Introduction

The Supreme Court’s upcoming confrontation with the Chevron doctrine invites an audit: how have Chevron U.S.A. Inc. v. Natural Resources Defense Council, Inc.’s¹467 U.S. 837 (1984). rationales developed since 1984, and how might they endure after 2024? This Article focuses on Chevron’s expertise rationale, especially as applied to energy and environmental law—Chevron’s birthplace, and Loper Bright Enterprises v. Raimondo’s²45 F.4th 359 (D.C. Cir. 2022), cert. granted in part, 143 S. Ct. 2429 (argued Jan. 17, 2024) (mem.). lineage.³See id. at 363 (granting summary judgment based on the Agency’s reasonable interpretation of its statutory authority). For simplicity, I refer throughout to Loper Bright Enterprises v. Raimondo, acknowledging that Relentless, Inc. v. U.S. Department of Commerce, 62 F.4th 621 (1st Cir. 2023), cert. granted in part, 144 S. Ct. 325 (argued Jan. 17, 2024) (mem.), offers the same question presented. Ultimately, the waning of Chevron invites a richer, more nuanced understanding of agency expertise as we grapple with some of the most pressing issues of our time.

This outcome is certainly not a given. To the extent that Loper Bright invokes only a superficial understanding of expertise, it will undermine the many virtues of engaging deeply with agencies’ rich expertise. These virtues include fostering better decisionmaking within, and governance by, the agencies, which enhances external accountability. Second, while undermining expertise is troubling for any number of fields, it is especially salient for energy and environmental law, where expertise features so heavily in the agencies’ missions.⁴Were space to allow, I would also engage the concern that undoing Chevron would create considerable uncertainty about past step-two decisions upholding agency interpretations, especially those related to energy and environmental law where expertise plays a significant role in the decision to accept the agency’s reasonable interpretation. See Brief for the Respondents at 32–33, Loper Bright Enters. v. Raimondo, 143 S. Ct. 2429 (filed Sept. 15, 2023) (No. 22-451) (arguing that overruling Chevron would upset reliance interests on a “settled body of law” given the thousands of prior step-two decisions). Even a period of uncertainty is especially of concern in those fields where significant infrastructure investment for climate change is imperative, and legal clarity and reliance are important factors for attracting investors. See Emily Hammond & Jim Rossi, Stranded Costs and Grid Decarbonization, 82 Brook. L. Rev. 645, 645–47 (2017) (discussing challenges for attracting investors in major decarbonization energy projects). Thus, a look at those particular fields offers a focused lens on what is at stake.

Because there remains considerable potential to engage a rich conception of agency expertise, agencies, advocates, and courts should resist any weakening of such engagement that Loper Bright may invite. To develop these points, Part I includes some thoughts on what is meant by agency expertise, joining Professors Elizabeth Fisher and Sidney Shapiro in contending that a robust conception is both most accurate and best poised to do work in the theory of judicial review.⁵See generally Elizabeth Fisher & Sidney A. Shapiro, Administrative Competence: Reimagining Administrative Law (2020) (arguing that the capacity and authority of expert agencies is crucial to ensure the legitimacy and accountability of the administrative state). Next, Part II extends my prior work on what is meant by deference to expertise, emphasizing the theoretical rationales for such deference and arguing that hard-look review best leverages the comparative strengths of courts and agencies while ensuring fidelity to statute. Finally, Part III offers a cautionary tale: a recent Court of Appeals for the D.C. Circuit (“D.C. Circuit”) decision conflates arbitrary-and-capricious deference standards with the second part of Loper Bright’s question presented. The problematic analysis presented there illustrates why getting deference to expertise right matters.

I.      What Do We Mean By Agency Expertise?

“Expertise” is a word that is used quite loosely in administrative law, often leaving the reader to fill in their own understanding of the term.⁶See id. at 35 (“In writing about, thinking about, and talking about administrative law, the substance of expertise figures little in the collective imagination of administrative lawyers.” (citing Sidney A. Shapiro, The Failure to Understand Expertise in Administrative Law: The Problem and Its Consequences, 50 Wake Forest L. Rev. 1097 (2015))). Sometimes, it simply refers to an agency’s general familiarity with administering a statute—the sort of experience-based knowledge that the Court thought potentially useful in Skidmore v. Swift & Co.⁷323 U.S. 134 (1944); see id. at 140 (“We consider that the rulings, interpretations and opinions of the Administrator under this Act, while not controlling . . . , do constitute a body of experience and informed judgment to which courts and litigants may properly resort for guidance.”); see also id. at 139 (“[T]he Administrator’s policies are made in pursuance of official duty, based upon more specialized experience and broader investigations and information than is likely to come to a judge in a particular case.”); United States v. Mead Corp., 533 U.S. 218, 235 (2001) (noting Skidmore deference applies even in the absence of Chevron, in part because of the agency’s experience administering a detailed regulatory scheme of tariff classifications). It might also refer to an agency’s unique expertise in understanding the statute’s meaning given its participation in original implementation.⁸See, e.g., Entergy Corp. v. Riverkeeper, Inc., 556 U.S. 208, 224 (2009) (explaining that longstanding Environmental Protection Agency (“EPA”) interpretation of a statute, from time of enactment, tended to show reasonableness). Or it could reference the agency’s policy-laden decisionmaking, which was at play in the Chevron decision itself.⁹See, e.g., Chevron U.S.A. Inc. v. Nat. Res. Def. Council, Inc., 467 U.S. 837, 858 (1984) (summarizing the EPA’s rationale for its bubble policy under the Clean Air Act); id. at 863 (noting the EPA implemented “policy decisions in a technical and complex arena”). At other times, it seems to relate to comparative institutional competence regarding scientific and technical matters, as the Court emphasized in Baltimore Gas & Electric Co. v. Natural Resources Defense Council, Inc.¹⁰462 U.S. 87, 103 (1983) (considering a Nuclear Regulatory Commission generic rulemaking regarding releases of radionuclides into the environment, “a reviewing court must generally be at its most deferential” when reviewing an agency decision “within its special expertise, at the frontiers of science”). Often, these understandings are blended together with little differentiation. Take this passage from Justice Kagan’s plurality opinion in Kisor v. Wilkie,¹¹139 S. Ct. 2400 (2019). which explains how expertise can provide a potential reason to defer to an agency’s interpretation of its ambiguous regulation:

Next, the agency’s interpretation must in some way implicate its substantive expertise. Administrative knowledge and experience largely “account [for] the presumption that Congress delegates interpretive lawmaking power to the agency.” So the basis for deference ebbs when “[t]he subject matter of the [dispute is] distan[t] from the agency’s ordinary” duties or “fall[s] within the scope of another agency’s authority.” This Court indicated as much when it analyzed a “split enforcement” scheme, in which Congress divided regulatory power between two entities. To decide “whose reasonable interpretation” of a rule controlled, we “presum[ed] Congress intended to invest interpretive power” in whichever actor was “best position[ed] to develop” expertise about the given problem. The same idea holds good as between agencies and courts. “Generally, agencies have a nuanced understanding of the regulations they administer.” That point is most obvious when a rule is technical; think back to our “moiety” or “diagnosis” examples. But more prosaic-seeming questions also commonly implicate policy expertise; consider the TSA assessing the security risks of pâté or a disabilities office weighing the costs and benefits of an accommodation. Once again, though, there are limits. . . . When the agency has no comparative expertise in resolving a regulatory ambiguity, Congress presumably would not grant it that authority.¹²Id. at 2417 (citations omitted).

This passage seems to point to any and all of the forms of expertise as potential reasons a reviewing court might grant deference. But what guidance does that provide?

As Professors Fisher and Shapiro have explained, any of these understandings alone offers a “thin” conception of expertise, promoting a “narrow vision of accountability” that over-emphasizes judicial review as the only method of constraint.¹³Fisher & Shapiro, supra note 5, at 84. First, these conceptions of expertise overlook how agencies typically function by failing to include, for example, discursive processes among experts from various fields, coordinating and synthesizing disparate information across technical and policy analyses, and the role of professionalism in cabining behavior.¹⁴See id. at 90. Second, Fisher and Shapiro draw from the fields of organizational behavior and public administration to argue that the narrow vision of accountability is one that focuses especially on judicial review, overlooking the many forms of internal accountability that are embedded in agencies’ work.¹⁵See id. at 85–86, 90–93, 95 (describing several such forms).

This concept of internal accountability¹⁶See Sidney A. Shapiro & Ronald F. Wright, The Future of the Administrative Presidency: Turning Administrative Law Inside-Out, 65 U. Mia. L. Rev. 577, 578–79 (2011) (defining inside-out legitimacy). is deeply linked to a rich conception of expertise. For example, as Professor David Markell and I have documented, this robust expertise can inform the fairness, transparency, and ultimate public acceptability of agency decisionmaking—even for actions that are unlikely ever to be judicially reviewed.¹⁷See generally Emily Hammond & David L. Markell, Administrative Proxies for Judicial Review: Building Agency Legitimacy from the Inside-Out, 113 Harv. Env’t L. Rev. 313 (2013) (presenting an empirical study of EPA behavior in responding to petitions to revoke state authority to implement environmental laws—a matter that is typically not judicially reviewable). Even though judicial review is only part of the accountability picture for agencies, it reinforces these same attributes of legitimacy across agency culture. Thus, it is worth attending to how courts review agency actions regardless of Chevron’s future because how the courts approach review can reinforce—or undermine—agency expertise well beyond the action being reviewed.¹⁸E.g., Elizabeth Fisher, Pasky Pascual & Wendy Wagner, Rethinking Judicial Review of Expert Agencies, 93 Tex. L. Rev. 1681, 1715 (2015) (describing symbiotic relationship between courts and agencies furthering improved administration of the National Ambient Air Quality Standards (“NAAQS”) program under the Clean Air Act); cf. Sidney A. Shapiro, Law, Expertise and Rulemaking Legitimacy: Revisiting the Reformation, 49 Env’t. L. Rev. 661, 681–82 (2019) (expressing concern that the demise of Chevron would limit the contribution of expertise because “[o]nce judges take it upon themselves to resolve statutory ambiguities or define vague terms, we lose the contribution that expertise makes towards resolving the policy issue or issues that underlie the definition of the term or terms”).

II.      What Do We Mean By Deference?

Although in popular conception, “Chevron deference” might sometimes refer to the entire doctrine, students and practitioners of administrative law alike know that the deferential part of Chevron takes place only in step two.¹⁹See Chevron U.S.A. Inc. v. Nat. Res. Def. Council, Inc., 467 U.S. 837, 843 n.9 (1984) (emphasizing courts’ “final authority” as to step-one construction); Antonin Scalia, Judicial Deference to Administrative Interpretations of Law, 1989 Duke L.J. 511, 521 (describing a textualist approach to step one). Still, as Professors Kent Barnett and Christopher Walker have argued, step two is relatively under-theorized, especially as compared to steps one and zero.²⁰Kent Barnett & Christopher J. Walker, Chevron Step Two’s Domain, 93 Notre Dame L. Rev. 1441, 1443, 1446 (2018). In any event, focusing on Chevron as the be-all and end-all of deference overlooks the vast array of cases involving substantive review of agency actions that do not apply Chevron at all.²¹A host of decisions might have been Chevron cases but were instead reviewed under the arbitrary-and-capricious standard. E.g., Massachusetts v. EPA, 549 U.S. 497, 534–35 (2007) (applying arbitrary-and-capricious standard); id. at 552–53 (Scalia, J., dissenting) (arguing issue should have been decided under Chevron framework); New York v. FERC, 535 U.S. 1, 26–28 (2002) (upholding, after evaluating agency’s reasoning, the FERC decision to not require unbundled transmission to comply with FERC Order 888); id. at 38–40 (Thomas, J., concurring in part and dissenting in part) (arguing Chevron should have determined the issue). Instead, other principles of substantive review have deeply informed Chevron,²²Professor Lisa Schultz Bressman makes this point in her contribution to this symposium. See Lisa Schultz Bressman, Lower Courts After Loper Bright, 31 Geo. Mason L. Rev. 499, 505 (2024) (“State Farm is an elaboration of the arbitrary and capricious test . . . and it establishes strong norms of judicial behavior, which Chevron essentially reinforced for agency interpretations.”). Chevron is also frequently cited in boilerplate standards of review along with other principles like super deference. See, e.g., City and County of San Francisco v. EPA, 75 F.4th 1074, 1088 (9th Cir. 2023) (citing Chevron, Kisor, and Baltimore Gas in a general recitation of standards of review). and Chevron at step two often merges with arbitrary-and-capricious review.²³See infra text accompanying notes 55–61. Integrated as Chevron is with generalized norms of the relative roles of courts, agencies, and Congress, we should pause to consider the spectrum of judicial deference to agency actions, especially as that spectrum relates to expertise.

What is meant by the word “deference” in this context? In legal parlance, the word refers to “[a] polite and respectful attitude or approach, esp[ecially] toward . . . [a] venerable institution whose action . . . or judgment should be presumptively accepted.”²⁴Deference, Black’s Law Dictionary (11th ed. 2019). When a court reviews an agency, therefore, the term suggests at the very least that the court will not consider the matter de novo. Provided the agency has met certain requirements, the court will not substitute its judgment for that of the agency. The devil is in the details (what requirements must be met?), but it is worth reflecting on the reasons for this stance. As Alexander M. Bickel famously expounded, it largely boils down to the counter-majoritarian difficulty: as between democratically accountable agencies and unelected courts, the latter should be reluctant to wield their power without considerable justification.²⁵See generally Alexander M. Bickel, The Least Dangerous Branch: The Supreme Court at the Bar of Politics (2d ed. 1986) (explaining why the Court must act in a way that does not undermine the justification of its power); see also Mark Seidenfeld, Chevron’s Foundation, 86 Notre Dame L. Rev. 273, 289 (2011) (“Implicit in Article III’s assignment of the judicial power is the premise that political powers, being extrajudicial, belong to the other branches of government, and that the judiciary therefore should avoid interfering with those branches’ exercise of their powers where such interference would require the courts to exercise the outer bounds of judicial power.”). This rationale—which places most policy decisions in the elected branches—becomes intertwined with a rationale of comparative institutional competence when scientific and technical complexity are layered into the matters under review.²⁶See Emily Hammond Meazell, Super Deference, the Science Obsession, and Judicial Review as Translation of Agency Science, 109 Mich. L. Rev. 733, 734–35 (2011) (noting agencies’ superior technical competence and stating, “if agency science is mostly about policy, and the politically accountable executive controls agencies, then agencies are the more legitimate institution with respect to science”). See generally Neil K. Komesar, Imperfect Alternatives: Choosing Institutions in Law, Economics, and Public Policy (1994) (describing choices among imperfect institutions). These rationales, of course, also underlie Chevron.²⁷Chevron U.S.A. Inc. v. Nat. Res. Def. Council, Inc., 467 U.S. 857, 863–65 (1984). Their origins, however, are both longstanding and deeply principled, and there is no reason why they should not continue even if Chevron is overruled or limited.²⁸See Bressman, supra note 22, at __ (arguing courts are still likely to give controlling weight to agencies’ statutory interpretations “when the interpretive dispute amounts to a policy disagreement”).

Respecting comparative institutional competence also serves legitimizing functions both for the agency and for the courts. After all, judicial deference to expertise reduces the risk that a court will get the science or technology wrong and undermine its legitimacy.²⁹Judicial struggles with science have been critiqued in other literatures, notably those involving admissibility standards for scientific evidence. See, e.g., Carl F. Cranor, The Dual Legacy of Daubert v. Merrell-Dow Pharmaceutical: Trading Junk Science for Insidious Science, in Rescuing Science from Politics: Regulation and the Distortion of Scientific Research 120, 120–22 (Wendy Wagner & Rena Steinzor eds., 2006) (urging that when legal institutions endorse mistaken science, the legitimacy of law as an institution is threatened). Instead, judicial review should seek to maximize scientific and technical accuracy across the branches, understanding that a robust application of expertise will be employed by an agency to fill the gaps of uncertainty.³⁰See also Emily Hammond Meazell, Scientific Avoidance: Toward More Principled Judicial Review of Legislative Science, 84 Ind. L.J. 239, 242 (2009) (“Judicial review of statutes should seek to maximize scientific accuracy in both branches; failure to do so undermines the legal system’s values of fairness and legitimacy.”). This approach serves congressional intent in entrusting agencies with implementing statutory policies, especially in the environmental and energy context. Consider the ramifications when expertise is undermined; the Court’s decision in Sackett v. Environmental Protection Agency³¹143 S. Ct. 1322 (2023). offers an example.³²Id. at 1329, 1336–41. In narrowing the meaning of “waters of the United States” in the Clean Water Act, the Court adopted an approach to characterizing jurisdictional wetlands that wholly defies hydrological principles.³³See id. The Court’s requirement of a continuous surface connection between waterways and wetlands is absurd when compared with actual principles of wetlands hydrology, which among other things considers groundwater interactions. See Emily Hammond Meazell, Emerging Trends in Clean Water Act Litigation: Drifting Away from Scientific Principles?, in Virginia Water Research Symposium 2005: Balancing Water Law and Science 114 (2005). Instead of engaging with agency expertise in considering whether given waters come under the Act’s jurisdiction, the Court’s bright-line rule left no room for expertise at all—making the decision look unprincipled and purely political.³⁴Another example might be the Court of Appeals for the Fifth Circuit’s decision in Alliance for Hippocratic Medicine v. FDA, 78 F4th 210, 247–48, 253 (5th Cir. 2023), which preliminarily enjoined the FDA’s relaxation of rules governing prescriptions of mifepristone after second-guessing the agency’s safety findings. See also Adam Unikowsky, The Fifth Circuit’s Mifepristone Decision Is Wrong, Part 2, Substack: Adam’s Legal Newsletter (Aug. 20, 2023), https://perma.cc/LVK6-LLU2.

As Professor Shapiro noted, a decision that does away with Chevron risks a similar decimation of opportunities to engage with agency expertise.³⁵Shapiro, supra note 18, at 681–82. But that is not the only option. If expertise remains a factor in deciding whether to extend deference—á la Kisor or Skidmore—a robust conception of expertise can still play a role in judicial review. The same would be true for considerations of how to conduct a deferential review, whether in an invigorated step two or a shift to more review under the hard-look approach.³⁶As described in Part III, the “hard-look” approach is associated with ensuring an agency has considered all important aspects of a problem, relied on factors Congress intended, and articulated a reasonable connection between the information before it and the decision reached. See Motor Vehicles Mfrs. Ass’n of the U.S. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983). The next section, therefore, takes stock of how deference principles relate to agency expertise in the case law.

III.      Effectuating Deference to Expertise

The pinnacle of deference to expertise is so-called super deference, which is most often attributed to the Supreme Court’s 1983 Baltimore Gas decision.³⁷Balt. Gas & Elec. Co. v. Nat. Res. Def. Council, Inc., 462 U.S. 87, 103 (1983). As I have fully detailed elsewhere,³⁸See Hammond Meazell, supra note 26, at 760–64. this case relates to the infamous Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council,³⁹435 U.S. 519 (1978). in which the Court held that courts may not impose additional procedural requirements on agencies beyond those set forth in statute or adopted by the agencies themselves.⁴⁰Id. at 543–45. In essence, this pair of cases arose as the Court was engaged in a project of reining in the D.C. Circuit.⁴¹See Hammond Meazell, supra note 26, at 758–59. And while deference to expertise both predate and survive that time period,⁴²See id. at 756–57 (describing post-New-Deal expertise model of administrative law); FERC v. Elec. Power Supply Ass’n, 577 U.S. 260, 292 (2016) (referencing “great deference” to “a technical area like electricity rate design”). it is worth noting that the Court decided its foundational hard-look decision, Motor Vehicle Manufacturers Ass’n of the U.S. v. State Farm Mutual Automobile Insurance Co.,⁴³463 U.S. 29 (1983). the same year as Baltimore Gas and only a year before Chevron.⁴⁴See id.; Balt. Gas & Elec. Co. v. Nat. Res. Def. Council, Inc., 462 U.S. 87 (1983); Chevron U.S.A. Inc. v. Nat. Res. Def. Council, Inc., 467 U.S. 857 (1984).

I have previously argued that whereas super deference is not particularly justified, traditional hard-look review—which indeed involves deference to expertise⁴⁵This is because a court does not substitute its judgment for that of the agency, provided the agency has adequately supported and justified its action. See State Farm, 463 U.S. at 43.—reaches a good balance among a host of competing factors.⁴⁶I thus would not characterize my work as opposing any form of deference to expertise. Cf. Eli Nachmany, Deference to Agency Expertise in Statutory Interpretation, 31 Geo. Mason L. Rev. 587, 599 (2024). First, it helps sort through the reality that many suits arguing an agency used flawed science arise out of dissatisfaction with the ultimate policy decision.⁴⁷See Hammond Meazell, supra note 26, at 749. A court can avoid falling into the trap of presuming science would provide a policy answer by ensuring that the agency engaged in reasoned decisionmaking, for example, by explaining why it credited one study over another or reached a conclusion that seems contrary to some of the science in the record.⁴⁸Id. Even though this judicial attitude may seem narrowly focused on particular studies, models, or the like, it reinforces a more robust conception of expertise because it recognizes that those small parts of the record are contextualized in a much larger whole.

Moreover, judicial engagement with the parties’ contentions and the record provides important functions in the constitutional order. Judicial opinions can offer generalist descriptions of agency records that are rife with terms of art and concepts native to particular fields of expertise.⁴⁹Id. at 778. These generalist descriptions serve to “provid[e] generalist accounts of specialized information for largely nonscientific consumers.”⁵⁰Id. These descriptions in turn promote congressional and civil-society oversight, serving the same kinds of democracy-forcing values that showed up in cases like State Farm and Chevron itself. And they leverage the strength of generalist judges in translating complex concepts for generalist audiences.⁵¹See id. at 780. These opinions are far more accessible and visible than agency records, and when done well, they demonstrate a deep engagement with the executive branch. This dialogue is valuable in and of itself and goes beyond the adversarial stance of judicial review as a means of constraint (or lack thereof). Instead, it demonstrates a partnership in fostering good governance and the carrying out of congressional directives.⁵²As Fisher and Shapiro noted, it was once common for judges to conceive of themselves as partners with the agencies, furthering the public interest. Fisher & Shapiro, supra note 5, at 283–84. Indeed, Professors Fisher, Pascual, and Wagner have empirically documented this potential in their study of judicial review of EPA National Ambient Air Quality Standards standard-setting over the course of decades.⁵³See Fisher et al., supra note 18, at 1684. They describe a symbiotic effect: “Generalist courts presiding over expert battles—at least when operating at their best—may actually improve the rigor of science-intensive decisions by insisting on agency-generated yardsticks while in turn benefitting from those improved yardsticks in reviewing agency action.”⁵⁴Id. at 1715. Notice how this approach reinforces internal and robust agency expertise while also monitoring fidelity to statute.

Indeed, this conception of hard-look deference to expertise offers theoretical justifications for how contemporary deference often works in the courts. While as a descriptive matter, super-deference principles continue to show up in judicial opinions, the actual analysis courts employ is more typically deeply engaged with the agency’s exercise of its expertise.⁵⁵See Fisher & Shapiro, supra note 5, at 241 (“[T]hicker ideas of competence allow for a more nuanced and intensive form of review . . . . [which facilitates] a more meaningful form of accountability.”). The Supreme Court has not cited Baltimore Gas for its super-deference principle since 1989 in Marsh v. Oregon Natural Resources.⁵⁶490 U.S. 360, 377 (1989). Moreover, a WestLaw search for Supreme Court opinions citing Baltimore Gas after 1989 yields only two opinions, neither of which reference the super-deference principle. There, the Court rejected environmental plaintiffs’ arguments that an Army Corps of Engineers decision not to create a Supplemental Environmental Impact Statement should be reviewed de novo for conformity with law.⁵⁷See id. at 374–76. The Court used Baltimore Gas to help illustrate what kind of decisionmaking the Corps had engaged in; it was the sort of issue, reasoned the Court, of which resolution “requires a high level of technical expertise.”⁵⁸Id. at 377 (quoting Kleppe v. Sierra Club, 427 U.S. 390, 412 (1976)). The Court therefore concluded that the appropriate standard of review was section 706(2)(A)—the arbitrary-and-capricious standard.⁵⁹Id. Yet when the Court reviewed the actual substance of the decision, it followed a hard-look methodology, engaging carefully with the scientific studies at issue as well as the Corps’ reasoning on the matter to uphold the agency’s action.⁶⁰This example is more fully documented in Hammond Meazell, supra note 26, at 764–66. This approach has carried forward; lower court decisions, too, largely apply a traditional hard-look analysis even when they cite Baltimore Gas’s super-deference principle.⁶¹See id. at 772–78 (collecting and discussing numerous examples); see also City and County of San Francisco v. EPA, 75 F.4th 1074, 1088 (9th Cir. 2023) (citing Baltimore Gas in boilerplate standards for review but discussing each of the plaintiff’s contentions and factual evidence supporting agency’s actions); Nantucket Residents Against Turbines v. U.S. Bureau of Ocean Energy Mgmt., No. 21-CV-11390, 2023 WL 3510955, at *20 (D. Mass. May 17, 2023), appeal docketed, No. 23-1501 (1st Cir. 2023) (citing Baltimore Gas but engaging concretely with each of plaintiffs’ contentions, and agency record regarding biological opinion about right whales and offshore wind turbines); Safari Club Int’l v. Haaland, 31 F.4th 1157, 1173–74 (9th Cir. 2022) (citing Baltimore Gas and discussing each of plaintiffs’ claims as well as data supporting agency’s response), cert. denied, 143 S. Ct. 1002 (2023); Unite the Parks v. U.S. Forest Serv., No. 21-16238, 2022 WL 229172, at *2–3 (9th Cir. Jan. 25, 2022) (holding district court properly reviewed administrative record in determining that agency had considered but rejected plaintiffs’ preferred scientific study); Nat’l Audubon Soc’y v. U.S. Army Corps of Eng’rs, 991 F.3d 577, 583–89 (4th Cir. 2021) (citing Baltimore Gas but analyzing and explaining each of plaintiffs’ contentions and agency’s responses); Shafer & Freeman Lakes Env’t Conserv. Corp. v. FERC, 992 F.3d 1071, 1090–93 (D.C. Cir. 2021) (detailing each contention and evidence in the record and reasoning that Fish and Wildlife Service’s choice of stream modeling methodology was not arbitrary and capricious). Of course, courts also distinguish Baltimore Gas in declining to extend super deference, usually while rejecting an agency action under hard-look review. E.g., WildEarth Guardians v. U.S. Bureau of Land Mgmt., 870 F.3d 1222, 1236–38 (10th Cir. 2017) (noting agency relied on other agencies’ expertise—Energy Information Agency and Department of Energy—to complete its economic analysis, so the agency’s expertise was not its own); Ctr. for Biological Diversity v. Bernhardt, 982 F.3d 723, 740 (9th Cir. 2020) (citing Baltimore Gas but declining to extend super deference where the agency’s analysis was incomplete and not directly related to its core expertise); cf. Del. Riverkeeper Network v. FERC, 45 F.4th 104, 113 (D.C. Cir. 2022) (describing plaintiffs’ National Environmental Protection Act (“NEPA”) analysis concerns but stating they were “undeveloped and unsupported”); League of United Latin Am. Citizens v. Regan, 996 F.3d 673, 705 (9th Cir. 2021) (Bybee, J., dissenting) (arguing majority should have deferred to EPA judgment about the safety of the pesticide chlorpyrifos). But see Alabama v. U.S. Army Corps of Eng’rs, No. 15-699, 2023 WL 7410054, at *47–48 (D.D.C. Nov. 9, 2023) (citing Baltimore Gas and concluding “extreme deference” was owed to Corps’ choice of models, without demonstrating how the Corps responded to each of the concerns about its modeling raised by the EPA). Note that federal circuit court decisions often cite Baltimore Gas simply for the NEPA hard-look requirement. E.g., Sierra Club v. FERC, 38 F.4th 220, 226, 232–33 (D.C. Cir. 2022) (holding agency action not arbitrary and capricious); Food & Water Watch v. FERC, 28 F.4th 277, 285, 292 (D.C. Cir. 2022) (holding arbitrary and capricious one of various challenged aspects of agency action but declining to vacate).

Returning to the world of Chevron for a moment, this discussion has relevance to step two as well. It is true that courts have brought a variety of approaches to Chevron’s second step. Numerous decisions—especially early ones—treated step two almost as a rubber stamp once courts agreed with the agency that the statute was unclear.⁶²See Barnett & Walker, supra note 20, at 1444 (explaining that because it is “received wisdom that agencies nearly always prevail at step two,” it “has largely seemed like a fait accompli”). Later, scholars argued that step two should be treated as a special application of arbitrary-and-capricious review or that the approaches should merge altogether.⁶³See Mark Seidenfeld, A Syncopated Chevron: Emphasizing Reasoned Decisionmaking in Reviewing Agency Interpretations of Statutes, 73 Tex. L. Rev. 83, 129 (1994) (arguing step two should be treated as a special application); Ronald A. Levin, The Anatomy of Chevron: Step Two Reconsidered, 72 Chi.-Kent L. Rev. 1253, 1254 (1997) (arguing the two tests should be merged). Increasingly, this juxtaposition of the standards appeared in at least some circuit court decisions,⁶⁴Barnett & Walker, supra note 20, at 1466–68. consistent with the Supreme Court’s signals.⁶⁵Judulang v. Holder, 565 U.S. 42, 52 n.7 (2011). Regardless of approach, it is at step two that expertise should matter most because it bears on reasonableness and the reasons for deference.⁶⁶See Chevron U.S.A. Inc. v. Nat. Res. Def. Council, Inc., 467 U.S. 837, 863 (citing agency expertise at step two).

If the Court keeps Chevron in place but tightens the standard, further guidance on applying step two might be forthcoming.⁶⁷See Barnett & Walker, supra note 20, at 1473 (noting the Court has provided little step-two guidance, making it “fertile ground for more theoretical and doctrinal development”). Here, expertise should continue to be a consideration, even under a hard-look-type approach—much as I have described above. On the other hand, if the Court limits Chevron’s applicability in Loper Bright, scores of cases will remain—especially in the lower courts—that require courts to grapple with agency expertise.⁶⁸See Bressman, supra note 22, at __. Either way, there is a fresh opportunity to engage with a deeper understanding of expertise in and of itself, reinforcing agencies’ expert capacities while ensuring they maintain fidelity to statute.

IV.      Where Deference To Expertise Matters – And Where It Does Not

The D.C. Circuit’s complicated recent treatment of standards of review in Maine Lobstermen’s Ass’n v. National Marine Fisheries Service⁶⁹70 F.4th 582 (D.C. Cir. 2023). offers opportunities to think critically about what the bounds of deference to expertise in the post-Chevron world should be. There, the Maine Lobstermen’s Association challenged the Service’s approach to protecting endangered right whales, which included a multi-phase plan that restricted lobster fishing practices.⁷⁰Id. at 590–91. This approach avoided a finding of jeopardy to the right whales for purposes of the biological opinion. Id. at 590. In its underlying analysis, the Service confronted gaps in its information about the magnitude of the impact of the fishery’s practices on right whales’ survival.⁷¹Id. at 589. The conservation plan extended to Jonah crab fisheries as well, but for simplicity this discussion focuses on the lobster fishery, as does much of the court’s opinion. Id. at 601. To manage these gaps, it made conservative assumptions that would be more protective of the species, including that the lobster fisheries were responsible for considerable right whale deaths.⁷²Id. at 588–90. To ameliorate the impact, the agency adopted a multi-phase plan that required the lobstermen to implement various protective practices, to which the lobstermen objected.⁷³Id. at 590.

The court’s analysis—by which it vacated the agency action—reflects considerable chicanery across standards of review. The Service, citing Baltimore Gas, argued that its decision to make conservative assumptions was due deference because it was informed by the agency’s expertise and consistent with legislative history suggesting the “benefit of the doubt” should go to endangered species.⁷⁴Id. at 596–98. The court responded, seemingly out of nowhere:

We have seen this line of argument before. Without mentioning the case, the agency is, in substance, asking us to adopt an “aggressive reading of Chevron [that] has more or less fallen into desuetude.” Under this version of Chevron, “silence” gives an agency wide latitude. Loper Bright Enterprises, Inc. v. Raimondo, 45 F.4th 359, 368 (D.C. Cir. 2022), cert. granted, No. 22-451 . . . 143 S. Ct. 2429, 216 L.Ed.2d 414 (U.S. May 1, 2023).

. . . .

There are several problems with this sub rosa Chevron argument.⁷⁵Maine Lobstermen’s Ass’n, 70 F.4th at 597 (citation omitted).

The court continued to catalog the problems: the Service’s reliance on legislative history meant it had taken a legal-interpretive, rather than a policy approach, which presumably justified the court’s shift to Loper Bright;⁷⁶Id. at 597–98. Of course, many questions of agency expertise and policy are reviewed under the arbitrary-and-capricious standard, even if they also rightly involve consulting the statutory text. See Citizens to Pres. Overton Park, Inc. v. Volpe, 401 U.S. 402, 415–16 (1971) (stating agency must act within the scope of its authority and not be arbitrary or capricious). the conservative estimating approach was a change from the prior administration without a justification;⁷⁷Maine Lobstermen’s Ass’n, 70 F.4th at 598–99. Agencies are indeed expected to justify their policy changes. See FCC v. Fox Television Stations, Inc., 556 U.S. 502, 514–16 (2009). and the Service’s approach was contrary to law because the Endangered Species Act’s relevant language requiring best available science did not explicitly include any precautionary principle.⁷⁸Maine Lobstermen’s Ass’n, 70 F.4th at 599. As the court underscored, “[i]t is not the province of a scientific consultant to pick whales over people.”⁷⁹Id. at 600.

Still, the court recognized the Service’s job was challenging, and it offered:

In most realistic cases, however, the Service will be able to make a scientifically defensible decision without resort to a presumption in favor of the species. When it does so, the Service’s predictions will be entitled to deference. If brute uncertainty does make it impossible for the Service to make a reasoned prediction, however, the interpretive rules supply a ready answer: The Service lacks a clear and substantial basis for predicting an effect is reasonably certain to occur, and so, the effect must be disregarded in evaluating the agency action.⁸⁰Id.

Maine Lobstermen’s Ass’n should stand as a cautionary example of how Loper Bright could operate to infect judicial review of a host of agency actions, undermining agency expertise along the way. First, notice that the court invoked a point made by the dissent in the D.C. Circuit’s Loper Bright: the idea that statutory silence establishes a lack of agency authority and is a reason for the court to second-guess the agency’s expertise.⁸¹Loper Bright Enters., Inc. v. Raimondo, 45 F.4th 359, 374 (D.C. Cir. 2022) (Walker, J., dissenting), cert. granted in part, 143 S. Ct. 2429 (argued Jan. 17, 2024) (mem.). This issue is part of the question presented. See Petition for Writ of Certiorari at i–ii, Loper Bright, 143 S. Ct. 2429 (filed Nov. 10, 2022) (No. 22-451) (“Whether the Court should overrule Chevron or at least clarify that statutory silence concerning controversial powers expressly but narrowly granted elsewhere in the statute does not constitute an ambiguity requiring deference to the agency.”). Its citation to the majority opinion for this proposition is therefore misleading and increases the likelihood that the proposition will creep into future opinions to undermine agency authority.

Second, this approach is especially troubling for statutory schemes that establish regulatory protections for people and the environment. Such statutes typically do not tell agencies where to draw lines when data are lacking, yet it seems clear that Congress trusted the agencies’ full expertise in doing so.⁸²See, e.g., Whitman v. Am. Trucking Ass’ns, 531 U.S. 457, 473–74 (2001) (upholding, against nondelegation challenge, statutory instructions to set NAAQS “at a level that is requisite to protect public health”); Am. Coke & Coal Chems. Inst. v. EPA, 452 F.3d 930, 932 (D.C. Cir. 2006) (upholding EPA use of protective assumptions in setting nationwide limitations on water pollutant discharges from the coke-making industry). Especially in the realm of environmental law, a host of cases have recognized that when agencies rely on estimating techniques and make assumptions protective of human health and the environment, they are properly exercising a mix of policy and technical expertise for which deference is appropriate.⁸³See, e.g., Ethyl Corp. v. EPA, 541 F.2d 1, 6–7 (D.C. Cir. 1976) (en banc) (upholding EPA precautionary standards to phase out lead in gasoline). In fact, the kind of robust expertise brought to bear on such decisions is precisely the sort that is properly engaged in hard-look review and legitimately deferred to by reviewing courts. So in Maine Lobstermen’s Ass’n, Baltimore Gas would have been too much deference (despite the Service’s advocacy), but the court was wildly off-track when it leapt to the Loper Bright dissent.

Indeed, to categorize such statutory frameworks as involving statutory silence, raising a presumption that the agency lacks authority, would decimate the role of expertise that Congress legitimately anticipated in setting forth agency responsibilities. This result seems to far exceed the bounds of the judiciary’s role, and it should raise an alarm for followers of scientific and technical fields of law.

Conclusion

Despite this cautionary tale, this Article ends where it began. Whether or not Chevron survives Loper Bright’s scrutiny, it seems likely that deference will still matter and expertise—one of the fundamental justifications for deference in Chevron—will still factor into judicial review. There is an opportunity here, perhaps for development in the lower courts, to conceive of expertise richly and to deeply engage with it. This approach remains deferential but advances legitimacy for both the administrative state and reviewing courts by respecting comparative institutional competence while reinforcing accountability and maintaining fidelity to statute.