The Yale Law Journal


The Jurisprudence of “Degree and Difference”: Justice Breyer and Judicial Deference

21 Nov 2022

abstract. In 1986, then-Judge Stephen Breyer published an article, Judicial Review of Questions of Law and Policy, advocating a nuanced approach to judicial review of agency statutory interpretations. He was writing to counter the emerging understanding of Chevron as imposing a single, simple rule of judicial deference to agency interpretations of ambiguous statutory provisions. In the article, he outlined an alternative approach that treats judicial deference as “a matter of degree, not kind.” Depending on the particular legal question and statute at issue, a court might accord an agency interpretation binding deference, some deference, or no deference at all. While his approach would vary by case, he identified certain context-based factors for courts to consider in determining the appropriate degree of deference. The overarching aim of his approach was to best approximate when Congress wishes, and when it makes sense, for courts to defer to agency interpretations.

This Essay shows that Justice Breyer’s context-specific approach has prevailed in the Supreme Court’s decisions to a greater extent than standard accounts of judicial deference acknowledge. The Court has traditionally presented Chevron as the dominant deference doctrine and subsequent doctrines as limited exceptions or qualifications to its two-step test. But now that the doctrines have settled, it is possible to see that the Court has developed different degrees of deference, with a thumb on the scale for Chevron-level deference. Furthermore, the doctrines themselves reflect the same factors and basic aim that Breyer emphasized in his early scholarship. He did not get everything he wished for, but his approach carried the day more than it might initially seem.

As Justice Breyer retires from the Court, the new conservative majority is looking to impose another simple rule—no deference—via the major questions doctrine. This Essay offers a prediction about the likely success of this new rule based on observations that Breyer made in his 1986 article. From his vantage point as a judge, he saw that interpretive questions arise with too many substantive and procedural complexities for lower courts to address without taking account of context-based factors. Consequently, he speculated that courts would gravitate toward a context-specific approach, finding ways to accommodate different degrees of deference, despite the rule that the Court announced. The institutional realities that then-judge Breyer identified have not changed with the membership of the Court or the direction of the simple rule. In fact, a no-deference rule will only intensify the need for courts to locate other legal principles that justify deferring to agency resolutions of interpretive questions they cannot honestly answer otherwise. Thus, the majority’s new rule is ultimately unlikely to succeed where it matters most: in the lower courts. For as long as there are regulatory statutes and agencies to implement them, institutional pressure will likely drive courts toward a Breyer-esque approach.


In 1986, then-Judge Stephen Breyer published an article advocating a nuanced approach to judicial review of agency statutory interpretations.1 Breyer wrote that article, Judicial Review of Questions of Law and Policy, to counter the emerging understanding of the Supreme Court’s new decision in Chevron as imposing a “single, simple rule” of judicial deference to agency interpretations of ambiguous statutory provisions.2 He argued that the decision need not be understood so rigidly and warned that such a rule would be “a greater abdication of judicial responsibility to interpret the law than would seem wise, from either a jurisprudential or administrative perspective.”3 As an alternative, he outlined a context-specific approach that would treat judicial deference as “a matter of degree, not kind.”4 Depending on the particular legal question and statute under review, a court might accord an agency interpretation binding deference, some deference, or no deference at all. Although his suggested approach would vary by case, Breyer identified general context-based factors for courts to consider in determining the appropriate degree of deference.5 The overarching aim of his approach was to best approximate when Congress wishes, and when it makes sense, for a court to defer to an agency interpretation.6

While Breyer’s article is well known and has been cited approvingly by the Supreme Court,7 few believe that his context-specific approach has carried over to the law.8 This Essay shows the contrary—namely, that his approach has prevailed in the Supreme Court’s decisions more than standard accounts of judicial deference acknowledge. Those accounts have taken their cue from the Court, which has presented Chevron as the dominant doctrine governing judicial deference to agency interpretations.9 To be sure, the Court has issued other important decisions in this area. For example, United States v. Mead Corp. makes Chevron deference available only if Congress has authorized the agency to issue interpretations “with the force of law” and the agency has utilized such authority to issue the relevant interpretation.10 Likewise, Food & Drug Administration v. Brown & Williamson Tobacco Corp. withholds Chevron deference from interpretations of great “economic and political significance.”11 But the Court has characterized these subsequent decisions as providing mere qualifications or exceptions to Chevron’s operation.12 Though this impression accurately describes how the decisions appeared over time, it is misleading. Now that the decisions have settled, it is possible to see that the Court has not developed a lead doctrine and a supporting cast; it has instead created different degrees of deference, with a thumb on the scale for Chevron-level deference. Furthermore, the doctrines themselves, which largely emerged after Breyer joined the Court, reflect the same factors and achieve the basic aim that he set out in his earlier scholarship. Breyer did not get everything he wished for—most notably, a genuinely “blended” approach to judicial review.13 Still, he helped build a doctrinal approach that—as he remarked midway through his tenure on the Court—“[a]lthough seemingly complex . . . has proved a workable way to approximate how Congress would likely have meant to allocate interpretive law-determining authority between reviewing court and agency.”14

As Justice Breyer retires from the Court, this context-specific approach is breaking down. A new conservative majority is looking to impose another simple rule—no deference—via the major questions doctrine.15 This Essay makes a prediction about the likely success of this rule based on observations that Breyer offered in his earlier scholarship. From his vantage point as a judge, he saw first-hand that interpretive questions arrive with an endless variety of substantive and procedural complexities that courts cannot address without taking account of context-based factors.16 Consequently, he speculated that courts would gravitate toward a context-specific approach, finding other ways in the law to accommodate different degrees of deference, despite the rule that the Court announced.17 The institutional realities that Breyer identified have not changed with the membership of the Court or the direction of the rule. In fact, a no-deference rule will only intensify the need for courts to locate other principles of statutory interpretation or administrative law that justify deferring to the agency’s resolution of the legal questions they cannot honestly answer otherwise. Thus, the Court’s new rule is unlikely to ultimately succeed where it matters most: in the lower courts. For as long as there are regulatory statutes and agencies to implement them, institutional pressures will likely drive courts toward a context-specific approach. And Breyer’s sensibility will continue to guide judicial review.

This Essay has three Parts. Part I shows how the Supreme Court’s decisions can be understood to reflect Justice Breyer’s context-specific approach to judicial deference. Part II addresses how that approach is breaking down under the Court’s new conservative majority. Finally, Part III offers a prediction about the future of judicial deference in the lower courts, drawing on Breyer’s early scholarly insights.

I. breyer’s context-specific approach and the court’s doctrine

Viewed collectively and individually, the Supreme Court’s decisions can be understood to establish a Breyer-esque approach to judicial deference. The strongest indication of that approach lies in the very existence of important deference decisions other than Chevron. As a descriptive matter, it may seem that the Court maintains a dominant deference doctrine—Chevron—and certain limited doctrines or principles within “Chevron’s domain.”18 This account captures how the subsequent doctrines and principles emerged: they rolled out piecemeal over time as elaborations of or revisions to Chevron. But now that they are established, the picture looks quite different. It is possible to see that the Court maintains a constellation of deference doctrines and principles, in which Chevron’s star simply shines the brightest. The other doctrines and principles are the familiar ones—for example, the major questions doctrine (no deference on questions of economic and political significance),19 the Mead doctrine (no deference absent indicia that the agency used binding lawmaking authority),20 and Skidmore deference (respect for agency expertise on a question even in the absence of lawmaking authority).21 But perspective matters when characterizing the law. And, taken together, the Court's deference decisions can be understood to achieve Breyer’s basic aim of determining case by case whether Congress likely intended for an agency to resolve a given interpretive question, rather than inferring such an intent from the statute’s ambiguity. The persistent image of “Chevron and its progeny,” even if historically accurate, tends to obscure this updated view of judicial review.22

The Court’s deference decisions also manifest Justice Breyer’s approach in a more specific way—they incorporate the factors that he identified in his article. He wrote that courts should consider whether:

(a) “the agency has special expertise that it can bring to bear on the legal question”;

(b) “the particular question [is] one that the agency or the court is more likely to answer correctly”;

(c) the question is a “major” one that Congress might wish to answer rather than an “interstitial” one arising “in the course of the statute’s daily administration”;

(d) the statutory language is “inherently imprecise” or broad in a way that “invite[s] agency interpretation”;

(e) the answer to the question will “clarify, illuminate or stabilize” the law;

(f) “the agency can be trusted” not to suffer from “tunnel vision,” which tends to make agencies “expand their power beyond the authority that Congress gave them”; and

(g) the agency has addressed the question in a manner, both substantively and procedurally, that Congress would intend to be binding.23

According to Breyer, this nonexhaustive list of factors would help resolve whether courts should give an agency interpretation full, partial, or no deference.24 Breyer did not invent these factors as some perfected vision of the administrative state. Rather, he collected them from caselaw and experience as the right ones to approximate congressional intent on the deference issue.25

These factors are evident in the Court’s main deference decisions. Concurring in City of Arlington v. Federal Communications Commission in 2013, Justice Breyer sorted almost three decades of post-Chevron decisions into groupings that track many of these factors.26 But there is no need to take his word for it. Consider, as the following Sections do, some of the decisions that he included in his list, as well as more recent decisions that illustrate the same point.

A. Major Questions, Trustworthy Agencies, and Correct Interpretations

Start with a pair of decisions, one from the Rehnquist Court and one from the early Roberts Court, said to establish the major questions doctrine: Brown & Williamson27 and King v. Burwell.28 These are not the only major questions decisions from the Breyer era (also notable, for example, are Gonzales v. Oregon and Utility Air Regulatory Group v. Environmental Protection Agency (EPA)).29 But they exemplify the Breyer connection.

In Brown & Williamson, the Court held that the Food, Drug, and Cosmetic Act (FDCA) did not grant the Food and Drug Administration (FDA) authority to regulate nicotine and cigarettes as “drugs” and “devices,” in part because other legislation—protective of the tobacco industry—constrained the FDA’s regulatory approach.30 Moreover, subsequent legislation that specifically addressed tobacco products “effectively ratified” the FDA’s prior denial of its own jurisdiction.31 This statutory evidence led the Court to declare that the statute precluded the FDA’s nicotine and cigarette regulations. But it was another consideration—the nature of the question—that confirmed this conclusion. The Court said that “Congress could not have intended to delegate a decision of such economic and political significance in so cryptic a fashion.”32 Justice O’Connor, who wrote for the majority, attributed this consideration to Justice Breyer, citing his 1986 article and quoting his discussion of “important” and “major” questions.33

While the Court expressly invoked the major questions consideration, its decision also can be understood to reflect another factor from Breyer’s article: whether the agency could be “trusted” to resist “tunnel vision,” which tends to make agencies “expand their power beyond the authority that Congress gave them.”34 The Court intimated that the FDA’s regulations resulted from a kind of tunnel vision—or at least blinding urgency—brought on by the mounting evidence that tobacco products caused adverse health effects.35 (The agency was perhaps also motivated by the priorities of a new presidential administration, although Justice Breyer was the only one to acknowledge as much.36) Thus, the Court observed, the FDA chose to regulate tobacco products in pursuit of its broad statutory mission to “protect the public health,” even though the agency “apparently recognized [the] dilemma” created by the statutory text: it could neither render tobacco products safe nor ban them from the market, as the FDCA requires.37 The Court was sympathetic to the FDA’s intense need to regulate tobacco products, which posed “the single most pressing threat to public health in the United States.”38 The Court remarked, however, that “[i]n our anxiety to effectuate the congressional purpose of protecting public health, we must take care not to extend the scope of the statute beyond the point where Congress indicated it would stop.”39 The FDA’s regulations had another strike against them as far as Chevron deference was concerned: indications that the agency’s “anxiety” over addressing a “pressing threat” prompted it expand its authority beyond statutory boundaries.

In King v. Burwell, the Court, with Chief Justice Roberts writing for the majority, again considered the significance of the question, this time to uphold a significant piece of the Affordable Care Act (ACA).40The case involved a provision that provides tax credits to individuals who obtain health-insurance plans from “an Exchange established by the State.”41 This provision made no mention of plans obtained from a federal exchange, though the statutory scheme expressly incorporated such exchanges elsewhere.42 The Internal Revenue Service (IRS) had issued a rule extending the tax-credit provision to federal as well as state exchanges.43 In reviewing the rule, the Court quickly bypassed Chevron’s usual two-step inquiry (“whether the statute is ambiguous and, if so, whether the agency’s interpretation is reasonable”).44 It then proceeded straight to the major questions doctrine, without using that label, as an alternative inquiry for “extraordinary cases” in which “there may be reason to hesitate before concluding that Congress has intended [the] implicit delegation” upon which Chevron is premised. Declaring, “This is one of those cases,” the Court stated that “[t]he tax credits are among the Act’s key reforms, involving billions of dollars in spending each year and affecting the price of health insurance.” According to the Court, “Whether those credits are available on Federal Exchanges is thus a question of deep ‘economic and political significance’ that is central to this statutory scheme; had Congress wished to assign that question to an agency, it surely would have done so expressly.”45 The Court continued, “It is especially unlikely that Congress would have delegated this decision to the IRS, which has no expertise in crafting health insurance policy of this sort.”46The Court then analyzed the provision in the context of the statutory scheme and concluded that it includes both state and federal exchanges.47

Like Brown & Williamson, King involved a major question that can be understood to pull in other Breyer-type factors. King involved a certain kind of interpretive problem: whether the literal language of the statutory provision should control despite other signs that Congress had a different intention.48 It was, at bottom, a case of “inartful” legislative drafting.49 The Court suggested that, in such cases, the nature of the question intersects with other factors that help determine which interpreter—court or agency—should decide whether the literal language controls or whether the provision permits or even requires a different reading in context. One relevant, Breyer-type factor that the Court considered was the agency’s specialized expertise (or lack thereof). Also relevant was which interpreter was more likely to answer the question correctly, another Breyer factor.50In this instance, the Court chose itself because the language of the provision, though clunky, was fairly accessible to the average reader, and the statutory scheme required a definitive answer—one that would not be susceptible to change in the future, as might happen if left to the agency.51 This understanding of King helps explain the Chief Justice’s terse conclusion that “this is not a case for the IRS.”52 And it helps explain why the Court used the major questions doctrine for the first time to essentially rewrite the language of a statutory provision. Once the Court took the question for itself by invoking the major questions doctrine, it determined that the provision was actually ambiguous in context, which meant it had to pick the “correct reading” and effectively edit the text accordingly.53

This understanding of King also ties the decision, somewhat counterintuitively, to Zuni Public School District No. 89 v. Department of Education, a case from before Chief Justice Roberts’s time, in which Justice Breyer wrote the majority opinion.54 Zuni involved a highly technical question, not a major one. The Secretary of Education had issued an interpretation of a provision in the federal Impact Aid Act that provides subsidies to school districts on certain lands with “a federal presence.” The statutory provision permits a state to offset its school funding by the amount of federal aid that the affected districts receive if it can show that all districts in the state are financially “equalize[d].”55 As originally enacted, the statute delegated the authority to define “equalization” to the Secretary.56 The Secretary’s regulation interpreting this provision provided that a state’s school districts were equalized if the per-pupil expenditures of the district with the highest per-pupil spending exceeded those of the district with the lowest per-pupil spending by no more than twenty-five percent.57 The regulation permitted states to exclude from their calculations the districts below the 5th and above the 95th percentile in total number of students.58 The result was that “districts whose students account[ed] for the 5 percent of the State’s total student population that [lay] at both the high and low ends of the spending distribution” were excluded.59

Twenty years after the regulation was first promulgated, Congress amended the Act to provide a calculation method. In relevant part, Congress directed the Secretary to disregard from the equalization calculation districts “with per-pupil expenditures . . . above the 95th percentile or below the 5th percentile of such expenditures.”60 The Secretary then issued a new rule implementing the amendment’s “disregard” instruction, including the same formula the agency had employed under the original version of the statute.61 The Secretary’s formula produced a different equalization determination than would a formula that strictly followed the amendment’s text to exclude “on the basis of the number of school districts (ranked by their per-pupil expenditures), without any consideration of the number of pupils in those districts.”62

Was the Secretary’s interpretation entitled to Chevron deference? The Court said yes: “[T]he matter at issue—i.e., the calculation method for determining whether a state aid program ‘equalizes expenditures’—is the kind of highly technical, specialized interstitial matter that Congress often does not decide itself, but delegates to specialized agencies to decide.”63 Perhaps this conclusion was predictable based on the author of the opinion and the mind-numbing complexity of the statute. But there is a deeper explanation. As in King, the core issue in Zuni was whether the literal language of the provision squared with what Congress likely intended the provision to mean. Which interpreter—the Court or the Secretary—was in the best position to decide whether the literal language should control or not? In King, the Court chose itself because the question was major, the text was relatively accessible, the agency lacked any specialized expertise, and the statute could not operate without a definitive resolution (not one susceptible to change with subsequent administrations). Here, the Court favored the Secretary because the question was highly technical, the text was arcane, the Secretary had specialized expertise, and the interpretation was definitive as a practical matter because it had been stable over a long period.64 As a doctrinal matter, the Court deferred to the Secretary’s interpretation under Chevron. But the formal deference standard was fairly unimportant in this case—as then-Judge Breyer said it might be in some instances.65 More important was that the Court looked to the Secretary for a workable, determinate interpretation of the statutory provision.

Although the major questions cases reflect Breyer-type factors, Justice Breyer did not get everything he wished for. The Brown & Williamson majority credited Breyer with bringing the major questions doctrine into the deference analysis,66 but he dissented in that case.67 Breyer disagreed with the Court’s statutory analysis, contending that tobacco products fit within the language and purpose of the statute.68 He did not disagree that tobacco regulation had “enormous social consequences” or that the importance of the question ought to be a relevant factor.69 Rather, he believed that the statutory language and purpose were “the most important indicia of statutory meaning”70 and doubted the existence of a “canon of interpretation” based on the question’s significance that tipped the balance against deference in “close cases.”71 For Breyer then, as before, the significance of the question was only one relevant factor among others, not a tiebreaker. Furthermore, he said, any such canon “do[es] not . . . control[] the outcome . . . [i]nsofar as the decision to regulate tobacco reflects the policy of an administration” because “that administration, and those politically elected officials who support it” will supply the “degree of accountability” necessary to “alleviate any concern that Congress, rather than an administrative agency, ought to make this important regulatory decision.”72

By joining the majority opinion in King, Justice Breyer acceded to an even more extreme form of the major questions doctrine. That form was stronger, capable not only of tipping the balance against deference in a close case but of overriding other indications of congressional intent on the deference issue in any case involving a major question. Furthermore, the Court treated the doctrine as a canon unto itself that serves as an outright alternative to Chevron in such cases. In Brown & Williamson, the Court considered the magnitude of the question at the first step of the Chevron test when asking whether “Congress had directly spoken to the precise question at issue.”73 In King, the Court asserted the magnitude of the question as the test.74 Perhaps Breyer reluctantly accepted this alteration of the major questions factor—a battle he had already begun to lose in Brown & Williamson—to secure the Chief Justice’s vote to fix the ACA in a decision that was otherwise consistent with his overall approach.

In any event, Breyer was more successful in preventing another factor that he had earlier emphasized—whether an agency interpretation is both “substantively” and “procedurally” the sort that Congress would likely intend to carry the force of law75—from becoming a simple rule. If the major questions cases basically align with Breyer’s approach, the cases focusing on the “procedural formality” or more broadly, congressional delegation of agency lawmaking authority, are the fullest realization of his vision. The next Section takes up this discussion.

B. Lawmaking Procedures

The first of the “procedural formality” cases is United States v. Mead Corp.76 That case arose when the Customs Service sent Mead a “ruling letter” advising the company that its day planners were considered “diaries” subject to tariff under the Harmonized Tariff Schedule, reversing their prior tax-free status.77 The Court held that the agency interpretation did not qualify for Chevron deference because a ruling letter is not a proper format for making binding law.78 In order for an agency interpretation to qualify for Chevron deference, Congress must delegate the authority to make binding law to the agency, and the agency must use that authority in making the interpretation at issue.79 The Court said the delegation of certain procedures—notice-and-comment rulemaking or formal adjudication—was a “good indicator” that Congress intended to give the agency lawmaking authority.80 Although the Secretary of the Treasury possessed the authority to exercise notice-and-comment rulemaking procedures for customs duties, the Customs Service instead sent a ruling letter to Mead.81 The Court found that “[t]he authorization for classification rulings, and Customs’s practice in making them, present a case far removed not only from notice-and-comment process, but from any other circumstances reasonably suggesting that Congress ever thought of classification rulings as deserving the deference claimed for them here.”82 At most, an interpretation contained in a Customs ruling letter could qualify for Skidmore deference83—that is, a court might “resort [to the interpretation] for guidance” based on “the thoroughness evident in its consideration, the validity of its reasoning, its consistency with earlier and later pronouncements, and all those factors which give it the power to persuade, if lacking power to control.”84 While closing the door on ruling letters, the Court left open the possibility that an interpretation could qualify for Chevron deference without procedural formality under other circumstances.85

The next Term, Barnhart v. Walton86 filled that gap. The Court, with Justice Breyer writing for the majority, held that the Social Security Administration’s (SSA’s) interpretation of the Social Security Act qualified for Chevron deference even though the agency initially provided the interpretation in an employee handbook and other informal sources before ultimately issuing it through its delegated notice-and-comment rulemaking authority.87 The Court said:

In this case, the interstitial nature of the legal question, the related expertise of the Agency, the importance of the question to administration of the statute, the complexity of that administration, and the careful consideration the Agency has given the question over a long period of time all indicate that Chevron provides the appropriate legal lens through which to view the legality of the Agency interpretation here at issue.88

Unsurprisingly given its author, Barnhart was the Court’s most forthright application of Justice Breyer’s factor-based, context-specific approach to judicial deference. To many, it was also the most confusing of the Court’s deference opinions. How were the Barnhart factors different from the ones in Skidmore, which directed courts to regard an agency interpretation not as a source of binding law but only as a source of “guidance” for their own independent interpretations?89 How would a court know when to apply the procedural rule from Mead or the standards from Barnhart?90 But these questions miss the bigger picture: the Court per Breyer was embracing a context-specific approach. It was not offering Mead, Barnhart, and Skidmore as alternative rules or on/off deference switches. It instead was creating space for different degrees of deference. If the SSA’s interpretation would qualify at a minimum for “persuasive” Skidmore deference based on the thoroughness evident in its consideration, the validity of its reasoning, [and] its consistency with earlier and later pronouncements,”91 why not give it “binding” Chevron deference when other important factors also were present: the agency possessed delegated lawmaking authority, issued its interpretation almost contemporaneously with the enactment of the statute, and had adhered to that interpretation since?92 Under these circumstances, withholding Chevron-level deference just because the agency had not used its delegated notice-and-comment rulemaking procedures (as it eventually did in anticipation of this litigation) would render those procedures an empty formality.93 The agency would have arrived in exactly the same place as before (which is precisely where it landed after notice-and-comment rulemaking).94 In Mead, the Court never said that procedural formality was required or necessary for its own sake—that was Justice Scalia’s characterization of the opinion in his dissent.95

Barnhart clarifies that Mead did not establish a simple rule requiring procedural formality and displays a context-based approach that echoes many of the factors that Justice Breyer earlier identified in his scholarship. Understanding Barnhart as the best evidence of a context-specific approach clears up some of the confusion about the “procedural formality” decisions. In addition, it demonstrates how such an approach works to approximate the degree of deference that Congress likely intends, and that makes sense, for a particular interpretation.

* * *

While the Court has decided other important cases consistent with a context-specific approach to judicial deference, the examples I have discussed show the extent to which the Court incorporated factors like those then-Judge Breyer identified in 1986. The complex doctrinal framework that Justice Breyer helped build on the Court substantially achieved what he earlier envisioned. I am not claiming that Breyer is responsible for the Court’s doctrines, the direction they took, or the direction they will take in the future. As previously noted, he did not agree with the Court’s formulation and application of the doctrines in every case. Perhaps most importantly, Barnhart notwithstanding, the Court never fully appreciated the music of his approach. Breyer was interested in a “blended” approach, harmonizing the Court’s individual doctrines to a much greater extent.96 Although there are strong indications that Breyer won the day more often than he lost, few would say that the law of deference reflects a genuinely blended approach.97

Regrettably for Justice Breyer, the Court’s approach is breaking down as he retires from the bench. Moreover, the new conservative majority appears to be moving toward another single, simple rule: no deference. Ironically, some are doing so to prevent “the abdication of the judicial duty.”98 Then-Judge Breyer thought that a rule of automatic deference to agency interpretations under Chevron would amount to more of an abdication than necessary, and Justice Breyer helped prevent that result. In contrast, some conservative Justices seem to view almost any deference as contrary to the Court’s “job of saying what the law is.”99 From this perspective, interpreting statutes is the courts’ constitutional role to perform, not Congress’s choice to assign or agencies’ prerogative to assume. But even the conservative Justices who have not expressed so forceful an objection to deference have harnessed a new tool for undermining it, as the next Part details.

II. the court’s new rule

The new conservative majority has been transforming the major questions doctrine in a way that strongly suggests they are moving toward a no-deference rule.100 The conservative majority started rather stealthily, responding to applications for emergency relief on the so-called shadow docket. In a per curiam opinion in Alabama Ass’n of Realtors v. Department of Health & Human Services, the Court blocked the Centers for Disease Control and Prevention (CDC) from instituting a nationwide eviction moratorium for certain tenants living in counties with high COVID-19 transmission levels.101 The Court described the statutory basis for the moratorium as a “wafer-thin reed upon which to rest such sweeping power,” and the agency’s “claim of expansive authority” under the relevant provision as both “unprecedented” and “breathtaking.”102 Quoting Utility Air Regulatory Group, which quoted Brown & Williamson, the Court stated, “We expect Congress to speak clearly when authorizing an agency to exercise powers of vast economic and political significance.”103 Justice Breyer dissented, joined by Justices Sotomayor and Kagan.104 He did not find it “demonstrably clear” that the CDC lacked authority to issue the moratorium,105 as the Court’s standard for emergency relief requires; indeed, he said it was “undisputed that the statute permits the CDC to adopt significant measures such as quarantines, which arguably impose greater restrictions on individuals’ rights and state police powers than do limits on evictions.”106 But his main point was that lower courts were split on the legal questions in the case. And the answers to those questions were very important, “impact[ing] the health of millions.”107 He believed that “[t]hese questions call[ed] for considered decisionmaking, informed by full briefing and argument,” and the Court should not “set aside the CDC’s eviction moratorium in this summary proceeding.”108

In the COVID-vaccine-mandate case, National Federation of Independent Business (NFIB) v. Occupational Safety & Health Administration (OSHA), the Court again applied the major questions doctrine to block a challenged agency regulation on its shadow docket.109 In a per curiam opinion, the Court prevented OSHA from imposing a vaccine mandate in workplaces covered by the Occupational Safety and Health Act because (1) such mandates were a matter of great “economic and political significance”; and (2) OSHA regulates “workplace safety,” not “public health measures.”110 If an agency’s authority over a new subject is significant and “untethered” from its core duties, then Congress must grant that authority clearly.111 Justice Gorsuch, joined by Justices Thomas and Alito, wrote a concurrence that went even further.112 He called the major questions doctrine a surrogate for the constitutional nondelegation doctrine: while the clear-statement rule gets the job done in some cases by blocking an agency from exercising significant authority, in many others Congress should not be able to delegate such authority at all.113 In Gorsuch’s view, the major questions doctrine is a clear-statement super rule: no deference and no delegation. Justices Breyer, Sotomayor, and Kagan jointly dissented.114 They acknowledged that the question was important: “The [mandate] responds to a workplace health emergency unprecedented in the agency’s history.”115 And they appreciated that the mandate might be “far-reaching,” as would reflect “the scope of the crisis.”116 But they did not find these factors controlling because the mandate “perfectly fit[] the language of the applicable statutory provision.”117 Consequently, the dissenting Justices said, the Court lacked justification to prevent the mandate from taking effect.118

The Court reserved its most dramatic makeover of the major questions doctrine for a case on the merits docket, decided on the last day of the 2021 October Term.119 In West Virginia v. EPA, Chief Justice Roberts—writing for himself and the five other conservative Justices—concluded that the EPA lacked authority to issue a rule shifting electric-power generation from higher-pollution-emitting sources to lower-pollution-emitting ones, rather than imposing a source-specific requirement.120 The Court offered four main reasons for rejecting this rule. First, “as EPA itself admitted . . . ‘[u]nderstand[ing] and project[ing] system-wide . . . trends in areas such as electricity transmission, distribution, and storage’ requires ‘technical and policy expertise not traditionally needed in EPA regulatory development.’”121 Second, “[t]he basic and consequential tradeoffs involved in such a choice” (i.e., “the decision how much coal-based generation there should be over the coming decades”) “are ones that Congress would likely have intended for itself.”122 Third, “[t]he last place one would expect to find [such congressional authorization] is in the previously little-used backwater of Section 111(d).”123 Finally, the Court could not “ignore that the regulatory writ EPA newly uncovered conveniently enabled it to enact a program that, long after the dangers posed by greenhouse gas emissions ‘had become well known, Congress considered and rejected’ multiple times.”124

With respect to the major questions doctrine itself, the Court for the first time endorsed it as a full-blown canon of statutory construction. The Court confirmed its “label” and status as a “doctrine”:

[The label] took hold because it refers to an identifiable body of law that has developed over a series of significant cases all addressing a particular and recurring problem: agencies asserting highly consequential power beyond what Congress could reasonably be understood to have granted. Scholars and jurists have recognized the common threads between those decisions. So have we.125

Among the major questions cases, the Court highlighted Brown & Williamson, Gonzales, Utility Air, and NFIB.126Underscoring the force of the doctrine, the Court specified that it was not only a “practical” indicator of congressional intent but also rooted in “separation of powers principles.”127 As for its operation, the Court said that an agency cannot exercise authority over a major question without “something more than a merely plausible textual basis” for its action.128 Instead, the agency must identify “‘clear congressional authorization’ for the power it claims.”129 Although the Court said the doctrine only applies in “extraordinary cases,”130 it never mentioned Chevron or any alternative framework for “ordinary” cases.

Justice Gorsuch concurred, joined by Justice Alito, writing to offer some further “observations” about the major questions doctrine.131 He explored in detail its constitutional foundation and practical function as “a clear-statement rule.”132 He also surveyed the Court’s decisions to demonstrate when the doctrine comes into play. For example, it applies “when an agency claims the power to resolve a matter of great ‘political significance’ . . . or end an ‘earnest and profound debate across the country.’”133 It applies “when Congress has ‘considered and rejected’ bills authorizing something akin to the agency’s proposed course of action,”134 and it applies when an agency “seeks to regulate ‘a significant portion of the American economy’ . . . or require ‘billions of dollars in spending by private persons or entities.’”135

Justice Kagan dissented, joined by Justices Breyer and Sotomayor. She said there was no such thing as the major questions “doctrine.”136 On her reading, the Court had previously approached major questions in a more contextual, “purposive” fashion.137 Specifically, “the Court simply insisted that the text of a broad delegation, like any other statute, should be read in context, and with a modicum of common sense.”138 Under this “ordinary method” of statutory interpretation, the Court asked whether “[f]irst, an agency was operating far outside its traditional lane, so that it had no viable claim of expertise or experience[,] [a]nd second, the action, if allowed, would have conflicted with, or even wreaked havoc on, Congress’s broader design”—in Kagan’s view, neither applied to the EPA’s interpretation in the case.139 In departing from this method, Kagan argued, the majority “replace[d] normal text-in-context statutory interpretation with some tougher-to-satisfy set of rules,” namely a “two-step inquiry” under which “[f]irst, a court must decide, by looking at some panoply of factors, whether agency action presents an ‘extraordinary case[],’ [and second,] if it does, the agency ‘must point to clear congressional authorization for the power it claims,’ someplace over and above the normal statutory basis we require.”140 Not only was this two-step inquiry newly minted—”[t]he Court has never even used the term ‘major questions doctrine’ before,” Kagan noted141—it also defied common sense. Congress, she observed, delegates decisions of “significant ‘economic and political magnitude’” to agencies “all the time,” but in a “sensible way” that must be assessed in consideration of “(among other potentially relevant factors) the nature of the regulation, the nature of the agency, and the relationship of the two to each other.”142 Applying these factors to the EPA’s rule, Kagan found that it “f[ell] within EPA’s wheelhouse” and “fit[] perfectly” with the statute.143

Justice Kagan’s protests aside, West Virginia establishes that the major questions doctrineeven if once only a factor among others in the deference mixis now an official canon of construction that prohibits agencies from issuing interpretations that address matters of significance absent clear congressional authorization. Furthermore, this new canon is positioned to become a simple rule depriving agencies of deference for any interpretation of nontrivial consequence. The Court claimed in West Virginia that the doctrine applies only in “extraordinary cases.”144 However, it has provided little more than general categories (e.g., the political nature of the question or its effect on the national economy) to separate such cases from “ordinary” ones. Given the nature of regulatory statutes and the problems they address, these categories are easily satisfied by any agency action worthy of attention. Meanwhile, the Court has little institutional incentive to restrain application of the doctrine to truly extraordinary cases. The major questions doctrine is a mechanism for the Court to reclaim judicial authority and cabin congressional delegation without overruling entrenched precedents, notably Chevron and the permissive nondelegation decisions.145

That is not to say blanket application of the doctrine is inevitable. Already there are cases involving important questions that evince a more regular approach. An example is Biden v. Missouri,146 which the Court decided on the orders docket the same day as NFIB. The case also involved a COVID-vaccine mandate, issued by the Secretary of Health and Human Services (HHS) and the Medicaid and Medicare Programs.147 A majority comprised of Chief Justice Roberts, the three liberal justices, and Justice Kavanaugh refused to block a vaccine mandate for healthcare providers because it was likely lawful for the Secretary to issue such a mandate, which “fit[]” within the relevant text and statutory scheme.148 Justices Thomas, Alito, Gorsuch, and Barrett dissented.149 They would have decided the two vaccine-mandate cases alike and against the relevant agencies, blocking both mandates from taking effect.150 Perhaps, then, a majority of the Court would refrain from applying the major questions doctrine unless a case involves an extraordinary “misfit” between an agency’s asserted authority and its statutory “wheelhouse,” to use Justice Kagan’s terms from her West Virginia dissent.151

Another example is Becerra v. Empire Health Foundation,152 which the Court decided a week before West Virginia. The case had a remarkably similar feel to Zuni—in Empire Health Foundation, the Court approved an agency interpretation of inartful, highly technical text. At issue was the Medicare fraction, which is used to pay hospitals for providing inpatient treatment to qualifying individuals with disabilities. HHS had maintained a consistent interpretation of the formula for nearly two decades (though, admittedly, it read the provision differently for as many years before that).153 Justice Kagan wrote for the majority, which included Justices Thomas, Breyer, Sotomayor, and Barrett.154 She characterized the question at issue as “technical but important” and the statutory language as an interpretive maze.155 She introduced the statutory scheme at “a high level of generality” to provide a way into the text.156 Nevertheless, reading the text as written was nearly impossible.157 She set it out, inserting in brackets basic grammar, program names, and other missing referents to assist the reader. Still, she acknowledged, it was “a mouthful (and without the brackets it’s even worse).”158

How should the Court approach such unfortunate language, upon which major healthcare dollars depended? Justice Kagan stated up front that “HHS’s regulation correctly construes the statutory language at issue.”159 Unlike the majority in Zuni, she did not mention Chevron. Instead, she offered what might be understood as a different pair of steps. First, she determined that “the ordinary meaning of the fraction descriptions, as is obvious to any ordinary reader, does not exactly leap off the page.”160 Second, she invoked “Justice Frankfurter’s injunction that when a statute is ‘addressed to specialists, [it] must be read by judges with the minds of the specialists.’”161 “[R]ead in that suitable way,” Kagan found that “the fraction descriptions disclose a surprisingly clear meaning—the one chosen by HHS.”162 Although she found a clear meaning and did not defer to HHS, as the Court had in Zuni, the exact level of deference made no difference in this case. She chose the correct interpretation “with the mind” of specialists at HHS, not a judge wed to words. With this mindset, she worked through the text in the context of the statutory structure, rejected the arguments on the other side, and concluded that “[t]he text and context support[ed] the agency’s reading: HHS has interpreted the words in those provisions to mean just what they mean throughout the Medicare statute.”163 Her message was clear: the Court did not simply rely on HHS in interpreting an ambiguous statutory interpretation, Skidmore-style; rather, HHS led the Court through the statutory maze to the “surprisingly clear meaning.” Justice Kavanaugh dissented, joined by the Chief Justice and Justices Alito and Gorsuch.164 In Kavanaugh’s view, the case was easily resolved “by the most fundamental principle of statutory interpretation: Read the statute.”165 He believed that the literal language of the relevant provision controlled and cut against HHS.166 But Kagan (with the support of Justices Thomas and Barrett, as well as her liberal colleagues!) devised a new strategy for cases involving technical-but-important questions and inside-baseball text: Read the statute with the mind of the agency.

If these cases are any indication, the major questions doctrine need not become a simple rule directing courts to withhold deference from an agency interpretation any time they perceive the question to be a significant one. But if the Justices do take the doctrine too far, there will be cause to recall what then-Judge Breyer said when the Court was on the precipice of embracing Chevron as a simple rule: the context-specific approach likely will not disappear from the law but will instead reemerge in the lower courts through other outlets.167 The next Part explains why.

III. justice breyer’s legacy in the lower courts

In his 1986 article, then-Judge Breyer explained that a simple rule of judicial deference was unlikely to prevail in the long run because it does not account for the difficult task of judicial review.168 Legal questions come “in an almost infinite variety of sizes, shapes and hues” and arise in an innumerable variety of procedural postures.169 Some questions are bound up with a “web of existing interpretations, including interpretation of rules, standards, statutory meanings and interpretive practices.”170 Some involve voluminous records, some pop up in the middle of already-protracted litigation, and some arrive without proper agency consideration.171 Meanwhile, judges feel “institutional pressure” to make sense of the law before them and dispose of cases in a fair, expeditious manner.172 In short, “the way[s] in which questions of statutory interpretation may arise are too many and too complex to rely upon a single simple rule to provide an answer.”173 Breyer speculated that, if the Court did understand Chevron to impose a simple rule of deference, lower courts would likely gravitate toward a “jurisprudence of ‘degree and difference’” regardless.174 Specifically, lower courts would not automatically defer to an agency interpretation of an ambiguous statutory provision without considering whether that degree of deference was appropriate under the circumstances.175

One might imagine that the inverse is also true—courts will not automatically withhold deference from an agency interpretation, even if the question is important, without considering whether doing so is appropriate in context. Rather, lower courts will work out the “agency-court relationship” by considering which interpreter can best make sense of the particular question and statute at issue.176 The reason is that none of the institutional realities that then-Judge Breyer confronted daily have changed. Legal questions are no less varied—either substantively or procedurally. Furthermore, courts are likely to experience additional pressure given the direction of the Court’s new rule toward banning deference. Neither the lower courts nor Congress have suddenly developed the expertise necessary to address the innumerable “technical but important” issues that arise in the operation of a complex regulatory statute.177 Feeling considerable strain just trying to do their jobs, lower courts will look for other principles of statutory interpretation or administrative law that justify deferring to the agency’s resolution of those questions.

Lower courts will not have to look far. For example, they might take the tack that the majority took in Empire Health Foundation, finding a clear meaning of a highly technical statutory provision “with the minds of the specialists.”178 Or, if the relevant statutory provision is ambiguous, courts might heavily rely on the agency interpretation in reaching their own interpretation, Skidmore-style, even if the Court never cites that case again. In many instances, the precise degree of deference does not matter as much as achieving a sensible result. But Chevron-level deference does matter when an agency seeks the flexibility to change its interpretation in light of changed circumstances, whether technological, economic, social, or political. If a court locks in the meaning of a statutory provision, generally only Congress may change it.179 When strong-form deference remains important, courts might see a path forward by shifting their focus. If Justice Kagan can read a statute with the mind of a specialist, lower courts can see a statutory interpretation through the lens of an administrator. Thus, they might characterize an agency’s action not as interpretation, but as a policy decision implementing the statute, subject to more accommodating arbitrary-and-capricious review under the Administrative Procedure Act.180

The upshot is that, as long as regulatory statutes exist, a more context-specific approach to judicial review of agency statutory interpretations is likely to live on in the lower courts. No matter how simple the rule, lower courts do not have a simple job. As then-Judge Breyer wrote, “Inevitably, one suspects, we will find the courts actually following more varied approaches, sometimes deferring to agency interpretations, sometimes not, depending upon the statute, the question, the context, and what ‘makes sense’ in the particular litigation.”181


Well before he joined the Court, Justice Breyer advocated an approach under which the degree of judicial deference to an agency interpretation of an ambiguous statutory provision varies with the particular interpretation and statute at issue. Courts would consider a number of relevant factors—such as the nature of the question, the agency’s related expertise, and the procedural format of the interpretation—in determining the level of deference that Congress likely would intend, and that makes practical sense, for the interpretation at issue. This Essay has shown that Breyer’s context-specific approach emerged in the Court’s deference decisions more than many might appreciate. My claim is not that these decisions are attributable to Breyer or that they achieve the exact balance he wanted. Rather, he helped build an approach to judicial deference that, though complex, proved workable for decades.

Now the conservative majority on the Court has begun to dismantle that approach by positioning the major questions doctrine as a rule of no deference to any agency interpretation of real consequence. When the major questions doctrine comes into play, Chevron drops from the analysis. The Court demands clear textual authorization for the agency interpretation rather than implying such authorization from statutory ambiguity. Although the Court has said the doctrine applies only in “extraordinary” cases, it has offered no metric for assessing a question’s significance beyond general, easily satisfied criteria. And it has an institutional incentive to apply that doctrine extensively to restore its own authority as well as to restrict congressional delegation. What remains of deference is uncertain.

But Justice Breyer may have the last word, though he would take no pride and find no solace in offering it. Even if the Court takes its new rule too far, it is ultimately unlikely to see that rule have a significant effect in the lower courts. Judicial review is not a simple matter amenable to a simple rule. Legal questions come in an endless variety, and these complexities generate institutional pressure that may lead judges back toward a context-specific approach, despite any simple rule. These realities have not changed with the membership of the Court. Only the direction of the rule has changed—from automatic deference to no deference—which intensifies pressure on courts to find a workaround given the sheer volume of important technical questions that arise under regulatory statutes. No simple rule is likely to succeed in the long run. The new conservative majority might believe that a context-specific approach constitutes an abdication of judicial responsibility to say what the law is. For judges in the trenches and others who carry on Breyer’s legacy, it is the very definition of responsible judging.

Associate Dean for Academic Affairs and David Daniels Allen Distinguished Chair in Law, Vanderbilt Law School. This Essay is written with admiration and affection for Justice Stephen Breyer, for whom I clerked during his first year on the Supreme Court of the United States. For generous comments and support, I thank Michael Bressman, Justin Driver, and Kevin Stack. For the opportunity to contribute to this Essay collection and excellent editorial assistance, I thank Sarah Walker and the Yale Law Journal Forum team.