How the Administrative State Got to - Am MIT spezialisierte KI-Forschung

How the Administrative State Got to
This Challenging Place

Peter L. Strauss

Written for a dispersed agrarian population using hand tools in a local economy,
our Constitution now controls an American government orders of magnitude larg-
er that has had to respond to profound changes in transportation, communication,
Technologie, economy, and scientific understanding. How did our government get to
this place? The agencies Congress has created to meet these changes now face pro-
found new challenges: transition from the paper to the digital age; the increasing
centralization in an opaque, political presidency of decisions that Congress has as-
signed to diverse, relatively expert and transparent bodies; the thickening, sowie,
of the political layer within agencies themselves; and the increasing judicial use of
analytic techniques invoking the expectations of those who wrote the Constitution
so long ago and in such different circumstances. Never easy, finding the appropriate
balance between law and politics presents major challenges today.

A s the United States enters the third decade of the twenty-first century,

almost two-and-a-half centuries after its Constitution was written, its
federal government employs more than two million civilian employees.1
Of these, mehr als 1,800 work directly for the President, in the Executive Office
of the President (EOP). Virtually all the remainder–outside the seventy thousand
or so employed by Congress and the federal judiciary–work in hundreds of gov-
ernment agencies and other institutions, performing tasks assigned to them by
congressional legislation.2

Our Constitution’s text addressing America’s government (as distinct, Das
Ist, from the particular institutions of Congress and the presidency itself ) hat
not been amended since the founding. Although conservative and libertarian
voices increasingly insist that, absent amendment, only the founders’ under-
standings can be honored, our Constitution must somehow be understood in
relation to today’s dramatically different circumstances, if our government is
to continue functioning. In 1791, the first American census reported a popula-
tion of 3,929,214 inhabiting an area of 864,746 square miles, roughly one percent
of today’s population, and one-quarter its present area with, dazugehörigen-
ly, a much lower population density. Its economy was predominantly agrarian,

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u
D
A
e
D
A
R
T
ich
C
e
–
P
D

F
/

1
5
0
3
1
7
2
0
6
0
4
7
2
D
A
e
D
_
A
_
0
1
8
5
7
P
D

B
j
G
u
e
S
T

Ö
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

leavened by small, local artisans and other businesses dealing directly with cus-
tomers. Both travel and communication were impeded by distance, the means
of transportation, and the available communication technology. The first Con-
gress to meet once the Constitution was ratified created a Post Office and De-
partments of War, Navy, Foreign Affairs, and Treasury, each in unique ways suit-
ed to its responsibilities; this new government employed few civil servants to
manage all its affairs. The first serious count of federal civilian employees, In
1816, reported that they numbered 4,837.3

While the Constitution has not changed, Congress has repeatedly created new
Departments and new administrative agencies to meet problems arising as the na-
tion and its economy matured. Its reactions to steamboat boiler explosions and
fires on navigable American waters, with their high cost in lost lives and property,
early illustrated its resourcefulness. An Act of 1838 created a licensing scheme in
the Department of the Treasury, requiring various safety measures and providing
for twice-a-year inspections by engineers appointed by U.S. district court judges.
When this proved inadequate, Congress in 1852 created a Steamboat Inspection
Service (SIS) headed by nine presidentially appointed regional inspectors em-
powered to oversee local inspectors the Secretary of the Treasury could discipline
and to adopt implementing regulations. To refine this administrative structure, ein
1871 law created a central office and emphatically reframed SIS authority to adopt
governing regulations. Measures around the turn of the century placed all service
employees except those presidentially appointed into the Civil Service, moved
the SIS from the Treasury into the new Commerce and Labor Department, Und
again heightened its regulatory authority. The result, wrote leading legal scholar
Jerry Mashaw, was to combine “something of the ‘New Deal’ independent, reg-
ulatory commission and ‘Great Society’ health and safety regulation by delegat-
ing administrative authority to a multimember Board that combined licensing,
rulemaking, and adjudicatory functions.”4

As community-based artisans were replaced by factories and new forms of
transportation and communication created a national economy, Congress re-
peatedly expanded federal administration, establishing government bodies to
respond to such risks as discriminatory railroad freight charges, railroad equip-
ment causing workplace carnage on the Civil War’s scale, impure foods that sup-
plied national markets, unethical behaviors by large manufacturers and distant
suppliers affecting those markets, and actions presenting unacceptable risks to
the national economy. The states created public utility commissions, often sep-
arate from the elected executive, to control the behaviors of natural monopolies
like electric utilities, telephone companies, oder (in the countryside) railroad lines.
Congress sometimes placed the regulatory bodies it created in conventional Cab-
inet Departments; but increasingly it created multimember bodies–the Inter-
state Commerce Commission and the Federal Reserve Board, for example–that

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u
D
A
e
D
A
R
T
ich
C
e
–
P
D

F
/

1
5
0
3
1
7
2
0
6
0
4
7
2
D
A
e
D
_
A
_
0
1
8
5
7
P
D

B
j
G
u
e
S
T

Ö
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

Dädalus, das Journal der American Academy of Arts & SciencesHow the Administrative State Got to This Challenging Place

it placed outside the conventional executive government structure dominated by
the President and Cabinet Secretaries.

At the beginning of the twentieth century, “administrative law” emerged as
a distinct public law discipline in response to these societal changes. The federal
Constitution presumes the existence of a government, yet it defines the powers
and responsibilities of only the three institutions at its head: Congress, the Presi-
dent, and the Supreme Court. This was deliberate. The draft sent to the committee
concerned with Article II in mid-August of 1787 proposed summarily to define a
handful of particular Departments and their responsibilities, and to create a coun-
cil modeled on parliamentary lines, while explicitly reserving to the President the
right of decision after receiving its advice.5 The draft of Article II returned to the
Constitutional Convention, and adopted by it, rejected this approach. It empow-
ered Congress to create all executive institutions below the President as well as
any federal courts below the Supreme Court.

Anticipating those creations, the Constitution’s spare text refers both to De-
partments and to their heads, and requires the Senate’s consent to presidential
appointment of the latter. It vests all executive power in a single elected President,
charged with seeing that Congress’s laws would “be faithfully executed.” Yet in
defining the President’s power in relation to the domestic government Congress
was to create, and in contrast to the draft it rejected, the Constitution does not
provide that the actions that government takes are to be the President’s; it says
only that he may “require the Opinion, in writing, of the principal officer in each
of the executive Departments, upon any subject relating to the duties of their re-
spective Offices.” Like the “faithful execution” clause, this language accepts that
actual administrative duties will be placed in others than the President himself.
Just what Departments there would be and how they would be organized–and
in what relationship to the President, Congress, and the courts–was unstated.
Our government is, in effect, the hole in our Constitution, a hole Congress has
been filling with a remarkable variety of public and quasi-public institutions, Pos-
sessing varying powers and responsibilities and in varying relationships with the
President, Congress, and our courts, ever since.

Studying the institutions that the Constitution defines, Dann, could no lon-
ger suffice. Administrative law emerged as the discipline concerned with the ac-
tions of these manifold institutions. Congress, vested with legislative power,
quickly understood that it was incapable of foreseeing the hazards the changes
were bringing or providing for their control with the necessary speed and detail.
Courts, looking at past events through spectacles fashioned by the prior gener-
ations’ perspectives, were poorly equipped to meet contemporary social needs.
If the President ever had been capable of exercising personal control over all im-
portant government actions, that time quickly passed, and it early came to be un-
verstand (as the “Opinion, in writing” and “faithful execution” clauses entail)

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u
D
A
e
D
A
R
T
ich
C
e
–
P
D

F
/

1
5
0
3
1
7
2
0
6
0
4
7
2
D
A
e
D
_
A
_
0
1
8
5
7
P
D

B
j
G
u
e
S
T

Ö
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

150 (3) Summer 2021Peter L. Strauss

that governmental duties were the direct responsibility of the institutions Con-
gress had created to perform them.6 In 1920, following the creation of the Federal
Reserve and the Federal Trade Commission earlier in the twentieth century, neun
Cabinet Departments (many housing within themselves discrete administrative
bodies like the Agriculture Department’s Forest Service) and at least two dozen
distinct federal governmental bodies with regulatory responsibilities employed
um 691,000 civil servants–now organized into a permanent Civil Service cho-
sen for merit, not political connection–under the direction of a much smaller
number of politically appointed officials.

The Great Depression of the 1930s brought in its wake the New Deal, reflect-
ing new ambitions and activities, and greatly enlarging the national government.
One consequence was the creation of the Executive Office of the President, quite
small initially, to advise the President in his relations with the expanding network
of government Departments and agencies. Ein anderer, spurred by the organized
bar’s pressure for more formal administrative procedures, was a remarkable em-
pirical study of the procedures the federal government’s many administrative
agencies actually followed. This study informed the drive for greater uniformity,
Transparenz, and control of agency actions that led, at the end of World War II,
to the unopposed congressional enactment of the federal Administrative Proce-
dure Act (APA) to govern the most formal elements of administrative action. Das
happened at a time when these actions were generally considered to be objective
means of applying expertise to social issues, apolitical in their fundamental na-
tur. The APA has since endured without significant amendment of its most cen-
tral elements, but today, as the possibilities of apolitical expertise have come into
question, its processes and their subjects have become highly politicized. The ex-
tent of national regulation is being hotly contested, the APA’s procedures have
been brought back to Congress’s attention (albeit without, to date, significant leg-
islation actually to change them), and the Trump administration took dramatic
steps to politicize administrative processes.

When the APA was enacted, the principal focus of federal regulation was on
high-consequence government actions involving regulation of individual actors,
often economic in nature: Zum Beispiel, setting railroad rates, or choosing the
routes an airline would be permitted to fly. These actions had long been taken af-
ter trial-like administrative procedures of considerable formality that judicial de-
cisions essentially treated as a constitutional necessity (on-the-record adjudica-
tion, in the APA’s terms, including a formal process for rate-making that, although
denominated “rulemaking,” strongly resembles what it requires of formal adju-
dication). Much of the political momentum the New Deal changes generated to
define federal administrative procedures focused on these high-consequence de-
cisions, which would directly affect the economic well-being of a particular rail-
road, airline, or telephone carrier. For almost two decades after the APA’s adop-

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u
D
A
e
D
A
R
T
ich
C
e
–
P
D

F
/

1
5
0
3
1
7
2
0
6
0
4
7
2
D
A
e
D
_
A
_
0
1
8
5
7
P
D

B
j
G
u
e
S
T

Ö
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

Dädalus, das Journal der American Academy of Arts & SciencesHow the Administrative State Got to This Challenging Place

tion, economic regulation associated with trial-like procedures was the central fo-
cus of its use.

Yet the APA also provided less formal “notice-and-comment” public proce-
dures to govern agency adoption of regulations having a more general impact than
would a single decision about a particular license, rate, or route. Such rules are, In
Wirkung, secondary legislation. If valid, they have the force of statutes, yet they are
adopted by executive agencies, not by Congress. Rulemaking within the frame-
work of enabling statutes had long been judicially tolerated, as long as those stat-
utes provided a framework of intelligible standards that permitted courts to as-
sess their legality. (Early in the twentieth century, Zum Beispiel, the Supreme Court
had upheld a statute authorizing the Secretary of Agriculture to adopt regulations
to secure the objectives of the national forest lands under his administration, Und
permitting criminal enforcement of one of those regulations, which the Secre-
tary had adopted to control the grazing of sheep there.) For a quarter-century,
rulemaking was little studied by either students or scholars of administrative law.
The late 1960s and 1970s brought profound changes. New statutes discarded
or dramatically restructured much economic regulation and closed the agencies
responsible for it (Zum Beispiel, the Interstate Commerce Commission, the Civil
Aeronautics Board, the Federal Power Commission, and elements of the Feder-
al Communications Commission), as economists persuaded Congress that such
regulation inappropriately constrained the operation of economic markets and
the entry of new competitors into them. Increasing concerns about the transpar-
ency of government records, in the wake of McCarthyism and developing civil
rights struggles, produced the Freedom of Information Act (FOIA) and then the
Privacy Act that would bloom beyond all expectations; they contributed as well
(along with significant concerns about the administration of welfare programs) Zu
focused attention on the procedural rights of individuals caught up in both crim-
inal and administrative disputes with the government. Now courts were persuad-
ed that citizen-government relationships potentially involved entitlements, nicht
merely beneficiary-benefactor relations; this “due process explosion” dramati-
cally expanded both the caseloads of agencies dealing with individual relation-
ships with government and the formality of the decision processes those agencies
employed. In the wake of these developments came dramatic growth in the public
provision and subvention of legal services.

At the same time, courts found in the importance of interests that statutes
called on government to protect–such as aesthetic, recreational, or similar bene-
ficiary interests–sufficient reason to permit judicial challenges to administrative
decisions affecting them by anyone suffering their concrete impairment. Diese
findings considerably expanded the set of persons having standing to challenge
government actions. Combined with the possibility of challenging government
regulations immediately upon their adoption, before their enforcement, it was

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u
D
A
e
D
A
R
T
ich
C
e
–
P
D

F
/

1
5
0
3
1
7
2
0
6
0
4
7
2
D
A
e
D
_
A
_
0
1
8
5
7
P
D

B
j
G
u
e
S
T

Ö
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

150 (3) Summer 2021Peter L. Strauss

now possible for citizens or non-governmental organizations (NGOs) vertreten-
ing them to challenge regulations for having done too little, not too much, to pro-
tect the interests Congress had made an agency responsible to regulate. Regula-
tors thought to have been tamed (“captured”) by the “daily machine-gun-like
impact” of their interactions with the regulated now had to be concerned, sowie,
with the possibility of challenge from others.7 The Audubon Society and the Sier-
ra Club first appeared as litigants in federal court in 1969; by mid-June 2020, Die
number of their appearances stands at 2,335, having steadily increased decade af-
ter decade.8

Perhaps the most dramatic changes resulted from new public concerns about
Gesundheit, safety, and the environment, leading both to the enlargement of some ex-
isting regulatory authorities, such as the Food and Drug Administration, und zu
the creation of new ones, including the National Highway Traffic Safety Admin-
istration, the Occupational Safety and Health Administration, and the Environ-
mental Protection Agency. Rulemaking was often the most influential procedure
these agencies employed, and they used it in ways profoundly affecting whole in-
dustries (Und, through them, the national economy). All automobile manufactur-
ers would now have to equip their vehicles in prescribed ways; all factories us-
ing benzene would have to control their workers’ exposure to it; all coal-burning
electric utilities would have to reduce the pollutants their smokestacks emitted.
These high-impact rulemakings and their associated rulemaking procedures rap-
idly drew the attention of scholars, the courts, and “public interest” litigators as-
serting that agencies had failed adequately to protect the interests that statutes
made them responsible to secure.

Although the courts eventually discredited efforts to convert the procedures
used in these important rulemakings into a species of trial process (on the judicial
Modell), they nonetheless interpreted the APA’s sparse language about rulemaking
in ways that substantially embroidered its transparency and its demands. Perhaps
building on FOIA’s clear commands, the courts now required agencies to expose
scientific reports and similar data as elements of the statutory comment process.
Although the APA’s language permits notice for comment of merely “a descrip-
tion of the subjects and issues involved,” courts required a new round of commen-
tary for regulations that were not a “logical outgrowth” of the proposal made. And
although Congress in 1946 would likely have expected judicial review of rulemak-
ing to be like the light-fingered touch its statutes ordinarily received, now courts
undertook to assure themselves that the agencies had taken “hard looks” at the
issues they resolved: addressing significant comments filed by interested per-
sons, demonstrating sound reasoning, and revealing a reasonable connection to
the materials available to them. Richard Stewart, in influential scholarship, aptly
characterized these developments as requiring a “paper hearing” comparable to
legislative hearings, and as appropriately recognizing the differing claims on ju-

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u
D
A
e
D
A
R
T
ich
C
e
–
P
D

F
/

1
5
0
3
1
7
2
0
6
0
4
7
2
D
A
e
D
_
A
_
0
1
8
5
7
P
D

B
j
G
u
e
S
T

Ö
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

Dädalus, das Journal der American Academy of Arts & SciencesHow the Administrative State Got to This Challenging Place

dicial respect owing to legislative action and administrative action.9 By the 1980s,
these developments had all become firmly established in the legal framework.
Few voices were to be heard challenging their appropriateness.

A s early as the Nixon administration, the model of administrative bodies

as objective, essentially apolitical actors came into intellectual question,
as neoclassical economic views and associated political science “public
choice” theories took hold. Administrative agencies–and consequently their
processes–have become considerably more political, and formalism and origi-
nalism have become more characteristic of judicial approaches to the issues of ad-
ministrative law. Before dealing with these changes, Jedoch, which considerably
predate the Trump administration, it is useful to give brief attention to another,
whose consequences for the administrative state and regulation are only begin-
ning to be felt: the transition from the paper to the digital age.

When agency adjudications and rulemakings had only paper records, partic-
ular items were discrete and existed in limited copies. Filing cabinets were phys-
isch, and their searchability depended on their organization and, vielleicht, index-
ing. Parties to an adjudication would be entitled to receive copies of each doc-
ument filed, and that filing would occur in a ritual order generally providing an
opportunity for response. Notice-and-comment rulemakings, andererseits,
lacked discrete parties; all interested were entitled to comment. Comments were
filed only with the proposing agency, and all comments–in support or in opposi-
tion–could be filed at the one deadline the agency had set to receive them. Dort
was no provision for seeing others’ comments or responding to them. Obwohl
FOIA permitted anyone to ask to see all filed comments, this right was indepen-
dent of the rulemaking itself, and hardly practical for any proposal inviting wide
participation. Save how an agency might choose to engage with the outside world
while processing comments–a process itself constrained by the paper record–
the agency essentially had a monopoly on the information that had come to it. To
the extent information is power, the agency was where the power was.

The transformation of government records from paper to digital formats has
worked extraordinary changes. FOIA searches have been complicated by the new
phenomenon of email chains combining many documents in one stream, aber die
capabilities of electronic search have also greatly eased them. Much more impor-
tant, now that desired words, concepts, or references can be found almost instan-
taneously where they occur, searching government records generally has been
transformed. As statutes now command, agencies have placed data and docu-
ments online in public electronic agency libraries–a veritable explosion in the
transparency of governmental work and work-product. Regulations.gov, a unified
site for notice-and-comment agency rulemaking, has simplified public participa-
tion, and now anyone interested can review filed comments and respond to them.

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u
D
A
e
D
A
R
T
ich
C
e
–
P
D

F
/

1
5
0
3
1
7
2
0
6
0
4
7
2
D
A
e
D
_
A
_
0
1
8
5
7
P
D

B
j
G
u
e
S
T

Ö
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

150 (3) Summer 2021Peter L. Strauss

One consequence may be a certain loss of effective agency power in relation to the
White House; since what is in the government “cloud” can be as easily viewed in
the EOP as in the agency itself, agencies have lost any informational advantage the
paper age had given them.

Governmental sharing of data sets and research results has fostered new possi-
bilities for public-private actions: use of its geologic data permitted a private NGO
to demonstrate the possible impacts of rising sea levels; a public database report-
ing toxic substance discharges, searchable by ZIP code, has encouraged discharge
reductions that regulations do not yet require; and agency safety ratings influence
consumer and manufacturer behaviors alike. If sensors embodied in waste dis-
charge outlets or complex machinery provide signals to agencies as well as to their
makers, agencies may be able to use artificial intelligence (AI) to identify more
rapidly any issues warranting their response. The filing now of required reports in
electronic form would also permit the automated creation of data sets. In der Tat, Die
possibilities of artificial intelligence for learning from data–whether rulemaking
comments or data collected from inspections, filed electronic reports, or other
available data sets–have only begun to be explored. Although these possibilities
are indeed exciting, one must remain aware that AI and algorithms are only as re-
liable as the humans monitoring and creating them.

On now to the issues of increasing political control and the associated dis-
placement of the view that administrative action is justified by its objective exper-
tise. The displacement was first evident in contexts of straightforward econom-
ic regulation. Bodies like the Interstate Commerce Commission (ICC) and Civil
Aeronautics Board (CAB) came to be seen as having been captured by the very en-
tities they were supposed to control, acting inefficiently in contexts where market
competition would produce efficient results.10 Pointing out mismatches in regu-
lation failing adequately to account for the possible impact of market operations
on corporate behaviors, then Harvard Professor, and now Supreme Court Justice,
Stephen Breyer’s Regulation and Its Reform11 underlay Congress’s choice to end the
CAB and then the ICC, and to alter the responsibilities of other bodies, such as the
Federal Maritime Commission, substantially. The consequence was significantly
diminished economic regulation. Hier, in eliminating “captured” regulators and
empowering competitive markets, the impact of defeating the “expert agency”
model was simple deregulation.

But in the realms of health, safety, and environmental protection, regulation
depended on science–that is, on expertise. Competition had not produced safer
cars, cleaner water or air, or workplace safety. Although the development of in-
formation regimes, marketable permits for pollutants, and the like might eventu-
ally provide the means of lessening direct regulatory commands–and regulators
would learn the virtues of framing standards to be met rather than issuing com-
mands defining precisely what must be done–none of these techniques would

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u
D
A
e
D
A
R
T
ich
C
e
–
P
D

F
/

1
5
0
3
1
7
2
0
6
0
4
7
2
D
A
e
D
_
A
_
0
1
8
5
7
P
D

B
j
G
u
e
S
T

Ö
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

Dädalus, das Journal der American Academy of Arts & SciencesHow the Administrative State Got to This Challenging Place

work well to provide accurate information, monitor the use of permits, or define
the standards to be achieved, in the absence of a regulatory apparatus. Despite
the occasional termination of agency mandates, Dann, administrative government
continued to grow, and the political opposition to regulatory measures denigrated
the possibility of objective science and promoted political controls.

In a brilliant 2008 Artikel, then Professor and current Judge David J. Barron
called attention to complementary trends that, since the administration of Pres-
ident Nixon, had steadily promoted the political control of ostensibly science-
based regulation: its centralization in the White House and the thickening of the
political layer within the agencies themselves.12

Centralization first, the phenomenon that has attracted the bulk of scholarly
attention in recent years. The Executive Office of the President, the White House
collective providing the President with his best means for understanding and in-
fluencing administrative action, has grown from the six advisors President Frank-
lin D. Roosevelt chose at its creation to more than 1,800 Menschen; Heute, as in the
Obama administration, it includes “czars” the President alone selects and charges
with overseeing choices that Congress has assigned to Senate-confirmed agency
heads. For internal agency political appointments, sowie, loyalty to White House
policy preferences has become the dominating consideration. One expression of
this decades-long development can be seen in President Trump’s apparent prefer-
ence to have “acting” officials responsible for administration, rather than appoin-
tees subject to the potentially conflicting loyalties that can come from the process
of Senate confirmation; reportedly, he had empowered a young White House as-
sistant simply to instruct agency heads whom to appoint to subordinate political
posts Congress authorized them to appoint, as constitutionally it may.13

Rulemaking’s emergence as an activity having major impacts on the national
economy has prompted steady growth in White House initiatives to gain control
over its outcomes. These initiatives first appeared under the rubric of presiden-
tial oversight and coordination, drawing directly on the President’s constitution-
al power to “require the Opinion, in writing, of the principal officer in each of the
executive Departments, upon any subject relating to the duties of their respective
Offices.” They moved inexorably from White House supervision and advice to
White House control. This development of White House direct engagement, Sei-
ginning with President Carter’s Executive Order 12044, was well captured in the
introduction to a 2017 Brookings Institution’s analysis, Evaluating the Trump Ad-
ministration’s Regulatory Reform Program:

The regulatory process has been the rare policy area in which presidents from the two
major parties have broadly agreed, building on each other’s efforts over the course of
decades:

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u
D
A
e
D
A
R
T
ich
C
e
–
P
D

F
/

1
5
0
3
1
7
2
0
6
0
4
7
2
D
A
e
D
_
A
_
0
1
8
5
7
P
D

B
j
G
u
e
S
T

Ö
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

150 (3) Summer 2021Peter L. Strauss

• President Carter formally launched White House oversight of major regu-
Beziehungen (those with an estimated annual economic impact of at least $100
Million) issued by executive branch agencies with Executive Order 12044,
which mandated that agencies conduct regulatory analyses before issuing
major rules, including a consideration of their economic consequences, Aber
did not require balancing costs against estimated benefits.

• President Reagan replaced Carter’s order with Executive Order 12291, welche
was the first to require that agencies explicitly balance estimated benefits
of major regulations against their costs, assuming their underlying statutes
permit it, stating that “regulatory action shall not be undertaken unless the
potential benefits to society for the regulation outweigh the potential costs
to society.”

• President Clinton replaced that order with Executive Order 12866, welche
shifted from the requirement that benefits “outweigh” costs to the require-
ment that benefits “justify” costs, stating that “each agency shall assess both
the costs and the benefits of the intended regulation and . . . propose or adopt
a regulation only upon a reasoned determination that the benefits of the in-
tended regulation justify the costs.”

• President George W. Bush lightly amended E.O. 12866 through Executive
Order 13422 (later revoked by President Obama), extending the White
House oversight requirements to guidance documents issued by executive
branch agencies.

• President Obama’s Executive Order 13563 reaffirmed the principles estab-
lished in E.O. 12866, including that agencies should propose or adopt a reg-
ulation only if “benefits justify its costs.”14

President Trump’s executive orders on rulemaking, and insistence on speedy
deregulation, strongly asserted presidential prerogatives of control. Consistent
with his project to lift the heavy hand of government off industry’s back, diese
executive orders stressed the elimination of existing regulations. They forbade
agencies to issue new regulations without, in effect, White House permission,
permission conditioned on a showing that the totality of costs the agency’s rules
imposed on the regulated would not then exceed a figure annually set by the Of-
fice of Management and Budget (the largest element of the EOP). What future
benefits the rules might confer–or, übrigens, what benefits rescinded rules
would have provided–were irrelevant. Perhaps unsurprisingly, the overwhelm-
ing majority of purported rescissions were found unlawful by courts in which they
had been challenged, often for the haste of their adoption and for failures of rea-
soning. Examples include the Supreme Court’s rejection of a citizenship question

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u
D
A
e
D
A
R
T
ich
C
e
–
P
D

F
/

1
5
0
3
1
7
2
0
6
0
4
7
2
D
A
e
D
_
A
_
0
1
8
5
7
P
D

B
j
G
u
e
S
T

Ö
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

Dädalus, das Journal der American Academy of Arts & SciencesHow the Administrative State Got to This Challenging Place

in the census, and of the attempted recission of President Obama’s program of de-
ferred action on “dreamers.” From the writer’s perspective, the more important
observation is that Congress has placed these rulemaking responsibilities in the
agencies, not the President, and that the steadily tightening presidential grip on
these judgments (especially taken together with the increasingly partisan road-
blocks in Congress) takes us back to George III, not to Philadelphia.

Politicization, Dann. The thickness of the political layer inside agencies has

grown as well. Political scientist B. Guy Peters recently observed that,

A president in the United States can appoint approximately four thousand people to
office, and four or even five echelons of political appointees may stand between a ca-
reer civil servant and the cabinet secretary. In the United Kingdom each ministry will
only have a few political appointments other than the minister or secretary of state
in charge–the largest number now is the Treasury with six appointments–but even
Dann, the major interface between political and administrative leaders occurs between
the minister and a single career civil servant, the permanent secretary.15

While political layering is rising in UK agencies too, a particularly dramatic
American shift occurred during the Carter administration, when Civil Service re-
forms moved essentially all civil servants with policy responsibility into a Senior
Executive Service (SES), subject to much greater levels of control by the agen-
cy’s political leadership than the Civil Service had permitted. The Trump admin-
istration’s Secretary of the Interior Ryan Zinke reassigned many in his Depart-
ment’s SES staff to jobs unsuited to their abilities. Presidents long regarded the
departmental and agency Inspectors General that Congress created in the same
Civil Service reform statute as desirably nonpartisan, apolitical internal monitors
of agency action, and permitted their service to span changing administrations.
For President Trump, Jedoch, the signs of “disloyalty” suggested by inquiries
into the actions of agency political leadership repeatedly became an occasion for
dismissal.

Yet if the President’s “taking control over the national administrative process
. . . gets things done [Und] brings coherence where none existed before,” Profes-
sor Barron asks, “then what of social learning? What of alternative regulatory ap-
proaches? What then of the long view?” He continues:

The concern reflected in such questions . . . lies at the heart of what makes increased
centralization and politicization so potentially troubling. These developments . . . have
made the federal agencies increasingly ill-suited to perform their customary role of
providing a mechanism for social learning. . . . [A] powerful institutional logic has
increasingly made the federal bureaucracy a fully committed member of the White
House regime. . . . [W]e should . . . be looking for ways to ensure that alternative voices
are brought into the mix nonetheless.16

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u
D
A
e
D
A
R
T
ich
C
e
–
P
D

F
/

1
5
0
3
1
7
2
0
6
0
4
7
2
D
A
e
D
_
A
_
0
1
8
5
7
P
D

B
j
G
u
e
S
T

Ö
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

150 (3) Summer 2021Peter L. Strauss

T urn now briefly to the courts and to the remarkable range of debates in

and about them currently roiling the world of administrative law. Wann
the APA was adopted, law school instruction about administrative law
was largely concerned with the use of courts to control administrative process-
es, not political controls; courts, like agencies, were generally viewed as a collec-
tion of experts trained to act on the basis of objective and apolitical factors (“the
law” and “justice”). The emergence of legal realism in the academies and prom-
inent Supreme Court actions with high political valence (President Roosevelt’s
Court-packing plan, defeated by the New Deal’s “switch in time”–in itself, eins
might think, a commitment to that apolitical view–and the civil rights decisions
of the 1950s) may have contributed to an erosion of that view. Yet the academ-
ic framework of administrative law instruction was captured in the title of Louis
Jaffe’s magisterial work Judicial Control of Administrative Action.17

The emergence of rulemaking brought the politics of administrative action to
the forefront and contributed (alongside reactions to the liberalization of crimi-
nal procedures, civil rights litigation, and the abortion decisions) to the steadily in-
creasing politicization of the judicial appointments process. The Senate’s increas-
ingly partisan behaviors resulted in the abandonment of safeguards that had long
controlled presidential ambitions to project their administration’s influence far
into the future: respect for the inputs of Senators from states where vacancies had
occurred and for the views of the organized bar, and the effective need to secure a su-
permajority in the face of opposition to any given appointment. During the Trump
administration, Senate Majority Leader Mitch McConnell consistently gave the
highest priority to confirming the President’s nominations to the federal courts.18
Given the relative youth of the appointments made, the views of those judges may
influence the outcomes of judicial decision-making for decades to come.

Perhaps not coincidentally, the legal framework for administrative law devel-
oped over the past decades has come into sharp question. Increasingly, courts are
reasoning with formality, relying on dictionaries to determine the “plain mean-
ing” of statutory terms, not attention to the political history of legislation, Und
generally favoring the original understandings of statutes and the Constitution.
Serious questions are now being voiced about the lawfulness of Congress’s autho-
rizations of agency rulemaking and agency adjudication. Rulemaking authority
is characterized as a delegation of the “legislative power” only Congress can con-
stitutionally enjoy, not the authorization of executive actions of a character to be
found in every developed legal system. Agency adjudication is challenged as the
exercise of the “judicial power” the Constitution reserves to federal courts, nicht
executive action subject to judicial review. Long-established doctrine calling on
the courts to respect agency policy choices made within the scope of the authority
their statutes imperfectly define is being replaced by judicial decision about the
meaning of statutes for whose administration they are not responsible, and with

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u
D
A
e
D
A
R
T
ich
C
e
–
P
D

F
/

1
5
0
3
1
7
2
0
6
0
4
7
2
D
A
e
D
_
A
_
0
1
8
5
7
P
D

B
j
G
u
e
S
T

Ö
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

Dädalus, das Journal der American Academy of Arts & SciencesHow the Administrative State Got to This Challenging Place

whose complexity they are not familiar. The proposition that statutes can only
mean what their words could have been understood to mean at the time of their
enactment threatens the universally accepted “paper hearing” courts articulated
in response to the emergence of rulemaking’s significance decades after the APA’s
enactment. The titles and substance of two colleagues’ recent publications may
suggest the tension: Professor Gillian Metzger’s Harvard Law Review Foreword,
“1930s Redux: The Administrative State Under Siege,” and Professor Philip Ham-
burger’s book, Is Administrative Law Unlawful?19

These challenges have long underlain the world of American administrative
law and the realities with which it deals, and they can be expected to endure. In
recent times, a firestorm of other challenges has arisen that underscores both the
necessity of a functioning government capable of dealing with the perils of the
natural world, the economy, and human behaviors, and the political difficulties
of achieving these ends in our constitutional republic. Partisanship has rendered
Congress the “Broken Branch.”20 A rise of renewed populism, threatening de-
mocracies across the world,21 brought America the presidency of Donald Trump,
with its repeated seeming indifference to the rule of law and “unprecedented, sein-
toric corruption.”22 The President’s indifference also to the world of science, ev-
ident enough in his administration’s repeated rescissions of environmental stan-
dards and its refusals to take seriously the prospects created by climate change,
propelled the United States to the forefront of nations suffering from the scourge
of COVID-19, with its extraordinary challenges both to science and to an economy
it has brought to its knees. Gleichzeitig, the police killing of George Floyd in
Minneapolis has generated an understanding of institutional racism–of the fra-
gility that obscures from Whites the ways in which their economic place and their
perceptions have been built on a history of successful oppression of others–that
may transform the ways in which the landmarks of American administrative law
are understood.23

Jacques Lipschutz’s monumental “Bellerophon Taming Pegasus” towers four

stories high over the portico of Columbia Law School, whence come these
words written in my fiftieth year there. Symbolically, it represents reason tam-
ing unreason: In der Tat, because Bellerophon’s head in the sculpture merges with
the wild horse’s body, it is man taming his own unreason. What a powerful meta-
phor for the work of law and perhaps, insbesondere, for the work of public law! Der
growing imbalance between reason and unreason in American administrative law
is the occasion for deep concern, and a major challenge for our collective future.

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u
D
A
e
D
A
R
T
ich
C
e
–
P
D

F
/

1
5
0
3
1
7
2
0
6
0
4
7
2
D
A
e
D
_
A
_
0
1
8
5
7
P
D

B
j
G
u
e
S
T

Ö
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

150 (3) Summer 2021Peter L. Strauss

author’s note

Like the other essays in this collection, work on this essay was essentially complete
before President Biden’s inauguration, so that the impacts of that transition, actu-
al and potential, are not considered. Ahmed Mabruk provided invaluable research
assistance.

about the author

Peter L. Strauss, a Fellow of the American Academy since 2010, is the Betts Pro-
fessor Emeritus of Law at Columbia Law School. Seit 1980, he has been Lead Ed-
itor of Gellhorn and Byse’s Administrative Law: Cases and Comments (Foundation Press,
2018). He is the author of Administrative Justice in the United States (Carolina Academic
Drücken Sie, 3rd ed., 2016), and Legal Methods: Understanding and Using Cases and Statutes
(Foundation Press, 3rd ed., 2014), among many other publications.

Endnoten

1 In 2019, the number was 2.1 million if one counted the individuals it employed and about
fifteen percent higher if one calculated “full time equivalents.” Julie Jennings and Jared
C. Nagel, “Federal Workforce Statistics Sources: OPM and OMB,” R43590 (Washing-
Tonne, D.C.: Forschungsdienst des Kongresses, 2019).

2 Jennifer Selin and David Lewis, Sourcebook of United States Executive Agencies, 2nd ed. (Wash-
ington, D.C.: Administrative Conference of the United States, 2018), 12, provide num-
bers of federal agencies ranging from 118 Zu 600, depending on definitions.

3 See Deanna Malatesta, “Evolution of the Federal Bureaucracy,” in A History of the U.S.
Political System: Ideas, Interessen, and Institutions, Bd. 1, Hrsg. Richard A. Harris and Daniel J.
Tichenor (Santa Barbara, Calif.: ABC-CLIO, 2010), 373, 380.

4 Jerry L. Mashaw, “Administration and ‘The Democracy’: Administrative Law from Jack-
son to Lincoln, 1829–1861,” The Yale Law Journal 117 (8) (2008); a more limited view of
the SIS’s functions is expressed in Joseph Postell, Bureaucracy in America: The Administra-
tive State’s Challenge to Constitutional Government (Columbia: University of Missouri Press,
2017), 96–102.

5 “The President of the United States shall have a privy council, which shall consist of the
president of the Senate, the speaker of the House of Representatives, the chief justice
of the Supreme Court, and the principal officer in the respective departments of for-
eign affairs, domestic affairs, Krieg, marine, and finance, as such departments of office
shall from time to time be established; whose duty it shall be to advise him in matters,
respecting the execution of his office, which he shall think proper to lay before them;
but their advice shall not conclude him, nor affect his responsibility for the measures
which he shall adopt.” Morris draft of August 20, found in Thomas H. Calvert, The Fed-
eral Statutes Annotated: Containing All the Laws of the United States of a Contained and Permanent
Nature in Force on the First Day of January, 1903, Bd. 8 (New York: Edward Thompson Com-
pany, 1905), 200–202.

6 “If the laws, Dann, require a particular officer by name to perform a duty, not only is that
officer bound to perform it, but no other officer can perform it without a violation of

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u
D
A
e
D
A
R
T
ich
C
e
–
P
D

F
/

1
5
0
3
1
7
2
0
6
0
4
7
2
D
A
e
D
_
A
_
0
1
8
5
7
P
D

B
j
G
u
e
S
T

Ö
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

Dädalus, das Journal der American Academy of Arts & SciencesHow the Administrative State Got to This Challenging Place

the law; and were the President to perform it, he would not only be not taking care that
the laws were faithfully executed, but he would be violating them himself.” William
Wirt, Office of the Attorney General, Opinions of the Attorneys General of the United
Zustände, Oktober 20, 1823, 624, 625. Roger Taney, as President Jackson’s Attorney Gener-
al, gave him the same advice.

7 James Landis, Report on Regulatory Agencies to the President-Elect (Washington, D.C.: UNS.

Government Printing Office, 1960), 69.

8 Lexis search of combined federal cases for “Name(Sierra Club) or Name(National Audu-

bon Society),” conducted June 15, 2020.

9 Richard B. Stewart, “The Development of Administrative and Quasi-Constitutional Law
in Judicial Review of Environmental Decision-Making: Lessons from the Clean Air
Akt,” Iowa Law Review 62 (1977): 713, 731–733.

10 Sehen, Zum Beispiel, Marver H. Bernstein, Regulating Business by Independent Commission (Prince-
Tonne, N.J.: Princeton University Press, 1955); and George Stigler, “The Theory of Eco-
nomic Regulation,” Bell Journal of Economics and Management Science 2 (1) (1971): 3–18.
11 Stephen Breyer, Regulation and Its Reform (Cambridge, Masse.: Harvard University Press,

1982).

12 David J. Barron, “From Takeover to Merger: Reforming Administrative Law in an Age of

Agency Polarization,” The George Washington Law Review 76 (5) (2008): 1095.

13 Only “principal” officers need be named by the President and, for them, Senate confir-

mation is also required.

14 Ted Gayer, Robert Litan, and Phillip Wallach, Evaluating the Trump Administration’s Regula-
tory Reform Program (Washington, D.C.: The Brookings Institution, 2017), https://www
.brookings.edu/wp-content/uploads/2017/10/evaluatingtrumpregreform_gayerlitan
wallach_102017.pdf.

15 B. Guy Peters, The Politics of Bureaucracy: An Introduction to Comparative Public Administration,

7th ed. (New York: Routledge, 2018).

16 Barron, “From Takeover to Merger,” 1151–1152.
17 Louis Jaffe, Judicial Control of Administrative Action (Boston: Little-Brown, 1965).
18 By the middle of 2020, President Trump had made two hundred Article III appoint-
gen. See Russell Wheeler, “Trump’s 200th Judicial Appointment: Less Than Meets
the Eye,” Brookings Institution, Juni 26, 2020, https://www.brookings.edu/blog/
fixgov/2020/06/26/trumps-200th-judicial-appointment-less-than-meets-the-eye/.
19 Gillian Metzger, “1930s Redux: The Administrative State Under Siege,” Harvard Law Re-
view 131 (1) (2017); and Philip Hamburger, Is Administrative Law Unlawful? (Chicago: Uni-
versity of Chicago Press, 2014). As may be evident, his answer is “yes.” For a rejoinder,
see Adrian Vermeule’s book review, "NEIN,” Texas Law Review 93 (2015): 1547.

20 Thomas E. Mann and Norman J. Ornstein, The Broken Branch: How Congress Is Failing Amer-
ica and How to Get It Back on Track (Oxford: Oxford University Press, 2006); Mann and
Ornstein produced two sequels as the problem deepened over the years.

21 Steven Levitsky and Daniel Ziblatt, How Democracies Die (New York: Crown, 2018).

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u
D
A
e
D
A
R
T
ich
C
e
–
P
D

F
/

1
5
0
3
1
7
2
0
6
0
4
7
2
D
A
e
D
_
A
_
0
1
8
5
7
P
D

B
j
G
u
e
S
T

Ö
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

150 (3) Summer 2021Peter L. Strauss

22 This is how Republican Senator Mitt Romney characterized President Trump’s com-
mutation of the prison sentence of Roger Stone, a long-time political ally. Peter Baker,
“President Ignores Limit Honored Even by Nixon," Die New York Times, Juli 12, 2020.
23 Robin DiAngelo, White Fragility: Why It’s So Hard for White People to Talk about Racism (Bos-
Tonne: Beacon Press, 2018). An extended blog symposium on the impact of implicit bias
on administrative law, initiated July 13, 2020, may be found at “Symposium on Racism
in Administrative Law,” Yale Journal on Regulation, https://www.yalejreg.com/topic/
racism-in-administrative-law-symposium/.

D
Ö
w
N
Ö
A
D
e
D

F
R
Ö
M
H

T
T

:
/
/

D
ich
R
e
C
T
.

ich
T
.

e
D
u
D
A
e
D
A
R
T
ich
C
e
–
P
D

F
/

1
5
0
3
1
7
2
0
6
0
4
7
2
D
A
e
D
_
A
_
0
1
8
5
7
P
D

B
j
G
u
e
S
T

Ö
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

Dädalus, das Journal der American Academy of Arts & SciencesHow the Administrative State Got to This Challenging Place
PDF Herunterladen