Saving Agency Adjudication

Introduction

When it comes to the federal judiciary, commentators often fixate on those 800 or so “Article III” judges who are appointed by the President with the advice and consent of the Senate.^[1] During his first term in office, President Trump nominated and the Senate confirmed 234 judges to Article III federal courts, including three Supreme Court justices, 54 circuit court judges, 174 district court judges, and three judges on the U.S. Court of International Trade.^[2] President Biden appointed 235 judges to life-tenured positions in the Article III judiciary, including one Supreme Court justice, 45 circuit court judges, and 187 district court judges.^[3] Enormous attention and resources have been dedicated to these judicial confirmations, including millions of dollars and thousands of hours by interest groups such as the American Bar Association (ABA).^[4]

If we care about the federal judiciary, however, focusing on Article III courts alone is myopic. The federal judiciary today expands far beyond the small group of judges with Article III protections.^[5] The overwhelming bulk of federal adjudication today takes place in federal agencies. The federal administrative judiciary is expansive, comprised of at least 1,900 administrative law judges (ALJs)^[6] and more than 10,000 non-ALJ agency adjudicators who preside over evidentiary hearings.^[7] The latter have been given a range of titles, such as administrative judge, hearing officer, immigration judge, patent examiner, and presiding officer.^[8] Agency adjudicators scattered across the federal government collectively issue millions of decisions per year on subjects ranging from government contracts and veterans benefits to immigration and patent rights.^[9] And yet, as one of us has observed, “there is no ABA committee that rates proposed immigration judges or other agency adjudicators. There are no television ads run. The Senate plays no role in their selection—though Congress of course retains its oversight and appropriations authority.”^[10]

Despite the lack of public attention, scholars and agency adjudicators in recent years have raised alarms that the federal administrative judiciary may be reaching a crisis point.^[11] Following the Supreme Court’s separation-of-powers lead, federal courts have begun holding that how agency adjudicators are appointed and removed violates the Constitution because, despite being part of the Executive Branch, these officials are not sufficiently subject to the President’s control.^[12]

The Supreme Court’s decision to hear SEC v. Jarkesy^[13] last Term highlights the stakes. In Jarkesy, a divided Fifth Circuit panel held that how the Securities and Exchange Commission (SEC) is structured violates the constitutional rule from Free Enterprise Fund v. Public Co. Accounting Oversight Board^[14] that Congress cannot impose two levels of removal protection between certain Executive Branch officers and the President.^[15] After all, the Fifth Circuit reasoned, the President cannot remove SEC Commissioners at will, and those Commissioners in turn cannot remove the SEC’s ALJs at will, thereby—in the Fifth Circuit’s view—preventing the President from “tak[ing] care that the laws are faithfully executed.”^[16] In dissenting from the denial of rehearing en banc, Judge Catharina Haynes lamented that the panel decision “deviated from over eighty years of settled precedent.”^[17] Indeed, commentators openly feared that “[u]nless overturned, the [Fifth Circuit’s] decision will be a sea change in both the regulation of the financial industry and administrative law.”^[18] Ultimately, the Supreme Court affirmed the Fifth Circuit’s decision in Jarkesy on the Seventh Amendment jury trial issue, leaving for another day the dual-layer removal issue.^[19]

Although the Fifth Circuit’s “rigid, categorical” approach to the SEC’s structure is controversial in some circles,^[20] Jarkesy’s application of Free Enterprise Fund was no surprise. To the contrary, scholars have seen this development coming for more than a decade,^[21] and Judge Neomi Rao on the D.C. Circuit had already reached the same conclusion.^[22] Furthermore, in his dissent in Free Enterprise Fund itself, Justice Stephen Breyer warned about this potential implication for agency adjudication,^[23] and subsequent Supreme Court decisions have only heightened the risk.^[24]

Merely because the Fifth Circuit’s Jarkesy decision was predicted, however, does not diminish its importance. Rather, it vividly illustrates a truth about the modern administrative state: Agency adjudication—at least as understood since the enactment of the Administrative Procedure Act (APA) in 1946—is in danger.^[25] Taken to their logical conclusion, the Supreme Court’s recent cases require the President to play a significant role (either directly or through an agency head under the President’s control) in the selection of all important players in the Executive Branch, including agency adjudicators.^[26] Perhaps even more significantly, these decisions suggest that the President must be able to remove every significant Executive Branch official, again, including perhaps adjudicators.^[27] Yet reformers worry that such political control of what is supposed to be a neutral, individualized process may undermine a central tenet of fair adjudication: decisional independence.^[28] If an adjudicator can be hired or fired for not sharing the politics or even predilections of the White House, parties may fear that the adjudicator is not ruling based on law and a matter’s individual facts but instead out of concern of being fired. Indeed, such political control of adjudication, at least with respect to certain rights and interests, may itself violate due process.^[29]

Hence the dilemma: The Supreme Court’s reading of Article II mandates that the President be able to fully control the Executive Branch—otherwise, “the buck would stop somewhere else”^[30]—but such control in theory could prevent fair adjudication. How can political control and decisional independence in adjudication co-exist?^[31]

Recognizing that the unitary executive theory of Article II is potentially on a collision course with decisional independence, scholars and other reformers have proposed several ways to save agency adjudication.^[32] Three main approaches have emerged: creating a new “central panel” agency to house most or all agency adjudicators;^[33] expanding the Article I courts to include more regulatory areas;^[34] or placing more or perhaps even all agency adjudication in Article III courts.^[35] Each of these reforms is motivated by a desire to solve the potential dilemma between presidential control and decisional independence. Unfortunately, we are not confident that any of them will work, at least as a general manner. They either will not solve the constitutional problem, will create massive unintended consequences, or, often, both.

So, can agency adjudication be saved? Yes, but part of the saving must be avoiding these reform proposals, especially during this era of uncertainty as to the scope of the problem and the impact of the sweeping proposals themselves. Instead, saving agency adjudication requires embracing a key insight: While the Constitution may pose threats to decisional independence, it also offers solutions that do not require fundamentally overhauling agency adjudication.

To begin, Congress should use what two of us (Nielson and Walker) have dubbed its “anti-removal power,” i.e., the power to increase the president’s political costs of removal.^[36] By design, the Constitution provides Congress with instruments like the Appointments Clause that allow Congress (if it wishes to do so) to create a measure of de facto independence for Executive Branch officials. In the context of agency adjudication, Congress could make some adjudicators Senate-confirmed.^[37] It can also require agency heads to give reasons for removing adjudicators, including in targeted congressional hearings. Although this anti-removal power does not strip the White House of its power to remove agency officials, it often will discourage removal—especially for less-salient positions for which political norms against removal already exist, such as agency adjudicators.^[38]

The Presidency itself also has a role to play. As Kent Barnett has explained, the President’s Article II power to control the Executive Branch includes the power not to control it.^[39] This means the Executive Branch can unilaterally regulate itself to prevent political interference with adjudication, both in terms of appointment and removal.^[40] In other words, as the Trump Administration proclaimed during its first term, merely because agency leaders have a say in who can serve as agency adjudicators, it does not follow that such leaders should use that authority to undermine impartiality.^[41] Even though the White House could theoretically undo such impartiality regulations, the mere act of formalizing them would discourage interference.^[42]

Such an internal check often should be enough to preserve decisional independence,^[43] especially when combined with Congress’s anti-removal power. This combined approach therefore should safeguard the decisional independence of agency adjudicators without prompting significant constitutional objections and unintended consequences, thus saving agency adjudication from both Article II concerns and misguided reforms.

The Article proceeds as follows: Part I provides an overview of the standard or default model for agency adjudication established in the APA, which includes hearing-level adjudications by impartial, decisionally independent agency adjudicators, followed by the availability of de novo agency head review and deferential judicial review in an Article III court. Part II details the dual reasons for this perceived crisis of decisional independence: the expanding statutory and regulatory exceptions to APA-governed formal adjudication and the Roberts Court’s embrace of the unitary executive theory in its separation of powers precedents. Part III introduces and critiques the three main reform proposals to date: a centralized Article II administrative judiciary; the creation of more specialized Article I legislative courts; and the shift of agency adjudication to Article III federal courts. Part IV concludes by introducing our two-fold reform proposal: Congress’s use of its anti-removal power and the President’s use of internal administrative law to create impartiality regulations for the hiring and firing of agency adjudicators. These two reforms, we argue, could address the perceived dilemma in ways that avoid the costs of the other proposals while still producing similar benefits.

I. The Standard Model for Agency Adjudication

The standard model for agency adjudication, which is delineated by the APA, strikes a careful balance between decisional independence of adjudicators and political control over agency adjudication.^[44] As Emily Bremer has explained, the APA sought to rectify several deficiencies with pre-APA administrative hearings, including concerns with arbitrator neutrality and the need for agency head control over adjudicatory outcomes.^[45]

The APA distinguishes “formal” adjudication from all other types of “informal” adjudication, the former of which was envisioned as the standard model for hearing-based adjudication.^[46] Paradigmatic APA-governed formal adjudication requires an evidentiary hearing held before an ALJ in which parties are entitled to oral arguments, rebuttal, and cross-examination of witnesses.^[47] An ALJ conducting a formal adjudicatory hearing is functionally analogous to a trial judge presiding over a bench trial.^[48] The ALJ is the principal factfinder and initial decisionmaker in an agency adjudication, and the APA generally empowers ALJs to “regulate the course of the hearing.”^[49] The ALJ’s decision is then subject to de novo agency head review.^[50]

Although the standard model for agency adjudication has been detailed before, our focus here is on the balance between two key structural features: a hearing before an impartial agency adjudicator and a final decision by a politically accountable agency head. By impartial agency adjudicator, we mean an adjudicator who faithfully applies the law to facts and makes unbiased factfinding. Importantly, the second key feature—the potential for review and final decision by the agency head—serves as a mechanism by which political control is infused into agency adjudication. However, as this Part highlights, the standard model enables a specific method for political control of agency adjudication, which is both transparent and circumscribed—and which ensures that the administrative record is compiled, and initial findings and decisions are made by an impartial agency adjudicator.

A. Hearing-Level Adjudicator Decisional Independence

The first key structural feature of the standard model of agency adjudication is that the hearing-level adjudicator has decisional independence. This decisional independence enables the adjudicator to create an administrative record and make findings free from political interference.

At the outset, we note that we use the term “decisional independence,” as that is the term of art in the legal literature. But our conception of the term here of decisional independence extends beyond protection against political interference with ALJ decisionmaking—interference that can come from the threat of promotion, demotion, and firing. It also entails that the agency adjudicators are hired based on merit. In that sense, the “neutral competence” framing from the public administration literature may better capture what we mean by decisional independence.^[51]

When it comes to decisional independence, the APA explicitly declares that ALJs must be unbiased, stating “[t]he functions of presiding employees and of employees participating in decisions . . . shall be conducted in an impartial manner.”^[52] Moreover, the statute envisions the disqualification of ALJs that cannot meet the impartiality requirement.^[53] Importantly, the APA has several additional requirements that reflect the due process concern of ensuring a meaningful opportunity to be heard before an unbiased adjudicator.^[54] For instance, the APA prohibits the adjudicator from engaging in ex parte communications about the case “unless on notice and opportunity for all parties to participate.”^[55] Nor can the adjudicator “be responsible to or subject to the supervision or direction of an employee or agent engaged in the performance of investigative or prosecuting functions for an agency.”^[56] This latter restriction sought to counter the rampant comingling of such functions that occurred pre-APA.^[57]

The APA also requires the ALJ’s decision to be based exclusively on the record created at the hearing,^[58] supported by “reliable, probative, and substantial evidence,”^[59] and include a statement of “findings and conclusions, and the reasons or basis therefor, on all the material issues of fact, law, or discretion presented on the record.”^[60] Finally, the APA provides that ALJs “be assigned to cases in rotation so far as practicable,”^[61] limiting the ability of the agency to control the adjudicatory outcomes through adjudicator assignment decisions.^[62]

To further ensure decisional independence or neutral competence, Congress sharply limited agency control over selection, retention, and removal of ALJs. Until recently, ALJs were selected through a competitive appointments process that was overseen by the Office of Personal Management (OPM). OPM ranked ALJ candidates on several factors, including examination scores,^[63] and then created a list of the three highest scoring candidates from which the agency can select its ALJ.^[64] That changed in 2018 when an executive order exempted all ALJs from the OPM competitive selection process and civil service statutes more generally,^[65] the latter of which prohibited employment decisions to be based on partisanship.^[66] That order, however, also stresses that ALJs should remain “impartial” and that the order “shall be implemented in a manner consistent with applicable law.”^[67] “OPM immediately authorized heads of executive departments to make ALJ appointments without OPM approval,”^[68] thus concentrating ALJ hiring process fully within the agency.

ALJs also continue to be protected from some aspects of the employment relationship that might impair their independence as adjudicators. For example, ALJs are still exempt from the Civil Service Reform Act’s performance appraisal requirements, which apply to most federal employees.^[69] As a result, ALJs’ pay is not tied to performance reviews but instead “set by statute and OPM regulations.”^[70] ALJs are not eligible for bonuses.^[71] Finally, the APA limits the agency’s ability to remove or demote an ALJ to only “for good cause,” which is determined by the Merit Systems Protection Board (MSPB), whose own members may only be removed for cause as well.^[72]

Why does the standard model of agency adjudication insist on decisional independence of ALJs? Decisional independence of ALJs is important for the same reason we prize judicial independence in the federal Article III judiciary. Perhaps the most fundamental principle of judging is that the adjudicator be free from outside influence in decisionmaking. The core tenet of due process is an impartial decisionmaker.^[73] The fear is that adjudicators cannot resolve issues impartially if members of the legislature lobby them for a particular result or if donors threaten to withhold support contingent on a particular outcome of a case.^[74] As a result, at the federal level, the Article III guarantees of lifetime tenure and protection against salary diminution stand as fortifications of decisional independence.^[75] By statute (and executive order), similar protections exist for agency adjudicators, based on the theory that members of Congress or political appointees in the agency should not be able to meddle in case-specific resolution of individual cases, and financial incentives should not prejudice the proceedings.

Decisional independence also enables the hearing-level adjudicator to create an administrative record—or the paper trail that documents the ALJ’s decisionmaking process and the basis for the ALJ’s decision—that is focused on case-specific facts and is free from outside influence. The administrative record itself has important benefits, which are substantially enhanced when it is compiled by an impartial adjudicator. First, the record helps to provide legitimacy to agency adjudication. The administrative record serves as the basis of the ALJ’s decision and outlines the legal reasoning for the outcome reached. Without such a record, it is impossible to know that decisions are fairly achieved. If parties to the adjudication do not believe decisions are fairly reached, they are less likely to accept the outcome.^[76]

Second, the administrative record provides the documentation needed for higher-level review, both within the agency and by federal courts. When reviewing an agency’s decision, courts will determine whether the agency’s action is reasonable and consistent with the applicable legal requirements.^[77] Importantly, whether a court or agency head seeks to overturn the hearing-level adjudicator’s decision, the higher-level reviewer must explain why the initial decision that outlines the adjudicator’s legal reasoning is incorrect.

B. Agency-Head Final Decisionmaking Authority

The second key structural feature of the standard model of agency adjudication is that the agency head has final decisionmaking authority. This feature is the way in which political control is infused into agency adjudication. Importantly, however, the standard model injects political control over agency adjudication in a transparent and constrained manner.

The Supreme Court has interpreted the APA to provide that the ALJ’s initial decision is not entitled to deferential administrative review.^[78] Plenary review stems from a critical difference between agencies and federal courts. As Harold Krent and Lindsay DuVall note, “[l]egislatures have directed the agency, not the ALJ, to issue a decision reflecting the agency’s position.”^[79] In fact, the critical difference between an ALJ adjudication and a civil bench trial is that the agency head has de novo review authority, while an appellate court defers to the trial court’s factual findings.^[80] Indeed, federal courts scholars have long distinguished Article III federal courts and Article I legislative courts from agency adjudicatory tribunals on the theory that the agency head has final policymaking authority.^[81]

As a result, an agency has complete freedom, as though it had heard the initial evidence itself, when reviewing the decision of the ALJ. Nevertheless, the agency is typically required to explain why it has rejected an ALJ’s findings, and courts examine the evidence more critically when an agency’s reversal of an ALJ ruling turns on the credibility of the witnesses who testified at the hearing.^[82] Thus, the standard model envisions the agency head exercising political control over agency adjudication but requires this power to be implemented through a transparent mechanism.^[83] This point cannot be overstated. The agency head has wide latitude to reverse the ALJ’s initial decision, including for policy considerations, but must explain her reasons for the reversal in a written decision. The agency head’s decision becomes part of the administrative record that is subject to judicial review by a federal court and scrutiny by Congress, the President, and the public more generally. As a result, a reviewing court has the benefit not only of the hearing-level ALJ decision, which is based on an impartial application of law to facts, but also the agency head’s reasoned decision, which may be imbued with express policy-based or political preferences.

Importantly, this second structural feature—an agency head’s final decisionmaking authority—provides numerous benefits that improve agency performance.^[84] Perhaps most saliently to this discussion, it ensures that agency heads control the regulatory structure they supervise. That is, it provides for political control and accountability over agency adjudication. Agency heads—who can comprise a single director, secretary, or administrator; or a commission, board, or other multi-member body—oversee the agency’s activities and set the agency’s policy preferences. It is widely accepted that agency heads have a comparative advantage in policy expertise relative to agency adjudicators.^[85] Generally, agency leadership has greater access to experts and staff that provide input and partake in the deliberative process that leads to better informed policy decisions than adjudicatory officers.^[86] In contrast to agency heads, adjudicatory officers often have significant caseloads that prevent them from having the time necessary to think deeply about policy matters.^[87]

In addition, agency heads possess direct review authority of adjudications to help ensure consistency in adjudicative outcomes. Jerry Mashaw, in his seminal book Bureaucratic Justice, expounded a theory of agency adjudication in which agency-head control sought to increase consistency and accuracy in adjudicative outcomes.^[88] From a normative perspective, consistency in adjudicatory outcomes is important to fairness arguments underlying equal enforcement, as well as encouraging confidence and hence ex ante compliance with agency policy.^[89] Yet despite these goals, inconsistency in adjudicatory outcomes is a reality of the modern day administrative state.^[90] By enabling an agency head to review and reverse adjudicatory outcomes, final agency head decisionmaking authority helps ensure that agency policy preferences are consistently applied and that similarly situated parties receive similar results across decisionmakers. An agency head’s use of precedential decisionmaking, moreover, increases consistency in subsequent adjudications as well.^[91]

Finally, agency-head review “helps the agency head gain greater awareness of how a regulatory system is functioning.”^[92] Such awareness assists the agency head when considering whether adjustments to the regulatory scheme are necessary via standard policymaking forms—such as rulemaking, adjudication, or guidance documents—or less formal mechanisms—such as quality assurance programs or inputs to performance evaluations. Prior work by one of us (Wasserman) and Michael Frakes demonstrates how agency heads influence agency culture. In turn, agency culture has tremendous impact on shaping how agency employees approach and develop their practice style.^[93] Frakes and Wasserman find that the Director of the U.S. Patent & Trademark Office (PTO) utilized initial training and quality assurance mechanisms, among other means, to help shape the granting culture of the PTO.^[94] They find strong evidence that the culture of the PTO (either pro-patent or anti-patent) the year the patent examiner was hired had a lasting effect on her granting patterns over the course of her career.^[95]

* * *

In sum, the standard model of agency adjudication—which is utilized by a number of so-called independent federal agencies, such as the Federal Trade Commission (FTC),^[96] the Federal Communications Commission,^[97] the International Trade Commission,^[98] and the SEC,^[99] and is commonplace at a number of executive branch agencies, including at the Departments of Agriculture, Health and Human Services, Interior, and Labor^[100]—strikes a balance between decisional independence of hearing-level adjudicators and political control of the agency’s final adjudication decision. More specifically, APA-governed formal adjudication provides a series of protections that guarantee decisional independence of ALJs but at the same time enable the agency head to have almost unfettered final decisionmaking authority.^[101] Importantly, the APA strikes this balance between political control and decisional independence by requiring the agency head to provide written detailed reasons for why it is overturning the initial adjudicator and subjecting this decision to federal court review.^[102] Thus, the infusion of political control over agency adjudication is achieved in a highly circumscribed and transparent manner in order to also maintain hearing-level adjudicator decisional independence.

II. Agency Adjudication’s Perceived Crisis

A mounting chorus of scholars and policymakers contend that agency adjudication is reaching a crisis point as the standard model of agency adjudication has come under ever more attack.^[103] Indeed, there is a view in many administrative law circles, and especially among ALJs and other agency adjudicators, that the first feature of the standard model—the decisional independence of agency adjudicators—is increasingly threatened.^[104] Given the centrality of a neutral arbitrator to the standard model in agency adjudication, many believe that even the potential erosion of decisional independence is alarming.^[105]

The threat to adjudicator impartiality in agency adjudication stems from at least two separate sources. The first involves congressional choices to depart from the APA model for formal adjudication. This choice has resulted in the vast majority of agency adjudications being overseen by hearing-level adjudicators that have less decisional independence than ALJs in terms of hiring, supervision, and firing. The second involves the Supreme Court’s recent precedents on the Appointments Clause and separation of powers. These precedents have strengthened presidential control over agency adjudication, at least in theory, at the expense of the potential decisional independence of all agency adjudicators. We address each in turn.

A. The New World of Agency Adjudication

Despite APA-governed formal adjudication being the standard model that every administrative law student learns, the vast majority of agency adjudications and federal regulatory actions do not involve APA-governed formal adjudications before an ALJ.^[106] Agencies instead, increasingly regulate using adjudicatory means that still require evidentiary hearings but do not embrace all of the features set forth in the APA for formal adjudication. To borrow from Daniel Farber and Anne O’Connell, the predominance of “formal-like” agency adjudication outside of the APA is yet another departure from the “lost world of administrative law”—further revealing “an increasing mismatch between the suppositions of modern administrative law and the realities of modern regulation.”^[107]

This new world of agency adjudication comprises evidentiary adjudication proceedings presided over by non-ALJ agency personnel that have diverse titles, such as administrative judges (AJs), immigration judge, administrative patent judge (APJ), and hearing officer.^[108] This Article collectively refers to these non-ALJ adjudicators as AJs. While formal-like adjudication maintains agency-head final decisionmaking authority, the adjudicators that oversee these evidentiary hearings—AJs—have far less decisional independence than ALJs. As a result, this new world of adjudication disrupts the careful balance between impartiality of agency adjudicators and political control of agency adjudication present in the standard model of agency adjudication.

Formal-like adjudication that falls outside of the APA now predominates the federal judiciary. Consider, for instance, the immigration court system within the Department of Justice’s Executive Office for Immigration Review. By the end of 2023, there were more than 730 immigration judges hearing cases at the Department.^[109] To put that number in perspective, during the first three years of the Biden Administration, the Senate confirmed 145 Article III judges.^[110] In recent years, these immigration judges have decided between 100,000 and 550,000 cases per year.^[111] In fiscal year 2023, for instance, the Justice Department received more than one million new cases in the immigration court system, and the immigration judges completed more than 520,000 cases.^[112] The stakes in immigration adjudication are high, including whether to allow noncitizens to remain in the United States to avoid persecution in their countries of origin.^[113]

Increased reliance on formal-like adjudication that is outside of the APA by both Congress and federal agencies has contributed to the concerns about diminished decisional independence of agency adjudicators. There are a host of ways in which agency heads have more latitude to influence the outcomes of formal-like adjudication than APA-governed formal adjudication. Most of these methods stem from the greater ability of agency heads to either directly or indirectly influence AJs.

Unlike ALJs, most AJs are subject to agency performance appraisals. AJs’ salaries can be affected by these reviews. No statute bars agencies from giving bonuses to AJs. As a result, a number of AJs receive bonuses for hitting certain performance targets, such as productivity quotas or goals.^[114] Recent research highlights how influential performance appraisals are in shaping the incentives of agency employees.^[115] Performance appraisals and financial incentives of AJs thus have been criticized.^[116] And bonuses paid to APJs upon workload completion have been challenged, albeit unsuccessfully, as violating due process.^[117]

Agencies have utilized other methods to influence AJ outcomes. AJs, unlike ALJs, are not necessarily assigned cases by rotation or randomization. For example, the PTO used to panel stack its adjudicatory board to arrive at the agencies’ preferred adjudicatory outcomes.^[118] When a three-board-member panel arrived at a decision that the agency disagreed with, the aggrieved party would ask for a rehearing and the Director would enlarge the panel to include five or sometimes seven members to ensure the outcome she preferred.^[119] This controversial practice would appear to be inconsistent with the APA if the APJs were in fact ALJs, who are afforded more protections.^[120]

Finally, agencies have more latitude to remove AJs and historically to hire AJs. Unlike ALJs’ “good cause” standard of removal, AJs are not subject to any particular protection from removal beyond general civil service protections. As a result, an AJ may fear being fired if, for example, she does not apply the law to facts in the way the agency head prefers or if she makes factual findings that the agency head disagrees with. And AJs, unlike ALJs, have never been subject to a selection process that involves an outside agency like OPM.^[121] While many agencies have criteria for selecting AJs and civil service statutes prevent the hiring of AJs based on partisanship, there is no statutory requirement that AJs have any particular qualifications.^[122] Nevertheless, partisan hiring of AJs has occurred. Perhaps most famously, a 2008 Special Report by the Office of the Inspector General and the Office of Professional Responsibility found that the Bush Administration had violated Justice Department policy and federal civil service statutes when it hired immigration judges based on political affiliation.^[123] Half of the thirty-seven immigration judges hired during the Bush Administration had no experience in immigration law.^[124]

Importantly, in this new world of agency adjudication, the standard model of agency head review persists but the first central feature of the standard model has eroded. The degree of decisional independence of hearing-level adjudicators is in much greater doubt in formal-like adjudication that falls outside of the APA than in APA-governed formal adjudication. And in this new world, formal-like adjudication has risen to overtake APA-governed formal adjudication in both the number of agency adjudicators and case volumes.

B. The Roberts Court and Separation of Powers

The second reason, however, is more far-reaching and limits what even Congress can do. Since the appointment of Chief Justice John Roberts and Justice Samuel Alito, the Supreme Court has embraced a more unitary executive view of separation of powers and the Appointments Clause. This trend, moreover, has sped up with the appointments of Justices Neil Gorsuch, Brett Kavanaugh, and Amy Coney Barrett—and sometimes Justice Elena Kagan. The resurgence of the unitary executive theory, moreover, now crosses political lines. The theory used to have more purchase in Republican administrations, but even President Joe Biden in some ways became remover-in-chief, going so far as to remove the head of the Social Security Administration despite a statutory tenure protection.^[125]

The Supreme Court’s embrace of the unitary executive theory and, concomitantly, the strengthening of presidential control over agency adjudication has potentially come at the expense of decisional independence, or neutral competence, of agency adjudicators. The Court’s rulings have involved two interrelated issues: the hiring and the removal of agency adjudicators. The Court’s decisions associated with agency adjudicator hiring have resulted in increasing agency head control over who hears the cases in the first instance. Critics contend that such localized power enables agency heads to hire adjudicators based on political affiliation or policy preferences rather than merit, undermining decisional independence of agency adjudicators. The Court’s precedents associated with removal have increased the latitude of the President to remove an agency head and of an agency head to remove agency adjudicators. By rolling back removal protections, reformers fear that agency adjudicators may be more subject to outside pressures to reach a specific outcome. Importantly, these two sets of cases interact with one another to amplify impartiality concerns of adjudicators. That is, if agency heads can fire adjudicators at will and then hire their replacements, decisional independence seems to be imperiled. As a result, the fear is that both sets of cases, and their interactions with one other, may potentially increase political decisionmaking in federal agency adjudication at the expense of the impartiality of hearing-level agency adjudicators.

This constitutional sea change began (at least in recent years) with Free Enterprise Fund, which concerned the constitutionality of various features of the Public Company Accounting Oversight Board—a regulatory body housed within the SEC. The SEC appoints the Board’s five members but, importantly, could not remove them at will, but only “for good cause” and pursuant to various procedures.^[126] The parties stipulated, moreover, that the President could not remove SEC Commissioners at will.^[127] The D.C. Circuit held—based on the Supreme Court’s well-known Humphrey’s Executor v. United States^[128] and Morrison v. Olson^[129] decisions—that this arrangement did not violate the separation of powers, even though there were two layers of removal protection between the President and the Board members.^[130] The Supreme Court, however, disagreed, holding in an opinion by Chief Justice Roberts that even if some statutory removal restrictions are constitutional, two levels was one too many. After all, Roberts reasoned, “[a] second level of tenure protection changes the nature of the President’s review,” even though “[t]he Constitution requires that a President chosen by the entire Nation oversee the execution of the laws.”^[131]

Commentators quickly observed that the Court’s reasoning would seem to apply beyond a two-level context.^[132] If the Constitution requires presidential control (the premise of the argument), why would the number of levels between the President and the agency official matter? But even a prohibition on two levels of removal protection alarmed Justice Breyer, who—in dissent—worried about what it would portend for agency adjudication.^[133] As he explained, many agencies with leadership protected by statutory removal restrictions also use adjudicators with their own removal restrictions. In other words, many agency adjudicators are separated by two layers of removal protection from the President. In response, the majority observed that its opinion did not address adjudicators:

[O]ur holding also does not address that subset of independent agency employees who serve as administrative law judges. . . . And unlike members of the Board, many administrative law judges of course perform adjudicative rather than enforcement or policymaking functions or possess purely recommendatory powers. The Government below refused to identify either “civil service tenure-protected employees in independent agencies” or administrative law judges as “precedent for the PCAOB.”^[134]

Free Enterprise Fund arguably would not have imperiled the standard model of agency adjudication, if ALJs were employees rather than officers. However, eight years later, in Lucia v. SEC,^[135] the Court foreclosed that possibility by holding that ALJs are “Officers of the United States” who must be appointed pursuant to the Appointments Clause.^[136] With Justice Kagan writing, the Court reasoned that ALJs are officers because they hold “a continuing office established by law”^[137] and “exercise[] significant authority pursuant to the laws of the United States.”^[138] Justice Breyer (joined by Justices Ruth Bader Ginsburg and Sonia Sotomayor) wrote separately to warn of the potential implications for agency adjudication of holding that ALJs are officers:

If the Free Enterprise Fund Court’s holding applies equally to the administrative law judges . . . then to hold that the administrative law judges are ‘Officers of the United States’ is, perhaps, to hold that their removal protections are unconstitutional. This would risk transforming administrative law judges from independent adjudicators into dependent decisionmakers, serving at the pleasure of the Commission.^[139]

In the wake of Lucia, President Trump issued Executive Order 13,843, an order that “[e]xcept[ed] Administrative Law Judges [f]rom the [c]ompetitive [civil] [s]ervice,” noting that under Lucia “at least some—and perhaps all—ALJs are ‘Officers of the United States.’”^[140] To ensure that ALJ hiring did not run afoul of the Appointments Clause, the order placed ALJs’ appointments in the hands of the agency.^[141] It is easy to see how this new process could erode adjudicators’ decisional independence or neutral competence by easing the barriers for agency heads to hire ALJs whose policy preference align with their own.^[142] To try to guard against that possibility, however, the executive order opened by reemphasizing that “[t]he Federal Government benefits from a professional cadre of administrative law judges (ALJs) . . . who are impartial and committed to the rule of law.”^[143] Indeed, the executive order emphasized as its “[p]olicy” that “ALJs must display appropriate temperament, legal acumen, impartiality, and sound judgment,” and “must also clearly communicate their decisions to the parties who appear before them, the agencies that oversee them, and the public that entrusts them with authority.”^[144]

While Lucia gave agency heads greater freedom in the appointment of ALJs, subsequent Supreme Court cases eased the ability of agency heads to fire agency adjudicators.

The Court further solidified its entrenchment of the unitary executive theory in Seila Law v. Consumer Financial Protection Bureau^[145] and Collins v. Yellen,^[146] wherein the Court held that structures less restrictive than two levels of removal protections at issue in Free Enterprise Fund are also unconstitutional.

Seila Law concerned the structure of the Consumer Financial Protection Bureau (CFPB), a powerful regulating agency headed by a single individual whom the president could only remove for “inefficacy, neglect, or malfeasance in office.”^[147] The CFPB’s defenders argued that the rule from Humphrey’s Executor (which allowed such protection for a multi-member body of principal officers^[148]) combined with the rule from Morrison (which allowed such protection for some individual inferior officers^[149]) should extend to the CFPB Director.^[150] Writing for a five-justice majority, however, Chief Justice Roberts disagreed, holding that the CFPB structure was unconstitutional.^[151] Chief Justice Roberts explained that “Humphrey’s Executor [merely] permitted Congress to give for-cause removal protections to a multimember body of experts, balanced along partisan lines, that performed legislative and judicial functions and was said not to exercise any executive power,” and that precedent also only protects “inferior officers with limited duties and no policymaking or administrative authority.”^[152] Notably absent from this analysis was any first principles defense of statutory restrictions on removal. Indeed, the Court described the rule from Humphrey’s Executor so narrowly that it does not appear to even apply to the FTC today.^[153] Alarmed, Justice Kagan dissented, using unusually pointed language.^[154]

In Collins v. Yellen, the Court—by a vote of 7 to 2 (Justice Kagan joined the Court’s judgment because of stare decisis, but she did not join the majority opinion)—extended Seila Law’s holding to the Federal Housing Finance Agency (FHFA), which was also headed by a single individual.^[155] Writing for the majority, Justice Alito concluded that it does not matter that the FHFA Director by statute wields less executive power than the CFPB Director, nor that the removal restriction at issue was less demanding (requiring just some “cause”).^[156] Justice Sotomayor dissented, lamenting that “[t]he Court has proved far too eager in recent years to insert itself into questions of agency structure best left to Congress.”^[157]

The Court then seemed to take a different path in United States v. Arthrex.^[158] In Arthrex, the Court held that APJs cannot have both final decisionmaking authority and statutory removal protections.^[159] As part of the America Invents Act, Congress empowered APJs to resolve certain patent questions without plenary review by a principal officer.^[160] The constitutional wrinkle is that the Patent Act does not give the Director of the PTO final decisionmaking authority. The Federal Circuit reasoned that this was unconstitutional, and that the correct remedy was to sever these adjudicators of their tenure protection such that the agency head could remove APJs at will.^[161] This remedy risks transforming the APJs from independent adjudicators into dependent decisionmakers.

The Supreme Court agreed with the Federal Circuit that the structure of the adjudicatory board of the PTO violates the separation of powers, holding that “[o]nly an officer properly appointed to a principal office may issue a final decision binding the Executive Branch in the proceeding before us.”^[162] But the Court opted for a different remedy. Instead of making APJs removable at will by the agency head, the Court held unlawful the Patent Act’s prohibition on the Director’s review of APJs’ decisions.^[163] In other words, by giving the agency head the final say, the Court effectively preserved the decisional independence of the hearing-level agency adjudicators in exchange for more political accountability over the substantive outcomes in the agency adjudication system. In so doing, the Court emphasized that agency-head final decisionmaking authority “is the standard way to maintain political accountability and effective oversight for adjudication that takes place outside the confines of” the APA’s agency-head review provision.^[164]

Finally, the Fifth Circuit in Jarkesy v. SEC held—precisely as Justice Breyer feared some court eventually would—that having ALJs which are insulated by two layers of for-cause removal protections violates the Constitution.^[165] The Fifth Circuit reasoned that the President cannot remove SEC Commissioners at will, and those Commissioners in turn cannot remove the SEC’s ALJs at will, thereby—in the Fifth Circuit’s view—preventing the President from “tak[ing] care that the laws are faithfully executed.”^[166] Although the Fifth Circuit’s ruling is controversial,^[167] Jarkesy’s application of Free Enterprise Fund was no surprise. As this Part illustrates, policymakers and scholars have seen this development coming for more than a decade.^[168]

The Supreme Court granted certiorari in SEC v. Jarkesy, holding oral argument in the case in November 2023.^[169] Ultimately, the Supreme Court affirmed the Fifth Circuit’s decision in Jarkesy on the Seventh Amendment jury trial issue, leaving for another day the dual-layer removal issue.^[170] In so doing, the Supreme Court left the Fifth Circuit’s dual-layer removal holding undisturbed,^[171] such that dual-layer removal protections for SEC ALJs remain unconstitutional in the Fifth Circuit. Indeed, after the Supreme Court issued Jarkesy, a district court in the Fifth Circuit applied the Fifth Circuit’s dual-layer removal holding to grant a preliminary injunction against an ALJ adjudicating a matter at the National Labor Relations Board.^[172] And the Fifth Circuit itself has now indicated that all three of its holdings in Jarkesy remain circuit precedent.^[173]

Accordingly, it is only a matter of time until the Supreme Court decides this issue. Indeed, as this Article was going to press, the Trump Administration announced that it would no longer defend the constitutionality of dual-layer removal protections for ALJs. ^[174] We expect, given the logic of the Court’s decisions, that it will hold unconstitutional the two levels of statutory removal protections from the President with respect to ALJs. In fact, because Jarkesy implicates three layers of removal protection—for the ALJ, the SEC, and the MSPB—the United States urged that the appropriate remedy (if the Court finds a constitutional violation) is for the SEC to be able to remove ALJs at will.^[175] Recent Supreme Court decisions that facilitate challenges to agency adjudicators will only serve to catalyze more litigation on the issue.^[176]

To be sure, whether the Constitution provides the President with an unfettered removal power has long been the source of constitutional debate; indeed, the First Congress split on the subject.^[177] Free Enterprise Fund, which made the debate of more than mere academic interest, prompted an avalanche of literature.^[178] Even so, the trend lines are clear: restrictions on removal are in retreat. For purposes here, it is enough to observe that most of the Justices now endorse unitary-executive principles, and those principles—if followed to their logical conclusion—may pose challenges for agency adjudication.

III. Reform Proposals and Their Limitations

What can be done? If the Supreme Court continues its current path (and there is little reason to believe it will swerve), the Justices may soon conclude that agency adjudication—at least as it is currently structured—is unconstitutional because the President does not have sufficient control over it. In fact, based on the Court’s embrace of the unitary executive theory, the Court may well conclude that any restrictions on presidential removal of Executive Branch officers who exercise any “policymaking or administrative authority” are unconstitutional, which appears to be the logical conclusion of the Court’s reasoning in Seila Law and Collins.^[179] Moreover, Congress’s repeated decision to create adjudicatory structures that require an evidentiary hearing but fall outside of the APA further exacerbates concerns about the impartiality of agency adjudicators. Does this mean that agency adjudication is doomed?

Alarmed that such a question is even reasonable to ask, reformers have begun searching for a path to save agency adjudication. Several solutions have been proposed, three of which have received the most attention: a new agency (or “central panel”) to house administrative adjudication; greater use of Article I courts; and the transfer of most or all agency adjudication to Article III courts.

In this Part, we briefly outline these three proposals and discuss some of the costs and benefits of each. Our goal here is not to provide a comprehensive analysis; that would likely require a full law review article for each proposal. Instead, we aim to highlight how each proposal has significant limitations. For the reasons explained in this Part, these proposals either will not solve the constitutional problem, will have unintended consequences, or often both.

A. Article II Centralized Administrative Judiciary

The solution that has prompted the most real-world activity appears to be the creation of a new centralized adjudication agency within the Executive Branch. This proposal—often called a “central panel”—involves moving all, most, or at least some agency adjudicators from their current agencies and placing them in a new agency.^[180] Although the precise details of the central panel vary, the basic idea is a new agency headed by directors nominated by the President and confirmed by the Senate. The agency’s leaders would be protected by a “good cause” removal protection and would then appoint other agency adjudicators who could be removed at will by the central panel directors.^[181] Alternatively, the head or heads of the agency could be removed at will, but the individual agency adjudicators would enjoy “good cause” removal protection.^[182] Adjudicative decisions then would be subject to de novo appeals to the agency head(s)—either the directors of the central panel or the head(s) of the substantive agency from which the adjudicators were transferred.^[183] This proposal has been championed by the National Conference of Administrative Law Judges (NCALJ), an entity comprised of thousands of agency adjudicators.^[184] In 2022, the ABA’s House of Delegates adopted a resolution urging Congress to create a federal central panel for the subset of federal agency adjudicators who handle benefits claims.^[185]

Richard Levy and Robert Glicksman, two of the main scholarly proponents of this proposal, have nicely summarized how this central panel proposal could function:

A properly designed central panel can preserve the advantages of the APA’s approach to administrative adjudication, including agency control over policy and preservation of specialized adjudicatory expertise, while increasing protections for decisional impartiality and independence. As in most states, agencies would retain control over policy formulation through promulgation of binding legislative rules, precedential adjudications, and less formal (and therefore nonbinding) guidance documents. In addition, agencies would retain final decisional authority through de novo review of central panel decisions and the power to preside over a case as an original matter in lieu of referring it for resolution by the central panel. The concept of a central panel does not require that all of its judges would have the same status or decide all types of cases. Thus, the central panel can preserve specialized expertise through the creation of subject matter-specific divisions, such as a specialized division of Social Security judges. At the same time, and equally important, by removing administrative adjudicators from the direct oversight of the agency whose cases they adjudicate, the central panel model severs the direct lines of control that facilitate improper agency pressure, thereby promoting impartial adjudication.^[186]

The central panel proposal would be transformative for the federal administrative judiciary—and, in our view, not in a good way. Our sets of concerns are at least threefold.^[187] First, the proposal is incredibly disruptive to the adjudication system, yet we have little to no empirical evidence on the magnitude of the problem the proposal seeks to address. Second, the proposal would introduce a host of unintended consequences and produce unclear benefits. And third, it is far from evident that this solution would actually solve the Article II concern, especially if the reasoning from Seila Law and Collins is extended to its logical conclusion. We address each in turn.

1. This Proposal Is Disruptive and Lacks Empirical Evidence on the Problem It Seeks to Address

The central panel proposal would potentially involve moving more than 12,000 agency adjudicators from their current agencies into one mega central panel agency, where hundreds of thousands if not millions of individuals would be forced to navigate that new system without the benefits of decades of improvements that have been adopted in the current adjudication system. Before adopting such a disruptive proposal, we think it is imperative to understand the magnitude of the problem that it attempts to solve. To be sure, we agree with Glicksman and Levy (and others) that there is a growing potential threat to decisional independence of agency adjudicators based on the two developments detailed in Part II. But no one has empirically explored in any rigorous manner whether and to what degree there is a systemic, real-world threat to decisional independence—whether agency adjudicators make decisions based on political influence and out of fear of being fired or otherwise politically disciplined, instead of based on law and facts. In particular, do ALJs and AJs judge partially today based out of concerns of being fired or otherwise punished by political leadership? Would the move from good-cause removal by the MSPB to good-cause removal by the agency head—or, at multi-member commissions, at-will removal by the commissions—increase the likelihood of ALJs and other agency adjudicators not impartially deciding cases? If so, to what degree? Are the proposed solutions proportional to the scope of the problems?

Importantly, this concern applies with equal force to all three proposals discussed in Part III. Before creating an agency comprising of one mega central panel of adjudicators or transferring all or some part of agency adjudication into Article I tribunals or Article III courts, Congress must better understand the scope and magnitude of the problem that the reform proposals seek to address.^[188] Without understanding the magnitude or the extent that decisional independence of agency adjudicators is actually imperiled, it is impossible to know whether the benefits associated with any of the three proposals outweigh their costs. And the costs, which we discuss below, could be staggering.

2. This Bold Proposal Will Introduce a Host of Unintended Costs and Produce Unclear Benefits

Just as no one has systematically studied the extent of the problem which the central panel system attempts to ameliorate, no one has thoroughly studied the intended and unintended consequences of the bold reform proposals of relocating thousands of agency adjudicators to a new federal agency. We fear such studies would reveal that the central panel proposal would produce insubstantial benefits that would come nowhere near justifying staggering costs to the system. And the millions of individuals who try to navigate these adjudication systems each year would bear the brunt of those costs.

To be clear, Congress, agencies, and the public have spent decades trying to improve agency-specific adjudication systems—to increase inter-decisional consistency, to improve the quality of adjudicator decisionmaking, to speed up the adjudication process, to manage crushing backlogs, and to help individuals who often appear without legal counsel to effectively navigate those systems. To provide just one example, as Matt Wiener and one of us (Walker) chronicle in a recent study for the Administrative Conference, agencies have carefully developed appellate review systems to help address these systemic challenges in their high-volume adjudication systems.^[189] This same type of careful study is needed of a central panel system to better understand the costs and benefits associated with this proposal.

For example, in August 2022, the ABA passed a resolution urging “Congress to enact legislation establishing a tribunal, staffed by ALJs to decide cases airing under federal benefit programs that is independent of the federal agencies that manage these programs.”^[190] This new federal central benefits panel would involve the transfer of all disability cases adjudicated by the Social Security Administration (SSA). Relocation of SSA adjudications from their current home to a mega central adjudicative agency would affect more than 1,650 SSA ALJs—about 85% of ALJs nationwide^[191]—and, more importantly, more than half a million individuals (and their legal counsel) who go through the SSA adjudication system each year.^[192]

The risks of inefficiencies, procedural deficiencies, and inter-decisional inconsistencies would be enormous. While the effect of such a massive upheaval on the agency adjudicators would be profound, we are even more concerned about the potentially catastrophic impact it would have on the individuals and their legal counsel who would struggle to navigate this new agency. Many of these individual beneficiaries and claimants—including members of marginalized communities—cannot afford to retain counsel, and the disruptive effect on them would be especially severe.

The SSA example is just the beginning of the complications that a federal central panel vision implicates. At least the ABA’s proposal limits the central panel proposal to benefits adjudications. The more ambitious federal central panel proposal is intended to encourage Congress to consolidate SSA adjudicators (and adjudications) into the same mega-agency with APJs, immigration judges, and SEC ALJs—just to name a few.^[193] Yet Congress has already created unique adjudicative systems for regularized decisionmaking in a number of these regulatory fields, such as the Patent and Trademark Appeals Board^[194] and the Board of Veterans’ Appeals.^[195] These systems would be seriously undermined if adjudicators in these fields were moved into a federal central panel. The proposal does not come to grips with the disruptive implications of displacing these systems.

Ironically, scholars, agency officials, and researchers at the Administrative Conference and elsewhere have spent decades studying and recommending improvements to specific agency adjudication systems. These recommendations do address decisional independence, but they also address efficiency and procedural improvements to increase the likelihood that each individual subject to adjudication at the agency level has a prompt and fair process and accurate outcome. The amount of careful research and study that have gone into improving SSA adjudication, for instance, is staggering.^[196] And yet the central panel proponents encourage Congress to consider instituting a federal central panel system that has been subjected to no comparable study.^[197] The burden of empirical proof should be on the moving party. Here, we are doubtful the central panel proponents would come close to demonstrating that the benefits justify the costs.

Compounding our concerns about the costs and benefits and the lack of empirical evidence for reform, the central panel proposal raises serious doubts about its intrinsic value and workability. One concern is that, even though the traditional APA model gives rise to certain risks that agency heads may sometimes interfere with adjudicators’ decisional independence for political or arbitrary reasons, it is also possible that oversight by central panel directors could prove to be political or arbitrary for other reasons. A president, for example, could appoint a political associate or ideologically motivated person to fill this slot. In other words, this proposed reform may fail under the reformers’ own untested theory of what motivates political leadership. Even if the director were to turn out to be well-qualified and well-motivated, the central panel would be a very large entity (far larger than any existing state central panel), and the line between “supervision” and “undue pressure” would likely be indistinct. One can only speculate about the pressures that the director or directors would be in a position to exert in the name of supervision, and about whether there would be adequate institutional safeguards to prevent misuses of the director’s supervisory authority.

Moreover, it is unclear whether a federal central panel would be able to maintain the advantages of specialization that accrue under the current APA structural model. Under the latter model, ALJs who regularly work within one agency develop an understanding of the issues it faces, many of which can be extremely specialized and technical. They also become familiar with the agency’s written and unwritten policies and priorities. Indeed, agencies can benefit from the informed ground-level critiques that ALJs can provide because of their past experience in working with the agency’s caseload.

In contrast, the great majority of state central panels are staffed by generalist ALJs who may be assigned to a variety of cases over time.^[198] If the federal panel were to follow that model, opinions would often be written by ALJs who would be unfamiliar with the relevant subject matter and agency practices. One likely consequence would be many more appeals to agency heads and more time spent reworking ALJs’ conclusions when the cases are appealed. This development would tend to delay the agencies’ final dispositions regarding those disputes, in an environment in which case backlogs are already extreme and the wheels of justice are already turning far too slowly.^[199] To be sure, this concern could potentially be addressed by subdividing the central panel by specialty. But that argument does not answer the question of how such judges would be supervised. The director, lacking the agency’s substantive expertise and program responsibilities, might make decisions on an arbitrary basis; political accountability for such improvident decisions might be hard to maintain because the director would likely have less visibility than a typical agency head has.

Other oversight issues would raise similar concerns. For instance, many adjudication systems face crushing backlogs, such that the real concern for individuals in the system is not a biased adjudicator, but the lack of a timely decision.^[200] Justice delayed, they say, is justice denied. Agency-head management is essential to addressing the timely adjudication of matters. Indeed, federal courts have repeatedly held that reasonable productivity goals are permissible and do not infringe on the decisional independence of the agency adjudicator.^[201] “An agency head that undertakes to prescribe ‘reasonable productivity goals’ can be guided by program needs, but a [central panel] director with no responsibility for implementing the substantive statute would lack that baseline.”^[202] For all of their imperfections, the institutional relationships between the leadership of federal agencies and the ALJs who respectively hear cases in those agencies are structured by longstanding statutes, rules, policies, and norms of agency practice in order to advance administrative law’s rule-of-law values of predictability, consistency, fairness, and efficiency.^[203] The central panel proposal would risk undermining much of these benefits of the standard federal model—especially in high-volume adjudication systems.^[204]

3. This Proposal May Not Satisfy Article II Anyway

Finally, it is not at all clear how the central panel proposal would address the perceived threats to agency adjudicator decisional independence. To be sure, the central panel would avoid Free Enterprise Fund’s bar on dual-level removal restrictions—if ALJs are removable at will with the central panel directors removable for cause (or vice versa).^[205] But assuming the central panel directors could be removable only for cause despite Seila Law and Collins under a Humphrey’s Executor theory—which may not be the case—the President would still exercise political control over the agency and especially the ALJs who would enjoy no removal protection from the agency head.^[206] In other words, we may not be in a different place than the status quo. The same is true in the event that the Court holds that agency heads are removable at will but that ALJs enjoy good-cause removal protections.

Furthermore, for all the reasons explained above, it is not at all clear that the Court’s decisions will stop at the relatively limited holdings in Seila Law and Collins. The Court’s logic seems to extend beyond just single-headed agencies and may even capture anyone who exercises “administrative authority.”^[207] If so, then the central panel idea may be an exercise in futility, for at the end of the day, the President would still have the constitutional power to remove agency adjudicators notwithstanding Congress’s dramatic statutory reforms.

B. Article I Courts

Sounding a similar theme, other reformers argue that Congress should expand the nation’s system of “Article I courts”—tribunals like the Court of Federal Claims or the U.S. Tax Court that are staffed by presidentially appointed, Senate-confirmed judges who do not enjoy the same tenure and salary protections that Article III judges receive^[208]—to include more regulatory matters, thus moving adjudication out of enforcement agencies. In 2022, for instance, House Democrats introduced legislation to transfer immigration adjudication from the Justice Department to a new Article I immigration court system.^[209] Under this proposed solution, presidential interference in agency adjudication would supposedly be limited without expanding the Article III system.^[210]

Unfortunately, this solution suffers from serious flaws. It would be remarkably disruptive and produce a host of unintended costs, would create significant accountability problems, and, again, likely would not solve the constitutional problem. We again consider these points in turn.

1. This Solution Would Be Remarkably Disruptive and Produce Unintended Costs

As with the federal central panel proposal, this proposed reform would be remarkably disruptive for regulated individuals while threatening other administrative law doctrines. As with the other proposals, we lack systematic study of the proposal’s costs and potential benefits. Like the central panel system, we fear that the benefits are likely less than suggested and the costs may be staggering. Hundreds if not thousands of agency adjudicators would be moved to a new system, where hundreds of thousands if not millions of individuals would be forced to navigate that new system without the benefits of decades of improvements that have been adopted in the current adjudication systems. For all the reasons explained above in section III(A)(2), there is good reason to fear significant unintended consequences.^[211]

2. The Solution Would Also Create Accountability Problems

Unlike the central panel proposal, Article I courts would lack any agency-head review and supervision. That could result in a rise in inter-decisional inconsistencies, fewer tools to encourage efficiency and deal with case backlogs, and less efficient and effective systemic effects and awareness when it comes to the full regulatory apparatus—from policymaking and enforcement to adjudication and case management. Indeed, for good or ill, the Article I model would eviscerate the accountability rationale for judicial deference to interpretations of law announced in agency adjudication^[212] and an agency’s power to make policy by adjudication.^[213]

To be sure, the benefits of some narrow subject-matter transfers of agency adjudications to Article I courts may well justify the costs.^[214] We do not foreclose that possibility. The Tax Court, after all, has shown promise. An Article I court has the advantage over an Article III alternative by allowing Congress to still minimize the costs on the adjudicated by, for instance, tailoring the substance and procedures to allow individuals to appear without legal counsel and with more-limited discovery obligations.^[215] But even then, much more empirical work needs to be done to understand the benefits and the intended and unintended costs.

3. This Solution Is Constitutionally Vulnerable

Finally, the reality is that Article I adjudication is itself vulnerable to constitutional challenge, at least to the extent that the President does not have plenary control over it. Although labeled “Article I,” these judges plainly exercise Article II power.^[216] Indeed, as explained by Judge Sri Srinivasan in a 2014 separation of powers case about the constitutional status of the Tax Court, “Tax Court judges do not exercise the ‘judicial power of the United States,’ pursuant to Article III,” and also are not part of “the Legislative Branch”; therefore, “[i]t follows that the Tax Court exercises its authority as part of the Executive Branch.”^[217] Judge Srinivasan further concluded that although “Congress may afford the officers of those entities a measure of independence from other executive actors, but they remain Executive-Branch officers subject to presidential removal.” ^[218]

So, if the Supreme Court ultimately holds that the President has unilateral authority to remove essentially any executive officer who exercises “administrative authority,” as Seila Law and Collins suggest is on the table,^[219] it is hard to see why plenary removal would not extend to Article I courts. To the extent that this fairly straightforward constitutional analysis proves accurate, moving adjudication out of “enforcement” agencies like the SEC and placing it in pure “adjudicative” bodies like Article I courts to prevent constitutional objections to restrictions on presidential removal may rely on a false distinction that there is constitutional line between those categories. And unlike in the central panel model, the Article I judges have no principal officer overseeing them who has final decisionmaking authority or plenary removal authority of the Article I judges—one of which the Supreme Court seemed to require in Arthrex.^[220]

To be sure, Seila Law’s reference to “administrative authority” does not itself end the analysis. Seila Law only used that phrase to demark the limits of prior precedent upholding restrictions on presidential removal; the Court did not take the further step of saying that any removal restrictions that fall outside of prior precedent are necessarily unconstitutional.^[221] Instead, the Court reasoned that where precedent does not control, the next step is to consider whether the removal restriction at issue has a “foothold in history or tradition” and comports with “constitutional structure.”^[222] Perhaps removal restrictions in the context of Article I courts survive that test even as other removal restrictions do not.

Perhaps—but perhaps not. True, in Myers v. United States,^[223] a key pillar in Free Enterprise Fund and Seila Law,^[224] Chief Justice Taft reviewed the nation’s history of removal restrictions and argued that adjudicators (including, presumably, Article I judges) may not be subject to the same constitutional objections as other executive officials with removal restrictions.^[225] Specifically, he explained that “there may be duties of a quasi-judicial character imposed on executive officers and members of executive tribunals whose decisions after hearing affect interests of individuals, the discharge of which the President can not [sic] in a particular case properly influence or control.”^[226] Defenders of an expanded Article I court may lean on that language. Yet Taft was not done: In order to “discharge his own constitutional duty of seeing that the laws be faithfully executed,” “even in such a case” the President may still “consider the decision after its rendition as a reason for removing the officer, on the ground that the discretion regularly entrusted to that officer by statute has not been on the whole intelligently or wisely exercised.”^[227] If the adjudicator knows beforehand that the President can fire her for not acting (in the President’s view) “intelligently or wisely,” ^[228] the adjudicator rationally may attempt to predict what the President wants rather than exercise independent judgment—thus indirectly limiting decisional independence. In other words, Myers’ dicta about adjudication does not appear to meaningfully safeguard decisional independence.

Of course, it is possible that a court would conclude that Myers erred in this respect and that a more robust removal restriction preventing even ex post removal is also appropriate in light of the nation’s history and constitutional structure.^[229] It is not our point here to definitively resolve the question—a question that merits a deeper historical analysis. Instead, we merely observe that moving adjudication out of “agencies” into “Article I courts” may not even solve the constitutional problem to the extent that Article I courts are, notwithstanding any labels, agencies for purposes for Article II.

C. Article III Courts

Perhaps boldest of all, some reformers argue that some, most, or all agency adjudication should be moved into Article III courts, thus eliminating agency adjudication as a category.^[230] In this scenario, every matter currently handled by an agency adjudicator would be decided by an Article III judge, thus requiring a massive expansion of the Article III judiciary and a radical reimagining of what many agencies do.

Placing everything in Article III courts also is no panacea. Doing so would create the same risks of unintended consequences and accountability concerns as the other proposals, and—though it would head off some constitutional concerns—may also be subject to constitutional attack.

1. In Many Contexts, This Proposal Makes Little Policy Sense

To begin, moving all adjudication into the judicial branch often would be a downright horrible policy, especially for high-volume adjudication systems. There are good reasons why benefits programs, which prompt most agency adjudications, are administered by agencies rather than courts. We have already detailed many of those rule-of-law values associated with the standard APA model of agency-head review and supervision—such as inter-decisional consistency, efficiency, and access to justice.^[231]

There is, of course, also the question of political feasibility and congressional capacity. There are more than 12,000 agency adjudicators.^[232] One cannot possibly imagine the Senate having the political will or capacity to confirm thousands of new Article III judges to handle these adjudications. (This is a similar concern with expanding the Article I judiciary, which also requires Senate confirmation.) This proposal suffers from the same shortfalls as the other proposals, in that it would be hugely disruptive, especially to high-volume adjudication systems and likely result in unintended consequences. The intended and unintended costs of transferring these adjudications would be much greater than the federal central panel proposal or even the Article I courts initiative.

2. This Proposal May Prompt New Constitutional Concerns

There may also be constitutional concerns with this solution, at least for some categories of agency adjudication. To be sure, this proposal may avoid Article II concerns in many contexts; Article III judges are not subject to presidential removal. But in so doing, it could potentially raise new ones.

Most notably, not all action that falls within the broad category of “agency adjudication”—again, the application of law to particular facts—necessarily satisfies the “case or controversy” requirement of Article III.^[233] To be sure, when an agency or Article I court resolves a concrete dispute between opposing parties—such as a breach of contract claim—it is easy to see how that dispute could be placed in Article III courts. Indeed, at least some such claims arguably must be placed in Article III courts; Justice William Brennan, for example, argued that it may be “the very definition of tyranny” to allow Congress to place common-law suits outside of the Article III courts, a theme that may undergird more recent precedent.^[234] This is the issue the Supreme Court confronted in Jarkesy.^[235] We see no constitutional obstacle to placing in Article III courts most (if not all) disputes between private parties that currently are heard in other federal tribunals.

The constitutional difficulty may arise in adjudications that do not clearly present an adversarial relationship. Hayburn’s Case,^[236] decided in 1792, at least suggests that Congress lacks constitutional power to place a pension program in the federal courts precisely because it was non-judicial in character.^[237] The implications of that analysis for, say, the SSA are notable. As the Supreme Court has explained:

The differences between courts and agencies are nowhere more pronounced than in Social Security proceedings. Although “[m]any agency systems of adjudication are based to a significant extent on the judicial model of decisionmaking,” the SSA is “[p]erhaps the best example of an agency” that is not. Social Security proceedings are inquisitorial rather than adversarial. It is the ALJ’s duty to investigate the facts and develop the arguments both for and against granting benefits, and the Council’s review is similarly broad. The Commissioner has no representative before the ALJ to oppose the claim for benefits, and we have found no indication that he opposes claimants before the Council. ^[238]

Of course, it may be possible to refashion agency adjudication to put many matters in Article III courts—indeed, many social security cases are heard in federal court even today.^[239] If, for example, the SSA were to make initial, non-adversarial determinations, followed by litigation in an Article III court whenever SSA rejects the person’s view, it is hard to see the constitutional problem at that point because adverseness would exist.^[240] But that initial SSA assessment is itself a form of adjudication. Perhaps that problem can also be addressed by changing the statute. We do not evaluate a scheme that no one has fleshed out. Instead, our point, more modestly, is that moving benefits programs from the Article II agencies into Article III courts could prompt plausible constitutional objections—in addition to the obvious practical problems of efficient and cost-effective adjudication of claims discussed above.

There may also be other constitutional objections. Some matters—such as those touching on national defense—may implicate core Article II authority.^[241] It is not hard to imagine, for example, constitutional objections if Congress were to place matters of military justice for incidents on the battlefield itself in Article III courts.^[242] Proponents of moving administrative adjudication into Article III courts may respond that subjects like the military should be treated differently. But some lines are not easily drawn.

To be fair, we are aware of only one proposal—by the Competitive Enterprise Institute—that recommends that all agency adjudications be transferred to Article III courts.^[243] Recent reform proposals by Steven Calabresi and Michael Rappaport, by contrast, have zeroed in on a smaller subset of agency adjudication: those that affect private rights and quasi-private rights and/or impose civil monetary penalties. Professor Calabresi, for instance, suggested replacing just 158 ALJs—the ones he determined have the power to impose civil penalties—with Article III judges.^[244] If truly limited to the adjudications dealing with private rights and/or potentially imposing civil penalties, such removal may well make sense (and would be constitutional).^[245] But the answer becomes less clear if the proposal is to move all higher-volume agency adjudications to Article III courts. More to the point, moving 150 or so ALJ positions to the Article III judiciary does absolutely nothing to address the larger concerns about decisional independence with the 12,000 or so other agency adjudicators in the federal administrative judiciary.

IV. A Path Forward

For all the reasons explained above, it is doubtful that the leading proposals to save agency adjudication will do so. Indeed, until we more fully understand the scope of the claimed problem and the impact of each sweeping reform proposal, these proposed solutions have the potential to cause substantial and widespread problems for the millions of individuals subject to agency adjudication each year, while not even necessarily solving the motivating constitutional flaws. Perhaps this Article’s most important takeaway is that agency adjudication must be saved from these reform proposals.

But that does not mean there is no path forward. Here we advance two narrowly tailored proposals that Congress and the Executive Branch, respectively, could implement. These proposals avoid sweeping reforms and accompanying costs to the adjudication system, yet likely address the threats to decisional independence. Each reform could be pursued on its own, but their combination would be, in our view, even more effective.

A. Article I Solution: Congress’s Anti-Removal Power

Congress has an important role to play—but this role is not to enact dual-layer restrictions on removal. The APA purports to safeguard decisional independence by statute, but that sort of statutory safeguard only works in a world where the unitary executive theory has no teeth. In 1946, Congress perhaps could be forgiven for concluding that the Constitution allows statutory restrictions on removal, but the rule from Humphrey’s Executor^[246]—and similar adjudication cases like Wiener v. United States^[247]—is now in retreat.^[248] Thus, any attempt to save agency adjudication through dual-layer statutory removal restrictions is likely to fail. Reformers should accept that the Supreme Court now views restrictions on the President’s Article II removal authority with considerable skepticism if not outright hostility.^[249]

The Constitution, however, provides Congress with powerful tools to discourage removal without formally preventing it. These tools have a longstanding pedigree, and the Supreme Court’s most recent cases support, rather than cast doubt, on their use. And at least in the agency adjudication context, the use of these tools could attract bipartisan support if Congress considers the risk sufficiently serious and has the political will to use them.

In Congress’s Anti-Removal Power, two of us (Nielson and Walker) document how the Constitution empowers Congress to dissuade presidents from removing Executive Branch officials by making removal more politically costly.^[250] Alexander Hamilton identified the Appointments Clause as such a tool. As he explained, not only does Senate confirmation for principal officers prevent presidents from selecting individuals who would be little more than “the obsequious instruments of [the President’s] pleasure,” but it also inherently creates greater “stability in the administration.”^[251] After all, “[w]here a man in any station had given satisfactory evidence of his fitness for it, a new President would be restrained from attempting a change in favor of a person more agreeable to him, by the apprehension that a discountenance of the Senate might frustrate the attempt.”^[252] In other words, because the President cannot simply install someone he prefers more, the White House must pause before removing an incumbent and decide whether the benefits of removal—discounted by the possibility that the Senate will not confirm the replacement or at least will not do so instantaneously—exceed the costs of removal, including political costs. James Madison made a similar point. Although he vigorously defended the President’s plenary power to remove Executive Branch officers, Madison also explained that Congress could check that power through other means, including by threatening impeachment and political embarrassment.^[253]

In addition to the Appointments Clause, Congress has other tools to increase the costs of presidential removal. In Congress’s Anti-Removal Power, we categorize these tools in the following chart.^[254] These include higher cloture voting requirements for confirmation votes (which, by increasing the likelihood that the Senate will not confirm a replacement, force the White House to more steeply discount the benefits of removal), reason-giving requirements (which signal to the White House that if the reason for removal is not a good one, the Senate will not confirm a replacement), pre-commitments to holding hearings should removal occur (which increase the political costs of removal), and anti-evasion tools to prevent recess appointments or acting officials (which also require the White House to discount the benefits of removal more than usual). Both in principle and in practice, it appears that Congress’s use of these tools discourages some presidential removal, especially for offices that are not high priorities to the White House.^[255]

If Congress wants to create some decisional independence for agency adjudicators, it could begin to use creatively some of the tools from this anti-removal toolkit. As a preliminary matter, Congress could send statutory signals in favor of decisional independence by enacting legislative findings on the importance of decisional independence, the expert-based qualifications for hiring, and the appropriate (and inappropriate) reasons for firing.^[256] These enacted findings would not be statutory commands but instead expectations—signaling to the President, the agency, and the public that Congress cares about decisional independence, both in terms of hiring and firing, and will use its political tools to protect it. Formally, such signals may not be important. The President can ignore them and remove an official. In the real world, however, history teaches that such efforts create norms against and increase the political cost of presidential interference.^[257] Congress could take steps to develop similar norms in the agency adjudication context. Given that even pro-unitary executive jurists like Chief Judge Taft expressed discomfort with political interference in adjudication^[258] and that there is widespread political support for impartial case-specific adjudication,^[259] there may be more political will to establish and reinforce these norms in the agency adjudication context than in many others.

Moreover, for some agency adjudicators—perhaps appellate-level adjudicators like those at the Board of Immigration Appeals—Congress could, per the Appointments Clause, require them to be presidentially appointed on advice and consent of the Senate.^[260] Rebecca Eisenberg and Nina Mendelson have suggested this as a potential option to address patent adjudication after Arthrex—i.e., by creating a panel of Senate-confirmed final decisionmakers.^[261] This would affect political insulation in terms of hiring and firing. To be sure, it would be infeasible to require thousands of agency adjudicators to be Senate-confirmed, which is one problem with creating more Article I courts or replacing agency adjudication with Article III administrative courts.^[262] But one could imagine Congress exploring this anti-removal tool with respect to a subset of appellate adjudicators to create an additional measure of decisional independence. One important feature of the Appointments Clause is that it gives Congress certain discretion over the appointment process for inferior officers,^[263] thus allowing Congress to tailor appointment to particular situations. The mere possibility, moreover, that an unhappy Congress may require Senate confirmation may dynamically reinforce norms against political interference in the adjudicative process.

Short of designating agency adjudicator appointments for Senate confirmation, Congress could use its anti-removal “soft tools” to raise the stakes for agency heads in deciding whether to fire an agency adjudicator. For instance, it could require the agency head to notify Congress of any termination and to provide the reason for the firing.^[264] In this way, Congress could require the agency head to provide reasons—indeed, good reasons, at least de facto, given the political costs of offering a bad reason^[265]—for firing an agency adjudicator. This is not a new tool. For generations, Congress has required the President to provide the reasons for the firing of the Comptroller of the Currency and inspectors general.^[266] By design, this requirement raises the cost of removal by forcing the agency head to reveal the reasons for the firing publicly and to Congress, to be judged in the court of public opinion and by the agency’s congressional overseer. As the congressional record demonstrates, the political reality is that presidents generally will not remove officials so protected absent a “good reason[].”^[267] Such a reason-giving requirement should have even more force when the political head of an agency—rather than the President—is firing ALJs.

Congress could also strengthen the reason-giving requirement by pre-committing by statute to hold an oversight hearing if the agency head fires an agency adjudicator, or perhaps only if the agency head fails to provide the statutorily required (good) reason.^[268] At that hearing, the fired adjudicator and agency head would testify, as well as any other witnesses the committee wanted to call.^[269] This hearing would raise the removal costs even more, as agencies are quite receptive to congressional oversight.^[270] To be sure, the end result would not eliminate an agency head’s power to fire or discipline an agency adjudicator. But in the real world, it should discourage removals not based on merit.

Returning to the core “soft tools” proposal, however, the reason-giving and congressional-hearing approach has the added benefit of keeping Congress appraised and focused on the issue of decisional independence in agency adjudication. If the threat against decisional independence were grave or widespread, the congressional hearings and oversight would help uncover and assess it. This proposal would also avoid the massive costs of the bolder reform proposals. It would keep the agency adjudication systems in place, but with a political rather than legalistic safeguard against political interference. In other words, not only would it be on firmer constitutional grounds, but pursuing reform through Congress’s anti-removal power would also be a less disruptive option than those discussed in Part III, which is critical in an area with so much empirical uncertainty as to the scope of the problem and the impacts of the broader reform proposals.^[271] Furthermore, Congress’s ability to modulate its use of its anti-removal power—stronger protections for certain positions, weaker for others—should help mitigate the risk of unintended consequences.

B. Article II Solution: Impartiality Regulations

There is also a role for the Executive Branch itself. All too often administrative law scholarship focuses on the checks imposed by Congress and courts^[272]—perhaps driven by the temptation to focus on what is easiest to see.^[273] But internal administrative law—that is, “the internal directives, guidance, and organizational forms through which agencies structure the discretion of their employees and presidents control the workings of the executive branch”^[274]—can also impose discipline on agencies.^[275] In countless contexts, internal administrative law contributes to a more effective and fair regulatory process, including for agency adjudication.^[276]

This important insight can and should play a key role in saving agency adjudication. Notwithstanding the Court’s embrace of unitary executive principles, the Executive Branch can use internal administrative law to take a step to ensure decisional independence for agency adjudicators, both in terms of hiring and firing. In other words, the President’s broad authority over the Executive Branch includes the ability to not exercise control. Indeed, unitary executive principles support such uses of internal administrative law.^[277] One of the central explanations the Court has given for robust presidential control over the Executive Branch is the President’s ability to protect liberty and, importantly, the political incentive that the President has to do so. As Seila Law explains:

The resulting constitutional strategy is straightforward: divide power everywhere except for the Presidency, and render the President directly accountable to the people through regular elections. In that scheme, individual executive officials will still wield significant authority, but that authority remains subject to the ongoing supervision and control of the elected President. Through the President’s oversight, “the chain of dependence [is] preserved,” so that “the lowest officers, the middle grade, and the highest” all “depend, as they ought, on the President, and the President on the community.”^[278]

That logic supports presidential efforts to encourage confidence in the decisional independence of adjudicators, including by unilaterally foreswearing exercise of Article II authority to interfere with the impartial hiring and firing of individual adjudications. After all, merely because the President has a power does not mean that it must be used in every case. The President, for example, has a robust pardon power;^[279] it does not follow that the President should use that power in every case. To the contrary, there can be political costs associated with using the pardon power too casually, plus the facts of individual cases may well not merit such relief. The same is true for appointments; even where the President has unilateral appointment authority (such as during a recess),^[280] it does not follow that the President should always use that authority in every case. The President is Commander-in-Chief of the armed forces, yet often wisely lets battlefield commanders make operational calls. Instead, responsible use of power sometimes means standing aside while others act. The same sort of analysis could apply to the removal power; the President could unilaterally impose restrictions on political interference with adjudication, not because Congress has required it, but because it is intrinsically the wise thing to do or at least because the President does not want to incur political costs. The Trump Administration’s decision following Lucia to emphasize its policy that ALJs should be “impartial and committed to the rule of law” illustrates the point. ^[281]

Kent Barnett is the leading proponent of using internal administrative law to limit political interference in the hiring, supervision, and firing of agency adjudicators. In Regulating Impartiality in Agency Adjudication, Professor Barnett argues that to safeguard adjudicator decisional independence in a world of unrestricted presidential removal, “the White House and agencies should use executive orders and regulations to mimic and improve administrative adjudicators’ existing statutory protections from at-will removal.”^[282] Indeed, he urges not just internal restrictions on removal of adjudicators, but also formal entrenchment of the “tiered” removal structure that Free Enterprise Fund calls into doubt.^[283] He suggests similar impartiality regulations to preserve apolitical, meritocratic hiring of agency adjudicators.^[284] In his view, such a targeted use of internal administrative law not only has the capacity to stave off due process concerns, but also accords with historical practice.^[285]

As Professor Barnett acknowledges, however, one significant limitation of relying on internal administrative law is that it is less sticky than statutory law.^[286] What one administration does through regulatory tools, after all, can be undone by another administration using those same tools.^[287] Professor Barnett argues, however, that impartiality regulations can be made more permanent if they are promulgated through the APA’s notice-and-comment rulemaking process.^[288] He is correct that rules are generally stickier than less formal regulatory tools,^[289] but there are at least two problems with his proposal. First, the notice-and-comment process is not especially difficult for all rules.^[290] Second, and more fundamentally, it is doubtful that a regulation—especially one promulgated by a prior President—could prevent the President from freely removing an agency official. The Supreme Court has concluded that Article II itself provides the President with removal power, and a regulation—even one supported by statutory law—cannot override a constitutional power. Again, consider the pardon power: Does anyone think a regulation requiring a certain procedure before a President could issue a pardon would nullify a pardon issued outside of that procedure? Accordingly, no matter what a regulation says, an agency adjudicator is always at risk of being removed by the White House—thus casting a cloud over decisional independence.

Nonetheless, we agree with Professor Barnett that agencies should promulgate impartiality regulations and that the President should issue impartiality executive orders, even if regulations and executive orders are not formally binding on the President. Such measures do not offend Article II and provide at least some decisional independence, based on inertia if nothing else—the more steps that an administration must do to remove someone, the less likely it is that the administration will do so, at least at the margins.

But we also urge presidents to take credible steps to increase the political costs of interfering with adjudicatory decisions. One way to do that is to publicly and prominently proclaim that political interference in individual adjudication hearings is improper. Although the President could still retreat from a “good government” pledge, the more politically costly such retreat becomes, the less likely it is that a president will engage in it. Presidents often find value in “tying themselves to the mast”; this is another possible place where such an effort make sense.^[291] Such public commitments against undue interference, moreover, set expectations for future presidential administrations, the deviation from which would not be costless.

Similarly, the President can empower ombudsmen and other officials to cry foul to prevent undue interference with agency adjudication. Congress may be able to use its anti-removal power to create “offices that protect the public from administrative overreach, such as agency ombuds, privacy offices, and other ‘offices of goodness.’”^[292] But whether Congress acts, the President can use executive orders or other forms of presidential direction to entrust certain individuals with responsibility to prevent undue interference. If an agency ombuds sounds an alarm, the political salience will likely increase—the prospect of which should reduce the likelihood of political interference to begin with—especially because hearing-level agency adjudicator independence (within reasonable limits) is an issue that commands widespread support.^[293]

One important advantage of using political norms to safeguard impartiality is that they may be better able to adjust to unforeseen circumstances. There may be sound reasons for the President to intervene in an agency adjudication that are hard to identify ex ante; a law barring such actions may be overinclusive. A political check, however, grounded in social norms is more flexible. Given the public’s inherently asymmetric preferences in favor of fair adjudication (“unbiased” adjudication undoubtedly polls better than the alternative),^[294] the risk that the White House will unduly interfere with adjudication decisions is unlikely so long as the President’s decision is public; thus, the President would need a pretty good explanation before being willing to take the political heat.^[295]

To be sure, we do not claim that it is impossible to imagine situations where the President disregards political norms against interference with individual adjudications or that political checks will never fail. Our point, more modestly, is that in the real world, political norms and checks generally should be enough to protect agency–adjudicator decisional independence, and that whatever downsides this option poses pale in comparison with the downsides of other proposed solutions. Furthermore, when impartiality regulations are coupled with Congress’s targeted use of its anti-removal power, it becomes even less likely that political interference will be a serious problem. There are no silver bullets, but these approaches should help safeguard decisional independence in agency adjudication while avoiding the significant costs and legal uncertainty of competing proposals.

* * *

Both of our proposals, while modest as a procedural matter, are quite useful as a policy matter. The key is to build the constituency for the standard model of agency adjudication with its dual pillars of decisional independence for hearing-level adjudicators and political control over the final adjudication decision.^[296] Yes, the agency head should have the final say, but absent merit-based concerns, she should not be able to bias through personnel pressure on hearing-level adjudicators the creation of the record and factfinding at the evidentiary stage. The public should demand—and Congress and the President should commit to—this fundamental principle of administrative governance of neutral, unbiased hearing-level agency adjudicators.^[297]

This political constituency can be mobilized across the ideological spectrum—from progressives concerned about benefits claimants and health, safety, and welfare, to libertarians fearful of government overreach, to conservatives seeking stability in law and regulation. It is hard to imagine many defending the idea of biased adjudicatory hearings and factfinding. As Justice Gorsuch (and Chief Justice Roberts) has worried in a related adjudicatory context, “[p]owerful interests are capable of amassing armies of lobbyists and lawyers to influence (and even capture) politically accountable bureaucracies.”^[298] The standard model for agency adjudication, when properly understood, seeks to address those concerns.

Conclusion

Modern developments in administrative law have weakened the footing of decisional independence in agency adjudication. Indeed, the Supreme Court may soon conclude that laws preventing the President from exercising plenary control of all Executive Branch officers, including adjudicators, violate Article II. At the very least, we expect the Court to eventually conclude that the agency head must be able to remove hearing-level adjudicators for cause. The Court is aware of the downsides of political interference with agency adjudications, but the structure of its constitutional analysis about the appointment and removal of Executive Branch officers is such that it is far from obvious how to carve out adjudication. Given the centrality of agency adjudication to federal operations, this potentially significant constitutional development should be squarely addressed.

Unfortunately, some cures are worse than the disease. We agree that a measure of decisional independence is at risk, but the leading reforms are not the answer. Massively overhauling the world of agency adjudication will trigger unintended consequences that will harm the millions of individuals who depend on agency adjudication, while not fully solving the deeper constitutional and policy concerns. Instead, saving agency adjudication will require both political branches to act: Congress can begin systematically using its anti-removal power, and the Executive Branch can self-impose internal regulations to preserve decisional independence of agency adjudicators. Such targeted reforms should allow the nation to retain the important benefits of agency adjudication without running afoul of the Court’s commitment to robust presidential control of the Executive Branch.

1. . U.S. Cts., Authorized Judgeships 8 (2018), http://www.uscourts.gov/sites/default
   /files/allauth.pdf [https://perma.cc/4T3W-G4NT]. ↑
2. . John Gramlich, How Trump Compares with Other Recent Presidents in Appointing Federal Judges, Pew Rsch. Ctr. (Jan. 13, 2021), https://www.pewresearch.org/fact-tank/2021/01/13/how-trump-compares-with-other-recent-presidents-in-appointing-federal-judges/ [https://perma.cc
   /9KD5-EP58]; Diversity of the Federal Bench, Am. Const. Soc’y, https://www.acslaw.org
   /judicial-nominations/diversity-of-the-federal-bench/ [https://perma.cc/CT95-9EXP]. ↑
3. . Judicial Appointment Tracker, Heritage Found., https://www.heritage.org/judicialtracker [https://perma.cc/NDF8-R7HR]; Judicial Nominations, Am. Const. Soc’y (last visited Nov. 5, 2024), https://www.acslaw.org/judicial-nominations/ [https://perma.cc/4MEZ-SUGG]. ↑
4. . See, e.g., Paul Kane, Senate Democrats Vastly Outspent by Right in Gorsuch Fight, Wash. Post (Mar. 18, 2017, 3:57 PM), http://wapo.st/2n1dcrc?tid=ss_tw [https://perma.cc/433H-6QS7] (reporting Republican Party estimates that $3.3 million were spent on ads to support the confirmation of now-Justice Gorsuch); see also Am. Bar Ass’n, Standing Committee on the Federal Judiciary: What It Is and How It Works 1–3 (2017), https://www.americanbar
   .org/content/dam/aba/administrative/federal_judicary/2024-backgrounder.pdf [https://perma.cc
   /SGY6-9536] (detailing the ABA’s judicial-nominee evaluation process that has operated since 1953). ↑
5. . Not to mention “the 50 or so Article I judges who populate the territorial courts, the Court of Federal Claims, the Tax Court, the Court of Appeals for the Armed Forces, and the Court of Appeals for Veterans Claims.” Christopher J. Walker, Charting the New Landscape of Administrative Adjudication, 69 Duke L.J. 1687, 1687–88 (2020). ↑
6. . ALJs by Agency, U.S. Off. Pers. Mgmt., https://www.opm.gov/services-for-agencies/administrative-law-judges/#url=By-Agency [https://perma.cc/GC62-XDKU]. ↑
7. . Kent Barnett & Russell Wheeler, Non-ALJ Adjudicators in Federal Agencies: Status, Selection, Oversight, and Removal, 53 Ga. L. Rev. 1, 22 n.119, 32 (2018). ↑
8. . Kent H. Barnett, Some Kind of Hearing Officer, 94 Wash. U. L. Rev. 515, 544 (2019). ↑
9. . And there are tens of thousands of other agency personnel that engage in the “hundreds of thousands of less-formal adjudications in countless regulatory contexts.” Walker, supra note 5, at 1688. ↑
10. . Id. ↑
11. . See, e.g., Adam B. Cox & Emma Kaufman, The Adjudicative State, 132 Yale L.J. 1769, 1782 (2023) (noting how the conflict between the unitary executive theory and the separate functions theory “threatens to pull the field [of administrative law] apart,” particularly in the context of administrative courts). ↑
12. . The Supreme Court has also recently held that litigants can challenge agency structure—including the role of agency adjudicators—directly in federal court. See Axon Enter., Inc. v. FTC, 143 S. Ct. 890, 897 (2023) (allowing constitutional challenges directly in federal district court); see also Carr v. Saul, 141 S. Ct. 1352, 1362 (2021) (rejecting issue-exhaustion requirement for structural challenges in one statutory scheme). ↑
13. . 144 S. Ct. 2117 (2024). ↑
14. . 561 U.S. 477 (2010). ↑
15. . Jarkesy v. SEC, 34 F.4th 446, 463 (5th Cir. 2022) (citing Free Enter. Fund, 561 U.S. at 498), aff’d on other grounds, 144 S. Ct. at 2139. The Fifth Circuit panel also held that because enforcement of securities law involves private—rather than public—rights, SEC adjudication offends the Seventh Amendment right to a jury trial and, further, that Congress violated the nondelegation doctrine by not providing the SEC with an “intelligible principle” when deciding whether to pursue an action before an ALJ. Id. at 449, 451. These holdings are also significant, but they do not apply to the entirety of agency adjudication. ↑
16. . Id. at 465. ↑
17. . Jarkesy v. SEC, 51 F.4th 644, 647 (5th Cir. 2022) (Haynes, J., dissenting from denial of rehearing en banc). ↑
18. . Benjamin M. Daniels & Trevor L. Bradley, Fifth Circuit Decision Threatens to Upend SEC’s Use of Administrative Proceedings, Nat’l L. Rev. (June 7, 2022), https://natlawreview.com
    /article/fifth-circuit-decision-threatens-to-upend-sec-s-use-administrative-proceedings [https://
    perma.cc/4XLX-99D5]; see also, e.g., Noah Rosenblum, The Case That Could Destroy the Government, Atlantic (Nov. 27, 2023), https://www.theatlantic.com/ideas/archive/2023/11/
    securities-and-exchange-commission-v-jarkesy-supreme-court/676059/ [https://perma.cc/7TVM-2V8C] (“Unscrupulous presidents would use agencies to punish their opponents and reward their allies. This would do more than turn regulators into political handmaidens; it would destabilize markets, stifle growth, and inevitably lead to financial crises.”). ↑
19. . SEC v. Jarkesy, 144 S. Ct. 2117, 2139 (2024) (“A defendant facing a fraud suit has the right to be tried by a jury of his peers before a neutral adjudicator. Rather than recognize that right, the dissent would permit Congress to concentrate the roles of prosecutor, judge, and jury in the hands of the Executive Branch.”). The Fifth Circuit has recently reiterated that its dual-layer removal holding (as well as its nondelegation holding) remains circuit precedent. See Jarkesy v. SEC, slip op. at 1, No. 20-61007 (5th Cir. Nov. 12, 2024) (per curiam) (“[W]e reiterate our prior holdings in this case.”). ↑
20. . Jarkesy, 34 F.4th at 478 (Davis, J., dissenting); see generally Jack M. Beermann, The Never-Ending Assault on the Administrative State, 93 Notre Dame L. Rev. 1599 (2018) (detailing constitutional attacks on the administrative state). ↑
21. . See, e.g., Jerome Nelson, Administrative Law Judges’ Removal “Only for Cause”: Is That Administrative Procedure Act Protection Now Unconstitutional?, 63 Admin. L. Rev. 401, 402 (2011) (identifying the risk that Free Enterprise Fund’s logic dooms the constitutionality of ALJs); see also Linda D. Jellum, “You’re Fired!” Why the ALJ Multi-Track Dual Removal Provisions Violate the Constitution and Possible Fixes, 26 Geo. Mason L. Rev. 705, 741 (2019) (concluding, reluctantly, that the logic of the Supreme Court’s cases makes it “inevitable” that “ALJ multi-track removal provisions violate the Constitution”); Jackson C. Blais, Note, Mischief Managed? The Unconstitutionality of SEC ALJs Under the Appointments Clause, 93 Notre Dame L. Rev. 2115, 2116 (2018) (anticipating the Fifth Circuit’s holding in Jarkesy); Jeffrey S. Lubbers, (If the Supreme Court Agrees) The SG’s Brief in Lucia Could Portend the End of the ALJ Program as We Have Known It, Yale J. on Regul.: Notice & Comment (Apr. 10, 2018), https://www.yalejreg.com/nc
    /if-the-supreme-court-agrees-the-sgs-brief-in-lucia-could-portend-the-end-of-the-alj-program-as-we-have-known-it-by-jeffrey-s-lubbers/ [https://perma.cc/47HZ-4KFR] (predicting the possible end of the ALJ program). ↑
22. . See, e.g., Fleming v. U.S. Dep’t of Agric., 987 F.3d 1093, 1115 (D.C. Cir. 2021) (Rao, J., concurring in part and dissenting in part) (concluding that ALJ dual for-cause removal provisions conflict with Free Enterprise Fund); see also Neomi Rao, A Modest Proposal: Abolishing Agency Independence in Free Enterprise Fund v. PCAOB, 79 Fordham L. Rev. 2541, 2544, 2550 (2011) (arguing that Free Enterprise Fund’s logic extends to all removal restrictions). ↑
23. . See Free Enter. Fund, 561 U.S. at 540–41, 543 (Breyer, J., dissenting) (“Does every losing party before an ALJ now have grounds to appeal on the basis that the decision entered against him is unconstitutional?”). ↑
24. . See infra subpart II(B) (discussing, among other things, Seila Law LLC v. Consumer Fin. Prot. Bureau, 140 S. Ct. 2183 (2020) and Collins v. Yellen, 141 S. Ct. 1761 (2021)). ↑
25. . See, e.g., Jennifer Nou, Dismissing Decisional Independence Suits, 86 U. Chi. L. Rev. 1187, 1187 (2019) (“Administrative adjudication is poised for avulsive change.”). ↑
26. . See Seila Law, 140 S. Ct. at 2198 n.2 (explaining that “under our constitutional structure” agency adjudication is an “exercise[] of . . . the ‘executive Power’” (quoting City of Arlington v. FCC, 569 U.S. 290, 305 n.4 (2013))). ↑
27. . See Collins, 141 S. Ct. at 1787 (holding unconstitutional a removal provision even assuming it allowed the President to remove the officer for policy disagreements because the Constitution requires “at will” employment). ↑
28. . See, e.g., Richard E. Levy & Robert L. Glicksman, Restoring ALJ Independence, 105 Minn. L. Rev. 39, 45 (2020) (“Although other recent threats to the rule of law may deservedly garner the headlines, we should not lose sight of the critical role that impartial agency adjudication plays.”); Elizabeth Magill & Adrian Vermeule, Allocating Power Within Agencies, 120 Yale L.J. 1032, 1075 (2011) (“The bottom line is that senior agency officials can ‘manage’ adjudication only to the extent that that supervision does not relate to the resolution of particular cases.”). ↑
29. . See Kent Barnett, Regulating Impartiality in Agency Adjudication, 69 Duke L.J. 1695, 1698 (2020) [hereinafter Barnett, Regulating Impartiality] (“Another constitutional provision—the Due Process Clause—requires impartiality for agency adjudication, but whether impartiality entails protection from at-will removal is less clear.” (footnote omitted) (citing Schweiker v. McClure, 456 U.S. 188, 195 (1982))). That said, it is possible that agency adjudicators should only be able to resolve public (rather than private) rights, with public rights being defined as those rights to which due process protections do not attach, perhaps including benefits. See Kent Barnett, Due Process for Article III—Rethinking Murray’s Lessee, 26 Geo. Mason L. Rev. 677, 699 (2019) [hereinafter Barnett, Due Process] (discussing how agencies could adjudicate public-rights related matters, such as benefits, without running afoul of Article III); cf. SEC v. Jarkesy, 144 S. Ct. 2117, 2132, 2139 (2024) (also concluding that ALJs cannot resolve certain private rights). Here, we do not address the precise constitutional point where political interference in adjudication violates due process. ↑
30. . Seila Law, 140 S. Ct. at 2191 (quoting Free Enter. Fund v. Pub. Co. Acct. Oversight Bd., 561 U.S. 477, 514 (2010)). ↑
31. . See, e.g., Christopher J. Walker, Constitutional Tensions in Agency Adjudication, 104 Iowa L. Rev. 2679, 2703 (2019) (“As Oil States and Lucia illustrated, Justice Gorsuch and others are deeply concerned about the constitutional tensions between the importance of political accountability in the administrative state and the dangers of politics in agency adjudication.”); Cox & Kaufman, supra note 11, at 1771 (similarly identifying this tension in the Roberts Court’s jurisprudence). ↑
32. . Other scholars have argued that the Supreme Court should not extend the holdings of its recent removal cases (Collins, Seila Law, and Free Enterprise Fund) into the adjudication context. See infra subpart II(B). As we explain below, however, that sort of doctrinal move would be easier said than done; the Court’s logical syllogism (Article II requires that the President be able to remove all executive officers; officer X is an executive officer; therefore, the President can remove officer X) does not distinguish between types of executive officers. See, e.g., Barnett, Regulating Impartiality, supra note 29, at 1717 (“[A]n adjudicator-based exception would be inconsistent with the Court’s formalist doctrine . . . . If a functional exception exists for adjudicators under Article II, that exception at the very least conflicts with the Court’s separation-of-powers formalism and more problematically undermines the normative force of formalism altogether.”). ↑
33. . See infra subpart III(A). ↑
34. . See infra subpart III(B). ↑
35. . See infra subpart III(C). ↑
36. . Aaron L. Nielson & Christopher J. Walker, Congress’s Anti-Removal Power, 76 Vand. L. Rev. 1, 39–40 (2023). ↑
37. . See Aaron L. Nielson & Christopher J. Walker, The Early Years of Congress’s Anti-Removal Power, 63 Am. J. Legal History 219, 224 (2023) (marshalling evidence that the Framers intended Congress to discourage presidential removal, including from The Federalist Nos. 76, 77 (Alexander Hamilton) and 1 Annals of Cong. 517 (1789) (Joseph Gales ed., 1834) (statement of Representative James Madison)). ↑
38. . See infra subpart IV(A). ↑
39. . See Barnett, Regulating Impartiality, supra note 29, at 1700–01 (noting that internal control is a useful tool for the Executive). ↑
40. . See id. at 1700 (proposing internal regulations by the Executive branch to protect at-will removal practices without offending separation-of-powers principles). ↑
41. . See Exec. Order No. 13,843, § 1, 3 C.F.R. 844, 845 (2019) (explaining that, notwithstanding the President’s constitutional powers with respect to their appointment, “ALJs must display appropriate temperament, legal acumen, impartiality, and sound judgment”). ↑
42. . See infra subpart IV(B); see also Aaron L. Nielson, Sticky Regulations, 85 U. Chi. L. Rev. 85, 90 (2018) (explaining how notice-and-comment procedures create stickiness). ↑
43. . See infra subpart IV(A) (discussing Congress’s anti-removal power as a check to executive power to influence administrative decisionmaking). ↑
44. . See generally 5 U.S.C. §§ 551–59, 701–06 (permitting appeal of decisions to the agency head while also allowing for judicial review). This standard model is subject to congressional override in the agency’s governing statute. See, e.g., Dickinson v. Zurko, 527 U.S. 150, 154–55 (1999) (holding that to depart from the APA default rules, the agency’s governing statute must suggest “more than a possibility of a [different] standard, and indeed more than even a bare preponderance of evidence,” and stating that the exception “must be clear”); see also Stephanie Hoffer & Christopher J. Walker, The Death of Tax Court Exceptionalism, 99 Minn. L. Rev. 221, 243–45 (2014) (discussing the standards for departing from the APA’s default rules). ↑
45. . See Emily S. Bremer, The Rediscovered Stages of Agency Adjudication, 99 Wash. U. L. Rev. 377, 386, 415–16, 430, 448 (2021) (discussing the APA’s conceptual foundation); see also Barnett, Due Process, supra note 29, at 681 (explaining Congress has permitted agencies to adjudicate claims over public rights, which allows an agency head discretion without Article III involvement, and private rights, which requires some Article III oversight). ↑
46. . Ronald M. Levin, Administrative Judges and Agency Policy Development: The Koch Way, 22 Wm. & Mary Bill Rts. J. 407, 412 (2013). ↑
47. . Fact Sheet, U.S. Off. of Pers. Mgmt., https://www.opm.gov/services-for-agencies/administrative-law-judges/#url=Facts [https://perma.cc/7GBR-GKUN]. ↑
48. . Christopher J. Walker & Melissa F. Wasserman, The New World of Agency Adjudication, 107 Calif. L. Rev. 141, 143 (2019). ↑
49. . 5 U.S.C. § 556(c)(5). ↑
50. . 5 U.S.C. § 557(b); see also Levin, supra note 46, at 410 (explaining the application of the agency head being the authoritative source of policy within the agency). ↑
51. . See, e.g., William F. West, Neutral Competence and Political Responsiveness: An Uneasy Relationship, 33 Pol’y Stud. J. 147, 148–50 (2005) (reviewing public administration literature on neutral competence). ↑
52. . 5 U.S.C. § 556(b). ↑
53. . Id. ↑
54. . Id.; see also id. § 3105 (“Each agency shall appoint as many administrative law judges as are necessary for proceedings required to be conducted in accordance with sections 556 and 557 of this title.”). The APA provides that parties may move to exclude administrative law judges for “personal bias or other disqualification of a presiding or participating employee.” Id. § 556(b). ↑
55. . Id. § 554(d). Indeed, the APA has detailed prohibitions on ex parte communications “relevant to the merits of the proceeding,” requirements to make any such communications part of the public record of the proceeding, and authority for the agency to require the offending party “to show cause why his claim or interest in the proceeding should not be dismissed, denied, disregarded, or otherwise adversely affected on account of such violation.” Id. § 557(d)(1). ↑
56. . Id. § 554(d). ↑
57. . Emily S. Bremer, Presidential Adjudication, 110 Va. L. Rev. 1749, 1759 (2024). ↑
58. . 5 U.S.C. § 556(e). ↑
59. . Id. § 556(b). ↑
60. . Id. § 557(c)(3). If a party believes the agency’s decision is based on a material fact outside of the record, the party must have an opportunity to make a timely request for reconsideration. Id. § 556(e). ↑
61. . Id. § 3105. ↑
62. . For a discussion of panel stacking that occurred at the Patent & Trademark Office with administrative patent judges, see Walker & Wasserman, supra note 48, at 178–87. ↑
63. . These factors include experience, Vanessa K. Burrows, Cong. Rsch. Serv., RL34607, Administrative Law Judges: An Overview 3 (2008), the results of an OPM-administered exam, id. at 2–3, and veteran status, 5 U.S.C. § 3309; 5 C.F.R. § 302.201 (2024). For a discussion on the controversial practice of veteran status and its impact on the final list of ALJs, see Jeffrey S. Lubbers, Federal Administrative Law Judges: A Focus on Our Invisible Judiciary, 33 Admin. L. Rev. 109, 115–16 (1981). The Administrative Conference of the United States has repeatedly recommended that Congress modify this preference in an effort to increase the number of qualified ALJ candidates. E.g., Admin. Conf. of the U.S., Recommendation 92-7, The Federal Administrative Judiciary 6 (1992), https://www.acus.gov/sites/default/files/documents/92-7.pdf. [https://perma.cc/4GJQ-UAJP]. ↑
64. . Burrows, supra note 63, at 2–3. To gain more control over the appointments process, agencies can hire ALJs who already work in another agency, Paul R. Verkuil, Reflections Upon the Federal Administrative Judiciary, 39 UCLA L. Rev. 1341, 1361 n.82 (1992), or wait until several vacancies exist so that they obtain a larger list of candidates from OPM, Admin. Conf. of the U.S., supra note 63, at 5. Agencies also may borrow an ALJ from another agency with that agency’s consent. 5 C.F.R. § 930.208(a) (2024); 5 U.S.C. § 3344. ↑
65. . Exec. Order No. 13,843, §§ 2–3, 3 C.F.R. 844, 844–45 (2019). ↑
66. . 5 U.S.C. § 2302(a)(2)(B), (b)(1)(E), (b)(3). ↑
67. . Exec. Order No. 13,843, §§ 1, 4(b), 3 C.F.R. at 844, 847. ↑
68. . Recent Guidance, Guidance on Administrative Law Judges After Lucia v. SEC (S. Ct.), July 2018, 132 Harv. L. Rev. 1120, 1122 (2019). ↑
69. . 5 U.S.C. §§ 4301(2)(D), 4302(a); Admin. Conference of the U.S., supra note 63, at 2. ↑
70. . Admin. Conference of the U.S., supra note 63, at 2; see also Harold J. Krent, Presidential Control of Adjudication Within the Executive Branch, 65 Case W. Rsrv. L. Rev. 1083, 1108 (2015) (observing that agencies cannot set or dock pay for ALJs or require performance reviews). ALJs’ pay is set out in significant detail in 5 U.S.C. § 5372, with three levels of basic pay. Notably, Congress moved from a two-tiered pay grade for ALJs—which was supposed to account for the difficulty of the kinds of cases that ALJs heard—and raised their pay. See Verkuil, supra note 64, at 1352 (“Both of these developments suggest that Congress is not fully comfortable with the more limited role of hearing examiners originally contemplated by the APA.”). ↑
71. . 5 C.F.R. § 930.210(b) (2024). ↑
72. . 5 U.S.C. § 7521(a); Jennifer L. Selin & David E. Lewis, Admin. Conf. of the U.S., Sourcebook of United States Executive Agencies 97 (2d ed. 2018). ↑
73. . The Supreme Court has explained that “due process requires a ‘neutral and detached judge in the first instance.’” Concrete Pipe & Prods of Cal., Inc. v. Constr. Laborers Pension Tr. for S. Cal., 508 U.S. 602, 617 (1993) (quoting Ward v. Vill. of Monroeville, 409 U.S. 57, 61–62 (1972)); see also Amanda Frost, The Limits of Advocacy, 59 Duke L.J. 447, 460–61 (2009) (explaining that one of the “core elements” of due process holds that “because the decisionmaker must remain impartial, he cannot serve as an advocate for the interests of either party”). ↑
74. . Harold J. Krent & Lindsay DuVall, Accommodating ALJ Decision Making Independence with Institutional Interests of the Administrative Judiciary, 25 J. Nat’l Ass’n Admin. L. Judges 1, 2 (2005). ↑
75. . See U.S. Const. art. III, § 1, cl. 2. (providing lifetime tenure and salary protection for all federal judges). ↑
76. . The same holds for federal courts. For example, the Supreme Court has been criticized by some commentators for its “shadow docket,” in which the Court rules on cases that do not receive full briefing by issuing succinct orders, generally issued without legal justification. See, e.g., William Baude, Foreword: The Supreme Court’s Shadow Docket, 9 NYU J.L. & Liberty 3–5, 9–11 (2015) (arguing that while the Supreme Court’s merit cases follow “procedural regularity,” which helps support the legitimacy of those decisions, the Court’s “shadow docket” may raise “questions of consistency and transparency” by following an “ad hoc or unexplained” process). ↑
77. . See, e.g., Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 416–17 (1971) (evaluating whether the Secretary of Transportation properly and reasonably authorized the use of federal funds for the construction of a highway through a public park under the relevant statutes). ↑
78. . See FCC v. Allentown Broad. Corp., 349 U.S. 358, 364 (1955) (rejecting deferential review of ALJ decisions by administrative agencies); 5 U.S.C. § 557(b) (stating that an agency reviewing an initial decision “has all the powers which it would have in making the initial decision”). ↑
79. . Krent & DuVall, supra note 74, at 29. ↑
80. . See generally Thomas W. Merrill, Article III, Agency Adjudication, and the Origins of the Appellate Review Model of Administrative Law, 111 Colum. L. Rev. 939 (2011) (describing how modern administrative law is based on the appellate review model, in which reviewing courts defer to trial courts on questions of fact but not questions of law). The administrative review model further digresses from the appellate model because[t]he appellate review model in the civil litigation context is based on the record from the prior proceeding, and the reviewing court does not engage in independent fact-finding. Likewise, the standard of review reflects the comparative expertise of the various institutions, with more or less deferential review depending on whether the issue is more factual or legal, respectively.Christopher J. Walker, Against Remedial Restraint in Administrative Law, 117 Colum. L. Rev. Online 106, 111 (2017). ↑
81. . See, e.g., Richard H. Fallon, Jr., John F. Manning, Daniel J. Meltzer & David L. Shapiro, Hart & Wechsler’s The Federal Courts and the Federal System 379–80 (7th ed. 2015) (noting the policymaking function in agency adjudication); Richard H. Fallon, Jr., Of Legislative Courts, Administrative Agencies, and Article III, 101 Harv. L. Rev. 915, 923 (1988) (“[Agencies are] entities established by Congress to administer statutory schemes of federal regulation. Although agencies typically have rulemaking power, their functions also include adjudication.” (footnote omitted)). ↑
82. . Universal Camera Corp. v. NLRB, 340 U.S. 474, 496 (1951). Most frequently, the agency head may reverse the ALJ’s initial decision for policy reasons. However, when the agency reverses for factual disputes, the Court has stated that “evidence supporting a conclusion may be less substantial when an impartial, experienced [ALJ] who has observed the witnesses and lived with the case has drawn conclusion different from the [agency’s].” Id.; see also Aylett v. Sec’y of Hous. & Urb. Dev., 54 F.3d 1560, 1561–62 (10th Cir. 1995) (reversing Department’s decision that housing discrimination occurred where ALJ had believed the landlord’s testimony). ↑
83. . Of course, the agency head can also influence the decisionmaking process of adjudicators through setting precedent for both substantive and procedural issues that ALJ must apply and follow when making their initial decisions. Agencies have long sought to minimize the discretion of adjudicators and hence increase the consistency of adjudicatory outcomes by promulgating rules or issuing precedential decisions. See Walker & Wasserman, supra note 48, at 188–93 (exploring how the Patent Office could standardize Patent Trial and Appeal Board decisions through substantive regulations or precedential rulings). See generally Christopher J. Walker, Melissa Wasserman & Matthew Lee Wiener, Precedential Decision Making in Agency Adjudication (2022) (discussing efforts to increase consistency through agencies issuing precedential decisions). ↑
84. . See Levin, supra note 46, at 412 (discussing how agency-head review brings ALJ decisions within the ambit of “accountability to the democratic process”). ↑
85. . Paul R. Verkuil, Daniel J. Gifford, Charles H. Koch, Jr., Richard J. Pierce, Jr. & Jeffrey S. Lubbers, The Federal Administrative Judiciary, in 2 Administrative Conference of the United States: Recommendations and Reports 779, 1034 (1992). ↑
86. . Id. at 911. ↑
87. . Id. at 821. ↑
88. . See Jerry L. Mashaw, Bureaucratic Justice: Managing Social Security Disability Claims 25–26 (1983) (explaining how a rational adjudication scheme that properly minimizes error requires “a supervisory determination of whether any adjudicative action taken correspond[s] to a true state of the world”); Robert A. Kagan, Inside Administrative Law, 84 Colum. L. Rev. 816, 820 (1984) (book review) (detailing how Mashaw’s “bureaucratic rationality” is a model of agency adjudication that facilitates “[g]reater control and consistency” by placing the “overriding value” on “accurate, efficient and consistent implementation of centrally-formulated policies”); see also Hoffer & Walker, supra note 44, at 276, 282, 286 (applying Mashaw’s notions of consistency, efficiency, and equity to the Tax Court and the Internal Revenue Service). ↑
89. . E.g., Hoffer & Walker, supra note 44, at 278. ↑
90. . See, e.g., Michael D. Frakes & Melissa F. Wasserman, Are There as Many Trademark Offices as Trademark Examiners?, 69 Duke L.J. 1807, 1822 (2020) (documenting heterogeneity in trademark office outcomes); Jaya Ramji-Nogales, Andrew I. Schoenholtz & Philip G. Schrag, Refugee Roulette: Disparities in Asylum Adjudication, 60 Stan. L. Rev. 295, 373 (2007) (documenting heterogeneity in asylum grant rates). ↑
91. . See Walker et al., supra note 83, at 28 (discussing the success of this practice for creating consistency in PTAB decisions). ↑
92. . Russell L. Weaver, Appellate Review in Executive Departments and Agencies, 48 Admin. L. Rev. 251, 289 (1996). ↑
93. . Michael D. Frakes & Melissa F. Wasserman, Patent Office Cohorts, 65 Duke L.J. 1601, 1605 (2016). ↑
94. . Id. at 1614–15. ↑
95. . Id. at 1605. ↑
96. . 15 U.S.C. § 45. ↑
97. . 47 U.S.C. § 155(c). ↑
98. . 19 U.S.C. § 1337. ↑
99. . 15 U.S.C. § 78d-1. ↑
100. . ALJs by Agency, supra note 6 (providing an agency-by-agency breakdown of ALJs). ↑
101. . 5 U.S.C. § 557(b); Levy & Glicksman, supra note 28, at 50–53. ↑
102. . 5 U.S.C. § 704; Darby v. Cisneros, 509 U.S. 137, 152 (1993); 5 U.S.C. § 557(c). ↑
103. . See, e.g., Levy & Glicksman, supra note 28, at 53–54 (introducing the growing threats to ALJ independence); Bremer, supra note 57, at 1776–81 (discussing how recent Supreme Court opinions have placed the ALJ administrative structure in peril). ↑
104. . See Bremer, supra note 57, at 24–28 (noting the erosion of the current adjudicatory scheme). ↑
105. . See supra notes 73–77 and accompanying text. ↑
106. . Some experts estimate that as much as 90 percent of all agency adjudication occurs outside of APA formal adjudication proceedings. Am. Bar Ass’n, A Guide to Federal Agency Adjudication 176 (Jeffrey B. Litwak ed., 2d ed. 2012) (citing Paul R. Verkuil, A Study of Informal Adjudication Procedure, 43 U. Chi. L. Rev. 739, 741 (1976)). Subpart I(A) draws substantially from Walker & Wasserman, supra note 48, at 153–57, 162–73. ↑
107. . Daniel A. Farber & Anne Joseph O’Connell, The Lost World of Administrative Law, 92 Texas L. Rev. 1137, 1140 (2014). ↑
108. . See generally Michael Asimow, Admin. Conf. of the U.S., Adjudication Outside the Administrative Procedure Act 2 (2019) (describing the various types of administrative adjudication and their consequences). As Michael Asimow explains, distinguishing formal-like adjudications from both APA-governed formal adjudication and informal adjudications is not an exact science. See id. at 7, 9–11 (describing a blurred line between informal adjudications and formal rulemakings). ↑
109. . Exec. Off. of Immigr. Rev., Adjudication Statistics: Immigration Judge (IR) Hiring (2024), https://www.justice.gov/eoir/media/1344911/dl?inline [https://perma.cc/RD8K-7YNS]. ↑
110. . John Gramlich, Most of Biden’s Appointed Judges to Date Are Women, Racial or Ethnic Minorities—A First for Any President, Pew Rsch. Ctr. (Dec. 4, 2023), https://www.pewresearch
     .org/short-reads/2023/12/04/most-of-bidens-appointed-judges-to-date-are-women-racial-or-ethnic-minorities-a-first-for-any-president/ [https://perma.cc/6LK7-S5NU]. ↑
111. . Exec. Off. for Immigr. Rev., Adjudication Statistics: New Cases and Total Completions (2024), https://www.justice.gov/d9/pages/attachments/2018/05/08/2_new_cases_and_total_completions.pdf [https://perma.cc/4WD3-VVHT]. ↑
112. . Id. ↑
113. . Elsewhere, two of us explore in greater detail another Type B adjudication scheme: the Patent Trial and Appeal Board at the U.S. Patent and Trademark Office. See generally Walker & Wasserman, supra note 48, at 157–73 (comparing the Type B adjudication scheme in patent litigation to the Type B adjudication scheme in immigration litigation). ↑
114. . See, e.g., Matthew Bultman & Ian Lopez, Big Bonuses for Patent Appeals Judges Raise Fairness Questions, Bloomberg L. (Sept. 8, 2020, 5:31 AM), https://news.bloomberglaw.com/ip-law/big-bonuses-for-patent-appeals-judges-raise-fairness-questions [https://perma.cc/F2AF-9CYR] (explaining how in patent courts, the bonus system incentivizes higher “production” of cases at the expense of adjudicatory decision quality). ↑
115. . See Michael D. Frakes & Melissa F. Wasserman, Deadlines Versus Continuous Incentives: Evidence from the Patent Office 30 (Nat’l Bureau of Econ. Rsch., Working Paper No. 32066, 2024), https://www.nber.org/papers/w32066 [https://perma.cc/29S3-4N47] (finding the addition of a daily examination-pendency measure to patent examiners performance reviews resulted in substantial reduction in deadline effects and near complete temporal smoothing in examiner behavior). ↑
116. . See, e.g., Jill Family, We Have Nothing to Fear but “Sovereignty Fear” Itself, Yale J. on Regul.: Notice & Comment (Aug. 5, 2021), https://www.yalejreg.com/nc/we-have-nothing-to-fear-but-sovereignty-fear-itself/ [https://perma.cc/CB4V-ZEW7] (“Immigration judges felt pressure to take the fast route and deny a case or deny a continuance, rather than to take the longer time necessary to approve a case or to slow down to make sure a hearing is fair.”). ↑
117. . See, e.g., Mobility Workx, LLC. v. United Patents, LLC, 15 F.4th 1146, 1156 (Fed. Cir. 2021) (“Mobility has therefore failed to establish that [APJs] have an unconstitutional financial interest in instituting AIA proceedings”). ↑
118. . See generally Walker & Wasserman, supra note 48, at 178–88 (describing the practice of panel stacking at the PTO). ↑
119. . Id. at 178–80. ↑
120. . See 5 U.S.C. § 3105 (“Administrative law judges shall be assigned to cases in rotation so far as practicable . . . . ”). ↑
121. . Verkuil, supra note 64, at 1347 (“The selection and appointments procedures for administrative judges are controlled by the agencies themselves.”). ↑
122. . See Act of Jan. 16, 1883, ch. 27, 22 Stat. 403–07 (laying out early requirements for the civil service and not mentioning qualifications). ↑
123. . U.S. Dep’t of Just., An Investigation of Allegations of Politicized Hiring by Monica Goodling and Other Staff in the Office of the Attorney General 115 (2008), https://oig.justice.gov/sites/default/files/archive/special/s0807/chapter6.htm [https://perma.cc
     /9BFS-RKGG]. An immigration judge (IJ) is “an attorney whom the Attorney General appoints as an administrative judge within the Executive Office for Immigration Review.” Id. at 70 (quoting 8 U.S.C. § 1101(b)(4)). OPM has categorized career attorney positions as Schedule A. U.S. Dep’t of Just., supra, at 70; 5 C.F.R. §§ 213.3101, 213.3102(d) (2024). IJs are career Schedule A appointees, such that the civil service laws there apply. U.S. Dep’t of Justice, supra, at 70. ↑
124. . Susan Benesch, Due Process and Decisionmaking in U.S. Immigration Adjudication, 59 Admin. L. Rev. 557, 566 (2007). ↑
125. . See, e.g., Aaron L. Nielson, Three Views of the Administrative State: Lessons from Collins v. Yellen, 2021 Cato Sup. Ct. Rev. 141, 162 (explaining how the unitary executive theory allows more aggressive use of agency power); see also Ronald Krotoszynski, The Conservative Idea That Would Let Biden Seize Control of Washington, Politico (Dec. 10, 2020, 6:00 PM), https://www.politico.com/news/magazine/2020/12/10/nathan-simington-christopher-waller-fcc-federal-reserve-appointments-unitary-executive-authority-444136 [https://perma.cc/X5GS-PF2A] (explaining how the unitary executive theory allows the president to remove Executive branch personnel opposed to his administration’s policies). ↑
126. . Free Enter. Fund v. Pub. Co. Acct. Oversight Bd., 561 U.S. 477, 484, 486 (2010) (quoting 15 U.S.C. § 7211(e)(6), invalidated by Free Enter. Fund, 561 U.S. at 514). ↑
127. . Id. at 487. ↑
128. . 295 U.S. 602 (1935). ↑
129. . 487 U.S. 654 (1988). ↑
130. . Free Enter. Fund v. Pub. Co. Acct. Oversight Bd., 537 F.3d 667, 669, 679 (D.C. Cir. 2008), rev’d, 561 U.S. 477 (2010). But see id. at 685–86 (Kavanaugh, J., dissenting) (disagreeing with the majority’s application of the two precedents). ↑
131. . Free Enter. Fund, 561 U.S. at 496, 499. ↑
132. . See, e.g., Rao, supra note 22, at 2549 (challenging scholars who concluded that Free Enterprise Fund did not amount to a major change in the law). ↑
133. . See Free Enter. Fund, 561 U.S. at 537–41 (Breyer, J., dissenting) (arguing that the court’s holding would bar removal restrictions on many federal agencies, specifically those limited by statute to removal for cause). As Justice Breyer noted in dissent, no statute expressly provides SEC Commissioners with any tenure protection and the Supreme Court has never held that some statute implicitly provides such protection. Id. at 546. Following Collins, the argument against such an implied tenure protection has become stronger because the SEC unquestionably exercises regulatory authority. See Collins v. Yellen, 141 S. Ct. 1761, 1782, 1783 n.18 (2021) (explaining that “[w]hen a statute does not limit the President’s power to remove an agency head, we generally presume that the officer serves at the President’s pleasure,” especially if an entity “is not an adjudicatory body”). ↑
134. . Free Enter. Fund, 561 U.S. at 507 n.10 (citations omitted) (quoting Free Enter. Fund, 537 F.3d at 699 n.8 (Kavanaugh, J., dissenting)). ↑
135. . 138 S. Ct. 2044 (2018). ↑
136. . Id. at 2055. ↑
137. . Id. at 2053. ↑
138. . Id. at 2051 (quoting Buckley v. Valeo, 424 U.S. 1, 126 (1976)); see also id. at 2053 (applying this “significant authority” requirement). ↑
139. . Lucia, 138 S. Ct. at 2060 (Breyer, J., concurring in the judgment in part and dissenting in part). ↑
140. . Exec. Order No. 13,843, § 1, 3 C.F.R. 844, 844–45 (2019). ↑
141. . See Levy & Glicksman, supra note 28, at 59–68 (discussing the order, its legality, and its possible consequences). ↑
142. . See, e.g., Editorial, Trump Is Politicizing the Federal Government Even Further. Step In, Congress, Wash. Post (July 22, 2018, 7:07 PM), https://www.washingtonpost.com
     /opinions/trump-is-politicizing-the-federal-government-even-further-step-in-congress/2018/07
     /22/eb4ce8ee-8ac1-11e8-8aea-86e88ae760d8_story.html [https://perma.cc/853V-YQUX] (“One can envisage an anti-welfare Social Security chief selecting an ALJS who is skeptical of benefits claims.”); Jeffrey S. Lubbers, The Regulatory Accountability Act Loses Steam but the Trump Executive Order on ALJ Selection Upturned 71 Years of Practice, 94 Chi.-Kent L. Rev. 741, 747–48 (2019) (“Trump needlessly swung the pendulum too far and opened up the program, to potential cronyism of political favoritism in the hiring of new ALJs.”). ↑
143. . Exec. Order No. 13,843, § 1, 3 C.F.R. at 844. ↑
144. . Id. at 844–45. ↑
145. . 140 S. Ct. 2183 (2020). ↑
146. . 141 S. Ct. 1761 (2021). ↑
147. . Seila Law, 140 S. Ct. at 2191. ↑
148. . Humphrey’s Ex’r v. United States, 295 U.S. 602, 631–32 (1935). ↑
149. . Morrison v. Olson, 487 U.S. 654, 696–97 (1988). ↑
150. . See Seila Law, 140 S. Ct. at 2192 (declining to extend the reasoning from these cases to the CFPB as the agency advocated). ↑
151. . Id. ↑
152. . Id. at 2199–2200. ↑
153. . See id. at 2198 n.2 (“The Court’s conclusion [in Humphrey’s Executor] that the FTC did not exercise executive power has not withstood the test of time.”); Daniel Crane, Debunking Humphrey’s Executor, 83 Geo. Wash. L. Rev. 1835, 1838 (2015) (arguing that the FTC does not match Humphrey’s Executor’s narrow description). ↑
154. . See Seila Law, 140 S. Ct. at 2240 (Kagan, J., dissenting) (“[T]he majority’s ‘exceptions’ (like its general rule) are made up.”). In rejecting the majority’s reasoning, Justice Kagan wrote:[T]he Court’s precedents before today have accepted the role of independent agencies in our governmental system. To be sure, the line of our decisions has not run altogether straight. But we have repeatedly upheld provisions that prevent the President from firing regulatory officials except for such matters as neglect or malfeasance . . . . Nowhere do [the Court’s removal] precedents suggest what the majority announces today: that the President has an ‘unrestricted removal power’ subject to two bounded exceptions.Id. at 2233. ↑
155. . 141 S. Ct. 1761, 1784 (2021); id. at 1799–1800 (Kagan J., concurring in part and concurring in the judgment in part). ↑
156. . See id. at 1785 (“[T]he nature and breadth of an agency’s authority is not dispositive in determining whether Congress may limit the President’s power to remove its head.”); id. at 1787 (“[A]s we explained last Term [in Seila Law], the Constitution prohibits even ‘modest restrictions’ on the President’s power to remove the head of an agency with a single top officer.” (quoting Seila Law, 140 S. Ct. at 2205)). ↑
157. . Id. at 1809 (Sotomayor, J., concurring in part and dissenting in part). ↑
158. . 141 S. Ct. 1970 (2021). ↑
159. . Id. at 1982, 1985. ↑
160. . Id. at 1986–87. ↑
161. . Arthrex, Inc. v. Smith & Nephew, Inc., 941 F.3d 1320, 1338 (Fed. Cir. 2019), vacated sub nom. United States v. Arthrex, 141 S. Ct. 1970 (2021) (labeling this the “narrowest remedy”). ↑
162. . Arthrex, 141 S. Ct. at 1985. ↑
163. . Id. at 1987. ↑
164. . Id. at 1984 (citing Walker & Wasserman, supra note 48, at 157). ↑
165. . Jarkesy v. SEC, 34 F.4th 446, 463 (5th Cir. 2022), aff’d on other grounds, 144 S. Ct. 2117, 2139 (2024); see Free Enter. Fund v. Pub. Co. Acct. Oversight Bd., 561 U.S. 477, 542–43 (2010) (Breyer, J., dissenting) (arguing that administrative law judges were at risk because of the majority); see also supra notes 13–20 and accompanying text (discussing Jarkesy). ↑
166. . Jarkesy, 34 F.4th at 465. ↑
167. . Jarkesy, 34 F.4th at 478 (Davis, J., dissenting); Jarkesy v. SEC, 51 F.4th 644, 645–47 (5th Cir. 2022) (Haynes, J., dissenting from denial of rehearing en banc); see also Jason Willick, What Might Conservative Legal Minds Go After Next?, Wash. Post. (May 23, 2022) (arguing that Jarkesy is part of a broader effort to put “unaccountable administrative agencies in the crosshairs”). See generally Beermann, supra note 20 (explaining constitutional attacks on the administrative state). ↑
168. . See supra sources cited in note 21. In fact, Judge Neomi Rao on the D.C. Circuit reached the same conclusion in a dissent in Fleming v. United States Department of Agriculture, 987 F.3d 1093, 1104, 1114–15 (D.C. Cir. 2021) (Rao, J., concurring in part and dissenting in part). The majority in Fleming declined to decide the issue, holding that petitioners had waived the constitutional challenge by not first raising it before the agency. Id. at 1097, 1104. ↑
169. . Transcript of Oral Argument, SEC v. Jarkesy, 144 S. Ct. 2117 (2024) (No. 22-859). ↑
170. . Jarkesy, 144 S. Ct. at 2139. ↑
171. . See id. (“We do not reach the remaining constitutional issues and affirm the ruling of the Fifth Circuit on the Seventh Amendment ground alone.”). ↑
172. . Space Expl. Techs. Corp. v. NLRB, Civil No. W-24-CV-00203-ADA, 2024 WL 3512082, at *3 (W.D. Tex. July 23, 2024) (“The Court finds that under current Fifth Circuit law, there is a substantial likelihood that SpaceX succeeds on the merits with regards to showing that the NLRB ALJs are unconstitutionally protected from removal.”); see also id. at *2 n.1 (“SCOTUS did not voice disagreement with, much less overrule, the Fifth Circuit’s holdings with regards to Issue Three. Accordingly, this Court remains bound by Fifth Circuit precedent in this case.” (citing Jarkesy, 114 S. Ct. at 2139)). ↑
173. . See Jarkesy v. SEC, slip op. at 1–2, No. 20-61007 (5th Cir. Nov. 12, 2024) (per curiam) (“[W]e reiterate our prior holdings in this case.”). ↑
174. . See Charlie Savage, Trump Claims Power to Fire Administrative Law Judges at Will, N.Y. Times (Feb. 20, 2025), https://www.nytimes.com/2025/02/20/us/politics/trump-power-administrative-law-judges.html [https://perma.cc/2XQK-ZRFK]. ↑
175. . Brief for Petitioner at 66, Jarkesy, 144 S. Ct. 2117 (No. 22–859). ↑
176. . In Carr v. Saul, the Court held that, at least where there is no statute or regulation that requires administrative exhaustion, litigants do not need to administratively exhaust constitutional challenges to how ALJs are appointed. 141 S. Ct. 1352, 1360, 1362 (2021). And in Axon Enterprise, Inc. v. FTC, the Court held that removal-power challengers to the FTC’s and SEC’s structure need not first proceed through the administrative process but can immediately seek judicial review in federal district court. 143 S. Ct. 890, 897 (2023). ↑
177. . See, e.g., Saikrishna Prakash, New Light on the Decision of 1789, 91 Cornell L. Rev. 1021, 1034, 1042 (2006) (outlining the debate and arguing that the decision supports presidential removal); see also Jed H. Shugerman, The Indecisions of 1789: Inconsistent Originalism and Strategic Ambiguity, 171 U. Pa. L. Rev 753, 757 (2023) (challenging whether the “Decision of 1789” supports a removal power); John F. Manning, Separation of Powers as Ordinary Interpretation, 124 Harv. L. Rev. 1939, 2030–32 (2011) (similar). ↑
178. . The ever-growing literature is too expansive to catalog here. Suffice it to say, many important works have been published in recent years touching on the subject. E.g., Michael W. McConnell, The President Who Would Not Be King 162–63 (2020); Jed Handelsman Shugerman, Vesting, 74 Stan. L. Rev. 1479, 1481 (2022); Daniel D. Birk, Interrogating the Historical Basis for a Unitary Executive, 73 Stan. L. Rev. 175, 177 (2021); Julian Davis Mortenson, The Executive Power Clause, 168 U. Pa. L. Rev. 1269, 1273 (2020); Christine Kexel Chabot, Is the Federal Reserve Constitutional? An Originalist Argument for Independent Agencies, 96 Notre Dame L. Rev. 1, 15 (2020); Ganesh Sitaraman, The Political Economy of the Removal Power, 134 Harv. L. Rev. 352, 353 (2020); Ilan Wurman, The Removal Power: A Critical Guide, 2020 Cato Sup. Ct. Rev. 157, 157; Andrew Kent, Ethan J. Leib & Jed Handelsman Shugerman, Faithful Execution and Article II, 132 Harv. L. Rev. 2111, 2189–90 (2019); Aditya Bamzai, Tenure of Office and the Treasury: The Constitution and Control over National Financial Policy, 1787 to 1867, 87 Geo. Wash. L. Rev. 1299, 1304 (2019); Jack Goldsmith & John F. Manning, The Protean Take Care Clause, 164 U. Pa. L. Rev. 1835, 1858 (2016); David E. Lewis & Jennifer L. Selin, Political Control and the Forms of Agency Independence, 83 Geo. Wash. L. Rev. 1487, 1490 (2015); Kirti Datla & Richard L. Revesz, Deconstructing Independent Agencies (and Executive Agencies), 98 Cornell L. Rev. 769, 772 (2013); Aziz Z. Huq, Removal as a Political Question, 65 Stan. L. Rev. 1, 3 (2013); see also Jane Manners & Lev Menand, The Three Permissions: Presidential Removal and the Statutory Limits of Agency Independence, 121 Colum. L. Rev. 1, 28 (2021) (discussing how term-of-year provisions relate to the debate); Brian D. Feinstein & Daniel J. Hemel, Partisan Balance with Bite, 118 Colum. L. Rev. 9, 11 (2018) (discussing whether partisan balance impacts executive decisionmaking); Patrick M. Corrigan & Richard L. Revesz, The Genesis of Independent Agencies, 92 N.Y.U. L. Rev. 637, 639–40 (2017) (discussing historical conditions for the creation of independent agencies); Adrian Vermeule, Conventions of Agency Independence, 113 Colum. L. Rev. 1163, 1174–75 (2013) (discussing whether political norms in independent agencies match the intent of the underlying judicial doctrine). ↑
179. . Seila Law LLC v. CFPB, 140 S. Ct. 2183, 2199–2200 (2020); see Aditya Bamzai & Aaron L. Nielson, Article II and the Federal Reserve, 109 Cornell L. Rev. 843, 892 (2024) (explaining the breadth of the Court’s reasoning). ↑
180. . See, e.g., Levy & Glicksman, supra note 28, at 44–45 (arguing that Congress should enact a federal central panel agency). ↑
181. . See supra note 121 and accompanying text. ↑
182. . See supra notes 13–20 and accompanying text. ↑
183. . See Richard E. Levy & Robert L. Glicksman, Toward a Federal Central Panel for Administrative Adjudication, Yale J. on Regul.: Notice & Comment (Feb. 14, 2022), https://www.yalejreg.com/nc/symposium-decisional-independence-01/ [https://perma.cc/5VCU-4AFR] (arguing that a properly designed central panel would allow agencies to retain final decisionmaking authority through de novo review of central panel decisions). ↑
184. . See Ronald M. Levin, Doubts About a Federal Central Panel, Yale J. on Regul.: Notice & Comment (Feb. 17, 2022), https://www.yalejreg.com/nc/symposium-decisional-independence-04/ [https://perma.cc/B9X6-FGWT] (discussing an ABA resolution, which NCALJ introduced and then withdrew from consideration at the ABA House of Delegates’ August 2021 session). ↑
185. . Am. Bar Ass’n, Resolution 200, at 1 (2022) [hereinafter ABA Resolution 200], https://www.americanbar.org/content/dam/aba/directories/policy/annual-2022/200-annual-2022.pdf [https://perma.cc/9EYP-RLXY]. ↑
186. . Levy & Glicksman, supra note 183. ↑
187. . The following draws heavily from the ABA Section of Administrative Law and Regulatory Practice’s unpublished Statement of Opposition to Resolution 201, which one of us (Walker) as Section Chair and Ron Levin as Section Delegate to the ABA House of Delegates drafted in June 2021. Professor Levin has kindly agreed for our joint work to be incorporated herein. ↑
188. . We are sympathetic that it is difficult to conduct high-quality empirical studies on the extent to which decisional independence erodes as a function of removal protections or as a function of the agency head ability to freely hire adjudicators. Perhaps one of the most challenging aspects of work of this nature is identifying a counterfactual. In this context there are policy interventions that create counterfactuals: the Trump Administration’s executive order that gives agency heads greater latitude to hire ALJs (albeit while also emphasizing the “policy” that ALJs remain “impartial and committed to the rule of law”), and the growth of the formal-like adjudication that falls outside the APA. At the very least, qualitative work could be done to survey and interview the key stakeholders—e.g., agency adjudicators, agency leadership, and the adjudicated—to attempt to understand their perceptions of the problem. ↑
189. . See Christopher J. Walker & Matthew Lee Wiener, Agency Appellate Systems 2 (2020), https://www.acus.gov/report/final-report-agency-appellate-systems [https://perma.cc
     /SGP9-Q4RW] (introducing report on how agency appellate bodies should be structured to achieve the most optimal results). ↑
190. . ABA Resolution 200, supra note 185, at 1. ↑
191. . ALJs by Agency, supra note 6. ↑
192. . Hearings and Appeals, Soc. Sec. Admin., https://www.ssa.gov/appeals [https://perma
     .cc/3UHK-C6LC]. ↑
193. . Levin & Glicksman, supra note 183. ↑
194. . 35 U.S.C. § 6. ↑
195. . 38 U.S.C. § 7107. ↑
196. . See, e.g., Gerald K. Ray & Jeffrey S. Lubbers, A Government Success Story: How Data Analysis by the Social Security Appeals Council (with a Push from the Administrative Conference of the United States) Is Transforming Social Security Disability Adjudication, 83 Geo. Wash. L. Rev. 1575, 1585–88, 1601–04 (2015) (detailing the history of ACUS recommendations for SSA adjudication). ↑
197. . Although the range of cases that the federal central panel would handle remains unspecified, one can reasonably assume that its caseload would be much larger than that of any state central panel, on which the proposal is based. Thus, the apparently successful track record of the state central panels does not necessarily foretell equally good results in a panel with nationwide scope, overseeing federal programs that in many cases raise remarkably complex substantive issues. The effectiveness of the state central panel model, moreover, should be subject to more rigorous empirical assessment than has been done to date. Cf. Malcolm C. Rich & Alison C. Goldstein, The Need for a Central Panel Approach to Administrative Adjudication: Pros, Cons, and Selected Practices, J. Nat’l Ass’n Admin. L. Judiciary, Spring 2019, at 1, 51–67 (reporting results from a survey of only twenty-three state central panel directors). ↑
198. . Rich & Goldstein, supra note 197, at 60–61. ↑
199. . See Jeffrey S. Lubbers, A Unified Corps of ALJs: A Proposal to Test the Idea at the Federal Level, 65 Judicature 266, 275 (1981). As Professor Lubbers explains:The agency’s reviewing function might be altered in unforeseen ways. Some proponents argue that establishment of a corps should be linked to a restriction of the agency’s ability to review initial decisions of ALJs. The wisdom in this is debatable, but without such a change agencies likely would feel the need to review more initial decisions more intensively (in light of their reduced rapport or familiarity with the judges), leading to an overall lengthening of the decisional process.Id. (footnote omitted). ↑
200. . The SSA has over a million cases awaiting review, and the PTO has repeatedly stated its backlog of patent applications is its biggest challenge. U.S. Patent & Trademark Office, 2008 Performance and Accountability Report 33, https://www.uspto.gov/sites/default/files/about
     /stratplan/ar/USPTOFY2008PAR.pdf [https://perma.cc/TJX3-7VTE] (“The Patent organization’s biggest challenge is to address the growth of pendency and the backlog of patent applications waiting to be examined while maintaining high quality.”); One Million Claims and Growing: Improving Social Security’s Disability Adjudication Process Before the Subcomm. on Soc. Sec. of the H. Comm. on Ways and Means, 118th Cong. 1 (2023) (statement of Rep. Drew Ferguson, Chairman, H. Subcomm. on Soc. Sec.) (detailing how the SSA has over one million disability claims awaiting review and that disability benefits adjudication is one of the agency’s biggest challenges). ↑
201. . See, e.g., Ass’n of Admin. L. Judges v. Colvin, 777 F.3d 402, 404–05 (7th Cir. 2015) (“The aim of the quota is to speed up decision-making rather than to prod administrative law judges to grant more applications for disability benefits.”); Sannier v. Merit Sys. Prot. Bd., 931 F.2d 856, 858 (Fed. Cir. 1991) (“[A]s long as the agency actions are not alleged to affect the ability of the ALJ to function as an independent and impartial decisionmaker . . . .”); Nash v. Bowen, 869 F.2d 675, 680–81 (2d Cir. 1989) (same); cf. Abrams v. Soc. Sec. Admin., 703 F.3d 538, 538, 540–42 (Fed. Cir. 2012) (upholding discipline of judge for failing to comply with instructions related to productivity). ↑
202. . Levin, supra note 184. ↑
203. . See Exec. Order No. 12,866, § 1(b)(5), (7), 3 C.F.R. 638, 638–39 (1994) (setting out a regulatory policy centered on efficiency and cost–benefit analysis); Todd S. Aagaard, Agencies, Courts, First Principles, and the Rule of Law, 70 Admin. L. Rev. 771, 772, 777 (2018) (contemplating the proper balance between the “specialization and routinization” found at agencies and “Rule of Law” values of “fairness, privacy, due process, and liberty”). ↑
204. . See Walker & Wasserman, supra note 48, at 175–78 (discussing objectives and defending the vesting of “final-decision-making authority with the agency head”). ↑
205. . Free Enter. Fund v. Pub. Co. Acct. Oversight Bd., 561 U.S. 477, 514 (2010). ↑
206. . See supra note 179 and accompanying text. ↑
207. . See Seila Law LLC v. CFPB, 140 S. Ct. 2183, 2199–2200 (2020) (holding that the CFPB director is removable partly because of the position’s “administrative authority”); see also Collins v. Yellen, 141 S. Ct. 1761, 1784 (2021) (holding similarly for the FHFA director). ↑
208. . Kuretski v. Comm’r, 755 F.3d 929, 940 (D.C. Cir. 2014). ↑
209. . Real Courts, Rule of Law Act of 2022, H.R. 6577, 117th Cong. § 2(a) (2022). ↑
210. . Id. ↑
211. . See supra section II(A)(2). ↑
212. . See United States v. Mead Corp., 533 U.S. 218, 226–27 (2001) (presumptively affording deference to agency interpretations announced in some adjudications). Chevron deference is grounded in part in principles of political accountability—principles that may already be weaker for adjudication. Chevron U.S.A. Inc. v. Nat. Res. Def. Council, Inc., 467 U.S. 837, 865–66 (1984), overruled by Loper Bright Enters. v. Raimondo, 144 S. Ct. 2244, 2273 (2024); see, e.g., Kristin E. Hickman & Aaron L. Nielson, Narrowing Chevron’s Domain, 70 Duke L.J. 931, 967–68 (2021) (describing the political rationale for deference and explaining that accountability is already weaker in the adjudication context); Shoba Sivaprasad Wadhia & Christopher J. Walker, The Case Against Chevron Deference in Immigration Adjudication, 70 Duke L.J. 1197, 1203 (2021) (advocating for agency rulemaking over adjudication in immigration policymaking). ↑
213. . See, e.g., SEC v. Chenery Corp. (Chenery II), 332 U.S. 194, 207 (1947) (allowing agencies to announce and apply new policies in adjudication). To be sure, there are strong reasons to be wary of such policymaking, especially retroactively, Hickman & Nielson, supra note 212, at 972, but it is a longstanding feature of the legal system and is not always (especially) controversial, especially where the agency has provided fair notice of its interpretation. Id. ↑
214. . See, e.g., Paul R. Verkuil & Jeffrey S. Lubbers, Alternative Approaches to Judicial Review of Social Security Disability Cases, 55 Admin. L. Rev. 731, 752 (2003) (introducing arguments that Congress should consider creating an Article I Social Security Court). ↑
215. . E.g., U.S. Tax Ct. R. Prac. & P. 24(b). ↑
216. . There are many examples of such judges. The Tax Court is one example:The Tax Court’s status as an “Article I legislative court” does not mean that its judges exercise “legislative power” under Article I. The Tax Court is in the business of interpreting and applying the internal revenue laws, not in the business of making those laws. And the Tax Court’s Article I origins do not distinguish it from the mine run of Executive Branch agencies whose officers may be removed by the President. After all, every Executive Branch entity, from the Postal Service to the Patent Office, is established pursuant to Article I. The Tax Court no more exercises Article I powers than do those agencies.Kuretski v. Comm’r, 755 F.3d 929, 943 (D.C. Cir. 2014) (citations omitted). ↑
217. . Id. ↑
218. . Id. at 944. ↑
219. . Seila Law LLC v. CFPB, 140 S. Ct. 2183, 2200 (2020); Collins v. Yellen, 141 S. Ct. 1761, 1793 (2021). ↑
220. . United States v. Arthrex, 141 S. Ct. 1970, 1983, 1986 (2021). ↑
221. . Seila Law, 140 S. Ct. at 2199–2201. ↑
222. . Id. at 2201–02. ↑
223. . Myers v. United States, 272 U.S. 52 (1926). ↑
224. . Free Enter. Fund v. Pub. Co. Acct. Oversight Bd., 561 U.S. 477, 492–93, 518 (2010); Seila Law, 140 S. Ct. at 2197–99. ↑
225. . Myers, 272 U.S. at 129. ↑
226. . Id. at 135. ↑
227. . Id. ↑
228. . Collins v. Yellen, 141 S. Ct. 1761, 1787 (2020) (quoting Myers, 272 U.S. at 135). ↑
229. . Cf. Barnett, Regulating Impartiality, supra note 29, at 1719 (“I cannot say with any certainty whether the Supreme Court will provide ALJs and other agency adjudicators a carve-out . . . . It may well be that the Court would find a functional exception to its formalist jurisprudence that considers adjudicators’ functions within the executive branch and the competing due process values.”). ↑
230. . See, e.g., Stone Washington & Ryan Young, Conflict of Justice: Making the Case for Administrative Law Court Reform 7 (2023), https://cei.org/studies/conflict-of-justice/ [https://perma.cc/YG2G-ZNV3] (“Congress needs to reform ALCs, either by moving them to the judicial branch, or by funding a magistrate-style system where circuit courts oversee ALC-like bodies.”); Michael B. Rappaport, Classical Liberal Administrative Law in a Progressive World, in The Cambridge Handbook of Classical Liberal Thought 105, 107 (M. Todd Henderson ed., 2018) (explaining one possible method of institutional reform would be the “establishment of independent administrative courts” that would “ideally” be Article III courts); see also Memorandum from Steven G. Calabresi, Professor, Nw. Univ. Pritzker Sch. of L. & Shams Hirji, Nw. Univ. Pritzker Sch. of L., to the Senate and the House of Representatives on Proposed Judgeship Bill 31 (Nov. 7, 2017), https://archive.thinkprogress.org/uploads/2017/11/calabresi-court-packing-memo.pdf [https://perma.cc/S4Y7-8CDP] (arguing for 158 ALJs who impose civil monetary penalties to be replaced with Article III judges). ↑
231. . See supra subpart I(B). ↑
232. . See ALJs by Agency, supra note 6 (calculating that there are more than 1,900 administrative law judges); Barnett & Wheeler, supra note 7, at 32 (indicating that there are more than 10,000 non-ALJ agency adjudicators). ↑
233. . U.S. Const. art. III, § 2. ↑
234. . CFTC v. Schor, 478 U.S. 833, 859–60 (1986) (Brennan, J., dissenting) (quoting The Federalist No. 47, at 334 (James Madison) (Henry B. Dawson ed., 1876)). Chief Justice Roberts has made a similar argument:The Court today declines to resist encroachment by the Legislature. Instead it holds that a single federal judge, for reasons adequate to him, may assign away our hard-won constitutional birthright so long as two private parties agree. I hope I will be wrong about the consequences of this decision for the independence of the Judicial Branch.Wellness Int’l Network, Ltd. v. Sharif, 575 U.S. 665, 705 (2015) (Roberts, C.J., dissenting). ↑
235. . SEC v. Jarkesy, 144 S. Ct. 2117, 2139 (2024) (holding that some adjudication must be heard in Article III courts because of the jury-trial right). ↑
236. . 2. U.S. (2 Dall.) 409 (1792). ↑
237. . Id. at 411; see Fallon et al., supra note 81, at 82–94 (explaining Hayburn’s Case and its importance). ↑
238. . Sims v. Apfel, 530 U.S. 103, 110–11 (2000) (alterations in original) (citations omitted) (first quoting 2 Kenneth Culp Davis & Richard J. Pierce, Jr., Administrative Law Treatise § 9.10, at 103 (3d ed. 1994); and then quoting Bernard Schwartz, Administrative Law 469–70 (4th ed. 1994)) (first citing Schwartz, supra, at 470; and then citing Richardson v. Perales, 402 U.S. 389, 400–01 (1971)). ↑
239. . Federal Court Review Process, Soc. Sec. Admin., https://www.ssa.gov/appeals
     /court_process.html [https://perma.cc/Q9JJ-3CRM]. ↑
240. . See, e.g., Sims, 530 U.S. at 112 (distinguishing between procedures before the SSA from procedures challenging SSA decisions in federal court). ↑
241. . U.S. Const. art. II., § 2, cl. 1. ↑
242. . For example:[T]he President’s Article II authority as Commander-in-Chief and the text of the Fifth Amendment—which expressly exempts from the Grand Jury Indictment Clause ‘cases arising in the land or naval forces, or in the Militia, when in actual service in time of War or public danger’—[have been cited as] key ingredients to the constitutionality of adjudication by non-Article III federal military courts . . . .Stephen I. Vladeck, Military Courts and Article III, 103 Geo. L.J. 933, 952 (2015) (footnote omitted). To be sure, merely because some adjudication can be done outside of Article III courts does not necessarily mean that it must be so. The President’s Article II authority, however, may require some matters to be heard by those in the Executive Branch. We do not purport to answer this question but merely note that eliminating agency adjudication may raise constitutional concerns in certain categories of disputes. ↑
243. . Washington & Young, supra note 230, at 7. ↑
244. . Calabresi & Hirji, supra note 230, at 31–32. ↑
245. . Id. at 32. Elsewhere, one of us argues that the better solution at the SEC would be to give the regulated party the right to remove certain adjudications to federal court. See Christopher J. Walker & David Zaring, The Right to Remove in Agency Adjudication, 85 Ohio St. L.J. 1, 3 (2024) (arguing that, in some circumstances, a regulated party should be able to remove an enforcement action from agency adjudication to Article III federal court). ↑
246. . Humphrey’s Ex’r v. United States, 295 U.S. 602, 629 (1935). ↑
247. . 357 U.S. 349, 356 (1958). ↑
248. . See, e.g., Free Enter. Fund v. Pub. Co. Acct. Oversight Bd., 561 U.S. 477, 496 (2010) (refusing to expand restrictions on removal); Seila Law LLC v. CFPB, 140 S. Ct. 2183, 2192 (2020) (same). ↑
249. . See supra subpart II(B); Seila Law, 140 S. Ct. at 2192 (“Such an agency lacks a foundation in historical practice and clashes with constitutional structure by concentrating power in a unilateral actor insulated from Presidential control.”). ↑
250. . Nielson & Walker, supra note 36, at 39–40. ↑
251. . The Federalist No. 76, at 457–58 (Alexander Hamilton) (Clinton Rossiter ed., 1961). ↑
252. . The Federalist No. 77, supra note 251, at 459 (Alexander Hamilton). ↑
253. . See 1 Annals of Cong. 517 (1789) (Joseph Gales ed., 1834) (statement of Representative James Madison) (remarking that a rational president would hesitate to “displace from office a man whose merits require that he should be continued in it,” because “the wanton removal of meritorious officers would subject him to impeachment and removal from his own high trust”); id. at 518 (arguing that the prospect of political reprisal for removal “will excite serious reflections beforehand in the mind of any man who may fill the Presidential chair”). ↑
254. . Nielson & Walker, supra note 36, at 68. ↑
255. . See id. at 51 (explaining the various ways that “Congress can discourage the President from removing an officeholder”). ↑
256. . See generally Jarrod Shobe, Enacted Legislative Findings and Purposes, 86 U. Chi. L. Rev. 669 (2019) (explaining the role of legislative findings and arguing for courts to afford them more weight). ↑
257. . See Nielson & Walker, supra note 36, at 52 (noting that the Comptroller of the Currency and inspectors general are labeled “independent” by statute, creating some norms of independence). ↑
258. . Myers v. United States, 272 U.S. 52, 135 (1926). ↑
259. . See, e.g., Exec. Order No. 13,843, § 1, 3 C.F.R. 844, 844 (2019) (reiterating following Lucia the “policy” that ALJs should be “impartial”). ↑
260. . See Nielson & Walker, supra note 36, at 54–55 (demonstrating dynamic role that Senate confirmation plays in removal). ↑
261. . Rebecca Eisenberg & Nina Mendelson, Limiting Agency Head Review in the Design of Administrative Adjudication, Yale J. on Regul.: Notice & Comment (Feb. 21, 2022), https://www.yalejreg.com/nc/symposium-decisional-independence-06/ [https://perma.cc/X5VF-SK8A]. See generally Rebecca S. Eisenberg & Nina Mendelson, The Not-So-Standard Model: Reconsidering Agency-Head Review of Administrative Adjudication Decisions, 75 Admin. L. Rev. 1 (2023) (arguing that agency-head review may be an unwise choice, especially in the highly technical patent context, and courts should not “constitutionalize a one-size-fits-all approach” to agency structures and adjudication schemes). ↑
262. . See supra note 232 and accompanying text (discussing a lack of political will to confirm Article III judges). ↑
263. . See U.S. Const. art. II, § 2, cl. 2 (“Congress may . . . vest the Appointment of such inferior Officers . . . in the President alone, in the Courts of Law, or in the Heads of Departments.” (emphasis added)). ↑
264. . See Nielson & Walker, supra note 36, at 36–37 (describing effects of reason-giving requirement); see also 12 U.S.C. § 2 (“The Comptroller of the Currency shall be appointed by the President, by and with the advice and consent of the Senate, and shall hold his office for a term of five years unless sooner removed by the President, upon reasons to be communicated by him to the Senate.”). ↑
265. . See Nielson & Walker, supra note 36, at 37 (“When President Obama removed one—and only one—inspector general, his administration felt obliged to defend itself repeatedly and in some detail to Congress.”). ↑
266. . See Bamzai, supra note 178, at 1378–79 (describing history of Comptroller provisions); Nielson & Walker, supra note 36, at 35–36 (describing history of Comptroller and inspectors general provisions). ↑
267. . Bamzai, supra note 178, at 1379 (quoting Cong. Globe, 38th Cong., 1st Sess. 2122 (1864) (statement of Sen. William Fessenden)); Nielson & Walker, supra note 36, at 34 (recounting this debate). As this Article was going to press, President Trump fired inspectors general at seventeen federal agencies, and litigation has been filed to challenge many of those actions. See Charlie Savage, 8 Inspectors General Fired by Trump File Lawsuit Seeking Reinstatement, N.Y. Times (Feb. 12, 2025), https://www.nytimes.com/2025/02/12/us/politics/inspectors-general-trump-lawsuit.html [https://perma.cc/DNU3-DTMV]. It is too soon to assess the impact these firings will have on the potency of Congress’s anti-removal tools in this or related contexts. ↑
268. . One of the ways Congress could do this would be byenact[ing] a trigger (either by statute or rule) that requires a congressional hearing whenever the head of an independent agency (or other agency official Congress so designates) is fired. This provision could be triggered by any such firing, or perhaps only when the President fails to provide a statutorily required reason. At this hearing, the fired official would testify along with other witnesses the relevant committee chose to call. To be sure, a congressional hearing is by no means a perfect substitute for an adjudication in an Article III court, but it subjects the removal decision to a trial in the court of public opinion.Nielson & Walker, supra note 36, at 53–54 (footnote omitted). ↑
269. . Id. ↑
270. . Cf. Christopher J. Walker, Federal Agencies in the Legislative Process: Technical Assistance in Statutory Drafting 17 (2015), https://www.acus.gov/sites/default
     /files/documents/technical-assistance-final-report.pdf [https://perma.cc/2VJL-X6JS] (reporting that one agency official observed that, in explaining why federal agencies assist Congress in legislative drafting, “his agency feels particularly pressed to complete all technical drafting assistance requests before a senior agency official is scheduled to appear at a congressional hearing”); see also id. (quoting another agency official who said that “oversight is always in the back of our minds” when the agency provides technical drafting assistance). ↑
271. . See supra Part III (outlining and analyzing the three most popular potential reform solutions). ↑
272. . One example of such scholarship argues that[a]dministrative practice is an iceberg. Federal courts see only the tip peaking above the water—the judicial challenges to regulatory actions that make it to the courthouse. Administrative law scholars have dedicated much time to analyzing that small peak of judicial review of agency action and related judicial deference doctrines. Yet, below the water’s surface exists a mass of regulatory activity that escapes the judiciary’s purview.Christopher J. Walker & Rebecca Turnbull, Operationalizing Internal Administrative Law, 71 Hastings L.J. 1225, 1227 (2020) (footnotes omitted) (citing Nicholas R. Parrillo, Federal Agency Guidance and the Power to Bind: An Empirical Study of Agencies and Industries, 36 Yale J. on Regul. 165, 170 (2019)). ↑
273. . See Paul R. Verkuil, The Self-Legitimating Bureaucracy, 93 Yale L.J. 780, 780 (1984) (reviewing Mashaw, supra note 88) (“It is as if, when asked the question what (or where) is administrative justice, [we] look for that particular lost coin under the proverbial streetlight of judicial process, not because the coin is there, but because that is where the light is.”). ↑
274. . Gillian E. Metzger & Kevin M. Stack, Internal Administrative Law, 115 Mich. L. Rev. 1239, 1239 (2017); see also Emily S. Bremer & Sharon B. Jacobs, Agency Innovation in Vermont Yankee’s White Space, 32 J. Land Use & Env’t. L. 523, 523–24 (2017) (explaining how administrative law allows agencies to employ procedures above the APA’s judicially reviewable baseline); Neal Kumar Katyal, Internal Separation of Powers: Checking Today’s Most Dangerous Branch from Within, 115 Yale L.J. 2314, 2318 (2006) (urging a structuring of the Executive Branch with internal checks); Elena Kagan, Presidential Administration, 114 Harv. L. Rev. 2245, 2285–90 (2001) (explaining how executive orders can be used to direct how agencies function). ↑
275. . See, e.g., Christopher J. Walker, Administrative Law Without Courts, 65 UCLA L. Rev. 1620, 1624, 1638 (2018) (“It is a mistake for administrative law to fixate on judicial review as the core safeguard for our constitutional republic.”). ↑
276. . See Walker & Turnbull, supra note 272, at 1242–45 (collecting potential internal administrative law reforms for use in adjudication); see also David Ames, Cassandra Handan-Nader, Daniel E. Ho & David Marcus, Due Process and Mass Adjudication: Crisis and Reform, 72 Stan. L. Rev. 1, 67–72 (2020) (proposing reforms for adjudications where discrete individuals may not, inter alia, have adequate incentives to seek judicial review). ↑
277. . See, e.g., Katyal, supra note 274, at 2328–29 (describing how the Foreign Service’s “Dissent Channel” promotes dissent without a concomitant fear of discharge). ↑
278. . Seila Law LLC v. CFPB, 140 S. Ct. 2183, 2203 (2020) (alteration in original) (quoting 1 Annals of Cong. 518 (1789) (Joseph Gales ed., 1834) (statement of Rep. James Madison)). To be sure, the Court also observed that “[t]he President ‘cannot delegate ultimate responsibility or the active obligation to supervise that goes with it,’ because Article II ‘makes a single President responsible for the actions of the Executive Branch.’” Id. (emphasis added) (quoting Free Enter. Fund v. Pub. Co. Acct. Oversight Bd., 561 U.S. 477, 496–97 (2010)). An “active obligation to supervise,” however, surely can be satisfied by ensuring others within the Executive Branch are not interfering with agency adjudication; indeed, the President’s ability to coordinate the whole of Executive Branch is important. One purpose of supervision, moreover, is to ensure that the system is working well; for adjudication, that may require actively not interfering at the hearing level. Of course, if the adjudicator is doing a poor job, active supervision may also require stepping in. Independence should exist within a range. Even defenders of agency independence recognize that removal is justified under some circumstances. See, e.g., id. at 2239–40 (Kagan, J., dissenting) (describing how the removal power can be “standard fare” and how for-cause removal protections can fit within that scheme). ↑
279. . U.S. Const. art. 2 § 2, cl. 1. ↑
280. . U.S. Const. art. 2, § 2, cl. 3. ↑
281. . Exec. Order No. 13,843, § 1, 3 C.F.R. 844, 844 (2019). ↑
282. . Barnett, Regulating Impartiality, supra note 29, at 1700. ↑
283. . See id. at 1700–01 (discussing “Free Enterprise’s tiered-removal prohibition” and arguing that “impartiality and tiered insulation for adjudicators inure to the benefit of . . . agencies themselves”). ↑
284. . Id. ↑
285. . See id. at 1720–21 (“In fact, the executive branch has a long-standing, yet perhaps overlooked, history of providing civil service protections to improve the professionalism of executive officials through internal administrative law”). The fact that presidents themselves have applied internal administrative law in this way may be relevant in identifying and reinforcing presidential norms. Cf., e.g., Daphna Renan, Presidential Norms and Article II, 131 Harv. L. Rev. 2187, 2202 (2018) (explaining how “past practice” can affect “structural norms” in the context of the “duty to defend”). ↑
286. . See Barnett, Regulating Impartiality, supra note 29, at 1724 (“[C]ompared to statutory administrative law, internal administrative law’s significant disadvantage is that it has less permanence and permits easier repeal.”). ↑
287. . See, e.g., Aaron L. Nielson, Sticky Regulations and Net Neutrality Restoring Internet Freedom, 71 Hastings L.J. 1207, 1223 (2020) (“But policy made through the administrative process can be unmade through the administrative process, and to the extent that policies made through the administrative process lack bipartisan support, we should expect incoming administrations to undo what their predecessors have done.”). ↑
288. . See Barnett, Regulating Impartiality, supra note 29, at 1724 (“[A]gencies can create more regulatory permanence by using notice-and-comment procedures for the promulgation, amendment, or repeal of internal rules. Unless repealed, the regulations would likely have the force of law.” (footnote omitted)); see also Elizabeth Magill, Agency Self-Regulation, 77 Geo. Wash. L. Rev. 859, 874 (2009) (explaining value of commitment mechanisms to make internal administrative law credible). ↑
289. . See Nielson, supra note 42, at 90 (“Because regulated parties know that an agency must survive a procedural gauntlet to change a regulatory scheme, they can have more confidence in that scheme’s stability.”). ↑
290. . See, e.g., Richard J. Pierce, Jr., Rulemaking Ossification Is Real: A Response to Testing the Ossification Thesis, 80 Geo. Wash. L. Rev. 1493, 1498 (2012) (explaining that agencies can easily navigate notice-and-comment procedures except “in the context of the much smaller number of rulemakings that raise controversial issues where the stakes are high”). ↑
291. . See The Indicator from Planet Money, Congressional Game Theory, NPR, at 05:43–06:06 (Oct. 14, 2021, 6:25 PM), https://www.npr.org/transcripts/1046180183 [https://perma.cc/Y5S9-EZ75] (discussing how public commitments impact political tactics, which can be considered a form of game theory). ↑
292. . Nielson & Walker, supra note 36, at 79 n.347 (citing Margo Schlanger, Offices of Goodness: Influence Without Authority in Federal Agencies, 36 Cardozo L. Rev. 53, 65 (2014)). ↑
293. . See, e.g., Arthur R. Miller, Simplified Pleading, Meaningful Days in Court, and Trials on the Merits: Reflections on the Deformation of Federal Procedure, 88 N.Y.U. L. Rev. 286, 290 (2013) (arguing that “like apple pie, baseball, and the flag,” rules designed “to reach the right result after an adversarial contest on a level litigation field [is] a worthy raison d’être for a procedural system”). To be sure, presidents may have the authority to remove the agency ombuds too, but that would only add to the political costs. It surely is not a coincidence that presidents historically have been quite reluctant to remove inspectors general. See Nielson & Walker, supra note 36, at 7 (“Although presidents occasionally remove inspectors general despite having to provide reasons, it is remarkable how often presidents do not remove inspectors general, even after a presidential transition.”). ↑
294. . Cf. Sara C. Benesh, Understanding Public Confidence in American Courts, 68 J. Pol. 697, 703 (2006) (drawing conclusions from a survey about public opinion of state courts that perceptions of fairness in procedures are extremely influential on overall confidence in state courts and judges). ↑
295. . There may be circumstances where political costs prevent beneficial presidential action; not every good decision is popular, especially situations where the reason the action is beneficial is difficult to explain. Internal administrative law may not solve that problem, but it is difficult to see how internal administrative law would make that problem worse than the status quo. ↑
296. . See supra Part I (detailing the standard model). ↑
297. . Cf. Richard J. Pierce, Jr., Does the Constitution Require Agencies to Use Biased Judges?, Regul. Rev. (Oct. 2, 2023), https://www.theregreview.org/2023/10/02/pierce-does-the-constitution-require-agencies-to-use-biased-judges [https://perma.cc/88EH-C97N] (“The Supreme Court’s decision in Free Enterprise Fund does not support the Fifth Circuit’s decision [in Jarkesy] to force all parties to agency adjudications to have their cases adjudicated by biased decision-makers.”). ↑
298. . Oil States Energy Servs., LLC v. Greene’s Energy Grp., LLC, 138 S. Ct. 1365, 1381 (2018) (Gorsuch, J., dissenting). ↑