William & Mary
Law Review
VOLUME 59

NO. 2, 2017

DESIGNED TO FAIL: THE PRESIDENT’S DEFERENCE TO
THE DEPARTMENT OF JUSTICE IN ADVANCING CRIMINAL
JUSTICE REFORM

RACHEL E. BARKOW* & MARK OSLER**
ABSTRACT
One puzzle of President Obama’s presidency is why his stated
commitment to criminal justice reform was not matched by actual
progress. We argue that the Obama Administration’s failure to accomplish more substantial reform, even in those areas that did not
require congressional action, was largely rooted in an unfortunate
deference to the Department of Justice. In this Article, we document
numerous examples (in sentencing, clemency, compassionate release,

* Segal Family Professor of Regulatory Law and Policy and Faculty Director, Center on
the Administration of Criminal Law, New York University School of Law.
** Robert & Marion Short Distinguished Chair and Professor of Law, University of St.
Thomas School of Law (MN).
We are grateful for the excellent research assistance of Beth Caldwell, Cassandra
Deskus, and Hillary Smith. For helpful comments on earlier drafts, we thank Michelle
Alexander, Kevin Arlyck, Susan Bandes, Rochelle Dreyfuss, Barry Friedman, Paul Larkin,
Deborah Leff, Daryl Levinson, Steve Marcus, Anne Milgram, Trevor Morrison, Erin Murphy,
Rick Pildes, Christopher Robertson, Steve Schulhofer, and Tali Farhadian Weinstein. The
views expressed here, however, are solely attributable to the authors.

387

388

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

and forensic science) of the Department resisting common sense criminal justice reforms that would save taxpayer dollars, help reduce
mass incarceration, and maintain public safety. These examples and
basic institutional design theory all point in the same direction: real
criminal justice reform requires putting the right institutions in
charge of criminal justice policy making. This Article offers institutional changes that would help future Presidents make the system
less punitive and reduce prison populations to achieve the broad
transformation that Obama desired but did not attain. A critical
move is to place criminal justice policy making in the hands of
individuals who can advise the President independently of the
institutional interests of prosecutors.

2017]

DESIGNED TO FAIL

389

TABLE OF CONTENTS
INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
I. THE DEPARTMENT OF JUSTICE AGENDA . . . . . . . . . . . . . . . . . .
A. The Prosecutorial Mindset . . . . . . . . . . . . . . . . . . . . . . . . .
B. Structural Factors Creating Bias Against Reform . . . . . .
II. THE DEPARTMENT OF JUSTICE RESPONSE TO REFORM
PROPOSALS DURING THE OBAMA ADMINISTRATION . . . . . . . .
A. Department of Justice Resistance to Criminal Justice
Reforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Sentencing Reform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2. Clemency. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3. Compassionate Release . . . . . . . . . . . . . . . . . . . . . . . . . .
4. Forensic Science . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
B. The Department of Justice’s More Aggressive Pursuit
of State Reforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
III. REDESIGNING THE EXECUTIVE BRANCH AND USING THE
APPOINTMENT POWER FOR INSTITUTIONAL REFORM . . . . . .
A. Presidential Criminal Justice Advisory Commission . . . .
B. Presidential Clemency Board . . . . . . . . . . . . . . . . . . . . . . .
C. Applying Office of Information and Regulatory Affairs
Review to Criminal Justice . . . . . . . . . . . . . . . . . . . . . . . . .
D. Appointments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Department of Justice Appointments . . . . . . . . . . . . . . .
2. The Judicial Branch . . . . . . . . . . . . . . . . . . . . . . . . . . . .
CONCLUSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

390
395
397
400
404
405
406
425
441
449
454
456
459
461
463
466
467
471
473

390

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

INTRODUCTION
If you follow criminal justice policy in the United States, two
themes dominate the discourse of the past several years. First, large
portions of society have awakened to America’s record-breaking
levels of mass incarceration and criminalization. One out of every
three adults possesses a criminal record,1 and our levels of incarceration have no equal anywhere else in the world.2 Second, and
relatedly, there has been a bipartisan call to reform this state of affairs because it is not necessary for—and often undermines—public
safety and because its costs outweigh its benefits.3 News articles
continually tout criminal justice reform as one of the few areas to
bring together many Republicans and Democrats,4 even in Congress
where common ground is hard to find.5
If you move beyond the rhetoric and focus on the reality, however,
not much has changed, even during the Obama Administration
when the President had a stated commitment to getting smart on
1. THE SENTENCING PROJECT, AMERICANS WITH CRIMINAL RECORDS (2015), http://www.
sentencingproject.org/wp-content/uploads/2015/11/Americans-with-Criminal-Records-Povertyand-Opportunity-Profile.pdf [https://perma.cc/H7L5-8MH7].
2. See, e.g., Barack Obama, Commentary, The President’s Role in Advancing Criminal
Justice Reform, 130 HARV. L. REV. 811, 816 (2017) (noting that the United States has 2.2 million incarcerated people, “more than any other country on Earth”).
3. CHARLES COLSON TASK FORCE ON FEDERAL CORRECTIONS, TRANSFORMING PRISONS, RESTORING LIVES ix (2016), http://www.urban.org/sites/default/files/publication/77101/ 2000589Transforming-Prisons-Restoring-Lives.pdf [https://perma.cc/NUW7-6JB2] (“There is broad,
bipartisan agreement that the costs of incarceration have far outweighed the benefits, and
that our country has largely failed to meet the goals of a well-functioning justice system: to
enhance public safety, to prevent future victimization, and to rehabilitate those who have
engaged in criminal acts.”); Obama, supra note 2, at 817 (“There is a growing consensus across
the U.S. political spectrum that the extent of incarceration in the United States is not just
unnecessary but also unsustainable. And it is not making our communities safer.”).
4. See Lynn Adelman, Criminal Justice Reform: The Present Moment, 2015 WIS. L. REV.
181, 188-89.
5. See, e.g., Mark Holden, Criminal Justice Reform Is Ripe for Bipartisan Achievement,
HILL (Jan. 3, 2017, 2:26 PM), http://thehill.com/blogs/congress-blog/judicial/312492-criminaljustice-reform-is-ripe-for-bipartisan-achievement [https://perma.cc/8KCS-FK2X] (“Criminal
justice reform has been one of the few policy areas where Republicans and Democrats have
forged bipartisan consensus.”); Bill Keller, Will 2017 Be the Year of Criminal Justice Reform?,
N.Y. TIMES (Dec. 16, 2016), http://www.nytimes.com/2016/12/16/opinion/will-2017-be-the-yearof-criminal-justice-reform.html?_r=0 [https://perma.cc/4LZ6-44WH] (noting that criminal
justice reform with bipartisan support is “[o]n the short list of things Congress could do to
reassure voters that government is back in business”).

2017]

DESIGNED TO FAIL

391

crime and rolling back harsh punishments. Indeed, President
Obama’s support of reform was so strong that in the waning days of
his time in office, he published a law review article entitled The
President’s Role in Advancing Criminal Law Reform, which highlighted his accomplishments and what he saw as “the tools Presidents can use to effect meaningful change throughout the system.”6
Yet his achievements in this area were modest at best.
Because criminal justice is “administered at all levels of government and shaped by a range of actors,”7 the federal government can
only do so much to tackle the broad tragedy of mass incarceration.
But the federal system is one of the largest jurisdictions in terms of
total imprisonment,8 and it also often sets an example for states
through its policies. Many commentators are quick to place most of
the blame with Congress for stalled progress,9 but significant
criminal law reforms can occur with the use of executive power
alone. Gains were modest under President Obama not because he
lacked the power to do more, but because he followed an institutional model that was designed to fail.10 Obama’s failure to accomplish more substantial reform was largely rooted in the fact that his
efforts were less about the President’s role in advancing criminal
law reform and more about the Department of Justice’s role.11 If
6. Obama, supra note 2, at 815.
7. Id. at 814.
8. See DANIELLE KAEBLE ET AL., U.S. DEP’T OF JUSTICE , CORRECTIONAL POPULATIONS
IN THE UNITED STATES, 2014, at 7 fig.3 (2016), https://www.bjs.gov/content/pub/pdf/cpus14.pdf
[https://perma.cc/DG3Y-6Y6U].
9. See, e.g., Obama, supra note 2, at 829 (“Although the reform bills appeared to have majority support in the Congress, including among many conservatives, Republican leaders have
not yet allowed them to come to the floor for a vote.”); Seung Min Kim, Senators Plan to Revive Sentencing Reform Push, POLITICO (Jan. 4, 2017, 5:13 AM), http://www.politico.com/story/
2017/01/senate-criminal-justice-sentencing-reform-233071 [https://perma.cc/B9RQ-FJGN]
(“[D]espite support from President Barack Obama, powerful congressional Republicans, and
a sprawling network of groups from the left and right, the legislation never made it to the
floor ... [because of] the determined efforts of law-and-order conservatives to steamroll it.”);
Press Release, Office of the Press Sec’y, Press Briefing by Press Secretary Josh Earnest (Dec.
8, 2016), https://obamawhitehouse.archives.gov/the-press-office/2016/12/08/press-briefingpress-secretary-josh-earnest-12816 [https://perma.cc/X3LE-RBFQ] (“[T]he President would
have preferred a more sweeping solution to [criminal justice reform] that only legislation
could bring about, but that is not something that Congress was able to succeed in passing,
despite [bipartisan support] .... [Reform is] another victim of Republican dysfunction in Congress.”).
10. See infra Part I.B.
11. See infra notes 67-68 and accompanying text.

392

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

future Presidents who share President Obama’s hope for reducing
the sweep of criminal law and punishment follow the same model of
deferring to the Department to lead the way on progressive reforms,
they will achieve, at best, the same minimal successes because of
the Department’s hard-wired institutional resistance to reforms that
make things more difficult for prosecutors.12
It makes sense for the President to turn to the Department for
input on criminal justice issues given its collective expertise and
experience in the area, but we believe the Obama Administration is
a case study that shows why the Department is precisely the wrong
entity to put in charge of reform efforts. The Department is the
agency charged with prosecuting federal criminal laws, and its
views on reform are inevitably colored by its prosecutorial functions
and a bureaucratic preference for maintaining a status quo that
favors the interests of prosecutors. In many areas, reform would
require the Department to second-guess its own prior actions. It is
a classic conflict of interest, and it defies everything we know about
human nature as well as government incentives to expect prosecutors in the Department to perform that task objectively and without
prioritizing prosecutorial interests above other valid interests.13
This is not to fault them or cast aspersions. They dutifully carry out
the mission of the Department. But that is precisely why they are
the wrong actors to lead the way on criminal justice reform.
In this Article, we will document numerous examples of how this
conflict drove the Department’s largely successful efforts to weaken,
rather than further, the President’s criminal justice reform goals.
Notably, the areas in which the Department acted most boldly were
in contexts in which its prosecutorial interests dovetailed with
reform efforts or in which the Department itself would be unaffected
by reform proposals. Thus, the Administration was most active in
its efforts to foster state and local reforms that do not affect federal

12. A President who wants to move the Department in the opposite direction—to support
tougher criminal laws and more powers for prosecutors—will likely find the Department to
be a much more willing and cooperative partner.
13. Rachel E. Barkow, Prosecutorial Administration: Prosecutor Bias and the Department
of Justice, 99 VA. L. REV. 271, 307-19 (2013) (explaining why, “given everything we know
about agency design,” the Department will prioritize its prosecutorial functions at the expense
of other goals).

2017]

DESIGNED TO FAIL

393

prosecution.14 But in other areas involving the federal criminal
justice system—areas in which reform would have made the Department’s job harder—the Department resisted more significant
change.15 The Department was particularly resistant when a reform
effort would require it to revisit decisions it already made.16
Whether the decision involved retroactive adjustments to sentences
already handed down,17 commutations or compassionate release for
those currently serving out prison sentences,18 or rethinking the use
of forensic science that had already been employed to obtain convictions,19 the Department resisted second looks of its prior judgments. Yet that kind of retrospective reassessment is critical to
reducing the existing prison population. In all of these areas, the
Department placed its own interests ahead of the broader goals
stated by the President for criminal justice reform.
We provide this retrospective as an important cautionary tale for
any future President who wishes to change criminal justice policies
in the United States to make them less severe. Whether a President
is working with Congress or changing executive policies on his or
her own, the President should get assistance from individuals
focused on presidential priorities, not just prosecutorial ones. And
that means the Department is the wrong leader for most of these
efforts. Asking the Department to do so is asking it to be a judge in
its own cause—the very definition of conflict and bias.20 Even when
people act with the best of intentions—and we have no reason to
doubt anyone acted otherwise—it is asking too much of anyone to
judge situations that involve his or her own interests as if he or she
would be unaffected. While law enforcement views must be seriously
considered as part of any reform, the analysis of criminal justice
reforms must be made by people that represent all the relevant
interests. The final decision must be made by policymakers who
take a broader view and consider all the relevant interests, not just
those of prosecutors.
14.
15.
16.
17.
18.
19.
20.

See infra Part II.B.
See infra Part II.A.
See infra Part II.A.
See infra Part II.A.1.
See infra Parts II.A.2-3.
See infra Part II.A.4.
See supra note 13 and accompanying text.

394

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

Our analysis proceeds in three Parts. Part I begins by explaining
why, as a matter of design, the Department is institutionally illsuited to lead the way on criminal justice reforms because of the
structure of the Department and its prosecutorial mindset.
Part II then shows how the Department’s focus on its institutional self-interest has played out in practice. Section A documents
numerous examples (sentencing, clemency, compassionate release,
and forensic science) of the Department resisting common sense
criminal justice reforms that would save taxpayer dollars, help
reduce mass incarceration, and not undermine public safety. In
other words, these are areas in which an objective decision maker
with an open mind to reform would embrace changes (and indeed,
often did, in opposition to the Department), but in which the Department opposed changes out of its institutional self-interest. Section
B provides further support for the argument that self-interest
largely underlies the Department’s resistance to criminal justice
reforms with numerous examples of the Department’s eagerness to
embrace bolder actions when its own interests were not at stake or
when reforms would make its prosecutorial mission easier to
achieve.
These examples and basic institutional design theory all point in
the same direction: real criminal justice reform requires putting the
right institutions in charge of criminal justice policy making. Part
III thus offers institutional changes that would help future Presidents achieve the broader goal of reducing mass incarceration while
maintaining public safety. The critical move is to place criminal justice policy making in the hands of individuals who balance prosecutorial interests against other costs and benefits and do not simply
decide based on the institutional interests of prosecutors. We believe
the President is best served by creating a commission within the
Executive Office of the President to advise him or her on criminal
justice policy matters.21 An advisory body in the Executive Office of
the President could also perform an oversight function over the Department by requiring it to explain why its policies stand up to costbenefit analysis, promote public safety, and find support in scientific
knowledge and data.22 Part III also explains the importance of
21. See infra Part III.A.
22. See infra Part III.C.

2017]

DESIGNED TO FAIL

395

putting people with diverse experiences in criminal justice administration in key positions that affect criminal justice reform. This
includes not only appointing people to leadership positions within
the Department and agencies working on criminal justice issues,
but also making sure the judicial branch represents diverse perspectives and not just those of prosecutors.
By favoring the view of prosecutors over others involved in criminal justice, the Executive Branch unavoidably favors punishment
and a law enforcement model over rehabilitation; retribution and a
focus on harm over culpability and redemption; and the status quo
over innovation. While the voices of prosecutors must be part of any
reform discussion, other voices must also be heard or the resulting
policies will suffer from an imbalance.
I. THE DEPARTMENT OF JUSTICE AGENDA
Imagine a President who announced that he or she would take
advice on criminal law matters only from the federal defenders’
office. His or her primary advisor would be the Federal Public Defender for the District of Columbia, and experts within the defenders’ offices would speak for the Administration before Congress and
the Sentencing Commission. All pending legislation or proposed
sentencing guideline amendments would be supported by the
Administration only if the defenders bought into the proposal. In
addition, suppose this President intended to put the defenders in
charge of federal prisons, forensics, and the clemency process. The
President’s reasoning would be straightforward: the defenders
understand criminal law at the ground level, offer a national
breadth of understanding, and include experienced experts.23 They
offer experience equal to or better than anyone else within this
specialized world, and are the group of practicing attorneys most
responsible for safeguarding the crucial protections for criminal
defendants embedded in the United States Constitution.

23. The Federal Defenders National Sentencing Resource Council advises defenders, prepares appellate briefs, and offers expert testimony (among other functions). See About, SENT’G
RESOURCE COUNS. PROJECT FED . PUB. & COMMUNITY DEFENDERS, http://www.src-project.org/
about/ [https://perma.cc/X3YC-EJ9Q].

396

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

That President would face a firestorm of dissent, despite the truth
of the claims. Defenders, critics would cry, only represent one part
of the justice system, and are biased given that all of their work is
on behalf of defendants.
Yet, the mirror opposite24—a President whose principal advisor
on criminal law is his or her chief prosecutor;25 who sends prosecutors to speak for the Administration before Congress and the Sentencing Commission;26 who vets proposals through the Department
before determining a position; and who allows the Department to
run the prisons, set our forensics policies, and control the clemency
process27—is our current reality.
It matters—a lot—that the primary advisors to the President on
criminal justice issues have professional interests to protect,
regardless of administration.28 These interests include maintaining
the Department’s size and budget, protecting and enlarging the
power and discretion of its prosecutors, and expanding the menu of
options those prosecutors have by constantly growing the size of the
federal penal code and maintaining mandatory punishments.29 It
24. To be sure, the comparison is not entirely the same because defenders have a primary
duty to their client while prosecutors should have a primary duty to justice. That said, for the
reasons discussed, prosecutors often do operate in a culture that rewards convictions and long
sentences. See also Stanley Z. Fisher, In Search of the Virtuous Prosecutor: A Conceptual
Framework, 15 AM . J. CRIM . L. 197, 206-07 (1988) (explaining that long sentences and convictions are ways prosecutorial competence is measured); Daniel C. Richman, Old Chief v.
United States: Stipulating Away Prosecutorial Accountability?, 83 VA. L. REV. 939, 967-69
(1997) (discussing prosecutorial incentives to maximize convictions).
25. The primary contact between the Department and the President, of course, is the
Attorney General.
26. See, e.g., Attorney General Eric Holder Testifies Before the U.S. Senate Committee on
the Judiciary, U.S. DEP’T JUSTICE (Mar. 6, 2013), https://www.justice.gov/opa/speech/attorneygeneral-eric-holder-testifies-us-senate-committee-judiciary-0 [https://perma.cc/9WTL-GJA4]
(“I’m pleased to join the President, the Vice President, and countless Americans in calling on
Congress to enact legislation addressing gun violence.”); Attorney General Loretta E. Lynch
Testifies at the House Committee on the Judiciary Hearing, U.S. DEP’T JUSTICE (Nov. 17, 2015),
https://www.justice.gov/opa/speech/attorney-general-loretta-e-lynch-testifies-house-committeejudiciary-hearing [https://perma.cc/LC67-8YTK] (“[T]he Justice Department is eager to see
meaningful sentencing reform enacted during this Congress.”).
27. See Barkow, supra note 13, at 278-306.
28. The Department also has great influence with Congress. William Stuntz sets out this
powerful dynamic well, explaining that Congress gives great weight to prosecutors’ views,
even when those views advance the goals of the prosecutors rather than the public. William
J. Stuntz, The Pathological Politics of Criminal Law, 100 MICH . L. REV. 505, 544-46 (2001).
29. See infra notes 46, 52-53 and accompanying text.

2017]

DESIGNED TO FAIL

397

should not surprise us that the size and power of the Department
and the penal code have both exploded in recent years.30 These
things are not always good for the United States, but they are
always good for the Department of Justice.
The Department is also especially likely to bristle at any reform
that questions its prior actions, whether it be a retroactive adjustment to sentences, the early release of someone in federal prison, or
a new look at forensic science that the Department commonly relies
upon.31 In areas such as these, reform may appear to the Department as the nullification of its work. Everything we know about the
endowment effect tells us that the Department will be particularly
resistant to efforts to take back or undo what it has achieved.
The role of the Department, including the bias it brings with it to
discussions of policy, has been a hidden brake on reform efforts
across a striking array of issues within the field of criminal law.32
While prosecutors necessarily represent the Executive Branch in the
federal courts in criminal matters, it is not necessary that they be
the only significant input to the President on policy issues. A close
examination reveals the damage done by giving this dual role of
advisor and interested party to a federal agency that has used that
unique power to expand its own power, even when under the
leadership of self-proclaimed reformers like President Obama and
Attorneys General Eric Holder and Loretta Lynch.
In this Part, we will first set out the nature of the Department’s
inherent bias on policy issues, and then specifically describe the way
that this bias undermined the stated intentions of President Obama
to reform the criminal justice system.
A. The Prosecutorial Mindset
Prosecutors’ chief self-conception is as defenders of the rule of
law. They do not see their role as second-guessing legislative
judgments about criminal law;33 on the contrary, their job—their
30. See Barkow, supra note 13, at 276-77.
31. See id. at 278-306 (explaining how the Department has reacted to these reform
efforts).
32. See infra Part II.
33. See Rachel E. Barkow, Clemency and Presidential Administration of Criminal Law,
90 N.Y.U. L. REV. 802, 859-60 (2015).

398

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

duty—is to enforce that law.34 Prosecutors do, however, believe that
it is their responsibility to exercise discretion in cases, to be the
razor of morality in our society, to discern between the acceptable
and unacceptable, and to decide what charges to bring. It should not
surprise us that the people in that job often share a sense of moral
certainty in what they do.35 That certainty of purpose allows one to
do the job (in fact, that trait is essential to doing the job well), but
also makes it difficult to admit mistakes or to critique the power
structure one resides within.36
Prosecutors are unique among lawyers in that they have no
specific client and instead represent the people overall. Freed of the
need to advocate for another person’s interests, prosecutors instead
rely upon their own consciences and the consciences of their supervisors to direct the use of their great power. The comments to the
ethical rules tell prosecutors that they are “minister[s] of justice,”
but that is left for each prosecutor to define.37 That does not mean
prosecutors do not have a side in the fight that ensues—the fact that
they do is made clear by the oppositional construction of every case
caption and courtroom. Rather, within that oppositional setting,
prosecutors will tend to have a deep personal engagement with their
work and their side stemming from the possession of such tremendous agency, a merging of work and identity.38 It may be hard to
morally separate oneself from a case pursuing a particular defendant when one has affirmatively chosen that case against that
defendant.39 When your job is taking away freedom, fortunes, and
lives, the cost of second-guessing the laws that require it or being
wrong in a given case is very high—perhaps too high to contemplate.40

34. See U.S. CONST. art. II, § 3.
35. Mark Osler served as an Assistant United States Attorney in the Criminal Division
in Detroit from 1995 through 2000.
36. See Bruce Green & Ellen Yaroshefsky, Prosecutorial Accountability 2.0, 92 NOTRE
DAME L. REV. 51, 59 (2016).
37. MODEL RULES OF PROF’L CONDUCT r. 3.8 cmt. 1 (AM . BAR ASS’N 2014).
38. See Jeanne Bishop & Mark Osler, Prosecutors and Victims: Why Wrongful Convictions
Matter, 105 J. CRIM . L. & CRIMINOLOGY 1031, 1034-37 (2015).
39. See id. at 1034-35.
40. To be clear, we are not discussing issues of prosecutorial misconduct here, but rather
prosecutorial conduct in the normal course of business. See id.

2017]

DESIGNED TO FAIL

399

The resulting tenacity of prosecutors and the tunnel vision it can
create has been well documented.41 Susan Bandes has described
“the refusal of prosecutors to concede that the wrong person was
convicted,”42 even when DNA evidence has proven that person innocent.43 That certainty, even after it is disproven, has striking
results: Bandes points out that in about half the cases in which
DNA exonerated an imprisoned individual, investigators did not
even bother to submit the new evidence (showing the real offender)
to the national DNA database, with a common reason being that
they could not believe the wrong person was convicted.44
In a recent “Message from the Executive Director,” Scott Burns
of the National District Attorneys Association concluded by saying,
“The prosecutors I know are ethical, fair and solid members of their
respective communities; the prosecutors I know don’t keep score,
notch their belt or talk about conviction rates.... It is a noble calling
and we should always be proud to say, ‘I’m a Prosecutor.’”45
This statement is consistent with the work of most prosecutors as
they see it, and it is this celebration of prosecutors that lays bare
the prosecutorial mindset more clearly than any critic’s barbs.
Prosecutors, with some justification, see their calling as noble, and
go forward with pride. Given that mindset, we should not be
surprised to observe their resistance to those reforms that at least
imply a failure of their present or past decisions, or ones that
suggest they should have less discretion to exercise their best
judgment. If you see your calling as a noble one, then you should be
trusted to do the right thing in any given case because you are, after

41. See, e.g., DAN SIMON , IN DOUBT: THE PSYCHOLOGY OF THE CRIMINAL JUSTICE PROCESS
22-25 (2012); Rachel E. Barkow, Institutional Design and the Policing of Prosecutors: Lessons
from Administrative Law, 61 STAN . L. REV. 869, 883 (2009); Alafair S. Burke, Improving
Prosecutorial Decision Making: Some Lessons of Cognitive Science, 47 WM . & MARY L. REV.
1587, 1611 (2006); Keith A. Findley, Tunnel Vision, in CONVICTION OF THE INNOCENT: LESSONS
FROM PSYCHOLOGICAL RESEARCH 303, 316-17 (Brian L. Cutler ed., 2012); Bruce A. Green &
Ellen Yaroshefsky, Prosecutorial Discretion and Post-Conviction Evidence of Innocence, 6 OHIO
ST. J. CRIM . L. 467, 488 (2009).
42. Susan Bandes, Loyalty to One’s Convictions: The Prosecutor and Tunnel Vision, 49
HOW . L.J. 475, 475 (2006).
43. Id. at 475-78.
44. Id. at 476.
45. Scott Burns, Proud to Support America’s Prosecutors, 46 PROSECUTOR 10 (2012).

400

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

all, pursuing the public interest. The prosecutor’s beliefs and values
thus merge with his or her idea of the public interest.
This explains why prosecutors resist efforts to make their jobs
harder to do. As they see it, these efforts do nothing but impede
noble pursuits. For example, mandatory minimums are valuable to
prosecutors because they give prosecutors leverage to extract pleas
and obtain cooperation. That is why federal prosecutors have consistently pressed Congress to enact or preserve mandatory minimums.46 Eliminating mandatory minimums would transfer power
currently held by prosecutors to judges, and prosecutors do not trust
judges more than they trust themselves to reach the right outcomes.
Indeed, prosecutors do not want any of their current powers vested
in any other actor for the same reason. The result of this worldview
is that any reform effort to eliminate tools that prosecutors find
helpful or strip them of their authority is going to be met with
resistance precisely because prosecutors see those tools and the
exercise of their authority as furthering a noble cause. It is hard for
them to see that they might not be the best actors to make these
decisions.
B. Structural Factors Creating Bias Against Reform
Beyond the mindset of individual prosecutors, there are structural features of the Department that inexorably push it towards
expanding its own power, resisting reforms that would limit its
power, and fighting to maintain the status quo when it already
46. See, e.g., Child Abduction Prevention Act and the Child Obscenity and Pornography
Protection Act of 2003: Hearing on H.R. 1104 and H.R. 1161 Before the Subcomm. on Crime,
Terrorism & Homeland Sec. of the H. Comm. on the Judiciary, 108th Cong. 15-17 (2003)
(statement of Daniel P. Collins, Associate Deputy Att’y Gen. of the United States) (requesting
legislation that would limit downward departures); Federal Cocaine Sentencing Policy:
Hearing Before the Subcomm. on Crime and Drugs of the S. Comm. on the Judiciary, 107th
Cong. 16-33 (2002) (statement of Roscoe C. Howard, U.S. Attorney for the District of
Columbia) (arguing against a reduction in the penalties for crack cocaine, in part because it
would reduce incentives for defendants to cooperate with prosecutors); Letter from Robert Gay
Guthrie, President, Nat’l Ass’n of Assistant U.S. Attorneys, to Senator Patrick Leahy,
Chairman, and Senator Charles Grassley, Ranking Member, Senate Judiciary Comm. (Jan.
31, 2014), http://www.naausa.org/site/index.php/resources/letters/40-jan-2014-letter-opposingmandatory-minimums-legislation/file [https://perma.cc/6GCG-NJU6] (opposing the reductions
in mandatory minimum sentences because they are an “indispensable tool” to “secure offender
cooperation”).

2017]

DESIGNED TO FAIL

401

possesses authority in an area. One, perhaps the most important, is
the fact that the Department is a large bureaucracy.47 A second is
that the Department is heavily influenced by career prosecutors.48
A third is the coastal bias one finds in the Department, which may
lead the Department’s managers to believe that what happens in
Washington, D.C., or the Southern District of New York is the norm,
rather than what goes on in Waco or Memphis.49
With forty-one components and more than one hundred thousand
employees,50 the Department is, above all, an enormous bureaucracy. The nature of bureaucracies is to perpetuate themselves, and
the Department is no different. It will naturally act to protect its
power and stature by continuously seeking to expand its role. Max
Weber, who shaped much of the way we see bureaucracy, worried
about the “power position” of large bureaucracies and the danger
that their organizational sophistication and control over the dispensation of resources could give them power greater than that of the
sovereign.51 The Department’s preference to expand the penal code
is a perfect example.52 With each new federal crime, the role of the
Department—in investigation, prosecution, and incarceration—grew
further into what had traditionally been the realm of the states.53
A second powerful institutional force within the Department is a
large cadre of career prosecutors, both at Main Justice and spread
out among the offices of United States Attorneys. These career prosecutors fill the rank and file positions as well as leadership posts.
47. See infra notes 50-53 and accompanying text.
48. See infra notes 54-60 and accompanying text.
49. See infra notes 61-62 and accompanying text.
50. U.S. DEP’T OF JUSTICE, FY 2016 AGENCY FINANCIAL REPORT, at I-3 (2016), https://www.
justice.gov/doj/page/file/910486/download [https://perma.cc/Z7BE-ZU7S]; DOJ Organizational
Chart, U.S. DEP’T JUSTICE (June 5, 2015), https://www.justice.gov/sites/default/files/doj/pages/
attachments/2015/04/27/doj_june_2015_2.pdf [https://perma.cc/W62X-9GQY].
51. See MAX WEBER, ECONOMY AND SOCIETY 990-91 (Guenther Roth & Claus Wittich eds.,
Ephraim Fischoff et al. trans., 1978); JAMES Q. WILSON, BUREAUCRACY: WHAT GOVERNMENT
AGENCIES DO AND WHY THEY DO IT 377-78 (1989).
52. See, e.g., Paul J. Larkin, Jr., Public Choice Theory and Overcriminalization, 36 HARV.
J.L. & PUB. POL’Y 715, 761-62 (2013) (noting the Department “has a strong institutional
interest in having as much charging discretion as possible and as many charging options as
Congress can provide”); Stuntz, supra note 28, at 510 (“[T]he story of American criminal law
is a story of tacit cooperation between prosecutors and legislators, each of whom benefits from
more and broader crimes.”).
53. See Stuntz, supra note 28, at 509.

402

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

Attorney General Holder spent most of his career with the Department,54 as did his successor, Loretta Lynch, who started her career
with the Department in 1990 and spent eighteen of the next twentysix years there.55 The Deputy Attorney General under Lynch was
Sally Yates, a career Department employee who served for twentyseven years, primarily as an Assistant United States Attorney.56
Yates’s Associate Deputy Attorney General, David Margolis, was a
fixture at Main Justice for fifty years, primarily advising the upper
echelons of the leadership.57 It would defy everything we know about
human cognition to expect individuals with that kind of institutional loyalty to be objective when assessing reforms. While there is
value in placing people with experience in such posts, we should
also expect that long-term employees are more likely to protect what
is known and familiar rather than reverse course. It should not
surprise us that bureaucracies in fields prizing innovation feature
frequent mobility between employers—for example, the average
tenure of employees at Google and Amazon is only about one year.58
The line prosecutors in the Department share those same
tendencies—something that was especially noticeable in the open
opposition to reform found among Assistant United States Attorneys
in the field offices. The National Association of Assistant United
States Attorneys resisted sentencing reform at every turn,59 and
when President Obama began using clemency more actively, they

54. See Eric H. Holder, Jr., COVINGTON , https://www.cov.com/en/professionals/h/ericholder [https://perma.cc/5L3X-PXYJ]; Attorney General: Eric. H. Holder, Jr., U.S. DEP’T JUSTICE , https://www.justice.gov/ag/bio/attorney-general-eric-h-holder-jr [https://perma.cc/ 2F7DX29B] (indicating that Attorney General Holder also clerked for the Criminal Division of the
DOJ during law school).
55. Attorney General: Loretta E. Lynch, U.S. DEP’T JUSTICE, http://www.justice.gov/ag/bio/
attorney-general-loretta-e-lynch [https://perma.cc/WZ4Y-R4YC].
56. Veronica Stracqualursi, Everything You Need to Know About Former Acting Attorney
General Sally Yates, ABC NEWS (Mar. 29, 2017, 4:11 PM), http://abcnews.go.com/Politics/
acting-attorney-general-sally-yates/story?id=46444595 [https://perma.cc/S7P8-DAWX].
57. Eric Lichtblau, David Margolis, a Justice Department Institution, Dies at 76, N.Y.
TIMES (July 15, 2016), https://www.nytimes.com/2016/07/16/us/david-margolis-a-justice-depart
ment-institution-dies-at-76.html?_r=0 [https://perma.cc/GH54-GZ9Y].
58. Full List of Most and Least Loyal Employees, PAYSCALE, http://www.payscale.com/
data-packages/employee-loyalty/full-list [https://perma.cc/3H4V-GEL6].
59. See infra Part II.A.1.

2017]

DESIGNED TO FAIL

403

even held a rally in Washington, D.C., and protested that the federal system needed more imprisonment, not less.60
Finally, the leadership of the Department tends to come from just
a few coastal districts, largely ignoring the middle of the nation. Of
the last thirteen Attorneys General, only three (Jeff Sessions of
Alabama, Alberto Gonzales of Texas, and John Ashcroft of Missouri)
came from somewhere other than a coastal state.61 The fact that
Main Justice is located in Washington, D.C.—meaning that officials
will live amidst the political elite—creates a strong nexus between
the coasts and the management of the Department. This creates the
potential for blindness to the practices in smaller offices across the
middle of the country, where federal cases can be different than the
60. Steven Nelson, Prosecutors Rally Against Sentencing Reform, Say Build More Prisons,
U.S. NEWS & WORLD REP. (July 17, 2015, 3:36 PM), http://www.usnews.com/news/articles/
2015/07/17/prosecutors-rally-against-sentencing-reform-say-build-more-prisons [https://perma.
cc/7YH4-4LRR].
61. Cf. Attorney General: Alberto R. Gonzales, U.S. DEP’T JUSTICE, https://www.justice.
gov/ag/bio/gonzales-alberto-r [https://perma.cc/V9VZ-CYU8] (indicating that Attorney General
Gonzales was born in and practiced in Texas); Attorney General: Benjamin Richard Civiletti,
U.S. DEP’T JUSTICE, https://www.justice.gov/ag/bio/civiletti-benjamin-richard [https://perma.cc/
8HMJ-EQ2M] (indicating that Attorney General Civiletti was born in New York and practiced
in Maryland); Attorney General: Edwin Meese, III, U.S. DEP’T JUSTICE , https://www.justice.
gov/ag/bio/meese-edwin-iii [https://perma.cc/96DA-RCNS] (indicating that Attorney General
Meese was born in and practiced in California); Attorney General: Eric H. Holder, Jr., supra
note 54 (indicating that Attorney General Holder was born in New York and practiced in
D.C.); Attorney General: Griffin Boyette Bell, U.S. DEP’T JUSTICE, https://www.justice.gov/
ag/bio/bell-griffin-boyette [https://perma.cc/BP3G-QXBX] (indicating that Attorney General
Bell was born in and practiced in Georgia); Attorney General: Janet Reno, U.S. DEP’T JUSTICE,
https://www.justice.gov/ag/bio/reno-janet [https://perma.cc/KL23-XFLE] (indicating that Attorney General Reno was born in and practiced in Florida); Attorney General: John David
Ashcroft, U.S. DEP’T JUSTICE, https://www.justice.gov/ag/bio/ashcroft-john-david [https://
perma.cc/UQE7-3G7B] (indicating that Attorney General Ashcroft was born in Chicago and
practiced in Missouri); Attorney General: Loretta E. Lynch, supra note 55 (indicating that Attorney General Lynch was born in North Carolina and “considers [Brooklyn] her adopted
home”); Attorney General: Michael B. Mukasey, U.S. DEP’T JUSTICE, https://www.justice.gov/
ag/bio/mukasey-michael-b [https://perma.cc/4TF3-958X] (indicating that Attorney General
Mukasey was born in and practiced in New York City); Attorney General: Richard Lewis
Thornburgh, U.S. DEP’T JUSTICE, https://www.justice.gov/ag/bio/thornburgh-richard-lewis
[https://perma.cc/DB35-BDTU] (indicating that Attorney General Thornburgh was born in and
practiced in Pittsburgh and became Governor of Pennsylvania); Attorney General: William
French Smith, U.S. DEP’T JUSTICE, https://www.justice.gov/ag/bio/smith-william-french
[https://perma.cc/4J9R-CPEX] (indicating that Attorney General Smith was born in New
Hampshire and practiced in California); Attorney General: William Pelham Barr, U.S. DEP’T
JUSTICE, https://www.justice.gov/ag/bio/barr-william-pelham [https://perma.cc/SUV6-X2LP]
(indicating that Attorney General Barr was born in New York City and practiced in D.C.).

404

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

norm in a large coastal city, and where prosecutors often pursue
severe sentences for things that would be treated as minor on the
coasts.62
For all these reasons, like a boulder being pushed on an even
surface, the Department is a behemoth whose very mass creates
momentum only in the same direction it was already moving. This
means the Department rolls with reforms that increase its powers
(more criminal laws or mandatory minimums that give it leverage
to obtain pleas and cooperation) and is resistant to those reforms
that will take some of its existing powers away or that question its
prior decisions.63
II. THE DEPARTMENT OF JUSTICE RESPONSE TO REFORM PROPOSALS
DURING THE OBAMA ADMINISTRATION
So far, we have discussed why as a matter of theory, culture, and
institutional design, one should expect the Department to favor law
enforcement interests.64 But we are told—often by individuals within the Department—that the Department represents justice itself
and does not favor any particular view.65 The law enforcement
interest, in other words, is also the public’s interest, so there is no
conflict. We have no reason to doubt that the people working in the
Department are doing their best to achieve justice and what they
view as the right result, not only in individual cases, but as a matter
of policy. But the fact remains that their perspective is necessarily
influenced by the culture within which they work and the outlook
62. This regional variation has been widely documented. Barkow, supra note 33, at 854-55
& nn.287-93 (citing sources documenting “large disparities among the ninety-four different
U.S. Attorneys’ Offices in terms of what cases are prosecuted, what kinds of plea agreements
are offered, and whether prosecutors move for departures under the Sentencing Guidelines”).
63. This is true across federal bureaucracies, of course, and in many instances, the
stability of bureaucracies can be beneficial. One’s assessment of whether this inertia is good
or bad will largely depend on what one thinks of the proposed change. If one’s goal is to push
for criminal justice reforms that lessen the size of the punitive state, however, this
bureaucratic inertia and resistance will act as an impediment to achieving that goal.
64. This has also been true historically. See Barkow, supra note 13, at 306-07.
65. Deputy Attorney General Sally Yates, for example, reportedly told aides to remember
“that locking people up is not the department’s sole mission” because it is not “the Department of Prosecution” but “the Department of Justice.” Eric Lichtblau, Obama Legacy of
Freeing Prisoners May Come Under Trump Siege, N.Y. TIMES (Jan. 15, 2017), https://www.
nytimes.com/2017/01/15/us/politics/obama-prisoners.html [https://perma.cc/5GQH-XG9K].

2017]

DESIGNED TO FAIL

405

it brings. This Part shows how this outlook manifests itself in
practice. Section A explores those areas in which the Department
resisted reforms that would undermine its own authority, while
Section B analyzes the spaces where the Department was more
willing to encourage change because the reforms either expanded
the Department’s authority or did not interfere with it.
A. Department of Justice Resistance to Criminal Justice Reforms
There have been many instances—indeed, in the very areas President Obama highlighted as his primary achievements66—where the
Department resisted reform efforts, but not because they failed the
Obama Administration’s stated goals of pursuing “evidence-based
solutions” and bringing the “system more in line with the values
that define us” such as “fairness, equality, and justice.”67 Rather, the
Department resisted because the reforms would make the Department’s work more difficult, second-guess its prior decisions, or run
counter to its judgment based on experience (as opposed to data). To
be sure, some critical reforms came directly from within the Department, such as Attorney General Holder’s amendments to charging
practices.68 We do not discuss charging reforms here, however,
because we are focused on those areas of reform that are properly
directed by the President and need not rest within the Department’s
judgment. We believe questions about how to enforce criminal laws
in particular cases fall within the Department’s appropriate
sphere.69 This Section thus addresses those areas of policy that need
not rest with the Department but did so under President Obama.
We catalog the most important of those—sentencing reform, clemency, compassionate release, and forensics.
66. President Obama’s list of federal criminal law achievements includes sentencing,
clemency, prison policies, and forensics. Obama, supra note 2, at 824-38, 860-62.
67. Id. at 822, 824, 866.
68. Memorandum from Attorney Gen. Eric Holder, Jr., U.S. Dep’t of Justice, to the U.S.
Attorneys & Assistant Attorney Gen. for the Criminal Div. (Aug. 12, 2013) [hereinafter Holder
Memorandum], https://www.justice.gov/sites/default/files/oip/legacy/2014/07/23/ag-memodepartment-policypon-charging-mandatory-minimum-sentences-recidivist-enhancements-incertain-drugcases.pdf [https://perma.cc/24C3-CA9G].
69. See infra note 158. For areas that are squarely within the Department’s authority,
such as charging discretion, the key to reform will be through appointments. See infra Part
III.D.1.

406

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

1. Sentencing Reform
The Obama Administration witnessed the greatest progress we
have seen in rolling back the harshest policies of the 1980s and
1990s,70 so it may seem nothing short of ungrateful to even ask if his
Administration could have done more to support sentencing reform.
In fact, however, the Department during President Obama’s Administration often stood either as a prominent obstacle to getting
more done or sided with maintaining the status quo. This Section
recounts the Department’s resistance across a range of issues and
seeks to explain why it might have reached different conclusions
than other reformers inside and outside of government.
Begin with the first major push for sentencing reform during
President Obama’s tenure—the effort to eliminate the disparity in
sentencing between crack and powder cocaine. President Obama led
the charge on this front even before he became President, vocally
urging reform of crack cocaine sentences when he was a presidential
candidate.71 Previously, five grams of crack triggered a five-year
mandatory minimum sentence, and fifty grams of crack triggered
the ten-year mandatory minimum.72 One would need five hundred
and five thousand grams of powder cocaine, respectively, to hit those
same triggers, which is what became known as the 100-to-1 ratio
between the two drugs.73 Congress rushed to create the initial 100to-1 framework in a state of panic when crack came on the scene. As
one representative candidly admitted at the time, “We didn’t really
have an evidentiary basis for it.”74 Worse still, that differential
treatment produced perniciously disparate outcomes based on race,

70. See Obama, supra note 2, at 824-38, 860-62.
71. In 2007, then-Senator Obama told a Howard University audience that we should “not
make the punishment for crack cocaine that much more severe than the punishment for
powder cocaine when the real difference between the two is the skin color of the people using
them.” Barack Obama, Remarks at the Howard University Convocation in Washington, D.C.
(Sept. 28, 2007), http://www.presidency.ucsb.edu/ws/?pid=77014 [https://perma.cc/XQ74PBCB].
72. See NAOMI MURAKAWA, THE FIRST CIVIL RIGHT: HOW LIBERALS BUILT PRISON AMERICA
121 (2014).
73. See id.
74. Id. at 125.

2017]

DESIGNED TO FAIL

407

with the harsher punishments for crack falling disproportionately
on African Americans.75
President Obama kept his campaign promise and pushed hard for
the passage of the Fair Sentencing Act (FSA) which raised the quantity triggers for the five-year and ten-year mandatory minimums to
twenty-eight and two hundred eighty grams of crack, respectively.76
While the ratio is still lopsided at eighteen-to-one—the result of a
compromise in Congress that was necessary to pass the Act—it is a
significant improvement from the prior regime that would not have
likely been passed but for President Obama’s leadership and
support, including the Department’s advocacy in favor of the law.77
Yet the Department has often failed to support efforts that would
minimize the racially pernicious effects of the old regime. Consider
the Department’s position before the Sentencing Commission on
how to amend the Sentencing Guidelines in light of the passage of
the FSA. In setting the initial Sentencing Guidelines for drug offenses, the Commission made the decision to use statutory quantities as benchmarks for the guidelines.78 Thus, because Congress
used five grams of crack as the trigger for a five-year sentence in the
Anti-Drug Abuse Act of 1986,79 the Commission’s initial guidelines
75. In 2013, 83 percent of those charged with crack trafficking offenses were black and 5.8
percent were white. U.S. SENTENCING COMM ’N , QUICK FACTS: CRACK COCAINE TRAFFICKING
OFFENSES, https://www.ussc.gov/sites/default/files/pdf/research-and-publications/quick-facts/
Quick_Facts_Crack_Cocaine.pdf [https://perma.cc/5GFS-84V8]. By contrast, blacks made up
only 31.5 percent of powder cocaine trafficking offenders, while Hispanics made up 58 percent
and whites made up 9.4 percent. U.S. SENTENCING COMM ’N , QUICK FACTS: POWDER COCAINE
TRAFFICKING OFFENSES, http://www.ussc.gov/sites/default/files/pdf/research-and-publications/
quickfacts/Quick_Facts_Powder_Cocaine.pdf [https://perma.cc/97LK-BMMA].
76. Fair Sentencing Act of 2010, Pub. L. No. 111-220, 124 Stat. 2372 (codified as amended
in scattered sections of 21 and 28 U.S.C.).
77. In 2009, the Department advocated for eliminating the disparity between crack and
powder. Unfairness in Federal Cocaine Sentencing: Is it Time to Crack the 100 to 1 Disparity?:
Hearing Before the Subcomm. on Crime, Terrorism, and Homeland Sec. of the H. Comm. on
the Judiciary, 111th Cong. 2-3 (2009) [hereinafter Statement of Lanny A. Breuer] (statement
of Lanny A. Breuer, Assistant Att’y Gen., Criminal Division, U.S. Department of Justice),
https://www.justice.gov/sites/default/files/testimonies/witnesses/attachments/2009/
05/21/2009-05-21-crm-breuer-crack-100-to-1.pdf [https://perma.cc/X29M-RK56].
78. See U.S. SENTENCING COMM ’N , REPORT TO THE CONGRESS: COCAINE AND FEDERAL
SENTENCING POLICY iv (2002), http://www.ussc.gov/sites/default/files/pdf/news/congressionaltestimony-and-reports/drug-topics/200205-rtc-cocaine-sentencing-policy/200205_Cocaine_
and_Federal_Sentencing_Policy.pdf [https://perma.cc/L8PK-FBX4].
79. Pub. L. No. 99-570, § 1001, 100 Stat. 3207 (codified as amended at 21 U.S.C.
§ 841(b)(1) (1986)).

408

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

opted to make five grams of crack trigger a guideline offense level
of twenty-six, which yielded a sentencing range of sixty-three to
seventy-eight months (slightly above the five-year mandatory minimum).80 Similarly, fifty grams would trigger an offense level of
thirty-two, or a range of 121 to 151 months.81 The Commission then
extrapolated upward and downward for other drug quantities above
and below these statutory minimums using these levels as the
anchors.82
Because the Commission’s research showed that the 100-to-1
ratio was without basis and produced glaring racial disparities,83 the
Commission decided to take its own steps to limit the differential
treatment between crack and powder in the guidelines even before
the FSA. In 2007, the Commission reduced the guideline offense
level for 5 grams to 24 (yielding a sentence of 51 to 63 months) and
50 grams to 30 (97 to 121 months).84 In doing so, the Commission
kept the ranges in accord with the statutory thresholds but somewhat reduced the severity of crack offenses.85
After the FSA passed, the Commission had to decide whether to
continue to use the lower offense levels or whether it should go back
to the original, higher offense levels under the theory that Congress
had made its own determination of how to address the disparity.86
80.
81.
82.
83.

U.S. SENTENCING GUIDELINES MANUAL § 5A5.2 (U.S. SENTENCING COMM ’N 1987).
Id.
See U.S. SENTENCING COMM ’N , supra note 78, at iv.
U.S. SENTENCING COMM ’N , REPORT TO THE CONGRESS: COCAINE AND FEDERAL SEN TENCING POLICY 3, 8 (2007), http://www.ussc.gov/sites/default/files/pdf/news/congressionaltestimony-and-reports/drug-topics/200705_RtC_Cocaine_Sentencing_Policy.pdf [https://perma.
cc/U5JJ-RS3P]; U.S. SENTENCING COMM ’N , supra note 78, at 91, 102-03; U.S. SENTENCING
COMM ’N , SPECIAL REPORT TO THE CONGRESS: COCAINE AND FEDERAL SENTENCING POLICY 2, 8
(1997), http://www.ussc.gov/sites/default/files/pdf/news/congressional-testimony-and-reports/
drug-topics/19970429_RtC_Cocaine_Sentencing_Policy.pdf [https://perma.cc/A4MK-UVAF];
U.S. SENTENCING COMM ’N , SPECIAL REPORT TO THE CONGRESS: COCAINE AND FEDERAL SEN TENCING POLICY 153-54 (1995), https://www.ussc.gov/sites/default/files/pdf/news/congressional
-testimony-and-reports/drug-topics/199502-rtc-cocaine-sentencing-policy/CHAP5-8.pdf [https://
perma.cc/RVR6-JMCC].
84. See U.S. SENTENCING GUIDELINES MANUAL app. C, amend. 706 (U.S. SENTENCING
COMM ’N Supp. 2008).
85. The Commission decided that “the problems associated with the 100-to-1 drug
quantity ratio are so urgent and compelling” that it needed to do what it could to limit the
impact within the Guidelines. Id.
86. See Fair Sentencing Act of 2010, Pub. L. No. 111-220, § 8, 124 Stat. 2374 (codified as
amended at 28 U.S.C. § 994 (2012)) (vesting the Commission with emergency amendment
authority to promulgate guideline changes consistent with the directives of the statute within

2017]

DESIGNED TO FAIL

409

At the Department’s urging,87 the Commission decided to revert to
the higher, pre-2007 offense levels.88 The Department argued that
the FSA’s language was fairly read to mean that all drug types
should be treated comparably.89 It further argued that “a two-level
reduction [for all drugs] is not warranted until further information
is presented and can be considered.”90
While the Department’s reading was a fair one, it was neither dictated by the statute’s language nor by its legislative history. There
was room to take a different position. In fact, the Commission
received comments from the lead sponsors of the law stating that
they intended the Commission to retain the lower offense levels.
Senator Dick Durbin wrote to the Commission to emphasize that
Congress approved of the Commission’s 2007 changes to the offense
levels for crack and that, “[i]n debating and passing the Fair
Sentencing Act, Congress did not intend for the base offense levels
for crack cocaine to change,” adding that “nothing in the text or
ninety days of its passage).
87. See, e.g., Hearing on Proposed Amendments to the Federal Sentencing Guidelines
Before the U.S. Sentencing Comm’n 16-17, 19 (2011) [hereinafter Statement of Laura E. Duffy]
(statement of Laura E. Duffy, United States Attorney, Southern District of California),
http://www.ussc.gov/sites/default/files/pdf/amendment-process/public-hearings-and-meetings/
20110317/Testimony_DOJ_Duffy.pdf [https: //perma.cc/U8K8-KJ3M] (stating the Department’s position that the Commission should make permanent its temporary amendment to
the Drug Quantity Table, which increased the offense levels back to twenty-six and thirty-two,
respectively).
88. The Commission changed its guideline triggering quantities from 5 to 28 and 50 to
280, respectively, and those levels in turn triggered corresponding offense levels of twenty-six
and thirty-two. 3 U.S. SENTENCING GUIDELINES MANUAL app. C, amend. 748 (U.S. SENTENCING COMM ’N 2011) (explaining this change as “ensur[ing] that the relationship between the
statutory penalties for crack cocaine offenses and the statutory penalties for offenses involving
other drugs is consistently and proportionally reflected throughout the Drug Quantity Table”).
Judge Castillo dissented because he did not believe Congress intended the Commission to
change the offense levels when it passed the FSA. U.S. SENTENCING COMM ’N , PUBLIC MEETING
MINUTES 4 (Oct. 15, 2010), http://www.ussc.gov/sites/default/files/pdf/amendment-process/
public-hearings-and-meetings/20101015/20101015_Minutes.pdf [https://perma.cc/2T9XPLXV].
89. Letter from Jonathan J. Wroblewski, Dir., Office of Policy & Legislation, to the Honorable William K. Sessions III, Chair, U.S. Sentencing Comm’n 2 (Oct. 8, 2010) [hereinafter
Wroblewski Letter], http://www.ussc.gov/sites/default/files/pdf/amendment-process/publiccomment/20101013/DOJ_Comment_100810.pdf [https://perma.cc/4YUU-GVZM] (arguing that
the Act aimed to do a better job of differentiating among drug traffickers and that the
language of the Act “strongly suggests” its approach should be “appl[ied] in all drug trafficking
offenses, across all drug types”).
90. Statement of Laura E. Duffy, supra note 87, at 18-19.

410

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

legislative history [of the FSA] suggests otherwise.”91 Senator
Durbin reiterated this point in a subsequent letter he wrote with
Senator Patrick Leahy, noting that his initial letter “was not contradicted by a single member of Congress” and was in fact buttressed
by additional comments filed with the Commission by the thenChairmen of the House Committee on the Judiciary and the thenChairman of the House Subcommittee on Crime.92 Although the
Commission received a letter from a group of House legislators
opposed to the FSA that urged the Commission not to maintain the
2007 offense levels,93 Senator Durbin observed that “[u]nder traditional canons of statutory construction,” it is the views of sponsors,
not opponents, that are entitled to greater weight.94
The Department may have disagreed because it feared that going
too far in President Obama’s first term would impede the prospects
of future legislation.95 If so, it is worth asking whether the Department is the right entity to make the calculation of whether a reform
at the moment is worth the risk of losing further gains later. The
Commission sided with the Department’s more limited reading in
2010, so we will never know whether greater reform efforts then

91. Letter from Senator Richard J. Durbin, U.S. Senate, to the Honorable William K.
Sessions III, Chair, U.S. Sentencing Comm’n (Oct. 8, 2010), http://www.ussc.gov/sites/default/
files/pdf/amendment-process/public-comment/20101013/SenDurbin_comment_100810.pdf
[https://perma.cc/U6K7-FLUJ].
92. Letter from Senators Richard J. Durbin & Patrick J. Leahy, U.S. Senate, to the
Honorable Patti B. Saris, Chair, U.S. Sentencing Comm’n 1 (Apr. 5, 2011) [hereinafter Leahy
& Durbin Letter], http://www.ussc.gov/sites/default/files/pdf/amendment-process/public-comment/20110321/SenDurbin_Leahy_Comment.pdf [https://perma.cc/GS87-MGS3] (citing Letter
from Representatives John Conyers, Jr. & Robert C. Scott, U.S. House of Representatives, to
the Honorable William K. Sessions III, Chair, U.S. Sentencing Comm’n (Oct. 8, 2010), http://
www.ussc.gov/sites/default/files/pdf/amendment-process/public-comment/20101013/House_
CrimeSubcomte_100810.pdf [https://perma.cc/NW53-Z9PL]).
93. Letter from Representative Lamar Smith et al., U.S. House of Representatives, to the
Honorable Patti B. Saris, Chair, U.S. Sentencing Comm’n (Mar. 15, 2011), http://www.ussc.
gov/sites/default/files/pdf/amendment-process/public-comment/20110321/HR_Judiciary
Subcomte_Comment.pdf [https://perma.cc/5ZQ5-6V6C].
94. Leahy & Durbin Letter, supra note 92, at 1.
95. This theory is supported by a Department letter to the Commission urging it to
“reinforce and support the bipartisanship that led to the passage of the Act and its important
reforms by ensuring that the will of the nearly-unanimous Congress that passed the Act is
achieved in its implementation.” Wroblewski Letter, supra note 89, at 2. Additionally, the
Department supported a reduction of offense levels for all drugs in President Obama’s second
term. See infra note 125 and accompanying text.

2017]

DESIGNED TO FAIL

411

would have stymied future sentencing reform or bipartisan efforts.96
But there are persuasive reasons for thinking that the Department’s
assessment of the costs and benefits of pushing additional reforms
at any given time might be skewed by its institutional biases and
interests. All else being equal, the Department might prefer to err
on the side of caution and keep offense levels high, reasoning that
its prosecutors can agree to lower sentences in cases in which they
believe such reductions are merited. Thus, the Department may
hold the view that it can police injustices instead of giving judges
more room to do so, and not much is lost by doing so. In addition,
the Department might not be as likely to see the problems with the
prior regime because it prosecuted those cases under the old
statutory triggers.97 The Department may also be more inclined as
an institution to focus on its law enforcement mission and less
sensitive to the personal interests at stake on the other side of the
balance. Whereas the Department highlighted in its 2010 comments
to the Commission that it “has consistently advocated for federal
sentencing laws that ... ensure public safety by being tough on drug
crime,”98 Senators Leahy and Durbin paid more attention to the fact
that roughly two thousand additional people would benefit from
sentencing reductions if the Commission kept the lower base offense
levels and made its changes retroactive.99
The Department’s resistance to crack sentencing reform did not
stop there. When the Commission considered whether to make its
amendments retroactive (such that any individual sentenced before
2010 who would receive a lower guideline range under the new FSA
96. If a concern with future reform was what motivated the Department in its resistance,
then the fact that the Commission lowered the offense levels for all drugs in 2014 without
pushback from Congress does suggest that concern was likely misplaced. See Beth Schwartzapfel, Federal Prisons Could Release 1,000 Times More Drug Offenders than Obama Did,
MARSHALL PROJECT (July 23, 2015), https://www.themarshallproject.org/2015/07/23/ federalprisons-could-release-1-000-times-more-drug-offenders-than-obama-did#.hjZw4rGdq
[https://perma.cc/8ANL-B3G7].
97. Department leadership was often careful to point out that its endorsement of changes
to the crack laws should not be seen as indictment of the Department’s prosecutions under
the old laws. See, e.g., Statement of Lanny A. Breuer, supra note 77, at 2 (“In committing
ourselves to pursuing federal cocaine sentencing policy reform, we do not suggest in any way
that our prosecutors or law enforcement agents have acted improperly or imprudently during
the last 15 years.”).
98. Wroblewski Letter, supra note 89, at 2.
99. Leahy & Durbin Letter, supra note 92, at 2.

412

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

guideline amendments than the range that applied to him or her
could seek retroactive adjustment of his or her sentence),100 the Department again took a narrow view. The Department argued that
some people should be “categorically prohibit[ed]” from having a
judge even consider whether a retroactive adjustment would be
consistent with public safety.101 Specifically, it argued that anyone
who received a sentencing enhancement for using or possessing a
weapon should be barred from having a judge reconsider his or her
crack sentence under the old 100-to-1 ratio.102 The Department also
wanted to categorically bar those with more significant criminal
histories from gaining the opportunity to appear before a judge.103
Importantly, no one automatically receives a sentence reduction
even if the Commission makes a guideline change fully retroactive.
Judges are required to “consider the nature and seriousness of the
danger to any person or the community that may be posed by a
reduction in the defendant’s term of imprisonment” before making
a retroactive adjustment.104 In each case, then, a judge must first
determine that a retroactive reduction will not harm public safety
before a defendant receives a lesser sentence.105 The Department’s
position that some people should be categorically banned from even
100. Only the guideline changes could be made retroactive, not the statute because, there
was no authorization in the FSA itself to apply retroactively, whereas the Commission has
authority to make its guideline changes retroactive. So, anyone sentenced pursuant to a
mandatory minimum under the old crack laws would not receive retroactive relief that gave
him or her a sentence below the statutory minimum.
101. Hearing on Retroactive Application of the Proposed Amendment to the Federal Sentencing Guidelines Implementing the Fair Sentencing Act of 2010 Before the U.S. Sentencing
Comm’n 2-3 (2011) [hereinafter Statement of Eric H. Holder, Jr.] (statement of Eric H. Holder,
Jr., Att’y Gen. of the United States), http://www.ussc.gov/sites/default/files/pdf/amendmentprocess/public-hearings-and-meetings/20110601/Testimony_AG_Eric_Holder.pdf [https://
perma.cc/RM2Y-5MFY] (discussing the role of congressional intent in applying retroactivity
to violent offenders).
102. Hearing on Retroactive Application of the Proposed Amendment to the Federal Sentencing Guidelines Implementing the Fair Sentencing Act of 2010 Before the U.S. Sentencing
Comm’n 8-9 (2011) [hereinafter Statement of Stephanie M. Rose] (statement of Stephanie M.
Rose, United States Attorney, Northern District of Iowa), http://www.ussc.gov/sites/default/
files/pdf/amendment-process/public-hearings-and-meetings/20110601/ Testimony_Stephanie_
Rose.pdf [https://perma.cc/SJ8R-DCXP].
103. Id. (explaining that the Department would have barred anyone higher than criminal
history category III from asking a judge to reconsider his or her crack sentence).
104. U.S. SENTENCING GUIDELINES MANUAL § 1B1.10, cmt. n.1(B)(ii) (U.S. SENTENCING
COMM ’N 2015); see 18 U.S.C. § 3582(c)(2) (2012).
105. See U.S. SENTENCING GUIDELINES MANUAL § 1B1.10, cmt. n.1(B)(ii).

2017]

DESIGNED TO FAIL

413

going before a judge was presumably based on a distrust of judges
making these determinations on a case-by-case basis and a willingness to accept that, as a result of its categorical bar, many deserving
individuals would not receive relief. Most individuals convicted of
drug offenses with higher criminal history categories pose little
threat of violence.106 Nor does receiving a weapons enhancement
necessarily signify dangerousness. Defendants can receive the weapon enhancement even if they personally did not have a weapon as
long as someone with them did—and this is true even if the defendant had no knowledge the other person had a weapon.107 The
Department was willing to sacrifice individuals in situations like
these and restrict them from receiving less disparate crack sentences in order to prevent judges from making public safety determinations on an individualized basis.108 It is hard to imagine anyone
but a prosecutor deciding that trade-off makes sense, especially
when it means individuals must continue to serve crack sentences
universally recognized to be unjust and discriminatory.
Indeed, contrast the Department’s preferred approach with the
comments filed by a group of Democratic senators who argued that
all eligible defendants should be able to petition a judge to determine if retroactive adjustments were appropriate.109 The Senators
had confidence “that all judges will consider public safety as a
paramount concern when determining whether a sentence reduction
is appropriate.”110 They argued that “[g]iven decades of inequity in
crack sentencing, the ability of judges to discern on an individualized case-by-case basis who should benefit from a sentence reduction, and the requirement that public safety must be considered, we
106. U.S. SENTENCING COMM ’N , REPORT TO THE CONGRESS: CAREER OFFENDER SENTENCING
ENHANCEMENTS 42 (2016), http://www.ussc.gov/sites/default/files/pdf/news/congressionaltestimony-and-reports/criminal-history/201607_RtC-Career-Offenders.pdf [https://perma.cc/
NR7L-BPD4] (noting that, for repeat offenders whose criminal history includes only drug
trafficking, the most common subsequent offense was drug trafficking).
107. See, e.g., United States v. Woods, 604 F.3d 286, 290 (6th Cir. 2010) (holding that a
defendant need not be aware a firearm is present but instead “the possession of a firearm by
a coconspirator must (1) be connected to the conspiracy and (2) be reasonably foreseeable”).
108. See supra note 103 and accompanying text.
109. See Letter from Senator Patrick J. Leahy et al., U.S. Senate, to the Honorable Patti
B. Saris, Chair, U.S. Sentencing Comm’n (June 1, 2011), http://www.ussc.gov/sites/default/
files/pdf/amendment-process/public-comment/20110602/Durbin_Leahy_Franken_Coons_
Comment.pdf [https://perma.cc/UJX7-D66J].
110. Id. at 3.

414

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

strongly believe that no particular category of defendant should be
excluded from retroactive application.”111
Here again, we see the Department advocating for an approach
that is at odds with broader criminal justice reform goals and with
Democratic members of Congress who had been leaders for sentencing reform. Why the difference, especially when President Obama
and Attorney General Holder spoke urgently about the need to
reform the crack laws?112 One can imagine that, unlike Attorney
General Holder, the members of Congress did not have to deal with
career staff who disagreed with the broader push for reform, so their
resistance to retroactivity naturally followed suit. It is likely that
Attorney General Holder faced substantial pushback within the
Department. The Bush Administration’s Department of Justice
opposed any retroactive application of the 2007 offense level reductions,113 and many of those same line prosecutors were still at the
Department when it was considering what position to take before
the Commission in 2011. Indeed, the National Association of
Assistant United States Attorneys, the professional organization
that represents Assistant United States Attorneys (AUSAs), filed its
own comments opposing any retroactive application of the crack
sentencing guideline amendments.114 One gets a sense of the kinds
of arguments the Attorney General must have been facing within
the Department from the AUSA association’s letter, which noted
111. Id.
112. E.g., Eric H. Holder, Jr., Att’y Gen., U.S. Dep’t of Justice, Remarks at the Annual
Meeting of the American Bar Association’s House of Delegates (Aug. 12, 2013), https://www.
justice.gov/opa/speech/attorney-general-eric-holder-delivers-remarks-annual-meetingamerican-bar-associations [https://perma.cc/BMQ8-4TQ8] (commenting on the successful
“reduction of the unjust 100-to-1 sentencing disparity between crack and powder cocaine”
(emphasis added)); Eric H. Holder, Jr., Eric Holder: We Can Have Shorter Sentences and Less
Crime, N.Y. TIMES (Aug. 11, 2016), https://www.nytimes.com/2016/08/14/opinion/sunday/erich-holder-mandatory-minimum-sentences-full-of-errors.html?_r=0 [https://perma.cc/U3GZN73P] (“There is still a disparity in sentencing for offenses relating to crack and powder
cocaine.... [T]his disparity must go.”).
113. See, e.g., Letter from Alice Fisher, Assistant Attorney Gen., to the Honorable Ricardo
H. Hinojosa, Chair, U.S. Sentencing Comm’n 1 (Nov. 1, 2007), http://www.ussc.gov/sites/
default/files/pdf/amendment-process/public-comment/20071100/PC200711_001.pdf [https://
perma.cc/EHL7-X5JH].
114. Letter from Steven H. Cook, Nat’l Ass’n of Assistant U.S. Attorneys, to the Honorable
Patti B. Saris, Chair, U.S. Sentencing Comm’n (June 2, 2011), http://www.ussc.gov/sites/
default/files/pdf/amendment-process/public-comment/20110602/NAAUSA_Comment.pdf
[https://perma.cc/63SY-V736].

2017]

DESIGNED TO FAIL

415

that early release of any individual currently serving a crack
sentence “would be a slap in the face to the law enforcement officers
who literally risked their lives in order to bring those criminals to
justice.”115 In the view of these line prosecutors, reducing crack
sentences would be giving the affected individuals a “windfall,” not
remedying an injustice.116 They further emphasized the strain
retroactive review would place on the system because roughly
twelve thousand individuals could then be able to seek relief.117
The Department’s compromise approach becomes more understandable (at least as an institutional matter) in light of the internal
pressures the Attorney General likely faced. Attorney General
Holder personally appeared before the Commission at its hearing on
retroactivity in 2011 instead of sending a representative from the
Department—the more typical approach—and he spent most of his
time speaking to the concerns of those who would oppose all
retroactive relief.118 He noted that the retroactive application of the
2007 amendments did not cause an increase in crime.119 He also
emphasized that it would be administratively feasible to process the
cases.120 What he did not do was spend time defending the Department’s view that judges should not be trusted to process all of these
cases.121 One can imagine that Attorney General Holder had been on
the defensive in arguing for any retroactive relief at the Department, so he took the same posture before the Commission, explaining why the middle ground was not too bold instead of recognizing
and defending why it was so narrow.122
Yet the Department’s view was far more limited than the view of
the sentencing reformers in Congress and most other commenters
before the Commission. As it turns out, it was also too limited for
the bipartisan Commission, which unanimously applied its quantity-based reductions to the guidelines retroactively in all cases,
without placing a limit based on the presence or use of a weapon or
115.
116.
117.
118.
119.
120.
121.
122.

Id.
Id. at 3.
Id. at 2.
See Statement of Eric H. Holder, Jr., supra note 101, at 2-4.
Id. at 4.
Id.
See id. at 1-5 (failing to address if judges should be trusted with such cases).
Id. at 2-4.

416

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

criminal history, just as it had done in 2007 with its offense level
reductions.123 The fact that these more objective decision makers,
without an institutional history of prosecuting these cases or a duty
to deal with cases going forward, reached a different conclusion than
the Department shows that the institution making the decision
about reform matters. It also causes one to doubt the Department’s
ability to lead on sentencing reform.
The Department’s subsequent positions on sentencing continued
to follow the same basic pattern, even in President Obama’s second
term and even after the evidence mounted that the reductions in
crack sentencing were not undermining public safety or the Department’s ability to successfully prosecute cases.124 When the Commission asked for comment in 2013 on whether it should reduce all of
its drug sentencing guidelines by two offense levels (what came to
be known as drugs minus two), the Department agreed with the
reductions as a prospective matter.125 But even in the relative
political safety of a second term, with the President more outspoken
about criminal justice reform, and in light of the successful implementation of crack retroactivity, the Department once again balked
when it came time to deciding whether the drugs-minus-two
amendment should apply retroactively.126 The Department became
123. See 3 U.S. SENTENCING GUIDELINES MANUAL app. C, amend. 750 (U.S. SENTENCING
COMM ’N 2012).
124. See, e.g., U.S. SENTENCING COMM ’N , AMENDMENTS TO THE SENTENCING GUIDELINES
22-23 (Apr. 30, 2014), http://www.ussc.gov/sites/default/ files/pdf/ amendment-process/readerfriendly-amendments/20140430_RF_Amendments.pdf [https://perma.cc/GA4K-UZWT] (finding
that the reduction in base offense levels for crack offenses had no deleterious effects on plea
rates or cooperation); U.S. SENTENCING COMM ’N , RECIDIVISM AMONG OFFENDERS RECEIVING
RETROACTIVE SENTENCE REDUCTIONS: THE 2007 CRACK COCAINE AMENDMENT 3 (2007), http://
www.ussc.gov/sites/default/files/pdf/research-and-publications/research-projects-and-surveys/
miscellaneous/20140527_Recidivism_2007_Crack_Cocaine_Amendment.pdf [https://perma.cc/
5R7K-W5DW] (finding no statistically significant difference in recidivism rates of those offenders released earlier).
125. Letter from Jonathan J. Wroblewski, Dir., Office of Policy & Legislation, to the
Honorable Patti B. Saris, Chair, U.S. Sentencing Comm’n 16 (Mar. 6, 2014), http://www.ussc.
gov/sites/default/files/pdf/amendment-process/public-comment/20140326/public-commentDOJ.pdf [https://perma.cc/R24L-NA74].
126. See Hearing on Retroactive Application of the Pending Drug Guideline Amendment to
the Federal Sentencing Guidelines Before the U.S. Sentencing Comm’n 7-8 (2014) [hereinafter
Statement of Sally Quillian Yates] (statement of Sally Quillian Yates, United States Attorney,
Northern District of Georgia), http://www.ussc.gov/sites/default/files/pdf/amendment-process/
public-hearings-and-meetings/20140610/Testimony_DOJ.pdf [https://perma.cc/56J9-83LZ].

2017]

DESIGNED TO FAIL

417

even less generous in its view of retroactivity than it was in 2010.127
Whereas it would have allowed individuals in criminal history
category III to seek retroactive reductions in the FSA amendments,
now it would only allow those in categories I and II.128 And this time
around, it also sought to ban anyone with an aggravating role adjustment or someone who received an enhancement for obstruction
of justice129—something that can happen when a defendant testifies
at trial but the judge decides the defendant is not being truthful.
The Department stated that its limited approach to retroactivity
was driven by “public safety concerns that arise from the release of
dangerous drug offenders and from the diversion of resources
necessary to process over 50,000 inmates.”130 The Department’s
resource concerns had merit, of course, because such a large number
of cases would be time-consuming to process.131 But the assumption
that full retroactivity would result in the “release of dangerous drug
offenders” without the Department’s proposed categorical bars
lacked support, especially in light of the Commission’s experience
retroactively applying its previous crack guideline amendments
without any categorical bars.
In the end, the bipartisan Commission once again unanimously
rejected the Department’s categorical bars on retroactivity and
made its amendments fully retroactive. It delayed implementation
for one year to allow judges time to process the cases and to allow
probation officers and the Bureau of Prisons to prepare people for
reentry.132 The delay, not ill-fitting substantive limits on eligibility,
directly addressed the resource issue. Courts processed more than
forty-six thousand motions, with more than thirty thousand people
receiving retroactive sentencing reductions that lowered their
sentences by 17 percent, or roughly two years, on average.133 If the
127. See id.
128. Id.
129. Id.
130. Id. at 4.
131. See id. at 4-6 (recognizing that prosecutors must devote additional time to growing
caseloads).
132. U.S. SENTENCING COMM ’N , AMENDMENT TO THE SENTENCING GUIDELINES 1-2 (2014),
http://www.ussc.gov/sites/default/files/pdf/amendment-process/reader-friendly-amendments/
20140718_RF_Amendment782.pdf [https://perma.cc/E24D-NDYN].
133. See U.S. SENTENCING COMM ’N , 2014 DRUG GUIDELINES AMENDMENT RETROACTIVITY
DATA REPORT, tbls.1 & 7 (2016), https://www.ussc.gov/sites/default/files/pdf/research-and-

418

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

Department had its way, more than half of the people who received
relief would have been ineligible.134
One is again left to wonder why a bipartisan Commission unanimously agreed that the Department was being too conservative on
the topic of sentencing reform. And the answer again lies in the
Department’s institutional biases in favor of the status quo and
avoiding any risk that someone released early might commit a
dangerous offense. While those traits have their merits, they are illsuited to the task of sentencing reform aimed at tackling mass
incarceration. A one-dimensional focus on the risk that people released earlier might commit violent offenses can obscure attention
to other risks, including those associated with not reducing sentences. The longer people serve in prison, the more difficult reentry
becomes (increasing the risk of recidivism).135 Furthermore, the use
of limited law enforcement resources to house individuals who do
not pose a risk means fewer resources to pursue other cases, help
witnesses, and provide rehabilitation. The Commission compiled a
substantial record after the retroactive reduction of crack sentences
showing the feasibility of that approach without undermining public
safety.136 And of course, disproportionate sentences have palpable
human costs. The Department’s resistance thus owed less to its
focus on the evidence and more to its institutional culture.
There are many other examples where the Department pushed
for longer sentences even when the evidence did not support its
claims.137 One final example especially drives home this theme
publications/retroactivity-analyses/drug-guidelines-amendment/20170510-Drug-RetroAnalysis.pdf [https://perma.cc/39HP-J5MU]. Most motions that were rejected were simply
ineligible for relief (for example, a mandatory minimum applied or the sentence would expire
before the effective date of the Commission’s amendment), but judges denied more than three
thousand on the merits, showing that public safety evaluations were taking place. Id. tbl.8.
134. Only 49.7 percent of the individuals fell within criminal history categories I and II,
the Department’s first cut-off for relief. Id. Additional people would be separately ineligible
because of an enhancement for weapons possession, violence, aggravating role, or obstruction,
but those numbers are unavailable.
135. EXECUTIVE OFFICE OF THE PRESIDENT , ECONOMIC PERSPECTIVES ON INCARCERATION
AND THE CRIMINAL JUSTICE SYSTEM 39-40 (2016), https://obamawhitehouse.archives.gov/sites/
whitehouse.gov/files/documents/CEA%2BCriminal%2BJustice%2BReport.pdf [https://perma.
cc/G6TP-ZC8E] (“[A] growing body of work has found that incarceration increases recidivism.”).
136. See supra note 124 and accompanying text.
137. For instance, when the Commission evaluated which offenses should be treated as

2017]

DESIGNED TO FAIL

419

that the Department’s interests are not well aligned with the goals
of criminal justice reform. President Obama did not stop his efforts
enumerated crimes of violence in its guidelines in the wake of the Supreme Court’s decision
in Johnson v. United States, 135 S. Ct. 2551 (2015), the Department wanted burglary included
in the list. See Letter from Jonathan J. Wroblewski, Dir., Office of Policy & Legislation, U.S.
Dep’t of Justice, to the Honorable Patti B. Saris, Chair, U.S. Sentencing Comm’n 2 (Oct. 30,
2015), http://www.ussc.gov/sites/default/files/pdf/amendment-process/public-hearings-andmeetings/20151105/DOJ.pdf [https://perma.cc/5YFN-JKXQ]. But the empirical evidence showed that the overwhelming majority of burglaries, even those involving dwellings, involve no
violence. U.S. SENTENCING COMM ’N , AMENDMENT TO THE SENTENCING GUIDELINES 3 (2016),
http://www.ussc.gov/sites/default/files/pdf/amendment-process/reader-friendly-amendments/
20160121_RF.pdf [https://perma.cc/LM2C-T6AQ] (citing studies). The Commission thus rejected the Department’s argument and left burglary off its list, noting that judges could instead
give an upward departure in those relatively few cases of burglary where violence was involved. Id. at 4.
More recently, the Department pushed for an across-the-board increase in sentences for all
alien smuggling cases based on its view that smuggling had become increasingly dangerous
and tied to criminal organizations. Letter from Michelle Morales, Acting Dir., Office of Policy
& Legislation, U.S. Dep’t of Justice, to the Honorable Patti B. Saris, Chair, U.S. Sentencing
Comm’n 3-13 (Mar. 14, 2016), http://www.ussc.gov/sites/default/files/pdf/amendment-process/
public-hearings-and-meetings/20160316/20160316_DOJ.pdf [https://perma.cc/4L48-6TMX].
Again, the data did not back up the Department’s request. The Department itself sought a
sentence below the guidelines in 37.2 percent of smuggling cases as part of its fast-track disposition program, which undercuts the view that all the cases are as serious as the Department claimed. U.S. SENTENCING COMM ’N , PUBLIC MEETING MINUTES 11 (Apr. 15, 2016) [hereinafter PUBLIC MEETING MINUTES], http://www.ussc.gov/sites/default/files/pdf/amendmentprocess/public-hearings-and-meetings/20160415/meeting_minutes.pdf [https://perma.cc/ARA7QU66]. Commission data also showed that the majority of the defendants convicted of this
offense serve low-level functions in the smuggling operation. Id. The Commission thus denied
the Department’s request for a blanket increase because the guidelines already accounted for
the increased risk in certain cases by providing for enhancements when there is a risk of
physical injury and other dangers. Id. at 11-12.
Then “[t]he Department opposed any inflationary adjustment to the guidelines’ monetary
tables.” Letter from Jonathan J. Wroblewksi, Dir., Office of Policy & Legislation, U.S. Dep’t
of Justice, to the Honorable Patti B. Saris, Chair, U.S. Sentencing Comm’n 12 (Mar. 9, 2015),
http://www.ussc.gov/sites/default/files/pdf/amendment-process/public-hearings-and-meetings/
20150312/DOJ.pdf [https://perma.cc/LUC8-5W89]. The Department saw this as “an unwarranted reduction” in fraud sentences because, in “the Department’s experience and judgment,
the harm from economic crimes is generally not being overstated.” Id. at 12, 14. But as the
Commission noted, none of the monetary amounts in the Guidelines had been adjusted for
inflation since adoption in 1987, and “[a]s a result, monetary losses in current offenses reflect,
to some degree, a lower degree of harm and culpability than did equivalent amounts when the
monetary tables were established or last substantively amended.” Sentencing Guidelines for
United States Courts, 80 Fed. Reg. 25,782, 25,789 (submitted to Congress Apr. 30, 2015).
There are other examples like these, in which the Department asked for sentencing guideline increases without a basis in empirical facts, and contrary to the Obama Administration’s
overriding goal of being smart on crime and reducing incarceration. In the interest of space,
we are not providing an exhaustive list, but these examples give the general flavor.

420

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

at legislative reform with the FSA. He continued to push for broader
sentencing reform through legislation, specifically encouraging additional reforms to mandatory minimum drug sentencing laws.138
The bill with the best chance of passage ultimately turned out to be
the Sentencing Reform and Corrections Act of 2015.139 That bill was
the outgrowth of an earlier bipartisan proposal introduced by Senators Dick Durbin and Mike Lee, known as the Smarter Sentencing
Act of 2014,140 which would have reduced mandatory minimums for
specified and defined nonviolent individuals from twenty years to
ten years, ten years to five years, and five years to two years; along
with entirely eliminating the mandatory sentence of life imprisonment in drug cases.141 Additionally, the bill would have made the
FSA retroactive and modestly expanded the safety valve to allow
judges to sentence defendants outside of mandatory minimums in
some cases.142
The Obama Administration backed those provisions of the Smarter Sentencing Act, but the Administration’s support was not enough
to get it through the Senate, so the bill was ultimately modified in
various ways to make it less ambitious, including eliminating the
reduction of the five-year mandatory minimum to two years.143
“[E]ven though the bill was imperfect,” President Obama supported
it144 and lamented that Republican leaders in Congress did not let
the bill come up for a vote.145
While it is true that the legislation stalled because leaders in the
Senate refused to bring it forward for a vote, one should not be left
with the impression that the Senate bears sole responsibility for the
fact that the bill turned out as watered down as it did, or that it
failed to get a vote. Here again, it is important to observe the Department’s role in advancing (or as the case may be, not advancing)
sentencing reform legislation.

138.
139.
140.
141.
142.
143.
144.
145.

Obama, supra note 2, at 827.
S. 2123, 114th Cong. (2015).
See Obama, supra note 2, at 827 (citing S. 1410, 113th Cong. § 4 (2014)).
See id.
Id.
See id. at 828.
Id.
See id. at 829.

2017]

DESIGNED TO FAIL

421

The President is correct that the Department supported the
Smarter Sentencing Act, but he neglects to mention that it was not
the only bipartisan bill being considered by the Senate. Senators
Patrick Leahy and Rand Paul proposed a more ambitious piece of
legislation to address the problems with mandatory minimum sentences. The Justice Safety Valve Act of 2013 would have allowed
judges to depart from mandatory minimums in individual cases
based on the nature and circumstances of the offense, the history
and characteristics of the defendant, and the purposes of punishment.146 The Justice Safety Valve Act would thus have effectively
eliminated mandatory minimums as a constraint on judges.147 There
was some reason to believe President Obama would share Senators
Leahy and Paul’s interest in this legislation based on his statement
in a speech at an NAACP Conference that “[f]or nonviolent drug
crimes, we need to lower long mandatory minimum sentences or get
rid of them entirely” and we should “[g]ive judges some discretion.”148
But the Justice Safety Valve Act was a nonstarter with the
Department. The Department has repeatedly endorsed the need for
mandatory minimums because of the leverage they give the Department to extract pleas and cooperation.149 Indeed, one reason the
reduction in the five-year mandatory minimum to two years
disappeared is that unnamed federal prosecutors in New York told
Senator Chuck Schumer that reducing the mandatory minimum
would make it too difficult for prosecutors to obtain guilty pleas and
cooperation in those cases.150 To be sure, it was likely that the
Justice Safety Valve Act lacked sufficient Senate support, so we are
146. S. 619, 113th Cong. § 2 (2013).
147. Id.
148. Remarks at the NAACP Annual Convention in Philadelphia, PA, 2015 DAILY COMP.
PRES. DOC. 5 (July 14, 2015) (emphasis added).
149. See, e.g., Letter from Robert Gay Guthrie, President, Nat’l Ass’n of Assistant U.S.
Attorneys, to Senator Patrick Leahy, Chairman, and Senator Charles Grassley, Ranking
Member, Senate Judiciary Comm. 2 (Jan. 31, 2014), http://www.naausa.org/site/index.php/
resources/letters/40-jan-2014-letter-opposing-mandatory-minimums-legislation/file [https://
perma.cc/ 6GCG-NJU6].
150. Executive Business Meeting (Committee on the Judiciary Jan. 30, 2014) (statement of
Sen. Chuck Schumer, Member, S. Comm. on the Judiciary), https://www.judiciary.senate.gov/
meetings/executive-business-meeting-2014-01-30 [https://perma.cc/76CR-G4TQ] (“[P]rosecutors across New York ... told [him] ... low-level [criminals] ... choose to cooperate rather than
face the mandatory minimum and so if we reduce that lower-level minimum from five to two,
many prosecutors are worried that they can no longer use this tool effectively.”).

422

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

not claiming that it failed to advance just because the Department
did not endorse it. But the position of the Administration matters
not only because it can help secure legislative victories, but also
because it sends a message about what kind of system the President
thinks we should have. By supporting the Justice Safety Valve Act,
the President could have sent the same message that Senators
Leahy and Paul wanted to send, which is that judicial discretion is
critical to a well-functioning criminal justice system. Perhaps
President Obama does not share that view. But if his only input
about what position to take came from the Department, that option
would have been off the table from the start.
The Department’s resistance to reform legislation did not stop
there. Even though the Department was willing to get behind the
modest reductions in the mandatory minimums because those
provisions did not undermine its ability to prosecute cases, it viewed
things differently when a key trade-off that emerged for getting the
legislation passed involved accepting so-called mens rea reform.
Specifically, some key legislators refused to support mandatory minimum reform unless the legislation also included a provision stating
that, if a federal statute did not require the government to prove
that the defendant acted with a particular state of mind, the default
state of mind the government should prove is “knowing.”151 Notably,
the mens rea reform provision had its own bipartisan support in the
House,152 and most criminal law scholars and professional bar
associations have lamented for years that strict liability laws have
no place in the criminal sphere.153 The Model Penal Code has its
151. H.R. 4002, 114th Cong. § 11(1) (2015); see Matt Apuzzo & Eric Lipton, Rare White
House Accord with Koch Brothers on Sentencing Frays, N.Y. TIMES (Nov. 24, 2015), http://
www.nytimes.com/2015/11/25/us/politics/rare-alliance-of-libertarians-and-white-house-onsentencing-begins-to-fray.html [https://perma.cc/8S9L-EG9C]; see also Jim Copland & Rafael
A. Mangual, Let’s End Criminalization Without Representation, NAT’L REV. (Jan. 14, 2017,
4:00 AM), http://www.nationalreview.com/article/443856/criminal-justice-reform-over-criminal
ization-federal-agencies-congress-lawmaking [https://perma.cc/BNW4-WNE3] (noting “[t]he
lack of legislative progress” is due in part to the Administration’s unwillingness to address
mens rea reform).
152. See Apuzzo & Lipton, supra note 151; Copland & Mangual, supra note 151.
153. See, e.g., Laurie L. Levenson, Good Faith Defenses: Reshaping Strict Liability Crimes,
78 CORNELL L. REV. 401, 402 (1993) (“[Strict liability] is a doctrine that contradicts the most
basic principles of modern criminal law.”); Richard A. Wasserstrom, Strict Liability in the
Criminal Law, 12 STAN . L. REV. 731, 731 (1960) (“The proliferation of so-called ‘strict liability’
offenses in the criminal law has occasioned the vociferous, continued, and almost unanimous

2017]

DESIGNED TO FAIL

423

own default mens rea requirement of recklessness, precisely because
of the consensus view among those who study criminal law that it
is inappropriate to punish people if they lack the requisite culpable
state of mind.154 Accepting mens rea reform would likely have led to
passage of the mandatory minimum reforms as well.155
But whatever the views of Democrats in Congress or criminal law
scholars, mens rea was a dealbreaker for the Department because
it believed such a requirement would make it too difficult to prove
its cases, particularly those involving environmental and regulatory
offenses.156 From the Department’s perspective, its resistance made
sense. It gained little as an institution from the mandatory minimum reductions, but it would lose a great deal in its ability to perform its tasks if mens rea requirements were added to its workload.
It is, however, far from clear that a disinterested observer would
make the same calculation. A trade-off that would give retroactive
relief to the almost five thousand individuals serving crack sentences under the old 100-to-1 ratio and limit the damage caused by
mandatory minimums in exchange for the adoption of a criminal
law principle that has been a bedrock element of culpability does not
seem so bad. Indeed, for many it is a win-win.
Not everyone agrees, of course, with how to balance the various
interests at stake in a piece of criminal justice legislation, and
reasonable people can disagree on the trade-off between mens rea
requirements and a reduction in mandatory minimum sentences.157
It may be that the White House itself, even without the Department’s viewpoint, did not want to trade some mandatory minimum
sentencing reductions for a greater burden in regulatory crime
cases. As outsiders, we cannot know where the opposition arose.
criticism of analysts and philosophers of the law.”).
154. See Levenson, supra note 153, at 452 (explaining how the Model Penal Code “requires
proof of a culpable mental state” in order for something to be labeled a crime); Wasserstrom,
supra note 153, at 731 (noting that the Model Penal Code provisions on culpability are
considered a “frontal attack” on strict liability).
155. See Paul J. Larkin, Jr., “A Day Late and a Dollar Short”: President Obama’s Clemency
Project 2014 at 11 (Mar. 8, 2017) (unpublished manuscript) (on file with authors).
156. See, e.g., Apuzzo & Lipton, supra note 151 (listing examples provided by the
Department). President Obama seemed to agree with the Department’s views, noting that
“mens rea reform ... could undermine public safety and harm progressive goals.” Obama,
supra note 2, at 829 n.89.
157. See id.

424

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

What we do know is that, in making an informed decision, it is
critical for the President to be guided by a diverse group of perspectives and not simply the Department, whose position will always
favor law enforcement interests above all else.
It is particularly important for the President to be advised on
legislative proposals from sources other than just the Department
because the President cannot control the exercise of prosecutorial
discretion directly. It is not the President’s place to interfere in
individual cases or tell the Department how to select which targets
to pursue and which ones to leave alone.158 But it is the President’s
duty to consider what criminal law legislation should be pursued,
and that inquiry should take into account—among other things—
how the legislation will interact with prosecutorial discretion.
Prosecutors will rarely—if ever—see a problem with laws that
expand their powers (either by imposing mandatory sentencing or
reducing the elements that must be proven beyond a reasonable
doubt) because they will always trust themselves to exercise discretion and will prefer to make those decisions without hindrance.159
But if one is balancing and considering other interests—the protection of individual liberty, the risk of overcharging, the danger that
sentencing enhancements will be used as leverage to extract pleas
even when cases are weak or to punish those who exercise the right
to trial, the threat of disproportionate sentences in particular
cases—then the value of a judicial check becomes obvious. It is
better for the jury to make the assessment about mens rea and to
have judges play a role in determining the sentence in cases, and
not just the prosecutor. But a careful analysis of these values will
not take place if only prosecutors do the weighing because any
discretion given to another actor takes away theirs.

158. The Department’s decisions about charging policies are thus one area that we believe
appropriately falls within the Department’s purview. The Department changed several of its
charging policies as part of the Smart on Crime push. Obama, supra note 2, at 824-26.
159. See supra notes 45-46 and accompanying text.

2017]

DESIGNED TO FAIL

425

2. Clemency
Perhaps none of President Obama’s criminal law projects received
as much attention as his clemency initiative, which resulted in more
than 1700 commutations of sentence.160 President Obama properly
took credit for “[r]einvigorating [c]lemency,”161 and encouraging
people in prison to apply for commutations,162 while remembering
his gracious meeting with five clemency recipients.163 Other claims
he has made—that his efforts were “unprecedented,” and resulted
in more grants “than the previous eleven Presidents combined”164—
selectively count previous presidential clemency efforts165 and
should be viewed in a broader context. While President Obama’s
grants were laudable, it is also important to note that he denied
nearly twice as many petitions as at least the past five Presidents
combined—a statistic that would likely be even more lopsided if
data on denials alone were available before the Carter Administration.166 More fundamentally than these historical comparisons,
160. Clemency Statistics, U.S. DEP’T JUSTICE, https://www.justice.gov/pardon/clemencystatistics [https://perma.cc/8AHG-DTAL] (statistics through January 19, 2017).
161. Obama, supra note 2, at 835.
162. Id. at 836.
163. Id. at 837.
164. Id. at 836-37. Those claims grew larger as the number of grants increased. See Neil
Eggleston, The Reinvigoration of the Clemency Authority, WHITE HOUSE: BLOG (Jan. 19, 2017,
2:25 PM), https://obamawhitehouse.archives.gov/blog/2017/01/19/reinvigoration-clemencyauthority [https://perma.cc/EEA4-VREP] (“[T]he President has granted more commutations
than any president in this nation’s history and has surpassed the number of commutations
granted by the past 13 presidents combined.”).
165. Specifically, these claims focus on commutations but ignore pardons, and thus, ignore
the efforts of President Ford, who granted more than thirteen thousand pardons to draft
evaders and army deserters. See Casey Tolan, The Bold Step President Obama Could Take
to Let Thousands of Federal Inmates Go Free, FUSION (May 4, 2016, 3:44 PM), http://fusion.
net/story/298158/obama-clemency-board-gerald-ford/ [https://perma.cc/4ESR-CYJK].
166. U.S. DEP’T JUSTICE, supra note 160; Current Fiscal Year Clemency Statistics, U.S.
DEP’T JUSTICE, https://www.justice.gov/pardon/current-fiscal-year-clemency-statistics [https://
perma.cc/U3PU-EL5Y] (statistics through January 19, 2017). According to these statistics,
President Obama denied 18,749 clemency petitions. U.S. DEP’T JUSTICE, supra. Presidents
Carter through George W. Bush denied 11,305 combined. U.S. DEP’T JUSTICE, supra note 160.
The Department of Justice keeps records differently before President Carter’s Administration.
For the earlier period, the Department consolidates denials with petitions that were closed
without presidential action, so it is not possible to know what percentage of those were
denials. Id. President Obama’s denial rate is also comparable to previous Presidents.
President Obama received 33,149 clemency petitions and denied 57 percent of them, whereas
Presidents Carter through George W. Bush received a combined total of 17,150 petitions and

426

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

President Obama’s efforts largely failed if the goal was to (as he put
it) “address particularly unjust sentences in individual cases”
because so many deserving people were denied or never received an
answer.167 A primary cause of this failure was again the central
involvement of the Department.168
Obama failed to heed the advice of his first White House Counsel,
Gregory Craig, who devised a plan to move clemency out of the
Department and into the hands of an independent commission.169
Consistent with his general deference to the Department, Obama
chose to leave clemency primarily in its hands, despite the obvious
conflict of interest inherent in allowing prosecutors to control the
review of their own work.170 Predictably, the pardon power languished, with only one sentence commuted in Obama’s first term.171
Indeed, “President Obama granted fewer clemency applications
than any full-term President since George Washington” in his first
term.172
Eventually, people began to notice that President Obama’s clemency record did not match his rhetoric on criminal law. Beginning
in 2011, ProPublica reporters Dafna Linzer and Cora Currier embarked on a series of articles outlining some of the failings of the
Administration in using the pardon power.173 The headline Obama
denied 66 percent of them. Id.
167. Obama, supra note 2, at 836.
168. We follow in the footsteps of other experts who have urged moving clemency out of the
Department. See Albert W. Alschuler, Bill Clinton’s Parting Pardon Party, 100 J. CRIM . L. &
CRIMINOLOGY 1131, 1167-68 (2010); Daniel T. Kobil, How to Grant Clemency in Unforgiving
Times, 31 CAP. U. L. REV. 219, 231 (2003); Margaret Colgate Love, Justice Department Administration of the President’s Pardon Power: A Case Study in Institutional Conflict of Interest,
47 U. TOL. L. REV. 89, 103-06 (2015); Jonathan T. Menitove, The Problematic Presidential
Pardon: A Proposal for Reforming Federal Clemency, 3 HARV. L. & POL’Y REV. 447, 456-60
(2009); Evan P. Schultz, Does the Fox Control Pardons in the Henhouse?, 13 FED . SENT’G REP.
177, 178 (2001); Charles Shanor & Marc Miller, Pardon Us: Systemic Presidential Pardons,
13 FED . SENT’G REP. 139, 144 (2001).
169. Bill Keller, The Bureaucracy of Mercy, MARSHALL PROJECT (Dec. 14, 2015), https://
www.themarshallproject.org/2015/12/13/the-bureaucracy-of-mercy#.OWbVOK106 [https://
perma.cc/G4ED-3GKP].
170. See Love, supra note 168, at 103, 105; Margaret Colgate Love, Obama’s Clemency
Legacy: An Assessment, 29 FED . SENT’G REP. 271, 272-73 (2017).
171. See U.S. DEP’T JUSTICE, supra note 160.
172. Paul J. Larkin, Jr., Revitalizing the Clemency Process, 39 HARV. J.L. & PUB. POL’Y 833,
855 (2016).
173. The complete series is available at the ProPublica web site: Presidential Pardons,
PROPUBLICA, https://www.propublica.org/series/presidential-pardons [https://perma.cc/2KXD-

2017]

DESIGNED TO FAIL

427

Has Granted Clemency More Rarely than Any Modern President174
is typical of their coverage. By late 2013, other writers had begun to
follow ProPublica’s lead. Perhaps stung by headlines like Judging
from His Clemency Record, Obama Likes Turkeys 10 Times as Much
as People,175 and Obama Neglects His Power to Pardon,176 there was
finally some movement from the Administration on clemency in
January 2014177—inexcusably, too late to make clemency a real
mechanism of reform.
Unfortunately, the first tentative step came from the Department
and signaled a fundamental misunderstanding of what had gone
wrong with the clemency process. On January 30, Deputy Attorney
General James Cole gave a speech at the New York State Bar Association Annual Meeting.178 There he correctly noted that the federal
prison population had grown by nearly 800 percent in thirty years,
before moving onto issues of fairness, and then pivoting to clemency.179 After describing the eight people who had received commutations the previous month from President Obama, Cole made a
request of the bar association members, telling them, “This is where
98HX]. Many of the pieces in the series were featured in the Washington Post. See, e.g., Dafna
Linzer & Jennifer LaFleur, ProPublica Review of Pardons in Past Decade Shows Process
Heavily Favored Whites, WASH . POST (Dec. 3, 2011), https://www.washingtonpost.com/invest
igations/propublica-review-of-pardons-in-past-decade-shows-process-heavily-favoredwhites/
2011/11/23/gIQAElnVQO_story.html?utm_term=.36bc6ba03e3e [https://perma.cc/8YAMZ3HM].
174. Dafna Linzer, Obama Has Granted Clemency More Rarely than Any Modern President,
PRO PUBLICA (Nov. 2, 2012, 9:00 AM), https://www.propublica.org/article/obama-has-grantedclemency-more-rarely-than-any-modern-president [https://perma.cc/XNN6-NR5W].
175. Jacob Sullum, Judging from His Clemency Record, Obama Likes Turkeys 10 Times as
Much as People, FORBES (Nov. 27, 2013, 12:19 PM), http://www.forbes.com/sites/jacobsullum/
2013/11/27/judging-from-his-clemency-record-obama-likes-turkeys-10-times-as-much-aspeople/ [https://perma.cc/M3SG-MZ6J] (noting that Obama had pardoned ten turkeys but
commuted the sentence of only one drug offender).
176. Editorial Bd., Obama Neglects His Power to Pardon, WASH . POST (Dec. 1, 2013),
https://www.washingtonpost.com/opinions/obama-neglects-his-power-to-pardon/2013/12/01/
283600c8-579c-11e3-8304-caf30787c0a9_story.html [https://perma.cc/VUC4-V47Q].
177. Kevin Johnson, States Asked to Identify Prison Candidates for Clemency, USA TODAY
(Jan. 30, 2014, 5:12 PM), http://www.usatoday.com/story/news/nation/2014/01/30/state-baroffenders-clemency/5052847/ [https://perma.cc/F83T-3EJU].
178. James Cole, Deputy Attorney Gen., Remarks as Prepared for Delivery by Deputy
Attorney General James Cole at the New York State Bar Association Annual Meeting (Jan.
30, 2014), https://www.justice.gov/opa/speech/remarks-prepared-delivery-deputy-attorneygeneral-james-cole-new-york-state-bar [https://perma.cc/98DG-G8R7].
179. Id.

428

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

you can help. We are looking to the New York State Bar Association
and other bar associations to assist potential candidates for
executive clemency.”180
Though the idea was—at that point—far from fleshed out, Deputy
Attorney General Cole’s speech made three things clear: an expansion of clemency was worthwhile, the problems with clemency could
be addressed in part through the representation of people in federal
prison by pro bono attorneys, and the clemency process itself would
remain embedded in the Department.181 The first was true, the
second was not, and the third was the mistake that would be the
ultimate stumbling block for the project.
In April 2014, the rest of the plan became clearer. Disgraced
Pardon Attorney Ronald Rogers was finally dispatched, a new pardon attorney came on board, and a broad project to solicit petitions
and find representation for people in prison was announced.182
However, even at this turning point—a point that could only have
been reached with the approval of the President—it was settled that
the Department was firmly in control. The initiative was not
announced by the President himself, or even the Attorney General;
that again was left to Deputy Attorney General Cole.183 The visual
image was striking: an initiative to undo harsh sentencing came
literally from the very building from which that harsh sentencing
had sprung, a stone edifice full of prosecutors who had put some of
those same people in prison.184
180. Id.
181. See id.
182. See Press Release, U.S. Dep’t of Justice, Announcing New Clemency Initiative, Deputy
Attorney General James M. Cole Details New Criteria for Applicants (Apr. 23, 2014), https://
www.justice.gov/opa/pr/announcing-new-clemency-initiative-deputy-attorney-general-jamesm-cole-details-broad-new [https://perma.cc/JC9U-QMGJ]; Sari Horwitz, Justice Department
Outlines Criteria for Clemency to Nonviolent Prison Inmates, WASH. POST (Apr. 23, 2014),
https://www.washingtonpost.com/world/national-security/justice-department-outlines-criteriafor-clemency-to-nonviolent-prison-inmates/2014/04/23/1c5e9932-cad7-11e3-95f7-7ecdde72d
2ea_story.html [https://perma.cc/H5RV-E95M].
183. Deputy Attorney General Cole’s in-person announcement was accompanied by a video
statement by Attorney General Eric Holder made earlier that week. Eric H. Holder, Jr.,
Attorney Gen., Expanding Executive Clemency Criteria, U.S. DEP’T JUSTICE (Apr. 21, 2014),
https://www.justice.gov/opa/video/expanding-executive-clemency-criteria [https://perma.cc/
3B64-2GJD].
184. Photos of the announcement show Deputy Attorney General Cole in the Department
briefing room. See Dafna Linzer, Justice Finally Comes to the Pardons Office and Perhaps to
Many Inmates, MSNBC (Apr. 23, 2014, 12:49 PM), http://www.msnbc.com/msnbc/justice-

2017]

DESIGNED TO FAIL

429

The plan Deputy Attorney General Cole announced focused on
two things: encouraging people in prison to apply for clemency and
establishing a system to assign them attorneys in the private sector.185 At the same time, it concisely described the criteria for the
project, which was intended to focus on people serving long sentences without a history of violence.186 To match up people in prison
with lawyers and vet the petitions, Deputy Attorney General Cole
had assembled a group of prominent outside organizations—the
American Civil Liberties Union, the National Association of
Criminal Defense Lawyers, the American Bar Association, Families
Against Mandatory Minimums, and the federal defenders—which
collectively became known as “Clemency Project 2014” or CP14 for
short.187 By making some of the institutional critics of the Department its partners in the initiative, the Administration was able to
ensure a relative absence of dissent within Washington criminal
reform circles.188
The criteria Deputy Attorney General Cole announced189 defined
the project in confounding ways. Though the primary focus was

finally-comes-the-pardons-office [https://perma.cc/XHH6-FFZW].
185. Press Release, U.S. Dep’t of Justice, supra note 182.
186. Deputy Attorney General Cole’s written press release announced that
[u]nder the new initiative, the department will prioritize clemency applications
from inmates who meet all of the following factors:
•They are currently serving a federal sentence in prison and, by operation of
law, likely would have received a substantially lower sentence if convicted of
the same offense(s) today;
•They are non-violent, low-level offenders without significant ties to large
scale criminal organizations, gangs or cartels;
•They have served at least 10 years of their prison sentence;
•They do not have a significant criminal history;
•They have demonstrated good conduct in prison; and
•They have no history of violence prior to or during their current term of
imprisonment.
Id.
187. Lorelei Laird, Massive Volunteer Effort Will Help with Obama Clemency Proposal,
ABA J. (May 30, 2014, 8:45 PM), http://www.abajournal.com/news/article/massive_volunteer_
effort_will_help_with_obama_clemency_proposal/ [https://perma.cc/TFZ5-4BKW].
188. Alone among experts in the field, leaders of the Clemency Project 2014 consortium
have heralded that Project as a noble success. See, e.g., Norman L. Reimer, The Commutation
Legacy of President Barack Obama: Reflections on Clemency Project 2014—The Legal
Profession’s Response to a Call for Help, CHAMPION , Dec. 2016, at 16.
189. See Press Release, U.S. Dep’t of Justice, supra note 182.

430

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

clearly narcotics crimes,190 the criteria failed to explicitly mention
this. The result was that the Bureau of Prisons (BOP) solicited
petitions from every person in federal prison,191 including child
pornography offenders, white collar criminals, and others who may
have met the expressly stated criteria but would ultimately have
little to no chance of getting clemency based on the de facto focus on
those incarcerated for drug offenses. At the same time, the criteria
tracked exactly some of the more miserly practices of the Department in other policies. For example, as with charging policies,192 the
Department broadly concerned itself with not allowing breaks for
“gang members,” despite the difficulty of understanding and defining that term and the ease with which any group of people (especially younger people and people of color) might find themselves
labeled a “gang.”193
Another criterion, perhaps the most important one, plainly
reflected the defensive crouch of the Department as it managed
clemency: the requirement that the petitioner “likely would have
received a substantially lower sentence if convicted of the same
offense(s) today.”194 With this limitation, the Department effectively
insulated itself from second-guessing the choices made by line
assistants, because the project would now be principally targeted at
changes in the law rather than the overzealousness of prosecutors.
While it created a scramble to determine what the new criteria
meant, the clemency initiative did nothing to change the existing
190. In his speech to the New York Bar, Deputy Attorney General Cole said, “We anticipate
that the petitioners potentially eligible for consideration would include: non-violent, low-level
drug offenders.” Cole, supra note 178. All of the clemency recipients were incarcerated for
drug crimes. U.S. SENTENCING COMM ’N , AN ANALYSIS OF THE IMPLEMENTATION OF THE 2014
CLEMENCY INITIATIVE 9 (2017), https://www.ussc.gov/sites/default/files/pdf/research-andpublications/research-publications/2017/20170901_clemency.pdf [https://perma.cc/Z2SZ-ZZ5H]
(“[A]ll the offenders who received commutations under the Initiative had committed a drug
trafficking offense.”).
191. See Press Release, U.S. Dep’t of Justice, supra note 182.
192. In his August 2013 memorandum on charging, Attorney General Holder stipulated
that mandatory minimums would continue to be enforced where a defendant has “significant
ties” to a gang. Holder Memorandum, supra note 68, at 2.
193. Even experts acknowledge that “we still know little about the psychology of gang
membership—street or prison” and “little about the links between street and prison gangs.”
Jane L. Wood et al., Predicting Involvement in Prison Gang Activity: Street Gang Membership,
Social and Psychological Factors, 38 LAW & HUM . BEHAV. 203, 205 (2014) (citation omitted).
194. See Press Release, U.S. Dep’t of Justice, supra note 182.

2017]

DESIGNED TO FAIL

431

structure for evaluating clemency petitions. It was—and remains—a
long vertical bureaucracy that winds through four federal buildings
and no fewer than seven levels of review.195 It begins with the
petition landing on the desk of a staffer at the Office of the Pardon
Attorney, the first of four Department employees who will perform
an investigation, evaluate the case, and draft a memorandum with
a recommended outcome.196 This is reviewed by the Pardon Attorney, who is responsible for the recommendation that will then go up
the chain.197 The file then goes over to Main Justice, where staff assigned to the Deputy Attorney General review the case.198 They
make their recommendation to the Deputy Attorney General,199 who
again goes over the file.200 At that point, the case is shuttled across
town to the staff of the White House Counsel,201 who review clemency cases amongst many other tasks.202 Next up is the White House
Counsel, who gives the case its penultimate review before (finally)
sending it to the President.203 The process often takes years.204
CP14 did nothing to change the stunningly inefficient Department bureaucracy described above. Instead, it just added more lay195. See Mark Osler, Clementia, Obama, and Deborah Leff, 28 FED . SENT’G REP. 309, 309
(2016). The Pardon Attorney is housed at a satellite Department building in Washington, the
Deputy Attorney General works at the “Main Justice” building at 950 Pennsylvania Avenue,
the White House Counsel staff is located at the Eisenhower Executive Office Building, and
the President, of course, is at the White House. Id.
196. See Samuel T. Morison, The Politics of Grace: On the Moral Justification of Executive
Clemency, 9 BUFF. CRIM . L. REV. 1, 36-38 (2005).
197. Id. at 39. Those at the levels above the Pardon Attorney—the Deputy Attorney
General, White House Counsel, and their staffs, as well as the President—normally only see
this recommendation, not the complete file. See id. at 40-41.
198. Osler, supra note 195, at 309.
199. 28 C.F.R. § 0.36 (2008).
200. See id. The Deputy Attorney General is the direct supervisor of the U.S. Attorneys
and their Assistants. Organization, Mission & Functions Manual: Attorney General, Deputy
and Associate, U.S. DEP’T JUSTICE, https://www.justice.gov/jmd/organization-mission-andfunctions-manual-attorney-general#dag [https://perma.cc/7VXQ-YVC8]. The Department of
Justice’s organizational chart illustrates the remarkable breadth of the job. See DOJ Organizational Chart, supra note 50.
201. Sari Horwitz, Struggling to Fix a “Broken” System, WASH . POST (Dec. 5, 2015), http://
www.washingtonpost.com/sf/national/2015/12/05/holderobama/ [https://perma.cc/CVB7-JN82].
202. Presidential Department Descriptions, WHITE HOUSE, https://obamawhitehouse.
archives.gov/participate/internships/departments [https://perma.cc/VFA5-DR8F].
203. See Osler, supra note 195, at 309.
204. Kimberly Lawson, How Significant Are Obama’s 1,385 Commutations?, BROADLY (Jan.
18, 2017, 7:29 PM), https://broadly.vice.com/en_us/article/how-significant-are-obamas-1385commutations [https://perma.cc/S5NJ-RX3Q].

432

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

ers. On top of the unwieldy process described above, CP14 created
its own process, with redundant reviews and part-time experts.205
Surveys from more than thirty-six thousand people in prison were
sent to CP14 for an initial screen for basic disqualifiers.206 From
there, four more points of review were established (for a total of five,
including the initial screen): a pro bono attorney assessed the case,
then a “screening committee” reviewed a summary prepared by the
attorney, followed by a similar, redundant review by a “steering
committee.”207 At this point, a petition would be drafted and reviewed by CP14,208 before being submitted to the Pardon Attorney
to then run the entire gauntlet within the Administration. In all, a
clemency case traversing CP14 and then the Administration would
face no less than a dozen sequential chokepoints with a varied array
of personnel and filters, with no safeguards for consistency.
This model was not the only option. Indeed, in choosing this
model, President Obama rejected the clearest precedent for a
successful large-scale clemency project: Gerald Ford’s 1974-1975
bipartisan Presidential Clemency Board, which resulted in more
than thirteen thousand pardons for draft evaders and deserters
from the armed services during the Vietnam era.209 President Ford
managed to give each case individual consideration while maintaining a scale Obama could not, all while avoiding controversy. As
writer Abby Rapoport noted, “Ford’s boards made it easier to give
draft-dodgers more just sentences; they considered each application
individually, but the sheer number of cases meant that the media
and public didn’t—couldn’t—scrutinize each pardon individually.”210
Critically, Ford did this by circumventing the normal process by
taking it out of the Department.211
205. See Keller, supra note 169.
206. Clemency Project 2014, NAT’L ASS’N OF CRIMINAL DEF. LAWYERS, https://www.nacdl.
org/cp2014/ [https://perma.cc/MTG2-G26P].
207. Keller, supra note 169.
208. Id.; Osler, supra note 195, at 309-10.
209. Tolan, supra note 165.
210. Abby Rapoport, Why Obama Should Take a Cue from Gerald Ford on Crack Pardons,
AM . PROSPECT (Feb. 6, 2014), http://prospect.org/article/why-obama-should-take-cue-geraldford-crack-pardons [https://perma.cc/NLK3-H4ZT].
211. See id. The Ford model was not a mystery to the Obama administration; advocates
presented the Ford Commission’s Report to several officials in the White House and the
Department of Justice. See Tolan, supra note 165.

2017]

DESIGNED TO FAIL

433

If CP14 was unsuccessful in altering the Department-centered
process that had served prior Presidents poorly,212 it was a resounding success at something else: creating new hope among people in
federal prisons. Attorney General Holder publicly predicted that ten
thousand people might be freed213 and the BOP sent a notice about
the project along with the criteria for eligibility to every person in
federal prison.214
Over the next two years, it became clear that those hopes would
not be realized, and that neither CP14 nor the Department were up
to the task. The Marshall Project reported the grim statistics about
twenty months after the announcement of CP14: more than thirtythree thousand applications had resulted in just 224 petitions
submitted to the Pardon Attorney, and only four cases had been
granted commutations after going through the CP14 gauntlet.215
Clearly, the tactics of adding bureaucracy and keeping the Department at the center of the action was not working.
Outside observers were not alone in reporting these struggles.
Deborah Leff, who became the Pardon Attorney in April of 2014,216
resigned less than two years after being appointed.217 A few months
later, her resignation letter to Deputy Attorney General Sally Yates
became public.218 One sentence in that letter laid bare the problem
212. See Margaret Colgate Love, The Twilight of the Pardon Power, 100 J. CRIM . L. &
CRIMINOLOGY 1169, 1195-1203 (2010). Within the Bush Administration, two White House
Counsels (Harriet Miers and Fred Fielding) had complained that good cases were not reaching
their desk. See Linzer & LaFleur, supra note 173.
213. Dara Lind, President Obama Was Supposed to Shorten 10,000 Prison Sentences. What
Happened?, VOX (Mar. 29, 2016, 12:40 PM), http://www.vox.com/2016/3/29/11325502/obamapardon-clemency-leff [https://perma.cc/S6CG-SLVD].
214. The text of that letter is available at Notice to Inmates: Initiative on Executive
Clemency, Fed. Bureau of Prisons, U.S. Dep’t Justice, https://www.bop.gov/resources/news/
pdfs/Notice_to_Inmates_Initiative_on_Executive_Clemency.pdf [https://perma.cc/585P-A3GX].
215. Keller, supra note 169. The final number of inmate applications later topped thirty-six
thousand. NAT’L ASS’N OF CRIMINAL DEF. LAWYERS, supra note 206.
216. Linzer, supra note 184.
217. See Letter from Deborah Leff, Pardon Attorney, U.S. Dep’t of Justice, to Sally Quilian
Yates, Deputy Attorney Gen., U.S. Dep’t of Justice (Jan. 15, 2016), reprinted in 28 FED .
SENT’G REP. 312 (2016) [hereinafter Leff Letter]. Leff was not a career prosecutor, unlike her
predecessor and successor. See Carrie Johnson, New Pardon Chief in Obama Justice Department Inherits a Huge Backlog, NPR (Feb. 3, 2016, 2:00 PM), http://www.npr.org/2016/02/03/
465429779/new-pardon-chief-in-obama-justice-department-inherits-host-of-problems [https://
perma.cc/Z9DA-VA99] (describing successor Robert Zauzmer’s background); Linzer, supra
note 184 (describing predecessor Ronald L. Rodgers’s and Deborah Leff’s backgrounds).
218. Gregory Korte, Former Administration Pardon Attorney Suggests Broken System in

434

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

with clemency’s circuitous route through the Department of Justice:
“In addition, as you know, I have been deeply troubled by the decision to deny the Pardon Attorney all access to the Office of White
House Counsel, even to share the reasons for our determinations in
the increasing number of cases where you have reversed our
recommendations.”219 The sequential chain of decision-making had
become so long and strained that one end could no longer communicate with the other, and the President never found out when the
Pardon Attorney recommended a grant but was overruled by the
Deputy Attorney General.220
Part of the problem was a lack of resources.221 In announcing the
initiative, Deputy Attorney General Cole promised that Department
lawyers would be detailed to the Office of the Pardon Attorney to
help deal with the predictable influx of cases.222 As Leff’s resignation
letter made clear, that help never came, despite a dramatic increase
in the office’s caseload.223
As President Obama’s second term came to an end and it was
clear that his clemency project was unlikely to reach all the eligible
candidates under his initiative, a group of experts and advocates
wrote an open letter to the President, urging him to “determine that
nonviolent offenders in certain extremely low-risk categories either
deserve expedited review or should be granted clemency absent an
individualized review.”224 The letter urged particular attention to
those who were sentenced under the old 100-to-1 crack sentencing
Resignation Letter, USA TODAY (Mar. 28, 2016, 5:37 PM), http://www.usatoday.com/story/
news/politics/2016/03/28/former-administration-pardon-attorney-suggests-broken-systemresignation-letter-obama/82168254/ [https://perma.cc/Z4HZ-RRTT].
219. Leff Letter, supra note 217.
220. When Leff’s letter became public, the Department changed that policy. See Sari
Horwitz, Lack of Resources, Bureaucratic Tangles Have Bogged Down Obama’s Clemency
Efforts, WASH . POST (May 6, 2016), https://www.washingtonpost.com/politics/courts_law/lackof-resources-bureaucratic-tangles-have-bogged-down-obamas-clemency-efforts/2016/05/06/
9271a73a-1202-11e6-93ae-50921721165d_story.html [https://perma.cc/BPV9-GQR4] (noting
that the Department decided to allow Leff’s replacement to have contact with the White
House).
221. Cf. Leff Letter, supra note 217 (claiming she cannot fulfill her responsibilities as
Pardon Attorney without the previously promised additional staff).
222. Press Release, U.S. Dep’t of Justice, supra note 182.
223. Leff Letter, supra note 217.
224. Letter from Glenn E. Martin, President, JustLeadershipUSA, et al. to President
Barack Obama (Nov. 29, 2016), http://www.law.nyu.edu/sites/default/files/upload_documents/
Expanded%20Clemency%20Letter%20to%20POTUS.pdf [https://perma.cc/2CVV-4RVW].

2017]

DESIGNED TO FAIL

435

structure.225 Despite the fact that by the last few months of the
Obama presidency, it was clear that statutory reform was not going
to happen and that no other avenues existed for thousands of easily
identified people who would be free if sentenced under the current
crack law, Attorney General Loretta Lynch told reporters that a categorical approach “would be hard to craft.”226 That statement is hard
to understand, because it was always an option for the Attorney
General to ask the Sentencing Commission to provide him or her
with lists of people who met the different criteria deemed important
to the Department.227 And of course the Department’s rejection of
categories of people who merited sentencing relief stands in sharp
contrast to its endorsement of using categorical bars to limit who
could get retroactive adjustments under the Sentencing Guidelines.228 In the latter context, the Department had no trouble crafting categories in barring people from accurate-length sentences.229
Certainly, many in the Obama Administration worked hard on
clemency,230 but in part that was because of the inefficiency of the
process—you have to pedal hard to get a rusty bicycle to move. The
Administration got off to a late start even thinking about clemency
and then struggled to keep up with the number of petitions filed.231
225. See id. President Obama could have given all the individuals serving sentences under
the old crack laws the opportuntity to petition a judge for resentencing, allowing them to
make an argument for being sentenced under the Fair Sentencing Act of 2010 terms, while
also allowing judges to check to be sure a public safety risk did not exist. See Larkin, supra
note 155, at 11-12 (describing potential provisions a wide-scale commutation order could
contain). Paul Larkin lays out how this process could have been constructed. See id.
226. Laura Jarrett, AG Loretta Lynch Says Don’t Expect Blanket Pardons from Obama,
CNN (Dec. 15, 2016, 5:25 PM), http://www.cnn.com/2016/12/15/politics/lynch-no-blanketpardons/ [https://perma.cc/KF3G-KVGC].
227. See 28 U.S.C. § 994(w)(1), (4) (2012) (requiring the Commission to make data files
regarding the reasons for imposing a sentence available to the Attorney General upon request).
228. See supra note 101 and accompanying text.
229. See supra note 102 and accompanying text.
230. White House Counsel Neil Eggleston alluded to this when he hyperbolically claimed
that he told his staff, “No more eating, sleeping or drinking until we get all these commutations done.” Sarah Wheaton, White House Promises to Speed up Clemency Program, POLITICO
(Apr. 1, 2016, 1:23 PM), http://www.politico.com/blogs/under-the-radar/2016/04/ white-houseclemency-speeding-up-221467 [https://perma.cc/L4JW-HM7P].
231. See Jacob Sullum, Obama Issued More Commutations than Any Previous President but
Neglected Pardons, REASON : HIT & RUN (Jan. 20, 2017, 7:30 AM), http://reason.com/blog/2017/
01/20/obama-issued-more-commutations-than-any [https://perma.cc/NP3P-UVSM] (claiming
“the most striking aspect of Obama’s commutations” are their concentration at the end of his

436

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

A full 89 percent of the clemency grants took place in President
Obama’s last ten months in office, a stark contrast with most
previous Presidents who granted clemency continuously from the
beginning of their terms.232 Thirty-one percent of these grants came
in the final month.233 Deputy Attorney General Sally Yates told the
Washington Post that she reviewed “hundreds of petitions” over the
2016 holidays while working “24-7 over the Christmas break.”234 We
applaud Yates’s commitment, but a process where a key official is
reviewing hundreds of petitions with lives at stake over a holiday is
simply not part of a process that is well designed. It also might
explain the large number of people who never got an answer from
the Obama Administration on their clemency application, even
though they filed their petitions by the Administration’s stated
deadline.235
Obama’s process failed to reach all the people who met the
Administration’s stated criteria,236 much less all the people who
deserved clemency. The clearest measure of success is not the total
number of clemencies granted (as the Administration touted
throughout his presidency),237 but the rate of grants against the

Administration).
232. See id.
233. See U.S. DEP’T JUSTICE, supra note 160.
234. Sari Horwitz, Obama to Commute Hundreds of Federal Drug Sentences in Final
Grants of Clemency, WASH . POST (Jan. 16, 2017), https://www.washingtonpost.com/world/
national-security/obama-to-commute-hundreds-of-federal-drug-sentences-in-final-grants-ofclemency/2017/01/16/c99b4ba6-da5e-11e6-b8b2-cb5164beba6b_story.html?hpid=hp_hpbanner-low_criminal-justice-240pm%3Ahomepage%2Fstory&utm_term=.75d5a3672387
[https://perma.cc/6RQY-K4G3]; see also U.S. DEP’T JUSTICE, supra note 160 (indicating that
President Obama granted 539 commutations between December 19, 2016, and January 19,
2017).
235. See U.S. DEP’T JUSTICE, supra note 160 (indicating that at the end of the Obama
Administration, 4250 clemency petitions were closed without presidential action, while 8880
were pending—neither acted upon nor administratively closed); Sean Nuttall, Inside the
Clemency Lottery, MARSHALL PROJECT (Jan. 26, 2017, 10:00 PM), https://www.themarshall
project.org/2017/01/26/inside-the-clemency-lottery?utm_medium=email&utm_campaign=
newsletter&utm_source=opening-statement&utm_term=newsletter-20170127-681#.
G7SOYsLCd [https://perma.cc/BMD7-7XSB] (noting that approximately 10 percent of the two
hundred petitions filed by the Clemency Resource Center remained pending without an
answer, despite being filed by the stated Department deadlines).
236. See Nuttall, supra note 235 (claiming “petitions that clearly appeared to meet the
criteria” were ultimately denied, while others never received a decision at all).
237. See supra notes 160-65 and accompanying text.

2017]

DESIGNED TO FAIL

437

total number of meritorious petitions submitted.238 A report by the
United States Sentencing Commission identified 2687 individuals
convicted of drug trafficking offenses who met the 6 criteria, yet only
92 of them received clemency—just a little more than 3 percent.239
And that is holding to the initial announced criteria. If one takes a
broader view, there are many more left behind. The grants offer
only the numerator; to understand what happened to the hope that
President Obama and Attorney General Holder generated, we must
consider the denominator as well, especially given the super-sized
prison population. As Deputy Attorney General Cole acknowledged
in his speech to the New York Bar, federal prison populations had
gone up nearly 800 percent in the thirty years leading up to that
moment. Those serving crack sentences under the 100-to-1 laws
alone amounted to more than five thousand people.240 Moreover, if
one compares President Obama’s grant rate with other Presidents
and not simply his total number of grants, the modesty of his
clemency efforts becomes even clearer. His final grant rate relative
to petitions received for both pardons and commutations was 5
percent.241 By contrast, President Kennedy granted 36 percent of the
petitions filed, President Johnson 31 percent, President Nixon 36
percent, President Ford 27 percent,242 President Carter 21 percent,
President Reagan 12 percent, President George H.W. Bush 5
percent, President Clinton 6 percent, and President George W. Bush
2 percent.243 President Obama’s overall clemency grant rate was 5.3
percent, barely better than President Clinton and far behind
238. Cf. supra notes 166-67 and accompanying text (claiming that despite the high number
of grants, President Obama failed to meet his goal to address particularly unjust sentences).
239. U.S. SENT’G COMM ’N , supra note 190, at 2.
240. See Attorney General Confirmation Hearing Before the S. Comm. on the Judiciary, Day
1, Part 1, CSPAN (Jan. 10, 2017), http://cs.pn/2iaTFPr [https://perma.cc/BS42-4NXK] (statement of Sen. Dick Durbin) (referring to the “almost five thousand still serving under this
unfair 100-1 standard”).
241. U.S. DEP’T JUSTICE, supra note 166 (indicating that at the end of the Obama Administration, he had granted 212 pardons and 1715 commutations out of 36,544 total petitions
received).
242. See Barkow, supra note 33, at 816-17. This figure does not include the over thirteen
thousand pardons Ford issued outside of the regular clemency process to draft evaders and
Army deserters. See supra note 165 and accompanying text. Presumably, the Pardon Attorney’s statistics only include cases that went through that office, and the Ford grants came
through an alternative process.
243. Barkow, supra note 33, at 816-17.

438

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

Presidents Reagan, Carter, Ford, Nixon, and Johnson.244 The
numbers are even worse if one looks solely at pardons, in which
President Obama’s grant rate of 6 percent places him among the
most frugal pardon-granters in history.245 Moreover, shortly after
President Donald Trump’s inauguration, a pile of 11,374 clemency
petitions were still pending; the Department and the White House
never got to them.246 Obama cranked the Department machine as
hard as he could, but it was not capable of delivering on the promise
of hope that he had made.
Beyond the failure reflected in the grant rates, deep tragedy
occurred at a personal level. Some of those denied clemency were
unworthy of it, of course. But others were prime candidates under
the stated criteria and still did not receive relief.247 Sean Nuttall248
writes about his client, Tom, who at twenty-one years old became
addicted to heroin.249 Tom’s girlfriend was also a drug user, and
when she asked for heroin from Tom, he gave it to her, not realizing
she had also taken a large quantity of cocaine.250 Although Tom’s
girlfriend took only about one-twentieth the quantity needed for a
lethal overdose of heroin, she died of an overdose because of the
cocaine in her system.251 Tom was charged with her death and with
drug distribution.252 The jury acquitted him of causing her death,
but found him guilty of drug distribution.253 The judge then sentenced Tom to eighty years in prison because he concluded—at odds

244. U.S. DEP’T JUSTICE, supra note 160.
245. See id. (indicating that at the end of the Obama Administration, he had granted 212
pardons out of 3395 petitions received).
246. See id. Obama inherited 1943 commutation and pardon petitions, received another
36,544, and resolved only 27,144. See id.
247. See Nuttall, supra note 235 (claiming petitions that clearly met the clemency criteria
were eventually denied).
248. Nuttall worked for the Clemency Resource Center, the organization we created to help
get more clemency applicants legal help with their petitions. See Clemency Resource Center,
CTR. ON ADMIN . CRIMINAL L., http://www.law.nyu.edu/centers/adminofcriminallaw/clemency
[https://perma.cc/84D7-3SKF]; see also Nuttall, supra note 235 (noting the author’s work with
the Clemency Resource Center).
249. See Nuttall, supra note 235 (claiming Tom became addicted to drugs after leaving
home at age 16).
250. Id.
251. Id.
252. Id.
253. Id.

2017]

DESIGNED TO FAIL

439

with the jury—that Tom had contributed to her death.254 The judge
wrote in support of Tom’s clemency petition before President
Obama, explaining that the decision had haunted him ever since he
made it.255 Yet Nuttall writes that he had to give Tom “the same
answer I had to give so many of our clients; [t]hat as deeply unsatisfactory as it must sound, the Obama administration’s clemency
initiative was often a crapshoot.”256 Indeed many of those denied
seem indistinguishable from others who received their freedom.257
USA Today reporter Gregory Korte highlighted the puzzling case
of Harold and DeWayne Damper, two brothers from Mississippi who
“were indicted together, tried together, given the same sentence
and, until recently, served their sentences at the same minimumsecurity prison.”258 DeWayne had the more serious criminal record
(two prior felony convictions as opposed to Harold’s one), yet it was
DeWayne who got clemency and Harold who was denied.259
Deference to prosecutors had an additional dimension within the
clemency initiative as well. Not only do prosecutors in Washington
control the review process, but prosecutors in the office of conviction
are also consulted regarding each petition viewed as promising.260
Paul Larkin of the Heritage Foundation has written that in the
waning days of the Obama initiative, the reliance on local prosecutors was even enhanced.261
Deference to those prosecutors—the local ones who put the
defendant in prison for a long term in the first place—are written
into the Pardon Attorney’s internal standards in language taken
directly from the United States Attorney’s Manual, dictating that
“[t]he views of the United States Attorney are given considerable
254. Id.
255. See id. (noting the sentencing judge wrote a letter to the pardon attorney in support
of Tom’s clemency petition).
256. Id.
257. Gregory Korte, Two Brothers, Two Petitions for Clemency, Two Different Outcomes,
USA TODAY (Jan. 9, 2017, 6:47 PM), http://www.usatoday.com/story/news/politics/2017/01/
09/two-brothers-two-petitions-clemency-two-different-outcomes/96297020/ [https://perma.cc/
VZ36-AYXY] (noting that the Damper brothers were given different commutation decisions
despite the similarities of their cases).
258. Id.
259. See id.
260. See Morison, supra note 196, at 36-38.
261. See Paul J. Larkin, Jr., Delegating Clemency, 29 FED . SENT’G REP. 267, 267 (2017).

440

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

weight in determining what recommendations the Department
should make to the President.”262 Allowing local offices to have “considerable weight” in the reconsideration of sentences they sought
will inevitably put a thumb on the scale for denying requests.263 The
mindset in favor of the status quo can be seen in a letter to the Pardon Attorney from a United States Attorney, recommending denial
of clemency to a petitioner with a particularly sterling prison record:
[The] clemency application details what appears to be a remarkable turn-around during his time in custody. I cannot offer any
insight into the representations about his progress. Nor can I be
of much assistance in evaluating whether the changes, if true,
warrant clemency. “Jailhouse conversions” are not uncommon,
and I have no tools or experience with which to evaluate which
ones are genuine or may endure beyond the confines of imprisonment.264

The petition was denied.265 Within this one paragraph—and its
remarkable skew against freedom—rests the problem with deference to prosecutors, particularly within the context of the Constitution’s specific grant of the clemency power to the President.
In failing to listen to his first White House Counsel or the advice
of history to take clemency out of the hands of the Department,266
the President neglected to reach thousands of worthwhile petitioners for clemency.267 That failure cannot be blamed on Congress.
President Obama had the clear authority to do more. But his decision to delegate this task to the Department sealed its fate, whether
intended or not. While his clemency efforts resulted in more than a
thousand grants, it left many thousands more behind, including
those still serving 100-to-1 crack sentences.268 Those denied relief
262. Standards for Consideration of Clemency Petitioners, U.S DEP’T JUSTICE § 1-2.111,
https://www.justice.gov/pardon/about-office-0 [https://perma.cc/4G7R-VBVY].
263. Id.
264. Letter from Richard L. Durbin, Jr., U.S. Attorney, to Deborah Leff, Pardon Attorney,
Office of the Pardon Attorney, U.S. Dep’t of Justice (Aug. 30, 2015) (on file with authors).
265. See Commutations Denied by President Barack Obama, U.S. DEP’T JUSTICE,
https://www.justice.gov/pardon/obama-denials/commutations-denied-president-barack-obama
[https://perma.cc/JZF7-Z9BZ] (listing the names of persons whose petitions were denied).
266. See supra note 169 and accompanying text.
267. See supra note 247 and accompanying text.
268. See supra note 240 and accompanying text.

2017]

DESIGNED TO FAIL

441

have suffered two grave injustices. First, they received a disproportionately long drug sentence. Second, they had their hopes legitimately raised to believe they would get relief if they met the
Department’s criteria, only to find that their cases were denied for
some unexplained reason. In both cases, the decisions rested with
prosecutors,269 but the second one did not have to rest in prosecutorial hands, and it should not have been placed within an institution
that is structurally biased against second-guessing its own decisions. For the thousands who remain in prison, this choice reflected
a brutal timidity.
3. Compassionate Release
Clemency is not the only avenue for releasing people earlier than
the date their federal prison sentence is set to expire. When it
abolished parole in the federal system, Congress passed a law that
gives the Director of the BOP, who sits within the Department, the
authority to seek reductions in sentence for “extraordinary and
compelling reasons,”270 or what has more commonly been referred to
as compassionate release. Congress then vested authority in the
Sentencing Commission to “describe what should be considered extraordinary and compelling reasons for sentence reduction, including the criteria to be applied and a list of specific examples.”271 The
Commission’s initial list included defendants with terminal illnesses
and significant health problems.272 These factors are unfortunately
269. See Barkow, supra note 13, at 289-90 (describing how prosecutors have significant
roles in the current clemency system).
270. 18 U.S.C. § 3582(c)(1)(A)(i) (2012). The law specifies two grounds for filing a motion:
either “(i) extraordinary and compelling reasons warrant such a reduction” or (ii) the
defendant is serving a life sentence because he or she has a previous violent felony and is now
“at least 70 years of age, has served at least 30 years in prison, ... and a determination has
been made by the Director of the Bureau of Prisons that the defendant is not a danger to the
safety of any other person or the community.” Id. § 3582(c)(1)(A)(i)-(ii). In both cases, the
statute requires that the reduction be “consistent with applicable policy statements issued by
the Sentencing Commission.” Id. § 3582(c)(1)(A).
271. 28 U.S.C. § 994(t) (2012).
272. Until its recent amendments in 2016, the Commission’s list included defendants
suffering from a “terminal illness,” those “suffering from a permanent physical or medical
condition, or ... experiencing deteriorating physical or mental health because of the aging
process, that substantially diminishes the ability of the defendant to provide self-care within
the environment of a correctional facility and for which conventional treatment promises no
substantial improvement,” and situations involving “[t]he death or incapacitation of the

442

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

common occurrences for an increasingly aging federal prison
population,273 yet the BOP during President Obama’s time in office
rarely granted compassionate release. With a prison population
averaging around two hundred thousand people, the BOP granted
compassionate release to an average of only twenty-four people each
year during his first term in office.274 Even after critical reports of
the BOP’s practices from the Department’s independent Inspector
General (IG) prompted the Department to change some of its
compassionate release policies,275 the number of grants in the President’s second term remained relatively slight compared to the
overall population of elderly and sick individuals.276
This makes no sense if one were interested, as President Obama
said he was, in “evidence-based solutions” to address “massive levels
of incarceration [that] have not made us safer.”277 Elderly individuals and those with serious illnesses and health problems pose

defendant’s only family member capable of caring for the defendant’s minor child or minor
children.” U.S. SENTENCING GUIDELINES MANUAL § 1B1.13 cmt. n.1(A)(i)-(iii) (U.S. SEN TENCING COMM ’N 2015). The Commission also provided that the Bureau of Prisons could release individuals for other extraordinary and compelling reasons, making clear its list was not
exhaustive. See id. § 1.B.1.13 cmt. n.1(A)(iv).
273. People aged fifty and over are the fastest growing segment of the federal prison
population, increasing by 27 percent from 2009 to 2014. See Public Hearing on Compassionate
Release and Conditions of Supervision Before the U.S. Sentencing Comm’n 63-65 (2016)
[hereinafter U.S.S.C. Hearing], http://www.ussc.gov/sites/default/files/pdf/amendment-process/
public-hearings-and-meetings/20160217/Transcript.pdf [https://perma.cc/S28B-49FA] (statement of Michael E. Horowitz, Inspector General, United States Department of Justice).
Between 2009 to 2013, the number of incarcerated individuals age 65-69 increased by 41
percent and the number of individuals age 80 and older increased by 76 percent. OFFICE OF
THE INSPECTOR GEN ., U.S. DEP ’T OF JUSTICE , THE IMPACT OF AN AGING INMATE POPULATION
ON THE FEDERAL BUREAU OF PRISONS 2 (rev. ed. 2016), https://oig.justice.gov/reports/2015/
e1505.pdf [https://perma.cc/9T6H-5GGN].
274. See OFFICE OF THE INSPECTOR GEN ., U.S. DEP’T OF JUSTICE, I-2013-006, THE FEDERAL
BUREAU OF PRISONS’ COMPASSIONATE RELEASE PROGRAM 1 (2013), https://oig.justice.gov/
reports/2013/e1306.pdf [https://perma.cc/KQX2-R5GC].
275. See, e.g., OFFICE OF THE INSPECTOR GEN ., supra note 273, at 42 (discussing the working
group created to expand the use of compassionate release after the release of the 2013
compassionate release program report).
276. Between 2013 and 2015, 261 people received compassionate release. See Letter from
Michelle Morales, Acting Dir., Office of Policy & Legislation, U.S. Dep’t of Justice, to the Honorable Patti B. Saris, Chair, U.S. Sentencing Comm’n 17 (Feb. 12, 2016) [hereinafter Morales
Letter], http://www.ussc.gov/sites/default/files/pdf/amendment-process/public-hearings-andmeetings/20160217/DOJ.pdf [https://perma.cc/Q6WX-MVP3]. The Bureau received more than
3,100 requests for compassionate release during this time. Id.
277. Obama, supra note 2, at 819, 822.

2017]

DESIGNED TO FAIL

443

almost no public safety risk if they are released.278 Moreover,
releasing these individuals saves money that can be used for other
public safety initiatives.279 Elderly individuals and those with serious health ailments are the most expensive group of people to
incarcerate because of their increased medical needs.280 Costs for
medical care have exploded at the federal level, with medical care
expenses reaching $1.1 billion in fiscal year 2014, a 30 percent increase from the costs five years earlier.281 In fiscal year 2013, the
BOP spent a full 19 percent of its budget on incarcerating aging
individuals,282 and 17 percent on medical expenses.283
These empirical facts are why the IG “concluded that an efficiently-run compassionate release program combined with modifications to the program’s eligibility criteria could expand the pool of
eligible candidates, reduce overcrowding in the federal prison system, and result in cost savings for the BOP” without compromising
public safety.284 The IG issued two comprehensive reports on the
Department’s use of compassionate release and found it wanting in
both reviews.285 The first report, released in 2013, found that BOP’s
management and implementation of compassionate release “likely
278. People released pursuant to the BOP compassionate release program during the fiveyear period from 2006 to 2011 had a recidivism rate of 3.5 percent. OFFICE OF THE INSPECTOR
GEN ., supra note 273, at 49-50. Indeed, if one looks at all prisoners released over the age of
sixty, without even requiring an assessment of their physical health, the recidivism rate is
only 16 percent. See U.S. SENTENCING COMM ’N , RECIDIVISM AMONG FEDERAL OFFENDERS : A
COMPREHENSIVE OVERVIEW 23 fig.11 (2016), http://www.ussc.gov/sites/default/files/pdf/
research-and-publications/research-publications/2016/recidivism_overview.pdf [https://perma.
cc/CC8G-4QHW].
279. See TINA CHIU , VERA INST. OF JUSTICE, IT’S ABOUT TIME: AGING PRISONERS, INCREASING COSTS, AND GERIATRIC RELEASE 5 (2010), https://storage.googleapis.com/vera-web-assets/
downloads/Publications/its-about-time-aging-prisoners-increasing-costs-and-geriatric-release/
legacy_downloads/Its-about-time-aging-prisoners-increasing-costs-and-geriatric-release.pdf
[https://perma.cc/E6N9-7T7H] (discussing the expenses involved in caring for elderly inmates).
280. See id. (“[P]risons spend about two to three times more to incarcerate geriatric individuals than younger inmates.”).
281. See U.S.S.C. Hearing, supra note 273, at 64-65.
282. OFFICE OF THE INSPECTOR GEN ., supra note 273, at i.
283. Id. at 10.
284. Hearing on Compassionate Release and Conditions of Supervision Before the U.S. Sentencing Comm’n 3, 6 (2016), https://www.ussc.gov/sites/default/files/pdf/amendment-process/
public-hearings-and-meetings/20160217/IG.pdf [https://perma.cc/F8C2-4K7J] (statement of
Michael E. Horowitz, Inspector General, United States Department of Justice).
285. See generally OFFICE OF THE INSPECTOR GEN ., supra note 274 (discussing the BOP’s
Compassionate Release Program); OFFICE OF THE INSPECTOR GEN., supra note 273 (discussing
the effects of the aging inmate population on the BOP).

444

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

result[ed] in eligible inmates not being considered for release and in
terminally ill inmates dying before their requests were decided.”286
In the wake of the IG’s 2013 Report and as part of the Smart on
Crime Initiative, the Department formed a working group to address ways compassionate release could be improved.287 The group
ultimately “concluded that aging inmates do not pose a significant
public safety threat.”288 It also concluded the Department could
release individuals over the age of sixty-five “suffering from chronic
or serious medical conditions related to the aging process” who are
also “experiencing deteriorating mental or physical health that
substantially diminishes their ability to function in a correctional
facility,” and “for whom conventional treatment promises no substantial improvement to their mental or physical condition” if those
individuals had served half their initial sentence.289
The IG studied these new criteria for compassionate release in
2015 in its second major report on the subject, but still found the
Department falling short.290 The IG concluded that the Department
could lower the eligibility age from sixty-five to fifty for those suffering from chronic or severe health problems to save more money
and without compromising public safety.291 The IG also found fault
with another category of elderly people recognized as eligible for
compassionate release by the Department: people sixty-five and
older who “have served the greater of 10 years or 75 percent of their

286. OFFICE OF THE INSPECTOR GEN ., supra note 274, at i.
287. OFFICE OF THE INSPECTOR GEN ., supra note 273, at 42. The BOP’s complete list of
criteria are laid out in Program Statement 5050.49. See generally FED . BUREAU OF PRISONS,
U.S. DEP’T OF JUSTICE, No. 5050.49, CN-1, COMPASSIONATE RELEASE/REDUCTION IN SENTENCE:
PROCEDURES FOR IMPLEMENTATION OF 18 U.S.C. §§ 3582(C)(1)(A) AND 4205(G ) (2015), https://
www.bop.gov/policy/progstat/5050_049_CN-1.pdf [https://perma.cc/ZM7M-SUDC] (listing criteria for compassionate release eligibility). Its factors largely mirror the original criteria set
out in § 1B1.13 of the Guidelines, though in addition to recognizing the death of a caregiver
of a minor child to be an extraordinary circumstance, the BOP guidelines also “include the
incapacitation of an inmate’s spouse or registered partner when the inmate would be the only
available caregiver for the spouse or registered partner.” Id. at 7 (implementing Program
Statement 5050.49).
288. OFFICE OF THE INSPECTOR GEN ., supra note 273, at 47.
289. Id. at 43.
290. See id. at 46-50.
291. Id. at 48. The IG noted that its recommendation was consistent with the National
Institute of Corrections, which also “recommend[s] that inmates be considered aging” starting
at age fifty. U.S.S.C. Hearing, supra note 273, at 69.

2017]

DESIGNED TO FAIL

445

sentence.”292 The Bureau interpreted this latter provision to mean
that individuals had to serve both a minimum of 75 percent of their
sentence and at least ten years, thus rendering anyone with a
sentence of less than ten years ineligible for consideration.293 The IG
argued that this interpretation seemed contrary to the Department’s
concern with both overcrowding and public safety because it makes
almost half the elderly population ineligible. The Department’s
interpretation cut off those “likely to be the best candidates for early
release [because] ineligible inmates who received a shorter sentence
[of less than ten years] are more likely to have committed a less
serious offense, and present less danger to the public.”294 The IG
further noted the limited effect the Department’s 2013 changes were
having on grant rates. Of the 203 elderly people with medical
conditions who applied—thirty-three of whom had their requests
approved by the institution that housed them—none were ultimately approved for compassionate release by the BOP Director.295
And only two of the ninety-three elderly individuals who applied
without medical conditions received compassionate release.296
It is telling to compare the IG’s assessment of compassionate
release with the way the Department sees it. Whereas the IG
emphasized—based on empirical research—that a focus on elderly
and sick people in prison who no longer pose a threat to society
could ease prison overcrowding and save some of the BOP’s taxed
resources, the Department refused to accept the premise that the
release of elderly and seriously ill people should be seen as part of
the effort “to reduce prison overcrowding or the prison population
generally.”297 In the Department’s view, “These objectives are
addressed by the Department’s support for lowered guideline
penalties for drug trafficking offenses, the Department’s Smart on
Crime charging policy, by the Clemency Initiative, and by the
Department’s support for legislation currently pending before the
United States Congress.”298 It is hard to understand why those other
292.
293.
294.
295.
296.
297.
298.

OFFICE OF THE INSPECTOR GEN ., supra note 273, at 7.
Id. at 46.
Id. at 49.
Id. at 45 tbl.9.
Id.
Morales Letter, supra note 276, at 15.
Id.

446

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

initiatives—which we have already explained were pursued with
less than full effort by the Department—would somehow obviate the
need to have an additional program dealing with a growing segment
of the prison population that poses the unique combination of high
costs and a low risk to public safety. Moreover, none of the other
initiatives highlighted by the Department address this population,
so it is odd to see them as somehow replacing a more robust use of
a mechanism that Congress put in place specifically to address
changed circumstances of particular individuals.
The Department’s narrow view is all the more difficult to
understand in light of the fact that legislative history highlighted
terminal illness as one of the examples of a changed circumstance
that would be extraordinary and compelling.299 Furthermore, the
text of the statute itself provides, as one eligible category, elderly
people serving life sentences because of a violent criminal record
who no longer pose a threat.300 This indicates that Congress expected age to be a key factor and provided for it as a grounds for
early release, even in cases in which defendants had violence in
their past.301 If the statute and its history are not clear on this issue,
they are at a minimum open to an interpretation that sees this
provision more broadly than the Department does. With a prison
population busting at the seams and a budget straining under high
medical costs, one would think that broader interpretation would
make ample sense to a Department working for a President committed “to highlight[ing] the compelling economic and policy arguments for justice reform as well as the human toll of a failing
system.”302 Keeping people with failing health makes no sense as a
matter of economics or public safety, and the human toll of not
releasing them is considerable.
The Department did not just disagree with the IG’s assessment;
it also found itself at odds with the Sentencing Commission.
Whereas the Department required individuals to demonstrate a
prognosis of death within eighteen months or less to show a terminal illness,303 the Commission agreed with the medical community’s
299.
300.
301.
302.
303.

See S. REP. NO. 98-225, at 121 (1983), as reprinted in 1984 U.S.C.C.A.N. 3182, 3304.
18 U.S.C. § 3582(c)(1)(A)(ii) (2012).
See id.
Obama, supra note 2, at 821-22.
See, e.g., OFFICE OF THE INSPECTOR GEN ., supra note 274, at ii (stating the Depart-

2017]

DESIGNED TO FAIL

447

assessment that because such predictions were not made by doctors,
using that timeline impeded individuals’ ability to get compassionate release.304 Instead, the Commission defined a terminal illness
consistent with the medical testimony that it should be “a serious
and advanced illness with an end of life trajectory” and that “[a]
specific prognosis of life expectancy (i.e., a probability of death within a specific time period) is not required.”305 Additionally, while both
the Department and the Commission agreed that individuals with
serious health problems presented extraordinary and compelling
circumstances, there were key differences. First, the Commission
would allow anyone suffering from serious health problems to seek
compassionate release,306 whereas the Department made only those
over sixty-five eligible.307 Second, the Commission did not require
individuals to serve a minimum prison term of any length.308 In contrast, the Department’s program statement concluded that people
“should serve a minimum of 50 percent of the sentence to justify the
resources that the Department spent to prosecute the inmate.”309
Finally, the Commission included those with “serious functional or
cognitive impairment” among those who are eligible for medical
reasons.310
The two agencies also differed in how they treated the elderly.
The Sentencing Commission agreed with the IG that those sixty-five
and older311 should be eligible for release if they serve either ten
years or 75 percent of their sentence, whichever is less, thus making
those with sentences of less than ten years eligible.312 Finally, the
ment’s intent to “expand the compassionate release program by making inmates with a life
expectancy of up to 18 months eligible for consideration”); U.S. SENTENCING COMM ’N , supra
note 78, at 4 (emphasizing that the BOP only considers a terminal illness to be serious if the
inmate’s life expectancy is eighteen months or less).
304. See PUBLIC MEETING MINUTES, supra note 137, at 4.
305. U.S. SENTENCING GUIDELINES MANUAL § 1B1.13 cmt. n.1(A)(i) (U.S. SENTENCING
COMM ’N 2016).
306. See id. cmt. n.1 (“Provided the defendant meets the requirements ... extraordinary and
compelling reasons exist under any of the circumstances set forth below.”).
307. OFFICE OF THE INSPECTOR GEN ., supra note 273, at 43.
308. See U.S. SENTENCING GUIDELINES MANUAL § 1B1.13.
309. OFFICE OF THE INSPECTOR GEN ., supra note 273, at 43.
310. U.S. SENTENCING GUIDELINES § 1B1.13 cmt. n.1(A)(ii)(II).
311. One of the commissioners would have lowered this requirement to age sixty based on
the testimony of medical experts and the Commission’s recidivism data. PUBLIC MEETING
MINUTES, supra note 137, at 5.
312. The Commission did not allow age alone to make someone eligible; the person addi-

448

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

Commission sought to “encourage[ ] the Director of the Bureau of
Prisons to file such a motion if the defendant meets any of the
circumstances” and reminded the Bureau that “[t]he court is in a
unique position to determine whether the circumstances warrant a
reduction.”313
Why is the Department so much more reluctant than the IG or
the Sentencing Commission when it comes to reductions in sentence
for extraordinary and compelling circumstances, especially when its
position is seemingly at odds with President Obama’s stated goals?
The Department’s comments before the Sentencing Commission
hold the answer. It told the Commission that it believed Congress
intended to create “a system whose fundamental premise is that offenders should serve most of the sentences imposed by the courts.”314
That view explains both the limited clemency initiative and the
cramped view of compassionate release—even though both clemency
and compassionate release are also part of the “system.”
The Department further noted that the Bureau should be
“cautious” in granting compassionate release because individuals
over fifty commit a range of offenses (listing among the notable
offenses drug trafficking and fraud) and adding that 36.3 percent of
them have a criminal history category of IV or higher.315 It further
added that “BOP data also suggests that older inmates convicted of
drug offenses cannot always be described as low-level participants.”316 The Department, in other words, does not care much
about what extraordinary and compelling circumstances have developed in an individual’s life—whether it is a terminal illness or
incurable and serious health problem; the Department cares about
the initial offense. If that offense is serious enough—with seriousness assessed by the Department—it does not see a reason to revisit
the sentence. This view flies in the face of the empirical evidence
amassed by the IG about recidivism rates for the sick and the elderly and the broader goals of conserving limited prison resources.317
tionally had to show he or she was experiencing a serious deterioration in his or her health
because of his or her age. See U.S. SENTENCING GUIDELINES MANUAL § 1B1.113 cmt. n.1(B).
313. Id. cmt. n.4.
314. U.S.S.C. Hearing, supra note 273, at 30 (statement of Jonathan Wroblewski, Principal
Deputy Assistant Att’y Gen., United States Department of Justice).
315. Morales Letter, supra note 276, at 18.
316. Id. at 18-19.
317. See supra notes 278-83 and accompanying text (discussing the results of the IG’s 2013

2017]

DESIGNED TO FAIL

449

But it is entirely consistent with how the Department approached
just about every criminal justice reform issue involving federal
offenders. The Department allowed for a limited category of relief
for what it viewed as the lowest level offenders with the most minimal criminal history, but reaffirmed the status quo for everyone
else—no matter how disproportionate their sentences, how extraordinary their life circumstances today, or even how much views on
sentencing or drug policies have themselves changed over the
years.318
4. Forensic Science
Throughout his time in office, President Obama embraced the
importance of science,319 including “a wide range of research and
policy initiatives to strengthen the forensic sciences.”320 Yet when
Obama reviewed this issue at the end of his time in office,321 he
neglected to mention that the Department rejected the findings and
recommendations of a report he commissioned from his scientific
advisors, and he did not demand that it take a different action based
on the report.
The President had asked the President’s Council of Advisors on
Science and Technology (PCAST) in 2015 to consider “whether there
are additional steps” that could usefully be “taken on the scientific
side” to strengthen the forensic science disciplines and “ensure the

and 2015 reports on compassionate release).
318. Compassionate release was not the only area where the BOP seemed to ignore
empirical evidence. The American Bar Association has complained that the BOP lags behind
state systems in developing innovative programs, is haphazard in preparing people for
release, and offers unduly limited drug treatment options for people in prison. See Letter from
Bruce Green, Chair, ABA Criminal Justice Section, to Eric Holder, Attorney Gen., U.S. Dep’t
of Justice (May 6, 2011) (on file with authors). The Obama Administration did not address
these issues until November 2016, when it announced a plan to enhance educational
opportunities in prisons and to improve the standards for halfway houses under contract to
the BOP. See Obama, supra note 2, at 830-32. Because these initiatives were announced so
late—after President Trump’s election—the Obama Administration never implemented them,
and it remains to be seen whether they will survive.
319. See Gardiner Harris, Obama to Leave the White House a Nerdier Place than He Found
It, N.Y. TIMES (July 31, 2016), https://www.nytimes.com/2016/08/01/us/politics/obama-to-leavethe-white-house-a-nerdier-place-than-he-found-it.html?_r=0 [https://perma.cc/6RHM-B2TF].
320. Obama, supra note 2, at 860.
321. See id. at 862.

450

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

validity of forensic evidence used in the Nation’s legal system.”322
This request came on the heels of a 2009 report by the National
Academy of Sciences (NAS) that found deficiencies in numerous
forensic methods.323 The NAS report concluded that aside from DNA
analysis, “no forensic method has been rigorously shown to have the
capacity to consistently, and with a high degree of certainty,
demonstrate a connection between evidence and a specific individual or source.”324 The NAS then called for the creation of an
independent federal agency that would oversee scientific research
to establish the validity of various forensic methods, national testing
standards, and the certification of forensic labs.325 The NAS pushed
for an independent agency because it believed any agency responsible for forensics needed “a culture that is strongly rooted in science”
and not “principally beholden to law enforcement.”326 The NAS
expressly cautioned that the FBI and National Institute of Justice
were ill-suited to lead the way on reform because they were housed
within the Department and thus “part of a prosecutorial department
of the government,” making them susceptible “to subtle contextual
biases that should not be allowed to undercut the power of forensic
science.”327 In the NAS’s view, the Department could not be in
control of forensic science because its “principal mission is to enforce
the law” and “[t]he potential for conflicts of interest between the
needs of law enforcement and the broader needs of forensic science
are too great.”328
The Department resisted the NAS findings and its calls for
reform.329 Instead of the creation of an independent entity, the
322. PRESIDENT’S COUNCIL OF ADVISORS ON SCI. & TECH ., EXECUTIVE OFFICE OF THE
PRESIDENT, REPORT TO THE PRESIDENT: FORENSIC SCIENCE IN CRIMINAL COURTS : ENSURING
SCIENTIFIC VALIDITY OF FEATURE-COMPARISON METHODS x (2016) [hereinafter PCAST], https:
//obamawhitehouse.archives.gov/sites/default/files/microsites/ostp/PCAST/pcast_forensic_
science_report_final.pdf [https://perma.cc/HWX6-EEDD].
323. See NAT’L ACAD . OF SCIS., STRENGTHENING FORENSIC SCIENCE IN THE UNITED STATES
4-7 (2009) [hereinafter NAS].
324. Id. at 7.
325. Barkow, supra note 13, at 301.
326. NAS, supra note 323, at 17-18.
327. Id. at 16.
328. Id. at 17.
329. Barkow, supra note 13, at 303; Erin Murphy, What “Strengthening Forensic Science”
Today Means for Tomorrow: DNA Exceptionalism and the 2009 NAS Report, 9 LAW , PROBABILITY & RISK 7, 14 n.50 (2010) (suggesting that the Department’s research division “actively

2017]

DESIGNED TO FAIL

451

Department partnered with a technology division within the Commerce Department to create a National Commission on Forensic
Science in 2013.330 In 2015, the Department took up recommendations by that Commission to require its labs to obtain accreditation
and mandate that its prosecutors use only accredited labs within the
next five years.331
The President’s request to PCAST came as a follow-up to this
effort, and PCAST found that much more still needed to be done. Its
analysis was focused on “‘feature-comparison’ methods—that is,
methods that attempt to determine whether an evidentiary sample
(e.g., from a crime scene) is or is not associated with a potential
‘source’ sample (e.g., from a suspect), based on the presence of similar patterns, impressions, or other features in the sample and the
source.”332 This kind of forensic methodology would include commonly used methods for analyzing DNA, fingerprints, ballistics,
bitemarks, handwriting, shoeprints, tire tracks, and hair samples.333
PCAST conducted a thorough review of a range of pattern matching
disciplines and found many “lack[ed], any scientific support for their
basic assumptions,”334 while others rested on thin scientific foundation and were prone to gross exaggeration. PCAST thus urged
scientific research to establish the validity of these methods—what
it called foundational validity—where feasible; once that was established, it advocated for clear limits of validity as applied which
would ensure the reliability of any testimony offered.335 PCAST’s
concern here was with overclaiming, and it emphasized that
“[s]tatements claiming or implying greater certainty than demonstrated by empirical evidence are scientifically invalid.”336 Radley
Balko helpfully compares DNA sample interpretation with these
other methods to show why the latter are lacking. Someone
opposed congressional efforts to study forensic sciences through a research entity such as the
National Academy of Sciences rather than a law enforcement entity such as itself ”).
330. Obama, supra note 2, at 860.
331. Id. As President Obama noted, the FBI also undertook a review of more than twentyone thousand cases involving the use of microscopic hair analysis. Id. at 861.
332. PCAST, supra note 322, at 1.
333. Id.
334. Radley Balko, When Obama Wouldn’t Fight for Science, WASH . POST (Jan. 4, 2017),
https://www.washingtonpost.com/news/the-watch/wp/2017/01/04/when-obama-wouldnt-fightfor-science/?utm_term=.65894f81a862 [https://perma.cc/B8U2-GXLE].
335. See PCAST, supra note 322, at 6.
336. Id.

452

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

analyzing a DNA sample “can give precise calculations of the odds
of the suspect being the person whose DNA was left at the crime
scene.”337 In contrast, other feature comparison methods lack a standard methodology or reference set for interpretative purposes, and
have much less scientifically established rates at which characteristics are distributed across a population. Thus, when an expert
testifies, “we don’t know the rate at which those characteristics are
distributed across the population,” this makes it hard to know how
probative it is that there is a match between the defendant and the
sample from the crime scene.338 If thousands of people have the
same kind of hair microscopy, then the presence of a match might
not tell us very much. Anyone with even a passing familiarity with
science—indeed, with common sense—would recognize this as a
significant limitation to the use of this kind of evidence.
PCAST, unsurprisingly, could not have been clearer that the
Department’s response to the 2009 NAS report was deficient because it failed to fill this gap in empirical knowledge. The PCAST
report cautioned “that neither experience, nor judgment, nor good
professional practices (such as certification programs and accreditation programs, standardized protocols, proficiency testing, and codes
of ethics) can substitute for actual evidence of foundational validity
and reliability.”339 It noted that
an expert’s expression of confidence based on personal professional experience or expressions of consensus among practitioners about the accuracy of their field is no substitute for error
rates estimated from relevant studies. For forensic featurecomparison methods, establishing foundational validity based on
empirical evidence is thus a sine qua non. Nothing can substitute for it.340

Despite the fact that the PCAST report was authored by nineteen
preeminent scientists, that its logic and grounding in scientific
methods is irrefutable, that it was commissioned by the President,
and that its results were touted in a press release by the White
337.
338.
339.
340.

Balko, supra note 334.
Id.
PCAST, supra note 322, at 6.
Id.

2017]

DESIGNED TO FAIL

453

House,341 the Department simply refused to accept it. Attorney
General Loretta Lynch curtly and quickly responded to PCAST’s
release that, although the Department “appreciate[s] their contribution to the field of scientific inquiry, the [D]epartment will not be
adopting the recommendations related to the admissibility of
forensic science evidence.”342 The Department’s refusal to accept
bedrock scientific analysis is troubling enough as applied to the
federal system, but it is all the more disconcerting because forensic
science is an area in which the federal government often takes on a
leadership role that influences the states. So the message the
Department sent was loud and clear: whatever the science says, we
find this evidence valuable to our law enforcement agenda, so we
will continue to use it.343 The Department likely believed that doing
so was the right thing because of its experience using this evidence
in cases where its prosecutors believed defendants to be guilty, thus
validating the value and utility of these methods for the Department, whatever the empirical evidence might say.344
Balko, who has written several probing articles on forensic
science, sums up well the frustration with this outcome. He notes
that President Obama “deserves credit for” calling attention to criminal justice reform.345 But then he laments that “[all] too often those
statements have been followed by policy changes that have been
341. See Eric Lander et al., PCAST Releases Report on Forensic Science in Criminal Courts,
WHITE HOUSE: BLOG (Sept. 20, 2016, 5:59 AM), https://obamawhitehouse.archives.gov/blog/
2016/09/20/pcast-releases-report-forensic-science-criminal-courts [https://perma.cc/GZ58D75M].
342. Jordan Smith, FBI and DOJ Vow to Continue Using Junk Science Rejected by White
House Report, INTERCEPT (Sept. 23, 2016, 2:10 PM), https://theintercept.com/2016/09/23/fbiand-doj-vow-to-continue-using-junk-science-rejected-by-white-house-report/ [https://perma.cc/
JK2N-G8GB]. The National District Attorneys Association and police organizations also dismissed the report. See Balko, supra note 334.
343. Even when the Department endorses best practices, as it recently did with eyewitness
testimony, it is careful to add a caveat that if it fails to follow best practices, it should not
suffer any consequences. See Memorandum from Sally Q. Yates, Deputy Attorney Gen., U.S.
Dep’t of Justice, to Heads of Department Law Enforcement Components 2 (Jan. 6, 2017),
https://www.justice.gov/file/923201/download [https://perma.cc/4PZM-TNE3] (“[N]othing in
this memorandum implies that an identification not done in accordance with these procedures
is unreliable or inadmissible in court.”).
344. See Barkow, supra note 13, at 313 (“Prosecutors may therefore place greater faith in
existing forensic science methods than empirical evidence would justify because they have
used this information in cases where they believed the defendant to be guilty.”).
345. Balko, supra note 334.

454

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

largely symbolic.”346 Balko’s target was the response to forensic
reform, but he could have been describing any of the previous issues
discussed in this Part. While the Administration gave strong statements about various issues and did take some concrete steps to
improve policies, the measures were limited and inadequate because
the President did not back up his rhetoric by making sure things
were appropriately implemented, and left it up to the Department
to decide what should really happen. And all too often, nothing
much did.
B. The Department of Justice’s More Aggressive Pursuit of State
Reforms
Tellingly, there were areas where the Department advocated for
more robust reforms. Their common thread was that they did not interfere with the Department’s law enforcement mission, or, in fact,
assisted it.
It is in the Department’s interest as a law enforcement agency
charged with preventing police violence to do its part to remedy the
underlying conditions that create this violence.347 It is not surprising, then, that the Department would be vigilant in pursuing cases
and reforms that help to deter future cases of excessive police
violence. In the course of its investigation into the Ferguson Police
Department after the death of Michael Brown, for example, the
Department found “that revenue generation is stressed heavily
within the police department” and that “[p]atrol assignments and
schedules are geared toward aggressive enforcement of Ferguson’s
municipal code, with insufficient thought given to whether enforcement strategies promote public safety or unnecessarily undermine
community trust and cooperation.”348 This caused the officers to see
the citizens they policed “less as constituents to be protected than
as potential offenders and sources of revenue,” which undermined

346. Id.
347. See Civil Rights Division, About the Division, U.S. DEP’T JUSTICE, https://www.justice.
gov/crt/about-division [https://perma.cc/ZL9C-8Q6M].
348. CIVIL RIGHTS DIV., U.S. DEP’T OF JUSTICE, INVESTIGATION OF THE FERGUSON POLICE
DEPARTMENT 2 (2015), https://www.justice.gov/sites/default/files/opa/press-releases/attach
ments/2015/03/04/ferguson_police_department_report.pdf [https://perma.cc/94HM-65NS].

2017]

DESIGNED TO FAIL

455

police-community relations and in turn created a culture that breeds
excessive force.349
The result is that the Department was at the forefront in addressing excessive fines and fees in state and local systems.350 Doing
so furthered the Civil Rights Division’s mission and did nothing to
threaten federal practice. Arguing against localities seeking to
extract fines and fees from impoverished defendants posed no
conflict with the Department’s mission because unlike these
municipalities, the Department has no need to use the criminal
process to help keep its budget afloat.351
The same is true of the Department’s efforts to spur states to enact sentencing reforms. The Department could urge states to expand
parole eligibility or good time credits because it was not bound by
the reforms it was advocating.352 The Department can resist
sentencing reductions and retroactive relief at the federal level
while asking states to go bolder because one situation involves its
own self interest and the other does not. In the latter context, the
reform has no effect on the Department itself, but in the former, it
does. Likewise, the President’s Data-Driven Justice Initiative and
Police Data Initiative could ask state and local jurisdictions to
reduce jail stays and adopt other cost-effective solutions because
none of those requests applied at the federal level.353
The Department can also support reforms that apply to other
parts of the federal government as long as those changes do not
require it to change its own practices. It took part in an Obama
cabinet-level working group on reentry and endorsed reforms that
made it easier for people with criminal records to get jobs, housing,
and education.354 Notably, while the Department helped other departments figure out how to change their policies, it did not do much
349. Id.
350. Obama, supra note 2, at 843-44.
351. Similarly, no conflict occurred when the Obama Administration decided to stop selling
certain military equipment to local police departments. Gregory Korte, Obama Bans Some
Military Equipment Sales to Police, USA TODAY (May 18, 2015, 5:04 PM), https://www.
usatoday.com/story/news/politics/2015/05/18/obama-police-military-equipment-sales-newjersey/27521793/ [https://perma.cc/YY5R-TQPQ].
352. Cf. Obama, supra note 2, at 846 (describing these efforts).
353. Cf. id. at 848-51 (discussing how these programs help the Department advance reform
at the state and local levels).
354. See id. at 833-34, 853.

456

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

within its own Department on reentry issues. Although some U.S.
Attorneys fostered innovative reentry practices within their districts, the Department itself did little to spearhead changes in
prosecution practices that would aid reentry other than asking each
office to put a reentry coordinator in place and trying out pilot
programs to divert small numbers of people (typically fewer than a
dozen in a given district) who presented the most sympathetic cases
in which the initial charging decision was itself questionable. Even
in those cases, it was typically federal judges pushing for the diversion efforts with prosecutors reluctantly agreeing to them.
III. REDESIGNING THE EXECUTIVE BRANCH AND USING THE
APPOINTMENT POWER FOR INSTITUTIONAL REFORM
President Obama’s results in the realm of criminal law did not
match his stated ambitions, even in areas completely under his
control that did not require congressional cooperation. President
Obama may have agreed with the Department that a more modest
approach was preferable, despite his broader proclamations. But if
a future President wants to achieve more, he or she needs to create
new structures that allow decision-making and oversight to be more
independent of the Department. Similar constructs have been
created in the area of national security (in which the National
Security Advisor floats as an independent advisor between the
security agencies and the President),355 foreign policy (in which the
Secretary of State is just one of many advisors weighing in on
crucial questions),356 and trade (in which the U.S. Trade Representative operates independently of the Department of Commerce).357
Examples exist, too, of Presidents creating important advisory bod355. See ALAN G. WHITTAKER ET AL., THE NATIONAL SECURITY POLICY PROCESS: THE NA7-10 (rev. ed. 2011), http://www.landon
rowland.com/NSC-Report-2007.pdf [https://perma.cc/CJ2N-CQSW].
356. See id. (describing role of Secretary of State within the National Security Council).
357. See Mission of the USTR, OFF. U.S. TRADE REPRESENTATIVE, https://ustr.gov/aboutus/about-ustr [https://perma.cc/L3SU-H799]. The White House Counsel does not serve this
function; most of her tasks are not connected to criminal justice. Nor does the Domestic Policy
Council, which is primarily focused on other issues, and thus naturally defers to the specialists at the Department. See Domestic Policy Council, WHITE HOUSE, https://obamawhitehouse.
archives.gov/administration/eop/dpc [https://perma.cc/VZ2W-P5K3]. Moreover, the White
House Counsel and Domestic Policy Council, as generalists, are not well suited to the task of
collecting and analyzing data about specific substantive areas.
TIONAL SECURITY COUNCIL AND INTERAGENCY SYSTEM

2017]

DESIGNED TO FAIL

457

ies by executive order, such as the “czars” for the environment and
Ebola response.358 There is even some precedent in criminal law, as
noted, because President Ford created a body independent of the
Department to advise him on clemency with respect to people who
had dodged the draft.359 In all these spaces, the White House recognized that it needed more independent assessments to counterbalance what it was hearing from agencies that might be biased by
their primary mission and not see the bigger picture.
While the problem of institutional bias is not unique to the
Department, there are additional features of the Department that
make it particularly important for the President to hear from a
more independent source on criminal justice policy. First, the
Department, unique among federal agencies, serves as the President’s legal advisor, and the President turns to it to support the
legality of his or her actions. The President may thus be particularly
reluctant to second-guess its views on criminal justice issues
because of a general deference to its legal judgments. Second, the
White House might be hesitant to act contrary to the Department’s
views if the Department’s objections to an alternative view are
grounded in public safety arguments.360 Without an independent
analysis of the costs and benefits—including a risk assessment—the
President may be even more likely to side with the Department
than he or she would be to side with other agencies.
No one path can correct for this, and it is beyond the scope of this
Article to provide all the options or all the details. Rather, our goal
is to provide two main paths available to a President interested in
getting less biased advice about criminal justice issues. Our focus is
on those things the President can do without congressional help, so
although it might be advisable to move the BOP and forensic re-

358. See, e.g., Jack Healy et al., Obama May Name “Czar” to Oversee Ebola Response, N.Y.
TIMES (Oct. 16, 2014), https://www.nytimes.com/2014/10/17/us/obama-may-name-ebola-czarafter-amber-joy-vinson-flight-causes-concern.html?_r=0 [https://perma.cc/U5UM-48LZ].
359. See supra notes 209-11 and accompanying text.
360. See, e.g., supra note 156 and accompanying text (describing how President Obama
deferred to the Department’s views on mens rea reform).

458

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

search outside of the Department entirely,361 that would require
statutory changes.362
One possible option is for the President to create a single Criminal Justice Advisory Commission (which would include Department
representation but would be comprised of other viewpoints) within
the Executive Office of the President that could advise the President
on criminal justice issues. This is akin to the “czar” model the President has followed in other contexts363 and would still give the President confidential advice, but with a commission structure so that
the President could receive a diversity of viewpoints. This Commission could advise the President on all the policy issues discussed,
thus providing a viewpoint separate from (but informed by) the
Department. Once created, this body could also take over the task
of reviewing clemency applications in light of the President’s policy
preferences, thus replacing the function of the Pardon Attorney.364
A second path would be to create separate bodies, one that
advises on policy questions and another that focuses on clemency
decisions. Under this option, in addition to the Criminal Justice
Advisory Commission, there would be something like a Presidential
Clemency and Reentry Board that could replace the Pardon Attorney and the seven-level, Department-centered review currently employed.365
Under either of these models, a President could also create a
process whereby the policy decisions endorsed by the Department
get reviewed by either the Office of Information and Regulatory
Affairs or the Criminal Justice Advisory Commission to make sure
361. For such an argument, see Barkow, supra note 13, at 335-41. No state follows the federal model of housing its corrections agency within the prosecutor’s office; they all developed
independent prison commissions or bureaus by 1929. See Federal Penal and Reformatory
Institutions: Hearing on H.R. Res. 233 Before the Spec. H. Comm. on Fed. Penal & Reformatory Insts., 70th Cong. 80-81 (1929) (statement of James Bennett, United States Bureau of
Efficiency).
362. Barkow, supra note 13, at 280, 295 (describing the statutory framework placing these
functions in the Department).
363. See supra note 358 and accompanying text.
364. If the body were to perform clemency review, then we believe it is in the President’s
interest to make bipartisan appointments to it so that it provides the President with a degree
of political cover for his decisions to reduce sentences and issue pardons. See Rachel E.
Barkow & Mark Osler, Restructuring Clemency: The Cost of Ignoring Clemency and a Plan
for Renewal, 82 U. CHI. L. REV. 1, 22 (2015).
365. See supra Part II.A.2.

2017]

DESIGNED TO FAIL

459

the proposed decisions are justified under traditional cost-benefit
analysis.
Finally, apart from these structural reforms, the President can
also further criminal justice reform with the use of his or her
appointment power, something President Obama did in practice but
spent little time highlighting in his reflections on his own accomplishments.
A. Presidential Criminal Justice Advisory Commission
The problems inherent in the Department’s dual role as a law
enforcement agency and as advisor to the President, particularly for
a reform-minded president, were made clear in the Obama years.
One way to allow the President strong and broad advice on criminal
law issues would be to form an advisory commission that would
include Department representation, while remaining independent
from the Department. This body should include individuals from
academia who specialize in criminal justice and criminology, policy
leaders on criminal justice reform, corrections officials and experts
who study reentry and recidivism reduction, those who work in law
enforcement, attorneys who work in criminal defense, and formerly
incarcerated people who can speak to their experiences while incarcerated and during reentry.366 Because federal programs often are
directed at state and local criminal justice groups, representation
from outside the federal system would be particularly helpful.
The Department would need to implement any of the President’s
policies, so it is important to give it a seat at the table to understand
its concerns on any given policy. But whereas now those implementation concerns often drive the Department to reject suggestions
outright, this new structure would allow the policies to be evaluated
based on all their costs and benefits, and implementation issues
would be but one of many issues under consideration. It is also
important for any such advisory body to craft policies to facilitate
Department compliance because the resistance to reform discussed
in Part I would not disappear simply because policies change. Thus,
366. For a discussion of the kinds of representation to include, see Rachel E. Barkow,
Administering Crime, 52 UCLA L. REV. 715, 772, 778, 783 (2005); Barkow & Osler, supra note
364, at 21-22; and P.S. Ruckman, Jr., The Obama Administration: Breaking Records in a
Broken Clemency System, 29 FED . SENT’G REP. 87, 90 (2017).

460

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

any such body should have a preference for clear rules instead of
vague standards that can be more easily evaded, because prosecutors are likely to dutifully follow clear rules. For example, if the
Commission were to offer a framework for the BOP to follow on
compassionate release, it should set a clear timetable for the BOP
to make decisions, set out clear eligibility criteria, and audit the
BOP’s decision-making for compliance.
The Criminal Justice Advisory Commission should also be set up
to be data-driven in focus, with a mission to promote public safety
in the least costly manner based on empirical evidence. This focus
on data will allow the President to better assess any Department
objections to a proposal based on the Department’s views about how
those reforms will affect public safety. The Commission will need a
staff to help with this analysis, though some of this could be done in
partnership with other agencies. The Sentencing Commission, for
example, can provide the Administration with a great deal of
information. Research institutions outside of government could also
assist with specific tasks, as criminology departments and law
schools already investigate many of the issues this body would
consider.
One would assume that the problems with reforming forensic
science367 would have been handled much differently if such a commission were in place, acting as a buffer between the self-interested
perspective of the Department and the consensus view of the
scientific community that forensic methods require validation and
more careful use in the courts.368 The BOP’s approach to compassionate release and the Department’s position on sentencing policy
may have also been different if it had first been analyzed by such a
commission—just as the independent Sentencing Commission
reached different conclusions than the Department after its analysis
of the data and research on recidivism and costs.369
It is also possible that the President, in the end, may have sided
with the Department on some of these issues even after hearing a
wider set of views. But there is no question that, on net, decision-

367. See supra Part II.A.4.
368. See supra notes 322-28 and accompanying text.
369. See supra Part II.A.1.

2017]

DESIGNED TO FAIL

461

making on criminal justice will be improved overall if the President
has full information before deciding which policies to endorse.
B. Presidential Clemency Board
While the Criminal Justice Advisory Board could also be tasked
with setting clemency standards and reviewing petitions, another
model would be to separate this work and give it to a different body
because it involves individual case processing. This entity would
handle clemency issues but coordinate with the Criminal Justice
Advisory Board on recidivism and reentry policy issues—as those
are relevant to clemency—as well as on prosecution practices to help
identify outlier sentences and practices in specific districts where
clemency could provide a unifying corrective check.370
The board model for clemency has already proven effective in the
states. It is not by accident that no state has embraced371 the two
key attributes of federal clemency: endless review and a central role
for prosecutors.372 Margaret Colgate Love has identified fourteen
states that demonstrate well-functioning systems that provided
“frequent and regular” pardon grants: Alabama, Arkansas, Connecticut, Delaware, Georgia, Idaho, Illinois, Iowa, Nebraska, Nevada,
Oklahoma, Pennsylvania, South Carolina, and South Dakota.373
This mix of red and blue states shows that it is not politics that
matters, it is process. And we know what kind of process works, too.
Five out of the six states where pardon decisions are made by highly
independent boards—Alabama, Georgia, Idaho, Connecticut, and
South Carolina—are also among the fourteen members of the “frequent and regular” list.374 Moreover, in each of the other states with
high-functioning systems, a board plays a significant role in deci370. See Barkow & Osler, supra note 364, at 22-25.
371. The Criminal Justice Policy Foundation maintains an online database for state clemency procedures. See State Clemency Guide, CRIM . JUST. POL’Y FOUND., https://www.cjpf.org/
state-clemency/ [https://perma.cc/ME8G-GNHL]. Margaret Colgate Love created a similar site
for state pardon procedures in conjunction with the National Association of Criminal Defense
Lawyers. See Margaret Colgate Love, Restoration of Rights Project, NAT’L ASS’N CRIM . DEF.
LAW ., https://www.nacdl.org/rightsrestoration/ [https://perma.cc/HY6N-LS79].
372. See supra Part II.A.2.
373. Margaret Colgate Love, Reinvigorating the Federal Pardon Process: What the President
Can Learn from the States, 9 U. ST. THOMAS L.J. 730, 755-66 (2012).
374. Id. at 756-59, 766. The sixth such jurisdiction, Utah, largely misses the cut because
the board in that state only gets three to five requests for pardons a year. Id. at 767.

462

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

sion-making.375 There is a clear lesson from the states: use clemency
boards, not redundant panels of prosecutors, if you want regular
and rational clemency grants to be a part of your system of criminal
law.
An example of this dynamic exists within the federal system, too:
the Presidential Clemency Board used by President Ford to identify
over thirteen thousand recipients of clemency in the wake of the
Vietnam War.376 Though the Ford Clemency Board was temporary
(lasting just one year),377 the structure he created could be replicated
as a permanent institution. That board featured a bipartisan and
diverse membership, which included representatives from academia, prominent members of the private bar, religious leaders,378 and
importantly—given its subject—representatives of the military.379
Beyond its diversity, though, the Ford Clemency Board had a related advantage that is equally distinctive relative to the current
federal clemency process: independence from the Department.380
While a clemency review board should include a representative from
the Department, it should be established outside of the Department
and report directly to the President. This Board should include
individuals with experience as federal defenders, judges, victims’
representatives, probation officers, and police officers, among others, to get a range of perspectives aired and discussed.381
In addition to the thousands of clemencies, the Ford Clemency
Board left behind a worthwhile legacy: a 400-page report on how it
accomplished the task.382 To be sure, the Ford Clemency Board had
a more narrow task given that it had only a subset of all federal
cases. But its methodology could apply more broadly. For example,
375. See id. at 756-66.
376. See supra notes 209-11 and accompanying text.
377. See PRESIDENTIAL CLEMENCY BD ., REPORT TO THE PRESIDENT 165, 178 (1975).
378. See id. app. A (providing the names and biographies of the Presidential Clemency
Board members).
379. Id.
380. See supra notes 209-11 and accompanying text. Others have suggested alternative
ways to achieve this independence. For example, Paul Larkin urges that the Pardon Attorney
be relocated to the White House. Larkin, supra note 172, at 905-06. While that approach
achieves the needed independence, we prefer a board model to give the President a broader
range of viewpoints and to provide the President with more political cover—by allowing him
or her to make bipartisan appointments—for broader reforms.
381. Barkow & Osler, supra note 364, at 21.
382. PRESIDENTIAL CLEMENCY BD ., supra note 377.

2017]

DESIGNED TO FAIL

463

to ensure uniformity, the Board paid NASA to engineer a computer
program that reviewed case dispositions for consistency with Board
precedent (something that even the Pardon Attorney does not seem
to have in place, forty-two years later).383 The White House has already created an office that specializes in using technology for better
data analysis.384 That office could put in place similar mechanisms
for the Clemency Board based on key factors for consideration
across cases, or it could have different templates for different
categories of cases.385
There is one other opportunity that would come with the formation of a clemency board that is absent in our current system: the
ability to accumulate and analyze data relative to clemency grants
and denials, tracking things such as racial disparities, recidivism
rates, and the success of reentry.386 That data could then be used to
guide subsequent clemency decisions as well as other policy
positions.
C. Applying Office of Information and Regulatory Affairs
Review to Criminal Justice
Other executive agencies in government face oversight from the
President through the Office of Information and Regulatory Affairs
(OIRA), which is charged with making sure that the benefits of
agency regulations outweigh the costs, and that less expensive
alternatives have been considered.387 Christopher DeMuth and
Judge Douglas Ginsburg explained that this review process is
necessary because a regulatory agency “will invariably wish to
spend ‘too much’ on its goals. An agency succeeds by accomplishing
the goals Congress set for it as thoroughly as possible—not by
383. See id. at 327.
384. See U.S. DIGITAL SERV., EXEC. OFF. PRESIDENT, https://www.usds.gov [https://perma.
cc/WFH2-S7TU].
385. It is also possible that the Clemency Board could subsume the staff and budget of the
Pardon Attorney if authorized by Congress, which would make the move revenue-neutral,
particularly if commutations increase and reduced prison costs are considered. But even if the
President cannot get Congress to agree, he or she could potentially use some of the existing
budget that is used for the Domestic Policy Council for this effort.
386. See Barkow & Osler, supra note 364, at 22-23.
387. See Exec. Order No. 12,866 § 6, 3 C.F.R. 638, 644-48 (1994), reprinted in 5 U.S.C. § 601
(2012); Exec. Order No. 13,563 § 1, 3 C.F.R. 215, 215-16 (2012), reprinted in 5 U.S.C. § 601
(2012).

464

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

balancing its goals against other, equally worthy goals.”388 They
therefore advocated for “some countervailing restraint” in the form
of review “by an office that has no program responsibilities” to
generate more balanced regulatory decisions.389
These same problems exist for prosecutors, who also will inevitably “wish to spend ‘too much’”390 on their goals of criminal law
enforcement. But no countervailing constraint was created for them
and their policy decisions because OIRA review is currently set up
for regulations, not general policy positions.391 But with parole gone
and clemency within the hands of prosecutors themselves, no other
executive actor takes a second look at prosecutorial decision-making
in individual cases, much less undertakes an analysis of the way in
which the broader policies are operating and interacting.
Presidents do not have to accept this model. The OIRA review
process was created by executive order, and another president could
decide to institute reviews of Department policies, either by placing
such review within OIRA or, if the President were to create a
Criminal Justice Advisory Commission as described above, within
that body. Specifically, the President could ask the Department to
submit its policy determinations for review, to provide data to support its positions when available, and to explain why it rejected alternatives. Although this kind of oversight will not work for case-bycase applications of policies, it will allow for analysis of the policies
themselves, which may in turn affect how the Department operates
in specific cases. In an area like forensics, for example, the Department would have to explain why it rejected the PCAST findings, and
the reviewing agency in the White House (either OIRA or something
like it) could assess those determinations. If the White House
disagreed, it could tell the Department to shift its policy decision.
The same type of review could take place for BOP programming, or
388. Christopher C. DeMuth & Douglas H. Ginsburg, White House Review of Agency
Rulemaking, 99 HARV. L. REV. 1075, 1081 (1986).
389. Id.
390. Id.
391. President George W. Bush extended OIRA review for guidance documents, see Exec.
Order No. 13,422, 3 C.F.R. 191 (2008), but President Obama reversed that policy, see Exec.
Order No. 13,497, 3 C.F.R. 218 (2010), reprinted in 5 U.S.C. § 601 (2012). The Department’s
policy positions may not meet the definition of a guidance document as Executive Order
13,422 defined it, but a President could define the scope of what should be submitted for
review.

2017]

DESIGNED TO FAIL

465

Department positions before agencies like the Sentencing Commission or its testimony before Congress. The idea would be to have the
Department explain to another actor within the Executive Branch
who brings a generalist perspective, and not strictly a law enforcement one, why its approach makes sense. This poses no problems to
the Department’s independence, because the White House would not
be interfering with any individual determinations about whom to
prosecute or investigate. It would be simply overseeing how the
Department sets its policies overall and asking the Department to
provide support. In the process of preparing such reports, the
Department might find that some positions should be revised
because they really do not have support in data.392
The President knows how to order specific policies when he or she
wants to, so this would be a mechanism for him or her to find out
when it is necessary to intervene. President Obama followed a
model along these lines after his Task Force on 21st Century Policing released its report. He “directed [his] team to adopt the Task
Force’s federal recommendations.”393 He apparently did not wait to
see how the FBI and other federal law enforcement agencies
thought they should respond. The issue was important enough to
him that he ordered the shift in practices immediately. The President took a similar approach with respect to solitary confinement,
issuing a directive to the Department in 2016 to implement certain
reforms.394
President Obama’s more aggressive commands in those contexts
—and the contrast with the other areas outlined here—suggests
that part of the Department’s unwillingness to go further may well
have been because it was not receiving sufficiently strong signals
from President Obama about how much political bandwidth he
would spend on a given issue.395 As outsiders, we cannot say if
392. To be sure, some policies might be more difficult to quantify in terms of cost-benefit
analysis for purposes of OIRA oversight. But that oversight need not be limited to purely
quantitative review. See Exec. Order No. 13,563 § 1(c) (“Where appropriate and permitted by
law, each agency may consider (and discuss qualitatively) values that are difficult or impossible to quantify, including equity, human dignity, fairness, and distributive impacts.”).
393. Obama, supra note 2, at 842.
394. See id. at 830.
395. This may also help explain why the Department took more aggressive steps in the
second term than the first, when the President may have signaled he, too, was more willing
to go out on a political limb for criminal justice. See supra note 125 and accompanying text.

466

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

President Obama privately urged the Department to do more. But
we do know that any President will be ill-equipped to push for
things if he does not know they are at issue. And this reviewing
mechanism would allow a President and White House staff to stay
apprised of key issues, should the President want to exercise political capital for a change.
Of course, it is possible that the Department resists the White
House’s requests for change, and the next decision point for the
President would be what to do about it. At that point, the President
would need to decide if the issue is important enough to ask for the
resignation of the relevant political appointee, which in some cases
could be the Attorney General. Radley Balko envisioned just this
type of scenario with respect to the Department’s rejection of the
PCAST report.396 He argued that the President should have
discussed the report with Attorney General Lynch, and if she
refused to implement the recommendations, asked for her resignation.397 He made the same argument with respect to Director James
Comey of the FBI.398 This is obviously a politically risky move for
any President, but one would expect that few disputes would come
to this point. Instead, the OIRA-like review process should itself
bring out the most rational path to follow, and one would expect the
White House and Department to coalesce around that option, with
few disagreements rising to the level of a standoff.
D. Appointments
When President Obama analyzed what Presidents can do to
achieve criminal justice reform, he focused mainly on substantive
policies without mentioning the critical role the appointment power
plays in furthering reform. To be sure, he footnoted how important
it was to have Attorneys General leading his efforts who supported

And it may account as well, in part, for the bolder push by the Civil Rights Division, which
also received higher-level backing from the White House in contrast to, say, clemency, where
those announcements were left to the Deputy Attorney General and White House Counsel
until the very end of President Obama’s time in office when he started to make claims about
clemency on his own. See supra notes 160-65 and accompanying text.
396. See Balko, supra note 334.
397. See id.
398. See id.

2017]

DESIGNED TO FAIL

467

his stated reform agenda.399 But the Attorney General is not the
only position that matters for criminal justice reform; there are
several other key appointments Presidents make that have a significant impact on the direction criminal justice reform takes. This
Part highlights these other critical appointment decisions and how
important it is to appoint people with the right backgrounds for
these posts.
1. Department of Justice Appointments
In addition to the Attorney General, other senior posts at the
Department—especially the Deputy Attorney General and the
Director of the BOP—are essential for the success of criminal justice
reform. The Assistant Attorney General for the Criminal Division
and head of the Office of Legal Policy also occupy crucial posts. If a
President is concerned with reform, then he or she should make
sure that individuals in those posts share his or her reform goals. In
addition, a President interested in reform should encourage the
Attorney General to dial-back the influence of United States Attorneys in setting policy within the Department. The Attorney General
has an Advisory Committee composed of United States Attorneys
that holds enormous sway in influencing Department policy.400 A
more balanced advisory committee—with defense lawyers, academics, and experts on criminal justice policy—would produce more
balanced policy advice than a group made up of chief prosecutors.
We want to highlight two other key positions that may be less
obvious here. The first is the Assistant Attorney General for the
Civil Rights Division. The Civil Rights Division works on a range of
critical issues including voting rights and discrimination cases
399. Obama, supra note 2, at 824 n.53 (“[T]he most consequential actions I took [with
respect to changing policies] were my decisions to entrust these dedicated public servants
[Eric Holder and Loretta Lynch] (and other senior officials in the Department of Justice) with
the authority to use their discretion wisely and to provide guidance and set an example for
the thousands of federal prosecutors across the United States.”). Although we have outlined
why even the most resolute leader of the Department will face obstacles in pushing the
institution toward reform—and may in fact become captured by the viewpoints from within
the Department—it is certainly true that getting someone interested in furthering the reform
agenda highlighted by President Obama is a necessary, if not sufficient step.
400. See Press Release, U.S. Dep’t of Justice, Six New Members Appointed to Attorney
General’s Advisory Committee (Apr. 30, 2015), https://www.justice.gov/opa/pr/six-newmembers-appointed-attorney-general-s-advisory-committee [https://perma.cc/Z3LD-EJSF].

468

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

outside of criminal justice,401 so Presidents may choose leaders of
that Division based on how they prioritize different aspects of the
Division’s work. For a President that cares especially about criminal
justice issues, it is helpful to choose someone with experience
working on reform in that area. President Obama did a particularly
good job selecting heads of the Civil Rights Division. The Division
worked on police reform from the start of the Obama Administration
under the leadership of Tom Perez,402 who came to the position with
a background in civil rights (including having previously worked as
a prosecutor in the Civil Rights Division as well as advising Senator
Kennedy on civil rights and criminal justice issues), consumer
advocacy, and state politics.403 The Division was even more active in
criminal justice reform in President Obama’s second term under the
leadership of Vanita Gupta. Gupta’s background was not as a
Department prosecutor. Instead she came to the Department having
previously worked as a civil rights lawyer, including serving as the
Deputy Legal Director of the ACLU, where she was responsible for
the ACLU’s national effort to further criminal justice reform.404
Under her leadership, the Civil Rights Division focused on how the
pernicious effects of fines and fees both corroded the relationship
between the police and the community, as well as undermined
401. See U.S. DEP’T JUSTICE, supra note 347.
402. See, e.g., David Corn, Why Picking Tom Perez for Attorney General Would Be a Smart
Move for Obama, MOTHER JONES (Nov. 3, 2014, 6:00 AM), http://www.motherjones.com/
politics/2014/10/tom-perez-attorney-general-obama [https://perma.cc/D557-VCRS] (“Perez
revitalized [the Civil Rights Division] by mounting voting rights cases and legal challenges
to discrimination against gays and lesbians. During his tenure, the division opened a recordbreaking number of investigations into police abuse and forged wide-ranging agreements to
clean up various police forces accused of misconduct.”); Justin Miller, The Subtle Force of Tom
Perez, AM . PROSPECT (June 22, 2016), http://prospect.org/article/subtle-force-tom-perez [https://
perma.cc/3PG2-S2SM] (noting Perez’s work on police misconduct—including investigations
into Seattle, Portland, Los Angeles, and Puerto Rico police forces and Arizona Sheriff Joe
Arpaio—and stating that “[b]y 2011, the division was investigating 17 police and sheriff departments—the most in the DOJ’s history”).
403. See, e.g., Meet the Assistant Attorney General, U.S. DEP’T JUSTICE, https://www.justice.
gov/crt/meet-aag-banner [https://perma.cc/QCQ3-ENPU] (discussing Perez’s background enforcing consumer rights laws and workplace safety laws); Miller, supra note 402 (discussing
the same and Perez’s experience advising Senator Ted Kennedy on civil rights and criminal
justice issues).
404. Gupta also worked for the NAACP Legal Defense and Educational Fund. See Former
Head of the Civil Rights Division Vanita Gupta, U.S. DEP’T JUST. ARCHIVES, https://www.
justice.gov/archives/crt/staff-profile/former-head-civil-rights-division-vanita-gupta [https://
perma.cc/WA9C-555F].

2017]

DESIGNED TO FAIL

469

efforts to reduce criminal behavior because of how those financial
obligations burdened people already struggling to get by. This is the
kind of insight one gets from having a range of experiences working
on criminal justice issues, not just as a prosecutor, and it is a good
blueprint for further appointees to head the Civil Rights Division if
the goal is to advance criminal justice reform.
President Obama’s pick for IG of the Department also proved to
be critical to his reform efforts. While IGs are typically known for
becoming captured by the very agencies they are charged with
overseeing,405 that did not hold true for Michael Horowitz, who was
appointed IG in 2012. Horowitz came to the IG position with extensive Department experience, including eight years as a prosecutor
in the Southern District of New York and three years at the
Department in the Criminal Division as Chief of Staff and Deputy
Assistant Attorney General.406 He came to the job with the perspective of a prosecutor and the skill set of a first-rate investigator. His
work as the Chief of Public Corruption in the United States Attorney’s Office for the Southern District of New York likely made him
sensitive to the misuse of government resources. He also served for
six years as a Commissioner at the United States Sentencing
Commission.407 The Commission, as noted, grounds its decisions in
data, and Horowitz took that same perspective as IG, amassing
critical data in service of all his recommendations. Perhaps most
importantly, IG Horowitz had a broader perspective of the IG’s role
than some of his predecessors, not simply looking for fraud and
misconduct, but looking for ways to encourage the Department to
more effectively and rationally spend and allocate its limited law
enforcement resources.408 He was not afraid to criticize the Department for falling short, and the result was that he influenced the
Department’s criminal justice agenda.
Perhaps most critical were Horowitz’s calls for the Department to
pay close attention to what he labeled as the “growing crisis in the
405. See Rachel E. Barkow, Insulating Agencies: Avoiding Capture Through Institutional
Design, 89 TEX. L. REV. 15, 21-24 (2010).
406. Meet the Inspector General, U.S. DEP’T JUSTICE, https://oig.justice.gov/about/meetig.htm [https://perma.cc/H78W-KY85].
407. Id.
408. See Rachel E. Barkow, Foreword: Overseeing Agency Enforcement, 84 GEO . WASH . L.
REV. 1129, 1178 (2016).

470

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

federal prison system” of both overcrowding and having BOP costs
eat up a greater percentage of the Department’s budget, thus
crowding out other law enforcement priorities.409 His reports and
testimony on the strain the BOP’s budget posed for other law
enforcement needs provided the springboard for the Department’s
Smart on Crime Initiative.
The IG’s analysis of the compassionate release practices of the
BOP also catalyzed the Department to change its policies. Even
though the Department did not go as far as the IG suggested, it is
doubtful the Department would have done anything at all without
the IG’s prompting. Similarly, the IG’s analysis of private prisons
(showing that they posed greater security and safety risks) led the
Department to announce it would reduce and ultimately end its use
of private for-profit facilities.410 Additional IG reports highlighted
the need for the BOP to do a better job preparing individuals for
reentry.411 Ultimately, at the end of President Obama’s time in
office, the Department focused more on the BOP’s programming.412
The IG’s data-driven emphasis on how the Department was falling
short in its law enforcement mission by spending too much money
on failed policies provided the foundation for change. The fact these
reports were publicly released also helped create some pressure

409. Memorandum from Michael E. Horowitz, Inspector Gen., U.S. Dep’t of Justice, to the
Attorney Gen. and the Deputy Attorney Gen. (Dec. 11, 2013, reissued Dec. 20, 2013), https://
oig.justice.gov/challenges/2013.htm [https://perma.cc/5YRW-Z9VH]; see also Oversight of the
Department of Justice: Hearing on Appropriations for 2014 Before the Subcomm. on Commerce, Justice, Science and Related Agencies of the H. Comm. on Appropriations, 113th Cong.
131-32 (2013) (statement of Michael E. Horowitz, Inspector General, U.S. Department of
Justice).
410. See OFFICE OF THE INSPECTOR GEN ., U.S. DEP’T OF JUSTICE, REVIEW OF THE FEDERAL
BUREAU OF PRISONS’ MONITORING OF CONTRACT PRISONS i-iii (2016), https://oig.justice.gov/
reports/2016/e1606.pdf [https://perma.cc/G3PV-NZ3V]; see also Obama, supra note 2, at 831.
411. See generally OFFICE OF THE INSPECTOR GEN ., U.S. DEP’T OF JUSTICE, REVIEW OF THE
FEDERAL BUREAU OF PRISONS’ RELEASE PREPARATION PROGRAM i-iii (2016), https://oig.justice.
gov/reports/2016/e1607.pdf [https://perma.cc/ACJ7-52G7].
412. See generally BOS. CONSULTING GRP., REDUCING RECIDIVISM THROUGH PROGRAMMING
IN THE FEDERAL PRISON POPULATION : FINAL REPORT: BOP PROGRAMS ASSESSMENT (2016),
https://www.justice.gov/dag/page/file/914031/download [https://perma.cc/6D66-CC87] (assessing BOP programming to determine its effects on reentry and recidivism); Prison Reform:
Reducing Recidivism by Strengthening the Federal Bureau of Prisons, U.S. DEP’T JUSTICE
ARCHIVES, https://www.justice.gov/prison-reform [https://perma.cc/D2AW-T54F] (listing recent
Federal Bureau of Prisons reforms to improve reentry).

2017]

DESIGNED TO FAIL

471

within the Department to make those changes and help break its
normal preference for the status quo.
As a result, Horowitz created a model for the Department’s IG
that could serve future Presidents well in their efforts to prompt
change in the Department’s practices. One of the key blocks to
reform efforts is getting the Department to go along with those
efforts—not only in principle, but also in implementation. Horowitz
policed the Department closely to make sure both its policies and its
implementation of those policies furthered the true mission of law
enforcement—protecting the public while respecting individual
rights—and used limited resources to achieve those goals most effectively. Future Presidents interested in criminal justice reform
would do well to find IGs who share Horowitz’s view that the scope
of the IG’s work includes both evaluating and implementing substantive policies to make sure they are the most effective use of the
Department’s limited resources.
2. The Judicial Branch
It is not just the President’s executive branch appointments that
matter for criminal justice reform. The judicial branch is a key
partner in efforts to improve the system. The Sentencing Commission, an independent agency within the judicial branch, issued some
of the most significant criminal justice reforms during President
Obama’s time in office by changing the sentencing guidelines on
drug offenses and making those changes retroactive.413 Appointments to the Commission are thus important for criminal justice
reform efforts.414
413. See supra Part II.A.1.
414. The Sentencing Commission is set up to be a bipartisan commission, so no more than
four of its seven members can come from the same political party. See Organization, U.S.
SENT’G COMMISSION , http://www.ussc.gov/about/who-we-are/organization [https://perma.cc/
N3UB-X6KP]. In addition, at least three of the seven commissioners must be federal judges.
Id. It is notable that President Obama appointed Ketanji Brown Jackson, an individual with
a background as an assistant federal public defender, to the Commission in 2009 (he later
appointed Jackson to the federal bench). See Del Quentin Wilber, Ketanji Brown Jackson
Nominated to Federal Bench, WASH . POST (Sept. 20, 2012), https://www.washingtonpost.
com/local/ketanji-brown-jackson-nominated-to-federal-bench/2012/09/20/744848dc-037d-11e28102-ebee9c66e190_story.html?utm_term=.3b95cfb29160 [https://perma.cc/NMU2-39YJ].
Previously, the Commission had largely been made up of individuals with prosecution experience, and few of its members have had experience as criminal defense lawyers, much less

472

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

Judges, too, are responsible for sentencing defendants and
overseeing the government’s law enforcement efforts to ensure that
these efforts are consistent with the Constitution and any relevant
statutes. In performing those tasks, it is critical to have a bench
that represents a diversity of viewpoints on criminal law. The
federal bench, however, is dominated by individuals with prosecution experience and has very few individuals with a public defense
background. Specifically, a full 43 percent of active Article III judges
have prosecutorial experience, compared with only 10.4 percent with
public defense experience.415
To his credit, President Obama’s judicial appointments outside of
the Supreme Court were somewhat less tilted toward prosecutors.
While 41 percent of his nominees had prosecution experience, 14
percent had public defense experience.416 Notably, of the seven
current Circuit Court judges with public defense backgrounds, five
were appointed by President Obama.417 But while President Obama
experience as federal defenders. See Barkow, supra note 366, at 764-65. In 2013, President
Obama appointed Rachel Barkow (one of the authors), a law professor whose scholarship is
largely devoted to criminal justice reform. Appointments like these help provide the Commission with important perspectives that balance the prosecutorial perspective it tends to have.
Because the majority of the Commission’s members at any one time typically have prosecutorial backgrounds and because the Department sends a person to serve as an ex officio
member of the Commission, the Department’s views are always presented at every meeting,
while the federal defenders lack any similar institutional presence on the Commission. See
id. at 764.
415. Another 13.7 percent have private criminal defense experience, but that typically
consists of white collar defense. While those experiences help practitioners see some areas
common to all criminal defendants, they do not typically expose lawyers to the plight of the
poorest criminal defendants who make up the bulk of the federal prison population. These
figures were compiled by downloading a database of federal judges. See History of the Federal
Judiciary: Biographical Directory of Article III Federal Judges, 1789-Present, FED. JUD . CTR.,
http://www.fjc.gov/history/judges [https://perma.cc/YP2E-5ULY]. This database was then
limited to active, non-senior district court and circuit court judges and queried for career histories. All judges with private experience were reviewed manually to determine if they did
defense work because it was not obvious from job title alone. These numbers were then crosschecked with information published in a study by the Alliance for Justice. See ALL. FOR JUSTICE , BROADENING THE BENCH : PROFESSIONAL DIVERSITY AND JUDICIAL NOMINATIONS (2016),
http://www.afj.org/wp-content/uploads/2014/11/Professional-Diversity-Report.pdf [https://
perma.cc/3YB2-Z8BA].
416. Twenty percent of his nominees had private defense experience. See ALL. FOR JUSTICE,
supra note 415, at 8-9.
417. These judges are G. Steven Agee, Bernice Donald, Jane Kelly, Edward Prado, Luis
Restrepo, Robert Wilkins, and James Wynn. See Biographical Directory of Article III Federal
Judges, supra note 415.

2017]

DESIGNED TO FAIL

473

did a better job than previous Presidents at appointing individuals
with more varied criminal justice experience (including appointing
people who had been public defenders) the overall numbers still
skew heavily toward people who had been prosecutors.
People can change their perspectives and outlooks with new roles,
so undoubtedly many people who take the bench after serving as
prosecutors may actively question government positions. But again,
it defies what we know about human nature and experience to think
that someone’s past experiences have no effect on his or her outlook.
Because real criminal justice reform requires judges to check
overreach by prosecutors and other law enforcement officials, it is
critical that the bench reflect a range of professional backgrounds,
and for any future President to put people with diverse backgrounds
in positions responsible for criminal justice administration. This
includes having individuals with defense or criminal justice reform
backgrounds on the Supreme Court.
Lower courts matter as well, particularly as federal judges start
to explore diversion programs and reentry courts to aid with criminal justice reform. In this regard, it is notable that perhaps the
most successful reentry federal program was started by Judge Luis
Restrepo, who before becoming a judge was a Philadelphia public
defender and then later a federal defender.418 Future Presidents
should similarly pay attention to this factor if criminal justice reform remains a priority.
CONCLUSION
Barack Obama made strong statements about his commitment to
criminal law reform, and yet his actual achievements were modest.
The Department’s control over criminal justice reform is a big part
of the reason why, as we have seen in our examination of sentencing, clemency, compassionate release, and forensic science. A President who relies on the Department to roll back mass incarceration
and lead the way on large-scale criminal justice reform is choosing
an institutional model that is designed to fail, because such an effort
will focus too much on the Department and not enough on justice.
Obama’s passion for reform was groundbreaking, authentic, and
418. See ALL. FOR JUSTICE, supra note 415, at 8.

474

WILLIAM & MARY LAW REVIEW

[Vol. 59:387

important. Our hope is that a successor will pair that passion with
institutional models that allow for true and lasting success.