Its title notwithstanding, Professor Josh Blackman’s Comment1 is not really about gridlock. It makes no attempt to ask what constitutes gridlock or how we might identify it,2 nor does it offer much by way of a theory of how we ought to respond to gridlock, if indeed we are experiencing it. Instead, Blackman takes the opportunity presented by two recent executive actions of which he disapproves to advance a certain theory of administrative law, one in which the “major questions doctrine” plays an increasingly outsized role. But the reasons Blackman gives in support of this new direction in administrative law do not stand up to scrutiny.
The two executive actions at which Blackman takes aim resulted in two of last Term’s High Court fizzles: Zubik v. Burwell3 and United States v. Texas.4 The former involved the structures of the exemptions and accommodations to the “contraceptive mandate” under the Affordable Care Act5 (ACA); the Court vacated the lower court judgments and remanded the cases with soothing words about the possibility of a negotiated compromise. The latter involved the Deferred Action for Parents of Americans6 (DAPA) immigration-law program; the equally divided Court affirmed the appellate court’s decision upholding the district court’s injunction preventing implementation of the program. Blackman nicely summarizes the underlying disputes and the Court’s (lack of) resolution; there is no need to replicate his efforts here.7
Rather than delving into the underlying substantive law, Blackman suggests that “these disputes can be resolved on the more neutral principle of whether the agency can take such novel actions in the first instance. . . . These ‘major questions’ should be returned to the political process — which is where they should have been decided to begin with.”8 Indeed, the core of Blackman’s Comment is a call for a more muscular major questions doctrine. The doctrine originated in Justice Scalia’s 1994 opinion in MCI Telecommunications Corp. v. AT&T Co.,9 in which he wrote: “It is highly unlikely that Congress would leave the determination of whether an industry will be entirely, or even substantially, rate-regulated to agency discretion — and even more unlikely that it would achieve that through such a subtle device as permission to ‘modify’ rate-filing requirements.”10 In MCI, the doctrine is best understood as a factor at Step One of the Chevron analysis: it goes to the “boundary of legitimate agency discretion” as a factor in the determination of whether the statute is, in fact, ambiguous in relevant part and therefore deserving of Step Two deference.11 Indeed, before getting to the language quoted above, Scalia wrote (citing Chevron), “Since an agency’s interpretation of a statute is not entitled to deference when it goes beyond the meaning that the statute can bear, the Commission’s . . . policy can be justified only if it makes a less than radical or fundamental change in the Act’s tariff-filing requirement.”12
But in a series of subsequent cases entirely written by, and largely joined by, the conservative bloc of the Court,13 the doctrine moved backward to Step Zero of the Chevron analysis — that is, it became a question of whether the Chevron framework would apply at all.14 In the words of one commentator, the major questions doctrine has come to stand for “a general presumption against implied delegation where the Court independently determines that the issue was simply too significant to be left to the agency.”15 Such issues are therefore taken wholly outside of the Chevron framework and reviewed de novo. As Chief Justice Roberts wrote in the most recent major-questions-doctrine case, King v. Burwell,16 a matter “involving billions of dollars in spending each year and affecting the price of health insurance for millions of people” constituted “a question of deep ‘economic and political significance’ . . . ; had Congress wished to assign that question to an agency, it surely would have done so expressly.”17 Accordingly, it was the Court’s job “to determine the correct reading” of the statutory language,18 not to determine whether the agency’s reading was reasonable.
In King, after performing the de novo review, the Court came to the conclusion that the agency’s reading of the statute was correct; put differently, because no deference applied, no other agency interpretation of the statutory text would have been permissible. Blackman’s version of the doctrine goes significantly further. The major questions doctrine, in his view, obviates the “need for judges to draw . . . difficult line[s]” with respect to statutory or constitutional law.19 Whereas the King majority used the major questions doctrine to force itself to wrestle with the question of how to interpret statutory text correctly, Blackman would have courts use his version of the doctrine to prevent such close engagement with substantive law. How? By returning major questions to Congress, “which is where they should have been decided to begin with.”20 In its most extreme version, the Court’s major questions doctrine denies deference whenever agencies address “major questions.” Blackman’s major questions doctrine, by contrast, posits that some questions are too major ever to be delegated to agencies, full stop. That is the only way that the doctrine could allow judges to avoid addressing questions of substantive law.
So what are we to make of Blackman’s major questions doctrine on steroids? It is, perhaps, most illuminating to begin with his claim that the doctrine is a “neutral principle” — or, at least, is “more neutral” than delving into questions of statutory interpretation or constitutional law.21 The language of “neutral principles” of course gestures toward Professor Herbert Wechsler’s famous treatment,22 but the gesture remains indistinct. Neutrality cannot be free-floating; otherwise all principles would fail the test. (“Apply neutral principles” is decidedly nonneutral as between neutral principles and nonneutral principles.) For Wechsler, what was required seemed to be neutrality with respect to the identity of the litigant: it should not matter to the outcome whether the claim is “put forward by a labor union or a taxpayer, a Negro or a segregationist, a corporation or a Communist.”23 Or, as he would much later put it, the judge should ask herself: “Would I reach the same result if the substantive interests were otherwise?”24 Of course, as critics began pointing out almost as soon as his article was published, Wechsler’s standard is both insufficiently nuanced and immensely difficult to apply.25 Nevertheless, Wechsler clearly understood that “neutrality” had to mean neutrality as between different values of some variable(s) identified in advance. A criminal court should be neutral as between blonde and brunette defendants; it should not be neutral as between innocent and guilty ones.
So what, exactly, does Blackman think that his major questions doctrine is neutral as between? He is not explicit, but the fact that he contrasts it with the courts’ having to confront difficult questions of statutory interpretation (of the Religious Freedom Restoration Act26 (RFRA) or the Immigration and Nationality Act27) or of constitutional law (with respect to the First Amendment or the scope of executive power)28 suggests that he thinks it is more neutral as a matter of substantive values. After all, if the act in question fails the major-questions-doctrine test, then “there is no need for judges to draw th[e] difficult line” posed by the statutory and constitutional questions: “[t]hese ‘major questions’ should be returned to the political process — which is where they should have been decided to begin with.”29 The major questions doctrine, then, is figured as institutionally partial, but substantively neutral. (I’ll leave for another day Blackman’s implicit claim that the bureaucracy stands outside of “the political process.”30)
Blackman is right in one regard. The doctrine is institutionally partial — although not quite in the way he imagines. Begin with what ought to be obvious: whether or not a particular question is “major” is a political judgment, not a fact about the world. Blackman suggests that, while the caselaw may be convoluted — it “suggests at least nine factors, none dispositive, to determine if a decision is major”31 — the core of the doctrine at least is clear: “Despite these uncertainties, the majorness of the question at issue in Zubik is entirely beyond question.”32 Blackman arrives at his majorness-beyond-dispute determination through a particular framing: “the principles of free exercise, enshrined in the First Amendment and RFRA, are of the highest order of magnitude. . . . Surely religious freedom is more important to Congress — and to the nation as a whole — than the regulation of snuff,”33 a reference to the major-questions-doctrine language in FDA v. Brown & Williamson Tobacco Corp.,34 which dealt with tobacco regulation. Likewise, “[t]he Women’s Health Amendment [of the ACA] vested HHS with authority over the ‘interstitial matters’ of what constitutes preventive care, without addressing the ‘major questions’ of how religious objectors should be accommodated.”35 The Department was exercising “a great substantive and independent power over free exercise,”36 but had “Congress wished to assign that question to an agency, it surely would have done so expressly.”37 Indeed, “Congress’s silence in the ACA demonstrates that Congress did not intend for the agencies to exercise such an awesome power.”38 (Blackman makes similar claims about the majorness of DAPA.39)
One can, for starters, question Blackman’s ipse dixit that the free exercise issue in Zubik is “surely” more major than the tobacco issue in Brown & Williamson. After all, in 2000 (when Brown & Williamson was decided), just under a quarter of American adults smoked cigarettes,40 and between 2000 and 2004, over 2.2 million American deaths were attributable to smoking.41 One might well think that the premature deaths of a population the size of Houston (the nation’s fourth-largest city) over a five-year period would be “more major” than the burden to the religious practice of a relatively small number of employers arising out of HHS’s contraceptive mandate under the ACA. Or one might not. The point is that a huge number of federal government programs affect large numbers of people in varied and complex ways. To reduce this complexity to a single spectrum of “majorness” and then assert with the utmost confidence the relative positions of various policies on that spectrum is to fall almost immediately into a querulous quagmire. And that’s before someone demands to know where on this spectrum the cutoff for triggering the major questions doctrine is located.
But, in fact, the problem is even more fundamental. Above, I accepted at face value Blackman’s assertion that the “question” at issue in Zubik was “the principles of free exercise” or “religious freedom.” But this is very much a contestable framing. As Blackman demonstrates in some detail, the “contraceptive mandate” arose from HHS’s interpretation of the ACA’s requirement that employer-provided insurance plans cover (without additional copayments) “preventive care” for women.42 HHS then (presumably to avoid violating RFRA) created a structure of exemptions and accommodations for religious employers.43 Employers who did not like the limitations and structure of those exemptions and accommodations filed suit, leading to the seven cases consolidated in Zubik. Blackman is not wrong to describe the issue here as one of religious freedom. But neither would it be wrong to describe the issue as one of health care, preventive health care, contraception, health insurance, the scope of the social safety net, reproductive rights, sex equality, etc. All of these are plausible frames on the same set of facts. And, depending on the substantive values one brings to the table, some of these issues may seem “major” and others not. Indeed, Blackman argues that the ACA “vested HHS with authority over the ‘interstitial matters’ of what constitutes preventive care, without addressing the ‘major questions’ of how religious objectors should be accommodated”44 — in other words, the rule was minor, but its exception was major. This is certainly possible, but it is by no means obviously correct.
And this brings us back to the institutional question. If the “majorness” of a question is necessarily a matter of (a) how that question is framed, and (b) what substantive values one understands as indicia of “majorness,” then the major questions doctrine empowers whoever gets to make those (necessarily normative) determinations. Blackman gives us a parenthetical hint of whom that might be: “Justice Kennedy has been in the majority of each major question doctrine case.”45 The major questions doctrine aggrandizes judges, who decide in any given case both how to frame the issue and how important that issue is. As Professor Lisa Heinzerling explains, far from empowering Congress, the doctrine actually “aggrandize[s] the courts at the expense of Congress and the executive” by “chang[ing] the ground rules of statutory interpretation after the other branches have acted, upsetting the reliance the other branches may have placed in the pre-existing interpretive regime and yet not replacing that regime with stable and predictable rules that could foster reliance moving forward.”46 The point is not simply that there is a transitional period of uncertainty; it is that the uncertainty is indefinite because the doctrine does no more than tell judges not to defer to agencies when the actions taken by those agencies are, in the view of those judges, a big deal. And, of course, Blackman’s version of the doctrine would go even further: instead of telling judges to withhold deference, it would tell them to vacate the administrative action whenever that action was, in their view, a big deal.
And what do the judges do with this power? The answer, it turns out, undercuts Blackman’s claim that the doctrine is substantively neutral. The major questions doctrine partakes of what Professors Cass Sunstein and Adrian Vermeule have termed “libertarian administrative law” — that is to say, “a form of administrative law that . . . invokes, implicitly or explicitly, libertarian goals to give a kind of strict scrutiny to agency decisions.”47 As Heinzerling notes, the major questions doctrine “require[s] clear congressional language to enable an ambitious regulatory agenda, but not to disable one. This asymmetry is the [doctrine’s] tell; it is the sign that [it] mask[s] a judicial agenda hostile to a robust regulatory state.”48 To be clear, a judge need not be consciously seeking libertarian outcomes for this to be the case: action is generally more salient than inaction,49 so it should not be surprising that a move to regulate would seem “major,” while nonregulation and deregulation both seem more minor. Consider what Blackman has to say about the contraceptive mandate: “There was a really, really easy way to avoid the Zubik controversy in the first place. The executive branch could have simply rejected the Institute of Medicine’s determination and excluded contraceptives from the definition of ‘preventive care.’”50 It turns out that not regulating is always a “really, really easy way” to avoid falling prey to the major questions doctrine.51 HHS’s regulations embodied “policy-laden judgments,”52 but apparently not issuing those regulations would have been the sound judgment to avoid policymaking — at least, it would have constituted the avoidance of major policymaking.
It is, perhaps, worth pausing to note here that, even in the absence of any major questions doctrine at all, agencies would hardly have carte blanche. As Sunstein has noted, “Chevron deference does not give agencies a blank check. It remains the case that agency decisions must not violate clearly expressed legislative will, must represent reasonable interpretations of statutes, and must not be arbitrary in any way.”53 The major questions doctrine adds an additional layer: when it is triggered, judges will not accord the administrative action Chevron deference at all (in the Court’s version) or they will vacate the action on sight (in Blackman’s more extreme version). But if nonregulation and deregulation receive Chevron deference (or stronger54) as a matter of course, while regulation receives it only if judges determine that the regulation does not address a topic of “major” concern, then the doctrine is putting a thumb on the scale against regulation.55 This is not a substantively neutral doctrine, even in the Court’s comparatively modest version.
In the Comment’s only real nod to its ostensible topic, Blackman attempts to justify this thumb on the scale in terms of status-quo bias: “Gridlock does not license the expansion of the executive’s power. Under our system of government, there is only one way to decide major questions, as difficult as it may be in our gridlocked polity. In the absence of consensus, the status quo remains.”56 It is true that status-quo bias is a feature of any plausible political system,57 but Blackman is mistaken to view the major questions doctrine as an agent of the status quo, full stop. After all, the doctrine only has bite if, under the normal rules (that is, Chevron), a court would determine that Congress had, in fact, delegated adequate authority to the agency to take the action in question. The doctrine comes in to forestall the Chevron analysis, instituting a special, status-quo-protecting norm only in cases in which the court deems a “major question” to be present. Because, as we have seen, judges are far more likely to find major questions in cases of active regulation, rather than nonregulation or deregulation, the doctrine privileges only nonregulatory baselines, while allowing for regulatory ones to be rolled back more easily.
There may well be a justification for a doctrine of administrative law that is institutionally partial towards the judiciary and substantively partial towards deregulatory outcomes. But neutrality is not it, and Blackman has not offered an alternative.
* Professor of Law, Cornell Law School. I am grateful to Marty Lederman, Jennifer Nou, Catherine Roach, and Jed Stiglitz for helpful and thought-provoking comments on earlier drafts. Any remaining errors or infelicities are, of course, my own.