The Court issued three decisions touching on administrative law this week.* There is no obvious theme uniting the three, which cover much of the waterfront of executive regulatory authority, from labor relations, to energy, to public health and immigration. Two decisions resolve petitions for review of final agency action, and the third, an interlocutory appeal from an order granting a preliminary injunction. But each one necessarily addresses the standard by which the D.C. Circuit reviews the decision before it. And each one raises questions about that standard that I think followers of administrative law will find interesting.
In Wendt Corp. v. NLRB, the Court granted in part and denied in part a petition for review of a National Labor Relations Board order finding that Wendt Corporation had committed a variety of unfair labor practices. (It also granted in part and denied in part a cross-application for enforcement.) The opinion by Judge Wilkins (joined by Judges Pillard and Edwards) walks through five incidents giving rise to the NLRB’s findings. I will focus on one incident here: the Board’s finding that Wendt assigned an employee to low-skilled work and denied him overtime because of his union activities. As with other claims of employment discrimination, the Board analyzes charges of anti-union animus under a burden-shifting framework: First, the government must make a prima facie showing that anti-union animus motivated the employer’s action. Then, the burden shifts to the company to “show that it would have taken the same action in the absence of an unlawful motive.” What struck me is the standard under which the Court reviews the Board’s determination. The Court generally reviews the Board’s findings to confirm that they are supported by substantial evidence, but the Court’s “review of the Board’s conclusions as to discriminatory motive is even more deferential than the substantial evidence standard because most evidence of motive is circumstantial.” It seemed odd to me that the Court would be more deferential to the Executive Branch’s finding of liability in circumstances in which liability is more difficult to ascertain. It appears that this more deferential standard of review has been around for a while. Judge Wald explained its rationale in the mid-nineties: “drawing inferences” from mostly circumstantial evidence “to assess an employer’s . . . motive invokes the expertise of the Board, and consequently, the Court gives substantial deference to inferences the Board has drawn from the facts, including inferences of impermissible motive.”
In Long Island Power Authority v. FERC, the Court granted in part and denied in part consolidated petitions for review challenging the Federal Energy Regulatory Commission’s decision to approve a settlement allocating the cost of high-voltage facilities that transmit electricity within the mid-Atlantic region. As the Court explains, the question how to allocate those costs “is difficult because high-voltage projects afford two different kinds of benefits—local benefits that accrue primarily to utilities close to the project at issue, and regional benefits that accrue throughout the grid.” The regional transmission organization whose settlement is at issue in this case has for years bounced between cost allocation methods that ignored one set of benefits or the other. The settlement sought to achieve a “reasonable balance” between the two. Two parties whose costs increased under the settlement objected and, when FERC approved the settlement over their objections, petitioned for review.
Because ratemaking is imprecise, the Court is careful not to require “exacting precision” in reviewing cost-allocations. But here, the Court was not reviewing a pure cost-allocation decision. Instead, it reviewed FERC’s approval of a settlement allocating costs. In approving settlements, FERC considers whether the “overall result of the settlement is just and reasonable, even if individual aspects of it may be problematic.” Judge Katsas (joined by Judges Henderson and Walker) suggests in a footnote that this standard—developed by FERC—may conflict with the Federal Power Act, which requires “all rates and charges made” to be just and reasonable. Because FERC found each rate within the settlement in this case to be just and reasonable—and the Court agreed—the potential conflict remained academic.
Also academic, but potentially interesting, is the question how the Court reviews certain determinations FERC makes when considering settlement agreements. For example, where individual aspects of a settlement are problematic, FERC must find “that the challenger would be in no worse position under the settlement than if the case were litigated.” And FERC’s regulations call upon it to consider whether there is a “genuine issue of material fact.” These are the sorts of determinations one would find courts uniquely competent to make, which makes me wonder whether their review would be more exacting.
Last but not least, in Huisha-Huisha v. Mayorkas, the Court affirmed in part a district court preliminary injunction barring enforcement of an executive order prohibiting “covered aliens”—namely, aliens without valid travel documents—from entering the United States by land from Canada or Mexico.
The order issued under a nineteenth century statutory provision that authorizes the President to prohibit “the introduction of persons and property from such countries or places as he shall designate” during public health emergencies. The provision—42 U.S.C. § 265—lay mostly dormant through a century of changes to immigration law. When the CDC dusted it off in the early months of the COVID pandemic, it was unclear how § 265 interacted with other substantive and procedural limitations on the President’s immigration powers.
The plaintiffs in this case represent a putative class of covered aliens subject to expulsion under the executive order. They challenged the expulsions on a number of grounds, but the most substantial challenges have to do with the relationship between § 265 and other immigration laws. In a thorough and informative opinion, Judge Walker (joined by Chief Judge Srinivasan and Judge Wilkins) concludes that the plaintiffs are unlikely to succeed on the merits of their claim that the President must afford them an opportunity to apply for asylum before they are expelled, as is ordinarily required under 8 U.S.C. § 1158(a)(1). But he concluded that they are likely to succeed on the merits of their claim that the President may not expel them to countries where they are likely to face persecution or torture, as is prohibited by 8 U.S.C. § 1231.
The interesting standard of review issue is one that recurs with some frequency in administrative law cases but is by no means exclusive to them: how does an appellate court review the lower court’s “likelihood of success on the merits” conclusion on appeal from a preliminary injunction? According to the typical formula, the Court reviews conclusions of law de novo, findings of fact for clear error, and the balancing of the preliminary injunction factors for abuse of discretion. But is “likelihood of success” a matter of law, fact, or discretion? Sometimes, as in this case, the merits present a pure question of law, but one that is difficult enough that the Court is not prepared to answer it definitively at the pace required by preliminary injunction motions and appeals. In such cases, is the “likelihood of success” determination a legal conclusion that is reviewed de novo, or does the district court enjoy a measure of discretion to decide how close is close enough? Here, the Court appeared to apply a de novo standard of review because it rejected the district court’s “likelihood of success” conclusion without expressly finding an abuse of discretion. This may be because the panel did not find the merits ground on which the district court ruled—that § 265 does not authorize expulsions—to be all that close a question. Yet the Court seemed to leave the door open for the lower court to rule for the plaintiffs even on that ground on remand, which suggests that the Court did not definitively find a legal error, either.
There is also a more classic administrative law issue in the opinion: Chevron deference. The Court declined to defer to the Government’s reconciliation of the various immigration statutes because “Chevron deference does not apply when an agency interprets a statute in a way that limits the work of a second statute that a different agency administers.” In this case, CDC administers § 265, while the Department of Homeland Security administers the other at-issue immigration statutes.
Finally, readers may also enjoy the opinion’s use of Justice Jackson’s Youngstown tri-partite framework to organize its analysis of executive action in an area of shared congressional-executive responsibility.
* United States v. Jackson, also decided this week, involves a sentencing challenge and an appeal from decisions denying compassionate release. In addition, the Court reissued a 2020 opinion in United States v. Hirani Engineering & Land Surveying, replacing the word “case” with the word “record” in several places. Aaron Nielson previously observed that there was “not a lot” of administrative law in the opinion, and there is none in recent procedural development. In 2020, the Court had intended to remand the record to the district for further factfinding to aid in the disposition of the other matters on appeal. It accidentally remanded the entire case, however, so now that the supplemental factfinding is complete, it is recalling the mandate to resolve the earlier appeal.
D.C. Circuit Review – Reviewed is designed to help you keep track of the nation’s “second most important court” in just five minutes a week.