The Not-So-Standard Model: Reconsidering Agency-Head Review of Administrative Adjudication Decisions

The Supreme Court has invalidated multiple legislative design choices for independent agency structures in recent years, citing Article II and the need for political accountability through presidential control of agencies. In United States v. Arthrex, Inc., the Court turned to administrative adjudication, finding an Appointments Clause violation in the assignment of certain final patent adjudication decisions to appellate panels of unconfirmed administrative patent judges. As a remedy, a different majority declared unenforceable a statutory provision that had insulated Patent and Trademark Office (PTO) administrative adjudication decisions from political review for almost a century. The Court thereby enabled the politically appointed PTO Director to review and change individual decisions, reasoning that this would provide political accountability while conforming PTO practice to “the standard model” for agency adjudication.

Descriptively, agency-head review of adjudication decisions is far from standard, either in current agency structures or in historical patent practice. Nor is it necessarily a panacea for achieving effective oversight. For many kinds of adjudication decisions political control can present dubious benefits and distinct risks. It may bring little accountability for highly technical, low salience decisions, while allowing political officials to reward friends and punish enemies, especially when agencies adjudicate high-value claims. It may compromise other long-valued adjudication features, including independence of adjudicators from enforcement officials. It may be a haphazard mechanism for achieving uniform decisions in high-volume adjudication regimes. Finally, it may tempt agencies to make policy through adjudication rather than through procedures that enable broader input and provide greater accountability such as rulemaking. In short, no single model is likely to be appropriate in all adjudication settings. The history of legislative design of administrative adjudication structures shows that the political branches are able to learn from experience, assess tradeoffs, and revise institutions to address public needs. When the Court redesigns agency structures, it intrudes on responsibilities traditionally and more appropriately exercised by the political branches.

Introduction

Administrative law has long supported legislative flexibility in creating bureaucratic structures with expert staffs, while simultaneously assuring accountability through public input, judicial review, and political oversight. In recent years, however, the Supreme Court has repeatedly stepped in to invalidate legislative design choices for administrative agencies, citing the need for political accountability. In particular, the Court has found violations of Article II’s Vesting and Take Care Clauses in independent agency structures that Congress has insulated to some degree from presidential control. This coincides with calls for greater review of agency decisions by independent, politically insulated courts, suggesting that it may reflect an anti-administrativist impulse to confine and check bureaucratic power as much as it reflects confidence in the benefits of political control.

The Court emphasized accountability through political control—as distinct from review by independent tribunals—when it turned its attention to agency adjudication structures in its 2021 ruling in United States v. Arthrex, Inc. A majority of the Court found an Appointments Clause violation in the assignment of final decision authority within the Patent and Trademark Office (PTO) to panels of unconfirmed administrative patent judges (APJs) serving on the Patent Trial and Appeal Board (PTAB). Although PTAB panels had final decision authority within the agency to resolve private challenges to the validity of previously issued patents, the statute gave the PTO Director unreviewable discretion to decide whether to institute proceedings in response to petitions, authority to designate the composition of panels for particular cases, and authority to prescribe regulations governing the proceedings. PTAB decisions could be reviewed by Article III courts, but not elsewhere in the agency. To remedy the Appointments Clause violation, a different majority effectively modified the statute to allow review of adjudicatory decisions by the politically appointed PTO Director, declaring unenforceable a legislative provision that had insulated administrative adjudication decisions in the PTO from political review for almost ninety-five years. Although Arthrex was formally an Appointments Clause case, the Court reflexively applied to the agency adjudication structures before it much of the same analysis that guided its decisions in the independent agency cases decided under the Take Care Clause and Vesting Clause of Article II. In effect, the Court chose political control rather than political independence for PTAB decisions. Some commentators have also advocated for across-the-board agency-head review of adjudicatory decisions as a matter of “best practices” for agency adjudications. We disagree.

Commentators have argued that political accountability is important when agencies resolve policy issues, whether through rulemaking or through adjudication. This framing harkens back to older arguments that agency adjudication should be seen as in part “political” rather than strictly “judicial.” Characterizing agency adjudication as “political” recognizes that it sometimes addresses matters of policy. This picture of agency adjudication has implications for the appropriate mix of judicial review and political review of decisions, as well as for how far adjudicators should be walled off from enforcement staff. Commentators today take that older debate further. They champion agency-head control over adjudication not only because agency adjudication might implicate policy concerns, but also to bring about more consistent, error-free adjudication decisions.

In our view, this one-size-fits-all position risks placing undue confidence in agency-head control to supply accountability and fails to acknowledge serious risks that political control over agency adjudication decisions can present. For some adjudications that resolve salient policy issues, political control of adjudication decisions may be an effective and accountable way for an administration to advance its policy preferences. But for many kinds of administrative adjudications—including the patent proceedings that the Court addressed in Arthrex—political control can present dubious benefits and distinct risks. Advocates of decisional independence and insulation of agency adjudicators have long identified the risk that adjudicators might otherwise unfairly favor agency enforcers, a risk that is aggravated by political control of adjudication decisions. Beyond this, political control of adjudication may be simply ineffective as a source of accountability for highly technical, low salience decisions, while providing an opportunity for political supervisors to reward friends and punish enemies. The option of political review of adjudication decisions may tempt agencies to use individual adjudications to establish new policies rather than engaging in more public-facing proceedings, such as rulemaking, that enable broader input into policy decisions and provide advance public notice of policy changes and new governing legal interpretations. And for high-volume adjudication regimes, review by a busy agency head may be a haphazard mechanism for achieving consistent and accurate adjudication decisions. These considerations point us to a more nuanced assessment of political accountability and structures of accountability in general.

Courts are in a poor position to observe these tradeoffs, which vary across administrative contexts and may come to light only over time through unsatisfactory experience with prior structures. When the Court, as it did in Arthrex, chooses its own preferred designs for agency adjudication as a remedy for constitutional violations, it truncates the political branch processes of learning from experience, considering competing interests, and reforming institutions to better address public needs. In the setting of administrative adjudication in particular, Congress has considered and reconsidered adjudication structures and processes, settling on legislation that presidents have signed into law. The political branches make these design choices with better information, greater institutional competence, and greater political accountability than the courts, and adjust them over time as they learn from administrative experience. These advantages are consistent with the Constitution’s authorization to Congress, not the courts, to make all laws necessary and proper for executing the powers the Constitution vests in the government. History demonstrates that Congress and the Executive Branch have made good use of their institutional competence in assessing and deliberating over these issues in the legislative process.

Although the Court has increasingly required greater executive control over agencies through its Article II Take Care Clause and Vesting Clause rulings, it has not decided how much political control over agency adjudicators these clauses require. Meanwhile, a divided Fifth Circuit panel recently held, in Jarkesy v. Securities and Exchange Commission, that the for-cause restriction on removal of Securities and Exchange Commission (SEC) administrative law judges (ALJs) violated the Take Care Clause. At the time of writing, a certiorari petition seems likely; the issue is increasingly pressing.

We beginwith a review of the growing popularity of political control of the administrative state in the Supreme Court and in scholarly literature. Part I traces the rise of the unitary executive theory in the courts in Article II decisions declining to enforce restrictions on the authority of the President to fire agency officials. Part II examines the Court’s more recent use of a similar analysis in Arthrex, where it relied on the Appointments Clause in declining to enforce a statutory restriction on agency-head review of administrative adjudication decisions. In Part III, we challenge the Arthrex majority’s assertion that agency-head review of adjudicatory decisions is a standard, “almost-universal” practice. Part III.A shows that Congress has provided a wide variety of structures for administrative review in different agencies, including many examples of restrictions on the authority of the agency-head to review adjudication decisions. Part III.B turns to historical practice. That section reviews the long but previously underexplored history of administrative adjudication in the patent system, contributing to the literature on the history of the administrative state. With respect to patent adjudication, we find that, for most of this history, Congress allocated final decision authority within the Executive Branch to administrative adjudicators rather than to the agency head, relying on the courts to provide further review.

Part III.C considers the normative case for agency-head review as well as its limitations and pitfalls. We argue that political control of agency adjudication may not promote political accountability for highly technical and low-salience decisions; that it can pose distinctive risks in some settings; and that other control mechanisms such as rulemaking may provide more political accountability for policy choices. Because adjudication structures vary greatly, we conclude that no single model is likely to be appropriate in all settings. In Part III.D, we argue that the political branches have a considerable advantage over the Court in assessing the relevant considerations, in designing appropriate structures for administrative adjudication in different agencies, and in modifying those structures over time. We conclude with a caution against ossification of constitutional constraints that arrest the ongoing political process of crafting a government that serves the needs of the public.

I. The Growing Enthusiasm for Political Control
in Vesting Clause and Appointments Clause Cases

The unitary executive theory is, of course, the notion that the President must have control over Executive Branch officials. Theorists have sometimes defended this notion on original intent grounds, and sometimes on the functional ground that presidential control provides electoral accountability for Executive Branch actions. Opponents contest both positions.

The Court has yet to endorse the theory in so many words in any majority opinion or to decide explicitly how far presidential control must extend. Without naming it outright, a series of recent opinions by Chief Justice Roberts have nonetheless given teeth to the unitary executive theory by declining to enforce legislative provisions that the Court thinks leave the President with too little authority over the administrative state. For now, these may seem like baby teeth, leaving minor dents in departments that continue to perform as before, but they point toward places where larger incisors may follow.

In each case, the challenged legislation has been controversial, although not necessarily on the grounds highlighted in the Court’s constitutional analysis. The formal focus of the Justices who find a constitutional violation is on the allocation of authority and oversight within the Executive Branch. While they may also seem troubled by the regulatory reach of the legislation, the problem they address is not that the legislation expands regulation per se, but that it allocates powers to unaccountable bureaucrats rather than to the elected President in whom the Constitution vests the executive power of the United States—“all of it.”

So far, a majority of Justices in each case has stopped short of invalidating the statutory scheme entirely. Instead, they have nibbled at the margins with a targeted remedy that severs an administrative design feature without directly invalidating the substantive extension of regulatory authority itself. But some justices have flashed their incisors in dissent, ready to invalidate the statute and leave Congress to figure out how to fix it in new legislation.

In the 2010 decision in Free Enterprise Fund v. Public Company Accounting Oversight Board, the Court considered a challenge to provisions in the Sarbanes-Oxley Act of 2002 that created a new five-member Public Company Accounting Oversight Board (PCAOB) within the SEC and gave it “expansive powers to govern an entire industry”—the accounting industry. A five-Justice majority of the Court left the PCAOB and its expansive powers in place. But the majority found the provision of “two layers” of removal restrictions protecting the PCAOB to violate Article II’s Vesting Clause and severed from the statute the “for good-cause” restrictions on the Commission’s ability to remove PCAOB members before the end of their terms.

A decade later in Seila Law LLC v. Consumer Financial Protection Bureau, a more fractured majority took a similar approach to the legislative design of the controversial Consumer Financial Protection Bureau (CFPB) in the Dodd-Frank Wall Street Reform and Consumer Protection Act. Congress created the CFPB in the wake of the 2008 financial crisis as an independent agency headed by a single director with a five-year term, providing for removal by the President only for “inefficiency, neglect of duty, or malfeasance.” The Court had previously upheld identical statutory tenure protection for Federal Trade Commission (FTC) commissioners, but the Seila Law majority distinguished the FTC from the CFPB, characterizing the former as multimember, balanced along party lines, and implementing duties that the Court characterized, somewhat disingenuously, as “neither political nor executive.” Noting that the CFPB Director “wields vast rulemaking, enforcement, and adjudicatory authority over a significant portion of the U.S. economy,” five Justices concluded that for-cause removal provisions, judicial review, and other checks on the CFPB Director’s authority were insufficient to satisfy the Constitution, and that the Constitution further requires that the President be free to remove the CFPB Director without cause. Two members of that majority were unwilling, however, to join the Chief Justice in limiting the remedy for the constitutional violation to severance of the tenure provision. To reach a majority for that remedy, the Chief Justice needed the votes of Justices who dissented on the analysis of constitutionality. The bottom line in Seila Law was similar to that in Free Enterprise Fund: the Court found constitutional violations in statutory provisions limiting grounds for removing the heads of agencies, and severed the offending tenure provision while leaving the rest of the statute in place.

The Court made a similar ruling the following year in Collins v. Yellen. Although the Federal Housing Finance Agency Director—who has the power to regulate and enforce the law against government-sponsored mortgage financing institutions, such as Fannie Mae—arguably exercised far less significant power than the CFPB Director, the Court nonetheless severed the for-cause restriction on the President’s power to remove the CFPB Director, stating that the Constitution prohibited even “modest restrictions” on the President’s power to remove such an agency head. Again emphasizing its concern for political accountability, the majority reasoned that Presidential control was essential to “subject Executive Branch actions to a degree of electoral accountability.”

II. The Potential Expansion of Political Control to Administrative Adjudication

In some respects, Arthrex followed the same playbook. Although dubiously framed as an Appointments Clause case, it reads like the Vesting Clause cases that preceded it. Yet, strikingly unlike those cases, Arthrex is about an adjudicatory regime. Once again, the Court considered a dispute controversial recent legislation—this time, provisions in the Leahy-Smith America Invents Act of 2011 (AIA) that expanded available proceedings within the PTO for challenging the validity of previously issued patents before an administrative tribunal. The AIA had made no change, however, to statutory provisions that, since 1927, had allowed final decisions of the tribunal to be appealed directly to an Article III Court without further review within the agency. A majority of the Court found a constitutional violation. And once again, a different majority held that the appropriate remedy was to sever a statutory provision while leaving the rest of the law in effect.

This time the Court’s objection was not to the independence of leadership in a new administrative department with a full range of functions, but rather to the independence of expert adjudicators in a very old department with a long history of administrative adjudication. Specifically, the issue was the status of hundreds of APJs who adjudicate validity challenges within the PTO as members of an administrative tribunal, recently renamed the Patent Trial and Appeal Board (PTAB). In earlier Article II Vesting Clause cases, the Court had explicitly distinguished administrative law judges as falling outside the scope of its decision. Moreover, in its earliest major case on presidential control, Myers v. United States, the Court identified adjudication as a distinct function not appropriate for political control.

In the lower court decision under review in Arthrex, the Court of Appeals for the Federal Circuit nonetheless concluded that the statutory rights and responsibilities of APJs made them “principal officers” for whom the Constitution required appointment by the President with the advice and consent of the Senate, rather than “inferior officers” whose appointment could be delegated to the Secretary of Commerce as provided in the statute. Following the remedial approach in Free Enterprise Fund, the Federal Circuit modified the portion of the statute that restricts removal of the PTO’s officers and employees (but only as applied to APJs), allowing the PTO Director to fire them without cause. Without for-cause job protection, the Federal Circuit held that APJs would qualify as “inferior officers” whose appointments Congress could delegate to the Secretary, thus fixing the constitutional violation with minimal disruption to the statutory scheme.

The Supreme Court ultimately chose a different remedy. In the merits portion of the decision, a majority led by Chief Justice Roberts noted that “Congress provided that APJs would be appointed as inferior officers, by the Secretary of Commerce as head of a department,” and then asked “whether the nature of their responsibilities is consistent with their method of appointment.” The majority thus took the legislative provisions on appointment and removal as fixed and shifted its focus to the design of the tribunal’s authorities to determine whether there was an Appointments Clause violation. Although the statute provided for significant administrative and substantive oversight by principal officers, authorizing the Director to control panel assignments of APJs and to make unreviewable decisions about whether to institute particular proceedings and authorizing the Secretary to remove APJs for cause, the majority focused on the APJs’ “power to render a final decision on behalf of the United States” without review by any principal officer in the Executive Branch. Statutory provision for judicial review of PTAB decisions was irrelevant, because “review outside Article II—here an appeal to the Federal Circuit—cannot provide the necessary supervision . . . [of] the ‘executive Power’ for which the President is ultimately responsible.”

To remedy this violation, a different majority modified the statutory scheme to provide for review of PTAB decisions by the PTO Director. Key to this portion of Chief Justice Roberts’ opinion was the repeated assertion that agency-head review of adjudication, not present in the statutory regime under review, is “the standard way to maintain political accountability and effective oversight for adjudication” and “the almost-universal model of adjudication in the Executive Branch.”

The majority lost Justice Gorsuch’s vote on its remedy for the violation, but Justices Breyer, Sotomayor, and Kagan, who dissented on the merits, concurred in the majority’s remedial holding, perhaps figuring they could thereby limit its scope. The result was a majority decision to sever a statutory provision specifying that only PTAB itself could grant rehearings. As the Court interpreted the statute, that change would allow the politically accountable, Senate-confirmed Director to set aside PTAB decisions in inter partes review proceedings (IPRs), allowing de novo review of adjudication decisions by a politically appointed official who was removable at will by the President.

Arthrex signals a significant expansion of separation of powers jurisprudence, both in the Court’s new willingness to find Appointments Clause violations in Congress’s choice of an appointments process for administrative judges, and in the Court’s willingness to rewrite legislative structures for administrative adjudication to fix the problem. That said, the split decision and unusual circumstances, including that Arthrex was decided under the Appointments Clause, leave the Court some wiggle room in future cases. Senate confirmation of affected officials would immunize them from Appointments Clause challenges, of course, although it may be impractical for the large tribunals that are necessary for high-volume adjudication regimes. Principal officer review below the level of the agency head, including in an appellate tribunal, might be another option. Further, for what it may be worth, the Court specifically limited its holding to IPR proceedings, commenting neither on other patent adjudication proceedings nor on adjudication frameworks in other agencies.

That wiggle room could disappear with continued acquiescence in agency-head review of adjudicatory decisions as a remedy in future Appointments Clause cases. It could also disappear if the Court were to hold in a future case that Article II’s Vesting or Take Care Clauses require agency‑head control. The remedy settled on by a fractured majority in Arthrex to minimize the greater disruption of invalidating the statute could potentially become ossified as an apparent constitutional constraint—or at least a constitutional safe harbor—for permissible designs of administrative adjudication regimes. That would represent a significant departure from current and historical practice.

Despite Chief Justice Roberts’ assertion in Arthrex that agency-head review is a “standard” means of providing political accountability and oversight and “the almost-universal model of adjudication in the Executive Branch,” agency-head review is far from standard and by no means universal across the varied landscape of administrative adjudication, either today or in the past, as we explain below. Both the variety of current practice across the administrative state and the long history of congressional tinkering with adjudication structures argue strongly against judicial embrace of agency-head review as a uniform constitutional constraint. Even if it may offer a useful mechanism of political control of policy choices in some administrative contexts, it is not standard and should not be required.

III. Examining the Varied Roles of Agency Heads in Administrative Adjudication

In Part III.A., we examine the role of agency heads in administrative adjudication regimes, both in current and historical practice. That Part takes on Chief Justice Roberts’ empirical claim that agency-head review of adjudication decisions is “standard” and “the almost-universal model” of adjudication today. In Part III.B., we then turn to historical practice, showing that agency-head review has been anything but standard in the long history of administrative adjudication in the patent system. In Part III.C., we consider the normative case for agency-head review of adjudication decisions as a mechanism for providing political control and accountability for Executive Branch policy choices, evaluating its limitations and pitfalls. We compare agency-head review of adjudication decisions to other mechanisms for providing political control and accountability that better serve those purposes, such as rulemaking and use of less formal agency guidance documents. Finally, in Part III.D., we consider the institutional advantages of the political branches over the courts in designing administrative structures for adjudication, and in learning from experience and reforming those structures over time.

A. Agency-Head Review of Individual Adjudication Decisions is Far From the Standard Model Today

Chief Justice Roberts asserted in Arthrex that agency-head review of adjudication decisions is “standard” and “almost-universal.” This type of descriptive claim sometimes takes on outsized legal implications in Supreme Court cases, making it important to assess its accuracy. In addressing separation of powers claims, the Court has repeatedly invoked practices that it asserts are typical or widespread to distinguish the administrative designs under review as “novel,” and thus more vulnerable to challenge, than those that have previously avoided or survived constitutional scrutiny.

Chief Justice Roberts cited an influential and provocative law review article by Professors Christopher Walker and Melissa Wasserman for the characterization of agency-head review as “standard” and “almost-universal.” But Walker and Wasserman are more cautious, placing quotation marks around the phrase “standard federal model” and attributing it to others. For their part, they recognize that agency-head final review has rarely been discussed in the administrative law literature, including in leading treatises, that it does not appear on a list of “core features” of formal adjudication governed by the Administrative Procedure Act (APA), and that it was not available in three out of ten in-depth case studies of “Type B” agency adjudication (i.e., adjudications not covered by APA requirements for formal adjudication but sharing some features of formal adjudications, including hearings) done by Professor Michael Asimow for the Administrative Conference of the United States. In arguing for embrace of agency-head review as a “best practice,” they aimed to raise the profile and tout the normative benefits of a practice that was common but by no means universal in the increasingly varied world of agency adjudication. Walker and Wasserman do assert that APA-governed formal adjudication incorporates agency-head final decisionmaking authority, although Congress is free to legislate different procedures in other statutes. Recognizing that the overwhelming bulk of adjudication in the administrative state is not formal adjudication governed by the APA, Walker and Wasserman suggest further that “in the vast majority of Type B adjudication models, the agency head has some degree of decisionmaking authority.”

In fact, across the administrative state, including in important and high-volume adjudication regimes, the picture of agency-head control—or even principal officer control—over adjudication decisions is far more complex, varied, and nuanced than the Arthrex Court recognized. Adjudication frameworks vary widely, as well-documented recent analyses by Professors Michael Asimow and Emily Bremer demonstrate.

Congress has not stuck to a single model for internal review of agency adjudication decisions, but instead has provided different schemes reflecting wide-ranging considerations that impact the validity and integrity of adjudication in different settings. Those design choices sometimes include specific restrictions on agency-head review.

Within the Department of Labor, for example, the Secretary designates five “especially qualified” individuals as the Benefits Review Board (BRB). The BRB reviews compensation and benefits claims initially decided by ALJs in formal adjudication hearings under the Longshoreman and Harbor Workers’ Act, the Black Lung Benefits Act, and several other statutes. The BRB can set aside the ALJ’s findings and conclusions only if they are not supported by substantial evidence. By statute, decisions of the BRB may be appealed directly to federal court; the statute gives the Secretary of Labor no further review authority.

The Contract Disputes Act formally authorizes four boards of government contract appeals—the Armed Services Board of Contract Appeals, in existence in some form since 1970, the Civilian Board of Contract Appeals, the Postal Service Board of Contract Appeals, and the Tennessee Valley Authority Board of Contract Appeals—each with jurisdiction to decide appeals from government contracting officer determinations. Government contracts can include lucrative and often sophisticated arrangements to supply goods, services, security, and even policy analyses. Disputes impact the contractor’s obligation to perform and right to be paid. These appeals boards are to be staffed “in the same manner as [ALJs],” from a register by the hiring agency, with the additional requirement that a member must have had at least five years of experience in public contract law. An interested party may appeal board decisions directly to federal court, subject to a deferential standard of review on the facts, but the statute does not provide for agency-head review. These provisions have been in the law for over forty years—at least since 1978.

The anti-discrimination provision of the Immigration and Nationality Act has barred, for at least thirty-five years, “unfair immigration-related employment practice[s]” based on (legal) immigration status or national origin. By statute, an ALJ’s determination of these claims is final—not subject to further agency review—and directly appealable to the courts. Decisions of the Department of Agriculture’s Judicial Officer are likewise not further reviewable by the agency head.

The Environmental Appeals Board at the Environmental Protection Agency has a similar review structure, with Board decisions appealable directly to federal court. In contrast to the preceding examples, in this particular case the statute reserves the possibility of agency-head review, but that possibility is merely a formality and reportedly never used.

Finally, in two of the largest federal adjudication schemes in the country, agency heads have no review of final decisions on either veterans’ benefits claims or Social Security claims. Lay adjudicators at the Department of Veterans Affairs (VA) handle over one million newly filed claims each year, which is only a portion of the agency’s overall claims load. The VA’s Board of Veterans’ Appeals is staffed by sixty-one Veterans Law Judges; its caseload exceeds 50,000 appeals per year. Meanwhile, of the roughly 2,000 ALJs in the federal system, over 1,600 are assigned to handle the massive workload of the Social Security Administration (SSA), with 110 non-ALJ administrative appeals judges responsible for hearing appeals.

In both of these high-volume adjudication schemes, an appellate panel has the final word within the agency on adjudicated claims. Congress has provided by statute that Board of Veterans’ Appeals decisions on veterans’ benefits are not further reviewable within the VA itself, although judicial review is available in an Article I court.

And since 1940, the SSA (consistent with the practice of its predecessor, the Social Security Board) has structured its retirement, survivors, and disability benefits adjudication system to give final authority over benefits decisions to its Appeals Council, staffed with administrative appeals judges (AAJs), rather than to the Social Security Commissioner. The SSA presents the Appeals Council to the public as the “last administrative decisional level.” By regulation, its decisions are final within the agency, appealable only to federal court. In 2020, when the SSA revised its regulations in light of caseload concerns to authorize the (non-ALJ) AAJs, in addition to its ALJs, to decide front-line cases, a central concern expressed in rulemaking comments was ensuring the “independence and integrity of our existing administrative review process.” Social Security disability decisions are heavily factual and medical in nature. Although the Social Security Commissioner possessed no further review authority, commenters still worried that AAJs, closer to the Social Security Commissioner than ALJs and eligible for potential bonuses, might be insufficiently neutral and impartial to properly carry out their factfinding duties.

The SSA’s adjudication framework may look like an example of agency-head control, since the statute formally delegates authority to the Social Security Commissioner, who in turn has structured adjudication to insulate adjudicators. But the Commissioner and the predecessor Board have, by regulation, consistently declined any formal role whatsoever in reviewing individual adjudication decisions. The Attorney General’s Committee on Administrative Procedure reported the finality of Social Security Appeals Council decisions within the agency to Congress—and, further, that the agency head had “completely divested itself” of review authority—when Congress enacted the APA in 1946. And even as it showed its awareness of this agency review feature in legislative history, Congress has left it alone as it has repeatedly amended statutory provisions relating to both agency and judicial review.

For this major adjudication regime—as well as for the VA—perhaps the political branches view agency-head review as unnecessary, infeasible, inefficient, or unhelpful given the factual and technical nature of the questions the ALJs must resolve. They might also view a firm limitation on agency-head review as a valid means of responding to concerns about separating enforcement and adjudication. Together with the other statutory examples given, the longstanding absence of actual agency-head review in these and other high-volume—and politically salient—adjudicating agencies shows that agency-head review is by no means “almost-universal,” and calls into question just how standard that model actually is.

One particular context in which review of agency adjudications does seem to be typical (and perhaps the Arthrex Court mistakenly took this context to be prototypical) is adjudication in the independent multimember commissions, including the SEC, FTC, Federal Communications Commission (FCC), and National Labor Relations Board (NLRB). These multimember commissions are indeed open, as a matter of course, to party requests to review frontline adjudicator decisions, though their caseloads are typically much smaller than those of the SSA or the VA. And in these particular agencies, adjudication can be a vehicle for major (and highly visible) policy announcements.

But even for the multimember commissions, where agency heads regularly review frontline adjudication decisions, courts have limited the scope of agency-head control in longstanding judicial review doctrine. For example, courts applying the substantial evidence review standard have long prioritized deference to frontline adjudicator findings over maximizing agency-head control. In the leading case of Universal Camera Corp. v. NLRB, the Supreme Court declared that substantial evidence review included ensuring that the agency heads had taken proper account of—and, indeed, deferred appropriately to—the findings of the hearing examiner.

Chief Justice Roberts also cited APA § 557 as embodying a “model of principal officer review” in asserting that agency-head review was “standard.” By its own terms, that section applies only to the rather small proportion of agency hearings that are governed by the APA’s formal adjudication provisions. With respect to those hearings, § 557 states that on appeal from an initial decision, “the agency has all the powers which it would have in making the initial decision[.]” While this language could be read to imply an expectation of high-level agency review in formal adjudication, on its face the text merely specifies what happens when agencies do conduct such review, rather than expressly requiring that review be available.

Historical context further clarifies that the APA formal adjudication provision’s reference to review by the “agency” does not necessarily mean what we now think of as the “agency head.” The 1941 Final Report of the Attorney General’s Committee on Administrative Procedure (the Report), widely agreed to be an “authoritative discussion” and the “intellectual foundation” of the APA, addressed the issue explicitly. In its adjudication chapter, after generally recommending an agency adjudication structure in which a hearing officer’s decision “will serve as the initial adjudication of most cases, and the final adjudication in many,” it stated that “agency heads . . . might take up any decision for review [from a hearing officer] upon their own motion,” to “preserve uniformity of decision and effective supervision.”

The Report then clarified, however, that “in this context . . . the term ‘agency heads’ connotes the highest adjudicatory body . . . even though the members of that body may not be the agency heads themselves.” The Report gave, as one example, the appellate framework for Social Security benefits decisions, noting that the then-called Social Security Board had “completely divested itself of jurisdiction to consider individual claims . . . .[D]elegat[ing] [final authority] to an Appeals Council.” The Report stated that for Social Security adjudication, “[I]t is the Appeals Council rather than the Social Security Board which is the ‘agency head.’” The Report went on to explain more generally that:

In single headed departments and agencies, like the . . . Departments of Commerce and Agriculture, the Committee recommends that all pretense of consideration of each case by the agency head be abandoned and that there be created either boards of review, as in immigration procedure, or chief deciding officers who shall exercise the final power of decision. . . . [because systematic individual secretarial consideration is] obviously impossible.

Apart from providing these examples, the Report did not discuss further whether review boards were to be created by Congress or instead by the agency under a general delegation of statutory authority. But since the Report’s “general recommendations” were aimed at Congress, it is reasonable to think the Report contemplated that Congress might create such agency review boards and give them the final say within the agency through legislation.

House and Senate Committee Reports on the APA add little on the question, though one House Report stated that agency heads should confine their review of front-line adjudications to “important issues,” casting doubt on an expectation that agency-head review would be the default mechanism.

In short, it is by no means clear that the APA sought to standardize a “model” of agency-head review, even for the small proportion of agency adjudications its formal adjudication provisions govern by their own terms. The APA, of course, does not preclude Congress from adopting other structures. As described above, Congress has frequently limited agency-head review in favor of other review structures, such as judicial review. Congress has extended these limits to bodies conducting formal adjudication that would otherwise be governed by the APA, such as the Department of Labor’s BRB.

Although the Court in Arthrex termed agency-head review “standard” and “almost-universal,” the reality is far more varied, complex, and nuanced. For the Court now to embrace agency-head review of adjudication decisions as a constitutional requirement would disrupt structures that have long governed administrative adjudications in important contexts and override the considered judgments of the political branches as expressed in legislation.

B. Agency-Head Review Has Not Been Available for Most of the Long History of Patent Adjudications

We now turn to the history of administrative adjudication in the patent system. We review this history at some length for two reasons. First, as one of our oldest adjudication regimes, patent adjudication sheds light on historical understandings of permissible administrative adjudication structures, as well as on the successes and failures that led Congress and the Executive Branch to reform some structural features while leaving others in place over the years. Second, in Arthrex the majority incorrectly characterized the historical practice in the patent system when it stated that beginning in 1836 “for nearly the next hundred years” the Commissioner of Patents, as head of the Patent Office, heard appeals from adjudicatory decisions of “the forebears of today’s APJs.” The Arthrex majority characterized agency-head review as both standard practice and historical practice in justifying its choice of a remedy for the Appointments Clause violation it found. Perhaps, then, correcting this misapprehension of the historical record will make a difference when the Court looks to history for guidance in future cases.

The history of the patent system shows use of a wide range of administrative adjudication structures at different points in time. Since the Founding, Congress and the Executive Branch have repeatedly designed, modified, and abandoned systems for administering patent rights in an ongoing process of learning and adaptation. Sometimes Congress has codified bureaucratic innovations that the Patent Office pioneered to address legislative gaps and emerging needs, and sometimes it has addressed complaints from outside the Patent Office. Although patent scholars have examined portions of this history, it remains an underexplored resource for understanding the history of the administrative state.

1. Early Experimentation in the Founding Era (1790–1802)

In addition to its broad powers to make all laws which are necessary and proper for carrying into execution the powers of the government, Congress has an explicit constitutional grant of power “[t]o promote the Progress of Science and useful Arts, by securing for limited Times to . . . Inventors the exclusive Right to their . . . Discoveries.” Congress might have exercised this power on its own by granting private petitions to issue patents, leaving the Executive Branch out of the process entirely. Instead, the first Congress turned to the nascent administrative state, making the history of patent administration an important reference point for discerning early understandings of the Constitution.

In 1790, Congress chose to delegate to Executive Branch officers the jobs of determining case-by-case which inventions merited a patent and for how long a term. Congress assigned these determinations to a three-member board consisting of the Secretary of State, Secretary of War, and Attorney General, any two of whom could decide to issue a patent. There was no provision for administrative or judicial review of decisions to deny patents, and only a limited one-year opportunity to challenge issued patents in court.

The 1790 patent board was the first of many three-member administrative tribunals that Congress authorized to make decisions about patents, a design that disperses authority and promotes deliberation while avoiding tie votes. Although the members of the first patent board were all principal officers, delegation of the challenging task of reviewing patent applications to such high-ranking officers with other pressing duties soon proved to be a poor design choice for managing the growing volume of patent applications.

After just three years, Congress radically redesigned the patent system in 1793. Abandoning its initial experiment with principal officer review and evaluation of patent applications prior to issuance, Congress authorized the Secretary of State to issue patents on all applications that complied with statutory formalities. The Patent Act of 1793 left the courts to adjudicate challenges to patent validity in litigation, a parsimonious design choice that allowed Congress to avoid the expense of a patent office for a few decades more. But the legislation recognized a need for an Executive Branch role in adjudicating inter partes priority disputes between “interfering applications”—i.e., multiple patent applications on the same invention from different inventors—a problem that vexed the 1790 patent board. The 1793 Act directed the Secretary of State to submit interfering applications to a three-member arbitration panel, with one member selected by each of the applicants and the third selected by the Secretary of State. The decision of any two of the arbitrators “shall be final, as far as respects the granting of the patent,” with no provision for further review in the Executive Branch.

Both the 1790 and 1793 Patent Acts gave final Executive Branch decision authority to a three-member panel. But the arbitration panels under the 1793 Act were not composed of principal officers and—much like PTAB panels today—they were tasked with adjudicating inter partes disputes rather than ex parte applications. Since 1793, patent law has continuously provided for inter partes Executive Branch adjudications of interference proceedings, although the statutory details have evolved over time.

2. Bureaucratic Innovation and Legislative Neglect (1802–1836)

The Patent Act of 1793 remained in force for forty-three years, although its longevity may reflect congressional neglect more than satisfaction with the status quo.

As the volume of patent applications grew, in 1802 the Secretary of State used funds from patent fees to fill a clerk position in the State Department with a full-time employee dedicated to processing patent applications. The first occupant of this position, William Thornton, claimed the title “Superintendent of Patents” and served from 1802 until his death in 1828 without presidential appointment or Senate confirmation. Although he was ostensibly overseen by an illustrious series of Secretaries of State and Presidents from the founding generation that included James Madison, Thomas Jefferson, Robert Smith, James Monroe, John Quincy Adams, and Henry Clay, these politically accountable officers largely held back from exercising supervisory powers as Thornton reached beyond the statute to make important decisions about the design of the patent system.

Thornton used his role in reviewing applications for compliance with formal requirements to avoid issuing patents that he thought were invalid, despite repeated admonitions from the Attorney General that he lacked statutory authority to reject patent applications. Thornton wrote and distributed a booklet on the form and content of applications and advised applicants not to waste their money on patents that he thought lacked novelty, in effect creating an advisory patent examination system after Congress had deliberately eliminated pre-issuance examination of patent applications. When forced to issue patents that he had no authority to reject, Thornton shared his views of their invalidity publicly and privately. He also repeatedly improvised new administrative proceedings beyond those provided by statute, some of which Congress eventually codified.

Not all of Thornton’s moves carried the day. Most notably, Thornton lost a long battle against public access to issued patents during the patent term without the consent of the patentee. Some inventors complained that Thornton used his position to claim rights in inventions for himself, sometimes as a joint inventor with an applicant and sometimes as a rival in an interference. Congress—not the Executive Branch principal officers from the founding generation who might have exercised political oversight—eventually disapproved and put an end to this practice in 1836.

More striking than the rare instances when this strong-willed bureaucrat was overruled were the many in which he was not. Thornton’s extra-statutory moves provoked repeated complaints to his Executive Branch superiors and to the courts, although never on the ground that such important decisions could only be made by a principal officer with a presidential appointment and Senate confirmation. In the early decades of the republic, political accountability seems to have provided little motivation for the Executive Branch to ensure faithful execution of the patent laws, because the patent system was not a salient political issue. But bureaucratic experience during this period nonetheless highlighted deficiencies in the 1793 design and needs for legislative change.

3. Designing and Overseeing a Real Patent Office (1836–1861)

Eventually Congress responded to calls for reform with a major revision of the patent statute in 1836. The Patent Act of 1836 provided for a patent office with professional staff to examine applications for patentability prior to issuance and to reject those that did not meet statutory standards. It authorized presidential appointment, with advice and consent of the Senate, of a Commissioner of Patents who in turn could appoint an “examining clerk” and various other office personnel with approval of the Secretary of State.

The 1836 Act did not give the Commissioner of Patents final review authority over Patent Office decisions. Instead, it provided for appeal of both rejections of applications and decisions of the Commissioner in interferences—inter partes administrative adjudications to resolve priority disputes between competing inventors—to a three-member “board of examiners” appointed by the Secretary of State, “one of whom, at least, to be selected, if practicable and convenient, for his knowledge and skill in the particular art, manufacture, or branch of science to which the alleged invention appertains.” Any two members of the board of examiners could reverse a decision of the Commissioner with binding effect on further proceedings in the Patent Office. Congress placed the Commissioner in charge of initial decisions to grant or deny patents or to resolve priority, while providing for prompt expert review before an independent board of examiners to correct errors of the Commissioner. An accompanying Senate Report explains:

In nineteen cases out of twenty, probably the opinion of the Commissioner, accompanied by the information on which his decision is founded, will be acquiesced in. When unsatisfactory, the rights of the applicant will find ample protection in an appeal to a board of examiners, selected for their particular knowledge of the subject-matter of the invention in each case.

In short, after forty-three years of unhappy experience with reliance on “endless litigation” in court as the only check on invalid patents, Congress sought to limit the need for resort to the courts by providing more expert administrative determinations at the outset, including administrative appeals to a board with pertinent expertise. The legislation notably did not give the Commissioner of Patents authority to review decisions of the board of examiners, but instead authorized the board of examiners to reverse decisions of the Commissioner, specifying that the Commissioner would be governed by board decisions in further proceedings within the Patent Office.

The Commissioner asked Congress to modify the administrative review process in his Annual Report to Congress for 1838, in part because of the inadequacy of the statutory compensation to persuade persons “of requisite qualifications” to serve on the board of examiners. In 1839 legislation, Congress transferred the authority of the board of examiners to review and reverse the Commissioner to “the chief justice of the district court of the United States for the District of Columbia,” acting in his individual capacity. This was not judicial review by a court, but an outsourcing of administrative appeals to an individual judge, who would be compensated for the service by a fixed annual payment out of “the patent fund.” As Commissioner Charles Mason later explained to Congress in his Annual Report for 1855, “That judge is only (for the occasion) a part and parcel of the Patent Office.” The decisions of the chief justice, like those of the board of examiners under the 1836 Act, were binding on the Commissioner but could be modified by a court in a bill in equity proceeding.

This system for administrative appeals became unstable when the aging chief justice was no longer able to handle the patent appeals on top of his regular caseload, yet remained on the bench. In 1852 Congress amended the statute to permit applicants to appeal to other judges on the same court, directing the Patent Office to compensate the judge for each appeal with a fee that the applicant paid to the Patent Office. But some parties exploited this odd statutory design to prolong delays by continuing to appeal to the infirm chief justice rather than to the others, with the result that no appeals were heard for several years.

4. Designing, Codifying and Revising Administrative Appeals (1861–1927)

Faced with a dysfunctional statutory appeals process, the Commissioners innovated. They created ad hoc boards of examiners within the Patent Office to hear appeals from examiner decisions. Meanwhile, they asked Congress to authorize a permanent board of examiners-in-chief in order to achieve uniformity in decisions without relying on the Commissioner to hear all the appeals, which was “not practicable.” Congress finally authorized appeals to an expert three-member administrative tribunal of principal officers in 1861. The “examiners-in-chief” were to review decisions of examiners rejecting patent applications, decisions in interferences, and applications for the extension of patents “when required by the Commissioner.”

The 1861 Act provided for appeal to the Commissioner of Patents from decisions of the examiners-in-chief, beginning a sixty-six-year period of agency-head review of administrative adjudication decisions in the Patent Office. In codifying these new appeals, the Act did not repeal previous provisions for appeal from decisions of the Commissioner to individual judges in the District of Columbia, nor for review by bill in equity of decisions of the Commissioner or of the judges. The result was a long appellate pathway that was an ongoing source of delays and grievances.

Congress addressed some of these complaints in 1870 legislation. The Patent Act of 1870 left the internal administrative appeals to the examiners-in-chief and to the Commissioner of Patents intact, but replaced the appeal from the Commissioner to individual judges with an appeal to the Supreme Court of the District of Columbia sitting “in banc.” The Act explicitly excluded from this appeal to the Supreme Court of the District of Columbia decisions in inter partes interference proceedings. As under prior law, an applicant refused a patent could still pursue a judicial remedy by bill in equity. But under an 1886 decision, a bill in equity suit could not be filed until after exhausting the appeal from the Commissioner to the Supreme Court of the District of Columbia, further lengthening the appellate pathway.

Disappointed patent applicants and losing parties in interferences tried one other non-statutory Executive Branch pathway to overturn Commissioner of Patents decisions: appeal to the Secretary of the Interior. At first, the Secretaries rejected these appeals on the ground that they lacked jurisdiction to review decisions of the Commissioner. Then, after an 1881 Attorney General opinion declared that final discretion in all matters relating to the granting of patents resided in the Secretary, the Secretary briefly considered appeals from Commissioner decisions. The Supreme Court ultimately rejected this approach in Butterworth v. United States ex rel. Hoe.

In contrast to the Arthrex Court’s hierarchical view of Executive Branch authority nearly 140 years later, the 1884 Supreme Court ended the practice of Cabinet Secretary review in Butterworth, holding that the Secretary’s supervisory authority did not extend to the judicial acts of the Commissioner of Patents. Rejecting a uniform rule across the Executive Branch that would give the Secretary, as head of the executive department within which the Patent Office is a bureau, the power to reverse the Commissioner, the Butterworth Court took direction from the governing statute: “Each case must be governed by its own text, upon a full view of all the statutory provisions intended to express the meaning of the legislature.” The Court figured that since the statute provided explicitly for review of patent office decisions by a court “in aid of the patent office,” it necessarily excluded another appeal to the Secretary on the same matter. Nor did the Secretary’s general authority of direction and superintendence permit appeals to the Secretary from “quasi-judicial” actions of the Commissioner of Patents “in cases in which he is by law required to exercise his judgment on disputed questions of law and fact, and in which no appeal is allowed to the courts.” The Court concluded that the Secretary’s powers of “supervision and direction . . . do not extend to a review of the action of the Commissioner of Patents in those cases in which, by law, he is appointed to exercise his discretion judicially,” explaining:

It is not consistent with the idea of judicial action that it should be subject to the direction of a superior, in the sense in which that authority is conferred upon the head of an executive department in reference to his subordinates. Such a subjection takes from it the quality of a judicial act. That it was intended that the Commissioner of Patents, in issuing or withholding patents, in reissues, interferences and extensions, should exercise quasi-judicial functions, is apparent from the nature of the examinations and decisions he is required to make, and the modes provided by law, according to which, exclusively, they may be reviewed.

With this strong affirmation from the Court that the relevant rules for review of administrative adjudication decisions are set by legislation rather than inherent in the lines of authority within the executive department, it fell to Congress and the Executive Branch to address complaints about the many layers of administrative and judicial review that delayed finality in patent adjudications through statutory reform.

5. Elimination of Commissioner Review (1927–2021)

Following years of complaints, President Taft’s Commission on Economy and Efficiency made an investigation of the Patent Office and issued a report in 1912 recommending that Congress eliminate the appeal from the board of examiners-in-chief to the Commissioner of Patents. The Report emphasized the many other demands on the Commissioner’s time, the growing volume of appeals that made it impossible to achieve uniformity in decisions through appeals to the Commissioner in person, the need to streamline patent appeals to reduce delays and expense, the superior technical expertise among the board members relative to that of any individual Commissioner, and the similar recommendations of many former Commissioners over the years. Rather than embracing political accountability as a justification for Commissioner review, the Report cited the status of Commissioners as political appointees as a reason why Commissioner “overthrow” of board decisions is “unreasonable” and “unwise.”

Following a long process of negotiations between the Patent Office and the bar, Congress eliminated Commissioner review of board decisions in 1927 legislation. The Patent Act of 1927 consolidated the previously separate appeals—first to the board and then to the Commissioner—into a single administrative appeal to an expanded board of appeals that included the Commissioner, first assistant commissioner, assistant commissioner, and examiners-in-chief as members. It further streamlined judicial proceedings by giving applicants a choice of either direct appeal to a court or a bill in equity proceeding, but not both. The result was to eliminate one layer of administrative appeal and one judicial review proceeding, sparing applicants from “delay, expense, and great uncertainty.”

After decades of redesigning the administrative appeals structure, the 1927 Act established a remarkably stable structure for administrative adjudications that lasted until the 2021 Arthrex decision. The statutory language enacted in 1927 with respect to the relationship of the board to the agency head was essentially unchanged in the version of the Patent Act before the Court in Arthrex, despite changes over time in the board’s name, the agency head’s title, and the procedures assigned to the board for adjudication. This structure was in its ninety-fifth year when the Arthrex majority declared a key statutory provision unenforceable.

The Arthrex majority attributed the problem with the statute to “the America Invents Act,” the 2011 legislation that created the IPR proceedings before it. But although by then the Senate-confirmed head of the relevant office had been redesignated as the “PTO Director” rather than the “Commissioner of Patents,” the AIA changed nothing about agency-head review of decisions. Instead, it left in place a venerable 1927 restriction on agency-head review, a structure that intervening Congresses had left undisturbed through repeated additions of new proceedings to the board’s assignments.

In legislative hearings held shortly before passage of the 1927 Act, congressional committees explicitly considered and rejected a proposal to eliminate the original version of the same sentence that the Arthrex majority decided to sever from the statute ninety-four years later: “The board of appeals shall have sole power to grant rehearings.” When a witness proposed deleting this sentence to maintain the Commissioner’s supervisory authority, legislators questioned whether this was necessary, distinguishing between the need for political control of the President’s cabinet and the need for decisional independence of agency adjudicators. The chairman of the Patent and Trademark Association of the American Bar Association noted the judicial character of the proceedings and warned that expanding rehearings would defeat the purpose of streamlining appeals:

Those questions [decided by the board] are essentially judicial and they ought to be. Mr. Fenning’s suggestion that the commissioner ought to grant rehearings also, in effect, goes back to the present system where appeals go up there, and there will be just as many petitions for rehearing as there are now direct appeals to him.

Another witness on behalf of the American Bar Association emphasized the need for an independent board with final decision authority. The same witnesses appeared before both the House and Senate Committees and made similar remarks.

With the issue thus fully aired in both houses of Congress, the bill passed and was signed into law, following decades of consideration within the Executive and Legislative Branches of how best to streamline patent adjudication proceedings.

Subsequent legislation has made substantial further changes to judicial appeals from final decisions of what is now the PTO, including creating specialized courts to hear patent appeals. Congress has also repeatedly modified procedures for appointing board members. But it never saw fit to restore the authority of the agency head to review board decisions in almost a full century between the Patent Act of 1927 and the Arthrex decision. That aspect of the administrative structure proved stable and did not generate calls for reform—even when Congress turned its attention to the constitutional requirements for appointments to the board.

6. Revising the Appointments Process (1975–2008)

A better argument for distinguishing the structure of administrative appeals under the 1927 Act from that in the regime before the Arthrex Court would have been that, in 1927, board members were appointed by the President and confirmed by the Senate. Indeed, that was the appointment process for board members beginning with the 1861 Act (which codified the administrative innovation of submitting appeals to a board of examiners-in-chief) until 1975, when Congress modified the statute to provide for their appointment by the Secretary of Commerce upon the nomination of the Commissioner. An accompanying House Report emphasized the legal and technical nature of the duties performed by the examiners-in-chief and the burden of the prior method of appointment in justifying the change. Congress evidently did not consider reinstating Commissioner review of board decisions when it changed the appointments process for examiners-in-chief, suggesting that, at the time, they did not see the relevance of Commissioner review to the appointments process for patent office adjudicators. It is striking that the Arthrex Court paid so little attention to this history in an opinion that is nominally about an Appointments Clause violation. Perhaps the majority was more interested in diminishing the responsibility of agency adjudicators than in correcting the manner of their appointments.

Even more remarkable is the Arthrex Court’s inattention to later 2008 legislation—passed for the purpose of correcting a possible Appointments Clause violation—that again changed the method of appointment of board members. I n 1999 legislation, Congress had vested the powers and duties of a reorganized U.S. Patent and Trademark Office in the newly created position of “Under Secretary of Commerce for Intellectual Property and Director of the United States Patent and Trademark Office” (known as the Director), to be appointed by the President with the advice and consent of the Senate. The “Commissioner for Patents,” who was to serve as chief operating officer with respect to patents, was to be appointed by the Secretary of Commerce. Congress gave the Director authority to appoint other officers and employees, including the APJs who served on the Board of Patent Appeals and Interferences (BPAI), without involving the Secretary.

Professor John Duffy questioned whether this arrangement for the appointment of APJs violated the Appointments Clause in an influential blog post that anticipated much of the Arthrex Court’s analysis years before that decision. But Professor Duffy did not conclude that APJs were “principal officers” who required Presidential appointments and Senate confirmations. His concern was that the “Under Secretary of Commerce for Intellectual Property and Director of the United States Patent and Trademark Office” did not qualify as a “Head of Department” within the meaning of the Appointments Clause, and was thus constitutionally ineligible to appoint “inferior officers.” If appointments of APJs by the Director (after 1999) rather than by the Secretary (between 1975 and 1999) were unconstitutional, a significant number of previous decisions of the BPAI might be invalid. Congress promptly addressed this concern in 2008 by restoring the 1975 system that assigned the appointment of APJs to the Secretary of Commerce in consultation with the Director.

Notably, in 2008 legislation for the sole purpose of bringing the appointments process for APJs in line with the Court’s Appointments Clause jurisprudence, nobody proposed amending the statute to provide for review of board decisions by the Director. Even Professor Duffy, who praised the Appointments Clause requirements as ensuring responsibility for Executive Branch decisions in the President and department heads, worried that de novo re-adjudication of issues decided in BPAI proceedings by the Director would, “at least with respect to individual factual issues, raise difficult issues of due process.” In other words, the more salient concern at the time was not that the Director had too little power over the BPAI, but rather too much.

Perhaps the Court missed this legislative history because it wanted to focus on the passage of the AIA in 2011 as the moment when Congress crossed a constitutional line. This framing harmonizes Arthrex with other recent decisions that set aside purportedly novel statutory designs of administrative structures as outside constitutional boundaries. To be sure, the AIA introduced new administrative proceedings for challenging patent validity that differed from previously available proceedings in significant ways. Challengers have used the new proceedings more extensively and more effectively than previously available ex parte and inter partes reexamination proceedings, as Congress hoped they would. But none of the features of these proceedings that the Arthrex Court sees as inverting the proper chain of command within the Executive Branch represents a break with precedent in the history of administrative adjudication in the patent system.

The Arthrex majority recognizes that that history is “more winding and varied than recounted here,” but nonetheless concludes—incorrectly—that history “has little to say about the present provision expressly ordering the Director to undo his prior patentability determination when a PTAB panel of unaccountable APJs later disagrees with it.” Indeed, if one takes the Court’s constitutional analysis seriously, history shows that nothing of any significance changed in 2011. If the relevant constitutional constraint prohibits board panels of APJs from undoing a prior patentability determination, Congress crossed that line at least by 1952, sixty-nine years before Arthrex, when it authorized a board of patent interferences to cancel patent claims following an adverse judgment in an interference. Interferences, like IPRs, are inter partes administrative adjudications that since 1939 have been finally decided within the Patent Office by three-member board panels, subject only to judicial review. If the relevant constraint is that agency adjudicators with authority to bind the Executive Branch in such decisions must be appointed by the President with Senate approval, Congress crossed that line at least by 1975, forty-five years before Arthrex, and arguably as early as 1836 when it authorized unconfirmed boards of examiners appointed by the Secretary to review decisions of the Commissioner with binding effect. If the relevant constraint requires agency-head authority to review final adjudication decisions, Congress crossed that line ninety-five years before Arthrex, in 1927. Indeed, as late as 2008, with its attention explicitly focused on ensuring that Board members were properly appointed, Congress saw no need to redesign the role of the agency head in the PTO’s adjudications, a clear sign that observers did not foresee how the Court would soon redraw the relevant constitutional lines. Perhaps the hasty innovator setting aside long-settled law was not Congress in 2011, but the Court in 2021.

The history of administrative adjudication in the patent system is indeed “winding and varied.” For more than 200 years, the patent system has challenged the political branches to adopt different administrative structures to manage a growing volume of patents, and to revise them in light of experience. For most of this time, legislation gave final decision authority within the Executive Branch over administrative adjudications to an administrative tribunal, while relying on the courts—not the agency head—for additional review. And during that period, the Court deferred to legislative allocations of review authority, refusing to find implicit review authority higher in the executive chain of command than where Congress placed it. Congress put agency-head review of administrative adjudication decisions to the test for a sixty-six-year period and then, after long review and deliberation, eliminated it in 1927. This history of informed legislative reconsideration of prior designs presents a striking contrast to the Arthrex Court’s abrupt decision to slash away longstanding statutory restrictions on agency-head review from the appellate bench.

The current mechanisms that Congress and the Executive Branch have chosen for the patent system may leave room for improvement, and they may not be optimal for other agencies or adjudication regimes. They are not cast in stone, but rather a work in progress that includes both old and new features. But the Court’s restrictions on constitutionally permissible administrative designs may prove harder to modify. To the extent that these restrictions rest on beliefs about prior legislative and administrative practice, the Court would do well to pay closer attention to the work done by the political branches in over two centuries of designing U.S. patent institutions.

C. The Normative Case for Agency Head Control of Adjudication Decisions Has Important Limits

We have shown that agency-head review is hardly standard and certainly neither universal nor “almost-universal” across agency adjudication regimes, which vary widely in design. Nor are restrictions on agency-head review novel, as shown by the history of patent adjudication, one of our earliest adjudication regimes. Nonetheless some scholars, including Professors Walker and Wasserman, argue that agency-head review should be standard, even if it presently is not. This Part considers those normative arguments.

The normative argument for political supervision of adjudication decisions that mattered most to the Court in Arthrex is that it supplies political accountability by giving elected Presidents and their appointees control over Executive Branch decisions. Chief Justice Roberts expressly focused on the link with the electoral process in Arthrex: “That [executive] power acquires its legitimacy and accountability to the public through ‘a clear and effective chain of command’ down from the President, on whom all the people vote.”

Because agency adjudication decisions may not be entirely case-specific and technical but may also involve policy choices, Walker and Wasserman further argue that agency heads are well-equipped to “set the agency’s policy preferences” and exercise control over “policy development.” Political appointees, with their indirect electoral ties through presidential supervision, face more incentives to align policy choices with the policy preferences of the administration that won the last election, so the argument goes, than career professionals who typically remain with the agency through changes in administration. Walker and Wasserman further claim that political supervisors may have greater “access to experts and staff that provide inputs and partake in the deliberative process” than front-line adjudicators, as well as more control over their time. In turn, review of adjudication decisions can support the agency head in learning what “adjustments to the regulatory scheme” may be necessary, even prompting an agency head to ask regulatory staff to initiate rulemaking. Defenders of agency-head review further argue that it can correct errors and bring greater consistency and transparency to adjudication decisions.

This normative case for agency-head control of individual adjudication decisions has traction in some contexts, but it also has important limits considered in Part III.C.2 below. Depending on the subject matter, adjudication structures, and saliency of the agency decision, agency-head review may offer significantly fewer benefits than advocates claim. Moreover, it can present distinctive pitfalls, particularly in certain types of garden-variety individual adjudication; these complexities underscore the need for deference to legislative judgments on the structure of adjudication. But first, we consider other mechanisms available to agency heads for controlling agency policy, providing a broader context for understanding the role of agency-head review of adjudication decisions.

1. Agency-Head Review of Adjudication Decisions in the Context of Other Policymaking Tools

Agencies have numerous means for supervision and policymaking beyond the review of individual adjudication decisions. At the time the APA was enacted, rulemaking was viewed as the procedural vehicle for making policy decisions, while adjudication was more narrowly “concerned with the determination of past and present rights and liabilities.” The APA specifically defines a “rule” as an agency statement of “general . . . applicability and future effect . . . designed to implement, interpret, or prescribe law or policy . . . .” Rulemaking followed a legislative rather than judicial model. Since that time, longstanding administrative law doctrine has clarified that agencies are not limited to using rulemaking to make policy. Under the Chenery II doctrine, they may also use adjudication to make policy decisions, and agencies have indeed developed policy through adjudication, as students of administrative law learn from leading cases involving the NLRB and the SEC. Nonetheless, policy matters are overall more central to rulemaking—including so-called “exempt” rules such as agency guidance documents—than they are to adjudication decisions.

In addition to relying on the responsiveness of political appointees who oversee rulemaking, the Executive Branch fortifies political control of rulemaking, including policy judgments, through centralized White House Office of Information and Regulatory Analysis (OIRA) review of significant Executive Branch notice-and-comment rulemaking decisions. This reflects the close connection of rulemaking to policy development and enforces political accountability for agency decisions. OIRA also reviews significant guidance documents, or so-called “exempt” rules, that agencies sometimes use to avoid more cumbersome notice-and-comment rulemaking and that may have similar effects. By contrast, the White House traditionally exercises no review authority over Executive Branch adjudication decisions. This is in part to avoid public concern over improper political interference, as discussed in greater detail below, but it also reflects that adjudication raises policy issues less frequently than rulemaking, and is therefore a less important focus of political oversight.

Rulemaking—especially significant rulemaking—is also much less frequent than adjudication and generally more far-reaching. The Executive Branch agencies issue several hundred significant notice-and-comment rules per year, and roughly three thousand rules overall (including hundreds of annual fish catch requirements and plane-model-specific airworthiness directives), all of which must be published in the Federal Register. One or two hundred are economically significant. This is an order of magnitude fewer than the typical number of adjudication decisions in the same time period, whether made by agency heads in agencies that have such review or by frontline adjudicators.

Finally, the “public-facing deliberative process” of notice-and-comment rulemaking is specifically aimed at broader public engagement. The public is entitled to file “data, views or arguments” in response to a notice of proposed rulemaking. Although the APA does not formally require notice-and-comment for “exempt rules,” such as guidance documents, agencies now typically offer similar public comment opportunities for significant guidance documents.

For all these reasons, individual agency rules are far more likely than adjudication to engage the public and to draw its attention. As a consequence, and because rules typically afford the public greater advance notice of their rights and obligations, agencies have long been encouraged to—and typically do—use rulemaking to resolve significant policy issues on a broad basis rather than relying on the narrower resolution of a particular case in adjudication.

By contrast, adjudications of the sort that concern agency-head review advocates—formal adjudications under the APA (“Type A” adjudication in the framework of Professor Asimow) or other adjudications in which the agency conducts a hearing (“Type B” adjudication in the Asimow framework)—are primarily about resolving individual controversies, following a “judicial” model of deciding individual cases by resolving disputed issues of fact and applying settled legal principles. To be sure, in the course of resolving individual controversies agency adjudicators can and sometimes do address unresolved policy matters, perhaps by interpreting statutes, regulations, or guidance. But individual adjudication decisions are unlikely to focus on policy decisions to the same extent as rulemaking. Sometimes, an agency with enforcement authority might choose to bring an enforcement proceeding before agency adjudicators to address a particular policy issue. But in other adjudication regimes aimed at resolving disputes among private parties, the agency’s ability under Chenery II to choose its policymaking method may be largely theoretical; the agency may lack authority to initiate an action before an adjudicator on its own to raise an unsettled legal or policy issue.

A narrow focus on agency-head control over individual adjudication decisions, as in the Arthrex decision, ignores the numerous other tools an agency head may have to make policy. These other tools may offer far more effective ways to constrain outcomes and to ensure accountability for lower-level decisions that implicate policy matters. Agency heads can issue rules that bind adjudicators or guidance documents that inform adjudication decisions. Even an agency like the PTO with only limited rulemaking authority may use rulemaking to resolve important procedural questions, or use guidance to announce policies and legal interpretations on important new questions.

Nor is agency-head review necessary when an adjudicator errs—or misbehaves. Errors in individual cases most often can be checked by appeals to an agency board or by judicial review, while the agency head can address recurring errors or departures from administration policy by removing a poorly performing adjudicator for cause or by announcing a different policy choice by regulation or guidance.

Another way that agency heads may control policymaking in the adjudication setting is by selecting which decisions become “precedential opinions.” For example, the SSA Appeals Council has issued a manual, the Hearings, Appeals, and Litigation Law Manual (HALLEX), to guide Social Security adjudication. The HALLEX manual includes authoritative interpretations of SSA law as well as procedural policies. The PTO now operates a Precedential Opinion Panel “to decide issues of exceptional importance,” with panel membership typically including the PTO Director. The U.S. Citizenship and Immigration Service also designates “precedent decisions” by the joint approval of the Secretary of Homeland Security, the Board of Immigration Appeals, and the Attorney General. The VA General Counsel issues written precedential legal opinions that are conclusive within the agency, and the Merit Systems Protection Board also identifies precedential decisions and non-precedential decisions.

Viewing adjudication in the context of these other tools has two critical implications for analysis of agency-head review. First, when the political branches, through legislation, have already settled on other mechanisms to enable control over and accountability for adjudication decisions, courts should give considerable deference to that legislative judgment and hesitate before overriding it. Second, recognizing that individual adjudications resolve case-specific issues even when they also address policy issues, lawmakers ought to consider whether review of these non-policy determinations by a politically appointed agency head is necessary or even appropriate. These questions require consideration of the limits and pitfalls of agency-head control of adjudication decisions, to which we now turn.

2. The Limits and Pitfalls of Agency Head Control

Adjudication regimes vary widely, and it would be inappropriate either to endorse any one model of agency adjudication or to oppose agency-head review across the board. Designing appropriate accountability mechanisms for the many varied adjudication regimes across the administrative state is a complex task that requires nuanced attention to context. Congress may have good reason to limit agency-head review of decisions in particular agency adjudication regimes. To begin with the political accountability on which agency-head review advocates focus, the benefits of agency-head review of individual adjudication decisions seem most plausible when two conditions are satisfied. First, agency-head review is most likely to be useful for advancing an administration’s policy goals, but only when a policy issue with implications beyond the immediate dispute dominates the questions an adjudicator must resolve. Agency-head review is less likely to supply useful accountability when adjudicators resolve more technical, case-specific matters. Second, agency-head review is most likely to provide political accountability for policy issues when the policy issues at stake in an adjudication are politically salient.

When adjudication is not primarily aimed at resolving policy matters, perhaps because the legal and policy questions at stake are settled and the remaining issues are essentially technical and case-specific, political review of the individual decision may do little to promote political accountability for policy choices.

As its advocates note, agency appellate review of adjudication serves wide-ranging functions in addition to policymaking. It may provide error correction, decisional consistency, and an alternative to court litigation. Error correction may be the main function of internal agency review in many adjudication schemes. But for agencies that handle a high volume of adjudications, such as the PTO and the SSA, review of decisions by a busy agency head (in contrast, say, to an agency’s appellate panel) may be haphazard at best, inadequate to assure accuracy, and a potential obstacle to uniformity and consistency. Moreover, an agency head’s status as a political appointee may confer no particular advantage in correcting errors consistently or well, especially for highly technical decisions. Politically appointed agency heads vary widely in their pertinent technical expertise and experience. Indeed, some commentators have noted that agency-head review of front-line adjudicators may serve to educate the agency head, rather than the other way around. When the agency head lacks relevant expertise, review of individual decisions by an expert appellate body within the agency may be a more effective mechanism for achieving error correction and consistency.

If the appointments process produces a politically responsive agency head with both some relevant expertise and an appreciation of the greater expertise necessary to make high quality adjudication decisions, perhaps agency-head review will provide useful oversight of decisions for accuracy and consistency. But this potential advantage is by no means assured. Political control may instead lead agencies to depart from applicable law and undermine confidence in Executive Branch decisions when political officials fail to respect the technical expertise of career officials within the agency. Both the Bush and Obama Administrations discovered this when they sought to modify decisions of Food and Drug Administration (FDA) experts that would have made the emergency contraceptive Plan B more readily available over the counter. In the Obama Administration Plan B controversy, FDA Commissioner Margaret Hamburg, a doctor and public health professional, stood behind the decisions of FDA experts to make the emergency contraceptive Plan B available over the counter. Controversy erupted when Health and Human Services Secretary Kathleen Sebelius overturned the FDA decision. As Professor Adrian Vermeule explained, because the “FDA had developed a reputation for impartial expertise, supported in part by its extensive network of expert advisory committees,” Sebelius’s decision was widely condemned as “the politicization of science.” In the earlier Bush Administration Plan B controversy, the Administration did not wait to overturn a decision made by professional staff. White House staff got involved earlier by directing the Commissioner to install new members on an Advisory Committee who could be counted on to vote against approval of over-the-counter sales of Plan B. The White House also let the Commissioner know that “there were a lot of constituents who would be very unhappy with [approval of] over-the-counter Plan B.” A reviewing court concluded that FDA lacked good faith in its actions on the Plan B applications based on evidence of “improper political influence.” In both episodes, far from celebrating political control as a source of accountability, critics lamented political interference in decisions that they believed should have followed the expert views of agency scientists.

In the patent setting, technological expertise can be critical to error correction on many questions of patentability. This is especially so for the IPR proceedings at issue in Arthrex. The statute limits these proceedings to technical questions of whether an invention meets statutory standards for novelty and nonobviousness in light of documentary prior art such as prior patents or publications in the field of the invention. Comparing pertinent prior art references to patent claims is in the wheelhouse of the APJs who serve on PTAB, who by law must be “persons of competent legal knowledge and scientific ability.” PTAB’s Standard Operating Procedures drill down further to match the technology discipline preferences of APJs assigned to a case to the technology discipline at issue in that case. While a PTO director might happen to have expertise in the field of the invention, ordinarily they will not.

Similarly, when the issues in an adjudication are essentially factual—as they often are, for example, in Social Security disability cases—agency-head review may offer few distinctive benefits compared to other modes of review, including so-called “quality assurance” approaches.

Even when individual adjudication decisions do raise policy issues, agency-head review is likely to promote electoral accountability only when the decision is transparent, and the issues are at least somewhat salient politically. Political accountability may be a powerful force if the electorate is anticipating a particular policy or set of policies, if interest groups are monitoring agency performance, or if news media are covering the issue. The more visible the issue, the more likely it is that political control will improve democratic responsiveness. Examples of highly visible policy choices embedded in administrative adjudication decisions include the FCC’s order imposing liability on television stations for broadcasting spoken expletives during the Billboard Music Awards, the FTC’s order holding Facebook liable for privacy violations involving deceptive claims about consumer control of personal data, and FDA’s orders directing Juul Labs, Inc. to stop selling and distributing their electronic nicotine products and delivery systems. Whatever the legal merits of these decisions, it is plausible that members of the public, news media, and interest groups will take note of them, and alert the electorate to hold the presidential administration accountable at the ballot box—although even these highly visible administrative decisions are unlikely to play more than a small role in public assessments of the president.

But it is utterly unrealistic to imagine that individual adjudicatory decisions on patent validity, immigration, or social security disability will have the kind of salience that facilitates politically accountable control over policy. For example, a decision on patent validity—and especially the basis for such a decision—is likely opaque to voters and unlikely to make it onto their list of concerns at the ballot box. Even with the guidance of interest groups, voters may be in no position to second guess administrative decisions about whether an invention meets patent law standards, or whether a drug is safe and effective.

At the same time, political control of these decisions can present distinct dangers to the faithful execution of the law. One long-recognized risk is that political supervision may threaten adjudicator impartiality, especially when one of the parties to the adjudication is the agency itself. The worry is that agency-head control may tempt the adjudicator to side with the enforcement officials in a case involving the agency. The SEC and the FCC have long faced criticism for the perception that they, including their insulated ALJs, too often side with their own enforcement divisions when violations are resolved administratively. When the agency is a party, this risk to impartiality has prompted calls for more, not less, independence and insulation of adjudicators from political control. For example, commentators have called for insulation of non-ALJ administrative judges that parallels the insulation currently required for administrative law judges.

More broadly, political control over individual decisions may make some adjudications vulnerable to inappropriate political influence in which “partisan politics undermin[es] the rule of law.” Professor Kate Andrias described President Nixon’s legendary efforts to deploy law enforcement for partisan ends, seeking Internal Revenue Service (IRS) audits of his political enemies’ taxes and directing the Attorney General to drop the government’s antitrust appeal against a Republican party donor, the International Telephone and Telegraph Corporation. More recently, President Trump was widely criticized for publicly condemning indictments of his political allies while urging the prosecution of political opponents.

Although political interference in enforcement proceedings may be particularly unseemly, political interference in adjudications of valuable rights such as patents or drug approvals presents similar risks of unfairness, lawlessness, and corruption. Now-Justice Kagan, in an otherwise strong defense of a muscular presidential role in administration, nonetheless acknowledged that adjudications call for a “fundamentally different” analysis because “[i]n this context, presidential participation in administration, of whatever form, would contravene procedural norms and inject an inappropriate influence into the resolution of controversies.”

The White House has adopted the general practice of refraining from contacting agencies either about specific enforcement actions or specific agency adjudications. Professor Vermeule has observed that a “network of tacit unwritten conventions” enforces this norm. Professor Andrias’s work favoring more direct presidential control of enforcement has similarly focused not on control of individual decisions, but rather on a transparent, coordinated effort to oversee overall Executive Branch enforcement direction, recognizing that control of individual decisions could prompt “unproductive” political involvement. In short, despite the formal presidential responsibility for enforcement actions, settled norms require political “hands off” individual enforcement decisions to reduce the potential for pernicious political influence. While advocates for agency-head control over adjudication have stopped short of arguing for presidential resolution of adjudication, review by Cabinet secretaries or agency heads close to the White House could pose many of the same risks.

As with the benefits of political control over adjudication decisions that raise broad policy concerns, the risks of inappropriate political influence are likely to vary. While public backlash may sometimes check inappropriate political intervention in high-visibility decisions, as with the Obama-era Plan B decision discussed above, the risks may be quite high when decisions are less visible.

Risks may be especially pronounced for low-visibility decisions with high financial stakes for well-funded and politically-connected interests. Political oversight of such decisions may give elected officials opportunities to help their friends or punish their enemies with little risk of exposure. This risk cautions against unbridled political control of adjudications concerning, for example, patents, government contracts, financial violations, or natural resource leases. Imagine a wealthy corporate patent holder facing a validity challenge (or bringing one) who finds it advantageous to bundle campaign contributions, or to fund a super PAC or lavish inaugural ball. The combination of technical complexity, low public visibility, and high financial stakes to the parties may make political control of these decisions distinctively vulnerable to inappropriate political interference or outright corruption. The same characteristics of low visibility and technical complexity undercut the potential for meaningful electoral accountability.

Political control might raise similar concerns for decisions of the Civilian Board of Contract Appeals within the General Services Administration, authorized to resolve individual government contract disputes with major corporations such as Dell, McKesson, and Xerox, or of adjudication processes for mineral leasing on public lands within the Department of the Interior. Consider, for example, the Department of the Interior’s treatment, in the 1980s, of the Navajo Nation’s coal leasing arrangement with Peabody Coal, a subsidiary of Peabody Group, then the largest coal mining company in the country. Tribal leasing arrangements are subject to federal approval, with an internal appeals process. Peabody had long paid what amounted to a 2% royalty rate, well below prevailing market rates. The Tribe invoked the internal adjudication processes in the Department of the Interior to set a fair market rate for the royalty. Three officials within the Department agreed, after considering multiple independent studies, that the Tribe should receive a royalty rate of 20% to bring it in line with market rates. The most senior official to review the decision was the Deputy Assistant Secretary for Indian Affairs, acting as the Commissioner of Indian Affairs. After Peabody officials learned of the decision, they hired lobbyists who made “numerous contacts” with Interior officials, including Interior Secretary Donald Hodel. Hodel ordered the Deputy not to inform the Tribe of his decision and instead to encourage the Tribe to continue negotiating. The economically strapped Tribe ultimately agreed to a far lower 12.5% royalty rate.

Agency heads may also face political pressure from outside the Executive Branch. For example, four New Jersey members of Congress brought enormous political pressure to bear on the 2008 decision of the FDA to approve a patch for injured knees called the Menaflex. The members of Congress had received significant campaign contributions from the manufacturer. In an extensive internal report, the Director of the FDA’s Office of Legislation described the “pressure from the Hill as the most extreme he had seen . . . and the agency’s acquiescence to the [applicant] Company’s demands for access to the Commissioner . . . as unprecedented.” In response to the pressure, the FDA Commissioner demanded an “expedited process [and] outcome in favor” of the manufacturer. The FDA’s scientific reviewers had already decided that the patch was unsafe because it often failed, requiring additional surgery, but the agency managers “overruled the scientists and approved the device for sale.” Political pressure to advance the interests of campaign donors is particularly worrisome when the public relies on an agency to make informed decisions affecting their health and safety that they are in no position to second-guess.

Requiring agency heads to announce such decisions publicly, and to explain their reasons, could improve transparency and provide a limited check on improper political influence. The Menaflex story did end up being publicly reported, and the FDA extensively analyzed and revisited both its process and the particular decision. More broadly, however, disclosure can only promote political accountability when it is thorough—which may sometimes be hampered by commercial or privacy interests in data or secrecy—and when the issue has enough salience to attract the attention of public monitors and voters. Requiring agency heads to disclose all of their meetings or other contacts with interested parties might be helpful, but it also might reduce the salience of problematic contacts by burying them in too much information, thus undermining rather than advancing accountability.

One final point: advocating across-the-board agency-head review of adjudication decisions may perversely discourage use of more politically accountable modes of agency policymaking. As discussed above, rulemaking provides advance public notice and broader public participation, and it is subject to more systematic White House supervision than adjudication. Thus, rulemaking generally resolves policy issues in a more transparent and accountable manner, in contrast to individual adjudications, where outside parties rarely participate. Guidance documents may also be superior to agency-head review of adjudication decisions because they are more visible and more participatory in their development. While agencies can elect adjudication as a flexible, low-cost means of developing policy in response to evolving conditions, normalizing agency-head control of individual adjudication decisions may undercut the agency’s incentive to use rulemaking, with the effect of diminishing overall political accountability for agency policy decisions.

In sum, Congress may have sound reasons for restricting agency-head review of individual adjudication decisions. Such restrictions should not be categorically prohibited, especially when agency heads may supervise adjudication in other ways, such as by issuing rules or guidance or by choosing which adjudication decisions to designate as precedential. Legislative choices might implicitly limit an agency’s discretion to select adjudication to make policy under the current administrative common law of Chenery II. But this doctrine has never been viewed as constitutionally compelled, and the burden is not substantial in any event. A legislative choice to restrict agency-head review of individual adjudication decisions is well within the range of discretion already exercised by the political branches in legislation that authorizes particular agencies to use only rulemaking or only adjudication.