UNITED STATES OF AMERICA
Federal Trade Commission
WASHINGTON, D.C. 20580
Office of Commissioner
Andrew N. Ferguson
1
Dissenting Statement of Commissioner Andrew N. Ferguson
Joined by Commissioner Melissa Holyoak
In the Matter of the Non-Compete Clause Rule
Matter Number P201200
June 28, 2024
On April 23, 2024, the Commission promulgated the Non-Compete Clause Rule (“Final
Rule”).
1
It bans all employee noncompete agreements—agreements in which an employee agrees
not to work for his or her employer’s competitor after his or her employment. It is by far the
most extraordinary assertion of authority in the Commission’s history. It categorically prohibits a
business practice that has been lawful for centuries. It invalidates thirty million existing
contracts. It redistributes nearly half a trillion dollars of wealth. And it preempts the law of forty-
six States. It does all of this on the basis of a few words in a 110-year-old statute—the Federal
Trade Commission Act (“FTC Act”)
2
—words that the Commission had never used to regulate
noncompete agreements until the day before this rulemaking began.
Whatever the Final Rule’s wisdom as a matter of public policy, it is unlawful. Congress
has not authorized us to issue it. The Constitution forbids it. And it violates the basic
requirements of the Administrative Procedure Act.
I therefore respectfully dissent.
I
A noncompete agreement is exactly what it sounds like. It is an agreement between two
parties limiting the extent to which they will compete with each other. These agreements take
many forms,
3
but we are here concerned with only one type: agreements in which an employee
1
Non-Compete Clause Rule, 89 Fed. Reg. 38,342 (May 7, 2024).
2
15 U.S.C. §§ 45(a), 46(g).
3
For example, noncompete agreements are a “classic ‘ancillary’” restraint accompanying the sale of a business, in
which the seller agrees not to compete with the purchaser of the business in order to ensure that the purchaser can
realize the full value of the business he or she is purchasing. Bus. Elecs. Corp. v. Sharp Elecs. Corp., 485 U.S. 717,
730 n.3 (1988). The Final Rule does not reach these agreements. Final Rule, 89 Fed. Reg. at 38,504.
2
agrees not to compete against his or her employer, including securing employment with a
competitor of his or her employer, after the conclusion of their employment relationship.
4
A
Noncompete agreements are not new. They are much older than the Republic.
5
And they
have been the subject of extensive and complex regulation for centuries. The reason for their
regulation is obvious. The Anglo-American legal tradition protected the right to ply one’s trade.
6
English “law abhor[red] idleness”
7
and viewed noncompete agreements as imposing idleness on
a society that could ill afford it.
8
Medieval English courts therefore uniformly proscribed
agreements that prohibited tradesmen from practicing their trades.
9
Regulatory hostility to noncompete agreements loosened as the economy became more
complex.
10
Then in 1711, the Queen’s Bench decided the seminal case of Mitchel v. Reynolds.
11
The noncompete agreement there forbade a baker who leased his business from competing
against the lessor for the duration of the lease.
12
The Mitchel court explained that such restraints
had traditionally been proscribed because of “the mischief which may arise . . . to the party, by
the loss of his livelihood, and the subsistence of his family,” and “to the publick, by depriving it
4
Final Rule, 89 Fed. Reg. at 38,502–03.
5
See Charles E. Carpenter, Validity of Contracts Not to Compete, 76 U. Pa. L. Rev. 244, 244–45 (1928) (discussing
reported cases addressing species of noncompete agreements dating back to the fourteenth and fifteenth centuries);
see also Harlan M. Blake, Employee Agreements Not to Compete, 73 Harv. L. Rev. 625, 626 (1960) (explaining that
noncompete agreements “comprise one of the traditional common-law ‘restraints of trade’ and present problems
which have kept them before the courts for more than five hundred years”).
6
See 1 William Blackstone, Commentaries *427 (“At common law, every man might use what trade he pleased.”);
Darcy v. Allein, 77 Eng. Rep. 1260, 1263 (K.B. 1603) (“[E]very man’s trade maintains his life, and therefore he
ought not to be deprived or dispossessed of it, no more than of his life.”); Corfield v. Coryell, 6 F. Cas. 546, 551–52
(C.C.E.D. Pa. 1823) (including the practice of one’s chosen trade among the privileges and immunities of “citizens
of all free governments”); Golden Glow Tanning Salon, Inc. v. City of Columbus, 52 F.4th 974, 982–84 (5th Cir.
2022) (Ho, J., concurring) (expounding history of Anglo-American protections of the right to “pursue one’s
occupation against arbitrary government restraint”).
7
The Case of Tailors of Ipswich, 77 Eng. Rep. 1218, 1219 (K.B. 1615) (“[A]t the common law, no man could be
prohibited from working in any lawful trade, for the law abhors idleness . . . .”).
8
Carpenter, supra note 5, at 245; see also Catherine L. Fisk, Working Knowledge: Trade Secrets, Restrictive
Covenants in Employment, and the Rise of Corporate Intellectual Property, 1800–1920, 52 Hastings L.J. 441, 455
(2001) (During the Middle Ages, “when the economy was still reeling from the death of half the workforce due to
the Plague, the inability to practice one’s trade was not only disastrous for the individual but a serious loss to the
public as well. Enforced idleness would also run afoul of the Statute of Labourers, which responded to the labor
shortage caused by the Plague by regulating wages and making it a crime for an able-bodied person without
independent means to refuse to work.”).
9
Carpenter, supra note 5, at 244–45.
10
See Alger v. Thacher, 36 Mass. 51, 53 (1837) (describing early seventeenth century cases approving of limited
noncompete agreements after “the most ancient rules of the common law” forbidding such agreements had
“continued unchanged and without exceptions” “[f]or two hundred years.”).
11
24 Eng. Rep. 347 (Q.B. 1711).
12
Id. at 347.
3
of an useful member.”
13
But noncompete agreements also conferred social benefits by, for
example, permitting business owners to sell their businesses profitably (because no one would
buy a business if the seller could immediately compete with the purchaser in the same field).
14
Although the Mitchel court continued to treat “general” restraints—those that applied
indefinitely throughout England—as categorically proscribed, “particular” restraints limited in
geographic scope and applicable only to certain “persons” were permitted.
15
So long as the
particular restraint was “reasonable” in scope, it could be enforced.
16
B
Mitchel’s approach—often described as the earliest application of what we today call the
“rule of reason”
17
—replaced the general medieval proscription of noncompete agreements.
18
“Mitchel established a multifactored analysis of reasonableness that has ever since dominated the
law’s approach to contractual restraints on the practice of a trade . . . .”
19
By the nineteenth
century, English courts upheld noncompete agreements if “the restraint is such only as to afford a
fair protection to the interests of the party in favour of whom it is given, and not so large as to
interfere with the interests of the public.”
20
American state courts similarly adopted a multifactor reasonableness test in the
nineteenth century, and generally enforced noncompete agreements under four conditions. First,
the agreement had to be “ancillary” to some other valid agreement, such as an employment
contract or business-sale agreement.
21
Second, the restraint had to be “no greater than is required
to protect the promisor.”
22
Third, the restraint could “not impose undue hardship on the
promisor.”
23
Finally, the restraint “must not [have] be[en] injurious to the public.”
24
This background is important for understanding the Final Rule. Not only have
noncompete agreements been around, and been regulated, for more than half a millennium, they
13
Id. at 350.
14
Ibid.; see also Nat’l Soc’y of Pro. Eng’rs v. United States, 435 U.S. 679, 689 (1978) (“The long-run benefit of
enhancing the marketability of the business itself—and thereby providing incentives to develop such an enterprise—
outweighed the temporary and limited loss of competition.”).
15
Mitchel, 24 Eng. Rep. at 348–49.
16
See, e.g., id. at 352.
17
Nat’l Soc’y of Pro. Eng’rs, 435 U.S. at 688.
18
See Blake, supra note 5, at 639–40.
19
Fisk, supra note 8, at 456.
20
Horner v. Graves, 131 Eng. Rep. 284, 287 (C.P. 1831).
21
15 Corbin on Contracts § 80.7 (2024).
22
Id. § 80.6.
23
Ibid.
24
Ibid.; see also Restatement (2d) of Contracts § 188 (1981) (proposing essentially the same reasonableness test).
4
have been, until today, the province almost exclusively of state legislative authority.
25
And the
States have vigorously exercised that authority to take a variety of different regulatory
approaches.
26
All fifty States regulate noncompete agreements extensively. The vast majority
retain the common-law reasonableness approach, either codified in statute or elaborated in
judicial decisions.
27
Some States apply the common-law rule to most restraints, but proscribe
them entirely for certain classes of workers.
28
Virginia is a good example of this approach. It has
long applied the common-law reasonableness test to noncompete agreements.
29
But its
legislature recently forbade noncompete agreements for employees who earn less than the
weekly average wage.
30
A handful of States replaced the common-law approach with a
comprehensive statutory scheme governing the scope and duration of noncompete agreements
and the employees subject to them.
31
Only four States have banned noncompete agreements
outright.
32
State regulatory regimes are not static. “States have been experimenting with non-
compete regulation for more than a century, with laws ranging from full bans to notice
requirements, compensation thresholds, bans for specific professions, reasonableness tests, and
more.”
33
Not all of these changes restricted noncompete agreements. Between 2014 and 2020,
state legislatures enacted “[n]ineteen changes [that] reduced [noncompete] enforceability and six
[that] enhanced it ….”
34
State legislatures continue to experiment today, “weigh[ing] the
25
Kenneth G. Dau-Schmidt, Xiaohan Sun & Phillip J. Jones, The American Experience with Employee Noncompete
Clauses: Constraints on Employees Flourish and Do Real Damage in the Land of Economic Liberty, 42 Compar.
Lab. L & Pol’y J. 585, 589 (2022) (“In the United States there is no federal law that is currently used to regulate
employment noncompetes.”).
26
See Gregory v. Ashcroft, 501 U.S. 452, 458 (1991) (“This federalist structure of joint sovereigns . . . assures a
decentralized government that will be more sensitive to the diverse needs of a heterogenous society” and “allows for
more innovation and experimentation in government.”); see also United States v. Lopez, 514 U.S. 549, 581 (1995)
(Kennedy, J., concurring) (“[T]he States may perform their role as laboratories for experimentation to devise various
solutions where the best solution is far from clear.”).
27
Dau-Schmidt, Sun & Jones, supra note 25, at 589–90 & n.26.
28
See Stewart J. Schwab, Report to the Study Committee on Covenants Not to Compete, Uniform Law Commission,
at 2–3 & Appendix Table A-1 (Dec. 13, 2019), available at https://tinyurl.com/yewcux9k.
29
See, e.g., Preferred Sys. Sols., Inc. v. GP Consulting, LLC, 732 S.E.2d 676, 681 (Va. 2012) (“Restraints on trade
are not favored in Virginia; hence, contracts in restraint of trade are enforceable only if ‘narrowly drawn to protect
the employer’s legitimate business interest, . . . not unduly burdensome on the employee’s ability to earn a living,
and . . . not against public policy.’” (quoting Omniplex World Servs. Corp. v. U.S. Investigations Servs., Inc., 618
S.E.2d 340, 342 (Va. 2005)).
30
Va. Code Ann. § 40.1-28.7:8.
31
Dau-Schmidt, Sun & Jones , supra note 25, at 589 & n.25.
32
Final Rule, 89 Fed. Reg. at 38,465.
33
Id. at 38,466.
34
Johnathan M. Barnett & Ted Sichelman, The Case for Noncompetes, 87 U. Chi. L. Rev. 953, 961 (2020).
5
competing policy interests at stake” and altering their regimes sometimes to expand the
enforceability of noncompete agreements, other times to curtail it.
35
Into this established system of sensible state regulation bursts the Final Rule.
C
There is no tradition of federal regulation of noncompete agreements. No federal statute
addresses them directly. The common law treated noncompete agreements as a “contract in
restraint of trade,”
36
however, and the Sherman Act regulates such contracts.
37
But in the 134
years since Congress passed the Sherman Act, there have been only “17 cases . . . in which
private plaintiffs or the federal government have challenged a non-compete clause between an
employer and a worker under either Section 1 or an analogous provision in a state antitrust
statute.”
38
Almost every single suit failed.
39
There has been even less action on noncompete agreements under Section 5 of the FTC
Act. The Commission has enjoyed unprecedented power to define the key term—“unfair
methods of competition”
40
—in our organic statute.
41
Yet, in the first 109 years of our existence,
35
Comment by West Virginia and 17 other States, FTC-2023-0007-20892, at 13–14.
36
Amasa M. Eaton, On Contracts in Restraint of Trade, 4 Harv. L. Rev. 128 (1890).
37
See 15 U.S.C. § 1 (declaring “[e]very contract . . . in restraint of trade or commerce among the several States . . .
to be illegal”).
38
Non-Compete Clause Rule, NPRM, 88 Fed. Reg. 3,482, 3,496 (Jan. 19, 2023) (hereinafter “NPRM”) (citing
cases).
39
Id. The Commission says that two of the challenges were “successful to some degree,” but even that is
overstating. Id. One example of “success” that the Commission provides is an unpublished order denying a motion
to dismiss on the ground that noncompete agreements are within the ambit of Section 1 of the Sherman Act.
Signature MD, Inc. v. MDVIP, Inc., No. 14-5453 DMG, 2015 WL 3988959, at *6–7 (C.D. Cal. Apr. 21, 2015). But
that is not “success.” That is blackletter law and is true of almost every contract. See, e.g., Newburger, Loeb & Co.,
Inc. v. Gross, 563 F.2d 1057, 1082 (2d Cir. 1977) (“[E]mployee agreements not to compete are proper subjects for
scrutiny under section 1 of the Sherman Act.”). The court then held that whether a noncompete agreement violates
Section 1 turns entirely on its particular competitive effects. Signature MD, 2015 WL 3988959, at *6.
The Commission also dramatically overstates the Supreme Court’s holding in United States v. American Tobacco
Co., 221 U.S. 106 (1911). The Commission says that American Tobacco “held that several tobacco companies
violated both section 1 and section 2 of the Sherman Act because of the ‘constantly recurring’ use of non-competes,
among other practices.” Final Rule, 89 Fed. Reg. at 38,343; see also NPRM, 88 Fed. Reg. at 3,496 (American
Tobacco “is the only case the Commission has identified in which a court analyzed the collective, rather than
isolated, use of non-compete clauses.”). But the Supreme Court did not address the lawfulness of noncompete
agreements at all. It held only that the use of “constantly recurring stipulations” requiring manufacturers,
stockholders, or employees not to compete against the tobacco trust was one of many facts that demonstrated that
the trust was an “illegal combination” under Sections 1 and 2. 221 U.S. at 182–83. The Court expressly abjured the
“legality” of those agreements when considered “isolatedly” from the many other acts and practices unrelated to
noncompete agreements in which the trust engaged. Id. at 183. American Tobacco therefore did not address under
Section 1 the question the Commission considers today under Section 5.
40
15 U.S.C. § 45(a).
41
See infra Section III.B.
6
we did not bring a single enforcement action against any noncompete agreement between an
employer and employee.
42
Notwithstanding federal law’s silence on noncompete agreements, the Commission
received an order from on high in 2021. As part of a “whole-of-government competition policy,”
President Biden “encouraged” the Commission to “exercise” its “statutory rulemaking authority
under the [FTC] Act to curtail the unfair use of non-compete clauses and other clauses or
agreements that may unfairly limit worker mobility.”
43
The majority got the message loud and clear. On January 4, 2023, the Commission
announced consent orders against three businesses enjoining them from enforcing noncompete
agreements against their employees.
44
The Commission did not adjudicate the alleged Section 5
violations, nor did it take the claims to federal court. It obtained no money from the alleged
offenders. And the orders contained no admission that the noncompete agreements violated
Section 5. Rather, the respondents agreed not to enforce their noncompete agreements in
exchange for the Commission leaving them alone.
45
These three consent orders embodied the
sum total of our “experience” with noncompete agreements under Section 5 in more than a
century.
The next day—before the settlements had even become final—the Commission issued a
55-page Notice of Proposed Rulemaking to ban all noncompete agreements.
46
All three members
of the majority invoked the previous day’s enforcement actions as a rationale for the proposed
rule.
47
42
We attempted to enforce Section 5 against one noncompete agreement in 1963. See Snap-On Tools Corp. v. FTC,
321 F.2d 825 (7th Cir. 1963). But the enforcement action did not involve an employment noncompete and would not
be covered by the Final Rule. And, inconveniently for the Final Rule, we lost on precisely the argument that the
Final Rule forbids an employer from making—that the particular terms of the noncompete agreement were
reasonable. See id. at 837.
43
Exec. Order No. 14,036 of July 9, 2021: Promoting Competition in the American Economy, 86 Fed. Reg. 36,987,
36,992 (July 14, 2021).
44
See Agreement Containing Consent Order, In the Matter of O-I Glass, Inc., FTC File No. 211-0182 (Jan. 4, 2023);
Agreement Containing Consent Order, In the Matter of Ardagh Group S.A. et al., FTC File No. 211-0182 (Jan. 4,
2023); Agreement Containing Consent Order, In the Matter of Prudential Security, Inc. et al., FTC File No. 221-
0026 (Jan. 4, 2023).
45
Settlements are a great thing, but they shed little light on the state of the law absent a confession of guilt.
46
See NPRM, 88 Fed. Reg. at 6,482. The Federal Register published the NPRM on January 19, 2023, but the
Commission announced it to the public on January 5, 2023. Fed. Trade Comm’n, FTC Proposes Rule to Ban
Noncompete Clauses, Which Hurt Workers and Harm Competition (January 5, 2023), https://www.ftc.gov/news-
events/news/press-releases/2023/01/ftc-proposes-rule-ban-noncompete-clauses-which-hurt-workers-harm-
competition.
47
See NPRM, 88 Fed. Reg. at 3,536–37 (Statement of Chair Lina M. Khan Joined by Comm’r Rebecca Kelly
Slaughter and Comm’r Alvaro M. Bedoya); id. at 3539 n.4 (Statement of Comm’r Rebecca Kelly Slaughter Joined
by Comm’r Alvaro M. Bedoya).
7
On April 23, 2024, the Commission issued the Final Rule. The Final Rule declares that
employee noncompete agreements—which it calls “non-compete clauses”—are unfair methods
of competition in violation of Section 5.
48
It defines a “non-compete clause” as any “term or
condition of employment that prohibits a worker from, penalizes a worker for, or functions to
prevent a worker from (i) [s]eeking or accepting work in the United States with a different
person where such work would begin after the conclusion of the employment” subject to that
agreement; or (ii) “[o]perating a business in the United States after the conclusion of the
employment” subject to such agreement.
49
The prohibition applies both prospectively and
retrospectively with only one exception: Employers may enforce noncompete agreements against
“senior executive[s]”—employees who were “in a policy-making position” and whose total
annual compensation was at least $151,164—so long as those agreements predate the Final
Rule.
50
II
Lawmaking by the administrative state sits uncomfortably in a democracy. Our
Constitution assigns Congress the legislative power because Congress answers to the People for
its choices.
51
We are not a legislature; we are an administrative agency wielding only the power
lawfully conferred on us by Congress.
52
Americans cannot vote us out when we get it wrong.
53
And Congress has tried to insulate us from the one person in the Executive Branch whom the
people can vote out,
54
separating us even further from those whose lives we claim to govern.
55
48
Final Rule, 89 Fed. Reg. 38,502–03.
49
Id. at 38,502.
50
Id. at 35,802–03.
51
U.S. Const. art I, § 1, cl. 1; see The Federalist No. 52, at 325 (J. Madison) (Clinton Rossiter ed., 1961) (The body
wielding the legislative power “should have an immediate dependence on, and frequent sympathy with, the people,”
and “[f]requent elections are unquestionably the only policy by which this dependence and sympathy can be
effectually secured.”); see also The Federalist No. 37, at 223 (James Madison) (“The genius of republican liberty
seems to demand on one side, not only that all power should be derived from the people, but that those intrusted
with it should be kept in dependence on the people . . . .”).
52
FEC v. Cruz, 596 U.S. 289, 301 (2022) (“An agency, after all, ‘literally has no power to act’ . . . unless and until
Congress authorizes it to do so by statute.” (quoting La. Pub. Serv. Comm’n v. FCC, 476 U.S. 355, 374 (1986)));
NFIB v. Dep’t of Labor, 595 U.S. 109, 117 (2022) (per curiam) (“Administrative agencies are creatures of statute.
They accordingly possess only the authority that Congress has provided.”).
53
See Free Enterprise Fund v. Pub. Co. Acct. Oversight Bd., 561 U.S. 477, 497–98 (2010) (“The people do not vote
for the ‘Officers of the United States’” (quoting U.S. Const. art. II, § 2, cl. 2)).
54
15 U.S.C. § 41 (“Any Commissioner may be removed by the President for inefficiency, neglect of duty, or
malfeasance in office.”); see Humphrey’s Ex’r v. United States, 295 U.S. 602, 629–30 (1935) (holding that Section
1’s limitations on the President’s power to remove commissioners were constitutional); but see Seila Law LLC v.
Consumer Fin. Protection Bureau, 591 U.S. 197, 239 (2020) (Thomas, J., concurring in part and dissenting in part)
(explaining that the “independence” guaranteed by Humphrey’s Executor “poses a direct threat to our constitutional
structure and, as a result, the liberty of the American people,” and “the Court has repudiated almost every aspect of
Humphrey’s Executor.”).
8
To be sure, the administrative state can act with greater dispatch than Congress; but the difficulty
of legislating in Congress is a feature of the Constitution’s design, not a fault.
56
The
administrative state cannot legislate just because Congress declines to do so.
57
Thus, whenever we undertake to make rules governing the private conduct of hundreds of
millions of people who do not vote for us, we should not begin with determining what the right
answer to the policy question is. Rather, we must first assure ourselves of the power to answer
the question at all.
We do not have the power to issue the Final Rule. I agree with Commissioner Holyoak
that Section 6(g) of the Federal Trade Commission Act
58
does not authorize the Commission to
make substantive rules regulating private conduct, and I join in full her scholarly dissent.
59
The
best interpretation of Section 6(g) is that it authorizes the Commission to make rules governing
its internal affairs and procedures rather than generally applicable rules governing private
conduct.
Even if Section 6(g) of the FTC Act grants us substantive rulemaking authority, it does
not grant us the authority to issue this rule. Under the major-questions doctrine, administrative
agencies may enact rules of great “economic and political significance” only if Congress has
clearly and unambiguously granted them the authority to do so.
60
Congress has not clearly and
55
Collins v. Yellen, 141 S. Ct. 1761, 1784 (2021) (“The removal power helps the President maintain a degree of
control over the subordinates he needs to carry out his duties as the head of the Executive Branch, and it works to
ensure that these subordinates serve the people effectively and in accordance with the policies that the people
presumably elected the President to promote.”); Seila Law, 591 U.S. at 222–24 (discussing the President’s removal
power as a key to ensuring the executive branch is accountable to the people through the President—“the most
democratic and politically accountable official in Government”); Free Enterprise Fund, 561 U.S. at 498 (Executive
officers “instead look to the President to guide the assistants or deputies subject to his superintendence. Without a
clear and effective chain of command, the public cannot determine on whom the blame or the punishment of a
pernicious measure, or series of pernicious measures ought really to fall.” (cleaned up)); Myers v. United States, 272
U.S. 52, 131 (1926) (discussing the First Congress’s concern that restricting the President’s removal power would
undermine “the great principle of unity and responsibility in the executive department, which was intended for the
security of liberty and the public good”).
56
Dep’t of Transp. v. Ass’n of Am. R.R., 575 U.S. 43, 61 (2015) (Alito, J., concurring) (“The Constitution’s
deliberative process was viewed by the Framers as a valuable feature, not something to be lamented and evaded.”
(citations omitted)); West Virginia v. EPA, 597 U.S. 697, 738 (2022) (Gorsuch, J., concurring) (“[T]he framers
deliberately sought to make lawmaking difficult by insisting that two houses of Congress must agree to any new law
and the President must concur or a legislative supermajority must override his veto.”).
57
The Federalist No. 73, at 442 (Alexander Hamilton) (“The injury which may possibly be done by defeating a few
good laws will be amply compensated by the advantage of preventing a number of bad ones.”).
58
15 U.S.C. § 46(g).
59
Melissa Holyoak, Comm’r, Fed. Trade Comm’n, Dissenting Statement regarding In the Matter of the Non-
Compete Clause Rule, Matter Number P201200 (June 28, 2024) (hereinafter “Comm’r Holyoak Dissent”).
60
West Virginia, 597 U.S. at 721, 725.
9
unambiguously granted us the authority we today claim. The Final Rule is therefore unlawful
even if Congress has conferred on us some substantive rulemaking power.
A
The major-questions doctrine is the name recently given to a longstanding principle
governing the interpretation of statutes conferring power on administrative agencies.
61
The
principle is simple. When an agency claims to have the power to issue rules of “extraordinary . . .
economic and political significance,” it must “point to ‘clear congressional authorization’ for the
power it claims.”
62
Whether understood as a clear-statement rule
63
or “as part of the context in
which a delegation occurs,”
64
the doctrine rests on “both separation of powers principles and a
practical understanding of legislative intent.”
65
Article I vests the legislative power in Congress.
66
This vesting is both exclusive and
indefeasible.
67
Congress, and Congress alone, may exercise the legislative power. No other
branch of the federal government may take that power away from Congress, nor may Congress
willingly cede it to anyone else.
68
The line between the exercise of legislative power and the
other government powers is not always perfectly clear.
69
But when a “particular function requires
the exercise of a certain type of power, . . . then only the branch in which that power is vested
can perform it.”
70
This principle of exclusive vesting is also known in our law as
“nondelegation,” a term describing the prohibition on Congress willingly delegating its exclusive
powers to some other entity.
71
Against the backdrop of exclusive vesting and nondelegation, the major-questions
doctrine has emerged. It may be understood in either of two ways. First, it may be understood as
a clear-statement rule enforcing the constitutional prohibition on the delegation of legislative
61
Id. at 724 (“As for the major questions doctrine ‘label,’ it took hold because it refers to an identifiable body of law
that has developed over a series of significant cases all addressing a particular and recurring problem: agencies
asserting highly consequential power beyond what Congress could reasonably be understood to have granted.”
(cleaned up)).
62
Id. at 721, 723 (quoting FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120, 159–60 (2000), and Util. Air
Regul. Grp. v. EPA, 573 U.S. 302, 324 (2014)).
63
See West Virginia, 597 U.S. at 736 (Gorsuch, J., concurring).
64
Biden v. Nebraska, 143 S. Ct. 2355, 2380 (2023) (Barrett, J., concurring).
65
West Virginia, 597 U.S. at 723.
66
U.S. Const. art. I, § 1, cl. 1.
67
Whitman v. Am. Trucking Ass’ns, Inc., 531 U.S. 457, 472 (2001).
68
Gundy v. United States, 588 U.S. 128, 135 (2019) (plurality op.) (“Congress, this Court explained early on, may
not transfer to another branch ‘powers which are strictly and exclusively legislative.’”).
69
Ass’n of Am. R.R., 575 U.S. at 69 (Thomas, J., concurring in the judgment) (“[I]t does not follow that there is no
overlap between the three categories of governmental power. Certain functions may be performed by two or more
branches without either exceeding its enumerated powers under the Constitution.”).
70
Ibid.
71
See infra Part III.
10
authority, thereby protecting the separation of powers.
72
In this sense, the doctrine operates just
like other clear-statement rules that protect important constitutional interests like state
sovereignty, state and federal sovereign immunity, Indian treaty rights, the powers of the federal
courts, or the protection against retroactive laws.
73
This rule does not forbid Congress from
conferring on agencies the power to make rules of vast economic and political significance;
rather, to protect the separation of powers from an accidental or thoughtless breach, the rule
requires Congress to state its intention to confer that power clearly and unambiguously.
74
Second, the doctrine may be understood as the “context” against which a statutory
delegation is enacted, and therefore “a tool for discerning—not departing from—the text’s most
natural interpretation.”
75
On this understanding, the doctrine is not a substantive rule it all.
Rather, it forms part of the backdrop against which Congress enacts statutes conferring authority
on administrative agencies.
76
Here, common sense—informed by constitutional structure—tells
us that “Congress normally ‘intends to make major policy decisions itself, not leave those
decisions to agencies.’”
77
The presumption that Congress reserves the answers to major policy
questions for itself “makes eminent sense in light of our constitutional structure, which is itself
part of the legal context framing any delegation.”
78
At bottom, this understanding of the major-
questions doctrine tells us that “in a system of separated powers, a reasonably informed
interpreter would expect Congress to legislate on ‘important subjects’ while delegating away
only ‘the details.’”
79
We therefore “should ‘typically greet’ an agency’s claim to ‘extravagant
statutory power’ with at least some ‘measure of skepticism.’”
80
72
See West Virginia, 597 U.S. at 737–40 (Gorsuch, J., concurring).
73
Gregory, 501 U.S. at 464 (state sovereignty); Pennhurst State Sch. & Hosp. v. Halderman, 465 U.S. 89, 99–100
(1984) (state sovereign immunity); United States v. Nordic Vill., Inc., 503 U.S. 30, 33, 35 (1992) (federal sovereign
immunity); South Dakota v. Bourland, 508 U.S. 679, 686–87 (1993) (Indian treaty rights); Chambers v. NASCO,
Inc., 501 U.S. 32, 47–-48 (1991) (inherent powers of the federal courts); Hughes Aircraft Co. v. United States ex rel.
Schumer, 520 U.S. 939, 946 (1997) (retroactive application). See also Amy Coney Barrett, Substantive Canons and
Faithful Agency, 90 B.U. L. Rev. 109, 118–19 (2010) (discussing a broad range of substantive canons of
construction and clear-statement rules); John F. Manning, Clear Statement Rules and the Constitution, 110 Colum.
L. Rev. 399, 402–04 (2010) (explaining that clear-statement rules “impose something of a clarity tax upon
legislative proceedings” that touch on sensitive constitutional values, thereby protecting those values by requiring
Congress to be particularly clear when it wishes to “sacrifice a specified constitutional value in pursuit of its
regulatory agenda”).
74
West Virginia, 597 U.S. at 736 (Gorsuch, J., concurring).
75
Nebraska, 143 S. Ct. at 2376 (Barrett, J., concurring).
76
Id. at 2388 (Barrett, J., concurring).
77
Id. at 2380 (Barrett, J., concurring) (quoting U.S. Telecom. Ass’n v. FCC, 855 F.3d 381, 419 (D.C. Cir. 2017)
(Kavanaugh, J., dissenting from denial of reh’g en banc)).
78
Ibid.
79
Id. at 2380–81 (Barrett, J., concurring) (quoting Wayman v. Southard, 23 U.S (10 Wheat.) 1, 43 (1825)).
80
Id. at 2381 (Barrett, J., concurring) (quoting Util. Air, 573 U.S. at 324).
11
No matter how one understands the doctrine, it requires us to make two determinations.
First, we must determine whether the Final Rule presents a “major question.” Second, if it does,
we must determine whether we have “clear congressional authorization” to issue the Final
Rule.
81
B
The Final Rule presents a major question because it is an administrative decision of vast
“economic and political significance.”
82
Indeed, if the Final Rule does not present a major
question, I cannot imagine what would.
Several factors guide the threshold major-question inquiry. First and foremost is the
economic significance of the rule, that is, whether the rule regulates “‘a significant portion of the
American economy.’”
83
This factor accounts both for how much economic activity the rule
regulates, as well as the costs it imposes on the public.
84
Second is the political significance of
the rule.
85
A politically significant rule requires “consequential tradeoffs” on “major social and
economic policy decisions,”
86
or it addresses issues that have “been the subject of an earnest and
profound debate across the country.”
87
The third factor is whether the rule “intrudes into an area
that is the particular domain of state law.”
88
All three factors cut decisively in favor of treating
the Final Rule as presenting a major question.
81
West Virginia, 597 U.S. at 743, 746 (Gorsuch, J., concurring) (explaining how to determine, first, “when an agency
action involves a major question for which clear congressional authority is required,” and second, “what qualifies as
a clear congressional statement authorizing an agency’s action”).
82
Id. at 721; see also Ala. Ass’n of Realtors v. Dep’t of Health & Human Servs., 594 U.S. 758, 764 (2021) (per
curiam) (“We expect Congress to speak clearly when authorizing an agency to exercise powers of vast economic and
political significance.” (quotation marks omitted)).
83
Id. at 722 (quoting Util. Air, 573 U.S. at 324)); accord id. at 744 (Gorsuch, J., concurring).
84
See West Virginia, 597 U.S. at 729–30 (considering the “consequential” nature to the economy of the tradeoffs
embodied in EPA rule); NFIB, 595 U.S. at 117 (considering both the “breadth of authority” claimed by the
government, as well as the costs imposed on regulated firms, in treating a workplace vaccine-mandate as a major
question); Ala. Ass’n of Realtors, 594 U.S. at 764 (considering the scope and breadth of regulation, as well as the
costs imposed on regulated entities).
85
See, e.g., NFIB, 595 U.S. at 117 (assessing the number of Americans subject to the rule); Ala. Ass’n of Realtors,
594 U.S. at 764 (similar).
86
West Virginia, 597 U.S. at 729–30 (quotation marks omitted); accord id. at 744 (Gorsuch, J., concurring).
87
Id. at 732 (quoting Gonzales v. Oregon, 546 U.S. 243, 267–68 (2006)); accord id. at 743 (Gorsuch, J., concurring).
88
Ala. Ass’n of Realtors, 594 U.S. at 764; West Virginia, 597 U.S. at 744 (Gorsuch, J., concurring). Rules that
intrude on the traditional legislative prerogatives of the States are already subject to a different clear-statement rule:
The requirement that Congress use “exceedingly clear language if it wishes to significantly alter the balance
between federal and state power.” U.S. Forest Serv. v. Cowpasture River Pres. Ass’n, 590 U.S. 604, 621–22 (2020).
This principle is grounded in longstanding “background principles of construction” that preserve “the relationship
between the Federal Government and the States under our Constitution.” Bond v. United States, 572 U.S. 844, 857–
58 (2014). “Federal statutes impinging upon important state interests ‘cannot . . . be construed without regard to the
implications of our dual system of government. . . . [W]hen the Federal Government takes over . . . local radiations
12
1
The Final Rule regulates “a significant portion of the American economy”—indeed,
nearly the entire economy.
89
The Final Rule’s breadth is sweeping. It purports to apply to every
for-profit business in every sector of the American economy except for those few that lie outside
our jurisdiction.
90
And the practice it prohibits in those sectors is pervasive.
91
Nearly one-fifth of
employees in the United States are currently subject to a noncompete agreement, meaning that
the Final Rule abrogates nearly thirty million existing contracts.
92
And nearly two-fifths of
employees will be subject to such an agreement at some point in their careers.
93
Abrogating
nearly thirty million existing contracts, irrespective of the costs, is a question of vast “economic
… significance.”
94
The Final Rule’s incredible costs confirm its economic significance. The Commission
estimates that the Final Rule could cost employers between $400 billion and $488 billion in
additional wages and benefits over the next ten years alone.
95
And that estimate wildly
understates the costs of the Final Rule for two reasons. First, it accounts neither for the increased
costs that employers will incur to protect trade secrets and other proprietary information, nor for
the loss in value that former employers will suffer when former employees disseminate trade
secrets and other proprietary information. The enforcement of non-disclosure agreements and
trade-secret laws will presumably deter some dissemination. But litigation is costly and
difficult—especially the enforcement of non-disclosure agreements
96
—and some dissemination
will go either undetected or unremedied in court.
in the vast network of our national economic enterprise and thereby radically readjusts the balance of state and
national authority, those charged with the duty of legislating [must be] reasonably explicit.’” BFP v. Resol. Trust
Corp., 511 U.S. 531, 544 (1994) (quoting Felix Frankfurter, Some Reflections on the Reading of Statutes, 47 Colum.
L. Rev. 527, 539–40 (1947)). Thus, the Supreme Court does not recognize abrogations of sovereign immunity or the
preemption of state law absent clear statements of congressional intent. Bond, 572 U.S. at 857–58. “Th[ese] plain
statement rule[s are] nothing more than an acknowledgment that the States retain substantial sovereign powers under
our constitutional scheme, powers with which Congress does not readily interfere.” Gregory, 501 U.S. at 461.
89
West Virginia, 597 U.S. at 722; accord id. at 744 (Gorsuch, J., concurring).
90
See Final Rule, 89 Fed. Reg. at 38,357–58.
91
Id. at 38,343.
92
Id. at 38,346 (“The Commission estimates that approximately one in five American workers—or approximately 30
million workers—is subject to a non-compete. . . . A 2014 survey of workers finds that 18% of respondents work
under a non-compete and 38% of respondents have worked under one at some point in their lives.”).
93
Ibid.
94
West Virginia, 597 U.S. at 721.
95
Final Rule, 89 Fed. Reg. at 38,470.
96
See Nathaniel Grow, Free Agency for the Front Office: How Data Analytics and Noncompete Agreements
Threaten to Disrupt Competitive Balance in U.S. Professional Sports Leagues, 58 Am. Bus. L.J. 121, 137 (2021)
(“However, even under the best of circumstances, enforcing a nondisclosure agreement against a former employee
can be difficult due to the inability to closely monitor how the ex-employee is using the knowledge he or she gained
while working for his or her former employer.”); David Lincicum, Note, Inevitable Conflict?: California’s Policy of
13
Second, the Commission does not even try to quantify the costs of nullifying almost
every single noncompete in force across the country.
97
Those costs are very real. Unlike the
prospective prohibition of new noncompete agreements, the nullification of existing contracts
leaves well-meaning and honest businesses exposed to the consequences of decisions made in
reliance on these agreements, such as liberally sharing valuable information with their
employees. Many employers likely would have made different hiring and operational decisions if
they had known that the noncompete agreements they signed would become unenforceable as a
matter of federal law.
And those are just the costs inflicted on employers. The Commission further estimates
that the Final Rule will impose costs on the general economy in the form of a 7.9% decrease in
capital investment in existing businesses.
98
Given that nonfarm capital investment across the
United States totaled $1.9 trillion in 2022,
99
the Commission’s forecasted decrease could easily
reach $100 billion. The Commission hopes that effect will be offset by an increase in new firm
formation and corresponding investment, but it concedes that might not be the case.
100
The transfer of value from employers to employees, from some competitors to other
competitors, and from incumbents to new entrants may very well be sound policy. But it is a
decision of undoubted economic significance—one we would expect Congress to make on behalf
of its constituents rather than unelected technocrats. We therefore should not presume to
undertake the “consequential tradeoffs involved in such a choice” unless Congress told us
unambiguously to do it.
101
2
The rule also addresses “the subject of an earnest and profound debate across the
country,”
102
and “seeks to ‘intrud[e] into an area that is the particular domain of state law.’”
103
The regulation of contracts, including employment contracts, is a core exercise of the States’
police power.
104
There has been a robust and lengthy debate among the States on how best to
Worker Mobility and the Doctrine of “Inevitable Disclosure,” 75 S. Cal. L. Rev. 1257, 1271 (2002)
(“[C]onfidentiality agreements, which theoretically prevent employees from disclosing trade secrets . . . are hard to
enforce because monitoring departed employees is difficult.”).
97
Final Rule, 89 Fed. Reg. at 38,433; see also id. at 38,470.
98
Id. at 38,470.
99
United States Census Bureau, U.S. Nonfarm Employer Businesses Capital Investment up to $1,899.9 Billion in
2022 (Feb. 28, 2024), https://www.census.gov/newsroom/press-releases/2024/nonfarm-employer-business-capital-
investment.html.
100
See Final Rule, 89 Fed. Reg. at 38,470.
101
See West Virginia, 597 U.S. at 730.
102
Id. at 732 (quoting Gonzales, 546 U.S. at 267–68 (2006)); see also id. at 743 (Gorsuch, J., concurring).
103
Id. at 744 (Gorsuch, J., concurring) (quoting Ala. Ass’n of Realtors, 594 U.S. at 764).
104
See, e.g., West Coast Hotel Co. v. Parrish, 300 U.S. 379, 397–98 (1937); Home Bldg. & Loan Ass’n v. Blaisdell,
290 U.S. 398, 434–35 (1934); Manigault v. Springs, 199 U.S. 473, 480 (1905).
14
regulate noncompete agreements.
105
The overwhelming majority of States permit noncompete
agreements subject to regulations of the types of professions who can be made to sign them, pay
thresholds, time-frames, geographic reach, and damages.
106
An exceedingly small minority ban
them outright.
107
More than half the States have altered their laws governing noncompete
agreements in the last two decades.
108
Noncompete agreements have also been the subject of considerable debate in Congress.
Since 2015, Congress has considered and rejected a host of bills that would have regulated
noncompete agreements,
109
including six that would have imposed the same policy the Final
Rule imposes.
110
This extensive state and congressional action on noncompete agreements demonstrates
that this issue is politically significant and the subject of a roiling debate across the country. The
Commission’s termination of the democratic process and preemption of the existing laws of
forty-six States presents a major question.
111
C
Because the Final Rule presents a major question, the Commission must have “clear
congressional authorization” to promulgate it.
112
Nothing in the FTC Act comes close to clearing
that bar. The Commission’s statutory argument combines two phrases from two different
provisions of the FTC Act to create the incredible rulemaking power it asserts today: Congress’s
105
See supra Section I.B.
106
Final Rule, 89 Fed. Reg. at 439; id. at 460 n.1098.
107
Final Rule, 89 Fed. Reg. at 38,472–73 (“Currently, non-competes are broadly prohibited in four States:
California, North Dakota, Oklahoma, and Minnesota.”).
108
See supra Section I.B.; see also Comment by West Virginia and 17 other States, FTC-2023-0007-20892, at 13
(“Since 2011, 29 States and the District of Columbia have passed bills changing their noncompete laws.”); id. at 13–
14 (noting that some States have recently loosened restrictions on noncompete agreements, while other States have
rejected proposals to tighten restrictions).
109
See, e.g., Mobility and Opportunity for Vulnerable Employees Act (“MOVE Act”), S. 1504, 114th Cong. (2015)
(forbidding noncompete agreements for low-wage workers); Limiting Ability to Demand Detrimental Employment
Restrictions Act (“LADDER Act”), H.R. 2873, 114th Cong. (2015) (same); Freedom for Workers to Seek
Opportunity Act, H.R. 4254, 114th Cong. (2015) (forbidding noncompete agreements for grocery-store employees);
Freedom to Compete Act, S. 124, 116th Cong. (2019) (preventing employers from using non-compete agreements in
employment contracts for certain employees).
110
See, e.g., Workforce Mobility Act of 2021, S. 483, 117th Cong. (2021); Workforce Mobility Act of 2021, H.R.
1367, 117th Cong. (2021); Workforce Mobility Act of 2019, S. 2614, 116th Cong. (2019); Workforce Mobility Act
of 2020, H.R. 5710, 116th Cong. (2020); Workforce Mobility Act of 2018, S. 2782, 115th Cong. (2018); Workforce
Mobility Act of 2018, H.R. 5631, 115th Cong. (2018).
111
See West Virginia, 597 U.S. at 732 (“‘The importance of the issue,’ along with the fact that the same basic scheme
EPA adopted ‘has been the subject of an earnest and profound debate across the country, . . . makes the oblique form
of the claimed delegation all the more suspect.’” (quoting Gonzales, 546 U.S. at 267–68)).
112
Ibid. (quotation marks omitted).
15
grant of authority to “prevent” the use of “unfair methods of competition” in Section 5,
113
together with Section 6’s grant of authority to “[f]rom time to time classify corporations and …
to make rules and regulations for the purpose of carrying out the provisions of” the FTC Act.
114
But this bank-shot statutory argument is not “clear congressional authorization”
115
to issue the
Final Rule.
1
Whether Congress has provided clear congressional authorization is, like any statutory-
interpretation question, primarily a textual one. Of course, text cannot be read in isolation. “It is a
fundamental canon of statutory construction that the words of a statute must be read in their
context and with a view to their place in the overall statutory scheme.”
116
Context is particularly
critical for the “extraordinary cases” where an agency invokes the power to issue rules of vast
“economic and political significance.”
117
In every major-questions-doctrine case, the agency’s
“regulatory assertions had a colorable textual basis.”
118
But resorting to “the outer limits of [the
text’s] definitional possibilities” does not fly when an agency is claiming vast regulatory
authority.
119
When an agency claims such power, the text on which it relies must be read with a
careful eye to its “place” in “the overall statutory scheme.”
120
“Where an agency relies on
“oblique or elliptical language,” or combines a series of “modest words, vague terms, or subtle
devices,”
121
it will not prevail even if its reading “would have been plausible if the relevant
statutory text were read in a vacuum.”
122
The doctrine’s emphasis on statutory context reflects
“common sense as to the manner in which Congress is likely to delegate a policy decision of
such economic and political magnitude to an administrative agency.”
123
When Congress intends
to give away a core part of its power, that intention will be screamingly obvious in the context of
the statutory scheme.
124
113
15 U.S.C. § 45(a)(2).
114
15 U.S.C. § 46(g).
115
West Virginia, 597 U.S. at 732 (quotation marks omitted).
116
Nat’l Ass’n of Home Builders v. Defs. of Wildlife, 551 U.S. 644, 666 (2007) (quotation marks omitted).
117
West Virginia, 597 U.S. at 721 (quotation marks omitted).
118
Id. at 722.
119
FCC v. AT&T, Inc., 562 U.S. 397, 407 (2011) (quotations marks omitted); see also West Virginia, 597 U.S. at 732.
120
Brown & Williamson, 529 U.S. at 133; see also id. at 132 (“In determining whether Congress has specifically
addressed the question at issue, a reviewing court should not confine itself to examining a particular statutory
provision in isolation. The meaning—or ambiguity—of certain words or phrases may only become evident when
placed in context.”).
121
West Virginia, 597 U.S. at 723 (quotation marks omitted).
122
Nebraska, 143 S. Ct. at 2382 (Barrett, J., concurring).
123
Brown & Williamson, 529 U.S. at 133.
124
West Virginia, 597 U.S. at 723 (“Congress [does not] typically use oblique or elliptical language to empower an
agency to make a ‘radical or fundamental change’ to a statutory scheme.”); Id. at 746 (Gorsuch, J., concurring)
(“[W]e look for clear evidence that the people’s representatives in Congress have actually afforded the agency the
16
Nothing about the Commission’s statutory argument demonstrates that “Congress in fact
meant to confer the power the agency has asserted.”
125
Section 5(a) is a general grant of authority
to “prevent … unfair methods of competition.”
126
The remainder of that section is not addressed
to rulemakings at all. It instead lays out the procedures for what was always understood to be the
Commission’s lone enforcement mechanism—case-by-case adjudication through a “quasi
judicial” process.
127
This highly general grant of authority comes nowhere near to the specificity
that the major-questions doctrine requires.
128
The Commission reasons that the specificity comes from Section 6(g). Section 6 is titled
“Additional powers of Commission”
129
—language should warn an ordinary reader that what
follows are “ancillary … gap filler[s],” rather than “clear congressional authorization” to
restructure the American labor market.
130
And sure enough, that is what Section 6(g) is. It reads:
“The Commission shall also have power to … [f]rom time to time classify corporations and
(except as provided in [the Magnuson-Moss Act]) to make rules and regulations for the purpose
of carrying out the provisions of this subchapter.”
131
So the Commission’s argument rests entirely
on the second half of a single sentence—contained in the “Additional powers” section of our
organic statute—the primary purpose of which is authorizing the Commission to classify
corporations. This text is far less “clear congressional authorization” for the immense power the
Commission today claims than the argument the Supreme Court rejected in West Virginia.
132
This
“oblique and elliptical” language in an “ancillary” provision of the FTC Act is simply not enough
power it claims.”); Nebraska, 143 S. Ct. at 2383 (“[A] reasonable speaker would not understand Congress to confer
an unusual form of authority without saying more.”) (Barrett, J. concurring).
125
West Virginia, 597 U.S. at 721.
126
15 U.S.C. § 45(a)(2).
127
A.L.A. Schechter Poultry Corp. v. United States, 295 U.S. 495, 533 (1935) (“What are ‘unfair methods of
competition’ are thus to be determined in particular instances, upon evidence, in the light of particular competitive
conditions and of what is found to be a specific and substantial public interest. To make this possible, Congress set
up a special procedure. A commission, a quasi judicial body, was created.” (internal citations omitted)).
128
West Virginia, 597 U.S. at 721 (“Where the statute at issue is one that confers authority upon an administrative
agency, that inquiry must be ‘shaped, at least in some measure, by the nature of the question presented’—whether
Congress in fact meant to confer the power the agency has asserted.” (quoting Brown & Williamson, 529 U.S. at
159)).
129
5 U.S.C. § 46.
130
West Virginia, 597 U.S. at 724.
131
15 U.S.C. § 46(g).
132
See West Virginia, 597 U.S. at 732–35 (holding that statute authorizing EPA to promulgate “standard[s] of
performance” for “sources” of “air pollutants” based on “the best system of emission reduction” that it finds “has
been adequately demonstrated” was insufficiently clear to authorize EPA to require coal-fired plants to outright
“reduce their own production of electricity, or subsidize increased generation by natural gas, wind, or solar sources,”
and not just increase their own pollution performance).
17
to justify the Final Rule, even if you believe that some substantive rulemaking is within the
“definitional possibilities” of Section 6(g).
133
2
The Commission’s prior rulemakings, as well the complete absence of any Section 5
enforcement actions against noncompete agreements before this rulemaking, confirm that
Sections 5 and 6(g) do not contain clear congressional authorization for the Final Rule.
Executive-branch practice has long been relevant to interpreting the scope of Congress’s grant of
authority to administrative agencies.
134
“[J]ust as established practice may shed light on the
extent of power conveyed by general statutory language, so the want of assertion of power by
those who presumably would be alert to exercise it, is equally significant in determining whether
such power was actually conferred.”
135
Put another way, “[f]ailure to use such an important
power for so long a time indicates . . . that the Commission did not believe the power existed.”
136
We have never invoked these sections of the FTC Act to do anything remotely similar to
the Final Rule—strong evidence that the Final Rule lies beyond our authority. The Commission
insults the reader by claiming that “non-competes have already been the subject of FTC scrutiny
and enforcement,” thereby making this rule “a more incremental—and thus less significant—step
than it would be for an agency to wade into an area not currently subject to its enforcement
authority.”
137
The truth is that, in the 109 years between Section 5’s adoption in 1914
138
and the
day before releasing the NPRM, the Commission did not bring a single enforcement action
against an employee noncompete agreement under Section 5.
139
Not one. The first time it did was
the day before issuing the NPRM—suspicious timing to say the least.
140
And none of the
enforcement actions filed on that day were litigated. In each case, the Commission accused an
employer of violating Section 5 through the enforcement of noncompete agreements, and the
employer settled without conceding a violation of Section 5.
141
It beggars belief to contend that
Section 5 contains within it a categorical prohibition on all noncompete agreements when the
Commission did not enforce Section 5 against a single noncompete agreement in its entire 109-
133
Id. at 723, 733.
134
See, e.g., Bittner v. United States, 598 U.S. 85, 97 (2023) (“[C]ourts may consider the consistency of an agency’s
views when we weigh the persuasiveness of any interpretation it proffers in court.”); West Virginia, 597 U.S. at 724–
25; id. at 747 (Gorsuch, J., concurring) (“[C]ourts may examine the age and focus of the statute” as well as “the
agency’s past interpretations of the relevant statute.”).
135
FTC v. Bunte Bros., 312 U.S. 349, 352 (1941).
136
Fed. Power Comm’n v. Panhandle Eastern Pipe Line Co., 337 U.S. 498, 513 (1949).
137
Final Rule, 89 Fed. Reg. at 38,353.
138
Federal Trade Commission Act, ch. 311, § 5, 38 Stat. 719 (1914).
139
See supra Section I.C.
140
See ibid.
141
See ibid.
18
year history until a day before this rulemaking.
142
“This ‘lack of historical precedent,’ coupled
with the breadth of authority that the [Commission] now claims, is a ‘telling indication’ that the
[Final Rule] extends beyond the [Commission]’s legitimate reach.”
143
The same is true when Sections 5 and 6(g) are considered together. The Commission has
deployed this bank-shot rulemaking theory only once in its history. In 1967, it issued a two-page
rule addressing the circumstances under which an “advertising payment or promotional
allowance” paid by a manufacturer of “men’s, youths’ and boys’ suits, coats, overcoats, topcoats,
jackets, dress trousers and uniforms” to resellers of those clothing items would be treated as
violating the antitrust laws.
144
This insignificant, and legally dubious, rule was issued with no
fanfare; was never enforced; was never subject to judicial review; and was repealed more than
thirty years ago.
145
That means that the Commission has invoked this statutory theory twice in
110 years: once to alert a single industry to the antitrust consequences of an isolated practice, and
again to redistribute nearly half a trillion dollars within the general economy by banning a
centuries-old contract deployed in every industry in the country. This is precisely the sort of
statutory theory the Court condemned in West Virginia: claiming to “‘discover’” the power to
“restructure the American [labor] market” in the “vague language of an ‘ancillary provision’ of
the” FTC Act, “one that was designed to function as a gap filler and had rarely been used in the
preceding decades.”
146
The Commission mines the historical record for additional support and comes up with
rules it promulgated in reliance on its power to prevent “unfair or deceptive acts or practices.”
147
This argument fails. As Commissioner Holyoak ably explains, the Commission did not have the
power to issue any rules, including those, until Congress enacted the Magnuson-Moss Act in
1975, and no one thought it had such power for nearly fifty years after Section 6(g) became
142
Noncompete agreements are contracts in restraint of trade, and therefore subject to the rule of reason under
Section 1 of the Sherman Act and Section 5 of the FTC Act. See supra note 37; see also infra notes 284, 286. But as
is true of all agreements that do not implicate one of the few per se rules, whether a given noncompete agreement
violates the antitrust laws will turn entirely on the particular circumstances and competitive effects of that
agreement. See infra Section IV.A. The Final Rule rests on a far more aggressive proposition: that Section 5
categorically forbids every single noncompete agreement irrespective of their particular effects. Section 5 contains
no such categorical prohibition, and the Commission’s failure to enforce this claimed prohibition against a single
noncompete agreement until the day before this rulemaking confirms the absence of such a prohibition.
143
NFIB, 595 U.S. at 119–20 (quoting Free Enterprise Fund, 561 U.S. at 505); see also Bunte Bros., 312 U.S. at
352; accord West Virginia, 597 U.S. at 725; id. at 747–48 (Gorsuch, J., concurring).
144
Discriminatory Practices in Men’s and Boys’ Tailored Clothing Industry, 32 Fed. Reg. 15,584, 15,585 (Nov. 8,
1967).
145
See Repeal of Trade Regulation Rule: Discriminatory Practices in Men’s and Boys’ Tailored Clothing Industry,
59 Fed. Reg. 8527 (Feb. 23, 1994).
146
West Virginia, 597 U.S. at 724 (quoting Util. Air, 573 U.S. at 324, and Whitman, 531 U.S. at 468).
147
See Final Rule, 89 Fed. Reg. at 38,349–50. Congress conferred that power on us in the Magnuson-Moss
Warranty—Federal Trade Commission Improvements Act, Pub. L. No. 93-637, 88 Stat. 2183 (1975).
19
law.
148
But even assuming Section 6(g) confers rulemaking power, the question is not whether
the Commission has the power to issue some rules; it is whether Congress has made clear our
power to make a “decision of such economic and political significance” as the one embodied in
the Final Rule.
149
That we have issued rules in the past does not answer that question. In NFIB, for
example, the agency issued dozens of rules on workplace safety under the statute it invoked to
issue the vaccine mandate.
150
But it had never promulgated a regulation “of th[e] kind” embodied
in the vaccine mandate, which reached much further than any previous rule.
151
Similarly, EPA
had issued multiple rules relying on the same statute it used to issue carbon-emissions standards
in West Virginia.
152
But it had never relied on the statute to issue the sort of market-transforming
rule the Court struck down in that case.
153
And in Nebraska, the agency had invoked the relevant
statute dozens of times to issue waivers and modifications of student loans “addressing . . .
specific issues.”
154
But the Court rejected the agency’s invocation of the statute “to release 43
million borrowers from their obligations to repay $430 billion in student loans” because the
agency “had never previously claimed powers of this magnitude.”
155
The same is true here. “No regulation premised on” Sections 5 and 6(g) “has even begun
to approach the size or scope” of the Final Rule.
156
The rules the Commission cites were part of a
rash of rulemakings in the 1960s and 1970s—more than fifty years after Congress enacted the
relevant language in Sections 5 and 6(g)—known as “Trade Regulation Rules.”
157
These rules
were overwhelmingly “minor” and “uncontroversial.”
158
It was initially unclear whether they
carried the force of law at all.
159
They applied to specific practices carried out in specific markets
and industries rather than general practices pervasive in the general economy. Almost every
single rule applied to representations or disclosures about specific products or practices. Even the
more controversial rules in the bunch bore no resemblance to the Final Rule, applying only to
148
Comm’r Holyoak Dissent, supra note 59, at Part II.
149
West Virginia, 597 U.S. at 729 (quoting Brown & Williamson, 529 U.S. at 160); see also id. at 721 (“Where the
statute at issue is one that confers authority upon an administrative agency, that inquiry must be shaped, at least in
some measure, by the nature of the question presented—whether Congress in fact meant to confer the power the
agency has asserted.” (quotation marks omitted)).
150
See NFIB, 595 U.S. at 133 (Breyer, Sotomayor, and Kagan, JJ., dissenting).
151
Id. at 119–20 (per curiam).
152
See West Virginia, 597 U.S. at 710–11; id. at 776–77 (Kagan, J., dissenting).
153
See id. at 726–30; id. at 749 (Gorsuch, J., concurring).
154
Nebraska, 143 S. Ct. at 2363.
155
Id. at 2372.
156
Ala. Ass’n of Realtors, 594 U.S. at 765.
157
Thomas W. Merrill & Kathryn Tongue Watts, Agency Rules with the Force of Law: The Original Convention, 116
Harv. L. Rev. 467, 552 (2002).
158
Id. at 554 & n.456; see also id. at 552–53 (“None of these early rules stirred much controversy.”).
159
Id. at 552 & n.442.
20
labels and warnings for two specific products.
160
None of these rules governed any aspect of the
labor market in any industry, much less the entire labor market across the whole country. They
provide no support for the proposition that Sections 5 and 6(g) contain the authority that the
Commission today claims.
161
Even accepting the Commission’s view of which rules are relevant to the historical
inquiry, that inquiry cuts decisively against the Final Rule. The Commission’s own conduct for
its first fifty years suggests that it understood Section 6(g) to contain “oblique or elliptical
language” that was “designed to function as a gap filler” rather than confer transformative
regulatory authority.
162
The Commission’s statutory theory is thus precisely what the major-
questions doctrine prohibits: the exploitation of “some gap, ambiguity, or doubtful expression in
Congress’s statutes to assume responsibilities far beyond those the people’s representatives
actually conferred on them.”
163
***
There are sound arguments in favor of regulating noncompete agreements. Every State
does. But “no matter how important, conspicuous, and controversial the issue,” and no matter
how wise the administrative solution, “an administrative agency’s power to regulate . . . must
always be grounded in a valid grant of authority from Congress.”
164
When an agency does
something as big as the Commission does in the Final Rule, the grant of authority it relies upon
must be more than oblique language tucked away in an ancillary subsection of its organic act.
But that is all the Commission has to justify the Final Rule. It is therefore unlawful.
III
Even if the Commission has the statutory authority to issue the Final Rule, it must
surmount another hurdle. A federal agency may act only upon a “valid grant of authority from
Congress.”
165
A grant of authority is “valid” only if it is a lawful exercise of Congress’s
legislative authority under the Constitution. If Sections 5 and 6(g) grant us the power to issue the
Final Rule, I believe that grant would be an unconstitutional delegation of legislative power.
160
See id. at 553–55 (describing the controversy surrounding, and congressional action repealing, the Unfair or
Deceptive Advertising and Labeling of Cigarettes in Relation to the Health Hazards of Smoking, 29 Fed. Reg. 8324
(July 2, 1964), repealed by 30 Fed. Reg. 9485 (July 29, 1965), and Posting of Minimum Octane Numbers on
Gasoline Dispensing Pumps, 36 Fed. Reg. 23,871 (Dec. 16, 1971), repealed by 43 Fed. Reg. 43,022 (Sept. 22,
1978)).
161
NFIB, 595 U.S. at 119 (“This ‘lack of historical precedent,’ coupled with the breadth of authority that the
[Commission] now claims, is a ‘telling indication’ that the mandate extends beyond the [Commission]’s legitimate
reach.” (quoting Free Enterprise Fund, 561 U.S. at 505)).
162
West Virginia, 597 U.S. at 723–24.
163
Id. at 742 (Gorsuch, J., concurring) (quotation marks omitted).
164
Brown & Williamson, 529 U.S. at 161 (cleaned up).
165
Id. (emphasis added).
21
A
Because the Constitution vests all legislative power exclusively in Congress, Congress
may not delegate it away.
166
Courts enforce this principle through the “nondelegation
doctrine.”
167
Although the courts sometimes speak as though the doctrine assesses whether a
particular delegation is constitutional,
168
the prohibition on delegation is categorical.
169
Congress
may never delegate the legislative power that the Constitution vests in it.
170
Not every grant of authority to the executive branch, however, is a delegation of
legislative power. “Certain functions may be performed by two or more branches without either
exceeding its enumerated powers under the Constitution.”
171
And “a certain degree of discretion
… inheres in most executive or judicial action.”
172
Conferring discretionary authority on the
executive branch to perform a function that is not an exercise of purely legislative power, even if
Congress could also perform it, is not a delegation of legislative power.
173
Congress thus has
power to “obtain[ ] the assistance of its coordinate Branches”
174
in the implementation of
national policy by, for example, directing the President to apply a statutory command only under
certain circumstances and investing him with the authority to determine whether those
circumstances in fact exist.
175
The “core of the legislative power,” which only Congress may exercise, is the power to
make “generally applicable rules of private conduct,”
176
or, put differently, to “prescribe general
rules for the government of society.”
177
But “classifying governmental power is an elusive
166
See supra Section II.A; Gundy, 588 U.S. at 135 (“Accompanying that assignment of power to Congress is a bar
on its further delegation.”).
167
Id. at 132 (“The nondelegation doctrine bars Congress from transferring its legislative power to another branch of
Government.”); Mistretta v. United States, 488 U.S. 361, 371–72 (“The nondelegation doctrine is rooted in the
principle of separation of powers that underlies our tripartite system of Government,” and “mandate[s] that Congress
generally cannot delegate its legislative power to another Branch.”).
168
See, e.g., Gundy, 588 U.S. at 135 (deploying the nondelegation doctrine to assess whether “a statutory delegation
is constitutional”); Big Time Vapes, Inc. v. FDA, 963 F.3d 436, 441–42 (5th Cir. 2020) (explaining circumstances
when “[d]elegations are constitutional”); United States v. Dhafir, 461 F.3d 211, 215 (2d Cir. 2006) (defining
circumstances when “[d]elegations of congressional authority are upheld”).
169
Whitman, 531 U.S. at 472 (“This text permits no delegation of those powers . . . .”).
170
Touby v. United States, 500 U.S. 160, 165 (1991) (“Congress may not constitutionally delegate its legislative
power to another branch of Government.”).
171
Ass’n of Am. R.R., 575 U.S. at 69 (2015) (Thomas, J., concurring in the judgment).
172
Mistretta, 488 U.S. at 417 (Scalia, J., dissenting).
173
Wayman, 23 U.S. (10 Wheat.) at 42, 43 (“Congress may certainly delegate to others, powers which the legislature
may rightfully exercise itself” so long as those powers are not “strictly and exclusively legislative.”).
174
Mistretta, 488 U.S. at 372.
175
See, e.g., Marshall Field & Co. v. Clark, 143 U.S. 649, 683–89 (1892) (explaining operation of conditional
legislation).
176
Ass’n of Am. R.R., 575 U.S. at 76 (Thomas, J., concurring in the judgment).
177
Gundy, 588 U.S. at 153. (Gorsuch, J., dissenting) (quoting Fletcher v. Peck, 10 U.S. (6 Cranch) 87, 136 (1810)).
22
venture.”
178
To draw “the true distinction . . . between the delegation of power to make the law,
which necessarily involves a discretion as to what it shall be, and conferring authority or
discretion as to its execution,”
179
the Supreme Court has turned to the “intelligible-principle
rule.” That “rule seeks to enforce the understanding that Congress may not delegate the power to
make laws and so may delegate no more than the authority to make policies and rules that
implement its statutes.”
180
The rule requires that Congress “lay down by legislative act an
intelligible principle to which the person or body authorized to [exercise the power conferred] is
directed to conform.”
181
If it does, then the law is not an unconstitutional delegation of legislative
authority.
182
B
“In a delegation challenge, the constitutional question is whether the statute has delegated
legislative power to the agency.”
183
The “answer requires construing the challenged statute to
figure out what task it delegates and what instructions it provides.”
184
The inquiry is not as
specific as whether Section 5 prohibits the noncompete agreements that the Final Rule bans.
Rather, it is whether Section 5 “lays down . . . an intelligible principle,”
185
that is, whether it
“provide[s] sufficiently ‘definite’ standards” to “guide[ ] executive discretion to accord with
Article I.”
186
Only if “unfair methods of competition” is a “definite” constraint on the exercise of
our discretion will it pass constitutional muster.
1
The operative word in the phrase “unfair methods of competition” is “unfair.” “[M]ethods
of competition” are business practices affecting competition in which businesses engage.
187
“[U]nfair” describes the standard, such as it is, to which private parties must conform when
engaging in business practices, and therefore is also the “congressionally mandated standard”
guiding the Commission’s discretion.
188
178
Ass’n of Am. R.R., 575 U.S. at 76 (Thomas, J., concurring in the judgment).
179
Marshall Field, 143 U.S. at 693–94 (quotation marks omitted).
180
Loving v. United States, 517 U.S. 748, 771 (1996).
181
J.W. Hampton, Jr., & Co. v. United States, 276 U.S. 394, 409 (1928).
182
Ibid.
183
Whitman, 531 U.S. at 472.
184
Gundy, 588 U.S. at 136 (quoting Whitman, 531 U.S. at 473; Am. Power & Light Co. v. SEC, 329 U.S. 90, 104–
105 (1946)).
185
Id. at 135 (quoting J. W. Hampton, 276 U.S. at 409).
186
Id. at 135–36.
187
See, e.g., E.I. du Pont de Nemours & Co. v. FTC, 729 F.2d 128, 138–39 (2d Cir. 1984).
188
See FTC v. Sperry & Hutchinson Co., 405 U.S. 233, 244 (1972) (The FTC Act instructs the Commission to
“measur[e] a practice against the elusive, but congressionally mandated standard of fairness”).
23
“Unfair” is hardly a definite constraint on our discretion. “The term ‘unfair’ is an elusive
concept, often dependent upon the eye of the beholder.”
189
A practice is unfair if it is “not
impartial,” “unjust,” “involving a trick or artifice,”
190
or “inequitable.”
191
On the theory
elaborated in the Final Rule, then, the Commission has the power to proscribe ex ante any
method of competition that is “unjust” or “inequitable.” That does not sound like the sort of
“intelligible principle”
192
that meaningfully constrains our rulemaking authority. It sounds
instead like the exercise of legislative power which only Congress may wield.
The legislative history sheds more light on the statute’s meaning, but compounds the
delegation problem. Congress adopted the phrase “unfair method of competition” because of its
incredible breadth.
193
Congress considered a narrower phrase with “a well settled meaning at
common law”—“unfair competition”—but rejected it as “too narrow.”
194
It instead chose “unfair
methods of competition” because it was “broader and more flexible.”
195
Section 5, a sponsor
declared, “covers every practice and method between competitors upon the part of one against
the other that is against public morals . . . or is an offense for which a remedy lies either at law or
in equity.”
196
Congress declined to define the term at all, worried that it would be “impossible to frame
definitions which embrace all unfair practices.”
197
Because “[t]here is no limit to human
inventiveness in this field,” Congress feared that any list of definitions it could create would
quickly become outdated and “it would be at once necessary to begin over again.”
198
It
concluded that the “better” course was to “condemn[ ] unfair practices” by a “general
declaration” and “leave it to the commission to determine what practices were unfair,” rather
than “attempt[ing] to define the many and variable unfair practices which prevail in
commerce.”
199
As one of the FTC Act’s legislative champions put it: “My belief is that [Section
5] will cover everything that we want, and will have such an elastic character that it will meet
189
E.I. du Pont de Nemours, 729 F.2d at 137.
190
Webster’s Complete Dictionary of the English Language 1442 (1886).
191
8 The Century Dictionary & Cyclopedia 6606 (1901).
192
Gundy, 588 U.S. at 135.
193
See generally Neil W. Averitt, The Meaning of “Unfair Methods of Competition” in Section 5 of the Federal
Trade Commission Act, 21 B.C. L. Rev. 227, 229–38 (1980) (describing legislative history of the Act and Congress’s
insistence on adopting a “vague and general language” to maximize Commission’s discretion).
194
FTC v. Raladam Co., 283 U.S. 643, 648 (1931); see also FTC v. R.F. Keppel & Bro., 291 U.S. 304, 311–12 &
nn.2–3 (1934).
195
R.F. Keppel & Bro., 291 U.S. at 311.
196
51 Cong. Rec. 11,112 (1914) (remarks of Sen. Newlands).
197
H.R. Rep. No. 63-1142, at 19 (1914) (Conf. Rep.).
198
Ibid.
199
S. Rep. No. 63-597, at 13 (June 13, 1914).
24
every new condition and every new practice that may be invented with a view to gradually
bringing about monopoly through unfair competition.”
200
Congress resolved this textual indeterminacy by leaving the scope of “unfair methods of
competition” to “an administrative body of practical men . . . who will be able to apply the rule
enacted by Congress to particular business situations, so as to eradicate evils with the least risk
of interfering with legitimate business operations.”
201
It created the Commission “with the
avowed purpose of lodging the administrative functions committed to it in ‘a body specially
competent to deal with them by reason of information, experience and careful study of the
business and economic conditions of the industry affected,’” and organized it “‘with respect to
the length and expiration of the terms of office of its members, as would ‘give to them an
opportunity to acquire the expertness in dealing with these special questions.’”
202
Congress gave
the Commission “wide latitude . . . to discover and make explicit those unexpressed standards of
fair dealing which the conscience of the community may progressively develop.”
203
The legislative history thus confirms what the text suggests. Congress enacted a
prohibition of tremendous “sweep and flexibility”
204
and declined to define the conduct it was
prohibiting in anything but the vaguest sense because it intended the Commission to do the
defining. It therefore left the key legislative task—the definition of the private conduct prohibited
by law—to the executive branch.
2
The nondelegation problem with Section 5 is obvious. Indeed, “[m]any congressmen had
expressed doubt about the constitutionality of Section 5.”
205
One Senator warned that the Act
was tantamount to “appoint[ing] a commission and authoriz[ing] that commission to pass such
laws as it pleases or to deal with business conditions as it pleases.”
206
Such a law “[o]bviously …
would be beyond [Congress’s] power, because that would be to delegate [Congress’s] power and
[Congress’s] authority.”
207
200
51 Cong. Rec. 12,024 (July 13, 1914) (remarks of Sen. Newlands).
201
Atl. Refining Co. v. FTC, 381 U.S. 357, 367 (1965) (quoting H.R. Rep. No. 63-1142, at 19 (1914) (Conf. Rep.)).
202
R.F. Keppel & Bro., 291 U.S at 314 (quoting S. Rep. No. 63-597, at 9, 11 (June 13, 1914)).
203
FTC v. Standard Educ. Soc., 86 F.2d 692, 696 (2d Cir. 1936) (Hand, J.), rev’d on other grounds, 302 U.S. 112
(1937); accord FTC v. Beech-Nut Packing Co., 257 U.S. 441, 453 (1922); see also S. Rep. No. 63-597, at 22 (June
13, 1914) (“The organization should be quasi judicial in character. We want traditions; we want a fixed policy; we
want trained experts; we want precedents; we want a body of administrative law built up.”).
204
Sperry & Hutchinson Co., 405 U.S. at 241.
205
Averitt, supra note 193, at 298 n.300.
206
See, e.g., 51 Cong. Rec. 12816 (remarks of Sen. Sutherland).
207
Ibid.; see also 51 Cong. Rec. 11113 (remarks of Sen. Reed) (“The people must, through their representatives, lay
down the law that defines what is legal and illegal. . . . Congress cannot create a tribunal and authorize it to create
25
Apparently aware of the nondelegation danger, the Supreme Court tried to articulate
some limitations on the Commission’s discretion.
208
In its first case considering the question, the
Court acknowledged that “[t]he words ‘unfair method of competition’ are not defined by the
statute and their exact meaning is in dispute.”
209
It defined the term as encompassing those
“practices . . . opposed to good morals because characterized by deception, bad faith, fraud, or
oppression, or as against public policy because of their dangerous tendency unduly to hinder
competition or create monopoly.”
210
In subsequent cases, the Court limited Section 5’s reach to
violations of the existing antitrust laws.
211
The Court began to abandon this limitation in the 1930s. In FTC v. R.F. Keppel &
Brother, Inc., the Court upheld a Commission order prohibiting a marketing scheme “of the sort
which the common law and criminal statutes have long deemed contrary to public policy,” but
which did not violate the antitrust laws.
212
Section 5’s application therefore now reached further
than the antitrust laws. For a while, however, the Supreme Court continued to link Section 5 to
the antitrust laws by treating Section 5 as a mandate “to stop in their incipiency acts and
practices which, when full blown, would violate” the antitrust laws, as well as to condemn “as
unfair methods of competition” existing violations of them.
213
laws. Congress cannot create a board and empower it to govern by its own reason, to rule by its own discretion, to
do what which to it seems proper, and penalize the citizen for refusing to render obedience.”).
208
Averitt, supra note 193, at 298 (“The Commission’s power over ‘unfair methods of competition’ could have been
a classic example of excessive delegation. . . . [T]he Supreme Court may have been endeavoring to shield Section 5
as a whole from constitutional infirmity.” (internal quotations omitted)).
209
FTC v. Gratz, 253 U.S. 421, 427 (1920).
210
Ibid.
211
Beech-Nut Packing, 257 U.S. at 453; see also FTC v. Curtis Publ’g Co., 260 U.S. 568, 579–82 (1923) (setting
aside Commission order based on a “vague general complaint charging unfair methods of competition” unrelated to
any existing statement of public policy); FTC v. Sinclair Oil Co., 261 U.S. 463, 473–76 (1923); see also Raladam
Co., 283 U.S. at 649 (“It is obvious that the word ‘competition’ imports the existence of present or potential
competitors, and the unfair methods must be such as injuriously affect or tend thus to affect the business of these
competitors.”).
212
291 U.S. at 243–44.
213
FTC v. Motion Picture Advert. Serv. Co., 344 U.S. 392, 394–95 (1953); see also FTC v. Brown Shoe Co., 384
U.S. 316, 322 (1966) (“We . . . hold that the Commission has power under § 5 to arrest trade restraints in their
incipiency without proof that they amount to an outright violation of § 3 of the Clayton Act or other provisions of
the antitrust laws.”); Fashion Originators’ Guild of Am. v. FTC, 312 U.S. 457, 463, 466 (1941) (“If the purpose and
practice of the combination . . . runs counter to the public policy declared in the Sherman and Clayton Acts, the
Federal Trade Commission has the power to suppress it as an unfair method of competition. . . . It was the object of
the Federal Trade Commission Act to reach not merely in their fruition but also in their incipiency combinations
which could lead to these and other trade restraints and practices deemed undesirable.”); Atl. Refining Co.,381 U.S.
at 369 (“As our cases hold, all that is necessary in § 5 proceedings to find a violation is to discover conduct that runs
counter to the public policy declared in the Act. But this is of necessity, and was intended to be, a standard to which
the Commission would give substance. In doing so, its use as a guideline of recognized violations of the antitrust
laws was, we believe, entirely appropriate.” (internal quotation marks and citations omitted)).
26
The Court has since abandoned even that limitation. In FTC v. Sperry & Hutchinson Co.,
the Court surveyed the Act’s legislative history and concluded that the Commission had authority
to act “like a court of equity” and “consider[ ] public values beyond simply those enshrined in
the letter or encompassed in the spirit of the antitrust laws” when it “measur[ed] a practice
against the elusive, but congressionally mandated standard of fairness.”
214
And with that, the
Commission’s Section 5 authority reached as far as its backers had hoped, and its critics had
feared. The Commission now enjoyed unfettered discretion to proscribe “not only practices that
violate the Sherman Act and the other antitrust laws . . . but also practices that the Commission
determines are against public policy for other reasons.”
215
3
The Commission’s views on the meaning of “unfair methods of competition” have
similarly evolved in just the past few years. Our interpretation of the text does not resolve the
constitutional question,
216
but it further demonstrates the text’s broad reach and indeterminacy.
In 2015, a bipartisan majority of the Commission issued a policy statement announcing
the Commission’s understanding of the meaning of “unfair methods of competition.”
217
It
repeated the familiar mantra—Congress did not define the words, but left it to an “expert
administrative body” to “develop” the prohibition on a “flexible case-by-case basis” in the light
of “changing markets and business practices.”
218
But it adopted the pre-1970s understanding of
the law by declaring that “Section 5’s ban on unfair methods of competition encompasses not
only those acts and practices that violate the Sherman or Clayton Act but also those that
contravene the spirit of the antitrust laws and those that, if allowed to mature or complete, could
violate the Sherman or Clayton Act.”
219
It then announced that in “standalone” Section 5 cases,
the Commission would “adhere[ ]” to the Sherman Act’s well-established rule-of-reason analysis
that weighted a practice’s “harm to competition or the competitive process” against its
“associated cognizable efficiencies and business justifications.”
220
It also declared that it would
“be guided by the public policy underlying the antitrust laws, namely, the promotion of consumer
214
405 U.S. at 244.
215
FTC v. Ind. Fed’n of Dentists, 476 U.S. 447, 454 (1986) (citations omitted).
216
Whitman, 531 U.S. at 473 (“The idea that an agency can cure an unconstitutionally standardless delegation of
power by declining to exercise some of that power seems to us internally contradictory. . . . Whether the statute
delegates legislative power is a question for the courts, and an agency’s voluntary self-denial has no bearing upon
the answer.”).
217
Fed. Trade Comm’n, Statement of Enforcement Principles Regarding “Unfair Methods of Competition” Under
Section 5 of the FTC Act, 80 Fed. Reg. 57056 (Aug. 13, 2015) (“Chairwoman Ramirez and Commissioner Brill,
Commissioner Wright, and Commissioner McSweeny voting in the affirmative, and Commissioner Ohlhausen
dissenting.”).
218
Ibid.
219
Ibid.
220
Ibid.
27
welfare.”
221
In short, the Commission announced its view that the words of delegation, more or
less, recapitulated the command of the Sherman and Clayton Acts. This view remained the
Commission’s governing interpretation until 2021.
In 2021, after a substantial change in personnel, the Commission rescinded the 2015
Policy Statement on a party-line vote.
222
In her concurring opinion, the Chair explained that the
majority had rescinded the 2015 Policy Statement because it “largely wr[ote] the FTC’s
standalone authority out of existence” by limiting the reach of Section 5 to violations of other
antitrust statutes.
223
In the majority’s view, Section 5 imposed a mandate “broader than the
Sherman or Clayton Acts.”
224
The Commission, once again in a party-line vote, issued a new policy statement the next
year interpreting Section 5 as “broader than, and different from, the Sherman and Clayton
Acts.”
225
It reasoned that Congress rejected “a static definition” of the conduct Section 5
proscribed because such a definition “would soon become outdated,” and “Congress wanted to
give the Commission flexibility to adapt to changing circumstances.”
226
Section 5 therefore
“reaches beyond the Sherman and Clayton Acts to encompass various types of unfair conduct
that tend to negatively affect competitive conditions.”
227
The Commission proposed limiting the
reach of the word “unfair” by requiring that the conduct (1) “may be coercive, exploitative,
collusive, abusive, deceptive, predatory, or involve the use of economic power of a similar
nature,” and (2) “must tend to negatively affect competitive conditions.”
228
But even these
limitations seem to do little more than synonymize the statutory terms.
221
Ibid.
222
Lina M. Khan, Chair, Fed. Trade Comm’n, Joined by Comm’r Rohit Chopra and Comm’r Rebecca Kelly
Slaughter, Statement on the Withdrawal of the Statement of Enforcement Principles Regarding “Unfair Methods of
Competition” Under Section 5 of the FTC Act (July 1, 2021); Noah Joshua Phillips, Comm’r, Dissenting Remarks
Regarding the Commission’s Withdrawal of the Section 5 Policy Statement (July 1, 2021); Christine S. Wilson,
Comm’r, Dissenting Statement, Open Commission Meeting on July 1, 2021 (July 1, 2021).
223
Lina M. Khan, Chair, Fed. Trade Comm’n, Joined by Comm’r Rohit Chopra and Comm’r Rebecca Kelly
Slaughter, Statement on the Withdrawal of the Statement of Enforcement Principles regarding “Unfair Methods of
Competition” Under Section 5 of the FTC Act, 1 (July 1, 2021); see also id. at 2 (“By tethering Section 5 to the
Sherman and Clayton Acts, the 2015 Statement negates the Commission’s core legislative mandate, as reflected in
the statutory text, the structure of the law, and the legislative history, and undermines the Commission’s institutional
strengths.”).
224
Id. at 4.
225
Fed. Trade Comm’n, Policy Statement Regarding the Scope of Unfair Methods of Competition Under Section 5
of the Federal Trade Commission Act, 3 (Nov. 10, 2022) (hereinafter “2022 Section 5 Policy Statement”); see
Christine S. Wilson, Comm’r, Fed. Trade Comm’n, Dissenting Statement regarding the “Policy Statement Regarding
the Scope of Unfair Methods of Competition Under Section 5 of the Federal Trade Commission Act” (Nov. 10,
2022).
226
2022 Section 5 Policy Statement, supra note 225, at 3.
227
Id. at 1.
228
Id. at 9 (citations omitted).
28
C
Given the way the Commission and courts have interpreted Section 5, it is well-nigh
impossible to identify what “intelligible principle” Congress provided to constrain our discretion.
Section 5 does not “lay[ ] down policies and establish[ ] standards, while leaving to” the
Commission “the making of subordinate rules within prescribed limits and the determination of
facts to which the policy as declared by the legislature is to apply.”
229
It does not “establish
primary standards” and leave it to the Commission “‘to fill up the details’ under the general
provisions made by the Legislature.”
230
There is instead “an absence of standards for the
guidance of the [Commission’s] action,” making it “impossible . . . to ascertain whether the will
of Congress ha[d] been obeyed” in a particular rulemaking.
231
This is not an intelligible
principle; it “is delegation running riot.”
232
The Commission responds, however, that “any doubt” about Section 5’s constitutionality
was “laid … to rest in A.L.A. Schechter Poultry Corp. v. United States.”
233
The Commission
reasons that Schechter Poultry expressly blessed “unfair methods of competition” as a lawful
conferral of executive authority rather than an unconstitutional delegation of legislative power.
234
The Commission is correct that Schechter Poultry eliminates any doubt about the
constitutionality of our adjudicative authority under Section 5. But it confirms the
unconstitutionality of rulemaking authority under Section 5.
As part of its suite of New Deal legislation, Congress enacted the National Industrial
Recovery Act (“NIRA”) and authorized the President to approve “codes of fair competition”
with the force of law for various industries.
235
The Court held that the provision conferring the
229
Panama Refining Co. v. Ryan, 293 U.S. 388, 421 (1935).
230
Id. at 426 (quoting Wayman, 23 U.S. (10 Wheat.) at 43)).
231
Yakus v. United States, 321 U.S. 414, 426 (1944).
232
Schechter Poultry, 295 U.S. at 553 (Cardozo, J., concurring). Justice Thomas has long criticized the intelligible-
principle doctrine as inconsistent with the Constitution’s original meaning and incapable of restraining
unconstitutional delegations, and has urged its reconsideration. See, e.g., Whitman, 531 U.S. at 487 (Thomas, J.,
concurring) (“I am not convinced that the intelligible principle doctrine serves to prevent all cessions of legislative
power.”); Ass’n of Am. R.R., 575 U.S. at 66–87 (arguing that the modern application of doctrine is inconsistent with
the original meaning of the Vesting Clauses). Other Justices now agree with him. See Gundy, 588 U.S. at 149 (Alito,
J., concurring in the judgment) (“If a majority of this Court were willing to reconsider the approach we have taken
for the past 84 years [in nondelegation cases], I would support that effort.”); id. at 152–69 (Gorsuch, J., with whom
Roberts, C.J. joins, dissenting) (arguing that the intelligible-principle doctrine is inconsistent with the Constitution’s
original meaning and ought to be discarded). Justice Thomas is correct. The intelligible-principle test is impossible
to square with the original understanding of the Vesting Clause. Because granting rulemaking authority under
Section 5 fails the less-exacting intelligible-principle test, I do not address the original meaning of the Vesting
Clause here.
233
Final Rule, 89 Fed. Reg. at 38,354 (citing Schechter Poultry, 295 U.S. at 495).
234
Ibid.
235
See National Industrial Recovery Act, Pub. L. No. 73–67, § 3, 48 Stat. 195, 196–97 (June 16, 1933).
29
power on the President to promulgate “codes of fair competition” was an unconstitutional
delegation. The Court framed the question as whether the phrase “fair competition . . . refer[s] to
a category established in the law” that “limit[s]” the President’s “authority to make codes . . .
accordingly,” or whether it is “a convenient designation for whatever set of laws . . . the
President” concludes are “wise and beneficent provisions” for that industry.
236
“Unfair
competition” was a traditional common law category.
237
And although “[t]he codes” could,
“indeed, cover conduct which existing law condemns,” they were “not limited to conduct of that
sort.”
238
Instead, the only statutory guidance given to the President on the content of the codes
were the precatory “general purposes” laid out in the statute’s “Declaration of Policy.”
239
The
Court concluded that the delegation of the power to issue “codes of fair competition,” undefined
and untethered from the common law of “unfair competition,” granted the President “unfettered
discretion to make whatever laws he thinks maybe needed” to ensure fair competition.
240
As a
result, NIRA was an unconstitutional delegation of Congress’s lawmaking power to the
President.
NIRA sounds awfully familiar. “Unfair methods of competition” in the FTC Act is
similarly undefined and divorced from existing legal categories.
241
The Supreme Court has
described it as “a broad delegation of power,”
242
as “sweep[ing],” “flexib[le],” “elusive,”
243
and
as constrained only by the FTC Act’s general “aim” of “protect[ing] the public from the evils
likely to result from the destruction of competition or the restriction of it.”
244
Indeed, “unfair
methods of competition” is even more capacious than “unfair competition.”
245
If “unfair
competition” is not an intelligible principle, “unfair methods of competition” is not either.
The Schechter Poultry Court, plainly aware of what its analysis might portend for Section
5, distinguished Section 5 from NIRA by the method each statute used to accomplish its
objectives. Under the FTC Act, the Commission determined whether a practice was an “unfair
method of competition . . . in particular instances, upon evidence, in the light of particular
competitive conditions and of what is found to be a specific and substantial public interest.”
246
To make this determination, “Congress set up . . . a quasi judicial body” and provided “for
formal complaint, for notice and hearing, for appropriate findings of fact supported by adequate
236
Schechter Poultry, 295 U.S. at 531.
237
Ibid.
238
Id. at 532.
239
Id. at 534–35.
240
Id. at 537–38.
241
See supra Subsection III.B.2.
242
Atl. Refining Co., 381 U.S. at 367.
243
Sperry & Hutchinson Co., 405 U.S. at 241, 244.
244
Raladam Co., 283 U.S. at 647.
245
Id. at 648; see also R.F. Keppel & Bro., 291 U.S. at 311–12 & nn.2–3.
246
Schechter Poultry, 295 U.S. at 533.
30
evidence, and for judicial review.”
247
NIRA, by contrast, “dispense[d] with the” FTC Act’s
adjudicatory procedures and authorized “code-making”—the “legislative undertaking” of
announcing ex ante generally applicable rules of private conduct.
248
The FTC Act’s grant of
authority to the Commission was constitutional, then, because the authority was confined to
case-by-case fact-finding and adjudication.
The Commission concedes “that Schechter Poultry approved of the FTC Act’s
adjudicatory process for determining unfair methods of competition without commenting on the
Act’s rulemaking provision.”
249
But the distinction makes no difference, the Commission
reasons, because “[i]f Congress may permissibly delegate the authority to determine through
adjudication whether a given practice is an unfair method of competition, it may also permit the
Commission to do the same through rulemaking.”
250
That argument is wrong. It advances precisely the opposite rule from the one announced
in Schechter Poultry. The primary distinction the Court drew between the two was that the FTC
Act made the Commission an adjudicator, and NIRA made the President a code-maker. If
making rules about “unfair methods of competition” enjoyed the same constitutional status as
making determinations through adjudication, then NIRA’s code-making should have presented
no constitutional problem.
The Schechter Poultry Court also suggested a difference in the “subject-matter” covered
by the two acts.
251
The Court rejected the government’s argument that NIRA was like Section 5,
which “merely” granted the authority “to deal with ‘unfair competitive practices’ which offend
against existing law, and could be the subject of judicial condemnation without further
legislation, or to create administrative machinery for the application of established principles of
law to particular instances of violation.”
252
Instead, it held that NIRA was unconstitutional
because it gave the President the authority to prescribe in the codes any provision that “may tend
to effectuate” any of a litany of general purposes listed in NIRA, the literal meaning of “codes of
fair competition” notwithstanding.
253
And anything in those codes was also treated as an unfair
method of competition under the FTC Act.
254
247
Ibid.
248
Id. at 533, 541–43.
249
Final Rule, 89 Fed. Reg. at 38,354. I would not have expected the Schechter Poultry Court to have addressed “the
Act’s rulemaking provision” since no one at the time thought that Section 6(g) conferred substantive rulemaking
authority at all. See supra Section II.C.; see also Comm’r Holyoak Dissent, supra note 59, at Part I.
250
Final Rule, 89 Fed. Reg. 38354–55.
251
Schechter Poultry, 295 U.S. 533–34.
252
Id. at 535.
253
Ibid.
254
Id. at 534 (“[W]hen a code is approved, its provisions are to be the ‘standards of fair competition’ for the trade or
industry concerned, and any violation of such standards in any transaction in or affecting interstate or foreign
31
That is the power the Commission claims in the Final Rule—the power to determine the
meaning of “unfair methods of competition” through ex ante rules having the force and effect of
law, without regard for the distinguishing features of individual circumstances. And that power is
procedurally and substantively more akin to NIRA-style code-making than even the most
expansive characterization of Section 5 offered by the Schechter Poultry Court—the power “to
create administrative machinery for the application of established principles of law to particular
instances of violation.”
255
When combined with the Supreme Court’s and the Commission’s
expansive definitions of “unfair methods of competition” since the 1970s, the Final Rule
obliterates whatever “subject-matter” distinction the Court identified in 1935.
256
There is another problem with the Commission’s assertion that it is merely doing through
rulemaking what it could already do through adjudication. It is not “appl[ying] … established
principles of law to particular instances of violation.”
257
It is proscribing every single iteration of
a practice no matter its unique effects. We would not, and could not, proscribe every single
noncompete agreement in America through adjudication.
Consider the three enforcement actions the Commission ginned up the day before this
rulemaking began. The complaints in all three actions alleged that the noncompete agreements
were unfair methods of competition because of the unique effects they had on the particular
employees subject to them and on potential rivals trying to enter the product markets in which
the employers operated.
258
The Commission’s theory in those matters had nothing to do with the
“cumulative” effects of other noncompete agreements for other employees in unrelated
industries. The Commission was focused only on the specific noncompete agreements that were
the subject of the enforcement action and their specific effects.
commerce is to be deemed ‘an unfair method of competition’ within the meaning of the Federal Trade Commission
Act.”).
255
Id. at 535.
256
See supra Subsections III.B.2, 3.
257
Schechter Poultry, 295 U.S. at 535 (emphasis added).
258
See Analysis of Agreement Containing Consent Order to Aid Public Comment, In the Matter of Prudential Sec.,
Inc., et al., FTC File No. 2210026 (Jan. 4, 2023) at 5–6 (noting that the employees “earned low wages, were not
permitted to negotiate the terms of the Non-Compete Restrictions, and did not consult attorneys,” that they “did not
receive any money, job security, or other compensation in exchange for” the noncompete agreements, that the
company enforced the noncompete agreements to “discourage, delay, and prevent them from” accepting other jobs,
and that the noncompete agreements covered a “broad geographic area”); Complaint, In the Matter of Prudential
Sec., Inc., et al., FTC File No. 2210026 (Jan. 4, 2023) at 3–5 (employees had little bargaining power, earned low
pay, received no additional compensation for the noncompete agreements, and sometimes remained subject to them
even after the defendant had exited the market); Analysis of Agreements Containing Consent Order to Aid Public
Comment, In the Matter of O-I Glass, Inc. and In the Matter of Ardagh Grp. S.A., Ardagh Glass Inc., and Ardagh
Glass Packaging Inc., FTC File No. 2110182 (Jan. 4, 2023) at 5 (concluding that the specific agreements had an
anti-competitive effect because “the ability to identify and employ personnel with skill and experience in glass
container manufacturing is a substantial barrier to entry and expansion”).
32
That is a far cry from what the Final Rule does. The Final Rule generalizes about the
“cumulative” effect of all noncompete agreements without assessing the specific effects of any
particular agreement, and it treats them all as unfair methods of competition. That mode of
analysis bears no relation to our adjudicative procedures. Instead, it moves the Commission out
of the “quasi judicial” process blessed in Schechter Poultry and into the sort of “virtually
unfettered … code-making authority” that the Court condemned.
259
We are no longer making
determinations about unfairness “in particular circumstances, upon evidence, in the light of
particular competitive conditions.”
260
Instead, we are “dispens[ing] with this administrative
procedure and with any administrative procedure of an analogous character,” and legislating
unchecked.
261
That is precisely what Schechter Poultry forbade.
262
D
Even if one disagrees with my reading of Schechter Poultry, the constitutional question is
undoubtedly a close one. That alone is sufficient reason to reject the Final Rule.
Since at least the early nineteenth century, courts have applied an interpretive principle
we today call the “avoidance canon.”
263
In its earliest form, the canon instructed that “as between
two possible interpretations of a statute, by one of which it would be unconstitutional and by the
other valid, [the interpreter’s] plain duty is to adopt that which will save the Act.”
264
But since
the early twentieth century,
265
the scope of the canon expanded beyond interpretations that in fact
violated the Constitution.
266
Now, the canon teaches that “[w]hen the validity of an act of the
Congress is drawn in question, and even if a serious doubt of constitutionality is raised, it is a
259
Schechter Poultry, 295 U.S. at 533, 542.
260
Id. at 533.
261
Ibid.
262
The Commission suggests that the Administrative Procedure Act’s rulemaking procedures eliminate this concern
by imposing procedures on Section 5 rulemaking similar to the adjudicative procedures on which Schechter Poultry
distinguished NIRA and the FTC Act. Final Rule, 89 Fed. Reg. 38,354. This is a strange argument. First, Congress
did not adopt the APA until 1946, so even on the Commission’s theory, Section 5 rulemaking suffered from a serious
constitutional problem for the first three decades we had it. See Administrative Procedure Act of 1946, Pub. L. No.
79-404, 60 Stat. 237. Second, the APA’s procedures for informal rulemaking bear no resemblance to the “quasi
judicial” procedures that the Schechter Poultry Court said distinguished the FTC Act from NIRA. There is no
“complaint,” no “notice and hearing,” and no “findings of fact supported by adequate evidence” in an informal
rulemaking procedure. Compare Schechter Poultry, 295 U.S. at 533, with 5 U.S.C. § 553. The APA therefore does
not solve the nondelegation problem.
263
See Barrett, supra note 73, at 139 (“The unconstitutionality canon . . . seems to have emerged in 1814 and
matured by the late nineteenth century.”).
264
Blodgett v. Holden, 275 U.S. 142, 148 (1927); see also Clark v. Martinez, 543 U.S. 371, 395 (2005) (Thomas, J.,
dissenting) (“Traditionally, the avoidance canon . . . commanded courts, when faced with two plausible
constructions of a statute—one constitutional and the other unconstitutional—to choose the constitutional reading.”).
265
United States ex rel. Attorney General v. Del. & Hudson Co., 213 U.S. 366, 408 (1909).
266
Rust v. Sullivan, 500 U.S. 173, 190–91 (1991) (explaining distinction between the two versions of the avoidance
canon).
33
cardinal principle that this Court will first ascertain whether a construction of the statute is fairly
possible by which the question may be avoided.”
267
A court need conclude only that one of the
potential interpretations raises constitutional doubts in order to avoid that interpretation.
268
The Supreme Court has relied on nondelegation concerns to “giv[e] narrow constructions
to statutory delegations that might otherwise be thought to be unconstitutional.”
269
In Kent v.
Dulles, for example, the Secretary of State refused to issue a passport to a communist, invoking a
statute that authorized him to “grant and issue passports . . . under such rules as the President
shall designate and prescribe.”
270
The Court declined to “reach the question of constitutionality”
of the denial, and instead read the delegation narrowly to avoid giving the Secretary the power to
withhold a passport because of a person’s beliefs.
271
The Court has similarly adopted narrowing
constructions of other delegation statutes to avoid interpretations “that might be unconstitutional
under the Court’s reasoning in [Schechter Poultry].”
272
We should have done the same here. The canon “comes into play only when, after the
application of ordinary textual analysis, the statute is found to be susceptible of more than one
construction.”
273
Where the statute is unambiguous, the canon does not apply.
274
Sections 5 and
6(g) unambiguously do not confer on us the authority to proscribe all noncompete agreements,
both because Section 6(g) does not confer any rulemaking authority at all and because, even if it
did, Sections 5 and 6(g) together do not confer power to proscribe every single noncompete
agreement.
275
But even if one disagrees, the statutes are at least ambiguous. The suggestion that the text
of Sections 5 and 6(g) unambiguously authorize the Final Rule—especially after applying the
major-questions doctrine—is not a serious one. And the Commission has before it two potential
267
Ashwander v. Tenn. Valley Auth., 297 U.S. 288, 348 (1936) (Brandeis, J., concurring).
268
See, e.g., United States v. Palomar-Santiago, 593 U.S. 321, 328–29 (2021) (“Courts should indeed construe
statutes ‘to avoid not only the conclusion that [they are] unconstitutional, but also grave doubts upon that score.’”
(quoting United States v. Jin Fuey Moy, 241 U.S. 394, 401 (1916)); Eric S. Fish, Constitutional Avoidance as
Interpretation and as a Remedy, 114 Mich. L. Rev. 1275, 1281–82 (2016) (recounting development of the two
versions of the canon).
269
Mistretta, 488 U.S. at 373 n.7.
270
357 U.S. 116, 123 (1958).
271
Id. at 129 (“Since we start with an exercise by an American citizen of an activity included in constitutional
protection, we will not readily infer that Congress gave the Secretary of State unbridled discretion to grant or
withhold it.”).
272
Indus. Union Dep’t, AFL-CIO v. Am. Petroleum Inst., 448 U.S. 607, 646 (1980) (Occupational Safety and Health
Act); Nat’l Cable Television Ass’n, Inc. v. United States, 415 U.S. 336, 342 (1974) (reading the Federal
Communications Act “narrowly to avoid constitutional problems” under “the requirement of Schechter and
Hampton”).
273
Jennings v. Rodriguez, 583 U.S. 281, 296 (2018) (quotation marks and citation omitted).
274
Iancu v. Brunetti, 588 U.S. 388, 397 (2019) (The canon of constitutional avoidance “applies only when ambiguity
exists”).
275
See supra Part II; see also Comm’r Holyoak Dissent, supra note 59, at Part I.
34
interpretations of Sections 5 and 6(g) to resolve the ambiguity—one that authorizes a rule that, at
the very least, raises grave constitutional concerns under Schechter Poultry, and one that does
not. The avoidance canon compels us to choose the latter, under which we lack authority to issue
the Final Rule. But the Commission has chosen the former, and thereby blundered into an
unconstitutional reading of our organic statute.
IV
Thus far, my objections to the Final Rule have all been about the Commission’s power to
issue it. But even if you think I am reading the FTC Act and the Constitution incorrectly, the
Final Rule is still unlawful. Not only must an agency have power to issue a rule,
276
it must
comply with the substantive and procedural requirements of the Administrative Procedure Act
(APA).
277
“The APA ‘sets forth the procedures by which federal agencies are accountable to the
public and their actions are reviewed by the courts.’”
278
Relevant here, the APA “requires
agencies to engage in ‘reasoned decisionmaking,’ and directs that agency actions be ‘set aside’ if
they are ‘arbitrary’ or ‘capricious.’”
279
Under this standard, the agency must “examine the
relevant data and articulate a satisfactory explanation for its action including a ‘rational
connection between the facts found and the choice made.’”
280
“Courts enforce this principle with
regularity when they set aside agency regulations which, though well within the agencies’ scope
of authority, are not supported by the reasons that the agencies adduce.”
281
To satisfy the APA,
then, the Commission must demonstrate that the evidence on which it relies in fact supports the
conclusion it has reached—that every single noncompete agreement
282
in America is unfair and
anticompetitive in violation of Section 5 of the FTC Act.
The Final Rule falls woefully short of satisfying this requirement. Noncompete
agreements are not the type of agreements that the antitrust laws categorically proscribe. The
antitrust laws do not treat vertical restraints
283
as inherently anticompetitive. Rather, whether a
276
See supra Section II.A.
277
5 U.S.C. § 551 et seq.
278
Dep’t of Homeland Security v. Regents of the Univ. of Cal., 591 U.S. 1, 16 (2020) (quoting Franklin v.
Massachusetts, 505 U.S. 788, 796 (1992)).
279
Ibid. (quoting Michigan v. EPA, 576 U.S. 743, 750 (2015), and 5 U.S.C. § 706(2)(A)).
280
Motor Vehicle Mfrs. Ass’n of U.S. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983) (quoting Burlington
Truck Lines v. United States, 371 U.S. 156, 246 (1962); see also FCC v. Fox Television Stations, Inc., 556 U.S. 502,
513 (2009) (quoting Motor Vehicle Mfrs. Ass’n, 463 U.S. at 43).
281
Allentown Mack Sales & Servs. v. NLRB, 522 U.S. 359, 374 (1998).
282
I recognize that the Commission has exempted an indeterminate number of existing noncompete agreements
involving senior executives from the Final Rule’s ambit. 89 Fed. Reg. at 38,502–03. But the Commission concluded
that these agreements are unfair methods of competition. Id. at 38,411. It excepted those agreements from the Final
Rule “because of . . . practical considerations.” Ibid.
283
A vertical restraint is an agreement between persons or firms at different levels of a supply chain—for example,
between a supplier and a retailer. United States v. Topco Assocs., Inc., 405 U.S. 596, 608 (1972). Employee
35
vertical agreement violates the antitrust laws depends on its unique terms and effects. That is
how federal courts have always analyzed vertical noncompete agreements under the antitrust
laws.
284
This antitrust principle aligns with the common law, which has for centuries approached
the validity of noncompete agreements on a case-by-case, rather than a categorical, basis.
285
Given the overwhelming tradition of assessing noncompete agreements on a case-by-case basis,
the Commission’s categorical approach must be buttressed by very compelling evidence. But the
evidence does not support a categorical rule. Rather, it demonstrates what experience teaches:
that sometimes noncompete agreements have anticompetitive effects, and other times they have
procompetitive effects.
A
The Final Rule treats noncompete agreements as categorical, or per se, violations of
Section 5.
286
Within the context of the antitrust laws, that is a remarkable thing. The antitrust
noncompete agreements are vertical agreements between the supplier of labor—the employee—and the purchaser of
labor—the employer.
284
See, e.g., Consultants & Designers, Inc. v. Butler Serv. Grp., Inc., 720 F.2d 1553, 1561–64 (11th Cir. 1983)
(rejecting application of per se rule to employee noncompete agreements and analyzing the particular effects of the
agreement under the rule of reason); Aydin v. Loral Corp., 718 F.2d 897, 900 (9th Cir. 1983) (“Employee covenants
not to compete or interfere with the employer’s business after the end of the employment relationship should not be
tested under the per se rule.”); Lektro-Vend Corp. v. Vendo Co., 660 F.2d 255, 264–69 (7th Cir. 1981); (applying rule
of reason to employee noncompete agreements); United States v. Empire Gas Corp., 537 F.2d 296, 308 (8th Cir.
1976) (“The difficulty with the government’s position is that there is no showing that any [noncompete agreement]
was unreasonable so as to violate the Sherman Act. . . . We agree with the district court’s conclusion that the mere
existence of a large number of covenants not to compete does not establish a 15 U.S.C. § 1 violation.”); Newburger,
Loeb & Co., Inc. v. Gross, 563 F.2d 1057, 1082–83 (2d Cir. 1977) (refusing to treat employee noncompete
agreements as per se Section 1 violations, instead analyzing their particular effects under the rule of reason); Golden
v. Kentile Floors, Inc., 512 F.2d 838, 843–46 (5th Cir. 1975) (same); Bradford v. N.Y. Times Co., 501 F.2d 51, 59–60
(2d Cir. 1974) (applying rule of reason to noncompete agreement and refusing ‘the invitation to classify such a
restraint as a per se violation of the Sherman Act”); see also Snap-On Tools, 321 F.2d at 836–37 (rejecting FTC
enforcement action against noncompete agreement between dealer and supplier on ground that it was not
unreasonable in its particular effects).
285
See infra Section I.B.
286
I recognize that the dichotomy between per se rules and the rule of reason arose in the context of Section 1of the
Sherman Act, 15 U.S.C. § 1. And I agree with the majority that Section 5 proscribes more conduct than is proscribed
by Sections 1 and 2 of the Sherman Act. Final Rule, 89 Fed. Reg. at 38,348. But that does not mean that when we
are dealing with Section 5 we throw out the window more than a century of Sherman Act precedent. Quite the
opposite. Courts have consistently linked Section 5 to the rest of the antitrust laws. See, e.g., Motion Picture Advert.
Serv. Co., 344 U.S. at 394–95 (“[T]he Federal Trade Commission Act was designed to supplement and bolster the
Sherman Act and the Clayton Act—to stop in their incipiency acts and practices which, when full blown, would
violate those Acts, as well as to condemn as ‘unfair method of competition’ existing violations of them.” (internal
citations omitted)); see E.I. du Pont de Nemours & Co. v. FTC, 729 F.2d 128, 136 (2d Cir. 1984) (“Congress’ aim
was to protect society against oppressive anti-competitive conduct and thus assure that the conduct prohibited by the
Sherman and Clayton Acts would be supplemented as necessary and any interstices filled.”); Chuck’s Feed & Seed
Co. v. Ralston Purina Co., 810 F.2d 1289, 1292–93 (4th Cir. 1987) (“[T]he FTC Act functions as a kind of penumbra
around the federal antitrust statutes. . . . [T]he scope of the FTC is . . . linked to the antitrust laws.”). “When conduct
36
laws have long distinguished between restraints subject to categorical prohibition—the per se
rule—and those subject to a case-by-case analysis—the rule of reason.
287
The rule of reason, or
restraint-specific approach, “presumptively applies” to every restraint.
288
Under the rule of
reason, “the factfinder weighs all of the circumstances of a case in deciding whether a restrictive
practice should be prohibited as imposing an unreasonable restraint on competition.”
289
That
means that the default antitrust analysis in American law involves a highly individualized inquiry
into the nature and effects of the “particular contract or combination.”
290
That is precisely the
approach that the common law, and a large majority of States, have applied to noncompete
agreements for centuries.
291
There are exceptions to this case-by-case approach. “Some types of restraints … have
such predictable and pernicious anticompetitive effect, and such limited potential for
procompetitive benefit, that they are deemed unlawful per se.”
292
“The per se rule, treating
categories of restraints as necessarily illegal, eliminates the need to study the reasonableness of
an individual restraint in light of the real market forces at work. . . .”
293
But per se treatment is
exceedingly rare. “It is only after considerable experience with certain business relationships that
courts classify them as per se violations of the Sherman Act.”
294
That “considerable experience”
must be sufficient for courts to “predict with confidence that [the restraint at issue] would be
invalidated in all or almost all instances under the rule of reason”; that is, that the restraint in
almost every iteration would be “manifestly anticompetitive” and “lack[ing] any redeeming
value.”
295
Given the complexity of business relationships and the unique effects those
relationships may have in different contexts, courts are “reluctan[t] to adopt per se rules with
regard to ‘restraints . . . where the economic impact of certain restraints is not immediately
obvious.’”
296
“Reluctan[t]” is understating it. Only horizontal restraints—price-fixing and market
allocation—are subject to per se categorization.
297
Courts once categorically prohibited some
does bear the characteristics of recognized antitrust violations . . . the Commission may properly look to cases
applying those laws for guidance.” Atlantic Refin. Co. v. FTC, 381 U.S. 357, 369–70 (1965).
287
Ohio v. Am. Express Co., 585 U.S. 529, 540 (2018); see also Standard Oil Co. of N.J. v. United States, 221 U.S. 1,
60 (1911) (announcing the rule of reason).
288
Texaco Inc. v. Dagher, 547 U.S. 1, 5 (2006).
289
Cont’l T.V., Inc. v. GTE Sylvania, Inc., 433 U.S. 36, 49 (1977).
290
Texaco, 547 U.S. at 5.
291
See supra Section I.B.
292
State Oil Co. v. Khan, 522 U.S. 3, 10 (1997).
293
Leegin Creative Leather Prods., Inc. v. PSKS, Inc., 551 U.S. 877, 886 (2007).
294
Topco Assocs., Inc., 405 U.S. at 607–08.
295
Leegin, 551 U.S. at 886–87 (cleaned up).
296
Khan, 552 U.S. at 10 (quoting Ind. Fed’n of Dentists, 476 U.S. at 458–459).
297
Copperweld Corp. v. Indep. Tube Corp., 467 U.S. 752, 768 (1984) (“Certain agreements, such as horizontal price
fixing and market allocation, are thought so inherently anticompetitive that each is illegal per se without inquiry into
37
vertical agreements too, but the Supreme Court has gotten rid of those precedents.
298
And one
court of appeals recently held that even horizontal price-fixing is subject to the rule of reason if
the two parties to the restraint were in a dual-distribution relationship.
299
Per se treatment in our
law is rare, and has gotten substantially rarer over the last thirty years.
The Final Rule charges headlong in the opposite direction by adding a third per se rule to
American antitrust law: noncompete agreements. It is the only vertical agreement in the bunch.
The question, then, is where we acquired the experience sufficient to justify departing so
radically from where courts have been taking the rest of antitrust law. The courts have disclaimed
sufficient experience to treat noncompete agreements as per se illegal.
300
The Commission has
very little experience with noncompete agreements.
301
Had it not cooked up some consent orders
on the eve of this rulemaking, the Commission would have no experience at all.
302
The States
have substantial experience with noncompete agreements. Their experience suggests that
noncompete agreements are often fair and procompetitive and should be permitted.
303
The Final
Rule reaches precisely the opposite conclusion.
The Commission justifies the Final Rule by relying on academic papers. A handful of
economic and sociological studies, it contends, demonstrates that noncompete agreements are
universally unfair and anticompetitive.
304
But the evidence on which the Commission relies is
nowhere near sufficient to justify this sweeping rule.
B
1
The Final Rule concedes that “some empirical evidence” reveals that noncompete
agreements “increase investment in human capital of workers, capital investment, and [research
and development] investment.”
305
Indeed, “the only study that attempts to identify the causal link
between non-competes and worker human capital investment” found that making noncompete
the harm it has actually caused.”); D. Francis & C. Sprigman, Antitrust: Principals, Cases, and Materials 274 (2d ed.
2024) (“Thus, today, with an asterisk for tying arrangements, no vertical restraints are per se illegal.”).
298
See Dr. Miles Med. Co. v. John D. Park & Sons Co., 220 U.S. 373 (1911) (categorically prohibiting minimum
resale price restraints), overruled by Leegin, 551 U.S. at 882, 907; Albrecht v. Herald Co., 390 U.S. 145 (1968)
(categorically prohibiting maximum resale price restraints), overruled by Khan, 552 U.S. at 7, 18, 22; United States
v. Arnold, Schwinn & Co., 338 U.S. 365 (1967) (categorically prohibiting vertical nonprice restraints), overruled by
GTE Sylvania, 433 U.S. at 58–59.
299
United States v. Brewbaker, 87 F.4th 563, 582–83 (4th Cir. 2023).
300
See, e.g., Bradford, 501 F.2d at 60 (“There is certainly a total absence of federal Sherman Act experience” with
noncompete agreements.).
301
See supra Section I.C.
302
Ibid.
303
See supra Section I.B.
304
Final Rule, 89 Fed. Reg. at 38,379–80, 38,388–89.
305
Id. at 38,422.
38
agreements unenforceable would substantially decrease companies’ investments in employee
training.
306
Another study found that noncompete “enforceability increases [research and
development] expenditure.”
307
These results are unsurprising. Economic theory predicts that, at
least in some circumstances, noncompete agreements promote an employer’s investment in its
employees by mitigating the risk that a rival will ride freely on those investments by luring the
employee away before the investing employer can recoup the return on those investments.
308
Investments in employee training and research and development are neither unfair nor
anticompetitive. They are critical to the success of employers, employees, and the general
economy. And some noncompete agreements promote both. These findings ought to be fatal to a
categorical rule that rests on the proposition that every noncompete agreement is unfair and
anticompetitive.
The Commission waves away these procompetitive effects, arguing that other
arrangements—namely, nondisclosure agreements (NDAs), intellectual property rights, and
invention-assignment agreements—will seamlessly fill the gaps left by the elimination of
noncompete agreements.
309
But the Final Rule provides no evidence to support this sweeping
conclusion. It merely declares ipse dixit that these alternatives will be just as good as
noncompete agreements.
310
It can make this sweeping declaration, it reasons, because a few
States have banned noncompete agreements and their economies seem to be doing fine.
311
That is
a slender reed on which to rest a rule obliterating thirty million contracts. The existence of a few
States with noncompete bans does not prove that alternatives can protect all noncompete
agreements’ potential procompetitive effects. For one thing, only two of the States that
purportedly ban noncompete agreements ban them anywhere near to the extent the Commission
does in the Final Rule.
312
And, in practice, those two States do not go nearly as far as the Final
Rule because, notwithstanding their bans, noncompete agreements remain common and
306
Id. at 38,423 (citing Evan Starr, Consider This: Training, Wages, and the Enforceability of Covenants Not to
Compete, 72 I.L.R. Rev. 783, 796–97 (2019)).
307
Ibid. (citing Matthew S. Johnson, Michael Lipsitz, & Alison Pei, Innovation and the Enforceability of
Noncompete Agreements (Nat’l. Bureau of Econ. Rsch., Working Paper No. 31487 2023)).
308
John McAdams, Noncompete Agreements: A Review of the Literature, Fed. Trade Comm’n Bureau of Econ.
Rsch. Paper 6 (2019) (“Non-compete agreements are one arrangement that can mitigate the hold-up problem . . . by
discouraging worker attrition before the firm has had time to recoup the cost of its upfront investment, and thus
permit firms to make investments in its workers that are mutually beneficial and that it otherwise may not decide to
do.”).
309
Final Rule, 89 Fed. Reg. at 38,424–33.
310
Id. at 38,424.
311
Id. at 38,424–25.
312
Barnett & Sichelman, supra note 34, at 1011–14 (2020) (identifying California and North Dakota and pointing
out that empirical studies have radically oversimplified other state regulatory regimes).
39
enforceable in some circumstances.
313
Further, this approach fails to account for other factors
that drive economic development and innovation in the States. For example, the Commission
argues that technology startups have thrived in California notwithstanding its ban on noncompete
agreements, “suggesting that employers have less restrictive alternatives for protecting trade
secrets.”
314
But this argument is just the post hoc fallacy. We have no idea what the California
economy would look like if it permitted noncompete agreements. And there are too many
confounding variables to account for any causation between the “less restrictive alternatives” and
the success of Silicon Valley.
Not only does the Commission fail to establish that alternative arrangements will
adequately replace noncompete agreements, the Final Rule, in fact, contains evidence that the
proposed alternatives may not be adequate replacements for noncompete agreements. The Final
Rule cites a number of studies that attempted to isolate noncompete agreements’ effects. If
alternatives could effectively replicate the benefits of noncompete agreements, these studies
should have shown no relationship between noncompete agreements and procompetitive
benefits. But, as the Final Rule acknowledges, that is not the case. Studies show that,
notwithstanding the purported alternatives, the use of noncompete agreements “increase[s]
investment in human capital of workers, capital investment, and [research and development]
investment.”
315
There is good reason to believe this evidence and doubt that alternative measures would
adequately preserve the procompetitive benefits of noncompete agreements. The proposed
alternatives are behavioral. That is, they preclude an employee from engaging in certain
behaviors with his or her former employer’s rival, but they do not prevent him or her from
working for that rival. Noncompete agreements, by contrast, are structural. They prevent former
employees with unique insight into a company from joining a competitor in the first place.
The Commission knows full well that structural protections are almost always more
effective than behavioral restrictions. When enforcing the antitrust laws, the Commission
typically rejects behavioral remedies and insists on more effective structural relief. In 2022, the
Commission told Congress that “we now strongly disfavor behavioral remedies and will not
hesitate to reject proposed divestitures that cannot fully cure the underlying harm.”
316
Indeed,
earlier that year, the Commission rejected a proposed behavioral remedy and blocked a vertical
313
Final Rule, 89 Fed. Reg. at 38,466–67 (“[W]orkers from States where non-competes are banned commented that
they faced enforcement of non-competes that selected the law of another State,” and “Colvin and Shierholz find that
45.1% of workplaces in California use non-competes even though they are unenforceable there.”).
314
Final Rule, 89 Fed. Reg. at 38,396.
315
Id. at 38,422.
316
Prepared Statement of the Fed. Trade Comm’n Before the United States Senate Committee on the Judiciary
Subcommittee on Antitrust, Competition Policy and Consumer Rights “Oversight of the Enforcement of the
Antitrust Laws,” 6 (Sept. 20, 2022).
40
merger.
317
The Commission’s preference for structural relief rather than behavioral relief reflects
straightforward concerns. Behavioral remedies are difficult to craft; violations are difficult to
detect; and the remedies are complicated and difficult to enforce.
318
The meager evidentiary
record in the Final Rule does not justify the dismissal of commenters’ analogous concerns about
the behavioral alternatives to noncompete agreements.
319
2
The evidentiary record is also far from conclusive on the potential benefits that a ban on
noncompete agreements would bring. The Commission’s principal justification for the Final Rule
is that banning noncompete agreements is likely to increase employees’ wages substantially.
320
But this claim is overstated. The Final Rule’s chief evidence for this claim is an unpublished
study which found that employees’ wages rose after restrictions on noncompete-agreement
enforceability increased.
321
The study analyzed changes to noncompete-agreement enforceability
317
Statement Regarding Termination of Lockheed Martin Corporation’s Attempted Acquisition of Aerojet
Rocketdyne Holdings Inc., Fed. Trade Comm’n Press Release (Feb. 15, 2022) (announcing termination of
Lockheed/Aerojet transaction); Lockheed Martin Reports Fourth Quarter And Full Year 2021 Financial Results,
Lockheed Martin News Release (Jan. 25, 2022), https://news.lockheedmartin.com/2022-01-25-Lockheed-Martin-
Reports-Fourth-Quarter-and-Full-Year-2021-Financial-Results (describing the FTC’s rejection of Lockheed Martin’s
proposed behavioral remedies).
318
Joint Letter from Dep’t of Just. AAG Jonathan Kanter and Fed. Trade Comm’n Chair Lina Khan to Canada
Ministry of Innovation, Science, and Industry, 3 (Mar. 31, 2023) (“Given the difficulties in crafting behavioral
remedies that adequately anticipate corporate incentives and reflect dynamic market realities as well as the
challenges of enforcing these provisions, behavioral remedies are particularly disfavored by the agencies”).
319
Comment by American Investment Council, FTC-2023-0007-21042, at 20 n.79 (“Even where it is possible to
demonstrate improper sharing of confidential information, at that point the genie cannot be put back into the bottle
and damages often cannot be reasonable calculated. Noncompete clauses seek to prevent the individual subject to it
from being in a position where he or she could intentionally or unintentionally disclose confidential information.
This is theoretically similar to the FTC’s preference for structural remedies in merger cases. Keeping parties apart
polices interaction better.”); Comment by HR Policy Association, FTC-2023-0007-20998, at 14 (“In the case of a
senior executive who is responsible for overall corporate strategy or product design, it is almost impossible to
‘ringfence’ the proprietary information that would do competitive harm if that executive immediately joined a direct
competitor with no cooling-off period.”); Comment by National Association of Manufacturers, FTC-2023-0007-
20939, at 4 (“NDAs do not address the reality that it is simply unrealistic to expect that employees will mentally
separate the trade secrets they have developed expertise on when they go to a competitor. The current remedies for
misappropriation of trade secrets under state laws . . . do not prevent misappropriation of trade secrets, but rather
only offer a remedy after the misappropriation has occurred. Noncompete agreements are, therefore, a necessary and
important complement to non-disclosure agreements and trade secret litigation for companies attempting to protect
their proprietary scientific, technical, and business information.”).
320
Final Rule, 89 Fed. Reg. at 38,474 (citing “an increase in earnings or earnings growth” as the first of the
“Benefits of the Rule”); id. at 38,460 (noting that “non-competes suppress wages for workers across the labor force”
and rejecting an alternative disclosure rule that would not achieve the same results); id. at 38,382–87 (section titled,
“Non-Competes Suppress Workers’ Earnings”); id. at 38,410–11 (section titled, “Non-Competes with Senior
Executives Suppress Labor Mobility and Earnings”).
321
Id. at 38,382 (discussing Matthew S. Johnson, Kurt J. Lavetti & Michael Lipsitz, The Labor Market Effects of
Legal Restrictions on Worker Mobility (Nat’l Bureau of Econ. Rsch., Working Paper No. 31929, 2023). The Final
41
that fall well short of the categorical ban the Final Rule imposes. Specifically, the study ranked
noncompete-agreement enforceability on a scale from 0.0 to 1.0.
322
The categorical ban that the
Final Rule imposes would be a dramatic shift for some States, taking them from near 1.0 to 0.0
based on the study’s methodology.
323
This is a far cry from the minor changes that the study
examined. Nearly all of the shifts in noncompete enforceability that it considered had “a score
change of 0.15 or less,” and the changes largely preserved most noncompete enforceability, with
seventy-five percent of the sample having a post-change enforceability score greater than 0.65.
324
This study therefore suggests that restrictions narrower than a complete ban could achieve the
wage increases the Commission says it wants, while preserving the procompetitive benefits the
Commission acknowledges exist. But the Commission nevertheless fires its blunderbuss and
bans them all—hardly the sort of “rational” response to this study that the APA requires.
The unpublished study also suffers from a key weakness. Even assuming that
noncompete agreements will produce short-run wage increases, it does not necessarily follow
that the unenforceability of noncompete agreements will benefit employees in the future.
Noncompete agreements help solve the “hold-up” problem. “Hold-up” describes the
phenomenon in which a firm may “forgo making certain investments in [its] workforce knowing
that [the] employees would be able to . . . quit and appropriate the value of the investment.”
325
Noncompete agreements mitigate the hold-up problem “by discouraging worker attrition before
the firm has had time to recoup the cost of its upfront investment, and thus permit firms to make
investments in its workers that are mutually beneficial and that it otherwise may not decide to
do.”
326
As of today, employers may have invested in training their employees, secure in the
knowledge that a noncompete agreement would prevent these employees from taking their new
Rule also invokes a study on the effects of the unenforceability of noncompete agreements for tech workers in
Hawaii. See id. at 38,382–83 (citing Natarajan Balasubramanian, Jin Woo Chang, Mariko Sakakibara, Jagadeesh
Sivadasan, & Evan Starr, Locked In? The Enforceability of Covenants Not to Compete and the Careers of High-Tech
Workers, 57 J. Hum. Res. S349, S351 (2022)). Whatever the merits of that study, its subject is so entirely
idiosyncratic that it cannot plausibly be used as evidence to sustain a ban on every single noncompete agreement in
America. The Final Rule further relies on a study of the unenforceability of noncompete agreements for “hourly
workers with relatively low earnings” in Oregon. See id. at 38,381 (citing Michael Lipsitz & Evan Starr, Low-Wage
Workers and the Enforceability of Noncompete Agreements, 68 Mgmt. Sci. 143, 144 (2022)). While this paper is a
valuable contribution to the literature, it examines a non-retroactive change affecting a limited population in a
limited area, which forms a poor basis for a national rule applying to the whole population. In addition, there are
questions about whether the paper’s results are confounded by macroeconomic changes from the Great Recession
and whether the timing of changes in mobility relative to changes in wages matches the authors’ interpretation. See
McAdams, supra note 308, at 17–18.
322
See Johnson, Lavetti & Lipsitz, supra note 321, at 10.
323
See ibid. (“Florida has the highest NCA Enforceability Score during our time period (which we normalize to 1),
and North Dakota has the lowest score (which we normalize to 0).”).
324
Id. at 11.
325
John M. McAdams, Non-Compete Agreements: A Review of the Literature 6 (SSRN, Dec. 31, 2019).
326
Ibid; see also Office of Econ. Pol’y, U.S. Dep’t of Treasury, Non-compete Contracts: Economic Effects and
Policy Implications 7–8 (Mar. 2016).
42
skills to a competitor. But with noncompete agreements no longer enforceable, employers may
make fewer similar training investments.
327
That decrease in training may undermine
productivity. The Final Rule discusses no empirical literature that examines the effects of
noncompete agreements on labor productivity. I am hard pressed to believe that less productive,
undertrained employees will be able to command higher wages in the long run than those in
whom firms have invested heavily in training.
The Commission’s claim that forbidding noncompete agreements will boost innovation is
also weak.
328
In support of this claim, the Final Rule cites three unpublished studies linking an
increase in patent filings to the unenforceability of noncompete agreements.
329
But patents are
not a good measure of innovation. First, many patent filings do not reflect any innovation at
all,
330
as the Commission has explained more than once.
331
Indeed, patents sometimes suppress
innovation rather than promote it.
332
Second, innovation does not always involve patents.
333
Finally, if firms cannot use noncompete agreements to protect their intellectual property, they
might use patents instead. This increase in patent filings would lead to a positive association
between patents and a ban on noncompete agreements but would tell us nothing about whether
firms are innovating any more or less than they were before the ban on noncompete
agreements.
334
While the papers on which the Final Rule relies attempted to account for some of
the problems with using patents as a proxy for innovation,
335
their success, and the overall link
between noncompete agreements and innovation, remains unsettled.
327
The Final Rule discusses the possibility that noncompete agreements may increase only “core” (basic) rather than
“advanced” training, because experienced workers might leave the industry due to noncompete agreements, which
would increase the need for basic training of new workers. See Final Rule, 89 Fed. Reg. at 38,487. But there does
not appear to be any evidence to support whether or not increased training associated with noncompete agreements
is core or advanced. The mere possibility that some training is inefficient should not be used to dismiss the increase
in an outcome (training) that is generally regarded as positive and as the primary benefit of noncompete agreements.
328
Id. at 38,394.
329
Ibid. (citing Zhaozhao He, Motivating Inventors: Non-Competes, Innovation Value and Efficiency 21 (SSRN,
2023); Johnson, Lipsitz & Pei, supra note 307; and Emma Rockall & Kate Reinmuth, Protect or Prevent? Non-
Compete Agreements and Innovation (SSRN, 2023)).
330
Barnett & Sichelman, supra note 34, at 1024.
331
See, e.g., Fed. Trade Comm’n Statement Concerning Brand Drug Manufacturers’ Improper Listing of Patents in
the Orange Book (Sep. 14, 2023).
332
Fed. Trade Comm’n, To Promote Innovation: The Proper Balance of Competition and Patent Law and Policy, at 5
(Oct. 2003) (“[P]oor quality or questionable patent[s] . . . can block competition and harm innovation in several
ways.”).
333
See Kewanee Oil Co. v. Bicron Corp., 416 U.S. 470, 483 (1974) (“[N]o patent is available for a discovery,
however useful, novel, and nonobvious, unless it falls within one of the express categories of patentable subject
matter of 35 U.S.C. § 101”).
334
See Barnett & Sichelman, supra note 34, at 1023–24.
335
In an attempt to address the fact that patents do not always reflect innovation, some papers assign more weight to
arguably higher relevance patents in various ways. See He, supra note 329, at 4 (using the value of each patent
measured by stock market reactions); Johnson, Lipsitz, & Pei, supra note 307, at 5 (assigning more importance to
43
3
The Final Rule also draws arbitrary lines. By and large, the Final Rule seeks to identify
the average effects of noncompete agreements. This approach creates two problems. First,
nothing in Section 5 forbids agreements that are themselves fair and procompetitive simply
because similar agreements in other contexts may be unfair and anticompetitive. By focusing on
averages, the Commission is necessarily, and arbitrarily, capturing in its dragnet perfectly fair
and procompetitive agreements in order to avoid “administrability concerns” about drawing lines
short of a complete ban.
336
Those agreements are outside of our Section 5 authority to proscribe
no matter how similarly they look to other agreements that may be unfair or anticompetitive.
337
Second, the Commission’s focus on averages elides critical details. There are likely particular
industries or employee types for which the procompetitive benefits associated with noncompete
agreements cannot be achieved by other means. The Commission notes that one paper estimates
“34%–39% increases in capital investment due to increases in noncompete enforceability at
knowledge-intensive firms.”
338
But the Commission estimates an increase of “7.9% across all
sectors.”
339
This wide variation shows that studies that aggregate results from many sectors could
miss concentrated effects, including harm, that a blanket ban might have on particular subgroups.
For example, commenters argued that a noncompete-agreement ban would deter investment that
broadcasters make “in on-air talent for news, sports, weather, and other programming,”
340
which
might hurt employees’ efforts to build the skills and reputation needed to jump to larger
television markets.
341
Even if the broad study cited in the Final Rule
342
proved that a
patents that have more forward citations by other patents in order to identify more important patents); Rockall &
Reinmuth, supra note 329, at 12 (assigning more importance to patents with fewer backward citations). Papers also
take different approaches to try to address concerns that noncompete agreement bans merely increase firms’
propensity to patent innovations rather than increasing actual innovation. One study examines propensity to patent
by replicating the overall results for the medical device and pharmaceutical industries considered in isolation, on the
theory that these industries are highly unlikely to have unpatented innovations. Johnson, Lipsitz & Pei, supra note
307, at 5. Though this result is intriguing, it is only for two industries in one paper and only reflects the effects on
patenting and not other forms of innovation. Another study assumes that product innovations are more likely than
process innovations to rely on patents due to greater concerns that competitors could reverse engineer products.
Rockall & Reinmuth, supra note 329, at 25. They find the ratio of process to product patents does not change with a
change in non-compete enforceability, which could suggest a stability of propensity to patent.
336
Final Rule, 89 Fed. Reg. at 38,458.
337
See, e.g., Boise Cascade Corp. v. FTC., 637 F.2d 573, 577 (9th Cir. 1980) (reversing the Commission’s finding of
FTC Act Section 5 liability and declining to enforce its order because “[d]espite these similarities [with conduct
previously found to violate Section 5], neither the Commission nor the administrative law judge (ALJ) purported to
find that the pricing system here was used with the price-stabilizing purpose and effect [as the prior conduct]”).
338
Final Rule, 89 Fed. Reg. at 38,423 n.746 (emphasis added) (citing Jessica S. Jeffers, The Impact of Restricting
Labor Mobility on Corporate Investment and Entrepreneurship, 37 Rev. Fin. Stud. 1, 28–29 (2024)).
339
Ibid. (emphasis added).
340
Comment by the National Association of Broadcasters, FTC-2023-0007-16268, at 2.
341
Comment by Block Communications, FTC-2023-0007-21041, at 4–5.
342
Final Rule, 89 Fed. Reg. at 38,382 (discussing Johnson, Lavetti & Lipsitz, supra note 321).
44
noncompete-agreement ban would increase average employee wages in the long run—and I do
not believe that it does—it cannot demonstrate whether the ban would specifically benefit or hurt
an extremely narrow group such as on-air broadcast talent.
Moreover, the Final Rule’s lone exception—for existing noncompete agreements in
which the employee is a senior executive—is likely the most arbitrary aspect of it all. The
Commission concedes that noncompete agreements “with senior executives may tend to
negatively affect competitive conditions in product and service markets to an even greater degree
than [noncompete agreements] with other workers, given the outsized role senior executives play
in forming new businesses and setting the strategic direction of firms with respect to
innovation.”
343
On the Commission’s theory, then, the only agreements it leaves in force are
more grievous violations of Section 5 than the agreements it nullifies. The Commission exempts
them from the Final Rule because of “practical concerns” about unwinding existing agreements,
but it concedes that similar concerns apply to other agreements.
344
The Commission’s careful
consideration of the effects of a particular species of noncompete agreement, while ignoring the
potentially unique effects of countless other varieties, demonstrates the Final Rule’s
overinclusion and arbitrariness.
4
The lack of evidence on noncompete agreements’ effects on innovation, wages,
productivity, and on the effects of potential alternatives to those agreements, points to a broader
problem that infects the Commission’s analysis: the economic literature on noncompete
agreements is new and still being developed. Indeed, one of our own economists pointed this out
just a few years ago, writing that “[f]urther research is needed in several areas” because “the
existing empirical literature on non-compete agreements suffers from several important
limitations that raise questions as to whether it has successfully estimated the causal effect of
such agreements on mobility, wages, entrepreneurship, and innovation.”
345
And just a few years
before that, one of the scholars on which the Commission heavily relies warned that “the data we
have currently is woefully inadequate and more research is needed to reach meaningful
conclusions about reforms.”
346
The Final Rule does not even cite these contrary views.
343
Final Rule, 89 Fed. Reg. at 38,404; see also id. at 38,407 (Noncompete agreements “for senior executives are
especially pernicious” because “[s]enior executives are relatively few in number, are bound by [noncompete
agreements] at high rates, and have highly specialized knowledge and skills. Therefore, it can be extremely difficult
for existing firms and potential new entrants to hire executive talent and to form the most productive matches.”).
344
Id. at 38,412.
345
McAdams, supra note 308, at 20.
346
Norman Bishara & Evan Starr, The Incomplete Noncompete Picture, 20 Lewis & Clark L. Rev. 497, 546 (2016).
45
Most of the papers that the Final Rule relies upon are either unpublished or were
published within the last five years.
347
Some are contradictory. For example, the two papers on
the only subject that has been examined by more than one group of authors—the effect of
noncompete agreements in financial advisory firms—reached opposite conclusions.
348
To be
sure, our knowledge of the effects of noncompete agreements has improved in recent years. But
we do not have anything close to the full picture of the procompetitive and potentially
anticompetitive effects of noncompete agreements. “[W]ithout the full picture of noncompete use
within and across firms, better measures of enforceability, worker perceptions, and employer
motivations, policymakers are still largely in the dark about what reforms, if any, are needed.”
349
Given that this literature is currently the subject of extensive ongoing research, the Commission
should not, and I believe cannot, prematurely use it to justify a wide-sweeping policy decision
that will foreclose the opportunity to further study the effects of noncompete agreements on the
American economy.
The APA “require[s] final rules to ‘articulate a . . . rational connection between the facts
found and the choices made.’”
350
It is possible that the evidence on which the Commission relies
may have justified a modest rule aimed at the specific industries or professions studied in the
academic literature. Instead, for reasons known only to the majority, the Commission decided to
go for broke and ban every single noncompete agreement it could find. The APA requires that a
vast trove of evidence sustain the vast sweep of the Final Rule. The handful of academic papers
cited in the Final Rule cannot justify its incredible reach, and relying on them to prohibit
noncompete agreements categorically is a “clear error of judgment.”
351
***
Because we have neither the authority nor the evidence to sustain the Final Rule, it is
unlawful. I respectfully dissent.
347
See, e.g., Johnson, Lavetti & Lipsitz, supra note 321.
348
See Umit G. Gurun, Noah Stoffman, & Scott E. Yonker, Unlocking Clients: The Importance of Relationships in
the Financial Advisory Industry, 141 J. Fin. Econ. 1218–43 (2021); Christopher P. Clifford & William C. Gerken,
Property Rights to Client Relationships and Financial Advisor Incentives, 76 J. Finance 2409–45 (2021).
349
Bishara & Starr, supra note 346, at 541.
350
Little Sisters of the Poor Saints Peter & Paul Home v. Pennsylvania, 591 U.S. 657, 682 (2020) (quoting State
Farm Mut. Auto. Ins. Co., 463 U.S. at 43).
351
Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 416 (1971), abrogated on other grounds by
Califano v. Sanders, 430 U.S. 99 (1977).
