Federal courts cited Chevron U.S.A. v. Natural Resources Defense Council over 18,000 times in the four decades before its downfall in Loper Bright Enterprises v. Raimondo. But Chief Justice Roberts promised that Loper Bright did “not call into question prior cases that relied on the Chevron framework,” which counseled courts to defer to reasonable agency interpretations of ambiguous statutes. “The holdings of those cases that specific agency actions are lawful . . . are still subject to statutory stare decisis” — Latin for “to stand by things decided” — “despite the Court’s change in interpretive methodology.” Justice Kagan in dissent, like many lawyers and scholars, was not convinced: “The majority is sanguine; I am not so much.”
Lower courts are now implementing that stare decisis instruction. In doing so, they have differed in their interpretations of Chief Justice Roberts’s sparse instruction. Those differences will determine how many long-settled regulations will be up for grabs, or even if Chevron can live on after death. In a forthcoming piece outlining possible approaches to Chevron precedents, Professor Jonathan Remy Nash argues that stare decisis for Chevron decisions in any form will produce inconsistency as those agencies lucky enough to have had their interpretations judicially reviewed under Chevron will enjoy different interpretive privileges than the rest.
So how have lower courts been approaching Loper Bright’s statutory stare decisis instruction? Two case studies from the Ninth and Sixth Circuits reveal the tension between a narrow reading of Loper Bright — that only “specific agency actions” upheld under Chevron are safe — and a broader reading of Chief Justice Roberts’s instruction that would account for Loper Bright’s worry about agency flip-flopping, the Court’s “congressional acquiescence” justification for statutory stare decisis, and the reliance rationale for regular stare decisis. Those three concerns are important to weigh against Professor Nash’s inconsistency concern. Although only one lower court has discussed congressional acquiescence in-depth at all post–Loper Bright, the appropriate application of the stare decisis instruction is a principled rather than mechanical one.
Where lower courts come out will also impact the incoming Trump Administration’s plans to roll back or modify a host of regulations. Applying the Ninth Circuit’s view in Lopez v. Garland (described below) would permit agencies to have one last chance to flip-flop on their interpretation of a statute, while locking in that switch as the “best” interpretation of the statute, as Judge Benjamin Beaton of the U.S. District Court for the Western District of Kentucky indicated at a conference last month.
Chevron and Statutory Stare Decisis
But first, a review of Chevron and statutory stare decisis. Chevron proceeded in two “steps”: At Step One, a court would ask whether the statute being interpreted was “silent or ambiguous with respect to the specific issue.” If it was, the court would proceed to Step Two and uphold the agency’s interpretation as long as it was within a range of “reasonable” interpretations. Importantly for understanding stare decisis for Chevron decisions, the Chevron-era decision National Cable & Telecommunications Association v. Brand X Internet Services permitted agency flip-flopping under Chevron, holding that agencies are free to switch between reasonable interpretations under Chevron as long as they adequately explain the change. In overturning Chevron, Loper Bright also castigated Brand X and Chevron as “destroy[ing]” “reliance interests,” arguing that permitting flip-flopping “fosters unwarranted instability in the law, leaving those attempting to plan around agency action in an eternal fog of uncertainty.”
As Professor Nash explains, the statutory stare decisis question is only relevant to holdings that reached Step Two — if the statute was unambiguous, the court’s unambiguous reading of the statute continues to control. Within that universe, Professor Nash hypothesizes that on one view, the Court’s “specific agency action” language “locks in” the initial interpretation upheld by the earlier Chevron decision. He argues it is unlikely that the Court intended that outcome, and posits three alternative interpretations:
Stare decisis could (i) insulate the entire pre-Loper Bright decision by providing the agency access to the entire range of reasonably possible interpretations, (ii) provide the agency with the choice between the interpretation previously upheld as lawful and the best interpretation (i.e., the interpretation that Loper Bright would otherwise mandate), or (iii) allow the agency to choose any reasonable interpretation between the interpretation previously upheld as lawful and the best interpretation [i.e., closer to the best even if not the best].
Before diving into those options, a brief summary of stare decisis: Stare decisis is a legal doctrine dictating that a court needs a special justification to overturn a prior decision made by the same court (horizontal stare decisis) or a higher court (vertical stare decisis) — they cannot do so just because the earlier decision was wrongly decided (though, post-Dobbs, perhaps they can if it is “egregiously wrong”). The doctrine “promotes the evenhanded, predictable, and consistent development of legal principles, fosters reliance on judicial decisions, and contributes to the actual and perceived integrity of the judicial process.” Nearly all circuits have horizontal stare decisis rules — many of which are even stronger than the Supreme Court’s — that bar a three-judge panel of the circuit from overturning a prior panel’s decision. Circuit courts sometimes, however, make exceptions in the case of an inconsistent Supreme Court decision.
Stare decisis carries “enhanced force” when the prior decision interpreted a statute rather than the Constitution, because “Congress can correct any mistake” the Court makes in statutory interpretation. Statutory stare decisis has added weight if the Court’s interpretation has stood for a long time without congressional correction, particularly if Congress has amended the statute without disturbing the Court’s interpretation.
However, the circuits are split over whether their own statutory precedents deserve enhanced force. A 2005 law review article by now-Justice Barrett, recently rediscovered by some circuit courts, argues that appellate courts should not apply statutory stare decisis. Justice Barrett contends that the acquiescence justification does not apply to courts of appeals given that lower courts cannot expect Congress to respond to appellate court decisions to the same extent that it responds to Supreme Court decisions. Two recent en banc panels have declined to apply super-strong statutory stare decisis to their precedents — the Fifth Circuit and, crucially for administrative law, the D.C. Circuit. To varying degrees, the First, Second, Ninth, Eleventh, and Federal Circuits have gone the other way, applying heightened statutory stare decisis to their precedent.
The number of regulations at risk will balloon if appeals courts decide that statutory stare decisis applies only to Supreme Court precedent. In Loper Bright, the government listed seventy Supreme Court cases that relied on Chevron — and the agencies lost in approximately one-quarter to one-third of those cases. A study of circuit court cases decided between 2003–2013 (a quarter of Chevron’s reign) found a little over 750 decisions that upheld agency action at Step Two. The answer to the question of whether Loper Bright’s statutory stare decisis instruction protects approximately fifty Supreme Court decisions, or additionally the likely thousand(s) more appellate decisions, will thus produce the biggest swings in outcomes for the stability of the administrative state post–Loper Bright.
Lopez v. Garland: “Specific agency action” and lock-in
The Ninth Circuit’s September 2024 decision in Lopez v. Garland exemplifies the second approach Professor Nash hypothesizes: that an agency can choose to stick with its prior interpretation or abandon it in favor of the “best” interpretation. The case also demonstrates how the lock-in approach might best align with the structural justifications for super-strong statutory stare decisis.
First, the Ninth Circuit upheld a Board of Immigration Appeals decision to stick with a prior interpretation upheld under Chevron. A noncitizen may be deported under 8 U.S.C. § 1227(a)(2)(A)(ii) if “convicted of two or more crimes involving moral turpitude, not arising out of a single scheme of criminal misconduct.” In 2019, the Ninth Circuit in Szonyi v. Whitaker deferred under Chevron to the Board’s interpretation of “single scheme.” Christian Lopez petitioned the Ninth Circuit to review a Board decision finding Lopez removable after the Board applied its definition of “single scheme” upheld in Szonyi. Lopez argued the Ninth Circuit should return to its pre-Szonyi definition of the phrase, which would have regarded his crimes as “arising out of a single scheme” and thus excluded him from removal under the statute. On a straightforward application of Loper Bright’s stare decisis instruction, the Lopez court rejected Lopez’s request to overturn the Ninth Circuit’s prior endorsement of the still-extant agency interpretation.
Second, the Lopez court permitted the Board to abandon its prior interpretation of the phrase “crimes involving moral turpitude” and independently assessed the Board’s new interpretation. A 2009 Ninth Circuit panel in Castillo-Cruz v. Holder accorded “substantial deference” to the Board’s then-interpretation that theft offenses involve “moral turpitude” only when committed with intent to permanently deprive an owner of their property. The Lopez court considered the question anew, however, after the Board added in 2016 that a crime can involve moral turpitude if committed with either intent to deprive permanently or “under circumstances where the owner’s property rights are substantially eroded.” That addition, the Board noted, marked a change from the permanent-deprivation-only approach that the Board had maintained from its “earliest days” — citing decisions going back to the 1940s. Though “evaluat[ing] the statute independently,” the Lopez court ultimately upheld the new interpretation under Skidmore v. Swift & Co., according it “due respect” given its “thoroughness [and] persuasive reasoning,” though not deferring to it as binding authority.
That approach could allow the Trump Administration to take a “first crack” at reinterpreting statutes. If a litigant challenges a new interpretation of a statute where a prior court had upheld an alternate interpretation under Chevron, the Lopez approach would consider the question anew without respect to the decision on the books. Theoretically, regardless of the litigants’ proposed interpretations, a court will apply the “best” interpretation of the statute, and thus which revised interpretation a court considers first should not matter. But if a court applies Skidmore and affords the agency’s proffered interpretation some respect, then which interpretation the court considers first will matter as the Skidmore analysis might introduce daylight between the “best” interpretation and the interpretation offered. Yet the court’s holding will still deem that new interpretation the “best” one, given Loper Bright instructed that Skidmore respect can speak only to the question of the “best” interpretation, thus permitting no further flip-flopping after that one final flop.
A lock-in approach to the “moral turpitude” issue might have instead looked like the following: Pre–Loper Bright, the Ninth Circuit read Castillo-Cruz as binding Ninth Circuit courts to all of the Board’s future reasonable interpretations of “moral turpitude,” rather than to the specific Board interpretation at issue in that case. If the Board switched to a different reasonable interpretation, the Ninth Circuit would change with it. But on the post–Loper Bright lock-in view, Castillo-Cruz adopted the substance of the underlying interpretation — i.e., Castillo-Cruz held, effectively: “For theft crimes, we interpret ‘moral turpitude’ to require intent to deprive permanently.” The Board’s adopting an alternate interpretation might be helpful reasoning for an en banc panel (rather than Lopez’s three-judge panel) considering whether to overrule fifteen-year-old precedent, but given the court’s recognition that Skidmore deference is non-binding, that new interpretation cannot serve as “higher authority” that wipes clean the precedential slate. The longstanding nature of the Board’s original interpretation, under the congressional acquiescence justification for statutory stare decisis, would also militate in its favor. The Lopez court did not, however, conduct such statutory stare decisis analysis.
Professor Nash argues that the lock-in approach would inappropriately elevate Chevron decisions to become “precedent on steroids,” enabling the “precedential value of [Chevron decisions to] sweep far beyond [their] . . . scope under Chevron itself,” given Brand X permitted agencies to change their mind even after a court had upheld their interpretation under Chevron. But, as noted, Loper Bright wanted to mitigate, not preserve, agency flip-flopping. Take Lopez: there are potentially very strong reliance interests in individuals residing in the United States who were not subject to removal for decades until the Board’s 2016 switch. Even during Chevron’s reign, Brand X was a uniquely disfavored opinion — its author, Justice Thomas, even sought to abandon it in 2020: “[I]t is never too late to ‘surrende[r] former views to a better considered position.’” Lopez demonstrates how the reliance principle behind stare decisis would support binding an agency to a prior interpretation, barring some special justification for changing course.
Tennessee v. Becerra: Range of reasonable interpretations
But, where Congress acquiesced in agency flip-flopping on a specific statute, and where reliance interests are weak, that acquiescence could serve as a principled reason to permit an agency to continue flip-flopping.
The Sixth Circuit’s decision in Tennessee v. Becerra is an example of Professor Nash’s first hypothesized approach, that agencies may adopt any interpretation deemed reasonable by the earlier court. Title X of the Public Health Service Act authorizes the Department of Health and Human Services (HHS) to “make grants to . . . public . . . entities” to operate “family planning projects,” and to “promulgate” “regulations” and establish “conditions” related to such grants — but any “funds appropriated” may not “be used in programs where abortion is a method of family planning.” In Rust v. Sullivan, the Supreme Court held in 1991 that the prohibition on the use of funds for “programs where abortion is a method” was ambiguous as to whether such programs can regardless provide abortion counseling and related services. The Court thus deferred under Chevron to a Reagan Administration HHS regulation that barred such abortion-related services in Title X–funded programs. Rust was an earlier version of Brand X — the HHS regulation at issue broke with eighteen years of agency practice and regulations that did not bar abortion-related services.
HHS continued flip-flopping post Rust. In 2023, the Sixth Circuit in Ohio v. Becerra relied on Rust in holding that a new Biden Administration HHS regulation requiring entities receiving Title X funding to provide such services (rather than barring them from doing so) was permissible. The Ohio court argued that because Rust concluded at Chevron Step One that the statute “does not speak directly to” the question of abortion-related services, “it must be permissible” for HHS to treat them as either barred or permitted by the statutory restriction on the use of funds “where abortion is a method.”
Under a different fact pattern, and now with Chevron in the dustbin, the State of Tennessee challenged the same HHS regulation at issue in Ohio after HHS sought to withhold Tennessee’s Title X funding because the state refused to provide abortion referrals. The district court denied Tennessee’s motion for a preliminary injunction against HHS enforcing the new rule. The Sixth Circuit affirmed. In evaluating Tennessee’s likelihood of success on the merits, the Sixth Circuit grappled with the stare decisis effect of both Rust and Ohio. The court acknowledged that both precedents were imperfect as binding authorities: Rust did not address the “specific agency action” at issue; and, citing Justice Barrett’s article, the court recognized that its own statutory precedent may also not be binding. However, the court ultimately seemed to find that the combination of the two did produce binding authority.
Judge Kethledge in dissent noted that Rust did not hold lawful the “specific agency action” at issue in Tennessee, and thus there was “no occasion to defer to that holding.” He critiqued the government’s more flexible stare decisis approach: “[I]n the Department of Justice’s view, apparently, Chevron lives on in perpetuity as to any statute that the Supreme Court has ever deemed ambiguous under that doctrine.”
The justifications for stare decisis could provide a more principled defense of the majority’s hand-wavy application of stare decisis. Title X is a nearly paradigmatic example of congressional acquiescence: Congress has considered hundreds of bills about Title X, including attempts to specifically foreclose the Rust issue, and enacted upwards of twenty into law, all without successfully amending the language at issue. Particularly given the general popular awareness of the flipping and flopping, Congress acquiesced in the Rust Court’s inferring an implicit delegation of authority to HHS to decide between reasonable interpretations. Reliance interests, moreover, do not counsel for the plaintiffs: Tennessee argued that the court should consider the state’s reliance on Title X funding that they had enjoyed for fifty years, which was at risk of being rescinded after Tennessee refused to provide abortion referrals. The Sixth Circuit reminded, however, that states have “no legally cognizable reliance interest in the receipt of a discretionary funding award on the conditions that it prefers.”
Interestingly, the Supreme Court turned down a chance to reject the “range of reasonable interpretations” approach. The Tenth Circuit in Oklahoma v. U.S. Department of Health & Human Services reached a similar conclusion to the Tennessee court just two weeks after Loper Bright was decided. In justifying its reliance on Rust, the Tench Circuit implied that the relevant part of the Court’s holding for stare decisis purposes was the finding that HHS enjoyed discretion: “[T]he Supreme Court held that HHS could enact requirements on counseling and referrals.” Oklahoma petitioned the Supreme Court for emergency relief, complaining in part that the Tenth Circuit “ignored” Loper Bright’s “specific agency action” language. But the Court denied Oklahoma’s petition — though Justices Thomas, Alito, and Gorsuch would have granted it.
Principles over Mechanics
Professor Nash worries that any stare decisis approach would produce disuniformity across administrative law: “Some agencies will have discretion to choose among statutory interpretations while other agencies will not, and indeed each agency will have that discretion only with respect to some provisions and not others.” But this is an inconsistency that only administrative scholars could hate. Viewed at the Chevron and Loper Bright level, administrative law is a procedural, not substantive, area of the law — governing a vast array of economic and social systems by generally applicable rules. Meanwhile, most of the practitioners of administrative law (notwithstanding administrative law professors and the forty-five career staff of the Office of Information and Regulatory Affairs) are domain experts across agencies, industries, and law firms. These folks, I imagine, as well as the Americans who are impacted by the substantive rules that make up administrative law, worry more about consistency in the substantive rules themselves, not whether an immigration rule was treated procedurally the same as an abortion rule. Consistency viewed at the procedural level, moreover, depends on which aspects of an agency or rule it seems fair to consider relevant when deciding whether one rule is “like” another — the principles undergirding statutory stare decisis can be consistently applied.
A faithful reading of the Loper Bright stare decisis instruction would presumably involve more consideration of the policy and structural justifications for stare decisis. Such an approach would focus less on formalist adherence to a particular formula for which prior interpretations bind and which do not, and would also make no brightline distinction between appellate precedent and Supreme Court precedent. Rather, it would investigate legislative history to determine whether Congress took notice of and acquiesced in a given regulation and consider reliance interests in the previously upheld regulation. The statutory stare decisis inquiry will thus overlap partly with the separate Skidmore inquiry regarding how longstanding an interpretation has been. But the former will focus on the precedent decision, and the latter on the regulation itself.
That, however, has not been the circuit courts’ approach. It is unsurprising that lower courts have been inconsistent in their reading of, and adherence to, the Chief Justice’s stare decisis instruction. The Court itself has set a poor example in this regard — as Professor Khiara Bridges observed in her Foreword assessing the 2021 Term: “[T]he Roberts Court does not appear to consider itself particularly bound by stare decisis.” Although the justifications for statutory stare decisis thus may offer a more principled way for lower courts to untangle difficult post–Loper Bright puzzles, lower courts so far have preferred formalism or hand-waving over wrangling with difficult policy or historical questions.
The post <strong>How Much of the Regulatory State is Safe Post–<em>Loper Bright</em>?</strong> appeared first on Harvard Law Review.