The Administrative Procedure Act1 (APA) provides that agencies must undergo notice-and-comment rulemaking to repeal, delay, or otherwise modify a finalized rule,2 but it does not specify when a rule becomes “final.”3 Incoming Presidents have taken advantage of this ambiguity to respond to leftover rulemakings from the prior presidency without notice and comment,4 both by ordering the withdrawal of completed rules that have yet to be published in the Federal Register and by delaying the effective date of published rules that have yet to go into effect.5 Recently, in Humane Society of the United States v. USDA,6 the D.C. Circuit foreclosed the first response, holding that a rule becomes final — and an agency must thus undergo notice and comment to repeal it — when the Office of the Federal Register (OFR) makes the rule available for public inspection prior to publication.7 Commentators have argued that the D.C. Circuit’s decision will have “broad implications” for presidential authority over the rulemaking process, making it more difficult for new administrations to reverse the last-minute decisions of their predecessors.8 But for an incoming President, Humane Society’s bark may be worse than its bite since new administrations remain empowered to suspend the implementation of newly finalized rules. Because such suspensions will likely increase as a result of the opinion, the D.C. Circuit’s holding risks prolonging uncertainty during political transitions for regulators and regulated entities alike.
Humane Society arose from efforts to stop the abuse of competitive show horses in the United States. To improve these horses’ gaits, some trainers employ an abusive practice known as “soring,”9 which Congress outlawed in the Horse Protection Act of 197010 (HPA). In 2016, the U.S. Department of Agriculture, which was charged with administering the HPA, proposed a rule providing that it would accredit and train inspectors to check for soring at horse shows.11 After five months of notice and comment, the USDA posted a finished, signed rule on its website12 and deposited it with the OFR, which subsequently made it available for public inspection on January 19, 2017.13 The following day, President Trump took office, and his Chief of Staff Reince Priebus immediately issued a memorandum ordering the withdrawal of all rules “sent to the OFR but not published in the Federal Register.”14 The USDA subsequently withdrew the horse-soring rule without notice and comment.15
The Humane Society of the United States challenged the USDA’s actions in the U.S. District Court for the District of Columbia, contending that the soring rule had been finalized, and thus notice and comment was required to repeal it.16 The USDA moved to dismiss, arguing that the soring rule was not final for APA purposes because it had not been published in the Federal Register.17 The district court agreed with the USDA and dismissed the suit.18 Observing that the APA “provide[s] little explicit guidance on the finality of agency rules,” the district court looked to context and case law to conclude that rules become final at publication in the Federal Register.19 Regarding context, one of the APA’s basic tenets is that “regulations do not take effect until they are published in the Federal Register,”20 and although the APA allows unpublished rules to be enforced against parties with actual notice of them, it still requires such rules to be published.21 Regarding case law, the D.C. Circuit in Kennecott Utah Copper Corp. v. U.S. Department of the Interior22 had concluded that a rule submitted to the OFR for “confidential processing” prior to being made available for public inspection had “never bec[o]me a binding rule requiring repeal or modification.”23 In sum, nothing “support[ed] a rule that prevents agencies from withdrawing rules prior to their publication in the Federal Register.”24
The D.C. Circuit reversed and remanded.25 Writing for the panel, Judge Tatel26 concluded that rules are final when they are filed for public inspection by the OFR and that the USDA’s unilateral repeal of the soring rule thus violated the APA.27 Beginning with statutory context, the court observed that the Federal Register Act28 (FRA), which governs the publication of all documents (including rules) in the Federal Register, “contemplates that a rule may be prescribed before publication”29 and provides that documents sent to the OFR become “valid” against the public when they are filed for public inspection.30 The FRA also allows for “prepublication enforcement against parties with actual notice,” which suggests that publication is not required for rules to gain legal effect.31 The APA contains similar provisions, such as a “good cause” exemption from its requirement that substantive rules be published thirty days prior to their effective date,32 implying that agencies can and do “prescribe rules with effective dates before publication.”33
The court then moved to case law, distinguishing Kennecott on the grounds that the draft rule at issue there was withdrawn at confidential processing,34 had not yet been made available for public inspection, and was thus never “valid” against the public under the FRA.35 The court also distinguished two out-of-circuit cases involving immigration rules because they did not involve procedural challenges.36 Finally, the court observed that the only out-of-circuit opinion to tackle the question at issue directly, Arlington Oil Mills, Inc. v. Knebel,37 had held that “lack of formal publication does not preclude the effectiveness of an otherwise valid agency action.”38 These precedents made the court “[c]onfident” that the APA requires agencies to undergo notice and comment before repealing a rule made available by the OFR for public inspection.39
Judge Rao dissented.40 She first took issue with the majority’s characterization of Kennecott, arguing that it “drew a sharp line between documents sent to OFR on the one hand” and “‘binding’ regulations published in the Federal Register on the other.”41 Judge Rao then moved to the APA’s text, observing that the statute requires substantive rules to be published and generally provides (subject to several exceptions) that “[s]uch rules cannot have legal effect . . . until publication.”42 For Judge Rao, these provisions suggested that “an agency’s rulemaking discretion continues up until the point of publication,” a conclusion bolstered by case law discussing publication as a “final agency action” for the purposes of judicial review.43 She also criticized the majority for relying so heavily on the FRA, which she argued was “limited by the APA’s more specific provision[s]” involving substantive rules.44 Ultimately, Judge Rao cautioned that the majority was impermissibly “impos[ing] additional procedural requirements on agencies,”45 which she warned would lead to “numerous disruptions for both agencies and courts.”46
The D.C. Circuit’s decision was described as highly consequential for executive authority during political transitions. Mark Febrizio, for instance, argued that the opinion gave “a newly inaugurated president . . . less control over rules published during their administration” and expressed concern that it would “invite more poorly justified rules” by the outgoing administration.47 Professor Jonathan Adler similarly observed that the case “implicate[d] broader questions about the ability of the [incoming] President to set policy priorities for the federal government.”48 But incoming administrations remain able to suspend finalized rules that have yet to go into effect, blunting Humane Society’s impact on presidential control. Such suspensions will likely increase as a result of the Humane Society rule, which risks prolonging uncertainty during political transitions for both regulators and the parties they regulate.
When Presidents leave office, they do not go quietly. Instead, in the last three months of an outgoing administration, “[t]here is a documented increase in the volume of regulatory activity,”49 a phenomenon known as “midnight rulemaking,” which has persisted for at least four decades.50 Not all finalized rules go into effect by inauguration on January 20, however, meaning that new administrations take office with a not-insignificant number of inoperative midnight rules left over from the outgoing presidency.51 Before Humane Society, new administrations typically responded to such rules in two ways: by withdrawing unpublished rules deposited at the OFR without notice and comment, and by suspending the effective dates of published rules that had yet to go into effect, also without notice and comment, by way of the APA’s various exceptions.52 When the Trump Administration took office in early 2017, it continued this trend.53
After Humane Society, future presidential administrations will likely treat unpublished rules sent to the OFR and made available for inspection the same way they have approached published rules with future effect — by suspending the implementation of all of them without public input. Incoming administrations have compelling reasons to continue suspending midnight rules: if a new administration dislikes a given rule, suspension can buy an agency time to gather data “affecting [the rule’s] factual underpinnings” or information about “changed economic or social circumstances,” both of which could eventually support repeal.54 Moreover, even if a new administration finds “no problems with the vast majority” of midnight rules,55 it might value suspensions as a means of weighing the costs and benefits of “chang[ing] the prior administration’s rules” or “focus[ing] on moving forward with the new agenda.”56 Thus, contrary to commentators’ fears,57 there is no reason to expect that Humane Society will wrest incoming Presidents of authority to deal with leftover midnight rules. Rather, suspensions will likely increase in number in the wake of Humane Society, encompassing the unpublished rules on file with the OFR that were affected by the D.C. Circuit’s ruling.
While suspensions ensure a degree of presidential control over midnight rules, an increase in their use risks creating uncertainty for both agencies and the entities they regulate. Prior to Humane Society, agencies considered withdrawal of rules sent to the OFR — when such withdrawals were allowed — to be practically more efficient and legally less risky than suspensions.58 Practically, withdrawals were an easy way for a new administration to immediately block the actions of the prior administration without notice and comment and get to work fashioning a new regulatory agenda.59 In other words, by withdrawing midnight rules, incoming administrations were able to make them disappear at negligible fiscal and political cost. Suspensions, on the other hand, ultimately require a costly choice between “expend[ing] political capital to reverse the prior administration’s rule, or . . . enforc[ing] a rule that is contrary to the incoming administration’s political preferences.”60
Legally, before Humane Society, regulators and commentators generally assumed that withdrawing rules made available for public inspection by the OFR was permissible.61 Suspensions, conversely, have always existed in a legal gray area.62 They are typically understood to be final agency actions, which under the APA generally require notice and comment and publication at least thirty days prior to going into force.63 Yet incoming administrations regularly invoke the APA’s two “good cause” exceptions64 to get around these requirements.65 Courts, in turn, have been inconsistent in answering the question of what constitutes good cause,66 as well as how long is “too long” of a unilateral suspension.67 Due to a lack of clear guideposts, agencies impose delays of varying lengths68 and provide variable explanations of “good cause” to justify them.69 For regulators, then, suspensions are practically and legally hazardous in a way that withdrawals, prior to Humane Society, were not.
For regulated parties, too, suspensions create more practical and legal uncertainty than withdrawals. Practically, withdrawals send a clear message to regulated parties that a given requirement will not be enacted.70 Suspensions, on the other hand, create extended unpredictability: a suspended rule may go into effect at some point in the future, or the agency might seek to repeal it. As Professors Bethany Davis Noll and Richard Revesz explain, “[r]egulated entities often need to make substantial investments in order to comply with regulatory requirements,” and “[w]ith too much regulatory vacillation, companies may put off investment decisions until the uncertainty is resolved.”71 Thus, the financial stability of regulated entities affected by a suspended rule may suffer, since such entities are in the dark about when, or if, they will have to come into compliance.72
Legally, the pathway to challenging withdrawals, as Humane Society itself showed, was relatively straightforward. Suspensions are another story: While regulated parties may be able to get them struck down on judicial review,73 many suspensions are short, which makes them more difficult to challenge.74 In addition, as noted, the legal standards governing suspensions are unclear, both with respect to what constitutes “good cause” to delay a rule’s effective date unilaterally and how long an “indefinite postponement” must be to make it “tantamount to a revocation,”75 thus necessitating notice and comment.76 Finally, suspensions risk incentivizing agency gamesmanship: if a regulated entity challenges a suspension via suit, the agency could simply lift the suspension, mooting the challenge.77 Taken together, these three factors mean that regulated parties do not have a clear path to legal recourse when challenging suspensions in court, and the judiciary will continue to struggle to develop standards governing their use.
At bottom, although Humane Society foreclosed an expedient method of addressing the leftover rulemakings of prior presidencies, new administrations remain able to unilaterally suspend the implementation of midnight rules with future effect, notwithstanding the APA’s procedural requirements for final agency action. This pathway brings with it practical inefficiencies and legal uncertainties for agencies and regulated parties during political transitions, underscoring the need for courts — or Congress — to craft a uniform standard for tackling agencies’ invocation of APA exceptions when delaying the implementation of finalized midnight rules.