Too often, Supreme Court opinions raise more questions than they answer. Those of us who painstakingly parse the Justices’ writings for insights often lament the explanations not offered. It would be so helpful if the Court’s opinions were more like law review articles, exhaus-tively articulating all the alternative lines of reasoning, surveying the counterarguments, justifying the superiority of the analytical path chosen, and perhaps throwing in a literature review as a bonus (thus giving bragging rights to the law professors whose work was cited). Sometimes we get those opinions, and then we criticize the shortfalls of their reasoning as well, and we complain about how longwinded the Justices have become.1
Professor Mila Sohoni is thorough and insightful in analyzing the limitations of and omissions from the reasoning in her titular “Quartet” of 2021–2022 Supreme Court decisions that advance the major questions doctrine as a limitation on the actions federal government agencies may take to resolve contemporary problems: Alabama Ass’n of Realtors v. Department of Health & Human Services,2 National Federation of Independent Businesses v. Department of Labor,3 Biden v. Missouri,4 and West Virginia v. EPA.5 Sohoni carefully traces the precedents from which the major questions doctrine emerged and its arguable evolution from Chevron deference exception to independent canon of statutory construction.6 She notes potential tensions with textualism and with other Court precedents presented by the new major questions canon.7 She contends that the Court neglected to provide an adequate theoretical justification for a major questions canon, whether as a variation on the canon of constitutional avoidance or as a clear statement rule under-girded by principles associated with the nondelegation doctrine and Article I, section 1 of the U.S. Constitution.8 Opinions from the Court acknowledging and addressing all of these concerns might have been helpful, perhaps, although one doubts that critics would be satisfied.
Yet Supreme Court opinions are not law review articles. Even Justices who share certain priors and premises, and who readily agree about the outcome of a case, may still disagree over the details and directions of the supportive reasoning. Substantial disagreements may yield concurring opinions, but the Justices often are willing to let lesser quibbles over dicta slide rather than write separately.9 Analytical omissions and ambiguous rhetoric in judicial opinions may represent strategic choices to make the compromises necessary to cobble together a majority.10
My goal with this Response is not to defend the Quartet as exemplars of judicial opinion writing. I am more sympathetic to the Roberts Court in this regard than many legal scholars, but I have my own quibbles with the Quartet and likely would have written the opinions differently, too. At a minimum, Sohoni is correct that the Court missed an opportunity to provide clearer guidance to lower court judges regarding the major questions doctrine. The Court’s failure to address at least some of Sohoni’s questions and concerns likely will lead to some amount of confusion and inconsistency among future lower court decisions.
Instead, this Response draws from Sohoni’s rather grudging acknowledgment of an argument with which she disagrees, that the Quartet reflects “merely responsible judicial minimalism . . . to cultivate the major questions doctrine on a rule-by-rule, statutory basis, gradually building out boundary lines that agencies must not cross, and to thereby allow the guardrails of nondelegation to accrete incrementally, in common law fashion.”11 In this manner, the Quartet and its reliance on the major questions doctrine align with a broader trend of Roberts Court decisions regarding separation of powers and administrative agencies: carefully narrow, calibrated to tweak the day-to-day of administrative governance incrementally, with plenty of carve outs and caveats, and with a preference for subconstitutional approaches, notwithstanding lofty flights of rhetoric about the Framers, liberty, and other constitutional values.12 Part of the Roberts Court’s incrementalism in separation of powers cases comes from its choice of remedies, such as its use of severability as a remedy for constitutional flaws in agency design decisions.13 But the Roberts Court’s doctrinal shifts in this area also are substantially more limited than they could be and seem designed more to accommodate rather than undermine the continued functioning of the administrative state — albeit with some new limitations. It’s a path designed to please almost no one, or at least not many of those who debate these issues most vocally. Regardless, for lower court judges seeking guidance for future cases, and perhaps inclined to apply the major questions doctrine aggressively, the overall message should be clear: Proceed With Caution!
I. Structural Formalism Meets Stare Decisis
Let us stipulate at the outset that, in addition to being more inclined toward originalist and textualist methods of interpretation, the Roberts Court also is more structurally formalist and more skeptical of agency action than any of its predecessors since at least the New Deal era.14 In 1978, renowned administrative law scholar Kenneth Culp Davis described formal separation of powers, rule of law, and nondelegation principles as “barriers to the development of the administrative process” and the modern administrative state (and judicial review thereof) as “focused primarily on particular substantive programs, not on broad and general conceptions about the nature and function of government or of law.”15 Irrespective of whether Davis accurately characterized the Supreme Court’s mid–twentieth century jurisprudence or merely read his own preferences into it, the Roberts Court by contrast takes seriously formalist conceptions of separation of powers, rule of law, and nondelegation principles. “Who decides?”16 matters at least as much as, if not more than, the substantive policies being pursued through agency action. The Roberts Court is troubled by judicial doctrines that have contributed to the gradual shift of governmental power away from Congress and the courts in favor of administrative agencies. After decades of Supreme Court decisions generally (although not always17) favoring structural functionalism and administrative flexibility,18 no one should be surprised that the pendulum is swinging the other way, with the Roberts Court shifting doctrine regarding the structural Constitution and administrative governance in a more formalist and less pro-agency direction.
It is equally obvious, however, that those among the Justices who are inclined toward structural formalism, originalism, etc., are not uniformly so.19 The Justices’ fealty to stare decisis and to judicial restraint also varies tremendously as regards constitutional questions. In Gamble v. United States,20 for example, Justice Alito defended stare decisis in constitutional interpretation as “promot[ing] the evenhanded, predictable, and consistent development of legal principles, foster[ing] reliance on judicial decisions, and contribut[ing] to the actual and perceived integrity of the judicial process.”21 According to Justice Alito, even when constitutional meaning is at issue, “a departure from precedent ‘demands special justification.’”22 By contrast, writing separately in the same case, Justice Thomas mostly rejected stare decisis in the constitutional context, explaining at length his view that the Court’s job is to interpret and apply the Constitution as written, not other Justices’ “erroneous” interpretations of it.23 In Ramos v. Louisiana,24 Justices Gorsuch,25 Sotomayor,26 Kavanaugh,27 Thomas,28 and Alito29 all authored opinions offering different ideas about stare decisis, when it applies, and what it requires.
Whether a particular Supreme Court decision actually follows stare decisis principles and adheres to precedent can be debatable. Holdings are precedential, while dicta is not, and which is which often is not clear.30 The Court may intend explanatory reasoning to guide lower courts in future cases, but not every judicial musing that catches the eye of commenters must be followed to the nth degree.31 On the other hand, today’s offhand summary of existing precedent may become tomorrow’s dominant two-part test.32 Interpretive canons and methods are not given precedential effect, but some argue that perhaps they should be.33
For that matter, if the Court purports to retain a precedent but narrows or recasts it in some way, whether under the guise of interpretation or otherwise, then has the Court really adhered to stare decisis?34 Administrative law is replete with examples of the Court interpreting its precedents in novel ways. In Kisor v. Wilkie,35 Justice Kagan preserved the doctrine of Auer or Seminole Rock deference to agency interpretations of regulations by transforming the simple and highly deferential “controlling weight unless . . . plainly erroneous or inconsistent with the regulation”36 into a much less deferential, five-stage inquiry, claiming precedent for every step along the way.37 Lest one think this kind of precedential retheorizing is unique to the present day, consider the 1935 case of Humphrey’s Executor v. United States,38 wherein the Court dismissed most of Chief Justice Taft’s seventy-one-page defense of an unfettered presidential removal power in Myers v. United States39 as “dicta” to uphold statutory restrictions on the President’s authority to remove Federal Trade Commissioners from office.40 Likewise, in Morrison v. Olson41 in 1988, the Court upheld statutory restrictions on the removal of independent counsels by interpreting away inconvenient reasoning from Humphrey’s Executor while purporting to respect that precedent.42
The Quartet’s approach to the major questions doctrine arguably reflects this sort of recasting of precedent in the name of adhering to it. In West Virginia v. EPA, Justice Kagan complained that “[t]he Court has never even used the term ‘major questions doctrine’ before.”43 That objection is not quite accurate. Justices had used the label previously in concurring and dissenting opinions.44 In doing so, the Justices merely picked up what legal scholars and others in the legal community for many years had dubbed the strand of Chevron and statutory interpretation jurisprudence on which the Court relied.45 Further, members of the Court have raised concerns for decades about agencies pushing the interpretive limits of ambiguous statutory delegations to pursue aggressive policy agendas, whether labeled as “jurisdictional” questions,46 as “elephants in mouseholes,”47 or otherwise.48 In King v. Burwell,49 Justice Kagan herself joined the Court’s majority in rejecting the availability of Chevron deference and claiming for itself de novo review of an Internal Revenue Service interpretation of the Affordable Care Act based in no small part on the interpretation’s “economic and political significance,”50 although the Court in that case sided with the agency.51
Justice Kagan’s broader point in West Virginia v. EPA, advanced as well by Sohoni, is closer to correct — that the version of the major questions doctrine advanced in the Quartet seemed meaningfully different from those precedents.52 How much so, and whether the shift falls outside the normal parameters of doctrinal evolution and stare decisis as described above, is where at least some of the scholarly debate over the major questions doctrine will lie.
II. Paths Proposed, Paths Not Taken
Beyond legal theory, what does it mean in practical terms for the Roberts Court to be more formalist, more originalist, and more skeptical of administrative agencies than its predecessors? If they truly wanted to pursue dramatic change and wreak havoc upon the administrative state, the Justices have no shortage of options. Legal scholars have been proposing doctrinal alternatives to overhaul the administrative law status quo for a long time. The Court could reinvigorate the nondelegation doctrine by replacing the intelligible principle standard with a sweeping and categorical prohibition against all agency regulations carrying the force of law or all regulations issued pursuant to statutory grants of general rulemaking authority.53 The Court could repudiate Chevron deference outright, calling into question the validity of thousands of agency statutory interpretations upheld by courts over the past four decades.54 Upon finding a constitutional deficiency in an agency’s structure, the Court could effectively nullify the agency’s existence, or at least its capacity to act in legally consequential ways, until Congress passes new legislation reconstituting the agency to satisfy constitutional constraints.55 The Roberts Court has done none of these things, and not for lack of opportunity.
A. The Nondelegation Doctrine Is Still (Mostly) Dead
The nondelegation doctrine holds that Article I, section 1 of the Constitution vests in Congress the legislative powers “herein granted” and that Congress may not delegate those legislative powers to the executive branch (or anyone else).56 For decades now, the nondelegation doctrine has been moribund.57 Since at least the 1970s, legal scholars and others have called for replacing the intelligible principle standard and reinvigorating the nondelegation doctrine.58 Notwithstanding several opportunities since then,59 the Court has not invalidated a statute on nondelegation grounds since 1935.60
Before resolving the Quartet, a majority of the Justices now on the Court had signaled interest in revitalizing the nondelegation doctrine. In Gundy v. United States,61 Justice Gorsuch outlined a proposed alternative to the intelligible principle standard and was joined by Chief Justice Roberts and Justice Thomas.62 Justice Alito concurred in Gundy’s outcome but called for reconsidering the intelligible principle standard on another occasion.63 In Paul v. United States,64 Justice Kavanaugh — who did not participate in Gundy — authored a statement in which he expressed a similar desire to rethink the intelligible principle standard and spoke favorably of Justice Gorsuch’s alternative approach.65 Justice Kavanaugh stopped short of endorsing Justice Gorsuch’s Gundy opinion outright, however. Instead, he spoke of “major policy question[s]” and cited some of the precedents from which the Quartet’s version of the major questions doctrine derives.66
At no time in its post–New Deal history has the Supreme Court seemed more likely to bring the nondelegation doctrine back to life than heading into the 2021 Term. As the lone case among the Quartet in which the Court actually granted certiorari and heard oral argument, West Virginia v. EPA was the most anticipated vehicle for revitalizing the nondelegation doctrine by replacing the intelligible principle standard with a more robust limitation on congressional delegations of policy-making authority to agencies.67 Pursuing that route, as the Justices seemed poised to do, would be a substantial doctrinal shift for the Court.68 It would overturn a longstanding and consistently applied legal standard and raise questions about the constitutional validity of many statutes previously upheld by the Court based on intelligible principles, including the Clean Air Act69 and the Occupational Safety and Health Act70 — although it is worth noting that Justice Gorsuch in Gundy took care to say that “[a]t least some” (and perhaps many) of those precedents might survive his proposed alternative as well.71
But the Court did not overturn the intelligible principle standard in West Virginia v. EPA. In fact, the Court did not approach any of the Quartet as a matter of constitutional interpretation at all, embracing instead the major questions doctrine as a subconstitutional canon of statutory interpretation.
Perhaps as a result, the vocal enthusiasm of some of the Justices for replacing the intelligible principle standard seems to be on hold. Even Justice Gorsuch, arguably the most vocal proponent of replacing the intelligible principle standard,72 did not suggest the nondelegation doctrine as an alternative basis for the Court’s holding in West Virginia v. EPA. Instead, in his concurrence, Justice Gorsuch merely claimed nondelegation and separation of powers principles as the conceptual bases for major questions as a substantive canon of statutory interpretation.73 He then dedicated most of his concurrence to providing a nonexclusive list of circumstances in which the Court might apply that major questions canon and “several telling clues” for “what qualifies as a clear congressional statement authorizing an agency’s action,”74 only some of which were suggested by the majority opinion.75 As restrained (for him) as Justice Gorsuch was in this regard, only Justice Alito joined him in pushing even so robust a vision for a major questions canon as a limitation on congressional delegations of policymaking discretion to agencies.76
Beyond leaving undisturbed almost a century of precedent applying the intelligible principle standard, the Quartet’s reliance on the major questions doctrine is distinct from reinvigorating the nondelegation doctrine in one very important respect. Declaring that a statutory delegation violates the nondelegation doctrine, and thus the Constitution, would preclude Congress from ever relying upon agency officials to resolve the associated policy matters absent a constitutional amendment. By comparison, with the Quartet and the major questions doctrine, the Court held that Congress, as a matter of statutory interpretation, did not give agencies the necessary authority to adopt the regulations at issue, not that Congress, as a matter of constitutional interpretation, could not give agencies that authority.
This distinction is important. To be sure, congressional action on controversial policy questions is always challenging, and the Court has not provided Congress with magic words or otherwise offered extensive guidance regarding just how much more statutory detail it expects Congress to provide to avoid the major questions doctrine. Nevertheless, persuading Congress to amend a statute remains an easier proposition than navigating the constitutional amendment process. And amending a statute to delegate more clearly to an agency the requisite rulemaking power to address a particular contemporary problem, perhaps with some amount of guidance regarding the tools Congress would like the agency to use or approaches that Congress would like the agency to follow, is much easier than amending the same statute with the specificity that would be necessary if the Court required Congress to resolve all of “the policy decisions” itself and leave only “the details” to agencies (as would be the case if the Court replaced the intelligible principle standard as Justice Gorsuch proposed in Gundy).77
Irrespective of the doctrinal effect of replacing the intelligible principle standard, I have argued elsewhere that the practical impact of such a move on administrative governance likely would be less dramatic than some expect for two reasons: (1) the sheer number and varying types of delegations contained in the U.S. Code, and (2) the incremental, case-by-case, provision-by-provision, statute-by-statute alternatives that Justice Gorsuch and others on the Court seemed to favor over more sweeping and categorical approaches suggested by some scholars.78 Any standard that evaluates delegations individually, one at a time, would simply be too limited in its reach to curtail either congressional delegations or agency policymaking very much. Just as most agency actions never see the inside of a courtroom,79 the same would be true of most delegations. As applied in the Quartet, the major questions doctrine similarly anticipates an individualized analysis of statutory text. Thus, whether it functions as an exception from Chevron deference or as an independent canon of statutory interpretation, the major questions doctrine has these same practical limitations. Agencies may be less inclined to pursue the very largest and most boundary-pushing regulation projects given the major questions doctrine, but such restraint still leaves agencies with a tremendous range of policymaking discretion.
Finally, as described by the Court in West Virginia v. EPA and applied throughout the Quartet, the major questions doctrine is not so standardless as some detractors suggest. Apart from the possibilities described by Justice Gorsuch’s concurrence in West Virginia v. EPA, Chief Justice Roberts for the Court suggested several guideposts and guardrails for applying the major questions doctrine, including (1) “the ‘history and the breadth of the authority that [the agency] has asserted’”;80 (2) “the ‘economic and political significance’ of that assertion”;81 and (3) the extent to which the agency is relying on “‘modest words,’ ‘vague terms,’ or ‘subtle device[s]’” rather than more direct delegations from Congress.82 Put more succinctly, one might say based on Chief Justice Roberts’s opinion that the major questions doctrine applies to curtail administrative discretion when an agency stretches the boundaries of statutory interpretation to claim new authority to address big problems that, previously, were not obviously within the agency’s purview.
To be sure, this description of the major questions doctrine is a mushy standard rather than a bright-line rule, which makes its application more subjective and uncertain than admirers of the administrative state might prefer. The drawbacks of standards are well known, including a lack of certainty ex ante and a potential for inconsistency of application.83 Yet, bright-line rules have their own drawbacks, and mushy standards are not new to constitutional law.84
None of this is to suggest that the evolution and application of the major questions doctrine is no big deal. The Quartet undoubtedly will prompt agency officials to restrain their rulemaking impulses on some occasions, especially if they are unsure of their authority to act. Courts can anticipate a new set of challenges to agency actions on major questions grounds. Nevertheless, the subconstitutional major questions doctrine is substantially more restrained and incremental, and arguably more consistent with stare decisis principles, than the constitutional alternatives on offer.
B. Chevron Still Lives
Like the nondelegation doctrine’s intelligible principle standard, the Chevron doctrine has been under attack and, arguably, in decline. Before his passing, Justice Scalia — long one of Chevron’s staunchest proponents85 — raised doubts regarding Chevron’s underpinnings.86 Since then, Justice Thomas has questioned whether Chevron deference violates constitutional separation of powers principles.87 Justice Gorsuch, too, has advocated abandoning Chevron deference, characterizing it as “permit[ting] all too easy intrusions on the liberty of the people.”88 Justice Kavanaugh also has called Chevron “an atextual invention by the courts” and “a judicially orchestrated shift of power from Congress to the Executive Branch” that “encourages the Executive Branch (whichever party controls it) to be extremely aggressive in seeking to squeeze its policy goals into ill-fitting statutory authorizations and restraints.”89
The Court also has not applied Chevron to defer to an agency interpretation of a statute since 2016,90 notwithstanding several opportunities to do so.91 Some scholars have pointed to the Court’s failure to cite Chevron in these cases as evidence that the doctrine is in decline, if not outright defunct.92 As James Kunhardt and Professor Anne Joseph O’Connell have documented, agencies may be adjusting their rule-making practices in response to the uncertainty created by the Court’s silence.93 Sohoni concludes that the Quartet has separated the major questions inquiry from Chevron analysis, and thus labels Chevron “a derelict.”94 Perhaps the Court one day will be explicit that it has rejected Chevron.
Or maybe it won’t. Many commentators anticipated that the Court might overturn Chevron deference outright in its 2021 Term,95 either in American Hospital Ass’n v. Becerra96 (where briefing expressly addressed whether the Court should overturn Chevron97., Brief Amicus Curiae of Pacific Legal Foundation in Support of Petitioners at 7, Am. Hosp. Ass’n (No. 20-1114), 2021 WL 4135099, at *7 (“[T]his Court should take this opportunity to reconsider Chevron deference.”); Brief of Amicus Curiae Americans for Prosperity Foundation In Support of Neither Party at 4, Am. Hosp. Ass’n (No. 20-1114), 2021 WL 4173461, at *4 (“The Court should squarely overrule Chevron here and now.”); Brief for Amicus Curiae the Chamber of Commerce of the United States of America in Support of Neither Party at 3, Am. Hosp. Ass’n (No. 20-1114), 2021 WL 4219174, at *3 (encouraging the Court at least to “take this opportunity to restate, clarify, and reinforce the limitations and requirements on Chevron deference”); Brief for the Respondents at 19–20, Am. Hosp. Ass’n (No. 20-1114), 2021 WL 4937288, at *19–20 (claiming Chevron deference). and the issue was discussed at oral argument98) or, alternatively, in Becerra v. Empire Health Foundation99 (where the government also asserted Chevron deference100). That did not happen. Instead, the Court again was merely silent regarding Chevron as it evaluated agency statutory interpretations by performing its own independent analysis using traditional tools of statutory interpretation as suggested by Chevron’s footnote nine.101
Ignoring precedent where it might apply can foreshadow the Court’s eventually overturning or abandoning that precedent.102 But silence does not a holding make. By merely remaining silent rather than overruling Chevron outright, the Court has preserved the doctrine’s availability for future cases.
Chevron has always had a more substantial effect on case outcomes in the lower courts than at the Supreme Court.103 The Justices may have cited Chevron in 238 cases from 1984 through the 2019 Term,104 but for all the hype, the Court only applied the Chevron standard to evaluate an agency’s interpretation of a statute in 107 cases105 and only deferred to the agency under Chevron in 72 of them.106 As Professor William Eskridge and Lauren Baer documented, from 1984 through the end of the 2005 Term, the Court applied Chevron in only 8.3%, and relied purely on its own reasoning without resort to any deferential canon, presumption, or standard in more than 50%, of 1014 cases evaluating agency statutory interpretations.107 Even when they limited their analysis to cases in which they felt Chevron clearly should apply, Eskridge and Baer found that the Court applied Chevron in at most 30% of cases.108 Professors Kent Barnett and Christopher Walker similarly have interpreted these findings to mean that, even before the Court’s present silence, its “questionable loyalty to Chevron suggests that the doctrine is not meant to discipline Supreme Court decisionmaking. Instead, the doctrine may better serve to control lower courts and provide nationwide uniformity.”109 In this vein, the government con-tinues both to claim and to receive Chevron deference in the lower courts.110
Regardless, although Sohoni asserts the contrary,111 the Supreme Court’s silence regarding Chevron in the Quartet does not necessarily prove that doctrine’s demise nor separate the major questions doctrine from its scope.112 The Court’s Chevron jurisprudence anticipates three steps: first, the Mead inquiry of whether Congress delegated authority to act with the force of law and the agency acted in the exercise of that authority;113 second, the Chevron step one inquiry of whether the statute’s meaning is clear or ambiguous;114 and third, the Chevron step two inquiry of whether the agency’s interpretation of an ambiguous statute is reasonable.115 The Court has never been consistent, however, in ordering those steps. Whereas the Court often has applied Mead’s inquiry as a “step zero” that comes first,116 other of the Court’s opinions have approached Mead as a “step one-and-a-half” that comes only after deciding that the statute’s meaning is ambiguous.117 Still other Court decisions and academic examinations rearrange or blend the steps in different ways.118 As the Court has observed repeatedly, Chevron step one entails the application of “traditional tools of statutory construction” to discern congressional intent, much like de novo review.119 Whether under the step one-and-a-half model or otherwise, it has been entirely common for the possibility of deference to come into play only after the Court decides for itself that traditional tools fail to resolve statutory meaning. Under such circumstances, the Court has no need to cite Chevron until such time as it decides that traditional tools of statutory interpretation fail to resolve statutory meaning. Throughout Chevron’s tenure, the Court’s mere silence regarding Chevron in one statutory interpretation case has offered little to no hint of the likelihood (or lack thereof) that the Court might apply Chevron in a subsequent case. The hostility that some of the Justices have shown toward Chevron deference simply gives the Court an additional reason to avoid mentioning Chevron if the deference question can be avoided by resolving statutory meaning using traditional tools.
Additionally, to the extent the Quartet casts the major questions doctrine as a substantive canon of statutory interpretation, it falls in the midst of the gray area of how substantive canons square with Chevron analysis, including whether they should be taken into account at Chevron step one or Chevron step two. Legal scholars have debated that question for some time,120 while the Supreme Court just has been inconsistent in its treatment. In some cases, the Court has applied the constitutional avoidance canon to bolster findings of statutory clarity and reject the agency’s interpretation at Chevron step one.121 On other occasions, however, the Court has postponed or outright rejected consideration of constitutional avoidance until Chevron step two, after a finding of statutory ambiguity.122 Absent a more explicit statement, little in the Quartet suggests that the major questions doctrine as a substantive canon would fare differently.
The Supreme Court’s approach to Chevron is and always has been malleable. One must take care not to read too much into what the Court does not say outright. The Court’s silences allow it to reserve the availability of Chevron deference for future cases. The Court also has preserved for itself whether and how the major questions doctrine as developed in the Quartet might interact with Chevron. Thus, irrespective of occasional sweeping rhetoric, the Roberts Court’s actual doctrinal pronouncements regarding Chevron are more restrained and incremental than dramatic and pivotal.
C. Agencies Soldier On
Agency design represents yet another area in which the Roberts Court has moved its separation of powers jurisprudence in a more formalist and less pro-agency direction, at least as a matter of doctrinal theory. Congress often prescribes by statute who will appoint the officers of the United States who in turn perform agencies’ legally critical rulemaking and adjudication functions; Congress also often imposes statutory limitations on the authority of the President (or other higher-ranking officers) to remove those officers from office.123 Precisely who is an officer of the United States and whether a particular officer is an inferior officer whose appointment Congress can vest in the President alone, the head of a department, or a court of law can be challenging questions given dozens of agencies and myriad government officials exercising varying types and degrees of authority. For most of the twentieth century, the Supreme Court was willing to go along with congressional preferences regarding how to structure an agency’s leadership. Such is no longer the case.
Unitary executive theory — the idea that the Constitution’s Article II requires “all federal officers exercising executive power [to] be subject to the direct control of the President,” including those at independent agencies124 — often goes hand in hand with structural formalism,125 although some functionalists support the argument as well.126 Consistent with unitary executive theory, one hallmark of Roberts Court separation of powers jurisprudence has been its whittling away of statutory limitations on the President’s authority to remove federal officers from office. In this way, the Court has substantially reduced Congress’s ability to pursue its own preferences for how federal agencies are structured. Yet it has done so while claiming quite explicitly that it is not overturning precedent,127 despite calls from some of the Justices to do precisely that.128 Instead, the Court generally has pursued an approach of recasting and minimizing its precedents to render them more compatible with the Court’s preferred direction. But the Court also has been enormously creative at fixing agencies’ constitutional infirmities while leaving the agencies otherwise intact to continue their work.
For example, in Free Enterprise Fund v. Public Co. Accounting Oversight Board,129 the Roberts Court declared that two layers of statutory for-cause removal restrictions for members of the Public Company Accounting Oversight Board (PCAOB) violated separation of powers principles.130 The Court resolved the constitutional difficulty, however, by simply severing one of the two statutory provisions and remanding the case for reconsideration by a now-constitutional agency.131 The accounting firm that challenged the PCAOB’s structure won its case before the Supreme Court, but its disagreement with the agency persisted for several months until the parties settled with the agency closing its investigation but reserving the authority to pursue future investigations of the accounting firm’s “past or future conduct.”132
In each of Seila Law v. Consumer Financial Protection Bureau133 and Collins v. Yellen,134 the Roberts Court held that an independent agency headed by a single director — in Seila Law, the Consumer Financial Protection Bureau (CFPB); in Collins, the Federal Housing Finance Agency — violated separation of powers principles.135 Again, however, the Roberts Court resolved the agencies’ structural problems with at best minimal interference.136 Given the presence of complicating factors in Collins, the Court remanded that case for the lower courts to decide what the remedy for the constitutional violation should be.137 In Seila Law, however, the Court again simply severed a statutory for-cause removal restriction, rendering the CFPB’s Director removable at will by the President and the agency no longer independent. The Court in Seila Law declined to dismiss outright the matter that gave rise to the case in the first instance, a civil investigative demand issued by the CFPB seeking documents and other information from Seila Law.138 Instead, the Court remanded the case, after which the Ninth Circuit decided that the CFPB’s demand was properly ratified later by a constitutionally accountable Director.139 In other words, although Seila Law won its case before the Supreme Court and the CFPB’s structure was altered to make its Director removable at will, the CFPB’s performance of its statutory functions — including its investigation of Seila Law — continued unabated.
Finally, United States v. Arthrex, Inc.140 ended with the Roberts Court using severability a little differently to align an agency’s structure more closely with unitary executive theory while simultaneously minimizing the impact on the agency’s performance of its statutory functions. In Arthrex, the Court determined that the Patent Trial and Appeal Board (PTAB) was unconstitutionally structured because administrative patent judges (APJs) made final decisions that no one else in the executive branch, including the Director of the Patent and Trademark Office, could review and overturn.141 The obvious conclusion from the Court’s analysis was that the APJs were principal officers who must be appointed by the President with the advice and consent of the Senate. The Court avoided that holding, however, on the ground that it was enough to say that “[o]nly an officer properly appointed to a principal office may issue a final decision binding the Executive Branch in the proceeding before us.”142 The Court then severed language in the statute preventing the Director from reviewing and overturning PTAB decisions, thus giving the Director sufficient supervisory authority over APJs to satisfy constitutional requirements by making APJs inferior officers, and otherwise allowing the PTAB to continue its work.143
In other work, I have questioned whether the Roberts Court’s willingness to use severability to rewrite statutory text is as restrained as might appear at first blush.144 Nevertheless, one cannot help but notice that, with those small statutory adjustments, the Roberts Court allowed the agencies in question to proceed in the day-to-day administration of their statutory responsibilities with nary a hiccup.
III. What Happens Next?
Along with the Roberts Court’s other administrative law cases, the Quartet brings some amount of change, but not the radical overhaul that some demand and others decry. Indeed, notwithstanding its trend toward greater structural formalism, the Roberts Court seems quite determined not to interfere too much with the overall functionality of the administrative state. Obviously, these decisions are dissatisfying to supporters of administrative governance and advocates of the policies advanced by the invalidated regulatory actions. They are also frustrating to critics of administrative governance who want more sweeping change.
The limitations of and omissions from the Quartet’s reasoning bring some amount of uncertainty regarding how to apply the major questions doctrine. Uncertainty leads to inconsistency.145 Some highly motivated judges will undoubtedly overread the Quartet, just as they will the Roberts Court’s other administrative law opinions, and push for more change than the Court has embraced. Other judges likely will underread the Court’s opinions and not apply them aggressively enough for the Court.
We also have not seen the last of the Roberts Court in any of these areas. Small changes in the short term are not incompatible with more significant ones later, nor with substantial upheaval over time. Today’s incremental moves could become tomorrow’s major alterations in the functionality of the administrative state. Of course, if the Court is deliberately pursuing major change on an incremental basis, the passage of time makes the possibility of reaching the end goal (whatever it may be) less likely as political and personnel changes occur.
One suspects the Court lacks the votes for grander doctrinal pronouncements, but we cannot know with certainty whether the Roberts Court’s incrementalism regarding separation of powers and agency action arises from genuine disagreement over how far doctrinal change should go or merely how best to get to a particular objective. The Court still could replace the intelligible principle standard,146 repudiate Chevron deference,147 and invalidate the very existence of some number of government agencies148 — thus reducing the ability of Congress to rely on agencies to accomplish policy priorities more than the Court has thus far. In the meantime, by shifting doctrine incrementally on a case-by-case basis without doing a more thorough job of explaining and justifying those moves and where they might be headed, the Roberts Court may run the risk of painting itself into a corner that it does not like and from which it has a hard time escaping.
Nevertheless, expecting the Roberts Court simply to maintain the doctrinal status quo despite changing times and Court personnel is ahistorical. With any movement in judicial doctrine, whether great or small, some amount of uncertainty and a whole lot of criticism will be inevitable, no matter how long and how thorough the Court makes its opinions. The real question all along has been how much change, not whether any would occur. At least for now, on administrative law issues, and as regards the day-to-day reality of administrative governance, that change is pretty limited.
* McKnight Presidential Professor in Law, Distinguished McKnight University Professor, and Harlan Albert Rogers Professor in Law, University of Minnesota Law School. I want to thank Jonathan Adler, Nick Bednar, Betz Bentley, Emily Bremer, Dick Pierce, Mila Sohoni, and Chris Walker for helpful comments and conversations and the student editors at the Harvard Law Review for their patience, professionalism, and hard work.