SSRN Id4458504

Washington and Lee Law Review
Volume 80 Issue 1 Article 6
Spring 2023
Federal Bureaucratic Studies

Jesse M. Cross
University of South Carolina School of Law, jmcross@law.sc.edu
Follow this and additional works at: https://scholarlycommons.law.wlu.edu/wlulr
Part of the Law Commons
Recommended Citation
Jesse M. Cross, Federal Bureaucratic Studies, 80 Wash. & Lee L. Rev. 229 ().
Available at: https://scholarlycommons.law.wlu.edu/wlulr/vol80/iss1/6
This Article is brought to you for free and open access by the Washington and Lee Law Review at Washington and
Lee University School of Law Scholarly Commons. It has been accepted for inclusion in Washington and Lee Law
Review by an authorized editor of Washington and Lee University School of Law Scholarly Commons. For more
information, please contact christensena@wlu.edu.
Electronic copy available at: https://ssrn.com/abstract=4458504

Federal Bureaucratic Studies
Jesse M. Cross*
Abstract
A vast literature has developed in legal scholarship on the

topic of bureaucratic governance. To date, this literature has
focused squarely on the executive branch. Yet a second
bureaucracy also exists in the federal government: the
congressional bureaucracy. Recent legislation scholarship has
brought this bureaucracy into focus—documenting its traits,
practices, and culture. In so doing, it has created a rich new
opportunity for cross-disciplinary dialogue—one where
executive-branch studies and legislative studies collaborate
toward a larger understanding of how bureaucracy operates, and
can operate, in a presidentialist system.
To begin that cross-disciplinary conversation, this Article
turns to five themes in the executive-branch literature. These are:
(i) the dual-allegiance problem, (ii) bureaucratic resistance, (iii)
dual advising-adjudicating roles, (iv) agency capture, and (v)
comparative understandings of the judiciary. In each case,
theories developed in the executive branch context enrich our
understanding of the congressional bureaucracy, while new
knowledge about the congressional bureaucracy also forces
revisions to those executive-branch theories. In many cases, the
congressional bureaucracy also reveals new governance solutions
in our tripartite system—solutions that are overlooked when
bureaucracy scholarship is confined to studies of a single branch.
Through an exploration of these and other lessons, the Article
illustrates the many possibilities inherent in a new
* Associate Professor, University of South Carolina School of Law. The

author wishes to thank Abbe Gluck, William Eskridge, and all the government
officials who contributed information.
229

230 80 WASH. & LEE L. REV. 229 (2023)
cross-disciplinary dialogue on the role of bureaucracy in our

federal system.
Table of Contents
INTRODUCTION .................................................................. 230

I. BACKGROUND: A TALE OF TWO BUREAUCRACIES ... 235
II. DUAL-ALLEGIANCE PROBLEM ................................. 242
III. BUREAUCRATIC RESISTANCE ................................... 249
A. Descriptive Questions: Bureaucratic
Protections ........................................................ 249
B. Descriptive Questions: Tools of Bureaucratic
Resistance ......................................................... 259
C. Normative Questions: Is Bureaucratic
Resistance Desirable? ....................................... 263
IV. ADVISING AND ADJUDICATING: ON DUAL ROLES .... 270
V. AGENCY CAPTURE .................................................... 273
VI. CONGRESSIONAL BUREAUCRACY AND THE JUDICIAL
BRANCH.................................................................... 279
A. Neutrality .......................................................... 280
B. Expertise ............................................................ 284
CONCLUSION ...................................................................... 294
INTRODUCTION
In legal scholarship, a vast literature has developed on the

topic of bureaucratic governance. Motivated by the rise of the
administrative state, it has sought to understand the
governmental structures that have predominated since the New
Deal settlement.1 Today, this literature spans a variety of fields,
1. On the idea of the “New Deal settlement,” see, e.g., Larry D. Kramer,
The Supreme Court 2000 Term Foreword: We the Court, 115 HARV. L. REV. 4,
122 (2001).

FEDERAL BUREAUCRATIC STUDIES 231
including administrative law,2 separation-of-powers law,3 and

presidential scholarship.4 Across these fields, it has taken on the
important challenge of understanding the balance of democratic
accountability and nonpartisan expertise in the bureaucracies
that populate the modern administrative state.
To date, this literature has focused squarely on the
executive branch.5 This is not surprising: the rise of
administrative agencies marked a momentous transformation
in federal governance, expanding its ranks by millions and
introducing new structures and actors.6 In response, scholars
have directed much attention toward the administrative
agency—studying its different organizational arrangements,
rules, and employees. Scholars also have devoted important
attention to the attorneys and other professionals who populate
specific executive branch offices, such as the Office of Legal
Counsel (“OLC”)7 and the Office of Management and Budget
2. For discussions involving scholarship generally considered

administrative law studies, see infra Parts IV, VI, & VII. Administrative law
has often been understood centrally as the study of bureaucracy and its
oversight by the judiciary. See DANIEL E. HALL, ADMINISTRATIVE LAW:
BUREAUCRACY IN A DEMOCRACY 19 (7th ed. 2020); C. EDLEY, JR.,
ADMINISTRATIVE LAW: RETHINKING JUDICIAL CONTROL OF BUREAUCRACY 33
(1990).
3. For discussions involving scholarship generally considered
separation-of-powers scholarship, see in particular infra Parts III, IV. See also
Bruce Ackerman, The New Separation of Powers, 113 HARV. L. REV. 633, 689
(2000) [hereinafter Ackerman, New Separation of Powers].
4. For discussions of scholarship generally considered
separation-of-powers scholarship, see in particular Parts IV, V. This research
has often focused on the tensions of a democratically accountable President
reliant upon a careerist bureaucracy for policy goals. See, e.g., JAMES Q.
WILSON, BUREAUCRACY: WHAT GOVERNMENT AGENCIES DO AND WHY THEY DO IT
257 (1989); Elena Kagan, Presidential Administration, 114 HARV. L. REV. 2245,
2272 (2001) (“Since the dawn of the modern administrative state, Presidents
have tried to control the bureaucracy only to discover the difficulty of the
endeavor.”).
5. See generally Neal Kumar Katyal, Internal Separation of Powers:
Checking Today’s Most Dangerous Branch from Within, 115 YALE L.J. 2314
(2006).
6. See Dennis Vilorio, Working for the Federal Government: Part 1, U.S.
BUREAU OF LAB. STATS. (Sept. 14, 2014), https://perma.cc/W7X2-N975.
7. See, e.g., BRUCE ACKERMAN, THE DECLINE AND FALL OF THE AMERICAN
REPUBLIC 88 (2010); Adrian Vermeule, Conventions of Agency Independence,
113 COLUM. L. REV. 1163, 1209–11 (2013); Katyal, supra note 5, at 2327;
Rebecca Ingber, Bureaucratic Resistance and the National Security State, 104

232 80 WASH. & LEE L. REV. 229 (2023)
(“OMB”).8 Through a study of these and other executive branch

institutions, this scholarship has made tremendous
contributions to our understanding of the modern presidency,
the administrative state, and modern bureaucratic power.
As Abbe Gluck and I have documented, however, a second
bureaucracy also exists in the federal government: the
congressional bureaucracy.9 As the executive branch expanded
over the twentieth century, Congress felt itself losing power to
the president—and it responded by creating its own legislative
bureaucracy.10 Today, Congress has thousands of nonpartisan
staffers spread across a dozen legislative offices.11 Much like the
civil servants studied in the executive branch context, this
congressional bureaucracy contributes expertise to democratic
governance, assisting partisans in their effort to govern in a
world of large, complex financial and bureaucratic institutions.12
Unlike executive branch agencies, however, the congressional
bureaucracy has received virtually no attention in the literature
on bureaucratic governance.13 Instead, comparative studies are
more likely to look to other nations—not realizing that a second
bureaucracy exists in our own federal government.14
IOWA L. REV. 139, 150 (2018); Daphna Renan, The Law Presidents Make, 103
VA. L. REV. 805, 808 (2017); Trevor W. Morrison, Stare Decisis in the Office of
Legal Counsel, 110 COLUM. L. REV. 1448, 1448 (2010).
8. See, e.g., John D. Graham, Valuing the Future: OMB’s Refined
Position, 74 U. CHI. L. REV. 51, 51 (2007); Donald R. Arbuckle, Obscure but
Powerful: Who Are Those Guys?, 63 ADMIN. L. REV. 131, 133 (2011); Jim Tozzi,
OIRA’s Formative Years: The Historical Record of Centralized Regulatory
Review Preceding OIRA’s Founding, 63 ADMIN. L. REV. 37, 40 (2011); Eloise
Pasachoff, The President’s Budget As A Source of Agency Policy Control, 125
YALE L.J. 2182, 2182 (2016).
9. I take the term “congressional bureaucracy” from our article, Jesse M.
Cross & Abbe R. Gluck, The Congressional Bureaucracy, 168 U. PA. L. REV.
1541, 1543 (2020).
10. See id. at 1555–60.
11. See id. at 1599–1600.
12. See id. at 1543–45.
13. As Bruce Ackerman observes, “comparative public administration is
not a well worked field.” Ackerman, New Separation of Powers, supra note 3,
at 710.
14. See, e.g., id. at 700 (comparing American and European approaches
to bureaucracy); Terry M. Moe & Michael Caldwell, The Institutional
Foundations of Democratic Government: A Comparison of Presidential and
Parliamentary Systems, 150 J. INSTITUTIONAL & THEORETICAL ECON. 171, 172
(1994).

It is not surprising that bureaucracy scholarship has

neglected this congressional counterpart. Even among
legislation scholars, the congressional bureaucracy has long
been overlooked.15 In recent years, however, a movement has
arisen to construct a more detailed, accurate, and modern
understanding of the institution of Congress.16 Termed by
Justice Amy Coney Barrett as the “process-based turn” in
legislative studies,17 this movement has brought the
congressional bureaucracy into focus for the first time.18
This scholarship on the congressional bureaucracy has
created a rich new opportunity for cross-disciplinary dialogue.
Under this project, legislative and executive branch scholars can
collaborate toward a larger understanding of how bureaucracy
operates—and can operate—in a presidentialist system. It is the
goal of this Article to begin that cross-disciplinary conversation,
showing the ways in which each field’s insights and discoveries
can transform the other.
To begin that cross-disciplinary conversation, this Article
turns to five persistent themes in the executive branch
literature. In each instance, it shows how awareness of the
congressional bureaucracy remakes conversations in both fields.
Those five themes are as follows:
 The dual-allegiance problem: Bureaucrats can have
allegiances to multiple principals, and scholars have
theorized about the institutional pathologies this can
create in a tripartite government (and about how to
minimize those pathologies).
 Bureaucratic resistance: Bureaucrats sometimes gain
autonomy from politically-accountable leaders,
creating opportunities for strategic resistance.
Scholars have examined the factors that foster this
autonomy, as well as the opportunities and risks it
creates.
15. See infra Part I.

16. See infra Part I.
17. Amy Coney Barrett, Congressional Insiders and Outsiders, 84 U. CHI.
L. REV. 2193, 2193 (2017). For her part, Justice Barrett is skeptical of the
interpretive utility of this process-based turn.
18. For the few studies before the Cross-Gluck study that looked together
at multiple offices of this bureaucracy, see infra note 55.

234 80 WASH. & LEE L. REV. 229 (2023)
 Dual advising/adjudicating roles: Some bureaucratic

offices must simultaneously perform advisory and
adjudicatory functions, and scholars have looked for
ways to address the risks that attend this institutional
structure.
 Agency capture: Private interest groups can gain undue
influence over bureaucratic decision-making, and
scholars have hypothesized about the origins of this
risk—and about institutional designs to minimize it.
 Comparative understandings of the judiciary:
Bureaucracies can provide illuminating comparisons to
the judiciary, particularly on metrics of neutrality and
expertise. Scholars have examined these two
institutions together for lessons on comparative
institutional competence and the proper relationship
between the branches.
For each of these topics, adding the congressional
bureaucracy to the discussion offers significant new lessons. For
legislation scholars, it enriches the understanding of the
congressional bureaucracy; here, executive-branch scholarship
provides theories of bureaucratic governance that can help
explain various dimensions of a previously under-theorized
bureaucracy. For executive-branch scholars, it provides a
testing ground for those same theories: the congressional
bureaucracy sometimes confirms and expands those theories,
other times revises or undermines them. At the same time, by
introducing the possibility of relocating bureaucracy not only
within but across branches, the congressional bureaucracy also
reveals new governance solutions that exist within our tripartite
system—solutions that are overlooked when bureaucracy
scholarship is siloed by governmental branch. Together, these
lessons illustrate the many possibilities inherent in a new
cross-disciplinary dialogue on bureaucracy in federal
government.
The Article proceeds in six Parts. Part I begins with a brief
background on the congressional bureaucracy and a review of
the relevant scholarship. Parts II through VI then introduce the
congressional bureaucracy into discussions in executive branch
theory, with each Part devoted to a different discussion. Part II
looks at the dual-allegiance problem, Part III turns to the topic
of bureaucratic resistance, Part IV examines the challenges
faced by offices that must both advise and adjudicate, Part V

looks at the problem of agency capture, and Part VI examines

comparative understandings of the judiciary. A brief conclusion
follows.
I. BACKGROUND: A TALE OF TWO BUREAUCRACIES
Since the inception of the modern administrative state,

scholarship on the federal bureaucracy has unfolded quite
differently for the two political branches. On the one hand, the
executive branch bureaucracy has received extensive study and
discussion from legal scholars.19 In the late nineteenth and early
twentieth centuries, many leading legal academics devoted
significant attention to understanding, debating, and
legitimating the administrative agencies that were emerging as
a pivotal tool of federal governance—figures including Woodrow
Wilson,20 Felix Frankfurter,21 James Landis,22 and others.23
Frankfurter also helped cement administrative law in the
academic curriculum of law schools, using his platform at
Harvard Law School to entrench it as an important field of legal
academic inquiry.24 It has remained there ever since: today, the
19. While typically associated with the New Deal, administrative

agencies date back to the Founding, and the rise of the modern administrative
state does not have a clearly defined start date, with important antecedents in
the late-1800s and before. See generally STEPHEN SKOWRONEK, BUILDING A NEW
AMERICAN STATE: THE EXPANSION OF NATIONAL ADMINISTRATIVE CAPACITIES,
1877–1920 (1982); Cass R. Sunstein, Constitutionalism After the New Deal,
101 HARV. L. REV. 421 (1987).
20. See Woodrow Wilson, The Study of Administration, 2 POL. SCI. Q. 197,
198 (1887) (“The science of administration is the latest fruit of that study of
the science of politics which was begun some twenty-two hundred years ago.”).
21. FELIX FRANKFURTER & J. FORRESTER DAVISON, CASES AND MATERIALS
ON ADMINISTRATIVE LAW (2d ed. 1935).
22. JAMES LANDIS, THE ADMINISTRATIVE PROCESS (1938).
23. See, e.g., WALTER GELLHORN, ADMINISTRATIVE LAW: CASES AND
COMMENTS (1940); LLOYD MILTON SHORT, THE DEVELOPMENT OF NATIONAL
ADMINISTRATIVE ORGANIZATION IN THE UNITED STATES 24 (1923); W.F.
WILLOUGHBY, AN INTRODUCTION TO THE STUDY OF THE GOVERNMENT OF MODERN
STATES 386 (1919); Robert L. Hale, Coercion and Distribution in a Supposedly
Noncoercive State, 38 POL. SCI. Q. 470, 478–81 (1923); FRANK J. GOODNOW,
COMPARATIVE ADMINISTRATIVE LAW: AN ANALYSIS OF THE ADMINISTRATIVE
SYSTEMS NATIONAL AND LOCAL, OF THE UNITED STATES, ENGLAND, FRANCE AND
GERMANY (1903).
24. See Mark Fenster, The Birth of a “Logical System”: Thurman Arnold
and the Making of Modern Administrative Law, 84 OR. L. REV. 69, 80 (2005).

236 80 WASH. & LEE L. REV. 229 (2023)
executive branch bureaucracy regularly generates

conferences,25 symposia,26 and journals.27
By contrast, the congressional bureaucracy has historically
received little attention. The topic of legislation itself, after a
brief promising period in the 1920s and 1930s, was not regarded
as a distinct area of inquiry (or a separate class in most law
schools) until the 1980s.28 As the field emerged in that decade,
it was shaped by formative figures such as Justice Antonin
Scalia, whose approach to legislation and statutory
interpretation displayed little interest in the realities of the
modern legislative process.29 Even among those with interest in
legislative process, focus typically was upon the textbook
“Schoolhouse Rock”30 legislative process and the partisan staff
that assisted it.31 Generally overlooked were the nonpartisan
25. See, e.g., Bureaucracy and Presidential Administration: Expertise and

Accountability in Constitutional Government, CTR FOR THE STUDY OF THE
ADMIN. STATE, ANTONIN SCALIA L. SCH. (Feb. 6, 2020), https://perma.cc/N3AY-
4CXL; The Administration of Immigration, CTR. FOR THE STUDY OF THE ADMIN.
STATE, ANTONIN SCALIA L. SCH. (Oct. 25, 2019), https://perma.cc/F38T-WRZJ;
Regulatory Change & the Trump Administrative State Conference, YALE J. ON
REG. (Mar. 22, 2019), https://perma.cc/ZXF3-8N7C; 2020 Administrative Law
Conference, AM. BAR ASS’N (Nov. 19–20, 2020), https://perma.cc/44UT-Z2NW.
26. See, e.g., Ctr. for the Study of the Admin. State, Symposium on
Federal Agency Adjudication (Aug.-Sept. 2020), https://perma.cc/8S5L-DJEG;
Ctr. for the Study of the Admin. State, The Administration of Democracy—
The George Mason Law Review’s Second Annual Symposium on
Administrative Law (Oct. 4, 2019), https://perma.cc/YLX6-K3VX: Reg. Rev.,
Constitutional Questions and the Administrative State (Dec. 16, 2019),
https://perma.cc/UC3C-6Q5T; Yale J. on Reg., Symposium on Racism in
Administrative Law (2020), https://perma.cc/BS5T-TTK2.
27. ADMIN. L. REV., https://perma.cc/88VV-53NP.
28. See William N. Eskridge, Jr., The Three Ages of Legislation Pedagogy,
7 N.Y.U. J. LEGIS. & PUB. POL’Y 3, 4 (2004) (chronicling the early promise of the
1920s and 1930s, the period until the 1980s when “legislation was basically a
dead area of legal academic inquiry,” and the 1980s “resurgence”); see also id.
at 6 (chronicling the assembly of Eskridge and Frickey’s landmark 1988
legislation casebook and subsequent return of Legislation classes to most law
schools).
29. See William N. Eskridge, Jr., The New Textualism, 37 UCLA L. REV.
621, 624 (1990) (describing Scalia’s role in ushering in the rise of “new
textualism” and its lack of interest in legislative process).
Schoolhouse Rock!: I’m Just a Bill (ABC television broadcast
30
Mar. 27, 1976), https://perma.cc/EN7K-Q27Q.

31. See, e.g., WILLIAM N. ESKRIDGE JR ET AL., STATUTES, INTERPRETATION,
AND REGULATION 33-80 (2014) (reviewing standard legislative process);

offices inside Congress, as well as the new legislative process

that was evolving to incorporate them and their expertise.32
However, Congress did indeed create a significant
nonpartisan legislative bureaucracy in the twentieth century. It
was a project with roots in the Progressive Era,33 when good
governance projects at the state level exerted an influence on
federal legislators34—and one that saw important expansions in
the 1940s and 1970s,35 when concerns about executive branch
power led Congress to resist by building its own nonpartisan
bureaucracy.36 The result was the congressional bureaucracy
Stephen Breyer, On the Uses of Legislative History in Interpreting Statutes, 65

S. CAL. L. REV. 845, 845 (1992) (arguing for relevance of “congressional floor
debates, committee reports, hearing testimony, and presidential messages”);
id. at 858–59 (observing role of partisan “staff members for legislators”); Cross
& Gluck, supra note 9, at 1636 (“Congress also has changed over time, and yet
barely a dent has been made from those changes in even those interpretive
theories and doctrine that are purportedly based on Congress’s own
operations.”).
32. See Cross & Gluck, supra note 9, at 1554 (noting the “description
deficit” for these offices in the literature).
33. See Amendment to H.R. 15279, Legislative, Executive, and Judicial
Fiscal Year 1915 Appropriations Act, June 13, 1914, U.S. CAPITOL VISITOR
CTR., https://perma.cc/KSR6-623D (providing funding and direction in 1914 to
establish the service that would become the Congressional Research Service);
Revenue Act of 1918, § 1303(a), Pub. L. No. 254, 40 Stat. 1057, 1141–42
(establishing Legislative Counsel in 1918).
34. See George K. Yin, Legislative Gridlock and Nonpartisan Staff, 88
NOTRE DAME L. REV. 2287, 2292–93 (2013) (discussing state antecedents to the
Congressional Research Service); WIS. LEGIS. REFERENCE BUREAU,
https://perma.cc/7RXS-PCZW; 56 CONG. REC. 10524 (1917) (statement of Rep.
Greene).
35. See Legislative Reorganization Act of 1946, Pub. L. No. 79-601, 60
Stat. 812; Legislative Reorganization Act of 1970, Pub. L. No. 91-510, 84 Stat.
1140 (codified at 2 U.S.C. § 166 (2012); Act of December 27, 1974, Pub. L. No.
93-552, 88 Stat. 1757 (establishing the Law Revision Counsel); Congressional
Budget and Impoundment Control Act of 1974, Pub. L. No. 93-344, 88 Stat.
297 (codified as amended at 2 U.S.C. §§ 601-688 (2012)) (establishing the
Congressional Budget Office); Legislative Branch Appropriation Act of 1978,
Pub. L. No. 95-94, 91 Stat. 653 (1977) (giving statutory foundation to the
House Parliamentarian).
36. See Cross & Gluck, supra note 9, at 1555–60 (documenting offices’
common origins in reclaiming power from executive).

238 80 WASH. & LEE L. REV. 229 (2023)
that exists today: a bureaucracy with over 4,000 expert

professional staff,37 spread across eleven nonpartisan offices.38
To be sure, this congressional bureaucracy is not of
comparable size to its executive branch counterpart. Its staff
number in the thousands, not the millions.39 Yet, thanks to its
design as a nonpartisan center of expertise within a political
branch, the congressional bureaucracy still shares important
traits with its executive counterpart—traits that provide a
valuable foundation for comparison.40
The eleven offices that now comprise the congressional
bureaucracy and contribute directly to the legislative process
are as follows:
 Congressional Research Service (CRS): Congress’s
“think tank,” a research service that provides legal and
policy analysis of legislation and other issues.41
 Offices of the House and Senate Legislative Counsel
(Legislative Counsel): Legislative drafting offices in
each chamber.42
37. See id. at 1599–1600.

38. This Article excludes nonpartisan congressional offices that do not
regularly participate in the legislative process, such as the Government
Printing Office, the Office of Senate Legal Counsel and the Office of the
General Counsel of the House of Representatives, the Architect of the Capitol,
the Capitol Police, each chamber’s Sergeant at Arms, and each chamber’s
chaplain. On these offices, see IDA. A. BRUDNICK, CONG. RESEARCH SERV.,
RL33220, SUPPORT OFFICES IN THE HOUSE OF REPRESENTATIVES: ROLES AND
AUTHORITIES 1 (2020); IDA A. BRUDNICK, CONG. RESEARCH SERV., R43532,
OFFICES AND OFFICIALS IN THE SENATE: ROLES AND DUTIES 1 (2015).
39. Compare Cross & Gluck, supra note 9, at 1599–1600 (listing employee
numbers in congressional bureaucracy offices), with Vilorio, supra note 6
(noting over two million federal civilian workers).
40. See Cross & Gluck, supra note 9, at 1543–44.
41. See Legislative Reorganization Act of 1946, § 203, Pub. L. No. 79-601,
60 Stat. 812; see also STEPHEN W. STATHIS, CRS at 100, in CRS AT 100: THE
CONGRESSIONAL RESEARCH SERVICE: INFORMING THE LEGISLATIVE DEBATE SINCE
1914, at 9, 25 (2014).
42. See Legislative Reorganization Act of 1970, Pub. L. No. 91-510, 84
Stat. 1140 (codified at 2 U.S.C. § 281-282e) (House office); Revenue Act of 1918,
§ 1303(a), Pub. L. No. 254, 40 Stat. 1057, 1141–42 (Senate office).

 Office of the Law Revision Counsel (OLRC): Staff who

turn Congress’s enacted public laws into the U.S.
Code.43
 Congressional Budget Office (CBO): Economists and
analysts who provide influential economic analysis,
including estimates of the cost of all significant
legislation.44
 Joint Committee on Taxation (JCT): Committee with
nonpartisan staff that assists with all aspects of tax
legislation, including policy analysis, drafting
assistance, and all revenue estimates.45
 Offices of the House and Senate Parliamentarians
(Parliamentarians): The arbiters of congressional
procedure and rules in each chamber.46
 Government Accountability Office (GAO): Congress’s
“watchdog” over the executive branch that conducts
audits, performs policy research, and informs Congress
about the implementation of its laws.47
 Medicare Payment Advisory Commission (MedPAC) &
Medicaid and CHIP Payment and Access Commission
(MACPAC): Commissions with nonpartisan staff that
function as Congress’s overseers and advisors on the
Medicare, Medicaid, and CHIP programs.48
The staff in these eleven offices are distinct from the
sizeable partisan staff that populates the modern Congress,
43. See Committee Reform Amendments of 1974, H.R. Res. 988, 93d
Cong. § 405 (1974); Pub. L. No. 93-554, 88 Stat. 1771, 1777 (1974) (codified at
2 U.S.C. § 285c (2018)).
44. See Congressional Budget and Impoundment Control Act of 1974,
Pub. L. No. 93-344, tit. II, 88 Stat. 297 (1974).
45. See I.R.C. § 8001; see also Cross & Gluck, supra note 9, at 1545.
46. See Legislative Branch Appropriation Act of 1978, Pub. L. No. 95-94
§ 115, 91 Stat. 653, 668 (1977). The Senate Parliamentarian’s Office has no
organic statute specifying its responsibilities. See Cross & Gluck, supra note
9, at 1584.
47. See 31 U.S.C. §§ 701–705, et seq; see also About GAO: Overview, GOV’T
ACCOUNTABILITY OFF., https://perma.cc/P8TJ-DRW6; Cross & Gluck, supra
note 9, at 1545.
48. See Balanced Budget Act of 1997, Pub. L. 105–33, § 5022(c), 111 Stat.
251 (1997) (MedPAC); Children’s Health Insurance Program Reauthorization
Act of 2009, Pub. L. No. 111-3, § 1900(a), 123 Stat. 8 (MACPAC).

240 80 WASH. & LEE L. REV. 229 (2023)
which are not the focus of this Article.49 This partisan staff
similarly could be viewed as a legislative bureaucracy, of course.
Partisan congressional staff, however, lack certain features that
have motivated much conversation about bureaucratic
governance in the executive branch context, including
nonpartisanship,50 professionalization in expert fields,51 and
long tenures.52 For purposes of this Article, the term
“bureaucracy” therefore refers specifically to staff who are
employed in a nonpartisan capacity to contribute
professionalized expertise to the legislative process.
In recent years, the scholarly neglect of this congressional
bureaucracy has finally begun to change. This shift has
coincided with the rise of what Justice Barrett has labeled the
“process-based turn”53 in legislation scholarship: a movement to
investigate the inner workings of the modern Congress and
theorize its implications for statutory interpretation and
legislative reform.54 In the past several years, members of this
49. For the size and structure of partisan staffs, see R. ERIC PETERSEN &
AMBER HOPE WILHELM, CONG. RSCH. SERV., R43946, SENATE STAFF LEVELS IN
MEMBER, COMMITTEE, LEADERSHIP, AND OTHER OFFICES, 1977–2016 (2016); R.
ERIC PETERSEN & AMBER HOPE WILHELM, CONG. RSCH. SERV., R43947, HOUSE
OF REPRESENTATIVES STAFF LEVELS IN MEMBER, COMMITTEE, LEADERSHIP, AND
OTHER OFFICES, 1977–2016 (2016).
50. See, e.g., SKOWRONEK, supra note 19, at 47 (citing “political neutrality”
as a “hallmark” of civil service); see also Jesse M. Cross, Legislative History in
the Modern Congress, 57 HARV. J. ON LEGIS. 83 (2019) [hereinafter Cross,
Legislative History]; Cross & Gluck, supra note 9.
51. See, e.g., SKOWRONEK, supra note 19, at 47 (citing “recruitment by
criteria of special training or competitive examination” as “hallmark” of civil
service). On professionalized expertise differences between partisan and
nonpartisan congressional staffs, see Cross, Legislative History, supra note 50,
at 102–22; Cross & Gluck, supra note 9. On role of professionalized expertise
in executive branch, see Bruce A. Green & Rebecca Roiphe, Can the President
Control the Department of Justice?, 70 ALA. L. REV. 1, 50 (2018) (“As Stephen
Skowronek and other historians have argued, the birth of the administrative
state during the last decades of the nineteenth century coincided with a new
faith in professionalism and expertise . . . .”).
52. See, e.g., SKOWRONEK, supra note 19, at 47 (citing “tenure in office” as
“hallmark” of civil service). On tenure differences between partisan and
nonpartisan staff, see Cross, Legislative History, supra note 50, at 106–08
(outlining differences between partisan and nonpartisan staffs); Cross &
Gluck, supra note 9, at 1552 (same).
53. See Barrett, supra note 17, at 2193.
54. This movement is widely viewed as having been inaugurated by a
two-part study in 2014 by Abbe Gluck and Lisa Bressman. See Abbe R. Gluck

movement have directed attention to individual offices in the

congressional bureaucracy, highlighting the role that one office
or another now plays in shaping legislation for Congress.55
These efforts have been buttressed as scholars in other legal
fields have directed attention to individual offices, such as
observations about OLRC in law library studies56 and JCT in
tax scholarship.57 A few scholars also have looked at multiple
& Lisa Schultz Bressman, Statutory Interpretation from the Inside—An

Empirical Study of Congressional Drafting, Delegation, and the Canons: Part
I, 65 STAN. L. REV. 901, 901 (2013) [hereinafter Gluck & Bressman, Part I];
Lisa Schultz Bressman & Abbe R. Gluck, Statutory Interpretation from the
Inside—An Empirical Study of Congressional Drafting, Delegation, and the
Canons: Part II, 66 STAN. L. REV. 725, 725 (2014) [hereinafter Bressman &
Gluck, Part II].
55. See, e.g., Cross, Legislative History, supra note 50; Jesse M. Cross,
The Staffer’s Error Doctrine, 56 HARV. J. ON LEGIS. 83 (2019) [hereinafter
Cross, The Staffer’s Error]; Jesse M. Cross, When Courts Should Ignore
Statutory Text, 26 GEO. MASON L. REV. 453 (2018) [hereinafter Cross, When
Courts Should Ignore]; Abbe R. Gluck, Congress, Statutory Interpretation, and
the Failure of Formalism: The CBO Canon and Other Ways that Courts Can
Improve on What They Are Already Trying to Do, 84 U. CHI. L. REV. 177 (2017);
Gluck & Bressman, Part I, supra note 54; Abbe R. Gluck et al., Unorthodox
Lawmaking, Unorthodox Rulemaking, 115 COLUM. L. REV. 1789 (2015);
Jonathan S. Gould, Law Within Congress, 129 YALE L.J. 1946 (2020); Jarrod
Shobe, Enacted Legislative Findings and Purposes, 86 U. CHI. L. REV. 669
(2019); Daniel B. Listwa, Comment, Uncovering the Codifier’s Canon: How
Codification Informs Interpretation, 127 YALE L.J. 464 (2017); Jarrod Shobe,
Codification and the Hidden Work of Congress, 67 UCLA L. REV. 640 (2020);
Tobias A. Dorsey, Some Reflections on Yates and the Statutes We Threw Away,
18 GREEN BAG 2D 377 (2015); Tobias A. Dorsey, Some Reflections on Not
Reading the Statutes, 10 GREEN BAG 2D 283 (2007); Rebecca M. Kysar,
Dynamic Legislation, 167 U. PA. L. REV. 809 (2019); Rebecca M. Kysar,
Interpreting by the Rules, 99 TEX. L. REV. 1115 (2021).
56. See Shawn G. Nevers & Julie Graves Krishnaswami, The Shadow
Code, 112 LAW LIBR. J. 213, 215 (2020); Mary Whisner, The United States
Code, Prima Facie Evidence, and Positive Law, 101 LAW LIBR. J. 545, 554
(2009).
57. See, e.g., Rebecca M. Kysar, Tax Law and the Eroding Budget Process,
81 LAW & CONTEMP. PROBS., no. 2, 2018, at 61; Clinton G. Wallace,
Congressional Control of Tax Rulemaking, 71 TAX L. REV. 179 (2017); Ellen P.
Aprill & Daniel J. Hemel, The Tax Legislative Process: A Byrd’s Eye View, 81
L. & CONTEMP. PROBS., no. 2, 2018, at 100; George K. Yin, Crafting Structural
Tax Legislation in a Highly Polarized Congress, 81 J.L. & CONTEMP. PROBS.,
no. 2, 2018, at 241, 251–54; George K. Yin, How Codification of the Tax
Statutes and the Emergence of the Staff of the Joint Committee on Taxation
Helped Change the Nature of the Legislative Process, 71 TAX L. REV. 723,
725–26 (2018) (describing the growth of the JCT staff and the staff’s work on
codifying the tax statutes); George K. Yin, James Couzens, Andrew Mellon, the

242 80 WASH. & LEE L. REV. 229 (2023)
offices of the congressional bureaucracy, although not

necessarily to study the offices themselves.58 Against this
backdrop, Abbe Gluck and I recently published a comprehensive
look at this bureaucracy (“Cross-Gluck study”).59 The
Cross-Gluck study worked to systematically document the
traits, practices, and culture that spanned these nonpartisan
offices. That study provides the foundation for much of the
comparative examinations in Parts II through VII.
II. DUAL-ALLEGIANCE PROBLEM
By turning to ongoing discussions in executive-branch

scholarship, we can begin to understand how the congressional
bureaucracy generates new cross-disciplinary insights. For
example, several scholars have contended that bureaucracy in
America is inevitably subject to what might be termed the
“dual-allegiance problem.” This scholarship observes that
bureaucracy in America—here, assumed to be an administrative
agency—is subject to competing allegiances to Congress and the
President.60 This is believed to create a host of vexing problems,
“Greatest Tax Suit in the History of the World,” and the Creation of the Joint
Committee on Taxation and its Staff, 66 TAX L. REV. 787, 788 (2013)
(explaining how JCT became intertwined with complex tax issues).
58. See generally, e.g., Yin, supra note 34; Jarrod Shobe, Intertemporal
Statutory Interpretation and the Evolution of Legislative Drafting, 114 COLUM.
L. REV. 807 (2014).
59. Cross & Gluck, supra note 9, at 1543–44.
60. See Bruce Ackerman, Good-bye, Montesquieu, in COMPARATIVE
ADMINISTRATIVE LAW. 38, 41 (Susan Rose-Ackerman & Peter L. Lindseth eds.,
2010) [hereinafter Ackerman, Good-bye] (“[P]residents must compete for
control with an independently elected Congress. Legislative leaders have their
own weapons for pushing the bureaucracy in their direction . . . . ”); Kagan,
supra note 4, at 2273; Kirti Datla & Richard L. Revesz, Deconstructing
Independent Agencies (and Executive Agencies), 98 CORNELL L. REV. 769, 816
(2013); Ali Farazmand, BUREAUCRACY & ADMIN. 185 n.9 (2009); Jon D.
Michaels, An Enduring, Evolving Separation of Powers, 115 COLUM. L. REV.
515, 554 (2015) [hereinafter Michaels, Evolving Separation of Powers];
Randall L. Calvert et al., A Theory of Political Control and Agency Discretion,
33 AM. J. POL. SCI. 588, 589 (1989). Kagan also cites political scientists on this
cross-pressure. See Kagan, supra note 4, at 2385; see also Synar v. United
States, 626 F. Supp. 1374, 1398 (D.D.C. 1986).

including loss of presidential control over the bureaucracy,61

insertion of presidential loyalists into the bureaucracy,62
opportunities for strategic behavior by bureaucrats,63
undermining of presidential policies,64 and problematic shifts of
power within Congress.65 Bruce Ackerman has argued that this
dual-allegiance problem is so severe that it renders bureaucracy
fundamentally incompatible with our tripartite system of
government.66 The dual-allegiance problem therefore appears,
in this literature, both serious and unavoidable.
However, the dual-allegiance problem is specific to the
executive branch. As the Cross-Gluck study documented, the
congressional bureaucracy is solely accountable to Congress.67
With respect to appointment68 and removal69 of office heads,
hiring and firing of staff,70 funding of offices,71 day-to-day
61. See Ackerman, Good-bye, supra note 60, at 41; see also ANDREW B.
WHITFORD & GARY MILLER, ABOVE POLITICS: BUREAUCRATIC DISCRETION AND
CREDIBLE COMMITMENT 102 (2016).
62. See Ackerman, Good-bye, supra note 60, at 41; Ackerman, New
Separation of Powers, supra note 3, at 700; Jon D. Michaels, Of Constitutional
Custodians and Regulatory Rivals: An Account of the Old and New Separation
of Powers, 91 N.Y.U. L. REV. 227, 246 (2016) [hereinafter Michaels, Of
Constitutional Custodians].
63. See Kagan, supra note 4, at 2273.
64. See, e.g., Aziz Z. Huq, The President and the Detainees, 165 U. PA. L.
REV. 499, 500 (2017) (observing a “bureaucratic-legislative alliance” that
obstructed President Obama’s efforts to close the Guantanamo Bay detention
facility).
65. See generally J.R. DeShazo & Jody Freeman, The Congressional
Competition to Control Delegated Power, 81 TEX. L. REV. 1443 (2003).
66. See Ackerman, New Separation of Powers, supra note 3, at 702;
Ackerman, Good-bye, supra note 60, at 41; Katyal, supra note 5, at 2346.
67. See Cross & Gluck, supra note 9, at 1606, 1613.
68. The lone exception is GAO. See 31 U.S.C. § 703(a)(1) (explaining how
GAO’s Comptroller General is appointed by the President and subject to
Senate confirmation after the congressional commission recommends a list of
at least three candidates). The commissioners of MedPAC and MACPAC are
also selected by the Comptroller General. SOURCE.
69. GAO again provides the exception. See 31 U.S.C. § 703(e)(1)
(providing that the head of GAO is removable only by impeachment or, for
specified reasons, by joint resolution); see also Cross & Gluck, supra note 9, at
1606.
70. See Cross & Gluck, supra note 9, at 1613.
71. The congressional bureaucracy is subject to a legislative branch
appropriations process that, while requiring presidential signature, has
historically received significant presidential deference. See L. Anthony Sutin,

244 80 WASH. & LEE L. REV. 229 (2023)
supervision,72 and regular use of its services,73 a convergence of

rules and practices has created a bureaucracy that is
single-mindedly responsive to Congress.74 As a result, the
congressional bureaucracy is generally immune to the
cross-pressures that worry executive-branch scholars.
This basic observation opens up a variety of useful lessons.
First, it reveals new bureaucratic possibilities in our tripartite
system. If we are troubled by the dual-allegiance problem, we
might consider relocating additional bureaucratic tasks and
roles from the executive branch into the legislative branch. As
the Cross-Gluck study demonstrated, this is precisely how most
of the existing congressional bureaucracy came into existence:
the performance of certain tasks by the executive branch was
seen as problematic for one reason or another, and so Congress
reclaimed their performance.75 In this way, the congressional
bureaucracy opens new and unexplored solutions to existing
separation-of-powers problems, such as dual-allegiance.
It is important not to overstate the promise of this solution,
of course. For several reasons, the federal bureaucracy could
never be entirely relocated into Congress. The subset of agency
work that involves actual execution of laws—awarding grants,
conducting inspections, and so on—obviously must be performed
by executive branch actors. In rulemaking, nuanced application
of the law often entails varied application across regions and
evolving application over time—both aspects that, due to the
general and prospective nature of legislation, would be difficult
to assign to the congressional bureaucracy.76 Institutionally, the
congressional bureaucracy also has retained much of its current
Check, Please: Constitutional Dimensions of Halting the Pay of Public

Officials, 26 J. LEGIS. 221, 242 (2000).
72. Cross & Gluck, supra note 9, at 1613.
73. Use of the bureaucracy is generally dependent upon congressional
demand, as its services are mostly optional. The nearest possible exceptions
are when enacted statutes mandate the involvement of an office, which raises
complicated questions about executive-branch authority to create points of
order in Congress. See id. at 1628–30.
74. See id. at 1546.
75. See generally Cross, The Staffer’s Error, supra note 55. See also Cross
& Gluck, supra note 9, at 1555–60.
76. See Ann Woolhandler, Public Rights, Private Rights, and Statutory
Retroactivity, 94 GEO. L.J. 1015, 1016 (2006) (observing that in its “ideal
form[], legislation is prospective and general” while noting exceptions).

influence by developing close working relationships with the

legislators and partisan staffs in Congress77—relationships that
would be difficult to establish and preserve at a size anywhere
near that of the executive branch bureaucracy.78 Ackerman
therefore is correct to observe that, to some extent, the
dual-allegiance problem is unavoidable in our system. That does
not mean, however, that fatalism is justified. If the
dual-allegiance problem is concerning, expansion of the
congressional bureaucracy provides a means to minimize it.
Second, prior scholarship on the dual-allegiance problem
provides new ways to understand the congressional
bureaucracy. For example, the Cross-Gluck study documented
that nonpartisan hiring practices have flourished in the
congressional bureaucracy, even beyond what is required by
law.79 Scholarship on the dual-allegiance problem provides a
partial explanation for this phenomenon (which will be
discussed further in Part III). In the executive branch,
Ackerman has argued, the dual-allegiance problem is a key
motivating factor for Presidents to insert partisan loyalists into
executive agencies.80 Since Congress is immune to this
dual-allegiance problem, it stands to reason that it would have
less incentive to politicize its own bureaucracy. In this way,
dual-allegiance scholarship provides some explanation for an
otherwise unexplained dimension of the congressional
bureaucracy. At the same time, the lesson also runs in the
opposite direction: the congressional bureaucracy provides new
evidence to support Ackerman’s diagnosis of executive agency
politicization.
Third, the dual-allegiance problem creates new ways of
thinking about statutory doctrine. For example, Magill and
Vermeule have argued that courts should be sensitive to the
ways that interpretive doctrines create incentives for shifts of

78. The counterpoint, and one that might be used to think about how a
significant expansion of the congressional bureaucracy might operate, is GAO,
which has approximately 3,000 employees (and once had nearly 15,000). See
id. at 1588.
80. See id.

246 80 WASH. & LEE L. REV. 229 (2023)
work and responsibility in the executive bureaucracy.81 The

congressional bureaucracy raises the question: should courts
similarly be sensitive to the impact of their doctrines on inter-
branch bureaucratic arrangements? For instance, should they
incentivize arrangements that avoid problems such as dual
allegiance?
Courts often do—many times unwittingly—take
approaches to interpreting legislation that create incentives for
Congress to assign tasks either to the congressional bureaucracy
or to an executive-branch agency.82 Sometimes, for instance,
courts punish Congress for creating lengthy and intricate
statutes—such as when they adopt an unforgiving approach to
statutory errors, thereby transforming every statutory detail
into a potential weapon to undermine the statute.83 In so doing,
courts effectively incentivize Congress to produce more brief,
open-ended statutes—and to thereby shift work to executive
branch agencies. Other times, courts might punish Congress for
vague statutory language that delegates a decision to an
administrative agency—for example, under a robust
nondelegation doctrine.84 In so doing, they incentivize Congress
to create long, detailed statutes, and to thereby shift work to the
congressional bureaucracy. In other words, the level of
specificity in a federal statute is, in part, a decision about which
bureaucracy should handle policy specification. When statutory
doctrine encourages a particular level of specificity, therefore, it
incentivizes use of one bureaucracy over another.
This is not how interpretive and other statutory doctrines
typically are understood. Consider the nondelegation doctrine.
Often viewed as a vessel for conservative frustrations with the
81. See Elizabeth Magill & Adrian Vermeule, Allocating Power Within
Agencies, 120 YALE L.J. 1032 (2011); see also Ingber, supra note 7, at 181 (“Like
Congress, the courts also allocate power to different actors within the
executive branch both implicitly and explicitly. They do so through a variety
of doctrinal mechanisms and canons of interpretation.”); Jody Freeman &
Adrian Vermeule, Massachusetts v. EPA: From Politics to Expertise, 2007 SUP.
CT. REV. 51; Michaels, Of Constitutional Custodians, supra note 62, at 273–74.
83. See id. See generally King v. Burwell, 576 U.S. 473 (2015).
84. For recent opinions signaling that the Court is interested in reviving
a strict nondelegation doctrine, see Gundy v. United States, 139 S. Ct. 2116,
2137–42 (2019) (Gorsuch, J., dissenting); id. at 2131 (Alito, J., concurring in
the judgment).

rise of the administrative state, the nondelegation doctrine is

assumed to be a judicial tool to return to a smaller—and perhaps
even pre-bureaucratic—version of federal government.85 Once
the congressional bureaucracy is taken into account, however, it
becomes clear that the much-anticipated revival of this
doctrine86 instead (or additionally) could have a hydraulic effect:
by forcing a reduced use (and perhaps concomitant shrinking) of
the executive branch bureaucracy, it may lead to increased use
and expansion of the congressional bureaucracy.
Given these dynamics, should courts design doctrines with
the aim of incentivizing a healthy federal bureaucracy? If we
define such a bureaucracy as one that is immune to the
dual-allegiance problem, for instance, should courts therefore
grant Congress greater leeway when errors appear in highly
detailed statutes,87 or craft a harsher nondelegation doctrine?
With respect to inter-branch bureaucratic allocations, it is
not clear that courts should be doing this. The Constitution
emphasizes insulating congressional practice from outside
influence.88 This is one way the Magill-Vermeule framework
changes when widened to the congressional context: it entails
judicial tinkering with internal congressional dynamics that
arguably are meant to be left alone.
If courts are going to account for these dynamics, however,
they must do so responsibly. Consider the concurring opinion in
a recent Sixth Circuit case.89 That opinion cited the Cross-Gluck
85. See, e.g., Noah Feldman, This Supreme Court Decision Should Worry
the EPA and FDA, BLOOMBERG (June 22, 2019); see also Gundy, 139 S. Ct. at
2117 (“Indeed, if SORNA’s delegation is unconstitutional, then most of
Government is unconstitutional—dependent as Congress is on the need to give
discretion to executive officials to implement its programs.”). But see Daniel E.
Walters, Decoding Nondelegation After Gundy: What the Experience in State
Courts Tells Us About What to Expect When We’re Expecting, 71 EMORY L. REV.
417 (2022).
86. See supra notes 84 & 85 and accompanying text; see also Nicholas R.
Parrillo, A Critical Assessment of the Originalist Case Against Administrative
Regulatory Power: New Evidence from the Federal Tax on Private Real Estate
in the 1790s, 130 YALE L.J. 1288, 1294 (2021).
87. For a methodology to accomplish this, see Cross, The Staffer’s Error,
supra note 55 (outlining a “staffer’s error doctrine” based on Court’s
interpretation in King).
88. See infra note 329 and accompanying text.
89. Tiger Lily, LLC v. U.S. Dep’t of Hous. & Urb. Dev., 5 F.4th 666, 675
(6th Cir. 2021) (Thapar J., concurring).

248 80 WASH. & LEE L. REV. 229 (2023)
study to suggest that its findings straightforwardly support a

robust nondelegation doctrine (since it showed that Congress
has the internal capacity to assume executive branch
bureaucratic tasks). Yet this argument has several issues. First,
as already mentioned, not all tasks delegated to agencies could
be performed by a congressional bureaucracy—and we have
seen no effort by courts to limit the doctrine to those that could
be so performed. This undermines the concurrence’s views about
the continuity of good governance that could be achieved under
a robust nondelegation doctrine. Second, courts have not
reckoned with the impacts that these bureaucratic shifts would
have on legislation. That relocation of expertise into the
legislative process could change the nature of federal legislation:
more statutes might resemble the Medicare statute90 (an
over-1,000 page labyrinth of hyper-detailed payment formulas)
than, for example, the Natural Gas Act (a more readable but
cryptic statute with open-ended terms, such as references to
“just and reasonable” rates).91 Such a change would significantly
reduce the readability of federal statutes for ordinary readers—
a value the Court has touted, as seen in its recent emphasis on
preserving fair notice in statutory interpretation.92 In this area,
in other words, judicial doctrine simultaneously impacts the
allocation of tasks across bureaucracies and the nature of
federal statutory law. As the courts consider wading back into
nondelegation jurisprudence, it is not clear that they have a firm
grasp on how to balance those effects to provide both a
functional bureaucratic government and a stable, accessible
legal regime.93 An introduction of the congressional bureaucracy
into the conversation, and a realistic understanding of its
distinctive traits and limitations, makes all this visible.
90. See generally Social Security Act of 1935, Pub L. No. 74-271, 49 Stat.
620 (codified at 42 U.S.C. §§ 301–1397mm).
91. 15 U.S.C. § 717c(a).
92. See, e.g., Niz-Chavez v. Garland, 141 S.Ct. 1474, 1481–82 (2021)
(“[A]ffected individuals and courts alike are entitled to assume statutory terms
bear their ordinary meaning.”); Bostock v. Clayton Cnty., 140 S. Ct. 1731, 1825
(2020) (Kavanaugh, J., dissenting) (“Citizens and legislators must be able to
ascertain the law by reading the words of the statute.”).
93. See supra notes 84, 85–86 and accompanying text.

III. BUREAUCRATIC RESISTANCE
In executive-branch scholarship, a wide-ranging discussion

has unfolded about bureaucratic resistance—that is, about the
capacity of nonpartisan bureaucrats to pursue an agenda at
odds with that desired by their partisan superiors.94 This
discussion has been both descriptive and normative.95 On the
descriptive side, it has examined the civil servant protections
that have enabled bureaucratic resistance, and it has catalogued
the toolkit that bureaucrats can use to resist the wishes of their
political bosses.96 On the normative side, it has debated the
merits and demerits of bureaucratic resistance, weighing
whether it should be viewed as a positive or negative aspect of
our system.97 Each of these discussions sheds light on, and is
illuminated by, the congressional bureaucracy.
A. Descriptive Questions: Bureaucratic Protections
First, consider the ongoing effort to catalogue and

understand the institutional protections that enable
bureaucratic resistance. Initially, this dimension of the
congressional bureaucracy appears confusing. Offices in the
congressional bureaucracy are not all afforded the same
statutory protections from partisan tampering.98 Of the eleven
offices, six have statutory requirements for nonpartisan
appointment of office heads,99 six have such requirements for
staff hiring,100 eight provide a role for political actors in the
94. I borrow the term “bureaucratic resistance” most directly from

Rebecca Ingber. See generally Ingber, supra note 7. For her part, Ingber
suggests that some accounts of this resistance are too simplistic in their
depiction of a powerful bureaucracy untethered from constraint and uniformly
positioned to resist the President. See id. at 54, 162–63.
95. Id.
96. Id.
97. Id.
99. See id. at 1613. Commissioners of MedPAC and MACPAC are
appointed by the Comptroller General, and staff are appointed by the
commissioners, subject to Comptroller General oversight. Id. Neither has
statutory requirements for nonpartisan appointment or hiring. Id.
100. Id.

250 80 WASH. & LEE L. REV. 229 (2023)
appointment of office heads,101 and five provide a role for such

actors in staff hiring, while three explicitly omit such a role.102
One office—the Senate Parliamentarian—simply has no organic
statute to protect it whatsoever.103 Nonetheless, interviewees in
the Cross-Gluck study reported consistent displays of
nonpartisan autonomy across all of these offices.104 The
Cross-Gluck study therefore presents a puzzle: why has the
congressional bureaucracy developed a relatively uniform
capacity for nonpartisan bureaucratic resistance, despite varied
statutory protections?
This puzzle is deepened by situations in which the
congressional bureaucracy has retained statutory protections
yet lost real-world powers. The GAO experience with demand
letters furnishes an example.105 GAO has statutory authority to
take agencies to court to access requested documents from them
and, prior to doing so, GAO typically issues a “demand letter”
requesting the documents and otherwise threatening to pursue
judicial remedy.106 While GAO previously exercised this power
with regularity, Vice President Cheney refused to comply with
one such demand letter—and GAO subsequently stopped
issuing demand letters.107 While GAO’s statutory authority was
unchanged, it seemingly lost a real-world tool of bureaucratic
resistance.108 This again begs the question: why the mismatch
between statutory protection and actual power?
Executive-branch scholarship can help solve this puzzle.
Not surprisingly, some executive-branch literature has
examined the hard statutory protections afforded to civil
101. Id.
102. Id.
103. See 2 U.S.C. § 6539. The Senate Parliamentarian exists under the
statutory authority of the Secretary of the Senate. Id. There also is a statutory
provision setting forth the maximum compensation for the Senate
Parliamentarian. Id. § 6535.
105. See Walker v. Cheney, 230 F. Supp. 2d 51, 66, 73 n.19 (D.D.C. 2002).
106. Interview with Staffer (on file with author).
107. See Walker 230 F. Supp. 2d at 57–58; Interview with Staffer (on file
with author).
108. But see U.S. GOV’T ACCOUNTABILITY OFF., GAO-19-55G, GAO’S AGENCY
PROTOCOLS 24 (2019), https://perma.cc/J8NH-S7QS (PDF) (asserting
continued demand letter authority and citing 2017 congressional reiteration
of it).

servants109—protections against partisan hiring, salary

determinations, and removal110 created in response to the spoils
system of the late 1800s,111 as well as additional statutory
protections for whistleblowers, among others.112 This
scholarship has noted, however, that statutory protections in
the executive branch have a similarly uneven quality: not all
agencies have them, or have them in same measure, and the
protections furnish insufficient explanations of the real-world
autonomy possessed by executive branch bureaucrats.113 In the
context of the Department of Justice, one study described this
as the “historical puzzle” and “historical paradox” of DOJ:
namely, that the Department is “structurally accountable to
presidential power . . . and yet it has developed strong norms of
109. Rachel Barkow notes, for example, that an “obsessive focus on

removal as the touchstone of independence” has “spawned countless law
review articles.” Rachel E. Barkow, Insulating Agencies: Avoiding Capture
Through Institutional Design, 89 TEX. L. REV. 15, 16–17 (2010); see also
SKOWRONEK, supra note 19, at 47 (citing Max Weber, Bureaucracy, in FROM
MAX WEBER: ESSAYS IN SOCIOLOGY 196, 196–244 (H.H. Gerth & C. Wright Mills
eds. & trans., 1958)). But see Michaels, Of Constitutional Custodians, supra
note 62, at 286.
110. For relevant protections, see Civil Service Reform Act of 1978, Pub.
L. No. 95-454, 92 Stat. 1111 (codified as amended at 5 U.S.C. §§ 7321–7326);
Pendleton Act, ch. 27, 22 Stat. 403 (1883) (codified as amended at 5 U.S.C.
§§ 3104–7212); Hatch Act of 1939, Pub. L. No. 76-252, 53 Stat. 1147 (codified
as amended at 5 U.S.C. § 118 & 18 U.S.C. § 61); 5 U.S.C. § 2102(a)(1); 5 U.S.C.
§§ 7513(a), 7521(a)–(b); 5 U.S.C. § 4303; 5 U.S.C. §§ 2301–2302. For
office-specific protections, see, e.g., 15 U.S.C. § 41 (removal protections for FTC
head); 29 U.S.C. § 153 (removal protections for NLRB head). See also JARED P.
COLE, CONG. RESEARCH SERV., R44803, THE CIVIL SERVICE REFORM ACT: DUE
PROCESS AND MISCONDUCT-RELATED ADVERSE ACTIONS 1 (2017),
https://perma.cc/K34S-D5WQ.
111. See COLE, supra note 110, at 1; Ingber, supra note 7, at 176; Jed
Handelsman Shugerman, The Creation of the Department of Justice:
Professionalization Without Civil Rights or Civil Service, 66 STAN. L. REV. 121,
at 143–59 (2014); Green & Roiphe, supra note 51, at 51.
112. See Civil Service Reform Act of 1978, Pub. L. No. 95-454, 92 Stat. 1111
(codified as amended at 5 U.S.C. §§ 7321–7326); Whistleblower Protection Act
of 1989, Pub. L. No. 101-12, 103 Stat. 16 (codified as amended in scattered
sections of 5 U.S.C.); 5 U.S.C. § 2302(b)(8) (protection against retaliation); see
also LOUIS FISHER, CONG. RESEARCH SERV., RL33215, NATIONAL SECURITY
WHISTLEBLOWERS (2005).
113. See Datla & Revesz, supra note 60, at 772; Andrew Kent, Congress
and the Independence of Federal Law Enforcement, 52 U.C. DAVIS L. REV. 1927,
1940 n.48 (2019); see also Daphna Renan, Presidential Norms and Article II,
131 HARV. L. REV. 2187, 2222 (2018); Green & Roiphe, supra note 51, at 37.

252 80 WASH. & LEE L. REV. 229 (2023)
professional independence.”114 Scholars also have noted the

inverse puzzle in the executive branch context: real-world
bureaucratic autonomy often can be undermined even as
statutory protections persist.115
To solve this historical puzzle, executive-branch scholars
have directed attention to the non-statutory forces that also
contribute to bureaucratic autonomy.116 To this end, they have
examined the “norms”117 and “conventions”118 that often
supplement statutory protections. This work has studied
agencies and offices where non-statutory protections play an
especially prominent role, including the DOJ,119 FBI,120 and
what Margo Schlanger terms “Offices of Goodness” inside
individual agencies.121
To appreciate the relevance of this work to the
congressional bureaucracy, consider the recent suggestion by
Jonathan Gould, in his outstanding study of the
Parliamentarians’ offices in Congress, of extending civil servant
statutory protections to the employees of these offices.122 This
114. See Shugerman, supra note 111, at 125.

115. See Katyal, supra note 5, at 2332.
116. See, e.g., Gillian E. Metzger, The Interdependent Relationship
Between Internal and External Separation of Powers, 59 EMORY L.J. 423, 430
(2009) (“Internal constraints can also take a ‘soft’ form, being rooted more in
agency traditions and culture than ‘hard’ structural features.”); Daniel Hemel,
President Trump vs. the Bureaucratic State, YALE J. ON REG.: NOTICE &
COMMENT (Feb. 18, 2016), https://perma.cc/KF42-3YW8 (“[W]e should think
not only in formal separation-of-powers terms but also in practical terms of
bureaucratic drift.”); Daniel A. Farber & Anne Joseph O’Connell, Agencies as
Adversaries, 105 CALIF. L. REV. 1375, 1393 (2017); Ingber, supra note 7, at 161;
Ackerman, New Separation of Powers, supra note 3, at 690.
117. See Renan, supra note 113, at 2189.
118. See Vermeule, supra note 7, at 1166.
119. See id. at 1201–03; Renan, supra note 113, at 2207–14; Ingber, supra
note 7, at 188 (noting the “powerful norms under which DOJ and the offices
within it expect—and are expected—to be shielded to differing degrees from
partisan politics”); Kent, supra note 113, at 1930; Shugerman, supra note 111,
at 125; Todd David Peterson, Federal Prosecutorial Independence, 15 DUKE J.
CONST. L. & PUB. POL’Y 217, 262 (2020); Green & Roiphe, supra note 51, at 6
(“[L]awyers’ professional norms provide a basis for internal limits on the
President’s power.”).
120. Renan, supra note 113, at 2210–11; Ingber, supra note 7, at 184–86.
121. Margo Schlanger, Offices of Goodness: Influence Without Authority in
Federal Agencies, 36 CARDOZO L. REV. 53, 54–55 (2014).
122. Gould, supra note 55, at 2026–27.

may indeed be wise, and some have suggested similar

extensions in the executive branch.123 Yet the executive-branch
literature also cautions that civil service protections may be
inadequate to generate an independent and energized
bureaucracy,124 and its scholars have done important work to
widen our understanding of what protects real-world autonomy
beyond civil service rules. Efforts to protect or expand the
congressional bureaucracy can benefit from this widened lens
and think beyond statutory protections.
For example, scholars have emphasized a structural feature
that can be as important as statutory protections: locating
bureaucracy in a centralized “institutional home” rather than
scattering it across partisan operations.125 Jed Shugerman has
argued that the Department of Justice was founded specifically
around this idea, with a centralized legal office that replaced
attorneys scattered across various agencies.126 The Cross-Gluck
study did not find similar intent in the design of the
congressional bureaucracy, but it did find a similar effect:127
Congress’s bureaucrats today are clustered into nonpartisan
offices rather than embedded in partisan structures such as
committees or Leadership,128 and this has contributed to
bureaucratic autonomy for the staffers in these offices.129 On the
one hand, therefore, the experience of the congressional
bureaucracy adds to the work done by Shugerman and others: it
suggests that their lessons about institutional structure apply
beyond the executive branch. On the other hand, their theories
help us understand bureaucratic autonomy in the congressional
bureaucracy—and they pinpoint elements beyond statutory
protections that should be borne in mind by those who wish to
preserve this autonomy, or to create additional spaces for
123. See, e.g., Kent, supra note 113, at 1974–82; Peterson, supra note 119,
at 285; Peter M. Shane, Prosecutors at the Periphery, 94 CHICAGO-KENT L. REV.
241, 241–45 (2019); see also Ackerman, New Separation of Powers, supra note
3, at 692.
124. Katyal, supra note 5, at 2332; Vermeule, supra note 7, at 1166.
125. See Schlanger, supra note 121, at 54–55.
126. Shugerman, supra note 111, at 150; id. at 163; see also Green &
Roiphe, supra note 51, at 49.
128. Id.
129. Id.

254 80 WASH. & LEE L. REV. 229 (2023)
nonpartisan expertise inside Congress. Such individuals may

want to think not in terms of creating positions in Congress, it
suggests, but of creating offices.
In these studies of norms that buttress bureaucratic
autonomy, it also has been observed that such norms are “more
robust” when supported by norm enforcers.130 Applied to the
congressional bureaucracy, this observation generates insights
regarding norm enforcers both inside and outside Congress.
Regarding those outside Congress: unlike the civil service, the
congressional bureaucracy benefits from almost no outside norm
enforcers. Each of the most likely candidates, lobbyists and the
executive branch, has a competitive interest in taking over the
work of the congressional bureaucracy, a fact that typically
prevents each from loudly insisting on the preservation and
steady use of this bureaucracy.131 The general public, which also
might serve as a norm enforcer (though perhaps a less effective
one),132 is largely unaware of the congressional bureaucracy, and
therefore is ill-equipped to enforce norms regarding its use.133
This defies some recent executive-branch scholarship, which has
argued that a norm-dependent bureaucracy cannot survive
without significant support from outside norm enforcers.134
What about norm enforcers inside Congress? Here, the
primary norm enforcers are powerful partisans. Typically, these
partisans have required—or formally urged—use only of those
offices in the congressional bureaucracy that play adjudicatory
roles. This is seen, for example, in formal directions to gather
130. Renan, supra note 113, at 2203 (“Structural norms may be more
robust, then, when they come to be expected and desired by pluralistic
communities or potential norm enforcers.”).
131. See Cross & Gluck, supra note 9.
132. Renan, supra note 113, at 2205. Renan does believe that media
education about breaches of norms can help the public play the role of norm
enforcement, however. See id. at 2215, 2241.
133. This ignorance is actively fostered by Members of Congress, who often
perform the steps of lawmaking from a generation ago—referring to bills they
“wrote,” acting as though hearings are spontaneous factfinding sites—and
thereby obscure the bureaucracy’s contributions. On the repurposed role these
legislative steps now fill, see Cross, supra note 50, at 139–50; Cross & Gluck,
supra note 9, at 1641–42.
134. See, e.g., Schlanger, supra note 121, at 59; see also Ingber, supra note
7, at 189.

and use CBO and JCT cost estimates.135 This trend is

understandable: nonpartisan experts are useful referees for
otherwise contentious partisan disputes.136 However,
unfavorable adjudications impose costs on partisans in these
situations,137 and this inevitably provokes partisans to examine
whether those costs are worth the benefits.138 By contrast, when
the congressional bureaucracy plays information-sharing139 and
advisory roles,140 it offers benefits with little cost, other than
possible delay. This highlights an additional pathway for
strengthening the norm of bureaucratic autonomy and
resistance in Congress: powerful partisans could more formally
urge the use of the congressional bureaucracy in its advisory and
information-sharing roles. This might resemble the instruction
issued by the House Appropriations Committee that
amendments should be drafted by House Legislative Counsel,141
and the formal urging by the House Rules Committee of the
same.142 Here, executive-branch scholarship once again
135. See Congressional Budget and Impoundment Control Act of 1974,

Pub. L. No. 93-344, § 402, 88 Stat. 297; RULES OF THE HOUSE OF
REPRESENTATIVES XIII.3(c)(3) (116th Cong.). For tax legislation, CBO
publishes the JCT estimate. See Pub. L. No. 99-177 § 273 (codified at 2 U.S.C.
§ 601(f) (2018)).
136. See Staffer Interview (on file with author) (noting the use of
Legislative Counsel as useful tool to adjudicate otherwise contentious
competing interpretations of bill language).
137. See Ingber, supra note 7, at 186; see also Jennifer Nou, Bureaucratic
Resistance from Below, YALE J. ON REG.: NOTICE & COMMENT (Nov. 16, 2016);
Schlanger, supra note 121, at 94–95; Vermeule, supra note 7, at 1210.
138. See, e.g., Alan Rappeport, C.B.O. Head, Who Prizes Nonpartisanship,
Finds Work Under Attack, N.Y. TIMES (June 19, 2017), https://perma.cc/GZ44-
WRAN.
139. CRS is an example of an office that plays a more straightforward
information-developing and information-sharing role. See 2 U.S.C. § 166(d)
(listing CRS functions).
140. Legislative Counsel is an example of an office with a more strictly
advisory role. See 2 U.S.C. § 281a (2018) (providing for the office to have an
“attorney-client relationship” with Members and committees).
141. See Amendment Resources, HOUSE OF REPRESENTATIVES COMM. ON
RULES, https://perma.cc/DE6Z-GHVE (“The assistance of the Office of the
Legislative Counsel . . . should be sought in drafting [all amendments
submitted to the House Committee on Rules].”).
142. See id.; Amending Appropriations Bills—A Basic Guide Presented by
the Committee on Rules, HOUSE OF REPRESENTATIVES COMM. ON RULES BLOG
https://perma.cc/2678-FUZY (last visited Jan. 28, 2021).

256 80 WASH. & LEE L. REV. 229 (2023)
highlights new pathways to bolstering congressional

bureaucratic resistance.
Rachel Barkow has noted another element that, in the
agency context, buttresses bureaucratic resistance.143 Much
scholarship on agency leaders, she notes, has focused on the
different statutory protections for appointees in independent
versus politically accountable agencies..144 In her effort to
broaden this lens, Barkow has observed: “One way to create
greater independence is to specify qualifications for appointees
so that the pool of potential candidates from which the President
picks is more limited and he or she cannot select solely on the
basis of partisan leanings.”145 She notes that the constraints can
be endemic to the position (for example, the President is
practically limited in selecting FDA heads to those with
scientific expertise) or statutorily imposed (for example, some
members of Surface Transportation Board statutorily must
have a professional background in transportation).146
This analysis partly explains the autonomy of office heads
in the congressional bureaucracy. Most of these office heads are
selected by a partisan actor. Only three office heads in the
congressional bureaucracy are not appointed directly by a
partisan actor; the majority are appointed by the House
Speaker,147 the Senate President Pro Tempore,148 or both.149 Yet
these appointments have remained largely resistant to partisan
pressure, and tellingly, they are subject to both statutory and
endemic constraints. Statutorily, selections are often subject to
a requirement to hire without regard to partisan
143. Barkow, supra note 109, at 47–48.

144. See id. at 17.
145. Id.
146. Id.
147. Those appointed by House Speaker: House Legislative Counsel; Law
Revision Counsel; House Parliamentarian.
148. Appointed by Speaker Pro Tempore: Senate Legislative Counsel.
149. Appointed by House Speaker and Senate President pro tempore, with
recommendations from Budget Committee: CBO. The head of GAO is
appointed by the President with Senate confirmation after congressional
submission of recommendations, the head of the JCT staff is appointed by the
Joint Committee, and the head of CRS is appointed by the Librarian of
Congress.

affiliation150—a requirement that presumably could not be

applied to many executive branch positions, though at least one
scholar has argued for their extension to the FBI Director and
to U.S. Attorneys.151 In the congressional bureaucracy, however,
it is difficult to see how these provisions would be meaningfully
enforced. The more illuminating constraints, therefore, may be
endemic.
In the congressional bureaucracy, a practical constraint
does often limit these appointments: in many cases, few outside
the office have the requisite expertise to perform—or
meaningfully oversee—the office’s work. In this sense, parts of
the congressional bureaucracy are subject to an even more
extreme version of the endemic constraints that Barkow
identifies. As a result, office heads in the congressional
bureaucracy often are selected through internal promotion of
existing employees.152 This is not uniformly true, and Barkow’s
analysis provides a rationale for the pattern: offices with skills
that can be cultivated elsewhere are more likely to sometimes
have outside leadership. For example, the economic analysis
performed by CBO is not unique to that office,153 and the
auditing and analysis practiced by GAO can be learned
elsewhere.154 By contrast, there is virtually nowhere to learn
congressional drafting practices other than Legislative Counsel,
to learn chamber procedure and precedent other than a
Parliamentarian’s office, or to master codification outside
OLRC. As a result, these latter offices particularly have
constraints endemic to the position.
This practical limitation changes the dynamics of
bureaucratic resistance. Consider the Senate Parliamentarian
in the 1980s and 1990s. This was a rare situation in which
partisans, due largely to political frustrations, exercised
removal power over an office head. When party control of the
Senate shifted in the 1980s and 1990s, the incoming leadership
150. For list of these offices and the applicable statutory requirements, see
Cross & Gluck, supra note 9, at 1613.
151. See, e.g., Kent, supra note 113, at 1975–76.
153. For a description of CBO economic analysis, see About CBO:
Processes, CONG. BUDGET OFF., https://perma.cc/22WN-Y9S5.
154. See Role as an Audit Institution, GOV’T ACCOUNTABILITY OFF.,
https://perma.cc/67F5-JZQD.

258 80 WASH. & LEE L. REV. 229 (2023)
regularly installed a new Parliamentarian.155 This practice

began when Republicans regained control of the Senate in 1981
after decades in the minority, during which they had come to
view the preceding Parliamentarian as overly congenial to the
Democrats.156 Democrats similarly would remove the
Republican-installed Parliamentarian upon regaining the
Senate in 1987, creating a practice that would continue through
the 1990s.157 In each instance, the replacement Parliamentarian
was taken from inside the (exceedingly small) Senate
Parliamentarian’s office.158 Admittedly, there was one
documented effort to install an outsider—a partisan aide—in
the position, with the outsider refusing.159 Yet the role would be
difficult for an outsider to perform in a meaningful way; mastery
of chamber procedural precedent is a skill cultivated almost
exclusively in the Parliamentarian’s office.160 As a result, a
trend of partisan removal in the 1980s led to the position simply
alternating between two long-tenured members of the office,
Robert Dove and Alan Frumin,161 a practice that did not seem to
significantly alter the office’s work product162 or undermine its
long-term credibility.163
155. See ANDREA C. HATCHER, MAJORITY LEADERSHIP IN THE U.S. SENATE 32

(2010); Gould, supra note 55, at 2005; James I. Wallner, Parliamentary Rule:
The U.S. Senate Parliamentarian and Institutional Constraints on Legislator
Behavior, 20 J. LEGIS. STUD. 380, 392 (2014); David E. Rosenbaum, Rules
Keeper is Dismissed by Senate, Official Says, N.Y. TIMES (May 8, 2001),
https://perma.cc/HP9X-9GBP. In 2001, Senate Majority Leader Trent Lott also
dismissed the Parliamentarian after disagreeing with rulings on tax and
budget bills and reinstalled a prior Parliamentarian. See STUART ALTMAN &
DAVID SHACTMAN, POWER POLITICS, AND UNIVERSAL HEALTH CARE (2011);
Gould, supra note 55, at 2006. Since then, the position has stabilized, with
each of the next two Senate Parliamentarians surviving party transitions. See
Gould, supra note 55, at 2006.
156. See Gould, supra note 55, at 2005.
157. See id.
158. The office currently consists of three people. See Cross & Gluck, supra
note 9, at 1600.
160. As one member of a Parliamentarian’s office has put it: “Our
knowledge isn’t replicated anywhere else.” Cross & Gluck, supra note 9, at
1583.
161. See supra note 155 and accompanying text.
162. See ALTMAN & SHACTMAN, supra note 155.

In this way, the congressional bureaucracy confirms and

extends Barkow’s thesis about the relationship between
endemic appointment limitations and functional independence.
Barkow’s thesis also helps explain why, unlike most presidential
appointments, the heads of the offices in the congressional
bureaucracy are widely viewed as nonpartisan.164 And Barkow’s
thesis provides important lessons for creators of future
legislative offices, who should not assume that a culture of
nonpartisan appointments will necessarily translate to future
offices if those offices will deploy forms of expertise more readily
cultivated elsewhere (for example, expertise in the form of
substantive policy knowledge found in agencies). For such
offices, more detailed statutory constraints on selection of office
heads, or at least informal commitments to promote leadership
from within, may be especially valuable.
B. Descriptive Questions: Tools of Bureaucratic Resistance
Scholars also have documented the different tools of

resistance that, in the executive branch, nonpartisan
bureaucrats have at their disposal.165 For the congressional
bureaucracy, several factors conspire to make one such tool
uniquely important: the persuasive power.166
164. See id. at 1613–16.

165. See, e.g., Ingber, supra note 7, at 205
The potential mechanisms of resistance . . . range from the anodyne (asking
a question, raising a concern), to the more assertive (seek to slow-roll, ‘build
a record,’ bring to the attention of superiors or ‘offices of goodness’ inside the
government), to the most aggressive (refuse to act, bring to the attention of
congressional overseers, leak).
id. at 163–65 (noting additional features identified by Albert Hirschman, and
by John Brehm and Scott Gates); Nou, supra note 137 (listing tools of slowing
down, record-building, leaking, enlisting internal inspectors general, using
allies, suing the agency, and resigning); Michaels, Of Constitutional
Custodians, supra note 62, at 237 (noting the tool of “exercising discretion
when implementing and administering programs on the ground”); Schlanger,
supra note 121, at 58–59 (noting tools of “Offices of Goodness” as “inclusion in
policy formulation working groups, clearance authority, giving advice,
providing training and technical assistance, undertaking program or
operational review, complaint investigation, outreach to outside groups,
generation of documents, and congressional reporting”).
166. See Cross & Gluck, supra note 9, at 1625–28 (discussing where
congressional bureaucracy offices fall on the authoritative-versus-permissive
scale).

260 80 WASH. & LEE L. REV. 229 (2023)
In part, this reliance on persuasion is because the

congressional bureaucracy does not have access to other tools
available in the executive branch context. For example, civil
servants often have the power to undertake unilateral action
that, as a practical matter, their partisan superiors cannot
review.167 This sometimes occurs because they are frontline
implementers, and therefore are the last relevant actors.168 It
also occurs because civil servants significantly outnumber their
political overseers, a fact which makes it functionally impossible
to perform oversight of all of their actions.169 Neither typically
is true of the congressional bureaucracy. In most instances, the
congressional bureaucracy provides an intermediate step before
final partisan action (legislative voting), so it is not the last
relevant actor.170 And the congressional bureaucracy is matched
167. See Michaels, supra note 62, Of Constitutional Custodians, at 236

Agency leaders cannot run agencies by themselves. Because an agency’s
responsibilities are sufficiently great, complex, and variegated, the
relatively small and often inexperienced group of appointed leaders must
necessarily rely on lower-level government employees to help with the
research, design, promulgation, implementation, and enforcement of
administrative policies.
Ingber, supra note 7, at 191 (“Constant oversight creates more work for
superior officers, who cannot possibly watch over the shoulder of every line
official beneath them as they go about their daily tasks.” (citing JOHN BREHM
& SCOTT GATES, WORKING, SHIRKING, AND SABOTAGE: BUREAUCRATIC RESPONSE
TO A DEMOCRATIC PUBLIC 25–46 (1997))); Kagan, supra note 4, at 2250 (“[N]o
President (or his executive office staff) could, and presumably none would wish
to, supervise so broad a swath of regulatory activity.”); STEPHEN BREYER,
MAKING OUR DEMOCRACY WORK: A JUDGE’S VIEW 110 (2010).
168. See Ingber, supra note 7, at 175 (“It is well understood that even
lower-level bureaucrats have a certain degree of functional power, whether or
not it is described as such, simply by virtue of their frequent position on the
front lines . . . ”); Michaels, supra note 62, Of Constitutional Custodians, at 237
(noting that “one way they help shape that policy is by exercising discretion
when implementing and administering programs on the ground”); Michaels,
supra note 62, at 542 (“[T]he coterie of agency leaders cannot actually
administer programs on the ground, where, once again, heavily relied-upon
civil servants have some say over how policy is actually implemented and
enforced.”).
169. See Michaels, supra note 62, of Constitutional Custodians, at 542 (“On
their own, agency leaders simply are not numerous enough or, in many cases,
experienced or sophisticated enough to conduct research or promulgate
rules.”).

by an even larger partisan staff inside Congress.171 While this

staff does not always have the expertise to review the work of
the congressional bureaucracy,172 it does typically have the time
and the manpower. These differences often eliminate the tool of
unilateral action.
Agency bureaucrats also can possess the tool of unilateral
action when Congress specifically empowers these bureaucrats
to act by statute, thereby removing the issue from presidential
discretion.173 By contrast, it is more complicated to remove the
congressional bureaucracy’s work from congressional discretion.
Insofar as these offices are woven into congressional procedure,
that procedure is constitutionally entrusted to Congress.174 This
further removes unilateral action as a tool.
Consequently, the tool of persuasion is particularly
important in the congressional context. In this sense, the
congressional bureaucracy resembles Margot Schlanger’s
“Offices of Goodness,” which she describes as offices internal to
a larger institution that operate primarily through the tool of
persuasion.175
For the congressional bureaucracy, this tool also serves a
secondary function not seen in the executive context—one
grounded in the different nature of the partisan principal. While
this bureaucracy typically assists all Members and staffs,176 it is
subordinate only to congressional leadership.177 That partisan
leadership differs from its executive-branch counterparts178 in
an important respect: congressional leaders typically have spent
decades working in their branch of government and interacting

172. See Cross, Legislative History, supra note 50, at 99–102 (describing
expertise differences between partisan and nonpartisan staffs); Cross & Gluck,
supra note 9, at 1612–13 (same).
173. See Ingber, supra note 7, at 289 (identifying congressional statute as
an authority permitting action “even against the President’s professed will”).
174. U.S. CONST. art. I, § 5.
175. See Schlanger, supra note 121, at 60.
177. See supra notes 147–150.
178. See Michaels, Of Constitutional Custodians, supra note 62, at 236
n.21 (noting and citing discussions of “the short tenure of most politically
appointed agency leaders”); see also Ackerman, New Separation of Powers,
supra note 3, at 706 (“[P]olitical appointees do not stay in office long enough
to operate productively.”).

262 80 WASH. & LEE L. REV. 229 (2023)
with its bureaucracy.179 This strengthens the opportunity for

persuasion by the congressional bureaucracy, which uses this
tool to perform and display its expertise for these partisans over
time, gradually illustrating for them the value their expertise
contributes to partisan work.180
In the 1995 Republican Revolution, this legitimation effect
of persuasion likely saved some of the congressional
bureaucracy from elimination.181 As Republicans retook control
of the House, there was significant desire among some newer
Republican staffers to eliminate much of the congressional
bureaucracy.182 Newt Gingrich and his staff, however, had come
to appreciate the bureaucracy’s value after seeing and
benefitting from its work over many years.183 In the end, only
one office was eliminated (the Office of Technology Assessment),
though others suffered cuts.184 Those that survived have
attributed their survival specifically to the cumulative effect of
many years of persuasive work with key high-ranking
partisans.185 This simply would not have been possible if
political leadership in Congress had no experience in the branch.
This role of persuasion offers lessons for executive-branch
research. In that research, it has been argued that the
transience of political appointees might be a source of
bureaucratic power, since it highlights the expertise accrued by
entrenched civil servants.186 The experience of the congressional
bureaucracy tells the other side of this story: the non-transience
of congressional political leadership (and of its staff)187 also is a
179. This is uniformly true at the Member level, and it also typically is
true at the staff level. See Cross, Legislative History, supra note 50, at 106–07.
180. See Schlanger, supra note 121, at 60.
181. See Cross & Gluck, supra note 9, at 1615–16 (“In fact, we were told
that this steadfast commitment to nonpartisanship is what saved the offices
of the House Legislative Counsel and Parliamentarian in 1995, when Speaker
Newt Gingrich and the new Republican majority revamped many other
congressional operations.”); see also Gould, supra note 55, at 2008.
182. See Cross & Gluck, supra note 9, at 1615–16; see also Interview with
Staffer (on file with author).
184. See id.
185. See id. at 1616.
186. See Ingber, supra note 7, at 190.
187. See Cross, Legislative History, supra note 50, at 106–07 (noting four
features of the legislative staff position indicating policy expertise).

source of bureaucratic power—at least, when it is approached as

an opportunity to build long-term reputations as indispensable
value-adds.
This analysis also offers lessons for legislative reformers. In
order for persuasion to accomplish the aforementioned
secondary goals for the congressional bureaucracy, offices
typically must be both small and advisory. If these offices are
too large, it becomes difficult for their many experts to establish
and maintain persuasive relationships with influential
partisans in Congress—though exceptions do exist, such as the
GAO workforce of 3,000.188 If they are not advisory, they lose the
ability to display their expertise in action, showing partisans
how their conclusions result from the application of analyses
and methodologies that add value along the way.189 These
factors may impose limits on the type and size of office that
reformers want to consider for addition or relocation into the
congressional bureaucracy.
C. Normative Questions: Is Bureaucratic Resistance

Desirable?
In the executive-branch literature, there also is ongoing

debate about whether bureaucratic resistance is desirable. For
some, it problematically undermines an energetic,190
accountable,191 unitary192 executive. According to others, it
188. For the size of each office in the congressional bureaucracy, see Cross
& Gluck, supra note 9, at 1599–60.
189. On the advisory role played by all offices in the congressional
bureaucracy, see infra Part V.
190. See, e.g., Kagan, supra note 4, at 2263; see also Ingber, supra note 7,
at 144–45.
191. See THEODORE J. LOWI, THE END OF LIBERALISM: THE SECOND
REPUBLIC OF THE UNITED STATES 311 (2d ed. 1979). On accusations of a “deep
state,” see MARC AMBINDER & D.B. GRADY, DEEP STATE: INSIDE THE
GOVERNMENT SECRECY INDUSTRY 4 (2013); MIKE LOFGREN, THE DEEP STATE:
THE FALL OF THE CONSTITUTION AND THE RISE OF A SHADOW GOVERNMENT 34–36
(2016); Peggy Noonan, The Deep State, WALL ST. J. (Oct. 28, 2013, 9:10 PM),
https://perma.cc/Y7VV-8KPM; Jack Goldsmith, Paradoxes of the Deep State, in
CAN IT HAPPEN HERE?: AUTHORITARIANISM IN AMERICA 105 (Cass R. Sunstein
ed., 2018).
192. See generally STEVEN G. CALABRESI & CHRISTOPHER S. YOO, THE
UNITARY EXECUTIVE: PRESIDENTIAL POWER FROM WASHINGTON TO BUSH (2008).

264 80 WASH. & LEE L. REV. 229 (2023)
creates an internal separation of powers193 inside the executive

branch that serves both traditional separation-of-powers
193. See Metzger, supra note 116, at 425 (discussing the rise of this
internal separation literature); Katyal, supra note 5, at 2318; Magill &
Vermeule, supra note 81, at 1059 (complicating unitary executive theory by
noting that agencies “are not unitary actors” and are “fractured internally”);
Michaels, Evolving Separation of Powers, supra note 60, at 551–60; Michaels,
Of Constitutional Custodians, supra note 62, at 229 (on the “tripartite
administrative separation of powers” in executive branch); Ingber, supra note
7; Morrison, supra note 7, at 1524; Green & Roiphe, supra note 51, at 5–6
(2018) (“[The article] adds to the voices of a growing number of scholars who
see the Executive not as unitary, but as a complex whole, whose parts serve as
checks on one another.”); Trevor W. Morrison, Constitutional Alarmism, 124
HARV. L. REV. 1688, 1692 (2011); see also Aziz Z. Huq & Jon D. Michaels, The
Cycles of Separation-of-Powers Jurisprudence, 126 YALE L.J. 346, 391 (2016)
(noting that the “complex ecosystem of intrabranch and entirely external
actors [is] not traditionally accounted for in the separation-of-powers
literature”); Metzger, supra note 116, at 428 (noting that “the focus of internal
separation of powers scholarship is overwhelmingly on the Executive
Branch”); Ackerman, Good-bye, supra note 60, at 39 (“Almost three centuries
later, it is past time to rethink Montesquieu’s holy trinity. Despite its canonical
status, it is blinding us to the world-wide rise of new institutional forms that
cannot be neatly categorized as legislative, judicial, or executive.”); Ackerman,
New Separation of Powers, supra note 3, at 689 (“[T]his second separationist
doctrine should begin by cordoning off vast areas of concrete decisionmaking
from those few questions that imperatively require the attention of democratic
statesmen.”).

values194 and additional values.195 For yet others, resistance is

mixed—existing on a spectrum that ranges from reasonable to
unreasonable, for example.196
194. See Huq & Michaels, supra note 193, at 420 (“As the internal political
surround of intra-agency lawyers, civil servants, and ALJs has become denser,
[and] as civil-service protections have taken root . . . the case for [Courts
creating rigid separation-of-powers rules] requiring strict, uniform, and
conforming practices within the executive has become weaker.”); Metzger,
supra note 116, at 426 (noting “the important separation of powers function
that internal constraints can serve”); Ackerman, Good-bye, supra note 60, at
39–40 (arguing that executive-branch bureaucracy advances the values of
functional specialization and of shielding law implementation from politics,
values sought by Founders through separation of courts into a separate
branch); Katyal, supra note 5, at 2319–20 (arguing that rise of bureaucracy in
executive branch creates new opportunity to realize separation-of-powers
values such as deliberativeness via intra-branch separation in an era when
interbranch checks are failing due to congressional acquiescence); M.
Elizabeth Magill, Beyond Powers and Branches in Separation of Powers Law,
150 U. PA. L. REV. 603, 605 (2001) (noting that internal separation
accomplishes the goal of fragmenting power); Anne Joseph O’Connell, The
Architecture of Smart Intelligence: Structuring and Overseeing Agencies in the
Post-9/11 World, 94 CALIF. L. REV. 1655, 1689 (2006) (“[T]he most effective
[national intelligence] structure probably would have redundant components
as well as components that coordinate and centralize certain efforts.”); LANDIS,
supra note 22, at 46; Green & Roiphe, supra note 51, at 5–6 (noting that the
executive branch’s “parts serve as checks on one another”); Michaels, supra
note 62, Of Constitutional Custodians, at 229; Daryl J. Levinson & Richard H.
Pildes, Separation of Parties, Not Powers, 119 HARV. L. REV. 2311, 2375 (2006)
(“[C]onstitutional engineering might focus on insulating the administrative
bureaucracy more fully from the partisan pressures of unified government.
The idea would be to take seriously the metaphor of the bureaucracy as a
‘fourth branch’ of government . . . .”). But see Bijal Shah, Response: Toward an
Intra-Agency Separation of Powers, 92 N.Y.U. L. REV. 101 (2017) (arguing
against the comparison to external separation of powers).
195. See, e.g., Metzger, supra note 116, at 440 (arguing that it can
depoliticize governmental administration, and noting its unique benefits as an
ex ante, continuous, and non-adversarial method of providing competing
views); Ingber, supra note 7, at 150 (noting it can provide continuity); see also
Nou, supra note 137 (noting that “the costs [of resistance] may help to ensure
that what resistance remains is more often evidence of a canary in a coal mine
than a bureaucracy run amok”); Matthew C. Stephenson, The Qualities of
Public Servants Determine the Quality of Public Service, 2019 MICH. ST. L. REV.
1177, 1180 (2019) (“I argue that we should want to attract and empower
bureaucrats who are not only technically competent, but who can function as
an effective counterweight to their agency’s politically-appointed leadership
and its overseers in the White House and Congress.”); Michaels, Of
Constitutional Custodians, supra note 62, at 230 (arguing that administrative
separation of powers makes agencies “responsive to the fuller range of
democratic, technocratic, and rule-of-law values we expect to inform State

266 80 WASH. & LEE L. REV. 229 (2023)
To date, these normative assessments of bureaucratic

resistance mostly have been grounded in the particular facts of
the executive branch.197 Extended to the congressional
bureaucracy, they reveal a different calculus—one that
generally is more approving of resistance in the congressional
context.
This revised calculus results from several factors. On the
one hand, many of the benefits touted by celebrators of internal
separation have indeed carried over into the congressional
context, as the Cross-Gluck study documented.198 The
deliberation-enhancing benefits and the salutary dispersal of
power do appear to have resulted from the creation of a
nonpartisan bureaucracy, for example.199 Admittedly, the latter
value may be less prized in the congressional context, which
lacks the intense concerns about consolidation of power that
worry some presidential scholars.200 Nonetheless, the virtues
power as exercised through administrative or any other set of actors”);

Schlanger, supra note 121, at 65 (arguing that “Offices of Goodness” in
agencies enhance considerations of civil rights and civil liberties); Brian D.
Feinstein & Abby K. Wood, Divided Agencies, 95 S. CAL. L. REV. 731, 737 (2022)
(arguing that civil servants may pull agencies toward views of the median
voter).
196. See generally Ingber, supra note 7.
197. For articles noting Congress’s internal separation of powers, see
Elizabeth Garrett & Adrian Vermeule, Institutional Design of a Thayerian
Congress, 50 DUKE L.J. 1277, 1314–16 (2001) (noting that internal separation
“leads to more balanced presentation of information” and allows Congress to
collectively fund and retain more skilled people, keep additional power from
already-powerful committee chairs, and “gain credibility with constituents”);
Mark Tushnet, The Ambiguous Legacy of Watergate for Separation of Powers
Theory: Why Separation of Powers Law Is Not “Richard Nixon” Law, 18 NOVA
L. REV. 1765, 1771–72 (1994) (arguing that the rise of bureaucracies, including
in Congress, has fractured the unitary branch interests that
separation-of-powers assumes); Huq & Michaels, supra note 193 (arguing that
internal separation across the branches creates a contestation over the values
that separation-of-powers itself should advance, and that the Court should
find ways to promote that ongoing contestation); Metzger, supra note 116, at
457 n.21 (noting internal separation in Congress while focusing on it in the
executive branch context); Magill, supra note 194, at 605–06 (arguing that,
due to internal separation, there is no single “interest” of each branch).
199. See id.
200. See, e.g., Katyal, supra note 5, at 2316 (framing the concern as one of
“an executive that subsumes much of the tripartite structure of government”);
Renan, supra note 113, at 2231 (“Authority-allocating norms in the main,

hypothesized in the executive branch context have rung true in

the congressional experience.
On the other hand, several observed downsides of
bureaucratic resistance either do not apply to the congressional
context or do not apply as strongly. First, some have argued that
bureaucratic resistance undermines an executive branch
envisioned as unitary by the Constitution.201 The merits of this
much-debated claim aside, this concern does not extend to the
congressional context. As legislation scholars are fond of noting,
Congress is not meant to be a unitary institution.202
Bicameralism explicitly fractures legislative power,203 and the
tradition of congressional committees furthers this internal
separation.204 As such, there is no equivalent downside to
potentially fracturing an otherwise unitary branch by
introducing bureaucratic resistance in Congress. And while the
legislative branch may have its own unique constitutional
imperatives and concerns, such as nondelegation concerns,205
the congressional bureaucracy has not to date triggered these
parallel concerns among legislation scholars.
however, have accreted power to the presidency.”); Green & Roiphe, supra note
51, at 49 (“The effort to install professional lawyers and rationalize an
increasingly splintered system provided a way to justify, or at least address,
concerns about the increase in power at the executive level.”); Metzger, supra
note 116, at 457.
201. See generally CALABRESI & YOO, supra note 192.
202. See Kenneth A. Shepsle, Congress Is a “They,” Not an “It”: Legislative
Intent as Oxymoron, 12 INT’L REV. L. & ECON. 239, 241–44 (1992) (coining the
oft-repeated phrase that “Congress is a ‘they,’ not an ‘it’”).
203. See I.N.S. v. Chadha, 462 U.S. 919, 951 (1983) (“The division of the
Congress into two distinctive bodies assures that the legislative power would
be exercised only after opportunity for full study and debate in separate
settings.”).
204. See JOSH CHAFETZ, CONGRESS’S CONSTITUTION: LEGISLATIVE
AUTHORITY AND THE SEPARATION OF POWERS 285 (2017) (noting that rise of
standing committees “naturally tended toward a certain diffusion of power,”
especially when party leaders did not control appointments); Magill, supra
note 194, at 652 (“State power is diffused . . . [w]ithin Congress: a house
committee chair; a ranking member of a Senate committee; and the deputy
whip in the Senate or the majority leader in the House.”); Metzger, supra note
116, at 457 n.21 (noting committees and parties achieve dispersion of power).
205. But see generally Nicholas Bagley & Julian Davis Mortenson,
Delegation at the Founding, 121 COLUM. L. REV. 277 (2021) (arguing that the
nondelegation doctrine did not actually exist to constrain Congress at the
founding); Parrillo, supra note 86 (same).

268 80 WASH. & LEE L. REV. 229 (2023)
Second, Justice Kagan in particular has worried that

bureaucratic resistance problematically undermines the energy
of the executive branch.206 And, despite its justified reputation
as an often-gridlocked institution, Congress often does act with
speed and energy when vetogate-proof majorities coalesce
around a legislative policy.207 Due to the congressional
bureaucracy’s largely advisory and non-binding character,
however, it lacks the power to undermine that energy.208 If
partisans in Congress decide they must act expeditiously and
energetically on legislation, and if the bureaucracy is not
providing input in line with the desired policy or its timelines,
Congress simply will act without its input. That, in turn, puts
pressure on the bureaucracy to align its energy with that of
partisans. In this way, the non-binding character of the
congressional bureaucracy’s work, and the fact that its work is
a prelude to formal partisan action rather than a consequence
of it, largely precludes it from creating the drag on branch
energy that worries Kagan.
In a similar vein, Justice Kagan also has argued that
bureaucratic resistance undermines the development and
implementation of effective policy in the executive branch.209 As
a result, Kagan argues that greater partisan control of, and
intervention into, the bureaucracy increases effectiveness.210 In
making this argument, Justice Kagan provocatively challenges
the consensus view that, while accompanied by downsides,
bureaucratic resistance at least increases effectiveness through
its infusion of expertise.211 Setting aside the validity of this
argument in the presidential context, it does not appear to
translate to the congressional bureaucracy. As the Cross-Gluck
study showed, partisan interventions into the congressional
206. Kagan, supra note 4, at 2263 (noting that “bureaucracy also has
inherent vices (even pathologies), foremost among which are inertia and
torpor”).
207. On the idea of vetogates, see ESKRIDGE ET AL., supra note 31, at 80.
208. See Cross & Gluck, supra note 9, at 1628–29 (observing that “one
difference across the board from the executive bureaucracy is that most of the
work of the congressional bureaucracy does not have formal legal effect
without some additional action from Congress”).
209. Kagan, supra note 4, at 2339.
210. Id.

bureaucracy typically have been attempts to shallowly politicize

it, not to render it more effective.212
Finally, when assessing the normative value of
bureaucratic resistance, scholars have debated the relationship
between internal and external separation of powers.213 Much
scholarship on internal separation of powers is premised on the
notion that bureaucratic resistance weakens or checks a
branch’s aggregate power relative to other branches. This is one
reason why internal separation is viewed as a replacement for
external checks.214 However, the opposite argument also has
been advanced: namely, that even as bureaucracies separate
power within a branch, they also bolster the overall power of
that branch.215 Which is correct?
In the congressional context, there is perhaps less
ambiguity on the relationship between internal and external
separation. The congressional bureaucracy was consciously
created to reclaim powers from the executive branch, so the
external separation-of-powers dynamic was clearer.216 While
outside studies may always reveal this congressional idea to be
misplaced, it at least is relevant that those in Congress widely
believed, and still believe, their interbranch power to be
enhanced by—or at least not eroded by—these internal
separations.
Still, this debate remains relevant. Here, interestingly, the
takeaway for Congress will be the opposite from that in the
executive branch. Since scholars’ overriding concern is that too
much power has accreted to the executive branch at Congress’s
expense, if bureaucratic resistance swells a branch’s power, that
is concerning for bureaucracy in the executive branch—but it is
welcome in Congress. And the opposite may also be true. The
212. See id. at 1615–16, 1626–28 (reviewing major instances of

intervention).
213. For a study focused specifically on this relationship, see generally
Metzger, supra note 116.
214. See, e.g., Katyal, supra note 5, at 2316 (“The first-best concept of
‘legislature v. executive’ checks and balances must be updated to contemplate
second-best ‘executive v. executive’ divisions.”).
215. Ingber, supra note 7, at 144–45 (citing Rebecca Ingber, The Obama
War Powers Legacy and the Internal Forces that Entrench Executive Power,
110 AM. J. INT’L L. 680, 687 (2016)); see also MICHAEL J. GLENNON, NATIONAL
SECURITY AND DOUBLE GOVERNMENT 91–99 (2015).

270 80 WASH. & LEE L. REV. 229 (2023)
findings of this debate therefore may not generate conclusions

that bureaucracy should be expanded or contracted, but rather
that it should be shifted across branches—an approach that
might preserve the expertise-generating function of
bureaucracy while redoubling the desired effects for external
separation of powers.
IV. ADVISING AND ADJUDICATING: ON DUAL ROLES
In executive-branch scholarship, concern also has been

raised over bureaucratic offices with dual roles. Neal Katyal
particularly has raised concerns about the dual role performed
by the Office of Legal Counsel (OLC).217 According to Katyal, the
combination of adjudicatory and advisory functions in OLC
undermines the office’s neutrality.218 He therefore advocates for
making OLC purely advisory, while creating a new adjudicatory
office (also housed in the executive branch).219 Similar
arguments have been made about the Office of Information and
Regulatory Affairs (OIRA).220 These arguments are grounded in
the premise that a hybrid advisory/adjudicatory office likely
cannot preserve a reputation of neutrality.
The congressional bureaucracy challenges this premise.
Many of its offices successfully navigate this dual
217. See Katyal, supra note 5, at 2337 (“The client-driven advisory

function, however, infects its adjudicatory role. Just as the trend in the
government has been to split the litigation function from the advisory function
(so that there is no longer a Solicitor General who both litigates and advises),
a new split between the advisory and adjudicatory functions of OLC is
necessary.”).
218. See id. (arguing that “in this climate, there is simply no way that
OLC’s aspiration to be a neutral decision-maker can play out in practice”).
219. See id. (“Instead of a compromised OLC, OLC should be stripped of
its adjudicatory role and permitted to function only as an adviser to the
administration. . . . The adjudication function would be transferred to a
separate official, a Director of Adjudication, who would resolve inter-agency
disputes.”). Cf. Renan, supra note 7, at 813–14 (arguing that the “shift to a
more informal, diffuse, and porous brand of legalism creates opportunities for
blended judgment” that “might ultimately present a more modest, but perhaps
also more honest vision of what law can and should achieve outside the
courts”).
220. See Stuart Shapiro, OIRA’s Dual Role and the Future of Cost-Benefit
Analysis, 50 ENV’T L. REP. 10385 (2020); Stuart Shapiro, Unequal Partners:
Cost-Benefit Analysis and Executive Review of Regulations, 35 ENVTL. L. REP.
10433 (2005).

advice-and-adjudication role. All four of its offices that

adjudicate disputes additionally function as advisors, guiding
partisans on how best to reach their desired results in any
ultimate adjudication.221 Despite this dual role, these offices
have generally retained strong reputations as neutral
adjudicators inside Congress.222 For the Parliamentarians’
offices, Jonathan Gould has posited that the dual role even has
had the opposite effect: it has buttressed their reputation for
neutrality.223 This suggests that a hybrid
advice-and-adjudication role actually is possible, and that other
elements of OLC and OIRA are potentially contributing to the
erosion of their adjudicatory neutrality.
For example, the challenge faced by OLC in particular may
be more fully explained by Rebecca Ingber’s distinction between
“entrenched” and “transient” bureaucracy.224 As Ingber
observes, OLC is a transient bureaucracy: its leader and staff
typically change with a new administration.225 By contrast, the
nonpartisan congressional offices with adjudicatory roles are
mostly entrenched bureaucracy. Their staff—and usually their
leadership—serve across multiple transitions of partisan
221. The four offices are: the House Parliamentarian, the Senate
Parliamentarian, CBO, and JCT. See generally Cross & Gluck, supra note 9
(documenting that all offices in congressional bureaucracy advise).
222. See Kevin R. Kosar, Legislative Branch Support Agencies: What They
Are, What They Do, and Their Uneasy Position in Our System of Government,
in CONGRESS OVERWHELMED: CONGRESSIONAL CAPACITY AND PROSPECTS FOR
REFORM 128 (Timothy M. LaPira et al. eds., 2020) (noting high rankings among
congressional staff of congressional bureaucracy offices on neutrality); Gould,
supra note 55, at 2008; Cross & Gluck, supra note 9, at 1613–16.
223. See Gould, supra note 55, at 2008 (“This example shows how the
advisory role can support the parliamentarians’ reputations as neutral brokers
who treat both parties fairly.”).
224. Ingber, supra note 7, at 161 (“This is the dynamic between what I will
call the entrenched and the transient bureaucracy, that is, those actors within
the bureaucracy who tend to remain in the government through
ideologically-opposed presidential administrations, and those who swap out
when their President or party is out of power.”).
225. Id. at 168–69
OLC itself is an office more likely to be on board with the President’s general
agenda than opposed . . . the office is typically led by a carefully chosen
political appointee and staffed with career lawyers who tend to change over
at a much higher frequency than those in agency general counsel offices,
and are thus more likely to have chosen to work in the office under the
contemporaneous President.

272 80 WASH. & LEE L. REV. 229 (2023)
power.226 This entrenched quality seems to have helped to

cement the reputation of these offices as neutral arbiters, with
each party recalling the fair dealing they received when in the
majority.227 Admittedly, this explanation may be partly
qualified by the experience of OIRA, which largely consists of
career staff, though important leadership positions are political
appointees.228 Still, when the congressional bureaucracy is
added into the analysis, an emphasis on the unique challenges
of transient bureaucracy may provide a more compelling
explanation than an emphasis simply on the inherent problems
of dual roles.229
Viewed in this light, it is not surprising that the most
noteworthy instance of politicization in the congressional
bureaucracy, the partisan removal of the Senate
Parliamentarian in the 1980s,230 arose when Republicans
assumed control of the chamber for the first time in twenty-six
years.231 In that instance, Republicans had come to see the
Parliamentarian as biased toward the opposing party.232 This
occurred partly because, after being in the minority so long,
Republicans had only witnessed the office’s staff assisting the
Democrats in the accomplishment of their legislative goals.233
The benefits of being an entrenched bureaucracy may also
be buttressed by an element of the congressional bureaucracy
226. There are some exceptions. The head of CBO serves for only a
four-year term, though the staff typically does not also turn over. 2 U.S.C.
§ 601(a)(3)–(4). As previously noted, the Senate Parliamentarian would turn
over with party change in the Senate during the 1980s and 1990s, though this
practically led simply to a switching of roles between two different figures.
Gould, supra note 55, at 2006.
227. See Cross & Gluck, supra note 9, at 1615–16 (noting the existence of
an ostensibly neutral “Congressional bureaucracy”); Gould, supra note 55, at
2008.
228. See Cass R. Sunstein, The Office of Information and Regulatory
Affairs: Myths and Realities, 126 HARV. L. REV. 1838, 1845 (2013).
229. See Ackerman, New Separation of Powers, supra note 3, at 698
(arguing that “the Weberian ethic of bureaucratic neutrality” can function “as
a vital insurance policy” for entrenched bureaucracies, allowing them to
plausibly claim to partisans that they will serve new governing majorities with
the same energy).
231. See Party Control, U.S. SENATE, https://perma.cc/QE32-UH2Y.
233. Id.

that presumably is not replicable in OLC or OIRA. Due to the

multi-principal structure of a legislature, the minority party is
present in Congress to receive assistance and advice from the
congressional bureaucracy.234 This means that, even when out
of power, partisans receive assistance from the bureaucracy and
witness its deployment of expertise.235 This can help support
impressions of its neutral character, including when these
partisans come into the majority.236 As previously noted, this
element helped save several offices of the congressional
bureaucracy in the Gingrich era, with Newt Gingrich’s inner
circle recalling the value added by the bureaucracy.237 Moreover,
assisting the minority party may help prevent bias and partisan
capture in these bureaucratic offices, as it consistently exposes
the bureaucracy to competing arguments and agendas. It may
therefore be that office neutrality can survive a hybrid
adjudicating-and-advising role, but that it requires specific
conditions that either have not, or cannot, be replicated in OLC
or OIRA.
V. AGENCY CAPTURE
Scholars also worry that the executive branch bureaucracy

is vulnerable to the problem of “agency capture” by interest
groups.238 In this view, administrative agencies are meant to
234. See Cross & Gluck, supra note 9, at 1621 (“Unlike the executive
branch bureaucracy, the congressional bureaucracy performs its tasks for both
the majority party and the party not in control.”).
235. See id.
236. See Kosar, supra note 222.
237. See Cross & Gluck, supra note 9, at 1615–16; Gould, supra note 55, at
2008.
238. See, e.g., Kagan, supra note 4, at 2264–65
One purpose, even if unfulfilled, of efforts to place institutional controls on
agency action relates to the prospect that, in the absence of these
safeguards, regulated entities and other organized interests themselves will
grasp the reins of regulatory authority. The view that firms subject to
regulation had ‘captured’ the agencies gained wide currency beginning in
the 1960s. Although the thesis often was stated too crudely, few could argue
with its basic insight—that well-organized groups had the potential to
exercise disproportionate influence over agency policymaking by virtue of
the resources they commanded, the information they possessed, and the
long-term relations they maintained with agency officials.
see also Datla & Revesz, supra note 60, at 816

274 80 WASH. & LEE L. REV. 229 (2023)
provide relatively autonomous zones for bureaucratic, expert

decision-making in pursuit of democratically-determined goals,
which may conflict with the goals of the private industries and
interest groups that are to be regulated.239 Structural features
of the executive branch bureaucracy, however, provide
opportunities for these interest groups to insert themselves into
the agency decision-making process and warp it to their ends.240
In this way, the design of the executive branch bureaucracy is
seen as containing vulnerabilities that allow concentrated,
well-funded interests to steer (or “capture”) agency
decision-making processes.241 This phenomenon of agency
capture has cast doubt on early aspirations by figures such as
James Landis, who believed that bureaucratic expertise might
create the possibility of pushing back against concentrated
corporate power.242
The congressional bureaucracy provides a useful,
unexplored comparison in the effort to understand agency
capture. Here, a fundamentally different relationship has
emerged between bureaucracy and special interests. Not only
has the congressional bureaucracy avoided capture by interest
groups—it functions as a counterweight against legislator
capture by these special interests.243
Some offices in the congressional bureaucracy do
occasionally make themselves available to interest groups, it
The phenomenon of interest group influence is commonly referred to as

agency capture. The capture thesis recognizes that because interest groups
representing regulated entities tend to be overrepresented in the agency
decision-making process compared to interest groups representing public
interests, the outputs of agencies will tend to be biased in favor of those
interests.
239. See Wendy E. Wagner, A Place for Agency Expertise: Reconciling
Agency Expertise with Presidential Power, 115 COLUM. L. REV. 2019, 2023–32
(2015).
240. See, e.g., Datla & Revesz, supra note 60, at 816. The most notable of
these is the requirement for notice-and-comment rulemaking. See 5 U.S.C.
§ 553(c).
241. See Datla & Revesz, supra note 60, at 816.
242. See LANDIS, supra note 22; see also Adrian Vermeule, Bureaucracy
and Distrust: Landis, Jaffe, and Kagan on the Administrative State, 130 HARV.
L. REV. 2463, 2470 (2017) (“The role of expertise in [Landis’s work] is twofold.
Trivially, causal, scientific, and policy knowledge are necessary for ongoing
administrative supervision of concentrated corporate power.”).

should be noted. JCT staff seeks out conversations with trade

associations, companies, and other taxpayers to develop an
understanding of the real-world impacts of tax policies, and they
also will respond to requests for conversations with these
actors.244 OLRC, in a process that mimics notice-and-comment
periods in agencies, will reach out to private actors when
assembling codification bills and invite feedback on whether
proposed language captures the consensus understanding of
existing legislation.245 And while the staffs at MedPAC and
MACPAC are permanent nonpartisan staffs, their
commissioners were meant to bring perspectives to Congress
that it lacked—a democratic corrective of the sort also
envisioned in agency practices.246 At least some corners of the
congressional bureaucracy therefore do interact with special
interest groups and account for their viewpoints.
Most of the congressional bureaucracy does not interact
with outside interest groups, 247 however, and the bureaucracy
typically functions as a key actor inside Congress that checks
interest group involvement in the legislative process.248 As the
head of the CRS employees’ union testified to Congress in 2019:
Members of this committee and your staff receive many
sources of facts and opinion. They are often useful. But
whether they come from the Alzheimer’s Association or Big
Pharma, they support the goals of the sources. Seasoned CRS
analysts can help sort through the facts and opinions and
provide an unbiased overview.249
Similarly, Legislative Counsel often is the key point of
resistance to straightforward insertion of legislative language

245. Positive Law Codification, OFF. OF THE L. REVISION COUNS.,
https://perma.cc/L2NC-94HQ.
247. Id. For example, Legislative Counsel does not respond to requests or
contacts from non-congressional actors, and even when partisan congressional
staff invite interest groups to attend legislative drafting sessions, Legislative
Counsel typically obliges only if the partisan staff also is present. Id.
249. Written Statement of Susan Thaul, President, Cong. Rsch. Emps.
Ass’n, Statement to Committee on House Administration 4–5 (June 20, 2019),
https://perma.cc/38G5-84JA (PDF).

276 80 WASH. & LEE L. REV. 229 (2023)
drafted by lobbyists.250 MedPAC and MACPAC were founded

specifically in the effort to resist interest groups,251 as was the
now-defunct independent technology agency, the OTA.252
Throughout the congressional bureaucracy, there is an
understanding that the bureaucracy is insulated from interest
group contact and is expected to provide an expert, nonpartisan
counterbalance to the partial information entering the
legislative process via interest groups.253
When we look at bureaucracy across Congress and the
executive branch, we therefore see two very different modes of
interacting with interest groups. This can offer lessons in both
directions.
For example, scholars have pointed to factors that may
contribute to the problem of agency capture.254 One key
contributing factor, they have suggested, has been the

251. See John Reichard, Make Way for MacPAC, the New Kid on
Washington’s Health Policy Block, COMMONWEALTH FUND (Aug. 2, 2010),
https://perma.cc/Q3D9-DAVJ (reporting Senator Rockefeller’s description of
MedPAC and MACPAC as a conscious effort to take decision-making “out of
the hands of Congress and the lobbyists”); see also Thomas R. Oliver, Analysis,
Advice, and Congressional Leadership: The Physician Payment Review
Commission and the Politics of Medicare, 18 J. HEALTH POL., POL’Y & L. 113,
149 (1993) (noting that “the commission’s own research and analysis made it
possible to test the empirical claims of the interest groups with greater rigor”
and that the “expertise of the PPRC commissioners and staff diminished the
informational power of lobbyists”); HOLLY STOCKDALE, CONG. RESEARCH SERV.,
R40915, AN OVERVIEW OF PROPOSALS TO ESTABLISH AN INDEPENDENT
COMMISSION OR BOARD IN MEDICARE 5 (2020) (noting that the two advisory
commissions that would be merged to form MedPAC operated to “buffer
members of Congress from pressures from interest groups”).
252. See CHRIS MOONEY, THE REPUBLICAN WAR ON SCIENCE (2005) (“In
OTA’s absence, however, the new Republican majority in Congress freely
called upon its own favorable scientific ‘experts’ and relied upon analyses
prepared by lobbyists and ideologically committed think tanks. . . .”);
Katherine Tully-McManus, House Members Call for Office of Technology
Assessment Revival, ROLL CALL (Apr. 2, 2019), https://perma.cc/WC67-ZSHV
(noting House members complaining about Congress’s use of
“non-governmental groups that are often advocating a position on
technological issues, rather than an unbiased perspective”).
253. See Written Statement of Susan Thaul, President, Cong. Rsch. Emps.
254. See, e.g., Kagan, supra note 4, at 2265 (citing “the resources they
commanded, the information they possessed, and the long-term relations they
maintained with agency officials”).

legitimation problem faced by agency rule-makers.255 According

to this account, career bureaucrats are troublingly unmoored
from the democratic accountability that legitimates most
decisions in our system. In response to this legitimation anxiety,
agency processes were created to mimic democratic
participation by interested parties. Agencies therefore were
required to invite outside groups to participate in their
rulemaking processes,256 which would function as a “perfected
legislative process for the formulation of policy”257 and as “a
surrogate political process within the bureaucratic sphere,”258 as
scholars have put it. This solution to agency legitimation
problems invited powerful interest groups into the agency
255. Ackerman, New Separation of Powers, supra note 3, at 697 (noting

that “regulatory decision-making needs special forms of legitimation that
enhance popular participation”); Katyal, supra note 5, at 2317 (“In short, the
executive is the home of two different sorts of legitimacy: political (democratic
will) and bureaucratic (expertise).”); id. at 2346 (“The President will play a
more powerful role than his European counterparts by being able to trump
bureaucratic decision-making. That approach suffuses the bureaucracy’s
expertise-laden legitimacy with political legitimacy.”); Metzger, supra note
116, at 454 (“[A]t the same time as they serve the constitutional goal of
checking excessive Executive Branch power, such constraints arguably
undermine political accountability . . . .”); Kagan, supra note 4, at 2252 (“I aver
that in comparison with other forms of control, the new presidentialization of
administration renders the bureaucratic sphere more transparent and
responsive to the public . . . .”); Vermeule, supra note 242, at 2463 (noting “the
administrative state’s legitimation problem” and discussing notable attempts
by Landis, Jaffe, and Kagan to solve it); Lisa Schultz Bressman, Beyond
Accountability: Arbitrariness and Legitimacy in the Administrative State, 78
N.Y.U. L. REV. 461, 465 (2003) (“It is no accident that the model emerging
contemporaneously with the countermajoritarian difficulty was the ‘interest
group representation’ model, which consciously characterized the
administrative process as a perfected legislative process for the formulation of
policy.”); id. at 475 (“Through an interest group representation model,
agencies’ decisions would gather legitimacy ‘based on the same principle as
legislation.’”); see also Kagan, supra note 4, at 2261 (explaining how the
narrative used to defend “the legitimacy of bureaucratic power” has evolved
over the past century). But see generally Jerry L. Mashaw, Prodelegation: Why
Administrators Should Make Political Decisions, 1 J.L. ECON. & ORG. 81
(1985).
256. As Justice Kagan puts it: “The interest group model . . . attempted to
ensure accountability by creating a surrogate political process within the
bureaucratic sphere, even if at some cost to administrative efficiency.” Kagan,
supra note 4, at 2331.
257. Bressman, supra note 255, at 465.
258. Kagan, supra note 4, at 2331.

278 80 WASH. & LEE L. REV. 229 (2023)
decision-making process, however, and thereby created the risk

of agency capture.
The contrasting experience of the congressional
bureaucracy supports this diagnosis of agency capture. Unlike
administrative agencies, the congressional bureaucracy
provides its input during a larger democratically-engaged
legislative process (that is, pre-enactment),259 and it provides
that input to a group of legislators that arguably are more
available than executive branch elected officials for direct
lobbying by interest groups and constituents.260 As a result, the
congressional bureaucracy would be expected to avoid
legitimation concerns that might press it to relinquish its
confidential relationships with legislators, and thereby expose
itself to possible agency capture.261 This is precisely what has
occurred.
Several lessons grow from this. First, as explained above, it
provides a better understanding of executive-branch agency
capture, providing support for the thesis that it is partly caused
by agencies’ unique legitimation challenges.262 Second, it
highlights another potential benefit of shifting some
bureaucratic work from the executive to the legislative
bureaucracy: that work does not suffer the same legitimation
concerns or risks of capture at the bureaucratic level.263 Of
course, it is possible that legislator capture is an even greater
concern.264 Insofar as the concern is specifically about insulation
of nonpartisan expertise from interest group pressure, however,
the congressional bureaucracy offers a promising alternative to
executive branch options.
Third, it provides some explanation for why the
congressional bureaucracy has largely avoided agency capture,
and thereby also provides lessons about how to ensure that it
continues to avoid it. Under this thesis, the congressional
bureaucracy’s connections to the legislative process—and that

260. On interest group involvement in legislative process, see generally
ESKRIDGE ET AL., supra note 31, at 77–79.
261. But see Cross & Gluck, supra note 9, at 1630.
262. See supra notes 255–258 and accompanying text.
263. See Ackerman, New Separation of Powers, supra note 3, at 697.
264. See generally, e.g., LAWRENCE LESSIG, REPUBLIC, LOST: HOW MONEY
CORRUPTS CONGRESS—AND A PLAN TO STOP IT (2011).

process’s larger democratic openness—have helped protect the

congressional bureaucracy.265 This suggests a potential limit on
the extent to which the congressional bureaucracy could be
expanded into post-enactment roles while still preserving its
beneficial traits. Interestingly, two of the offices in the
congressional bureaucracy that do post-enactment work—
JCT266 and OLRC267—also interact with private interests.
Fourth, it validates and extends James Landis’s notion that
bureaucratic expertise might create the possibility of pushing
back against concentrated corporate power.268 It suggests that
the setbacks American governance has experienced in this
project may be specific to executive branch institutional design,
rather than a bureaucratic inevitability. Again, this thesis
would need to be tested in a larger project to weigh the
overlapping problems of capture among political principals and
career bureaucrats—but, at a minimum, it shows that the
challenges administrative agencies have faced on this front do
not simply replicate across different bureaucratic structures in
our federal government.
VI. CONGRESSIONAL BUREAUCRACY AND THE JUDICIAL

BRANCH
The rise of the modern administrative state also has led

scholars to compare and contrast the bureaucracy and the
judiciary. This work has offered lessons about comparative
institutional competence, and about the proper relationship and
division of labor between the branches.269 This Part examines
two dimensions on which bureaucracy and the judiciary are
regularly compared: neutrality and expertise. As this Part
shows, introducing the congressional bureaucracy into the
265. See Written Statement of Susan Thaul, President, Cong. Rsch. Emps.
266. Most notably, JCT publishes the Blue Book, their general explanation
of the tax legislation enacted in the prior year. See Publication Search:
Bluebooks, JOINT COMM. ON TAXATION, https://perma.cc/MK29-JHTT.
269. See HENRY M. HART, JR. & ALBERT M. SACKS, THE LEGAL PROCESS:
BASIC PROBLEMS IN THE MAKING AND APPLICATION OF LAW 168–74, 1009–10
(William N. Eskridge, Jr. & Philip P. Frickey eds., 1994).

280 80 WASH. & LEE L. REV. 229 (2023)
conversation changes how we think about these

dimensions—and about the relationship of the political
branches to the judiciary.
A. Neutrality
It has been observed that, in both the bureaucracy and the

judiciary, there is a shared goal of institutional neutrality.270
Jon Michaels describes the civil service as the “closest
administrative ally” to the judiciary, for example, since both
“judges and civil servants tend to emphasize impartial, reasoned
decision making.”271 Michaels was not the first272 or last273 to
note this similarity, and it was self-consciously embraced at
times in the creation of the executive-branch bureaucracy.274
This observation has led to comparative assessments of the
branches based on their progress toward the goal of neutrality.
Two decades ago, for example, Bruce Ackerman diagnosed the
American political system as largely successful in achieving this
desired neutrality for the judiciary, but as having failed with
respect to bureaucracy.275 In making this diagnosis, Ackerman
was focused specifically on the executive branch bureaucracy,
which he saw as politicized by the dual allegiances discussed in
Part II.276
By contrast, the congressional bureaucracy appears more
successful on this dimension. In a recent survey of partisan
congressional staff, Kevin Kosar found offices in the
congressional bureaucracy repeatedly receiving the highest
270. This Article uses the term “neutrality” as a shorthand for the idea of
drawing conclusions via application of accepted professional methodologies
that are orthogonal to normative political considerations. For different
conceptions of neutrality, see Cross & Gluck, supra note 9, at 1621–24.
271. Michaels, Of Constitutional Custodians, supra note 62, at 250.
272. See, e.g., Ackerman, New Separation of Powers, supra note 3, at
715–16 (“The challenge [is] to define the conditions under which the claims of
functional specialization by judges and bureaucrats deserved constitutional
protection against . . . predictable efforts to erode the rule of law.”); see also
SKOWRONEK, supra note 19, at 47.
273. See, e.g., Ingber, supra note 7, at 188.
274. Shugerman, supra note 111, at 125.
275. See Ackerman, New Separation of Powers, supra note 3, at 641.
276. See id. (“The ongoing competition between House, Senate, and
Presidency for control over the administrative apparatus has created an
excessively politicized style of bureaucratic government.”).

ratings for trustworthiness among a variety of

institutions—with all three surveyed offices (GAO, CBO, and
CRS) easily outpacing the administration and bureaucratic
agencies, among others, in trustworthiness.277 The Cross-Gluck
study supported this conclusion, finding the congressional
bureaucracy to be widely perceived inside Congress as a highly
trusted source of professional and unbiased analysis.278 In the
pursuit of institutional neutrality, it seems that the
congressional bureaucracy may have been relatively successful.
In fact, although a provocative thesis, it is worth inquiring
whether the congressional bureaucracy might now be
considered a more reliably neutral institution than the federal
judiciary. Since Ackerman published his assessment twenty
years ago, the federal judiciary has undergone significant,
highly-visible efforts at politicization. In lower courts, the
appointment of judges deemed “not qualified” by the American
Bar Association has raised serious concerns that appointments
have been driven by political allegiance rather than neutral
competence.279 On the Supreme Court, the appointment of three
Justices who previously served on President Bush’s defense
277. Kosar, supra note 222, at 129. Professional committee staff ranked
very close to CBO—professional staff received a mean score for
trustworthiness on a scale from zero to three that was .01 higher than the
score for CBO for budget analysis, whereas CBO’s mean score was .02 higher
for healthcare. However, CBO still scored much more highly than all of the
other actors, and CRS and GAO both beat professional committee staff on both
budget and healthcare by large margins. Id.
278. See Cross & Gluck, supra note 9, at 1613–16 (“[T]he [congressional]
bureaucracy offices emphasized nonpartisanship as the defining characteristic
of their work.”).
279. See Philip Bump, How Unusual Are Trump’s ‘Not Qualified’ Judicial
Nominations?, WASH. POST (Dec. 15, 2017, 8:45 AM), https://perma.cc/3KT3-
37DY (noting that a higher percentage of President Donald Trump’s nominees
to the judiciary were determined “not qualified” within his first year of office
than his predecessors); see also Jordain Carney, Senate Confirms Trump
Judicial Pick Labeled ‘Not Qualified’ by American Bar Association, THE HILL
(Oct. 24, 2019, 3:40 PM), https://perma.cc/5FDM-SCQV (reporting that Justin
Walker was confirmed as a judge for the Western District of Kentucky in a
50–41 party-line vote despite being labeled “not qualified” by the American
Bar Association); Reis Thebault, Trump Nominee Who Is Anti-IVF and
Surrogacy Was Deemed Unqualified. She Was Just Confirmed., WASH. POST
(Dec. 4, 2019), https://perma.cc/8RM9-UEAM (reporting that Sarah Pitlyk was
confirmed as a judge for the Eastern District of Missouri despite being labeled
as “not qualified” by the American Bar Association).

282 80 WASH. & LEE L. REV. 229 (2023)
team in the politically-fraught case of Bush v. Gore280 have

raised similar concerns,281 as did Justice Kavanaugh’s openly
partisan testimony at his confirmation hearings.282 The removal
of the filibuster for judicial appointments has allowed more
appointments to proceed along party lines, deepening these
worries about a politicized appointment process.283 In the wake
of these changes, scholars have noted an erosion in the
trustworthiness and legitimacy of the courts as a neutral
arbiter.284 Recent public opinion polls have generally supported
280. 531 U.S. 98 (2000).

281. See Joan Biskupic, Supreme Court Is About to Have 3 Bush v. Gore
Alumni Sitting on the Bench, CNN (Oct. 17, 2020, 8:07 AM),
https://perma.cc/K557-P2BD (highlighting that Justices Amy Coney Barrett,
Brett Kavanaugh, and Chief Justice John Roberts assisted George W. Bush’s
legal team in Bush v. Gore).
282. See Tara Leigh Grove, The Supreme Court’s Legitimacy Dilemma, 132
HARV. L. REV. 2240, 2242 (2019) (reviewing RICHARD H. FALLON, JR., LAW AND
LEGITIMACY IN THE SUPREME COURT (2018)).
283. See Li Zhou, Senate Republicans Have Officially Gone “Nuclear” in
Order to Confirm More Trump Judges, VOX (Apr. 3, 2019, 4:42 PM),
https://perma.cc/DK8H-NFW7 (reporting that Republicans had invoked “the
so-called ‘nuclear option’ . . . [to amend] Senate rules in order to further limit
the amount of time lower-level [judicial] nominees could be debated on the
floor”).
284. See Grove, supra note 282, at 2250 (arguing that during “politically
charged moments [such as the present], the Justices may not be able to protect
the Court’s sociological legitimacy without sacrificing the legal legitimacy of
their decisions (or vice versa)”); Daniel Epps & Ganesh Sitaraman, How to
Save the Supreme Court, 129 YALE L.J. 148, 150 (2019) (warning that “[r]ecent
events have already taken a toll on perceptions of the Court’s legitimacy”); Aziz
Z. Huq, Why Judicial Independence Fails, 115 NW. U. L. REV. 1055, 1056 (2021)
(opining that “[t]o judge from polemics of the day, judicial independence is
beset by enemies upon all sides”); Lee Epstein & Eric Posner, Opinion, If the
Supreme Court Is Nakedly Political, Can It Be Just?, N.Y. TIMES (July 9, 2018),
https://perma.cc/T8C6-C2GQ (cautioning that soon “it will become impossible
to regard the court as anything but a partisan institution”); Erwin
Chemerinsky, With Kavanaugh Confirmation Battle, the Supreme Court’s
Legitimacy Is in Question, SACRAMENTO BEE (Oct. 7, 2018),
https://perma.cc/9XTR-R6RH (noting “the cloud over the court’s legitimacy”);
Justin Wise, Holder: Supreme Court’s Legitimacy Can Be Questioned After
Kavanaugh Confirmation, THE HILL (Oct. 8, 2018, 8:03 AM),
https://perma.cc/5TJ6-7T2D (reporting tweet by former Attorney General
Holder stating, “With the confirmation of Kavanaugh and the process which
led to it, (and the treatment of Merrick Garland), the legitimacy of the
Supreme Court can justifiably be questioned”); see also RICHARD H. FALLON,
JR., LAW AND LEGITIMACY IN THE SUPREME COURT 148 (2018) (noting that we

these observations.285 Ackerman himself warned in 2018 that,

without significant reforms, political partisanship “will
predictably destroy the court’s legitimacy in the coming
decade.”286 It is important not to conflate sociological and legal
legitimacy, of course, and perhaps the factors that have led to
perceptions of politicization of the judiciary have not actually
undermined its neutrality.287 If these assessments are correct,
however, and if the judiciary has indeed undergone a diminution
in neutrality in the last several decades, then it is plausible that
the congressional bureaucracy now actually surpasses the
judiciary on the metric of neutrality.
If true, this may have implications for statutory
interpretation. For example, several scholars have argued that
courts should adopt canons of statutory interpretation that defer
to prior interpretations or conclusions offered by offices in the
congressional bureaucracy, such as CBO,288 JCT,289 or the
live in “an era of hermeneutic suspicion” while arguing for a broadened

understanding of legal legitimacy).
285. See Jeffrey M. Jones, Approval of U.S. Supreme Court Down to 40%,
A New Low, GALLUP (Sept. 23, 2021), https://perma.cc/FDT6-4BFL
(“[O]pinions of the U.S. Supreme Court have worsened, with 40%, down from
49% in July, saying they approve of the job the high court is doing. This
represents, by two percentage points, a new low in Gallup’s trend, which dates
back to 2000.”); see also Positive Views of Supreme Court Decline Sharply
Following Abortion Ruling, PEW RSCH. CTR. (Sept. 1. 2022),
https://perma.cc/CW3J-5327 (“Americans’ ratings of the Supreme Court are
now as negative as—and more politically polarized than—at any point in more
than three decades of polling on the nation’s highest court.”).
286. Bruce Ackerman, Opinion, Trust in the Justices of the Supreme Court
Is Waning. Here Are Three Ways to Fortify the Court, L.A. TIMES (Dec. 20, 2018,
3:15 AM), https://perma.cc/UL6J-DHVB.
287. See FALLON, supra note 282.
288. See Bressman & Gluck, Part II, supra note 54, at 782 (suggesting that
a “CBO score could help courts reflect congressional expectations” for
statutes); Gluck, supra note 55, at 187–91 (discussing how the CBO canon can
improve statutory interpretation); see also Abbe R. Gluck, The “CBO Canon”
and the Debate Over Tax Credits on Federally Operated Health Insurance
Exchanges, BALKINIZATION (July 20, 2012), https://perma.cc/TV63-7CJX
(positing that the CBO “budget score offers better evidence of congressional
‘intent’ than other commonly consulted non-textual tools, including legislative
history”).
289. See Wallace, supra note 57, at 183.

284 80 WASH. & LEE L. REV. 229 (2023)
Parliamentarians’ offices.290 Typically, these arguments have

been grounded in an intent-based logic—that is, on the idea that
the congressional bureaucracy, because it is closer to Congress,
has a comparative advantage to the judiciary in its ability to
access and report congressional intent.291 These arguments,
however, might further be grounded in a neutrality-based
logic—that is, on the idea that the congressional bureaucracy
currently is the best-equipped actor to offer neutral, good-faith
assessments of legislation. Courts would be unlikely to
expressly adopt this logic, of course, since it entails recognition
of the erosion of neutrality within the courts themselves. Yet it
might provide an implicit reason for outsiders to more strongly
advocate for judicial use of, and deference to, these nonpartisan
congressional assessments.292
B. Expertise
The congressional bureaucracy also changes how we think

about the comparative capacity of the branches to cultivate and
use expertise. Traditionally, the executive has been viewed as
the outlier in federal governance: with its administrative
agencies, it has been viewed as uniquely equipped to develop
and use expertise.293 The congressional bureaucracy changes
this view. At a minimum, it reframes the branches along a
spectrum that ranks the branches as: executive (most expertise),
Congress (medium expertise), and courts (least expertise).
Arguably, it even supports a view of the judiciary as outlier: a
branch uniquely ill-equipped to develop and evaluate expert
290. See Gould, supra note 55, at 2023 (discussing the rationale for relying
on certain parliamentary procedures, such as rulings of the chair, when
judicially interpreting statutes).
291. But see Wallace, supra note 57, at 224–30.
292. On courts’ slow but growing use of these documents, see Cross &
Gluck, supra note 9, at 1650–51, 1676.
293. See, e.g., Wagner, supra note 239, at 2023 (describing the
“agency-as-expert” view and noting that “the basic concept that the agencies
should preside over specialized information is hard-wired into the design of
the administrative state”); William W. Buzbee & Robert A. Schapiro,
Legislative Record Review, 54 STAN. L. REV. 87, 122 (2001) (“[O]ne of the key
justifications for the administrative state is to allow expert agencies to exercise
their judgment and experience in resolving tasks delegated by Congress.”
(citing Sunstein, supra note 19)).

information.294 This revised understanding of comparative

institutional competence has several implications.
First, it suggests that courts should display greater
deference to congressional expertise. In the administrative law
context, courts broadly recognize that their comparative
expertise deficit provides reason for greater deference to the
executive branch.295 The same should be true when Congress
uses its bureaucracy. Recently, for example, the Supreme Court
faced the question of whether the individual mandate296 was
severable from the larger Affordable Care Act (ACA).297 The
Court avoided this question on standing grounds,298 but Justice
Alito’s dissent squarely addressed it—and argued that was not
severable.299 In reaching this conclusion, Justice Alito ignored
the fact that CBO concluded in 2017 that the provision was
294. On judiciary as an outlier, see A. Christopher Bryant & Timothy J.

Simeone, Remanding to Congress: The Supreme Court’s New “On the Record”
Constitutional Review of Federal Statutes, 86 CORNELL L. REV. 328, 391 (2001)
(“Regarding competency, there can be little doubt that federal judges, and
particularly appellate court judges, are singularly ill suited to make the kinds
of factual determinations and predictive judgments that must be made in
assessing whether legislation answers a real, rather than merely perceived,
national problem.”).
295. See, e.g., Babbitt v. Sweet Home Chapter of Communities for a Great
Or., 515 U.S. 687, 703–04 (1995) (“The latitude the ESA gives the Secretary in
enforcing the statute, together with the degree of regulatory expertise
necessary to its enforcement, establishes that we owe some degree of deference
to the Secretary’s reasonable interpretation.”); see also Bryant & Simeone,
supra note 294, at 371–72 (“[T]he modern administrative state’s claim to
legitimacy has been, from the outset, one of expertise rather than political
accountability. Accordingly, judicial review, and particularly review based on
the formal administrative record, is necessary to ensure that agency decision
making is genuinely a product of that expertise.”); Renan, supra note 113, at
2255–56
Underlying a range of doctrines of judicial deference is an antecedent
question of institutional choice: which is the more competent institutional
actor to decide the legal question at issue, the courts or the President? The
answer to that question is predicated (sometimes implicitly, sometimes
explicitly) on a set of assumptions about how these institutions actually
work.
296. The mandate is currently a zero-dollar tax. See I.R.C.
§ 5000A(c)(3)(A).
297. See California v. Texas, 141 S. Ct. 2104, 2113 (2021).
298. See id.
299. See id. at 2140 (Alito, J., dissenting).

286 80 WASH. & LEE L. REV. 229 (2023)
severable300—that is, the ACA could survive without it.301 Yet,

not only is CBO closer to Congress, it also has economic
expertise that Justice Alito (and the rest of the Court) lacks.302
In such situations, an expertise-based rationale supports
deference to the CBO assessment, thereby supplementing the
intent-based and neutrality-based rationales for deference
discussed in Subpart VI.A.
Deference to congressional expertise is additionally
important due to the judicial practice that Buzbee and Schapiro
have labeled “legislative record review.”303 Under this practice,
the Court requires legislation to be supported by sufficient
factfinding documented in the legislative record.304 This practice
made a notable appearance in Shelby County v. Holder,305 the
2013 case that struck down a key provision in the Voting Rights
Act.306 By adopting this practice in the last few decades,307 the
300. See generally CONG. BUDGET OFF., REPEALING THE INDIVIDUAL HEALTH
INSURANCE MANDATE: AN UPDATED ESTIMATE (2017), https://perma.cc/74HG-
3VV3 (PDF). See also Brief for Jonathan H. Adler et al. as Amici Curiae
Supporting Petitioners at 20, California v. Texas, 141 S. Ct. 2104 (2021) (Nos.
19-840, 19-1019) (“Before Congress acted in 2017, the Congressional Budget
Office had analyzed the effects both of repealing the individual mandate and
of eliminating the penalties while keeping the mandate in place.”).
301. On the severability standard, see Ayotte v. Planned Parenthood of N.
New Eng., 546 U.S. 320, 329 (2006) (noting that courts should “try not to
nullify more of a legislature’s work than is necessary . . . [since] a ‘ruling of
unconstitutionality frustrates the intent of the elected representatives of the
people’” (quoting Regan v. Time, Inc., 468 U.S. 641, 652 (1984)).
302. See Organization and Staffing, CONG. BUDGET OFF.,
https://perma.cc/UY5P-QPFW (“The [CBO] has about 275 staff members,
mostly economists or public policy analysts with advanced degrees . . . . Many
of them have extensive experience in their subject areas, including years of
work at CBO.”).
303. Buzbee & Schapiro, supra note 293, at 89 (noting that recent
jurisprudence has “raised legislative record review to new, dispositive
significance”).
304. See Bryant & Simeone, supra note 294, at 330 (“[T]he Court has, in
recent years, become increasingly aggressive in striking down federal statutes
because the formal legislative record inadequately supports a factual judgment
underlying congressional action.”).
305. 570 U.S. 529 (2013).
306. Id.
307. Most commentators trace the emergence of this practice to a string of
cases in the 1990s and early 2000s. See Ruth Colker & James J. Brudney,
Dissing Congress, 100 MICH. L. REV. 80, 122–23 (2001) (describing how the
Rehnquist Court “largely abandoned” its traditional deferential standard and

Court has dramatically narrowed its deference to

Congress308—and also has shifted its underlying logic. Once
anchored in ideas about democratic accountability,309 this
deference now appears to rely on ideas about publicly
documented exercises of expertise.310
Legislative record review strongly resembles a review
practice already applied to administrative agencies,311 and
replaced it with “a more constrained version in which Congressional

factfinding [is] subject to searching, skeptical review”). See generally, e.g.,
Buzbee & Schapiro, supra note 293; Bryant & Simeone, supra note 294.
308. See Colker & Brudney, supra note 307, at 99 (“Congress under
existing precedent merely needed to establish that there was a ‘rational basis’
for concluding that a regulated activity sufficiently affected interstate
commerce.”). For statement of typical approach under prior precedent, see
Hodel v. Va. Surface Min. & Reclamation Ass’n, Inc., 452 U.S. 264, 276 (1981)
(“The court must defer to a congressional finding that a regulated activity
affects interstate commerce, if there is any rational basis for such a finding.”).
309. See United States v. Lopez, 514 U.S. 549, 604 (1995) (Souter, J.,
dissenting) (“The practice of deferring to rationally based legislative
judgments . . . reflects . . . our appreciation of the legitimacy that comes from
Congress’s political accountability in dealing with matters open to a wide
range of possible choices.”); see also Colker & Brudney, supra note 307, at 119
[T]his dynamic and regularly recorded tension between Congress and its
constituencies contributes importantly to the legitimacy of the lawmaking
process, and helps explain the presumption of judicial deference to the final
product. The Court’s insistence on a type of pristine ‘substantial evidence’
approach slights another of Congress’s distinctive institutional virtues—the
politically accountable nature of its record building enterprise.
see also Maria Ponomarenko, Administrative Rationality Review, 104 VA. L.
REV. 1399, 1424 (2018) (“Often what bridges the gap—and justifies the highly
deferential standard of review–—is the idea that legislatures, unlike courts,
are accountable to the majority will.”); Renan, supra note 113, at 2256 n.383
(noting that “the Court defers to the legislature because of its democratic
pedigree” and removes deference at signs of democratic malfunction); Buzbee
& Schapiro, supra note 293, at 154 (noting that “the lack of democratic checks
on judges’ questioning of legislative judgments is problematic”). But see Turner
Broad. Sys., Inc. v. FCC, 520 U.S. 180, 196 (1997) (noting that “the deference
to Congress is in one respect akin to deference owed to administrative agencies
because of their expertise,” but courts also “owe Congress’ findings an
additional measure of deference out of respect for its authority to exercise the
legislative power”).
310. See Bryant & Simeone, supra note 294, at 385 (criticizing record
review because “Congress grounds its claim to legitimacy on knowledge of and
accountability to the citizens it represents, not on an assertion of neutrality or
technical expertise”).
311. See, e.g., Colker & Brudney, supra note 307, at 83 (noting that “the
new activist majority has treated the federal legislative process as akin to

288 80 WASH. & LEE L. REV. 229 (2023)
many prominent scholars have observed the inappropriateness

of its extension to Congress.312 Awareness of the congressional
agency or lower court decisionmaking”); Margaret B. Kwoka, Setting Congress

Up to Fail, 16 BERKELEY J. AFR.-AM. L. & POL’Y 97, 97 (2015) (“The Court has
increasingly required Congress to meet procedural standards akin to those
required of administrative agencies promulgating rules, in particular by
measuring legislation against the record Congress created.”); Harold J. Krent,
Turning Congress into an Agency: The Propriety of Requiring Legislative
Findings, 46 CASE W. RES. L. REV. 731, 733 (1996) (explaining how legislative
findings “might transform the legislature into a type of administrative agency,
monitored and controlled by the superintending judiciary”); Philip P. Frickey
& Steven S. Smith, Judicial Review, the Congressional Process, and the
Federalism Cases: An Interdisciplinary Critique, 111 YALE L.J. 1707, 1751
(2002) (describing how legislative record review “forc[es] Congress to behave
like an administrative agency”); Bryant & Simeone, supra note 294, at 331
(critiquing record review because “Congress is simply not an administrative
agency”); Buzbee & Schapiro, supra note 293, at 90 (asserting that “[r]eference
to administrative law . . . helps to clarify the scope and purpose of the new
legislative record review and to illustrate its flaws,” while also noting
important differences); Stephen Gardbaum, Rethinking Constitutional
Federalism, 74 TEX. L. REV. 795, 826 (1996) (comparing the judicial review
standard based on “the ‘hard look’ doctrine in administrative law” to the
“background principle of federalism that informs what is ‘proper’” that is
required of Congress before exercising its preemption power); see also Kwoka,
supra, at 97–98 (viewing Shelby County as creating a Chenery principle for
legislatures).
312. This review has been criticized because it: is unsupported by
precedent, see, e.g., Buzbee & Schapiro, supra note 293, at 90; violates the
separation of powers, see, e.g., Bryant & Simeone, supra note 294, at 372–73
(“notions of separation of powers suggest that the judiciary has little authority
to impose [such] requirements”); ignores the constitutional autonomy of
Congress in its procedures, see, e.g., Frickey & Smith, supra note 311, at
1749–50 (detailing how judicial influence over congressional review “seems in
tension with the Constitution itself, which provides that each house is
responsible for making its own rules”); is grounded in incorrect understanding
of legislative factfinding, see, e.g., Kwoka, supra note 311, at 98–99 (describing
how legislative record review “ignores the reality of the legislative process”
which “inherently occurs in large part off the record”); ignores that legislatures
can ground decisions in political factors, see id. at 101 (nothing legislative
record review as “inappropriate for decisions made by a body of elected
representatives who, by their nature, are not simply exercising objective
expertise, are not expected to act as neutral decision makers”); and
undermines other legitimate democratic functions served by congressional
procedures such as informing the public, see, e.g., Bryant & Simeone, supra
note 294, at 384 (suggesting that a court-imposed legislative record review
“threatens to limit the other legitimate purposes served by congressional
proceedings”).

bureaucracy deepens these critiques.313 As the congressional

bureaucracy reveals, the deployment of expertise changes across
the political branches—and does so in ways that legislative
record review problematically overlooks.
For example, expertise in the congressional bureaucracy
has a unique connection to legislative supremacy.
Administrative record review polices a bureaucracy that is
difficult for Congress to oversee and correct, given that
bureaucracy’s post-enactment role and dual allegiances. In this
context, legislative supremacy may require courts to ensure
bureaucratic fidelity to legislative directives, including via
record review.314 By contrast, legislative record review polices a
congressional bureaucracy that Congress is structurally
competent to oversee without assistance, as Part II explained.
Legislative record review therefore does not address a
comparable need to bolster legislative supremacy.
Expertise in the congressional bureaucracy also is subject
to unique statutory mandates. Congress has imposed
transparency requirements on the executive-branch
bureaucracy: agencies must publish records under the APA,315
and courts must review them.316 By contrast, statutory
mandates for the congressional bureaucracy often include
confidentiality requirements: several offices in this bureaucracy
313. For mention of the congressional bureaucracy in prior work on

legislative record review, see Bryant & Simeone, supra note 294, at 385–87.
314. See Buzbee & Schapiro, supra note 293, at 121 (“By exercising a
variety of reviewing functions in overseeing agency actions, courts ensure that
agencies abide by the authority conveyed by the legislature.”).
315. 5 U.S.C. § 553(c) (requiring a “concise general statement of [the] basis
and purpose” to accompany each final rule); see also 1 RICHARD J. PIERCE, JR.,
ADMINISTRATIVE LAW TREATISE § 7.4, at 596 (5th ed. 2010) (“The courts have
replaced the statutory adjectives, ‘concise’ and ‘general’ with the judicial
adjectives ‘detailed’ and ‘encyclopedic.’”). For relevant agency procedural
requirements generally, see 5 U.S.C. §§ 551–559, 701–706. Agencies may have
additional procedural requirements in their enabling acts. See Buzbee &
Schapiro, supra note 293, at 131 (“Judicial review of agency action is required
by the Administrative Procedure Act and most enabling acts.”); see also Bryant
& Simeone, supra note 294, at 370 (“[T]he Court’s emerging approach to the
legislative record resembles nothing so much as the careful, on-the-record
review of agency action that has developed under the APA.”).
316. Administrative Procedure Act (APA) § 2, 5 U.S.C. § 706; see also
Buzbee & Schapiro, supra note 293, at 141 (“To the extent suspicion is
justified, the suspicion stems from a congressional choice to confer a skeptical
role on the courts.”).

290 80 WASH. & LEE L. REV. 229 (2023)
are statutorily required to operate confidentially and no statute

requires judicial review of them.317 It is puzzling that the Court
could treat the resulting bureaucratic records as equally
comprehensive, given these different statutory foundations.
Expertise in the congressional bureaucracy also may be
dependent upon institutional dynamics that are uniquely
undermined by judicial review. Scholars have noted that agency
record review beneficially entrenches the executive bureaucracy
since it effectively requires public use of this bureaucracy’s
expertise.318 Legislative record review threatens to have the
opposite effect for the congressional bureaucracy, however. The
Cross-Gluck study found that confidentiality is key to the
legitimacy of the congressional bureaucracy, and its
institutional threats often have arisen from public performance
of its expertise.319 Legislative record review therefore threatens
to undermine the very expertise it ostensibly seeks in ways its
executive-branch counterpart does not.
317. Congress has written confidentiality obligations into the organic

statutes of CRS, House Legislative Counsel, and CBO. 2 U.S.C. § 166(i)
(requiring the Director of CRS to prepare and file a separate and special report
covering all phases of activity by CRS for the preceding fiscal year and any
efforts to make additional, nonconfidential products or services available to
the Librarian of Congress for publication); 2 U.S.C. § 281(a); 2 U.S.C. § 603(d);
see also Buzbee & Schapiro, supra note 293, at 90 (noting that legislative
record review differs in that it “lacks any legislative imprimatur”).
318. See Michaels, supra note 62, at 251
Judges can (and do) bolster the position of civil servants vis-à-vis their
administrative rivals by using various administrative tools and
doctrines . . . to encourage and endorse well reasoned agency actions
appropriately influenced by career staffers, and to more aggressively
scrutinize (if not categorically reject) agency actions undertaken without
much expert input or procedural rigor.
Ingber, supra note 7, at 181 (noting that “the courts also allocate power to
different actors within the executive branch both implicitly and explicitly”
through rules such as deference doctrines); Magill & Vermeule, supra note 81,
at 1056 (“Courts are inclined to defer to executive officials, especially the
President, and afford the barest rational basis scrutiny to administrative and
presidential action.”); Renan, supra note 113, at 2254–65 (noting and
critiquing the Court’s practice of “indirect enforcement” of agency norms).
319. See Cross & Gluck, supra note 9, at 1626–28 (illustrating how
congressional staff and bureaucratic entities will frequently rely on holistic
informal and confidential consultation, where “the more transparent the
bureaucracy work has been, the greater the public oversight and the more it
has come under . . . pressure”).

In its deployment of legislative record review, the Court

makes no effort to account for these differences in the operation
of bureaucratic expertise across the branches.320 The
congressional bureaucracy therefore adds to the catalogue of
problems with this judicial practice, further suggesting that the
Court should not continue it.321 If the Court does continue this
practice, however, it should take greater notice of the
congressional bureaucracy, which it has made little effort to
understand to date.322 The Court’s insistence on judicial review
of congressional factfinding, when paired with minimal interest
in understanding it, only reinforces concerns among scholars323
and Justices324 that legislative record review is a
320. If anything, it is more stringently applied to Congress. See Buzbee &

Schapiro, supra note 293, at 119–20 (“[I]n many ways, legislative record
review embodies a more probing, less deferential kind of scrutiny.”).
321. For a list of other problems with practice, see supra note 312 and
accompanying text.
322. See Cross & Gluck, supra note 9, at 1650–51 (listing the Court’s rare
citations to bureaucracy’s materials); id. at 1668–70 (noting the Court’s lack
of awareness of congressional bureaucracy in Yates v. United States);
Transcript of Oral Argument at 39–42, Lockhart v. United States, 577 U.S.
347 (2016) (No. 14-8358) (recording Justice Breyer citing the role of staff in
Congress and Justices Scalia and Roberts expressing skepticism of such staff
involvement); see also Bryant & Simeone, supra note 294, at 383 (“[I]n the last
thirty years the number and quality of these extra-record sources have
increased significantly. The Court’s failure to acknowledge these sources has,
not surprisingly, resulted in decisions characterized by an impoverished
appreciation of the legislative process.”); Colker & Brudney, supra note 307,
at 118 (demonstrating how legislative record review shows “the Court’s
unwillingness to recognize or respect how Congress educates itself”).
323. See Colker & Brudney, supra note 307, at 123–26 (noting that “the
Court’s rigorous approach to the legislative history may [sometimes] be
windowdressing”); id. at 136 (describing how recent Supreme Court decisions
reveal a standard that could portend “a Court determined to withdraw
authority from Congress”); Buzbee & Schapiro, supra note 293, at 154 (“The
prospect of such massive legislative documentation . . . supports the view that
the Court is actually masking broad political distrust with ostensible record
scrutiny.”); Mitchell N. Berman, Guillen and Gullibility: Piercing the Surface
of Commerce Clause Doctrine, 89 IOWA L. REV. 1487, 1499 n.51 (2004)
(asserting that “Morrison revealed the absence-of-findings criticism to be close
to a make-weight”).
324. See United States v. Lopez, 514 U.S. 549, 614 (1995) (Souter, J.,
dissenting) (“Such a legislative process requirement would function merely as
an excuse for covert review of the merits of legislation under standards never
expressed and more or less arbitrarily applied.”).

292 80 WASH. & LEE L. REV. 229 (2023)
politically-motivated project, not a good-faith effort to confirm

the factfinding of Congress.
Moving beyond the topic of deference, the revised view of
congressional expertise also counsels courts to adopt a different
understanding of legislation. Due largely to the congressional
bureaucracy, the crafting of legislation now is a formidable
bureaucratic undertaking that is informed by a variety of
specialists and experts. In the agency context, courts have long
recognized that this requires a shift in how it approaches agency
materials. In statutory interpretation, however, the Court
continues to advance doctrines which assume that federal
legislation is a generalist’s endeavor. These include its recent
emphasis on ordinary public meaning in statutory
interpretation325 and its reduced deference to expert agency
interpretations.326 There may be valid reasons for courts to
approach legislation this way, of course.327 In the era of the
congressional bureaucracy, however, this approach does have
325. See, e.g., Bostock v. Clayton County, 140 S. Ct. 1731, 1738 (2020)
(“This Court normally interprets a statute in accord with the ordinary public
meaning of its terms at the time of its enactment.”).
326. See, e.g., King v. Burwell, 576 U.S. 473, 485–86 (2015) (applying the
major questions doctrine to deny deference); see also Kisor v. Wilkie, 139 S. Ct.
2400, 2432–34 (2019) (Gorsuch, J., concurring) (arguing that Section 706 of
the APA forbids judicial deference to administrative interpretations);
Gutierrez-Brizuela v. Lynch, 834 F.3d 1142, 1152 (10th Cir. 2016) (describing
Chevron as an “abdication of judicial duty”); Marla D. Tortorice, Nondelegation
and the Major Questions Doctrine: Displacing Interpretive Power, 67 BUFF. L.
REV. 1075, 1089–1101 (2019) (discussing the major questions doctrine and
Justice Gorsuch’s rejection of Chevron deference during his judicial career).
327. For example, textualists cite the notice-providing function of statutes
as a reason to interpret from the vantage of ordinary citizens. See, e.g.,
ANTONIN SCALIA, Common-Law Courts in a Civil-Law System: The Role of
United States Federal Courts in Interpreting the Constitution and Laws, in A
MATTER OF INTERPRETATION: FEDERAL COURTS AND THE LAW 3, 17 (Amy
Gutmann ed., 1997) (emphasizing notice as an essential quality advanced by
statutes); Ryan D. Doerfler, Who Cares How Congress Really Works?, 66 DUKE
L.J. 979, 981 (2017) (observing scholars’ focus on notice and arguing on behalf
of it); OFF. OF LEGAL POL’Y, DEP’T OF JUST., J 1.96:H 62, USING AND MISUSING
LEGISLATIVE HISTORY: A RE-EVALUATION OF THE STATUS OF LEGISLATIVE
HISTORY IN STATUTORY INTERPRETATION 52 (1989) (making an argument for
notice by analyzing presumption of knowledge of the law and legislative
history). For a critique of this approach, see generally Jesse M. Cross,
Disaggregated Legislative Intent, 90 FORDHAM L. REV. 2221 (2022). On
nondelegation argument against Chevron deference, see Tortorice, supra note
326.

the collateral effect of widening the gap between Congress and

the courts. In so doing, it inhibits the construction of a shared
linguistic context between Congress and courts, a goal some
textualists have sought.328 While these textualists have argued
for Congress to bear the burden of constructing this shared
context,329 it is unclear why this burden should not fall instead
on the courts, given the Constitution’s emphasis on insulating
congressional practice from outside influence.330 One modest
way for courts to assist in the construction of shared context
would be to shift away from this focus on a generalist lens and
instead understand the ways that bureaucratic expertise shapes
modern legislation.
More drastic methods of constructing a shared linguistic
context exist as well. If the courts increasingly are outliers in an
otherwise specialized federal government, it raises the question
of whether there should be greater specialization in the courts,
for example.331 Agencies already specialize by subject matter,
and Congress has long divided work by topic through the
328. See John F. Manning, Textualism and Legislative Intent, 91 VA. L.

REV. 419, 433–36 (2005) (describing textualists’ desire to understand
legislative language and intent “as accurately as possible”); Doerfler, supra
note 327, at 983 (arguing that “context consists of information salient to both
author and audience”).
329. See Manning, supra note 328, at 433 (“If one cannot accurately
ascertain what the body as a whole would have done . . . then perhaps the best
one can do is to approximate the way a reasonable person in the legislator’s
position would have read the words actually adopted.”).
330. See Bryant & Simeone, supra note 294, at 369–73 (noting that “the
Constitution itself both expressly and implicitly imposes limits on judicial
intrusion into the workings of Congress” and citing the
bicameralism-and-presentment provision, Rules and Journal Clauses, the
Speech or Debate Clause, and the political question doctrine); Frickey &
Smith, supra note 311, at 1749–50 (“[T]he judicial intrusion into internal
congressional processes seems in tension with the Constitution itself, which
provides that each house is responsible for making its own rules . . . . It is
[deeply problematic] for the Court to impose procedural obligations upon
Congress going far beyond the Constitution or the houses’ own rules.”).
331. For the argument that generalist courts already use case assignment
to specialize somewhat, see Edward K. Cheng, The Myth of the Generalist
Judge, 61 STAN. L. REV. 519, 550–51 (2008) (detailing how judges are exposed
to different types of cases, allowing an informal system of opinion
specialization to develop at their discretion in particular areas and degrees of
specialization).

294 80 WASH. & LEE L. REV. 229 (2023)
committee system.332 The Cross-Gluck study showed that,

throughout the congressional bureaucracy, specialization by
subject matter also is the norm.333 Drafters in Legislative
Counsel specialize by topic, for example, allowing them to draft
with deep and detailed knowledge of the relevant statutory
regime.334 Would judicial interpreters benefit from a
correspondingly deep knowledge of the statutory regime in a
subject area, one that mirrored the knowledge of the statutory
drafters and agency implementers? This would be another way
to foster a shared linguistic context between Congress and
courts.335 Specialized courts may raise their own concerns, of
course.336 But in the era of a specialized congressional
bureaucracy, they also may have more benefits than previously
realized.
CONCLUSION
Legislation scholarship is only beginning to understand a

congressional bureaucracy that, for many years, has been
overlooked and under-theorized. Executive-branch scholarship
has developed a wealth of theories about bureaucracy—theories
that cry out for testing in contexts outside the executive branch.
Neither has devoted attention to the lessons each can offer the
other—or to the larger institutional design possibilities that
exist when we examine bureaucratic possibilities not only
within branches, but across them. Such a collaborative approach
332. See LIBR. OF CONG., Official U.S. Executive Branch Web Sites,
https://perma.cc/7R82-H5B5 (last updated Mar. 30, 2022) (listing agencies);
Committees of the U.S. Congress, https://perma.cc/EF7F-67AD.
334. See id.
335. See Manning, supra note 328, at 434; Doerfler, supra note 327, at 983.
336. See Cheng, supra note 331, at 554–57 (discussing politicization,
myopia, loss of prestige, and other concerns); Jed Rakoff, Judge, U.S. Dist. Ct.
(S.D.N.Y.), The Eleventh Annual Albert A. DeStefano Lecture on Corporate
Securities & Financial Law at the Fordham Corporate Law Center: Are
Federal Judges Competent? Dilettantes in an Age of Economic Expertise, in
17 FORDHAM J. CORP. & FIN. L. 4, 7–14 (2012) (noting that specialization in
courts can obscure reasoned elaboration, introduce problems of capture and
ideology, promote improper professional concerns among judges, promote
judicial tunnel vision, and prevent fresh examination of ideas); see also Sapna
Kumar, Patent Court Specialization, 104 IOWA L. REV. 2511, 2519–527 (2019)
(discussing problems arising from Federal Circuit specialization in patents).

holds significant promise for any who seek a larger

understanding of how bureaucracy operates, and can operate, in
our presidentialist system.

SSRN Id4458504

Uploaded by

Copyright:

Available Formats

SSRN Id4458504

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

SSRN Id4458504

Uploaded by

Copyright:

Available Formats

Washington and Lee Law Review

Volume 80 Issue 1 Article 6

Federal Bureaucratic Studies

Follow this and additional works at: https://scholarlycommons.law.wlu.edu/wlulr

Part of the Law Commons

Electronic copy available at: https://ssrn.com/abstract=4458504

A vast literature has developed in legal scholarship on the

* Associate Professor, University of South Carolina School of Law. The

Electronic copy available at: https://ssrn.com/abstract=4458504

cross-disciplinary dialogue on the role of bureaucracy in our

INTRODUCTION .................................................................. 230

In legal scholarship, a vast literature has developed on the

Electronic copy available at: https://ssrn.com/abstract=4458504

including administrative law,2 separation-of-powers law,3 and

2. For discussions involving scholarship generally considered

Electronic copy available at: https://ssrn.com/abstract=4458504

(“OMB”).8 Through a study of these and other executive branch

Electronic copy available at: https://ssrn.com/abstract=4458504

It is not surprising that bureaucracy scholarship has

15. See infra Part I.

Electronic copy available at: https://ssrn.com/abstract=4458504

 Dual advising/adjudicating roles: Some bureaucratic

Electronic copy available at: https://ssrn.com/abstract=4458504

looks at the problem of agency capture, and Part VI examines

I. BACKGROUND: A TALE OF TWO BUREAUCRACIES

Since the inception of the modern administrative state,

19. While typically associated with the New Deal, administrative

Electronic copy available at: https://ssrn.com/abstract=4458504

executive branch bureaucracy regularly generates

25. See, e.g., Bureaucracy and Presidential Administration: Expertise and

Mar. 27, 1976), https://perma.cc/EN7K-Q27Q.

Electronic copy available at: https://ssrn.com/abstract=4458504

offices inside Congress, as well as the new legislative process

Stephen Breyer, On the Uses of Legislative History in Interpreting Statutes, 65

Electronic copy available at: https://ssrn.com/abstract=4458504

that exists today: a bureaucracy with over 4,000 expert

37. See id. at 1599–1600.

Electronic copy available at: https://ssrn.com/abstract=4458504

 Office of the Law Revision Counsel (OLRC): Staff who

Electronic copy available at: https://ssrn.com/abstract=4458504

Electronic copy available at: https://ssrn.com/abstract=4458504

movement have directed attention to individual offices in the

& Lisa Schultz Bressman, Statutory Interpretation from the Inside—An

Electronic copy available at: https://ssrn.com/abstract=4458504

offices of the congressional bureaucracy, although not

II. DUAL-ALLEGIANCE PROBLEM

By turning to ongoing discussions in executive-branch

Electronic copy available at: https://ssrn.com/abstract=4458504

including loss of presidential control over the bureaucracy,61

Electronic copy available at: https://ssrn.com/abstract=4458504

supervision,72 and regular use of its services,73 a convergence of

Check, Please: Constitutional Dimensions of Halting the Pay of Public

Electronic copy available at: https://ssrn.com/abstract=4458504

influence by developing close working relationships with the

77. See Cross & Gluck, supra note 9, at 1615–16.

Electronic copy available at: https://ssrn.com/abstract=4458504

work and responsibility in the executive bureaucracy.81 The

Electronic copy available at: https://ssrn.com/abstract=4458504

rise of the administrative state, the nondelegation doctrine is