Download PDF

Axon Enterprises v. FTC[1] wasn’t supposed to be about labor law. In fact, it wasn’t supposed to be about any area of substantive law. It seemed to be about only a dry jurisdictional issue: does a plaintiff challenging an agency’s constitutionality have to exhaust the agency’s internal procedures before going to court?[2] The Supreme Court ultimately said no.[3] And had the case ended there, its effect might have been muted.

But it didn’t end therenot quite. In a spirited concurrence, Justice Clarence Thomas used the case as an opportunity to address a bigger issue: the very nature of judicial power.[4] He wrote that because of political, social, and historical developments, the rights of private parties have increasingly come to be adjudicated not by courts, but by administrative agencies.[5] These agencies take evidence, find facts, and interpret statutes.[6] More important, their decisions receive only the most cursory judicial review. In even serious cases, these decisions are effectively final.[7]

That arrangement, Thomas wrote, clashes with Article III of the U.S. Constitution.[8] Article III vests all “judicial power” in courts.[9] Historically, judicial power was understood as the power to issue binding decisions affecting “core private rights”i.e., life, liberty, and property.[10] Article III gave that power to federal courtsand only federal courts.[11] It left no room for the exercise of judicial power by agencies.[12] And yet, in modern practice, more and more cases have been shunted into internal agency processes, effectively allowing agencies to wield power over core private rights.[13] That is, agencies have come to wield the power denied them by Article III.[14]

Justice Thomas wasn’t the first to raise these concerns: he was building on a groundswell of legal scholarship.[15] In recent years, scholars have increasingly questioned whether agency adjudication can be squared with Article III.[16] And even more recently, litigants have seized on those doubts and challenged agency adjudication in court.[17] Those challenges have resulted in some startling victories, including a recent decision from the Fifth Circuit striking down aspects of the Securities and Exchange Commission’s adjudicatory procedures.[18] And while no direct Article III challenge has yet reached the Supreme Court, one could soon.[19] The stage seems set for a decisive ruling.

Yet amid this debate, a puzzle presents itself. One agency perhaps best illustrates the Article III problem. This agency exercises broad power over private rights, and it does so almost exclusively through case-by-case adjudication.[20] But so far, it has escaped the notice of litigants. None of the new crop of challengers has confronted it with a claim under Article III. And all the while, it has continued to decide the rights of private parties,[21] often in headline-grabbing cases.[22] It is, of course, the National Labor Relations Board.[23]

The Board is one of the oldest and, in some ways, most powerful independent agencies.[24] It sets labor policy for the entire country.[25] It decides which workers are protected, with whom they can organize, and when they can use their employer’s property for that purpose.[26] It makes those decisions through a quasi-judicial process, developing its own legal “precedent.”[27] And that precedent receives near-total deference in court.[28]

It is a puzzle, then, that the Board has so far avoided a challenge.[29] But the Board’s relative safety may be temporary. The more the legal community starts to question agency adjudication, the more glaring the Board’s status will become. Eventually, someone will realize that of all the federal agencies, the Board might be the one in most tension with Article III. And if Justice Thomas is right, that tension may be untenable.[30] The Board’s very structure may be unconstitutional.

I. Article III, Judicial Power, and Private Rights

The phrase “judicial power” is deceptively straightforward. It calls to mind the ordinary work of judges: applying law to facts.[31] But to the American founders, it meant something more specific.[32] It was the power to make binding decisions affecting “core private rights.”[33] Core private rights were the rights held by people as individuals.[34] These rights did not come from government; they were prior to government.[35] They existed in the state of nature and survived the creation of civil society.[36] They were sometimes called individual rights, sometimes fundamental rights, sometimes natural rights.[37] But whatever the label, they could be taken away only by due process of law.[38]

In contrast to core private rights were “public rights.”[39] Public rights were rights created by the government and belonged to the citizenry as a whole.[40] They included what we might today call licenses or privileges.[41] Classic examples were veterans’ benefits, patents, and public land grants.[42] When the government distributed these privileges, it was often applying law to facts.[43] After all, it had to determine who qualified for what privilege.[44] But it was not depriving anyone of core private rights.[45] And for that reason, it was not exercising “judicial power.”[46] Its activity could be more fairly described as legislative or executive.[47]

To the modern mind, that distinction may seem arbitrary. But the founders had good reasons for itreasons owing mostly to the relative competencies of the three branches.[48] The legislative and executive branches were by their nature political.[49] They answered to broad constituencies and so had an incentive to please the greatest number of people.[50] That incentive made them good at designing and enforcing general rules.[51] But it also made them dangerous arbiters of individual rights.[52] The interests of the public could sometimes clash with the interests of individual people.[53] For example, while redistributive laws might be popular in a general sense, arbitrary redistribution could violate individual rights.[54] More viscerally, a heinous crime might cry out for punishment.[55] The public might demand that someoneanyonetake the blame. But that impulse could lead to arbitrary prosecution.[56] There had to be an institution independent and powerful enough to resist popular will and protect individual interests.[57]

That’s where courts came in.[58] Courts were competent to adjudicate private rights precisely because they were outside the political process.[59] They considered individuals as individuals.[60] Lacking external constituencies, they were free to judge cases on the merits.[61] That’s why they could be trusted with judicial power.[62] They could provide the cool, independent, measured judgment needed to protect private rights.[63] And more to the point, the political branches could not.[64]

II. Agency Adjudication and the Appellate Model

That institutional model held for most of the nation’s first century.[65] Article III was understood to give courts all judicial power, and judicial power was understood to mean the power to adjudicate core private rights.[66]

But the model started to fray in the late 19th century.[67] After the Civil War, the nation experienced a boom of economic and social change.[68] Society was transformed by urbanization, immigration, and industrialization.[69] These new pressures produced new problems, which in turn prompted new calls for reform.[70] Lawmakers responded with a wave of statutory and regulatory schemesschemes that gave the nation its first glimpses of the administrative state. [71]

This new constitutional beasta headless “fourth branch”[72]quickly found its way into court. The earliest cases involved the Interstate Commerce Commission, or ICC.[73] The ICC was established mainly to regulate railroads.[74] Among other things, it set rates for common carriers moving freight across state lines.[75] At first, the Supreme Court reviewed the ICC’s orders closely.[76] The Court insisted that judges, as part of their constitutional duty, had to develop their own factual records.[77] They also had to exercise independent judgment over questions of law.[78] They owed the agency no deference.[79]

But in an age of progressive politics, that approach proved provocative.[80] People saw the Court as an obstacle to popular reform, and they demanded change.[81] Congress responded by passing the Hepburn Act.[82] Among other things, the Act made ICC orders self-executing if not challenged within 30 days.[83] And while it specified no standard of review, it implicitly instructed courts to take a back seat.[84]

The Supreme Court got the message. In a series of decisions in the early 20th century, it backed into what would become known as the appellate model.[85] The appellate model allowed the ICC to develop its own factual record.[86] Courts would limit their review to that record and treat the agency’s factual findings as presumptively conclusive.[87] On legal issues, however, courts would continue to have the last word.[88] They still owed the agency no deference on questions of law.[89]

That approach received its fullest articulation in Crowell v. Benson.[90] Crowell involved not the ICC, but workers’ compensation. Congress had adopted a workers’-compensation system for employees working on navigable waters.[91] The system was administered by an agency commissioner, who determined eligibility for benefits and issued binding compensation orders.[92] Crowell approved that approach, subject to appellate-style judicial review.[93] Courts would allow the commissioner to develop the record and determine ordinary facts.[94] But they would also supplement the record in certain respects.[95] In particular, they would consider additional evidence bearing on the commissioner’s jurisdiction or an individual’s constitutional rights.[96] And of course, they would still resolve all legal questions themselves.[97]

Crowell set the standard for judicial review going forward.[98] Its contours were absorbed into multiple statutes, including section 706 of the Administrative Procedure Act.[99] It was cited hundreds of times, often for the proposition that courts reviewed questions of law de novo.[100]

But that’s not to say the model was static. Over time, certain aspects changed or fell away. For example, courts mostly stopped developing their own records, even for “jurisdictional” facts.[101] They instead relied entirely on agency factfinding.[102] They also backed away from plenary review over questions of law. Rather than review those questions de novo, they increasingly deferred to agency interpretations.[103]

That trend famously reached its apogee in Chevron USA, Inc. v. Natural Resource Defense Council.[104] In Chevron, the Supreme Court announced a two-step process for reviewing agency interpretations.[105] First, the Court would determine whether the statute in question was ambiguous.[106] Second, if the Court found an ambiguity, it would defer to the agency’s interpretation as long as the interpretation was “reasonable.”[107]

The result was a model vastly different from the one Crowell envisioned. Crowell described the relationship between courts and agencies as something like the one between trial and appellate courts.[108] But by the end of the 20th century, that analogy no longer held. Courts accepted not only agency factfinding, but also agency interpretations of law.[109] That is, they deferred in nearly all respects. If they weren’t quite rubber-stamping agency decisions, they were doing something close to it.[110]

III. Deference and Its Discontents

That shift didn’t go unnoticed. As the years wore on, critics started to question the basic premises of the administrative state.[111] Much of the criticism centered on agency rulemaking and its apparent tension with Article I, which vests all legislative authority in Congress.[112] But scholars also started to question agency adjudication.[113] As the administrative state expanded, agencies often made decisions using quasi-judicial procedures.[114] They filed charges, took evidence, and issued binding orders.[115] Scholars struggled to square that model with Article III, which, again, vests all judicial power in courts.[116]

Some observers tried to resolve the tension by pointing to judicial review.[117] They argued that agency adjudication was constitutional as long as people still got their day in court.[118] Even if agencies made the initial decisions, courts would provide a backstop to protect individual rights.[119] That rationale made sense, however, only if you assumed that judicial review would be meaningful.[120] And as courts slipped deeper into deference, that assumption became harder to justify.[121] If courts deferred to agencies on every issue, what backstop did they really offer?[122] Weren’t agency decisions effectively final?[123]

Until recently, this debate remained cloistered in the halls of academe. It made for interesting law-review articles, but was rarely taken seriously in court.[124] Courts continued to apply the appellate model as a matter of course.[125] The discourse over agency adjudication seemed to be going nowhere.[126]

But the last few years have seen a shift. Multiple plaintiffs have challenged agency adjudication. They’ve targeted, among others, the Securities and Exchange Commission,[127] the Federal Trade Commission,[128] and the Department of Labor.[129] To be sure, some have only glanced at the Article III issue; they’ve focused instead on the status of administrative law judges and jury-trial rights.[130] But a few have teed up the Article III problem directly.[131] They’ve argued that Article III vests not just some, but all judicial power in courts.[132] And Congress cannot circumvent that requirement by providing for only cursory judicial review.[133]

Much of this litigation is still in its early stages. For example, as of this writing, the challenges to the Department of Labor’s process are still pending in district court.[134] But some lawsuits have already produced significant victories for private parties. Most notably, in 2022, the Fifth Circuit struck down some of the SEC’s administrative procedures.[135] The court didn’t rely on Article III per se; it focused instead on jury-trial rights and the nondelegation doctrine.[136] But its decision did show that courts are increasingly skeptical of agency adjudication.[137] And if that trend continues, it could produce even more challengeschallenges perhaps aimed at an even broader range of agencies.

IV. Private Rights, Labor Law, and the NLRB

One agency to escape these challenges has been, oddly enough, their most obvious target: the National Labor Relations Board. Established in 1935, the Board is one of the oldest “independent” agencies.[138] It consists of five presidentially appointed members, each serving a fixed term.[139] The members are charged with developing labor policy for the entire country.[140] They have statutory authority to develop that policy through rulemaking.[141] But more often, they do it through adjudication.[142] They issue opinions, draw on their own “precedent,” and spin out new rules as the circumstances require.[143] In other words, they mimic the methods of a common-law court.

Their decisions, of course, can be reviewed by real courts.[144] But since early in the Board’s history, it has been clear that judicial review would be limited. The original National Labor Relations Act required courts to accept the Board’s factual findings if supported by “evidence.”[145] The statute was later amended to require “substantial evidence,”[146] but factual review remained deferential.[147] And a similar standard held for questions of law. In 1944’s NLRB v. Hearst Publications, the Supreme Court instructed lower courts to defer to the Board’s legal interpretations.[148] The Court relied on a mix of congressional intent and agency expertise to reach this conclusion. It reasoned that Congress wanted to create a uniform national labor policy.[149] And the Board, by applying the law across hundreds of industries and thousands of labor disputes, had developed a certain expertise.[150] The Board knew the kinds of issues that could come up in an organizing campaign.[151] It knew what kinds of bargaining units were manageable.[152] And it knew what kinds of conditions were most likely to cause industrial strife.[153] If courts were to respect congressional intent, they had to also respect that expertise.[154] And that meant respecting the Board’s judgment about how to apply the statute.[155]

The result was an early example of agency dominance. Years before the APA and decades before Chevron, the Board wielded broad discretion over facts and law.[156] It had the power both to say what the law was and apply that law to private parties.[157] It was a court in all but name.[158]

Given that framework, it is hard to avoid the conclusion that the Board exercisedand continues to exercisejudicial power. Remember, the framers understood judicial power as the power to determine core private rights.[159] Among those rights were the right to private property.[160] And the Board’s decisions affect private property perhaps more directly than any other agency.[161] To take just a few examples, consider the Board’s rules on solicitation and distribution. The Board determines when, where, and under what circumstances union organizers can solicit employees on the employer’s property.[162] Likewise, the Board determines when organizers can use an employer’s equipment for union-related activity.[163] These rules require the Board to “balance” property rights against statutory rights.[164] And indeed, that kind of balancing runs through much of the Board’s internal doctrinea doctrine that makes up most of modern labor law.[165] It is therefore nigh impossible to understand labor law without also understanding its interaction with private property rights.[166]

The same could be said for private contract rights. Among the Board’s functions is certifying unions for collective bargaining.[167] And as soon as the Board certifies a union, contract rights are immediately curtailed.[168] Employers and employees can no longer bargain with each other directly.[169] Instead, an employer must bargain with the certified union.[170] And the union must bargain for all employees in the bargaining unit.[171] The resulting agreements set the terms of employment for everyone, even employees who refuse union membership.[172] That is, the collective agreement “extinguishes” the contract rights of individual workers.[173] A more direct limit on contract rights is hard to imagine.

V. The Enigma of the Status Quo

In that context, the conflict between Article III and the Board seems obvious. While Article III vests all judicial power in courts, modern labor law allows judicial power to be exercised by the Board. And that conflict has existed since the Board’s founding. So why has it gone overlooked for so long?

The main reason is probably adverse precedent. In the early days of the NLRA, multiple lawsuits were filed over the statute’s constitutionality. Those lawsuits were by no means frivolous; earlier efforts to mandate collective bargaining had been struck down.[174] But in NLRB v. Jones & Laughlin Steel Corp., the Supreme Court upheld the NLRA against a multi-pronged constitutional attack.[175] The challengers in Jones focused on the Interstate Commerce Clause; they argued that the NLRA was invalid because it regulated purely local (as opposed to interstate) activity.[176] Most of the Court’s opinion focused on that issue as well.[177] But the challengers also made a claim under Article III: they argued that the NLRA effectively delegated judicial power to the Board, and the Board was incompetent to exercise that power.[178]

The Court disagreedbut it didn’t dwell on the issue. Instead, in three thinly reasoned sentences, it dismissed the Article III argument as essentially frivolous:

We construe the [NLRA’s] procedural provisions as affording adequate opportunity to secure judicial protection against arbitrary action in accordance with well-settled rules applicable to administrative agencies set up by Congress to aid in the enforcement of valid legislation. It is not necessary to repeat these rules which have been frequently declared. None of them appears to have been transgressed in the instant case.[179]

In effect, the Court anticipated modern defenses of the appellate model. It reasoned that agency adjudication was fine as long as there was a judicial backstop.[180] But again, that rationale assumed that the backstop would be effective.[181] And given modern deference doctrines, it no longer is.[182] In the decades since Jones, courts have deferred not only to the Board’s factual findings, but also to its legal conclusions.[183] Cases like Hearst and Chevron have whittled judicial review down to a rump.[184] So in the modern context, Jones’s conclusions about Article III make no sense. The “judicial protection” it referred to no longer exists.[185]

Jones aside, another reason for the Board’s relative safety might be its limited remedies. Unlike some agencies, the Board cannot impose civil penalties. It can issue cease-and-desist orders, which require parties to comply in the future.[186] It can also make parties post notices admitting they violated the law.[187] And in some cases, it can impose make-whole remedies, such as backpay.[188] But it cannot levy freestanding civil fines.[189]

The Board’s remedies have been criticized as weak.[190] But their very weakness may have helped stave off a constitutional attack. Under the Seventh Amendment, private parties have a right to a jury trial for all actions at common law over $20.[191] Whether an action is “at common law” depends on whether it mirrors some common-law action that existed in 1791.[192] Because civil penalties are monetary remedies, they resemble classic common-law actions.[193] So at least some courts have found that they trigger the Seventh Amendment.[194]

But that rationale doesn’t work for the Board’s remedies. Make-whole remedies like back pay are inherently equitable.[195] They do not resemble common-law damages claims.[196] They therefore do not require a jurya fact that helps insulate the Board from a Seventh Amendment challenge.[197]

But that point does not solve the Article III problem. Even if the Board adjudicates no “actions at common law,” it still adjudicates core private rights. It still has effective final say over matters related to labor, contracting, and property. So the Board’s limited remedies can’t be the full answer.[198] Even if the Seventh Amendment is no issue, Article III still is.[199]

A final reason may be the Board’s sheer longevity. The Board has been around for nearly a century; and in that time, it has become a fixture of the legal firmament.[200] It is comparatively easy to attack an agency like the Consumer Financial Protection Bureau, which strikes some observers as both novel and dangerous.[201] But the Board is familiar; its strengths and weaknesses are well known and longstanding.[202] Its very age may seem to put it beyond question.[203]

But of course, age alone is a weak defense.[204] An error is no less an error because it has gone uncorrected for a long time.[205] The Board may predate the APA, Chevron, and modern anxieties about administrative creep. But it does not predate Article III. Article III assigns all judicial power to courts, without exception.[206] If the Board’s structure allows it to wield judicial power, it violates Article IIIno matter how longstanding that violation may be.[207]

VI. An Old Problem, Reborn and Magnified

And indeed, as the years have passed, the Board has illustrated the need for Article III more clearly than perhaps any other agency. It has modeled all the evils Article III was designed to prevent. And in some cases, it has modeled them spectacularly.

Again, the framers adopted Article III for a reason. They vested all judicial power in courts because they knew the alternative was worse.[208] They had seen the English crown abuse its power through “prerogative” courts, such as the Star Chamber and the High Commission.[209] These quasi-courts relied on flimsy evidence, enforced extra-legal standards, and required defendants to prove their own innocence.[210] Worse, they did all these things to serve political, rather than legal, ends.[211]

The Board does much the same thing.[212] It adopts one-sided presumptions, shifts the burden of proof, and imposes new rules retroactively.[213] It holds parties responsible for unannounced standards while pretending that those standards were the law all along.[214]

Those practices would be bad enough if they weren’t also infected with politics. But they are.[215] Indeed, the Board is notorious for flipping its positions from administration to administration.[216] Republican Boards reliably support management, and Democratic ones invariably support unions.[217] A particularly glaring example came during the Obama administration, when the Board reversed a series of precedents that had been in place for a combined 4,500 years.[218] The Trump Board then spent much of the next four years reverting to prior standards.[219] And now, the Biden Board is busy reversing the Trump Board’s rulings and returning to Obama-era precedents.[220] This cycle has recurred for decades and shows no sign of slowing down. It repeats from election to election, drowning private parties in wave after wave of “policy oscillation.”[221]

This is precisely the kind of political jockeying Article III was meant to avoid.[222] Article III gave judicial power to courts because individual rights should not depend on which party is in power.[223] They should depend on the lawa law declared in advance and enforced through fair procedures.[224] That, at least, is what the framers had in mind.[225] It’s what they envisioned when they wrote Article III.[226] If we’re serious about respecting their vision, we should reexamine how judicial power has come to be wielded by the political branches.[227] And if we’re looking places to start, there would be few better than the Board.


[1]  No. 21-86, slip op. at 1-2 (U.S. Apr. 14, 2023).


[2]  Id. at 2.


[3]  Id.


[4]  Id. at 2-9 (Thomas, J., concurring).


[5]  Id. at 4-6 (Thomas, J., concurring).


[6]  Id. (Thomas, J., concurring).


[7]  See id. at 1, 7-9 (Thomas, J., concurring).


[8]  See id. at 1 (Thomas, J., concurring) (expressing “grave doubts about the constitutional propriety of Congress vesting administrative agencies with primary authority to adjudicate core private rights with only deferential judicial review on the back end”).


[9]  U.S. Const. art. III § 1. See also Martin v. Hunter’s Lessee, 14 U.S. 304, 330 (1816) (“If, then, it is a duty of congress to vest the judicial power of the United States, it is a duty to vest the whole judicial power. The language, if imperative as to one part, is imperative as to all.”).


[10]  Axon, No. 21-86, slip op. at 3 (Thomas, J., concurring). See also Oil States Energy Servs., LLC v. Greene’s Energy Grp., LLC, 138 S. Ct. 1365, 1373 (2018) (explaining that Article III distinguishes between public rights and private rights, the latter of which may be adjudicated only by courts).


[11]  See Axon, No. 21-86, slip op. at 3-9 (Thomas, J., concurring). See also Oil States Energy, 138 S. Ct. at 1373; Exec. Benefits Ins. Agency v. Arkison, 573 U.S. 25, 32–33 (2014).


[12]  See Axon, No. 21-86, slip op. at 3-9 (Thomas, J., concurring). See also Arkinson, 573 U.S. at 33; Philip Hamburger, Is Administrative Law Unlawful? 154 (2014) (explaining that the Constitution vests all judicial power in courts and leaves no room for binding adjudication by the executive).


[13]  See Axon, No. 21-86, slip op. at 3-9 (Thomas, J., concurring). Cf. Hamburger, supra note 12, at 488 (“What once seemed a mere variation, however, has since become a central mode of governancea full-scale alternative to the constitutionally established forms of government.”).


[14]  See Axon, No. 21-86, slip op. at 3-9 (Thomas, J., concurring).


[15]  See, e.g., Thomas W. Merrill, Article III, Agency Adjudication, and the Origins of the Appellate Review Model of Administrative Law, 111 Colum. L. Rev. 939, 979-80 (2011); Caleb Nelson, Adjudication in the Political Branches, 107 Colum. L. Rev. 559, 569 (2007); Jennifer Mascott, Constitutionally Conforming Agency Adjudication, 2 Loyola U. Chi. J. Reg. Compliance 22, 45 (2017); Nathan S. Chapman & Michael W. McConnell, Due Process as Separation of Powers, 121 Yale L.J. 1672, 1727-70 (2012); Gary Lawson, The Rise and Rise of the Administrative State, 107 Harv. L. Rev. 1231, 1247 (1994). Cf. Adam B. Cox & Emma Kaufman, The Adjudicative State, 132 Yale L.J. 1769, 1788 (2023) (describing groundswell of scholarship aimed at reviving “Article III essentialism”).


[16]  See, e.g., Mascott, supra note 15, at 45; Lawson, supra note 15, at 1247; Merrill, supra note 15, at 980 (“The appellate review model, from this perspective, represents a major challenge: Is there a principled justification for what appears to be a violation of the plain requirements of the Constitution?”).


[17]  See, e.g., Jarkesy v. SEC, No. 20-61007 (5th Cir. May 18, 2022) (holding that SEC’s internal adjudicative process was unconstitutional); Sun Valley Orchards, LLC v. DOL, No. 1:21-cv-16625, 2023 BL 257772, at *6-7 (D.N.J. July 27, 2023) (considering and rejecting challenge to Department of Labor’s H-2B visa enforcement mechanisms, including adjudication by an administrative official, because in the court’s view that program involved public rather than private rights); Compl., C.S. Lawn & Landscaping, Inc. v. U.S. Dep’t of Labor, Case No. 1:23-cv-01533 (D. Md. May 30, 2023) (arguing that DOL’s administrative adjudication process for H-2B visa program violated Article III by giving a non-Article III decisionmaker power over core private rights).


[18]  Jarkesy, No. 20-61007, slip op. at 5-15.


[19]  Cf. Kalvis Golde, Another Federal Agency Challenges Adverse Ruling by 5th Circuit, SCOTUSBlog (Mar. 31, 2023), (noting that the Supreme Court will consider constitutional challenges to multiple agency structures next term, including the structures of the CFPB and the SEC).


[20]  See also William B. Gould IV, Politics and the Effect on the National Labor Relations Board’s Adjudicative and Rulemaking Processes, 64 Emory L.J. 1501, 1505 (2015) (noting that the Board was originally designed as an “effective substitution” for courts through the “mechanism” of an “expert” administrative tribunal).


[21]  See, e.g., Daniel Wiessner, NLRB Paves Way for Workers to Unionize Without Formal Elections, Reuters (Aug. 25, 2023), (reporting on Board decision authorizing union recognition without an election); Robert Iafolla, Unions Score Big as NLRB Eases Path to Representation, Bloomberg Law (Aug. 25, 2023), (same).


[22]  Compare Sean Redmond, NLRB’s Cemex Decision Denies Workers’ Rights to Make Free and Fair Choice About Unions, U.S. Chamber of Commerce (Aug. 31, 2023), (criticizing Board’s decision to deny workers free choice over unionization), with Tascha Shahriari-Parsa, Cemex Is a Big Change, but It’s Not Joy Silk, OnLabor (Aug. 26, 2023), (arguing that Cemex benefits workers but does not go far enough).


[23]  Cf. Nelson, supra note 15, at 601 (describing the Board as one of the more “adventurous” efforts to assign adjudicatory responsibility to agencies).


[24]  See Theodore St. Antoine, The NLRB, the Courts, The Administrative Procedures Act, and Chevron: Now and Then, 64 Emory L.J. 1529, 1538 (2015) (tracing judicial deference to Board’s decisions predating the APA and Chevron).


[25]  See Beth Israel Hosp. v. NLRB, 437 U.S. 483, 500 (1978) (“It is the Board on which Congress conferred the authority to develop and apply fundamental national labor policy.”).


[26]  See St. Antoine, supra note 24, at 1542-50 (surveying recent Board precedent and judicial review over a broad array of issues). See also The Atlanta Opera Inc., 372 N.L.R.B. No. 95, 2023 WL 4051664, at *20 (June 13, 2023) (adopting new test for determining when a worker is an “employee” protected by federal labor law and rejecting test adopted by federal courts); Cemex Constr. Materials Pac., LLC, 372 N.L.R.B. No. 130 (Aug. 25, 2023) (adopting new rule allowing union recognition without an election in a broader array of cases); NLRB Gen. Counsel Br. in Support of Exceptions, NLRB Case No. 10-CA-379843 (Apr. 28, 2023) (urging Board to overturn multiple recent precedents and arguing that the Board should require employers to allow workers to use employer email systems for organizing purposes).


[27]  See Gould, supra note 20, at 1505-06 (describing Board’s quasi-judicial proceedings).


[28]  See, e.g., 29 U.S.C. § 160(e), (f) (requiring courts to accept Board’s factual findings when supported by “substantial evidence”); Universal Camera Corp. v. NLRB, 340 U.S. 474, 488 (1951) (explaining that a court must defer to Board’s factual determinations if supported by substantial evidence even if the court would have weighed the evidence differently); NLRB v. Hearst Publications, 322 U.S. 111, 130 (1944) (treating the interpretation of statutory terms as merely part of the “administrative routine” of the Board and directing courts to defer), abrogation on statutory grounds recognized in Nationwide Mut. Ins. Co. v. Darden, 503 U.S. 318 (1992); NLRB v. Curtin Matheson Sci., Inc., 494 U.S. 775, 778 n.2 (1990) (deferring to Board’s interpretation of NLRA because the interpretation was “rational and consistent with the Act”); Stephens Media, LLC v. NLRB, 677 F.3d 1241, 1250 (D.C. Cir. 2012) (deferring to Board’s conclusion that employee’s secret tape recording was protected activity under the NLRA) (“As we have noted many times before, our role in reviewing an NLRB decision is limited.”) (quoting Wayneview Care Ctr. v. NLRB, 664 F.3d 341, 348 (D.C. Cir. 2011)).


[29]  Cf. Michael C. Harper, The Case for Limiting Judicial Review of Labor Board Certification Decisions, 55 Geo. Wash. L. Rev. 262, 317-18 (1987) (observing that a “broad” reading of Article III would call the Board’s adjudicatory authority into question).


[30]  Cf. Perez v. Mortgage Bankers Assn., 575 U.S. 92, 119 (2015) (Thomas, J., concurring) (explaining that judicial power was originally understood to require a court to exercise its independent judgment in interpreting and expounding the laws—without deference to the executive).


[31]  See NLRB v. Robbins Tire & Rubber Co., 161 F.2d 798, 802 (5th Cir. 1947) (Waller, J., concurring) (citing one “generally used” definition: “the power of a court to decide and pronounce judgment and to carry it into effect between parties who bring the case before it for decision” (quoting Gentry v. Fry, 4 Mo. 120 (1835))); Nelson, supra note 15, at 559 (rejecting “easy equation” of judicial power with binding adjudication).


[32]  See Axon, No. 21-86, slip op. at 3 (Thomas, J., concurring). See also Chapman & McConnell, supra note 15, at 1687 (tracing ban on executive exercise of judicial power to English common law and thought of Sir Edward Coke).


[33]  Axon, No. 21-86, slip op. at 3 (Thomas, J., concurring); Nelson, supra note 15, at 563. See also Hamburger, supra note 12, at 2 (explaining that administrative power conflicts with private rights only when it exercises binding power).


[34]  Axon, No. 21-86, slip op. at 3 (Thomas, J., concurring).


[35]  Id.; Nelson, supra note 15, at 565-67.


[36]  Nelson, supra note 15, at 567; Hamburger, supra note 12, at 330 (describing how natural-law theory influenced founders’ understanding of separation of powers). See also John Locke, Second Treatise on Government § 11.135 (Dover ed. 2002) (arguing that natural lawthe law existing in a state of natureremains in effect after people form civil societies).


[37]  See Axon, No. 21-86, slip op. at 4 (Thomas, J., concurring) (explaining that private-rights doctrine developed from Lockean social-contract theory of natural rights).


[38]  See id. (explaining that private rights, unlike public rights, could be abridged only through the exercise of judicial power by courts). See also John Harrison, Jurisdiction, Congressional Power, and Constitutional Remedies, 86 Geo. L.J. 2513, 2516 (1998) (“The measure of judicial involvement was private right.”). Cf. Hunter’s Lessee, 14 U.S. at 330 (“If, then, it is a duty of Congress to vest the judicial power of the United States, it is a duty to vest the whole judicial power. The language, if imperative as to one part, is imperative as to all.”).


[39]  See, e.g., Oil States Energy, 138 S. Ct. at 1373 (explaining that Article III distinguishes between public and private rights); Arkinson, 573 U.S. at 32-33 (same); Nelson, supra note 15, at 565-72 (same).


[40]  Axon, No. 21-86, slip op. at 4 (Thomas, J., concurring).


[41]  Id.


[42]  See id.; Merrill, supra note 15, at 950, 990; Nelson, supra note 15, at 557, 609; Hamburger, supra note 12, at 4 (listing examples of public rights).


[43]  See Nelson, supra note 15, at 609.


[44]  See id.


[45]  See Hamburger, supra note 12, at 293 (“But generally executive power—the executive’s ‘public rights’ as understood through much of the nineteenth century—was not a power to bind subjects, this being why it could be a realm of discretion defined and allowed by law.”).


[46]  See Axon, No. 21-86, slip op. at 4 (Thomas, J., concurring). See also Hamburger, supra note 12, at 191-92 (acknowledging that much proper executive and administrative activity involves applying law to facts, and that this activity does not conflict with the constitution).


[47]  See Oil States Energy, 138 S. Ct. at 1373 (explaining that because public rights can be disposed of without judicial power, Congress has “wide latitude” to assign their adjudication to executive agencies); Nelson, supra note 15, at 567-68 (describing original justifications for different treatment of public rights); Hamburger, supra note 12, at 2 (explaining that executive may distribute public benefits without binding private parties and without conflicting with private rights).


[48]  See Gutierrez-Brizuela v. Lynch, 834 F.3d 1142, 1150 (10th Cir. 2016) (Gorsuch, J.) (observing that the founders separated the branches into different spheres in part to “avoid the possibility of allowing politicized decisionmakers to decide cases and controversies”). See also Nelson, supra note 15, at 559, 624 (arguing that structural relationship among branches differed depending on which interests the government was acting on).


[49]  See Nelson, supra note 15, at 571-72.


[50]  Id.


[51]  See id. at 597-98 (explaining that legislative power extended to regulating prospective conduct, and legislature could properly authorize agencies to issue prospective orders (e.g., cease-and-desist orders); but that different considerations came into account when imposing liability on individuals for private conduct) (citing Ohio Valley Water Co. v. Ben Avon Borough, 253 U.S. 287, 289 (1920); St. Joseph Stock Yards Co. v. United States, 298 U.S. 38, 50-54 (1936)).


[52]  See id.; Appeal of Ervine, 16 Pa. 256, 268 (1851) (comparing relative competencies of legislature and judiciary and concluding that only judiciary could be trusted with protecting individual rights) (“But when individuals are selected from the mass, and laws are enacted affecting their property, without summons or notice, at the instigation of an interested party, who is to stand up for them, thus isolated from the mass, in injury and injustice, or where are they to seek relief from such acts of despotic power? They have no refuge but in the courts, the only secure place for determining conflicting rights by due course of law.”). Cf. Mascott, supra note 15, at 42 (noting that presidential control over agency officials is in “tension” with the kind of impartiality we expect in judicial adjudications); Hamburger, supra note 12, at 5, 339-490 (explaining that the power to bind is dangerous when vested in the executive because it allows the executive to constrain private liberty).


[53]  See Mascott, supra note 15, at 42-43 (arguing that adjudication by executive officials puts regulated parties at risk of having their rights determined according to political calculations).


[54]  See, e.g., Davidson v. New Orleans, 96 U.S. 97, 102 (1878) (drawing on English common law and declaring that a law taking property from A and giving it to B would violate due process); In re Smith’s Est., 607, 57 A. 37, 38 (Pa. 1904) (reasoning that a special law divesting a particular person of property would violate due process); Lawrence E. Tierney Coal Co. v. Smith’s Guardian, 203 S.W. 731, 736 (Ky. 1918) (“If the Legislature possessed an irresponsible power over every man’s private estate . . . all inducement to acquisition, to industry, and economy would be removed.”) (quoting Ervine’s Appeal, 16 Pa. at 256); Bowman v. Middleton, 1 S.C.L. 252, 252 (1792) (finding that act purporting to transfer title to property from one private party to another was “against common right and reason, as well as against Magna Charta; therefore, ipso facto, void”).


[55]  N. Pipeline Const. Co. v. Marathon Pipe Line Co., 458 U.S. 50, 70 n.25 (1982) (observing that criminal matters remain at the heart of adjudication required to be performed in court by Article III).


[56]  Cf. Hamburger, supra note 12, at 231 (noting that “where agencies adjudicate cases of a criminal nature, they tend to deny the associated constitutional rights”).


[57]  See Nelson, supra note 15, at 605 (describing criminal cases as the “paradigmatic example of a dispute that requires fully ‘judicial’ determination”); N. Pipeline, 458 U.S. at 70 n.24 (“Of course, the public-rights doctrine does not extend to any criminal matters, although the Government is a proper party.”).


[58]  See Nelson, supra note 15, at 562 (“When core private rights are at stake, the judiciary assumes an indispensable role.”).


[59]  See The Federalist No. 78 (Alexander Hamilton) (explaining that the judiciary must have “complete independence” from the political branches; otherwise, “all the reservations of particular rights or privileges would amount to nothing”) Hamburger, supra note 12, at 339 (explaining that the separation of judicial power from political branches protected people from interested decisionmaking).


[60]  See Harrison, supra note 38, at 2517; Nelson, supra note 15], at 590.


[61]  See Nelson, supra note 15, at 590. See also Thomas M. Cooley, Treatise on the Constitutional Limitations Which Rest upon the Legislative Power of the States of the American Union 109 (6th ed. 1890) (explaining the role judicial independence and exclusivity of judicial power played in protecting rights of individuals).


[62]  See Nelson, supra note 15, at 590; Ervine’s Appeal, 16 Pa. at 256. Cf. Axon, No. 21-86, slip op. at 9 (Thomas, J., concurring) (explaining that whether Article III requires adjudication in court depends on whether private rights are at stake).


[63]  See Ervine’s Appeal, 16 Pa. at 256; Hamburger, supra note 12, at 232-34. See also Commodity Futures Trading Comm’n v. Schor, 478 U.S. 833, 848 (1986) (explaining that Article III protects judicial independence not to promote the judiciary’s institutional interests, but to safeguard individual rights).


[64]  Axon, No. 21-86, slip op. at 3-4 (Thomas, J., concurring) (explaining that Article III allows core private rights to be abridged only through adjudication by courts); N. Pipeline Const., 458 U.S. at 77 (“It is, of course, true that while the power to adjudicate “private rights” must be vested in an Art. III court . . . .”); Parmelee v. Thompson, 7 Hill 77, 80, 1845 WL 4507 (N.Y. Sup. Ct. 1845) (explaining that “the legislature has no jurisdiction to determine facts touching the rights of individuals”); Gutierrez-Brizuela, 834 F.3d at 1149 (observing that if political branches could exercise judicial power, “[t]hey might be tempted to bend existing laws, to reinterpret and apply them retroactively in novel ways and without advance notice.”). Cf. Thomas v. Union Carbide Agr. Prod. Co., 473 U.S. 568, 583 (1985) (rejecting “absolute” reading of Article III and concluding that Congress’s authority to assign matters to agencies depends on nature of underlying rights and risks to judicial independence).


[65]  See Nelson, supra note 15, at 562 (reviewing 19th-century precedents).


[66]  See Merrill, supra note 15, at 944-52; St. Antoine, supra note 24, at 1532. See also Cooley, supra note 61, at 109 (stating that it was the “peculiar province” of the judiciary to “adjudicate upon, and protect the rights of individual citizens, and to that end to construe and apply the laws”). But see Cox & Kaufman, supra note 15, at 1791-94 (describing this view as formalist and questioning whether it accurately describes historical practice of courts and agencies, which were not always so neat in their categorizations).


[67]  Axon, No. 21-86, slip op. at 4-5 (Thomas, J., concurring) (“As notions of administrative efficiency came into vogue, courts were viewed less as guardians of core private rights and more as impediments to expert administrative adjudication.”).


[68]  See James W. Ely, The Contract Clause: A Constitutional History 147 (2016) [hereinafter The Contract Clause] (describing social and economic pressures that led to new and more aggressive regulatory approaches); James W. Ely Jr., The Guardian of Every Other Right: A Constitutional History of Property Rights 8 (3d ed. 2007) [hereinafter History of Property Rights] (same).


[69]  See The Contract Clause, supra note 68, at 147 (describing increased regulatory intervention in late 19th and early 20th centuries); History of Property Rights, supra note 68, at 8 (same).


[70]  See The Contract Clause, supra note 68, at 147; History of Property Rights, supra note 68, at 8. See also Herbert Hovenkamp, The Opening of American Law: Neoclassical Legal Thought, 1870-1970, at 277 (2014).


[71]  See, e.g., Norman Ware, The Industrial Worker, 1840-1860: The Reaction of American Industrial Society to the Advance of the Industrial Revolution 144-47 (Ivan R. Dee ed. 1990) (describing emergence of ten-hour legislation in states like Massachusetts and Pennsylvania in late 19th century in response to concern over working conditions in industrial workplaces); Herbert Hovenkamp, Federal Antitrust Policy: The Law of Competition and Its Practice §§ 2.1-2.2 (5th ed. 2020) (describing development of antitrust law in response to growing political and economic power of trusts).


[72]  See Randolph J. May, Defining Deference Down: Independent Agencies and Chevron Deference, 58 Admin. L. Rev. 429, 451 (2006) (“And it is odd in a constitutional system with three defined branches for courts to give controlling deference to agencies that, not without reason, are commonly referred to as ‘the headless fourth branch.’”).


[73]  See St. Antoine, supra note 24, at 1532.


[74]  Interstate Commerce Commission, U.S. National Archives, (last visited Aug. 27, 2023).


[75]  See Paul Stephen Dempsey, The Rise and Fall of the Interstate Commerce Commission: The Tortuous Path from Regulation to Deregulation of America’s Infrastructure, 95 Marq. L. Rev. 1151, 1152 (2012).


[76]  See ICC v. Ala. Midland Ry. Co., 168 U.S. 144, 175 (1897) (rejecting arguments that courts had to accept ICC’s interpretation of the Interstate Commerce Act and that courts had no authority to supplement the ICC’s evidentiary record).


[77]  Id. See also Merrill, supra note 15, at 951.


[78]  See Ala. Midland Ry. Co., 168 U.S. at 174-75 (explaining that courts not only had the power to “inquire into whether or not the commission has misconstrued the statute,” but that they could also accept “additional evidence” put forward by the parties and decide the case on “the entire body of evidence”).


[79]  See id. at 174 (explaining that the reviewing court’s role was to “proceed, as a court of equity, to hear and determine the matter, and in such manner as to do justice in the premises”).


[80]  See Merrill, supra note 15, at 953-54 (describing the “ICC crisis” of the late 19th century).


[81]  See id.


[82]  Pub. L. 59-337, 34 Stat. 584 (1906). See also Axon, No. 21-86, slip op. at 5 (Thomas, J., concurring) (reciting history that led to Act’s passage).


[83]  Id. ch. 3591, § 4. See also Axon, No. 21-86, slip op. at 5 (Thomas, J., concurring); Merrill, supra note 15, at 955-56.


[84]  See Axon, No. 21-86, slip op. at 5 (Thomas, J., concurring) (observing that Hepburn Act sent an implied message that courts should review ICC decisions less aggressively).


[85]  See ICC v. Ill. Cent. R. Co., 215 U.S. 452, 470 (1910); ICC v. Union Pac. R. Co., 222 U.S. 541, 547 (1912). See also Axon, No. 21-86, slip op. at 5 (Thomas, J., concurring) (explaining that the Court adopted a deferential standard of review for determinations of fact, but continued to insist on exercising plenary review over determinations of law).


[86]  Axon, No. 21-86, slip op. at 5 (Thomas, J., concurring).


[87]  Id.


[88]  Id.


[89]  See Ill. Cent. R. Co., 215 U.S. at 470 (explaining that while the Court would defer to the agency on matters within its “administrative functions,” it would continue to review the agency’s action to ensure it comported with the scope of congressional delegation and constitutional requirements).


[90]  285 U.S. 22 (1932).


[91]  Id. at 36-37 (citing Longshoremen and Harbor Workers’ Compensation Act, 44 Stat. 1424 (codified at 33 U.S.C. §§ 901-950)).


[92]  See id. at 42-44 (describing duties and powers of United States Employees’ Compensation Commission).


[93]  Id. at 47-55, 62-64.


[94]  Id. at 53-55.


[95]  Id. at 55-60.


[96]  Id.


[97]  See id. at 60 (“In cases brought to enforce constitutional rights, the judicial power of the United States necessarily extends to the independent determination of all questions, both of fact and law, necessary to the performance of that supreme function.”).


[98]  See Merrill, supra note 15, at 941 (pointing out that a “great preponderance” of administrative law is built on the appellate model articulated in Crowell).


[99]  See 5 U.S.C. § 706. See also Axon, No. 21-86, slip op. at 5-6 (Thomas, J., concurring) (noting that the appellate model was also built into the statutes creating the FTC and SEC); Merrill, supra note 15, at 965 (noting that appellate model had become “entrenched” by 1930s).


[100]  See, e.g., Union Carbide, 473 U.S. at 601 (explaining that the Crowell vision of appellate review “preserves the judicial authority over questions of law in the present context”); Stern v. Marshall, 564 U.S. 462, 508 (2011) (Scalia, J., dissenting) (explaining that in Crowell “the Court assumed that an Article III court would review the agency’s decision de novo in respect to questions of law”); Cf. FTC v. Gratz, 253 U.S. 421, 427 (1920) (“The words ‘unfair method of competition’ are not defined by the statute and their exact meaning is in dispute. It is for the courts, not the commission, ultimately to determine as matter of law what they include.”), overruled by FTC v. Brown Shoe Co., 384 U.S. 316 (1966).


[101]  See City of Arlington, Tex. v. FCC, 569 U.S. 290, 304 (2013) (rejecting “false dichotomy” between jurisdictional and nonjurisdictional issues); Hearst Publications, 322 U.S. at 130-31 (deferring to agency on statutory definition of “employee”a threshold issue dictating whether the agency had jurisdiction over the case). See also Merrill, supra note 15, at 966 (observing that courts soon forgot about the supposed distinction between ordinary and jurisdictional facts).


[102]  See Robbins Tire, 161 F.2d at 804 (Waller, J., concurring) (noting that by the late 1940s, there were “some half hundred boards and commissions” whose findings were reviewed only for substantial evidence).


[103]  See Hamburger, supra note 12, at 117, 318, 410 (describing and criticizing modern judicial deference to agency legal interpretations).


[104]  467 U.S. 837 (1984).


[105]  Id. at 842-43.


[106]  Id.


[107]  Id.


[108]  Merrill, supra note 15, at 940.


[109]  See, e.g., Stephens Media, 677 F.3d at 1250 (describing judicial deference to factual findings and legal determinations of NLRB); Wayneview Care Ctr., 664 F.3d at 348 (same). See also Lawson, supra note 15, at 1247 (pointing out that overlapping review doctrines have produced near-total deference).


[110]  See, e.g., Loper Bright Enters., Inc. v. Raimondo, No. 21-5166, slip op. at 12-16 (D.C. Cir. Aug. 12, 2022) (deferring to agency’s interpretation of statute simply because statute was ambiguous and agency’s position was “reasonable”); Clark Neily et al., Loper Bright Enterprises v. Raimondo, Cato Inst. (Dec. 9, 2022), (pointing to D.C. Circuit’s Loper opinion as an example of how courts now reflexively defer to agency interpretations); Hamburger, supra note 12, at 273-74 (criticizing weak judicial review of administrative orders as providing little protection against agency abuse). Cf. Harper, supra note 29, at 311 (arguing that in the 20th century, courts increasingly ignored the public/private rights distinction and accepted limitations on their power to review agency decisions).


[111]  See, e.g., Lawson, supra note 15, at 1231 (“The post-New Deal administrative state is unconstitutional, and its validation by the legal system amounts to nothing less than a bloodless constitutional revolution.”); Richard Epstein, The Classical Liberal Constitution loc. 936 (2017) (ebook) (arguing that modern administrative state is inconsistent with classical separation of powers).


[112]  See, U.S. Const. art. I § 1; Epstein, supra note 111, at loc. 936 (criticizing rulemaking by administrative agencies as a “complete inversion of the separation of powers”); Hamburger, supra note 12, at 38 (comparing agency rulemaking to Stuart crown’s abuse of royal proclamations outside ordinary lawmaking process).


[113]  See, e.g., Merrill, supra note 15, at 979 (“Modern constitutional law scholars frequently suggest that the appellate review model of administrative law violates the plain meaning of Article III of the Constitution.”); Hamburger, supra note 12, at 227 (“Scholars generally recognize that administrative adjudication is in tension with the Constitution’s grant of judicial power to the courts and its guarantee of due process and other procedural rights.”).


[114]  See Cox & Kaufman, supra note 15, at 1789 (describing “enormous” amount of “judicial” work now done by agencies); Mascott, supra note 15, at 43 (arguing that expansion of agency adjudication has exacerbated Article III problem: “the Constitution was not intended to permit executive agencies to resolve a number of the matters before them today”).


[115]  See, e.g., 29 C.F.R. pt. 18 (describing practices and procedures for Department of Labor’s Office of Administrative Law Judges); 16 C.F.R. pt. 3 (setting out rules of practice for FTC adjudicative proceedings); NLRB Casehandling Manual Part 1, Unfair Labor Practice Proceedings (2023) (describing agency’s procedures for investigating, prosecuting, and adjudicating unfair-labor-practice charges).


[116]  See Mascott, supra note 15, at 46-47 (reasoning that if private rights can be taken away only with judicial power, agencies should not adjudicate cases involving private rights or even act as adjuncts to courts in such cases); Lawson, supra note 15, at 1246 (observing that although one of the primary functions of modern agencies is to adjudicate disputes, administrative law judges lack all the features of Article III judges, such as lifetime tenure and salary protection). See also Harper, supra note 29, at 309 (observing that some “prominent scholars” have suggested that judicial review of Board decisions is necessary to preserve private rights).


[117]  See, e.g., Richard Fallon, Of Legislative Courts, Administrative Agencies, and Article III, 101 Harv. L. Rev. 915, 916-18 (1988); John Dickenson, Administrative Justice and the Supremacy of the Law 42 (1927) (“It is sometimes said or assumed that public officers have no jurisdiction to determine questions of law and that therefore, as to these, their action is not final, but is subject to court review.”); Harper, supra note 29, at 290, 317-18. See also Lawson, supra note 15, at 1247 (surveying views of pro-review scholars); Merrill, supra note 15, at 976 (attributing to Dickenson the idea that “[j]udicial review cures all”).


[118]  See Fallon, supra note 117, at 916-18 (arguing that Congress should have discretion over when to delegate adjudication to agenciesas long as it provides for judicial review). Harper, supra note 29, at 266, 317-18 (arguing that appellate-style review is not only sufficient to satisfy Article III, but even narrower review would also satisfy the Constitution).


[119]  See Fallon, supra note 117, at 918 (arguing that “adequately searching appellate review of the judgments of legislative courts and administrative agencies is both necessary and sufficient to satisfy the requirements of article 11”). Cf. Harper, supra note 29, at 309 (arguing that belief that judicial review is necessary to protect individual rights and comport with Article III has influenced judicial behavior toward agencies).


[120]  See Axon, No. 21-86, slip op. at 8-9 (Thomas, J., concurring); Lawson, supra note 15, at 1247.


[121]  Lawson, supra note 15, at 1247 (arguing that the possibility of review does not cure the Article III problem because review is too deferential to be meaningful).


[122]  See id. (“Article III would certainly not be satisfied if Congress provided for judicial review but ordered the courts to affirm the agency no matter what. . . . There is no reason to think that it is any different if Congress instead simply orders courts to put a thumb (or perhaps two forearms) on the agency’s side of the scale.”); Hamburger, supra note 12, at 282 (criticizing judicial-review justification because judges are not “apt to do much more than defer”).


[123]  See Axon, No. 21-86, slip op. at 8 (Thomas, J., concurring) (“It is no answer that an Article III court may eventually review the agency order and its factual findings under a deferential standard of review.”); Baldwin v. United States, No. 19-402, slip op. at 4 (U.S. Feb. 24, 2020) (Thomas, J., dissenting from denial of cert.) (“When the Executive is free to dictate the outcome of cases through erroneous interpretations, the courts cannot check the Executive by applying the correct interpretation of the law.”). Cf. Robbins Tire, 161 F.2d at 805 (Waller, J., concurring) (arguing that when courts refuse to examine agency factfinding, they render judicial review “impotent”).


[124]  Compare Lewis Jaffe, Judicial Control of Administrative Action 381-89 (1965) (arguing that courts must exercise plenary review over the matters falling within their jurisdiction, making deference inappropriate), and Henry Hart, The Power of Congress to Limit the Jurisdiction of Federal Courts: An Exercise in Dialectic, 66 Harv. L. Rev. 1362, 1374-86 (1953) (same), with City of Arlington, 569 U.S. at 304-07 (applying Chevron deference even to rules touching on agency’s jurisdiction), and NLRB v. Bell Aerospace Co. Div. of Textron, 416 U.S. 267, 294 (1974) (deferring to Board’s interpretation of statute as well as its judgment on what kind of procedure (rulemaking or adjudication) to use in adopting that interpretation). See also Harper, supra note 29, at 310-11 (noting that scholarly attacks on appellate model had, to that point, failed to gain traction).


[125]  See, e.g., City of Arlington, 569 U.S. at 304-07 (describing deferential, appellate-style review); Bell Aerospace, 416 U.S. at 294 (same).


[126]  Cf. Lawson, supra note 15, at 1232 (observing that the broad outlines of the administrative state had been accepted by administrations and high-level officials in both parties).


[127]  See Jarkesy, No. 20-61007, slip op. at 5-15.


[128]  See Compl., FTC v. Int’l Exchange, Inc., Case No. 3:23-cv-01710 (N.D. Cal. Apr. 10, 2023), ECF No. 1. See also Dan Papscun, Black Knight Sues FTC Over Constitutionality of In-House Judge, Bloomberg Law (Apr. 26, 2023), (reporting on lawsuit challenging constitutionality of agency’s adjudication procedures).


[129]  Sun Valley Orchards, No. 1:21-cv-16625, 2023 BL 257772, at *6-7; Compl., C.S. Lawn & Landscaping, Case No. 1:23-cv-01533, ECF No. 1.


[130]  See, e.g., Jarkesy, No. 20-61007, slip op. at 5-15.


[131]  See, e.g., Compl., C.S. Lawn & Landscaping, Case No. 1:23-cv-01533, ECF No. 1.


[132]  Id.


[133]  Id.


[134]  See Sun Valley Orchards, No. 1:21-cv-16625, 2023 BL 257772, at *6-7; Compl., C.S. Lawn & Landscaping, Case No. 1:23-cv-01533, ECF No. 1.


[135]  See Jarkesy, No. 20-61007, slip op. at 5-15.


[136]  Id.


[137]  See id. See also Golde, supra note 19 (placing Jarkesy in the context of a judicial trend of skepticism toward administrative power).


[138]  See 29 U.S.C. §§ 153-156 (describing establishment, powers, and duties of NLRB). See also Gould, supra note 20, at 1506 (describing Board’s creation as an “independent agency in the executive department”).


[139]  29 U.S.C. § 153(a).


[140]  Beth Israel Hosp., 437 U.S. at 500 (“It is the Board on which Congress conferred the authority to develop and apply fundamental national labor policy.”).


[141]  See 29 U.S.C. § 156 (authorizing the Board to issue regulations to implement NLRA).


[142]  See 29 U.S.C. § 160(c) (authorizing the Board to adjudicate unfair-labor-practice charges and issue written cease-and-desist orders); Bell Aerospace, 416 U.S. at 294 (recognizing that Board may develop policy either through rulemaking or adjudication); NLRB v. Wyman-Gordon Co., 394 U.S. 759, 766 (1969) (affirming Board’s decision to require employers to furnish names and addresses of employees to union through adjudication rather than rulemaking); See James J. Brudney, Isolated and Politicized: The NLRB’s Uncertain Future, 26 Comp. Lab. L. & Pol’y J. 221, 234 (2005) (“[O]ver its seventy year history the Board has chosen to operate virtually exclusively through adjudication, eschewing its rulemaking authority.”).


[143]  See, e.g., Gould, supra note 20, at 1506 (describing the Board as a “quasi-judicial agency”); Brudney, supra note 142, at 234-35 (describing Board’s case-by-case approach to policymaking). Cf. Int’l Org. of Masters, Mates & Pilots, ILA, AFL-CIO v. NLRB, 61 F.4th 169, 178 (D.C. Cir. 2023) (explaining that while Board can depart from its own “precedent,” it must offer “reasoned justifications” for doing so).


[144]  See 29 U.S.C. § 160(e), (f); 29 C.F.R. § 101.14.


[145]  National Labor Relations Act, Pub. L. 74-198 § 10, 49 Stat. 449, 454 (1935). See also Note, Effect of the Taft-Hartley and Administrative Procedure Acts on Scope of Review of Administrative Findings, 26 Ind. L.J. 406, 406 n.4 (1951) (discussing original standard and change made by Taft-Hartley Amendments).


[146]  See Labor-Management Relations Act of 1947, Pub. L. 80-101, 61 Stat. 136, 148 (amending 29 U.S.C. § 160(e), (f)).


[147]  See Universal Camera, 340 U.S. at 488-90. See also Robbins Tire, 161 F.2d at 804 (Waller, J., concurring) (concluding that substantial-evidence review fails to provide a real check on erroneous factfinding) (“Any evidence, however incredible, is substantial if it is adjudged to have been believed by the Examiner or the Board.”).


[148]  322 U.S. at 130-31.


[149]  See id. at 125-26 (explaining that national uniformity was essential to Congress’s scheme).


[150]  Id. at 130-31.


[151]  Id.


[152]  Id.


[153]  Id.


[154]  Id.


[155]  See id. (“[W]here the question is one of specific application of a broad statutory term in a proceeding in which the agency administering the statute must determine it initially, the reviewing court’s function is limited.”).


[156]  See id. (deferring to Board’s interpretation of “employee” under the NLRA); Universal Camera, 340 U.S. at 488-90 (explaining that courts must accept Board’s factual findings when supported by substantial evidence). See also St. Antoine, supra note 24, at 1531-37 (tracking deference to Board’s decisions before the APA and Chevron).


[157]  See id.


[158]  See Robbins Tire, 161 F.2d at 803 (Waller, J., concurring) (describing Board’s procedures as “judicial, or quasi-judicial”).


[159]  Axon, No. 21-86, slip op. at 5 (Thomas, J., concurring).


[160]  Id.


[161]  See Robbins Tire, 161 F.2d at 803 (Waller, J., concurring) (recognizing that the Board’s orders “involve the personal and property rights of citizens”).


[162]  See, e.g., David Saxe Prods., LLC, 370 N.L.R.B. No. 103, 2021 WL 1293347, at *5 (Apr. 5, 2021) (describing Board’s recent nonsolicitation precedent); UPMC, 368 N.L.R.B. No. 2, 2019 WL 2502063, at *3 (June 14, 2019) (same).


[163]  See, e.g., Register Guard, 351 N.L.R.B. 1110, 1114 (2007) (holding that employees had no right to access employer’s email system for protected activity); Purple Commc’ns, Inc., 361 N.L.R.B 1050, 1055 (2014) (reaching opposite conclusion and overruling Register Guard).


[164]  See Lechmere, Inc. v. NLRB, 502 U.S. 527, 537 (1992) (discussing Board’s balancing approach concerning organizers’ access to employer property).


[165]  See id. See also Jean Country, 291 N.L.R.B. 11, 14 (1988) (balancing organizers’ need to access employees against employer’s property rights), abrogated in part by Lechmere, 502 U.S. at 536; William R. Corbett, Awaking Rip Van Winkle: Has the National Labor Relations Act Reached A Turning Point?, 9 Nev. L.J. 247, 257 (2009) (surveying judicial and administrative attempts to balance statutory rights with property rights).


[166]  See Anne Marie Lofaso, Toward a Foundational Theory of Workers’ Rights: The Autonomous Dignified Worker, 76 UMKC L. Rev. 1, 27 (2007) (examining the role property rights have played in defining labor rights and attempts courts and agencies have made to balance the two).


[167]  29 U.S.C. § 159(a). See also Hearst Publications, 322 U.S. at 126-34 (noting that the “avowed purpose” of the NLRA was to promote collective bargaining, and that Congress delegated responsibility for advancing that purpose primarily to the Board).


[168]  Vaca v. Sipes, 386 U.S. 171, 182 (1967) (“The collective bargaining system as encouraged by Congress and administered by the NLRB of necessity subordinates the interests of an individual employee.”); Emporium Capwell Co. v. W. Addition Cmty. Org., 420 U.S. 50, 62 (1975) (“Congress sought to secure to all members of the unit the benefits of their collective strength and bargaining power, in full awareness that the superior strength of some individuals or groups might be subordinated to the interest of the majority. . . .”); J.I. Case Co. v. NLRB, 321 U.S. 332, 338 (1944) (holding that direct bargaining between employers and employees in workplace with certified bargaining representative violates the NLRA).


[169]  J.I. Case, 321 U.S. at 338.


[170]  See id.


[171]  Vaca, 386 U.S. at 182.


[172]  See id.


[173]  Emporium Capwell, 420 U.S. at 62.


[174]  See A.L.A. Schechter Poultry Corp. v. United States, 295 U.S. 495, 550 (1935).


[175]  301 U.S. 1, 47-48 (1937).


[176]  Id. at 29.


[177]  Id. at 29-43.


[178]  Id. at 47.


[179]  Id.


[180]  See id.


[181]  See id. (assuming that “all questions of constitutional right or statutory authority are open to examination by the court”).


[182]  See Lawson, supra note 15, at 1246 (arguing that excessive deference has caused the “death of the independent judiciary”).


[183]  See, e.g., Beth Israel Hosp., 437 U.S. at 501 (describing reviewing court’s role as “narrow”: “The rule which the Board adopts is judicially reviewable for consistency with the Act, and for rationality, but if it satisfies those criteria, the Board’s application of the rule, if supported by substantial evidence on the record as a whole, must be enforced.”); King Soopers, Inc. v. NLRB, 859 F.3d 23, 37 (D.C. Cir. 2017) (“The Board is entitled to considerable deference in crafting remedies for unfair labor practices . . . .”).


[184]  See Hamburger, supra note 12, at 410 (pointing out that overlapping deference doctrines have hollowed out judicial review) (“[W]hen judges defer to administrative interpretations, it becomes difficult to take seriously the idea that the judges are authoritative expositors of the law.”).


[185]  See id.


[186]  See 29 U.S.C. § 160(c) (authorizing Board to issue cease-and-desist orders); U.S. Postal Serv., 360 N.L.R.B. 181, 181 (2014) (issuing cease-and-desist order to remedy unfair labor practice).


[187]  See U.S. Postal Serv., 360 N.L.R.B. at 181 (ordering respondent to post notice of violations).


[188]  See 29 U.S.C. § 160(c) (authorizing Board to “take such affirmative action including reinstatement of employees with or without backpay, as will effectuate the policies of this Act”); Thryv, Inc., 372 N.L.R.B. No. 22, slip op. at 1 (Dec. 13, 2022) (holding that make-whole remedies for unfair labor practices should include “all direct or foreseeable pecuniary harms that . . . employees suffer as a result of the respondent’s unfair labor practice”).


[189]  See Protecting the Right to Organize Act, H.R. 842, 117th Cong. (1st Sess. 2021-22) (proposing to amend NLRA to add monetary penalties); James Paretti et al., U.S. House Poised to Add Civil Penalties to National Labor Relations Act, Littler Mendelson (Sept. 9, 2021), (describing failed attempt to authorize Board to impose civil penalties for unfair labor practices).


[190]  See, e.g., Steven Greenhouse, Will Starbucks’ Union Busting Stifle a Union Rebirth in the US?, Guardian (Aug. 23, 2023), (arguing that the Board’s remedies are too weak to deter companies from violating labor law); Hamilton Nolan, It’s Up to Unions to Make the NLRB Matter, In These Times (Aug. 28, 2023), (arguing that inability to impose financial penalties reduces Board’s effectiveness and incentivizes companies to disregard the Board’s standards).


[191]  U.S. Const. amend. VII.


[192]  See, e.g., N. Pipeline Const., 458 U.S. at 70 n.25; Jarkesy, No. 20-61007, slip op. at 5-15. See also Margaret L. Moses, What the Jury Must Hear: The Supreme Court’s Evolving Seventh Amendment Jurisprudence, 68 Geo. Wash. L. Rev. 183, 194-217 (2000) (surveying modern Supreme Court’s treatment of Seventh Amendment and the Court’s search for historical analogues).


[193]  Jarkesy, No. No. 20-61007, slip op. at 5-15.


[194]  Id. (holding that SEC could not constitutionally impose civil penalties through administrative proceedings because defendants had a Seventh Amendment right to defend themselves in court before a jury). But cf. Sun Valley Orchards, 2023 BL 257772 at *7 (concluding that availability of civil remedies did not itself mean that DOL’s H-2B visa program involved private rights within the meaning of Article III).


[195]  Broadnax v. City of New Haven, 415 F.3d 265, 271 (2d Cir. 2005).


[196]  Id. But see Nelson, supra note 15, at 602 (explaining that courts struggled with the distinction in the first half of the 20th century, with some even finding the Board’s structure unconstitutional under the Seventh Amendment (citing NLRB v. Mackay Radio & Tel. Co., 87 F.2d 611, 630-31 (9th Cir. 1937)).


[197]  See Jones, 301 U.S. at 48-49 (holding that the Board’s structure did not violate the Seventh Amendment because it did not allow the Board to adjudicate any “suit at common law”).


[198]  See Curtis v. Loether, 415 U.S. 189, 194-96 (1974) (holding that Seventh Amendment guaranteed access to jury for fair-housing claims under Civil Rights Act of 1968, even though those claims were statutory, because their monetary remedies resembled actions at common law). But cf. Atlas Roofing Co., Inc. v. Occupational Safety & Health Rev. Comm’n, 430 U.S. 442, 450 (1977) (suggesting that distinction drawn by Jones depends less on whether an agency can impose civil penalties than on the nature of the rights involvedpublic or private).


[199]  See Axon, No. 21-86, slip op. at 7 (Thomas, J., concurring) (explaining that the modern appellate-review model may violate Article III because it allows agencies to effectively adjudicate core private rights).


[200]  See John Higgins Jr., et al., The Developing Labor Law § 1.IV (8th ed. 2022) (describing historical development of judicial regulation of labor law, the perceived inadequacies of which led to statutory solutions and administrative responsibility).


[201]  See Amy Howe, Supreme Court Will Review Constitutionality of Consumer-Watchdog Agency’s Funding, SCOTUSBlog (Feb. 27, 2023), (reporting that Supreme Court will consider constitutional challenge to CFPB’s funding mechanisms in 2023-24 term).


[202]  See, e.g., Harper, supra note 29, at 313-18 (reviewing longstanding arguments for and against expanded judicial review of Board orders); Gould, supra note 20, at 1505 (noting that the Board has long acted as an “effective substitute” for judicial proceedings).


[203]  Cf. Oliver Wendell Holmes, The Path of the Law 167, 186 (1920) (“[I]f we want to know why a rule of law has taken its particular shape, and more or less if we want to know why it exists at all, we go to tradition.”).


[204]  See Hamburger, supra note 12, at 11, 24, 412 (arguing that constitutional problems associated with agency power cannot be dismissed simply because courts have acquiesced in them for years).


[205]  See id. at 488 (“Ultimately, time is no cure.”). Cf. Merrill, supra note 15, at 987 (observing that to the extent the private-rights issue is a constitutional problem, it has been one and has been ignored by the court since at least 1906).


[206]  Hunter’s Lessee, 14 U.S. at 330.


[207]  See Axon, No. 21-86, slip op. at 7 (Thomas, J., concurring) (stating that the appellate-review model “may violate Article III by compelling the Judiciary to defer to administrative agencies regarding matters within the core of the Judicial Vesting Clause”). Cf. Hamburger, supra note 12, at 397 (reasoning that agencies cannot properly exercise judicial power, even delegated power, because “[a]t common law . . . judicial power was inalienable”).


[208]  See Hamburger, supra note 12, at 8, 55 (tracing limits on executive power to founders’ fear of prerogative courts); id. at 132-33 (explaining that Article III “emphatically reiterated the constitutional bar to any extralegal adjudication”).


[209]  Id. at 5, 130, 133-42.


[210]  Id. at 157-59, 249-51.


[211]  Id.


[212]  Cf. id. (arguing that modern administrative agencies revive many of the prerogative courts’ abuses).


[213]  See, e.g., Raoul Berger, Retroactive Administrative Decisions, 115 U. Pa. L. Rev. 371, 385–86 (1967) (“Retroactive announcements by the NLRB of changes in existing rules respecting the jurisdiction it will exercise in the future exhibit so erratic and capricious a course as to shake confidence in its judgment that retroactivity was essential.”); The Atlanta Opera Inc., 372 N.L.R.B. No. 95, at *20 (requiring defendant to prove that certain workers were independent contractors, not employeesa question going to the heart of the Board’s jurisdiction over a labor dispute).


[214]  See NLRB v. Guy F. Atkinson Co., 195 F.2d 141, 148 (9th Cir. 1952) (refusing to enforce Board order reflecting a change in the Board’s jurisdictional policy, which would have penalized the respondent for engaging in conduct it reasonably believed was lawful at the time) (“The inequity of such an impact of retroactive policy making upon a respondent innocent of any conscious violation of the act, and who was unable to know, when it acted, that it was guilty of any conduct of which the Board would take cognizance, is manifest.”).


[215]  See, e.g., Brudney, supra note 142, at 223, 243-52 (tracing criticisms of Board’s politically tilted decision-making as far back as 1939 and noting an increase in politization in recent decades); St. Antoine, supra note 24, at 1529 (abstract) (noting that empirical studies have shown that the political backgrounds of Board members influence their decisions).


[216]  See, e.g., Gould, supra note 20, at 1506 (noting that because presidents have used their Board appointments to change labor policy quickly, observers have described the Board as a policy “seesaw”); Robert Iafolla, NLRB Dials Back Employers’ Authority to Act Unilaterally, Bloomberg Law (Aug. 30, 2023), (quoting Ginger Schroder, an employment lawyer representing management,, saying that “no employer or union can rely on NLRB precedent because the board is partisan and will flip-flop after control of the White House changes from party to party”).


[217]  See also Michael J. Lotito, Maury Baskin & Missy Parry, Coalition for a Democratic Workforce, Was the Obama NLRB the Most Partisan in History? 3 (2016), (observing that in “no case where the [Obama] Board overturned, or substantially modified, important principles did a Republican member join with the Democratic majority”).


[218]  Lotito et al., supra note 217, at 1-7.


[219]  See, e.g., The Boeing Co., 365 N.L.R.B. No. 154, slip op. at 2 (Dec. 14, 2017) (overruling prior standards for judging lawfulness of facially neutral employer work rules); Hy-Brand, 365 N.L.R.B. No. 156, slip op. at 1 (2017) (overruling standard announced in Browning-Ferris Industries, 362 N.L.R.B. No. 186 (2015) for determining when two entities will be considered joint employers).


[220]  See, e.g., Miller Plastic Products Co., 72 N.L.R.B. No. 134, slip op. at 1 (Aug. 25, 2023) (reversing Trump Board’s decision in Allstate Maintenance Insurance, LLC, 367 N.L.R.B. No. 68 (2019), and adopting a broader, all-relevant-circumstances standard for determining when an employee is engaged in conduct protected by the NLRA); Wendt Corp., Nos. 03–CA–212225, 03–CA–220998, and 03–CA–223594, slip op. at 1 (N.L.R.B. Aug. 26, 2023) (overruling Trump Board’s decision in Raytheon Network Centric Systems, 365 N.L.R.B. No. 161 (2017), and limiting an employer’s ability to make unilateral changes in accord with a past practice before a first contract takes effect or after a contract has expired); Tenocap LLC, 372 N.L.R.B. No. 136, slip op. at 1 (Aug. 26, 2023) (overruling a separate part of Raytheon and limiting an employer’s right to make unilateral changes consistent with an expired management-rights clause); Stericycle, Inc., 372 N.L.R.B. No. 113, slip op. at 1-2 (Aug. 2, 2023) (reversing Trump Board’s decision in Boeing Co., 365 N.L.R.B. No. 154 (2017), and adopting a stricter standard for determining when a work rule will be held unlawful because it may chill protected activity).


[221]  Lotito et al., supra note 217, at 5 (noting that defenders of the Board sometimes point to policy oscillation as not only legitimate, but also a justification of the Board’s shifts in doctrine). See also Brudney, supra note 142, at 227 (noting Board’s “ability and willingness to so readily depart from its own precedent” in response to political pressures); Lamons Gasket Co., 357 N.L.R.B. 739, 748 (2011 (Hayes, dissenting) (arguing that the Board’s decision to change its doctrine with respect to the opportunity to petition for an election after a voluntary recognition was a “purely ideological policy choice, lacking any real empirical support and uninformed by agency expertise”). Cf. Hamburger, supra note 12, at 433 (criticizing exercise of policy discretion by agencies on grounds that “administrative discretion leaves Americans insecure in their freedom”).


[222]  Cf. Harper, supra note 29, at 299 (noting that while it is hard for a president to change views of entire judiciary through appointments, it is relatively easy to do so with the Board, and pointing to that ease as a feature, not a bug).


[223]  Cf. Axon, No. 21-86, slip op. at 8 (Thomas, J., concurring) (reasoning that the appellate-review model may violate due process because it denies people the opportunity to have their core private rights determined in court).


[224]  See id. at 9 (“If private rights are at stake, the Constitution likely requires plenary Article III adjudication.”).


[225]  Hamburger, supra note 12, at 411 (explaining that structural assignment of powers to specialized branches was central to founders’ idea of due process and essential to their scheme for preventing consolidation of power).


[226]  See id. at 412 (explaining that the separation of powers reflects the founders’ ideal of the rule of law: fair, due process).


[227]  See Axon, No. 21-86, slip op. at 3-9 (Thomas, J., concurring) (observing that modern agency adjudication may violate separation of powers and, specifically, Article III).

Note from the Editor: The Federalist Society takes no positions on particular legal and public policy matters. Any expressions of opinion are those of the author. We welcome responses to the views presented here. To join the debate, please email us at [email protected].