This article was authored by Charles Tyler, a research fellow at the Constitutional Law Center at Stanford Law School and an appellate litigator at Orrick, Herrington & Sutcliffe.
This article first appeared in the Regulatory Review's series on "Improving Agency Procedure" that focuses on the ACUS Recommendations adopted at the June 2018 Plenary Session. Reposted with permission. The original may be found here, and the Regulatory Review's entire series may be found here.
Federal rulemaking under the Administrative Procedure Act (APA), by design, is cumbersome. To issue a new rule, an agency must avoid a litany of potential procedural pitfalls. For example, typically an agency must notify the public of the proposed rule, address any substantial and material comments that it receives, and justify the final rule based on a set of factors authorized by a congressional statute.
These requirements, however, can create substantial legal uncertainty for an agency seeking to issue a new rule. Including severability clauses in proposed rules could help fix this problem.
Agency officials are often unclear about whether the agency has done enough to assure a reviewing court that it has complied with its procedural obligations. It is sometimes difficult to know, for example, which factors a reviewing court will consider permissible, or which comments, among the millions that agencies sometimes receive, a court will consider substantial enough to warrant a response.
Courts that find violations of the APA’s procedural requirements typically vacate the entire rule, forcing the agency to start the rulemaking process over from scratch. Yet, it is far from clear that the APA requires this remedy. True, the APA directs a reviewing court to “hold unlawful and set aside” any “agency action” that violates various legal requirements. But the APA defines “agency action” to include “a part of an agency rule,” suggesting that the law may only require a reviewing court to set aside the unlawful portion of a challenged agency rule.
Under Chenery, a reviewing court may not affirm an agency decision on a ground different from the one the agency originally supplied. To do otherwise, it is thought, would potentially leave in place a policy that the agency never intended to endorse. Similarly, leaving the remainder of a rule in place after finding a part of the rule to be unlawful would potentially have the effect of leaving in place a rule that the agency never intended to issue, or at least that the agency never took through the proper rulemaking procedures. Hence the typical remedy of forcing the agency to take a mulligan.
When combined with the legal uncertainty inherent in rulemaking, the default remedy for unlawful agency actions is a big problem for agencies seeking to issue new rules. Vacating an entire rule costs the agency the time and expense of putting together a new rulemaking record—a process that can take years and thousands of hours of staff time. And the remedy deprives the public of the rule’s benefits, such as they are, at least for the time that it takes the agency to reissue a similar rule.
The study follows up on a 2015 article I co-authored with E. Donald Elliott, which argued that rules should more frequently include instructions to a reviewing court that the agency intends for the remainder of a rule to stay in effect even if certain portions of the rule are set aside. These instructions are commonly referred to as “severability clauses.”
We explained in the article that, before 2015, most federal agencies had never included a severability clause in any of their rules. And of the agencies that had, only a few had done so in more than 2 percent of rules. We urged federal agencies to consider including a severability clause in a rule when the agency has determined that the rule’s remainder would function sensibly in the absence of one of the rule’s provisions.
Including a severability clause in a rule would solve the Chenery-ish problem by assuring a reviewing court that the agency intended the remainder of the rule to remain in effect, even in the absence of one of the rule’s provisions. And it would also serve to notify the public of the agency’s position on the issue of severability far in advance of litigation. We also urged federal courts to review severability clauses contained in agency rules with deference, such that the remainder of a rule would generally stay in place, if that is what the text of the rule asks the court to do.
The recent ACUS study supplements this research with interviews of current and former officials in various agencies’ offices of general counsel and with several former officials in the U.S. Department of Justice. Respondents in that study indicated that several agencies had recently included severability clauses in their rules where it was appropriate to do so, and agency officials believed this made the rule’s potential implications clearer to affected parties. Several officials also observed that their agencies have regularly included severability clauses in their rules, and one agency official noted that doing so had not weakened his or her agency’s position on the merits in litigation.
Some respondents, however, warned that an agency’s ability to include severability clauses in its rules should depend on its subject matter and rulemaking processes. Not all rules are amenable to a severability clause because not all rules are composed of provisions that are intended to operate independently.
In the new recommendation, ACUS proposes that if an agency anticipates litigation over a new rule, then, early in the process of developing the rule, the agency should consider whether the rule can be divided into parts that would function independently, even if other parts are ultimately ruled unlawful. If so, then the agency should structure the text of the rule to reflect the independence of the rule’s parts.
Further, ACUS urges agencies to consider whether they should include a severability clause in their rules. If so, agencies should include the clause in the text of the final rule, rather than merely the rule’s statement of basis and purpose, to assure a reviewing court that the severability clause reflects the agency’s true legislative intent.
In addition, agencies should include the severability clause in the text of the proposed rule, rather than just the final rule, to allow the public an opportunity to comment. Finally, ACUS proposes that courts solicit the views of the parties on the issue of severability, if it becomes a relevant issue during judicial review.
By encouraging agencies to consider severability early in the process of developing a new rule and to ask for public comment on the issue of severability, ACUS’s recommendation promises to promote greater transparency and predictability in the development and implementation of federal policy.