Judge Kavanaugh on administrative law and separation of powers (Corrected)

Christopher J. Walker is a law professor at The Ohio State University Moritz College of Law.

Administrative law sets the ground rules for how federal agencies regulate and how courts review and constrain such agency action. Administrative law’s importance in our everyday lives has become even more pronounced in recent decades with the rise of regulation and the decline of legislation. To provide just one imperfect snapshot, in 2015 and 2016 federal agencies promulgated more than 7,000 final rules filling more than 60,000 pages in the Federal Register. During that same time, by contrast, the 114th Congress enacted just 329 public laws filling about 3,000 pages in the Statutes at Large.

By congressional design, the U.S. Court of Appeals for the District of Columbia Circuit is the nation’s pre-eminent administrative law court and arguably “the second most important court” overall, after the Supreme Court. And D.C. Circuit Judge Brett Kavanaugh — President Donald Trump’s nominee to the Supreme Court — is one of the most sophisticated, provocative and creative voices in the federal judiciary when it comes to administrative law.

During his dozen years of service on the D.C. Circuit, Kavanaugh has authored around 300 opinions, more than a third of which deal with administrative law. As detailed below, Kavanaugh has written numerous major administrative law opinions — including dissents and concurrences, which Aaron Nielson has collected here — and the Supreme Court has embraced his approach on a number of occasions.

In reviewing Kavanaugh’s robust record on administrative law, I find myself agreeing with Jonathan Adler’s conclusion that a Justice Kavanaugh would not bring to the Supreme Court a commitment, in Steve Bannon’s words, to the “deconstruction of the administrative state.” He cares deeply about administrative law and regulatory practice. But he would likely “put a tighter leash on the regulatory state” — a tightening that would generally apply to regulation and deregulation alike.

That is because Kavanaugh’s decisions on the D.C. Circuit, coupled with his other writings, reveal a judge who takes separation of powers seriously. For Kavanaugh, agency regulatory authority comes from and is constrained by Article I, in that “[a]ll legislative powers herein granted shall be vested in a Congress of the United States.” Similarly, the modern administrative state functions against the Article II backdrop that “[t]he executive power shall be vested in a President of the United States of America.” These constitutional separation-of-powers values deeply influence Kavanaugh’s approach to administrative law.

Although not a comprehensive survey of his administrative law jurisprudence, the following in-depth look at three areas captures some of Kavanaugh’s major contributions to administrative law and assesses his potential impact on the federal regulatory state if he is elevated to the Supreme Court.

1. Chevron (and Auer) deference

In recent years, there has been a growing call (mainly from those right-of-center) to eliminate — or at least narrow — administrative law’s judicial-deference doctrines regarding federal agency interpretations of law.

These reform efforts have been front and center at the Supreme Court. For example, in 2015, in Perez v. Mortgage Bankers Association, Justices Antonin Scalia, Clarence Thomas and Samuel Alito all questioned the wisdom and constitutionality of judicial deference to agency interpretations of their own regulations (Auer deference). And this term in Pereira v. Sessions, Justice Anthony Kennedy joined the prior calls by Thomas and then-Judge Neil Gorsuch to reconsider “reflexive deference” to agency statutory interpretations (Chevron deference).

Just as it was at Gorsuch’s confirmation hearing, I expect Chevron deference (and perhaps Auer deference) to be discussed at Kavanaugh’s hearing. A recent Mother Jones headline aptly summarizes one potential line of attack: “How Brett Kavanaugh Could Cripple the Next Democratic President. Two words: Chevron deference.”

So how would a Justice Kavanaugh affect Chevron deference’s future at the Supreme Court? The potential impact is threefold.

First, as a textualist, Kavanaugh would likely find statutes unambiguous more often than some of his more-purposivist peers who tend to interpret statutes in accordance with the statute’s purpose (and more often than his predecessor, Kennedy), and thus be less likely to defer to agency statutory interpretations. The role of ambiguity is critical to Chevron deference. After all, Chevron v. Natural Resources Defense Council commands that a reviewing judge defer to an agency’s interpretation of a statute the agency administers if (1) the statutory provision at issue is ambiguous and (2) the agency’s interpretation is reasonable.

As Kent Barnett and I have empirically explored in the circuit courts, the ambiguity inquiry at Chevron’s first step is far more exacting than the reasonableness inquiry at the second step. In our 11-year dataset of every published circuit-court decision that cites Chevron deference, we found that agencies prevailed under the Chevron doctrine 93.8 percent of the time when the court found the statute ambiguous and reached step two, but only 39.0 percent of the time when the court found the statute unambiguous and thus stopped at step one.

Kavanaugh has written extensively about the role of ambiguity in statutory interpretation. Most famously, he set forth his concerns in a Harvard Law Review essay reviewing Judge Robert Katzmann’s book on statutory interpretation. There, he argued that “judges often cannot make that initial clarity versus ambiguity decision in a settled, principled, or evenhanded way.” (For those interested in this debate, I highly recommend reading Katzmann’s purposivist reply.)

Kavanaugh himself has recognized that his textualist orientation will likely result in his finding fewer statutes ambiguous under Chevron than some of his judicial peers. As he observed in his Story Lecture last year, whereas some judges might require 90 percent certainty to declare a statute unambiguous, “I probably apply something approaching a 65/35 or 60/40 rule. In other words, if it is 60/40 clear, it is not ambiguous, and I do not resort to [Chevron deference].”

Accordingly, one should expect him to approach Chevron’s first step in a textualist fashion similar to Scalia’s, in which he exhausts all of the tools of statutory interpretation at Chevron step one to resolve the ambiguity. Or as Gorsuch framed it this term in Wisconsin Central Ltd. v. United States, a “clear enough” — as opposed to, perhaps, a crystal clear — Chevron step one inquiry. Kavanaugh’s opinion in Loving v. Internal Revenue Service is a good example of this approach. There, he relied on “the text, history, structure, and context of the statute” to reject the IRS’ interpretation of the statutory text “regulate practice of representatives of persons before the Department of Treasury” to include the authority to regulate tax preparers.

Second, Kavanaugh has advanced in his academic writings a more-systemic narrowing of Chevron deference based on concerns about uniformity of federal law and partisanship in judicial decision-making. As he explained in his Story Lecture, he finds the threshold ambiguity inquiry under Chevron problematic because his “goal is to help make statutory interpretation … a more neutral, impartial process where like cases are treated alike by judges of all ideological stripes, regardless of the issue and regardless of the identity of the parties in the case.”

So Kavanaugh proposes preserving agency deference “in cases involving statutes using broad and open-ended terms,” but perhaps eliminating it “where an agency is instead interpreting a specific statutory term or phrase.” That is because, in the latter instance, “[j]udges are trained to do that, and it can be done in a neutral and impartial way in most cases.” As Kavanaugh concludes, “the problem with certain applications of Chevron, as I see it, is that the doctrine is so indeterminate — and thus can be antithetical to the neutral, impartial rule of law — because of that initial clarity-versus-ambiguity decision.”

It is difficult to assess Kavanaugh’s proposal in the abstract. Perhaps he is suggesting a total elimination of Chevron deference when dealing with specific statutory ambiguities as opposed to open-ended delegations (that do not implicate major economic or political questions, more on that below) — though that line is often difficult to discern, much less draw. Or maybe this is just another way to articulate Kavanaugh’s textualist, Scalia-esque approach to Chevron step one. Or perhaps he is echoing Kennedy’s concerns from his Pereira dissent about how Chevron deference “has come to be understood and applied,” with “[t]he type of reflexive deference exhibited in some of these cases.” This strikes me as a fruitful line of inquiry for Kavanaugh’s confirmation hearing.

One final note: If Kavanaugh is concerned about administrative law’s political dynamics, the right prescription may be the opposite: Preserve a bright-line Chevron doctrine. In the latest article from our Chevron dataset, Kent Barnett, Christina Boyd and I find that, at least in the circuit courts, Chevron deference has a powerful effect on constraining partisanship in judicial decision-making and encouraging uniformity in federal law — the values that seem to motivate Kavanaugh in his academic writing. In our dataset (2003-2013), Kavanaugh largely applied the same approach to Chevron deference regardless of whether the agency interpretation under review was conservative or liberal. But that was not true for all conservative and liberal judges in our dataset. And perhaps this partisanship he sees in other judges’ application of Chevron deference is what is driving Kavanaugh’s concerns here.

Third, Kavanaugh has embraced a strong version of another means of narrowing Chevron deference: the major questions doctrine.

King v. Burwell, the statutory challenge to the Affordable Care Act, is a recent and prominent example of the major questions doctrine. In a 6-3 decision written by Chief Justice John Roberts, the Supreme Court found the statutory language ambiguous. But the court refused to apply any deference to the agency’s interpretation of the statutory ambiguity. Instead, the chief justice invoked the major questions exception to Chevron deference because the statutory provision implicated “a question of deep economic and political significance that is central to this statutory scheme” and for which the agency (the IRS) had no expertise.

Kavanaugh looked to King v. Burwell when the Federal Communications Commission’s net-neutrality regulation reached the D.C. Circuit in United States Telecom Association v. FCC. In his dissent from the denial of rehearing en banc, he argued that “[i]f the Supreme Court’s major rules doctrine means what it says, then the net neutrality rule is unlawful because Congress has not clearly authorized the FCC to issue this major rule.”

As Jeff Pojanowski has observed, Kavanaugh’s version of the major questions doctrine, which Kavanaugh relabeled the major rules doctrine, “came with a twist”:

After canvassing the Supreme Court’s jurisprudence and scholarly commentary, [Judge Kavanaugh] identified what he dubbed the “major rule” exception to Chevron deference. He saw this Chevron carve-out as holding that if “an agency wants to exercise expansive regulatory authority over some major social or regulatory activity … an ambiguous grant of statutory authority is not enough.” … Rather than hiding regulatory elephants in mouse holes, Congress can extend the reach of the administrative state only through clear statements.

Dan Deacon has argued that Kavanaugh’s approach is a “weaponized” version of the doctrine that, absent a clear congressional statement to the contrary, strips away not only Chevron deference for major questions but also any agency authority to regulate concerning those major questions. “[T]he ‘major rules’ doctrine might extend to actions that ‘de-regulate’ as well as regulate,” Deacon observes, “[b]ut the overall logic and tenor of [Kavanaugh’s] argument is largely anti-regulatory.” Eric Citron has a similar take in his contribution to SCOTUSblog’s coverage of Kavanaugh.

In that sense, Kavanaugh’s major rules doctrine is a second-order means of addressing nondelegation doctrine concerns. As noted at the outset, Kavanaugh’s view of separation of powers — and, in particular here, Article I’s nondelegation command that Congress cannot delegate legislative powers to federal agencies (or anyone else) — motivates his administrative law jurisprudence. Cass Sunstein, among others, has noted that the Supreme Court has seldom used the nondelegation doctrine to strike down a statute, largely because of line-drawing problems. Kavanaugh’s major rules doctrine attempts to address nondelegation concerns through a substantive canon of statutory interpretation instead of a constitutional doctrine, by establishing an interpretive presumption that Congress does not intend to delegate rulemaking authority over questions of major economic or political significance absent a clear congressional statement to the contrary.

Because Kavanaugh expounded this major rules doctrine in a dissent, its precise contours are understandably not fully developed. But I concur in Pojanowski’s assessment that “Judge Kavanaugh’s careful explication and reformulation of the ‘major questions’ exception is an important development in its own right, and a rich source for further reflection on the role of the courts in the administrative state.” It is certainly another fascinating line of inquiry for his confirmation hearing, especially when considered in conjunction with his proposal discussed above to limit Chevron deference to open-ended congressional delegations.

In sum, Kavanaugh’s approach to Chevron deference in practice would likely be quite similar to Scalia’s textualist approach at step one. He has also expressed concerns similar to his potential predecessor (Kennedy) about how the doctrine has become “reflexive deference” in practice, perhaps signaling a desire to cabin Chevron’s domain. In light of how he has embraced the major questions doctrine, it would be unsurprising to see a Justice Kavanaugh join the chief justice’s calls for a narrower, more context-specific Chevron deference.

Although Kavanaugh has not addressed the propriety of Auer deference to an agency’s interpretation of its own regulations, his concerns about interpretive doctrines that turn on ambiguity, coupled with his views on separation of powers, seem to suggest he would be receptive to calls to eliminate — or at least further limit — Auer deference. We may find out the answer as soon as this coming term, as a pending cert petition asks the court to overrule Auer v. Robbins.

Kavanaugh’s dissent in the net-neutrality regulation case, moreover, provides some fascinating clues for how a Justice Kavanaugh might address nondelegation and separation-of-powers concerns more generally. Again, we may learn more about Kavanaugh’s views on nondelegation doctrine as early as this term, when the Supreme Court decides Gundy v. United States, which raises a nondelegation challenge.

2. APA hard-look review

In SCOTUSblog’s coverage of Kavanaugh, Edith Roberts and Michael Livermore have already surveyed Kavanaugh’s approach to hard-look review of agency action under the Administrative Procedure Act.

Simply put, Kavanaugh has embraced the Supreme Court’s instruction in Motor Vehicles Manufacturers Association v. State Farm to invalidate an agency’s rule when:

the agency has relied on factors which Congress has not intended it to consider, entirely failed to consider an important aspect of the problem, offered an explanation for its decision that runs counter to the evidence before the agency, or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise.

Perhaps the most prominent example is Kavanaugh’s dissent in White Stallion Energy Center v Environmental Protection Agency. There he argued that the statutory language “appropriate and necessary” required the EPA to consider costs — an argument the Supreme Court ultimately embraced in its 5-4 decision in Michigan v. EPA.

As Roberts details, Kavanaugh’s approach to hard-look review does not always result in agency losses. One of his more prominent APA decisions is American Trucking Associations v. EPA, in which he authored the 2-1 majority opinion upholding the EPA’s authorization of California’s rule limiting emissions from in-use non-road engines. Judge Stephen Williams dissented, arguing that the EPA failed to engage in APA-required reasoned decision-making — an argument Kavanaugh rejected because he found the EPA had reasonably interpreted and considered the statutory criteria.

In Kavanaugh’s opinions concerning the APA arbitrary-and-capricious standard, one finds a judge who takes hard-look review seriously. He scrutinizes the agency’s rulemaking process to ensure that the agency has considered the relevant statutory factors, responded to counterarguments and evidence, and otherwise engaged in reasoned decision-making within the agency’s statutory authority. Agencies sometimes win and sometimes lose, and sometimes an agency loss is a “liberal” or “conservative” win.

Although I tend to view Kavanaugh’s approach to APA review as more principled and consistent than Livermore does, Livermore’s bottom line may well still be true: “[G]iven congressional gridlock, notice-and-comment rulemaking by environmental agencies has become the primary vehicle for environmental progress over the past several decades. [Replacing Kennedy with Kavanaugh] will make this path more difficult, halting and fraught with risk.”

And this prediction may be true not just for environmental cases but for administrative law more generally. To the extent federal agencies attempt to repurpose old statutes to address new problems in ways that stretch or distort those statutes, a Justice Kavanaugh is unlikely to vote to defer to the agency — and even less so, as discussed above, in matters of major economic or political significance. He has a long record of constraining agency action within what he perceives as the limits of the statutory text. Here, again, we see Kavanaugh’s separation-of-powers vision at play in that regulatory authority comes from Congress and is constrained by Article I nondelegation values.

There is one wrinkle that probably deserves a separate, extended write-up: Kavanaugh has shown some interest in APA originalism. In recent years, some scholars have called for a return to the text of the APA and an accompanying abandonment of administrative common law that is inconsistent with the original meaning of the APA. The Supreme Court has at times embraced this APA originalism, arguing that courts cannot require federal agencies to utilize procedures not expressly required by the APA itself. Vermont Yankee v. NRDC and Perez v. Mortgage Bankers come immediately to mind.

In light of his textualist approach to statutory interpretation and his separation-of-powers vision, it should come as little surprise that Kavanaugh has been relatively receptive to APA originalism. Consider American Radio Relay League, Inc. v. FCC. At issue was the D.C. Circuit’s Portland Cement doctrine, which requires federal agencies to disclose the technical data and studies on which they relied to draft the proposed rule. Although noting that this doctrine “may make sense as a policy matter in some cases,” Kavanaugh wrote separately to express his concerns that the doctrine is inconsistent with the text of the APA.

As a policy matter, I tend to agree with Dan Farber that the Portland Cement doctrine serves an important purpose in notice-and-comment rulemaking. (Though, it should be noted, some progressives have argued against the doctrine because it could put science on trial and discourage scientists from publicly sharing their datasets and models. Conversely, many regulated entities no doubt appreciate the doctrine as one more hurdle to trip up an agency on judicial review and thus further delay rulemaking.) There are compelling commonsense and nonpartisan policy reasons for the doctrine, and it is thus no surprise that the American Bar Association has urged Congress to amend the APA to include such a disclosure requirement.

Kavanaugh’s concern, however, is not about policy but law—in particular, whether a court can impose such a requirement that he perceives is lacking in the APA itself. The Portland Cement doctrine is just the tip of the iceberg when it comes to APA originalism. Administrative common law arguably includes a number of central administrative law doctrines, including Chevron and Auer deference, the presumption of reviewability and remand without vacatur — just to name a few. And, for present purposes, hard-look review under the APA.

It’s doubtful there are five (or even four, or three, or two) votes on the Supreme Court to engage in a full-fledged APA originalism project. It isn’t even clear that Kavanaugh would be committed to such a project, especially for bedrock doctrines like hard-look review and deference doctrines for which statutory stare decisis is arguably quite weighty. But this is another area to watch in the years to come, especially if Kavanaugh joins the Supreme Court.

3. Presidential control of independent agencies

For administrative law scholars and practitioners, it should come as no surprise which decision Kavanaugh lists in his Senate questionnaire as his most significant: his dissent in Free Enterprise Fund v. Public Company Accounting Oversight Board.

There, Kavanaugh argued that the PCAOB’s removal provisions were unconstitutional as contrary to the president’s Article II removal authority because of the double-insulation protection. In particular, PCAOB board members were only removable “for cause” by the Securities and Exchange Commission (not the president), and SEC commissioners were only removable “for cause” by the president. The 5-4 conservative majority on Supreme Court ultimately agreed with Kavanaugh’s dissent.

Kavanaugh may consider this dissent his most significant opinion on the D.C. Circuit, but it arguably isn’t even his most significant dissent when it comes to presidential control of independent agencies. As Jenn Mascott and Aaron Nielson detail elsewhere, “Judge Kavanaugh sounded similar themes in PHH v. CFPB, twice, in fact. There, he read Article II to prohibit another arguably novel agency structure—this one created by the Dodd-Frank Act.” In a panel opinion subsequently vacated by the D.C. Circuit en banc, Kavanaugh found unconstitutional the structure of the Consumer Financial Protection Bureau — an independent agency headed by a single director, as opposed to a multi-member commission, who was only removable for cause by the president. As he observed in the panel opinion, “That combination of power that is massive in scope, concentrated in a single person, and unaccountable to the President triggers the important constitutional question at issue in this case.” He thus severed the for-cause removal protection from the statutory scheme.

Last month a district judge in the Southern District of New York adopted Kavanaugh’s position, ruling that the CFPB cannot bring an enforcement action in district court because of its unconstitutional structure. Indeed, Judge Pleska went one step further and held that the for-cause removal provision is not severable. And earlier this month the U.S. Court of Appeals for the 5th Circuit found one of the few other single-director independent agencies (the Federal Housing Finance Agency) unconstitutional, on grounds similar to those articulated by Kavanaugh.

Kavanaugh’s views on Article II presidential control of federal agencies, and in particular of so-called independent agencies, could implicate a number of important administrative law issues that may reach the Supreme Court in the near future. Efforts to reconsider the political independence of administrative law judges — and perhaps the federal civil service more generally — are working their way through the administrative state. The White House may extend presidential review of agency rulemaking to cover independent agencies. And, of course, questions abound, at least as an academic matter, as to presidential control of a special counsel — a matter that exceeds the ambitions of this post but has been summarized by Kevin Russell in this SCOTUSblog series.

To be sure, Kavanaugh’s separation-of powers opinions do not directly address these issues — though some have strained to read “wolves” into Kavanaugh’s footnotes to cast doubt on the future of independent agencies writ large. If and when these issues reach the Supreme Court, however, one should expect a Justice Kavanaugh to have a unique and sophisticated take, one that takes into account the Article I and Article II separation-of-powers values that influence his administrative law jurisprudence.

Note: A prior version of this post erroneously stated that the Administrative Conference of the United States had joined the American Bar Association to urge Congress to amend the APA to include the Portland Cement doctrine disclosure requirement.

[Disclosure: Goldstein & Russell, P.C., whose attorneys contribute to this blog in various capacities, was among the counsel for the intervenors in United States Telecom Association. The author of this post is not affiliated with the firm.]