detennining
whether a defendant is subject to a mandatory minimum sentence. To be sure, the exercise of
discretion over charging decisions has always been an "integral feature of the criminal justice
system." United States v. LaBonte, 520 U.S. 751. 762 (1997). and is among the most important duties
ofa federal prosecutor. Current policy requires prosecutors to conduct an individual ized assessment
of the extent to which charges fit the specific circumstances of me case. are consistent with the
purpose of me federal crim inal code. and maximize the impact of federal resources on crime. When
making these individuali7.ed assessments, prosecutors must take into account nwnerous factors. such
as the defendant's conduct and criminal history and the circumstances relating to the commission of
the offense. the needs of the communities we serve. and federal resources and priorities.' Now that
our charging decisions also affect when a defendant is subject to a mandatory minimum sentence.
prosecutors must evaluate these factors in an equally thoughtful and reasoned manner.
It is with full consideration of these factors that we now refine our charging policy regarding
mandatory minimums for certain nonviolent. low-level drug offenders. We must ensure that our
most severe mandatory minimum penalties are reserved for serious. high-level . or vio lent drug
traffickers. In some cases. mandatory minimum and recidivist enhancement statutes have resulted in
unduly harsh sentences and perceived or actual disparities that do not reflect our Principles of Federal
Prosecution. Long sentences for low-level , non-violent drug offenses do not promote public safety,
detelTence. and rehabilitation. Moreover, rising prison costs have resulted in reduced spending on
criminal justice in itiatives, including spending on law enforcement agents. prosecutors, and
prevention and intervention programs . These reductions in public safety spending require us to make
our public safety expenditures smarter and more productive .

I These factors are set out more fully in my memorandum of May 19, 2010 ("Depamnent Policy on ChaTg,ing and
Sentencing") and Title 9 of the U.s. Attorneys' Manual, Chapter 27 .

53


Memorandum to the United States Attorneys and
Assistant Attorney General for the Criminal Division

Page 2

For all these reasons, t am issuing the following policy:
Continuation of Charging and Sentencing Policies: Pursuant to my memorandum of May
19,2010, prosecutors should continue to conduct .I an individuaJized assessment of me extent to
which particular charges fit the specific circumstances of the case, are consistent with the purpose of
the Federal criminal code. and maximize the impact of Federal resources on crime." While this
means that prosecutors "shou ld ordinarily charge the most serious offense that is consistent with the
nature of the defendant's conduct. and that is likely to result in a sustainable conviction," the charges
always shou ld renect an indiv idual ized assessment and fairly represent the defendant's crimina l
conduct.
Certain Mandatory Minimum Sentencing Statutes Based on Drug Quantity: Prosecutors
should continue to ascertain whether a defendant is eligible for any statutory mandatory minimum
statute or enhancement. However, in cases involving the applicability of Title 21 mandatory
minimum sentences based on drug type and quantity, prosecutors should decline to charge the
quantity necessary to trigger a mandatory minimum sentence if the defendant meets each of the
following criteria:]
The defendant's relevant conduct does not involve the use of violence, the credible threat of
violence. the possession of a weapon, the trafficking of drugs to or with minors. or the death
or serious bodily injury of any person;
The defendant is not an organizer. leader. manager or supervisor of others within a criminal
organization;
The defendant does not have significan t ties to large-scale drug trafficking organizations,
gangs, or cartels; and
The defendant does not have a significant criminal history. A significant criminal history
will normally be evidenced by three or more criminal history points but may involve fewer or
greater depending on the nature of any prior convictions.
Timing and Plea Agreements: If information sufficient to determine that a defendant meets
the above criteria is available at the time initial charges arc filed, prosecutors should decline to
pursue charges triggering a mandatory minimum sentence. However, if this information is not yet
available, prosecutors may file charges involving these mandatory minimum statutes pending further
information and a determination as to whether a defendant meets the above criteria. If the defendant
uhjmately meets the criteria, prosecutors should pursue a disposition that does not require a Title 21
mandatory minimum sentence. For example, a prosecutor could ask the grand jury to supersede the
indictment with charges that do not trigger the mandatory minimum, or a defendant could plead
guilty to a lesser included ofTense, or waive indictment and plead guilty to a superseding information
that does not charge the quantity necessary to trigger the mandatory minimum.

~ The poticy sel forth herein is not intended to create or confer any rights. privileges, or benefits in any matter, case,
or proceeding. See United Stules v. Caceres. 440 U.S. 741 (1979).

As with every case. prosecutors should determine, as B threshold maller, whether a case serves a substantial
federal interest. In some cases, satisfaction of the above criteria meant for low-level, nonviolent drug offenders may
indicate that prosecution would not serve a substantial federal interest and that lile case should not be brough,
federally.

1

54


Memorandum to the United States Attorneys and
Assistant Attorney General for the Criminal Division

Page 3

Advocacy at Sentencing: Prosecutors must be candid with the court, probation, and the
public as to the full extent of the defendant's culpability, including the quantity of drugs involved in
the offense and the quantity attributable to the defendant' s role in the offense. even if the charging
document lacks such specificity. Prosecutors also should continue to accurately calculate the
sentencing range under the United States Sentencing Guidelines. In cases where the properly
calculated guideline range meets or exceeds the mandatory minimum, prosecutors should consider
whether a below·guidelines sentence is sufficient to satisfy the purposes of sentencing as set forth in
18 U.S.C. § 3553(a). In determining the appropriate sentence to recommend to the Court,
prosecutors should consider whether the defendant truthfully and in a timely way provided to the
Government all information the defendant has concerning the offense or offenses that were pan of
the same course of conduct. common scheme, or plan.
Recidivist Enhancements: Prosecutors should decline to file an information pursuant to 21
U.S.c. § 851 unless the defendant is involved in conduct that makes the case appropriate for severe
sanctions. When determin ing whether an enhancement is appropriate. prosecutors should consider
the following factors:
Whether the defendant was an organizer, leader. manager or supervisor of others within a
criminal organization;
Whether the defendant was involved in the use or threat of violence in connection with the
offense;
The nature of the defendant's criminal history, including any prior history of violent conduct
or recent prior convictions for serious offenses;
Whether the defendant has significant ties to
gangs, or cane Is;

large~scale

drug trafficking organizations.

Whether the tiling would create a gross sentencing disparity with equally or more culpable
and

co~defendants ;

Other case-specific aggravating or mitigating factors.
In keeping with current policy. prosecutors are reminded that all charging decisions must be
reviewed by a supervisory attorney to ensure adherence to the Principles of Federal Prosecution. the
guidance provided by my May 19. 2010 memorandum, and the policy outlined in this memorandum

55


APPENDIX 6 


HOLDER MEMORANDUM, GUIDANCE REGARDING 

§ 851 ENHANCEMENTS IN PLEA NEGOTIATIONS 

SEPTEMBER 24, 2014 


@ffirr of t~r Attornrl! ®rnrral
]lJag~ington, ill, <11. 20.530
September 24, 2014

TO:

. C'~EPARTM ENT OF JUSTICE ATTORNEYS

FROM~ THE

RE:

ATTORNEY GENERAL

Guidance Regarding § 851 Enhancements In Plea
Negotiations

The Department of Justice's charging policies arc clear that in all cases,
prosecutors must individually evaluate the unique facts and circumstances and se­
lect charges and seek sentences that arc fair and proportipnai based upon thi s indi ­
vi duali zed assessment. " Department Policy on Charging and Sentencing," May
10, 20 I O. The Department provided more specific gu idance for charging manda,
tory minimums and rec idivist enhancements in drug cases in the August 12, 2013.
" Department Policy on Charging Mandatory Minimum Sentences and Recidivist
Enhancements in Certain Drug Cases." That mem orandum provides that prosecu­
tors should decline to seek an enhancement pursuant to 21 U.S.C. § 851 unless the
"defendant is involved in conduct that makes the case appropriate for severe sanc­
tions," and sets forth factors that prosecutors should consider in making that de­
termination. Whether a defendant is p leading guilty is not one of the factors enu­
merated in the charging policy. Prosecutors are encouraged to make the § 85 1 de­
termination at the time the case is charged, or as soon as possible thereafter. An
§ 85 1 e nhancement should not be used in piea negotiati ons for the sole or predom'
inant purpose of inducing a defendant to plead guilty. Thi s is consistent with
long,standing Department policy that "[c]harges should not be filed s imply to ex,
crt leverage to induce a plea, nor shou ld charges be abandoned to arrive at a pica
bargain that does not reflect the seriollsness of the defendant ' s conduct" " De­
partment Policy on Charg ing and Sentencing," May 19, 2010.
While the fact that a defendant mayor may not exercise hi s right to a jury
trial shou ld ordinarily not govern the determination of whether to file or forego an
§ 85 1 enhancement, certain circumstances -- such as new infonnation about the
de fendant, a reassessment of the strength of the government's case, o r recognition
of cooperation -- may make it appropriate to fore go or dismiss a previously filed
§ 851 information in connection with a guilty plea. A practice of routinely prem­
ising the decision to file an § 851 enhancement solely on whether a defendant is
entering a gui lty plea, however, is inappropriate and inconsistent with the spirit of
the policy.

56


APPENDIX 7 


U.S. SENTENCING COMMISSION FEDERAL SENTENCING 

GUIDELINES MANUAL, § 5C1.2., “SAFETY VALVE” 

§5C1.2.
limitation on Applicability of Statutory Minimum Sentences in Certain Cases
"the Safety Valve"
(a)
Except as provided in subsection (b), in the case of an offense under 21 U.S.C. § 841, §
844, § 846, § 960, or § 963, the court shall impose a sentence in accordance with the applicable
guidelines without regard to any statutory minimum sentence, if the court finds that the
defendant meets the crteria in 18 U.S.C. § 3553(1)(1)-(5) set forth below:
(1)
the defendant does not have more than 1 criminal history point, as determined under the
sentencing guidelines before application of subsection (b) of §4A 1.3 (Departures Based on
Inadequacy of Criminal History Category) ;
(2)
the defendant did not use violence or credible threats of violence or possess a firearm or
other dangerous weapon (or induce another participant to do so) in connection with the offense;
the offense did not result in death or serious bodily injury to any person;

(3)

(4)
the defendant was not an organizer, leader, manager, or supervisor of others in the
offense, as determined under the sentencing guidelines and was not engaged in a continuing
criminal enterprise, as defined in 21 U.S .C. § 848; and
(5)
not later than the time of the sentencing hearing, the defendant has truthfully provided to
the Government all information and evidence the defendant has concerning the offense or
offenses that were part of the same course of conduct or of a common scheme or plan , but the
fact that the defendant has no relevant or useful other information to provide or that the
Government is already aware of the information shall not preclude a determination by the court
that the defendant has complied with this requirement.
(b)
In the case of a defendant (1) who meets the criteria set forth in subsection (a) ; and (2)
for w hom the statutorily required minimum sentence is at least five years, the offense level
applicable from Chapters Two (Offense Conduct) and Three (Adjustments) shall be not less
than level 17.

Commentary
Application Notes:
1.
wMore than 1 criminal history point, as determined under the sentencing guidelines, ~ as
used in subsection (a)(1) , means more than one criminal history point as determined under
§4A 1. 1 (Criminal History Category) before application of subsection (b) of §4A 1. 3 (Departures
Based on Inadequacy of Criminal History Category).
2.
ffDangerous weaponff and fffirearm, ff as used in subsection (a)(2), and ffserious bodily
injury, as used in subsection (a)(3), are defined in the Commentary to §1B 1. 1 (A pplication
Instructions).
U

3.
ffOffense, H as used in subsection (8)(2)-(4), and Noffense or offenses that were part of the
same course of conduct or of a common scheme or plan, ~ as used in subsection (a)(5), mean
the offense of conviction and all relevant conduct.

57


4.
Consistent with §1B1.3 (Relevant Conduct) , the! term udefendant, as used in subsection
(a)(2), limits the accountability of the defendant to his own conduct and conduct that he aided or
abetted, counseled, commanded, induced, procured, or willfully caused.
U

5.
"Organizer, leader, manager, or supervisor of oihers in the offense, as determined under
the sentencing guidelines, " as used in subsection (a)( 4), means a defendant who receives an
adjustment for an aggravating role under §3B1. 1 (Ag9ravating Role).
6.
''Engaged in a continuing criminal enterprise, ~ as used in subsection (a)(4) , is defined in
21 U.S.C. § 848(c). As a practical maffer, it should not be necessary to apply this prong of
subsection (a)(4) because (i) this section does not apply to a conviction under 21 US.C. § 848,
and (ii) any defendant who Uengaged in a continuing criminal enterprise" but is convicted of an
offense to which this section applies will be an uorgaoizer, leader, manager, or supervisor of
others in the offense. "

7.
Information disclosed by the defendant with respect to subsection (a)(5) may be
considered in determining the applicable guideline range, except where the use of such
information is restricted under the provisions of §1B1. B (Use of Certain Information). That is ,
subsection (a)(5) does not provide an independent basis for restricting the use of information
disclosed by the defendant.
8.
Under 18 US.c. § 3553(f), prior to its determination, the court shall afford the government
an opportunity to make a recommendation. See also Fed. R. Crim. P. 32(f} , (i) .
9.
A defendant who meets the criteria under this section is exempt from any otherwise
applicable statutory minimum sentence of imprisonmf.mt and statutory minimum term of
supervised release.
Backaround: This section sets forth the relevant pro~'isions of 18 U.S.C. § 3553(f), as added by
section 80001(a) of the Violent Crime Control and Law Enforcement Act of 1994, which limit the
applicability of statutory minimum sentences in certain cases. Under the authority of section
B0001(b) of that Act, the Commission has promulgated application notes to provide guidance in
the application of 18 US.C. § 3553(f). See also H. Rep. No. 460, 103d Cong., 2d Sess. 3
(1994) (expressing intent to foster greater coordination between mandatory minimum
sentencing and the sentencing guideline system).
Hjstorical Note: Effective September 23, 1994 (~ A ppend ix C, amendment 509). Amended effective
November 1, 1995 (~Ap pendix C, amendment 515); Novembe r 1, 1996 (~Appe ndix C, amendment
540), November 1, 1997 (see Appendi x C, amendment 570); November 1, 2001 (see Appendix C,
amendment 624); October 27, 2003 ~ Appendix C, alTlE~nd ment 651); November 1, 2004 (~
Append ix C, amendment 674); November 1, 2009 ~ Appendix C, amendment 736).

58


APPENDIX 8 


THE DEPARTMENT’S RESPONSE TO THE DRAFT REPORT

U.S. Department of Justice

950 Pennsylvania Ave. , N. W.
Washing/an. D.C. 20530

JUN 0 6 2017
MEMORANDUM
TO:

FROM,

Nina S. Pelletier
Assistant Inspector General
Evaluation and Inspections
Office of the Inspector General
J tice

,s
Associate Deputy Attorney General
Office of the Deputy Attorney General

H4!ll'- ,

Monty Wilkinson
Director
Executive Office for United States Attorneys
SUBJECT:

Response to OIG' s Review of the Smart on Crime Initiative,
Assigrunent Nwnber A-20 15-008

The Office ofthe Deputy Attorney General (ODAG) and the Executive Office for United
States Attorneys (EOUSA) appreciate the opportunity to comment on the final draft report
(May 23 , 2017) of the review undertaken by the Office of the Inspector General (OIG) regarding
the Department' s implementation of prosecution and sentencing reform principles under the
Smart on Crime Initiative. We appreciate the effort the OIG undertook to review the in.itiative
and the acknowledgement of the progress that the Department made in implementing policies
under the initiative. The Department remains committed to the principles of prioritizing federal
prosecutions on the most serious threats to public safety and ensuring the fair and impartial
administration of justice. As the draft report notes, however, the Department issued a new
charging and sentencing policy on May 10, 2017, rescinding aspects of the Smart on Crime
Initiative. Nevertheless, the recommendations in the draft report are relevant to the
implementation of the new charging and sentencing policy, and the Department concurs with the
recommendations as fo llows. I

I Recommendations #1 and #2 are directed to the Department. Recommendation #3 is directed
to the Department and EOUSA. This response is submitted on behalf of the Department and
EOUSA.

59


-2Recommendation 1. Ensure tbat the U.S. Attorneys' Manual accurately renects
Department cbarging policies.
The Department concurs that the United States Attorneys' Manual (USAM) should
accurately reflect the policies set forth in the Attorney General's May 10,2017 memorandum,
"Department C harging and Sentencing Policy." The Department will make appropriate r.;hangcs

to the USAM .

Recommendation 2. Ensure that all U.S. Attorney's Offices consult with their law
enforcement partners to make sure that their policies are current and consistent with local
prosecution priorities.
The Department concurs that the United States Attorneys' offices should consult with
their law enforcement partners in establishing local prosecution priorities. As noted in the draft
report, most United States Attorneys' offices already work in conjunction with their law
enforcement partners in developing local prosecution priorities. Moreover, the Attorney General
has emphasizcd this principle in the context of violent crime. In his March 8, 2017
memorandum to all federal prostX:utors, "Commitment to Targeting Violent Crime," the
Attorney General directed the United States Attorneys to coordinate with federal, state, tribal,
and local law enforcement in identifying and pursuing violent criminals. Accordingly, the
Department will require United States Attorneys to confirm that they have consulted with their
law enforcement partners as required by the March 8, 2017 memorandum.

Recommendation 3. Require all U.S. Attorney's Offices to collect charging dala Ihal will
enable tbe Departmenl to determine wbetber its cbarging and sentencing policies are being
effectively implemented.
The Attorney General's May 10,2017 memorandum setting forth the Department' s
charging and sentencing policy directs United States Attorneys and Assistant Attorneys General
to ensure that deviations from the memorandum ' s general charging and sentencing directives be
approved by a delegated supervisor and documented in the case file. The Department has
traditionally ensured compliance with a charging policy in this manner, particularly by
examining whether the United States Attorneys' offices have instituted processes to facilitate
compliance, such as supervisory reviews of charging decisions, training on charging policies,
and case reviews between or among AUSAs and their supervisors. The Department's new
charging and sentencing policy directs all USAOs to document the file when an exception to the
policy is approved. Accordingly, the Department will require that such documentation be
maintained so that EOUSA can periodically request the USAOs to identify the number of
exceptions granted for comparison against the number of defendants charged. This method of
tracking compliance is consistent with the Attorney General's memorandum and with our
understanding of the intent of the recommendation.

60


APPENDIX 9 


OIG ANALYSIS OF THE DEPARTMENT’S RESPONSE
The OIG provided a draft of this report to the Department of Justice
(Department), including the Criminal Division, the Drug Enforcement
Administration, the Executive Office for United States Attorneys (EOUSA), the
Federal Bureau of Investigation, and the Office of the Deputy Attorney General
(ODAG) for comment. ODAG and EOUSA responded on behalf of the Department
(see Appendix 8) and stated that the Department remains committed to the
principles of prioritizing federal prosecutions on the most serious threats to public
safety and ensuring the fair and impartial administration of justice. While the
Department issued a new charging and sentencing policy on May 10, 2017,
rescinding aspects of the Smart on Crime initiative, the Department acknowledged
that the recommendations in the OIG’s report are relevant to the implementation of
the new policy. Recommendations 1 and 2 are directed to the Department, and
Recommendation 3 is directed to the Department and EOUSA. The Department
concurred with all of the recommendations as follows. We discuss the OIG analysis
of the Department’s response and the actions necessary to close the
recommendations below.
Recommendation 1: Ensure that the U.S. Attorneys’ Manual accurately
reflects Department charging policies.
Status: Resolved.
Department Response: The Department concurred with the
recommendation and stated that appropriate changes will be made to the U.S.
Attorneys’ Manual (USAM) to accurately reflect the policies set forth in the Attorney
General’s May 10, 2017, memorandum, “Department Charging and Sentencing
Policy.”
OIG’s Analysis: The Department’s actions are responsive to our
recommendation. By August 25, 2017, please provide copies of all updates to the
USAM to reflect the changes set forth in the Attorney General’s May 10, 2017,
memorandum.
Recommendation 2: Ensure that all U.S. Attorney’s Offices consult with
their law enforcement partners to make sure that their policies are current and
consistent with local prosecution priorities.
Status: Resolved.
Department Response: The Department concurred with the
recommendation and stated that the U.S. Attorney’s Offices (USAO) should consult
with their law enforcement partners in establishing local prosecution priorities. The
Department further indicated that the Attorney General emphasized this principle,
in the context of violent crime, in a March 8, 2017, memorandum to all federal
prosecutors, “Commitment to Targeting Violent Crime,” which directed the U.S.
Attorneys to coordinate with federal, state, tribal, and local law enforcement in

61


identifying and pursuing violent criminals. The Department will require U.S.
Attorneys to confirm that they have consulted with their law enforcement partners
as required by the Attorney General’s memorandum.
OIG’s Analysis: The Department’s actions are responsive to our
recommendation. By September 29, 2017, please describe how the Department
will require U.S. Attorneys to confirm that they have consulted with their law
enforcement partners to identify and pursue violent criminals, and whether the
Department’s requirement applies to consultations with law enforcement partners
regarding all local prosecution priorities.
Recommendation 3: Require all U.S. Attorney’s Offices to collect charging
data that will enable the Department to determine whether its charging and
sentencing policies are being effectively implemented.
Status: Resolved.
Department Response: The Department did not expressly concur with the
recommendation, but stated that the Attorney General’s May 10, 2017,
memorandum that set forth the Department’s charging and sentencing policy
directs U.S. Attorneys and Assistant Attorneys General to ensure that deviations
from the memorandum’s general charging and sentencing directives be approved
by a delegated supervisor and documented in the case file. The Department
indicated that it has traditionally ensured compliance with a charging policy in this
manner, particularly by examining whether the USAOs have instituted processes to
facilitate compliance, such as supervisory reviews of charging decisions, training on
charging policies, and case reviews between or among Assistant U.S. Attorneys and
their supervisors. The Department’s new charging and sentencing policy directs all
USAOs to document the file when an exception to the policy is approved.
Accordingly, the Department indicated that it will require that such documentation
be maintained so that EOUSA can periodically request the USAOs to identify the
number of exceptions granted for comparison against the number of defendants
charged. It indicated that this method of tracking compliance is consistent with the
Attorney General’s memorandum and with the Department’s understanding of the
intent of the recommendation.
OIG’s Analysis: The Department’s actions are responsive to our
recommendation. By September 29, 2017, please provide specific details on when
and how EOUSA plans to request the number of exceptions granted for comparison
against the number of defendants charged, as well as how it will track and use this
information.

62


APPENDIX 9
 
 
 

 

 
 
 
 
 
 
 
 
 
 
 
 

The Department of Justice Office of the Inspector General
(DOJ OIG) is a statutorily created independent entity
whose mission is to detect and deter waste, fraud,
abuse, and misconduct in the Department of Justice, and
to promote economy and efficiency in the Department’s
  be reported to the DOJ
operations. Information may
OIG’s hotline at www.justice.gov/oig/hotline or
(800) 869-4499.
 

 

Office of the Inspector General
U.S. Department of Justice
www.justice.gov/oig