The Administrative Procedure Act (APA) created notice-and-comment rulemaking as the paradigm for informal rulemaking. Scholars have recognized the benefits of this process for generating rules that reflect more inclusive public preferences and values and that are more carefully drafted to better serve the public interest. But the APA also permits an agency to adopt a rule without notice-and-comment when the agency shows good cause for why those procedures are impractical, unnecessary, or contrary to the public interest. Traditionally, the first and third ground for the good cause exception to notice-and-comment procedures were understood to be narrow, applying only when there is an unforeseen emergency and when advance notice of the rule would undermine the effectiveness of it. But the notice-and-comment rulemaking process has become protracted since the enactment of the APA in 1946, resulting in long lag times between when a rule is proposed and when it is finally adopted. Thus, this process imposes a cost of delay before the benefits of a new or amended rule are realized. This translates into a conundrum for use of the good cause exception because such use can alleviate the costs of delay but at the expense of the likely extent to which the rule serves the public interest.
Recently, agencies have used a mechanism called interim final rulemaking to adopt rules. Under this mechanism, the agency issues an interim final rule (IFR)—a rule that becomes effective before the agency receives public comments on it but on which the agency invites comments after the rule takes effect. The agency essentially commits to considering whether to amend the IFR in response to the postpromulgation comments that the IFR generates. Issuance of an IFR can solve the good cause exception conundrum because it both allows the IFR to go into effect quickly but does not preclude ultimately delivering the benefits of notice-and-comment procedures when the agency considers the post-promulgation comments and issues a final final rule. But IFRs can themselves alter the rulemaking process so that the final final rule might not be as good as the rule that would have resulted from prepromulgation notice-and-comment proceedings.
In light of the potential for IFRs to solve the good cause exception conundrum, this Article advocates that agencies more broadly invoke the exception by issuing IFRs, and that courts become more tolerant of that practice. At the same time, the Article reviews how use of IFRs might result in harm to the public interest and suggests some restrictions on the broadened use of IFRs to avoid their use resulting in less benefit to the public interest than would derive from the traditional notice-and-comment rulemaking paradigm.
The paradigmatic procedure for agencies to adopt substantive legislative rules under the Administrative Procedure Act (APA) is notice-and-comment rulemaking. Kenneth Culp Davis, an early and well-noted administrative law guru, went so far as to opine that notice-and-comment rulemaking is “one of the greatest inventions of modern government.” Nonetheless, in some circumstances, notice-and-comment procedures are costly and perhaps even interfere with the ability of agencies to regulate to best serve the public interest. Thus, the APA explicitly allows agencies to adopt substantive legislative rules without using notice-and-comment procedures when the agency shows good cause for why notice-and-comment procedures would be “impracticable, unnecessary, or contrary to the public interest.”
Traditionally, the good cause exception to notice-and-comment procedures has been construed narrowly. Recently, however, agencies have increasingly invoked the exception so that as of 2012, about 27% of major rules and 27% of nonmajor legislative rules were adopted without using notice-and-comment procedures and cited the exception. This reflects several developments that led agencies increasingly to rely on informal rulemaking to replace adjudication and formal rulemaking as the primary method of adopting agency policy. In what may have been reactions to increased use of informal rulemaking, Congress, the courts, and even the President have attempted to constrain agencies to ensure that agency rulemaking was careful and deliberative. As a result, the rulemaking process has become sometimes too cumbersome and protracted to address problems that, in a world of virtually instantaneous electronic communication, can arise and morph in short periods of time. Thus, the costs of delay that stem from the current practice of notice-and-comment as well as judicial review of agency rules can be great, which has prompted agencies to rely on the good cause exception in a variety of circumstances that do not meet the traditional standard to help speed up the rulemaking process.
Courts consistently have opined that the good cause exception is to be construed narrowly but have been inconsistent in how they review agency invocation of the exception. Commentators have generally advocated for the courts to hold the line on the use of the good cause exception to avoid the exception swallowing the general requirement of notice-and-comment procedures. However, given the slow pace of rulemaking—especially for matters that have a significant economic impact—the good cause exception can alleviate often long delays in adopting and implementing rules, and thereby provide the benefits of a rule sooner than if the agency follows notice-and-comment procedures. Thus, there is reason for courts to consider liberalizing the use of the good cause exception. At the same time, however, allowing agencies to invoke the good cause exception liberally forfeits the benefits that flow from the notice-and-comment procedures. In an attempt to balance these countervailing considerations, agencies have used a process known as “interim final rulemaking.” Under this process, the agency skips notice-and-comment procedures and issues an interim final rule (IFR), which is a legislative rule that usually takes effect immediately. At the same time, the agency invites comments on the IFR and promises to consider such comments to decide whether to maintain the IFR, or instead to issue a different “final final rule” (FFR) if the agency decides that comments warrant such additional action.
The thesis of this Article is that courts should recognize an expanded good cause exception to encourage agencies to issue IFRs except in circumstances where the issuance of an IFR is unlikely to result in a net increase in social welfare. This thesis essentially balances the benefit of an IFR in minimizing regulatory delay against any detrimental effects the IFR might have on the ultimate FFR adopted. The Article goes on to describe the factors that might lead to issuance of an IFR that results in a net loss of welfare and hence, if present, would counsel against use of the good cause exception even if the agency issues an IFR. These factors consider the benefits of the IFR as a substitute for the regulatory status quo ante that would otherwise continue unless and until the agency completed a notice-and-comment rulemaking, as well as the effects the IFR is likely to have on the quality of the ultimate FFR issued by the agency. In short, this Article’s bottom line recommends that courts consistently soften the traditional reluctance to allow agencies to use IFRs instead of prepromulgation notice-and-comment rulemaking when the issuance of an IFR and the ultimate FFR is likely to best serve the public interest.
To clarify precisely what the Article advocates for, it is imperative to note that the interim final rulemaking process is complex because it involves two agency actions: the adoption of an IFR that is meant to be temporary and the finalization of the rule into a permanent FFR. In addition, the IFR plays two different roles in the scheme: first, it is a rule that has the force of law that is meant to remain in effect until the agency finally resolves what is the best permanent rule; second, it substitutes for a notice of proposed rulemaking (NOPR) for the proceeding that results in the issuance of the FFR. This Article analyzes when an IFR should be found to be proper under the good cause exception to notice-and-comment rulemaking, which essentially addresses the procedural validity of the IFR. It also examines how the use of an IFR should affect the standard of review that the courts apply when considering a challenge to the FFR, given the influence the IFR has on the FFR due to the independent force of law an IFR carries unless and until it is replaced with an FFR.
Perhaps it is most clear to state what the Article does not address. First, an IFR might be a legitimate candidate for exercise of the good cause exception and still not be substantively justifiable as the best rule. That is, even if the nature of an IFR makes it appropriate for the agencies to issue it without notice-and-comment, the IFR might still be challenged as substantively invalid, for example, as being beyond the agency’s statutory authority or being arbitrary and capricious. That substantive review would be carried out using the usual standards courts apply to such challenges for rules that go through the notice-and-comment process and, hence, are not the focus of this Article. Second, an IFR might be a rule that should be considered procedurally valid under the good cause exception but might not provide adequate notice for the rulemaking that culminates with the FFR. Such challenges to the IFR as a substitute for the more traditional NOPR, for example, might claim that the IFR failed to include information that the APA requires be included in a NOPR or that the FFR was not a logical outgrowth of the IFR. Again, the standard for evaluating such challenges is the same for evaluating the adequacy of a NOPR in a traditional notice-and-comment rulemaking proceeding and, hence, is not addressed by this Article.
When the APA was adopted, Congress envisioned that the good cause exception would be narrow. The APA provides that agencies are required to justify invocation of the exception and thus puts the burden of showing good cause on the agency. According to the Attorney General’s Manual on the APA, issued in 1947, the three criteria that trigger the exception addressed three different concerns. Notice-and-comment was viewed as impracticable when it would delay a rule that was needed to address an emergency. An emergency had to be grounded in unforeseen circumstances that, if not addressed, would lead to threats to physical health, safety, or national security. Notice-and-comment was considered unnecessary when the rule was uncontroversial so that no person would object to it or so trivial that the cost of entertaining comments would not justify the concomitant expenditure of agency resources. Notice-and-comment was considered contrary to the public interest only if the rulemaking process itself would undermine the efficacy of the rule, such as a rule restricting withdrawal of funds from banks to guard against a bank run during an impending depression. The very notice would immediately prompt the forbidden behavior before the rule could take effect, causing a bank run, which is precisely what the rule was meant to prevent.
Invocation of the good cause exception when notice-and-comment is unnecessary generally is not controversial, and this Article does not address the use of the exception on this basis. In applying the exception based on the first and third criteria in particular cases, courts generally do not consider these criteria separately but rather evaluate overall whether the exigencies counseling for immediate adoption of the rule warrant bypassing the notice-and-comment procedures. This makes sense because the criteria themselves can overlap with respect to any particular rule; in fact, construed literally, the exception for rules for which notice-and-comment would be contrary to the public interest would include those rules for which notice-and-comment would be impracticable. Although courts almost universally state that the exception is to be construed narrowly, judicial review of the application of the exception occurs on a case-by-case basis, and judicial holdings seem to vary with respect to the breadth they afford.
One principle on which courts seem to agree, with which I fundamentally disagree, is that mere delay in effectively implementing the agency authorizing statute is usually not sufficient to justify bypassing notice-and-comment procedures. Essentially, courts uphold use of the exception to avoid delay only in those situations where an agency could not comply with its statutory obligations or achieve its regulatory objective at all if it had to go through notice-and-comment rulemaking. For example, courts have held that statutory deadlines for an agency to adopt regulations may justify use of the good cause exception, but only when it would be virtually impossible for the agency to meet the statutory deadline and proceed with notice-and-comment rulemaking, or when it is the clear intent of Congress that notice-and-comment procedures need not be followed.
An issue that has split the circuit courts is whether to apply the APA’s “de novo” standard or its more deferential “arbitrary and capricious” standard when reviewing whether an agency has justified invoking the good cause exception. As I will describe below, neither of these standards are appropriate for all aspects of such review. Whether a rule warrants adoption prior to the opportunity for the public to comment on it may involve questions about the effects of the rule that depend on evaluations of legislative facts and policy predictions. Courts are ill-suited to make such determinations without guidance from the agency. At the same time, simple deference to agency determinations will allow agencies to abuse the exception and apply it to adopt rules that disserve the public interest. Rather than simply advocate that courts defer or not, this Article lays out some factors that courts should consider when determining whether an agency is justified in invoking the exception.
Rules adopted under the good cause exception do not go through the notice-and-comment process and thus forfeit any benefits that accrue from that process. Advocates of notice-and-comment procedures contend that the procedures increase the information available to the agency and facilitate public deliberation about the proposed rulemaking. By increasing the breadth of sources of information available to agencies, both about the technical aspects of the rule and the various stakeholders’ preferences regarding it, comments are thought to lead agencies to structure their rules to better serve the public interest.
It is unclear, however, how much comments influence final rules that agencies ultimately adopt. Agencies frequently seem to have their minds set on the substantive fundamentals of rules by the time they specify a proposed rule in a NOPR. The lack of effect of comments seem to reflect two phenomena.
First, few members of the general public comment effectively on proposed rules, even if those rules will significantly affect them. To comment effectively, one must be aware that a rule has been proposed, have the incentive to comment, and have relevant information about how the rule will operate that goes beyond whether one simply prefers the proposed rule or not. Individuals affected by a rule rarely satisfy any of these three prerequisites for filing comments likely to affect the ultimate rule the agency adopts. Thus, most comments are filed by interest groups that represent a significant class of stakeholders, often a class that has a focused interest on the outcome of the rulemaking or are repeat players that represent interests that are generally affected by the rulemaking agency’s policies. Second, by the time a rule is proposed, which is before any comments are officially received, the rulemaking agency staff often has discussed the proposal with significant interest groups representing various stakeholders and already knows most of the information that the comments subsequently provide.
Nonetheless, because of the “threat” of judicial review, comments play an important role by encouraging agencies to vet proposed rules with interest groups who would be affected by the rule, especially those groups that regularly participate in agency regulatory matters.
[H]ard-look review has deeply influenced the organizational structure of contemporary administrative agencies. Agencies hire experts to study and corroborate their policy decisions, staff to review and respond to comments, economists to evaluate the costs and benefits of different policies, and lawyers to draft preambles explaining the reasons for policy decisions and to defend agency actions.
The psychological literature on accountability suggests that review of the agency’s reasoning by a reviewer whose views are not known to the agency is ideal for encouraging agency deliberation to ensure it adopts the best rule. Under the reasoned decisionmaking requirement that courts have imposed when engaging in arbitrary and capricious review, judges review the agency’s reasons for choosing or omitting the data it addresses, the alternatives it considers, and the arguments it provides for its ultimate rulemaking choice to the extent that the reviewing judges consider these matters relevant to the agency decision. Data, including suggested alternatives or arguments in the comments, are fair game for a court to find relevant. Thus, the prospect of judicial review leads agencies often to inform themselves on matters that are likely to be raised by comments even before they issue a NOPR. In fact, a recent study of rulemaking by three very different agencies demonstrated that comments are more likely directly to prompt changes from the agency’s proposed rule when the agency process does not include pre-NOPR discussion with significant stakeholders. Essentially, even if comments themselves do not induce the agency to alter the final rule, notice-and-comment procedures reinforce the incentive judicial review provides to an agency to consider data, alternatives, and arguments that it might otherwise ignore.
This becomes especially important when an agency faces political pressure to reach a preordained outcome, such as might occur when the White House has announced a preferred outcome early in the rulemaking process. Facing such pressure, an agency has great incentive to find data that supports the preordained outcome and to interpret the data it has as supporting that outcome. It might even go so far as to shortcut rational deliberation to reach the desired outcome. Notice-and-comment thus helps prevent the agency from engaging solely in self-confirming searches and interpretations of evidence and presenting questionable justifications for the rule it ultimately adopts. In short, the opportunity to comment is a crucial part of the process of creating a record for judicial review of an agency rule that forces agencies to engage in deliberative rulemaking.
Some scholars also claim that notice-and-comment procedures add democratic legitimacy to agencies’ action, which are headed by unelected officials and are only weakly accountable to Congress. At the simplest level, viewing the administrative process from a Pluralist perspective, one can argue that the notice-and-comment process provides information about the expected value of an agency rule in terms of the value the polity puts on it. Commenting requires that an individual learn of the rulemaking and inform themselves about any potential final rule at least sufficiently for the commenter to conclude whether they support the rule rather than some alternative. It also requires that the individual write and file the comment with the agency. Thus, the willingness to invest enough to overcome these costs and express an opinion on the rulemaking outcome can be seen as an indication of the value to commenters of their preferred rule. Taking this value into account and multiplying it by the number of commenters supporting a particular outcome provides agencies with a signal of the overall value of a particular rule to the entire polity. With the advent of electronic rulemaking, the costs of filing comments have decreased, which has vastly increased the number of individuals who participate in some rulemakings, and hence might be seen to render rulemaking potentially more democratic.
It is difficult, however, to maintain that notice-and-comment procedures render agency action meaningfully democratic in nature. As this Article already noted, participation in the process is biased in favor of those with focused interests. Second, even those who do participate as individuals in a rulemaking generally do not understand the technical issues and the ramifications of the various plausible alternative rules to ensure that their expressed rulemaking preferences accurately reflect their underlying values. Sometimes a NOPR does generate an overwhelming filing of comments by the public, but the ability of modern technology to facilitate the generation and filing of such comments renders such comments unreliable as an indication of the public’s preferences. Thus, notice-and-comment procedures are best understood not to ensure agency fidelity to the preferences or values of the general polity but rather to create a record that makes credible the threat of judicial reversal if an agency fails to deliberate adequately when promulgating a rule—that is, if the agency fails to identify likely effects of plausible final rules and consider the impact of those effects on all stakeholders.
Comments also can act as a “fire alarm” to warn members of Congress of the concerns of interest groups about a rule that might affect their chances of losing their seats, either because the rule adversely affects their constituents or, more often, because it adversely affects a group that can organize politically to oppose the members’ reelection. To politicians, comments signal stakeholders’ reactions (even if not well-informed reactions) to an agency-proposed rule. Hence, although the number of comments would not be relevant to justify a rule under the arbitrary and capricious standard of review, an agency would be well-advised to pay attention to a proposed rule that generates a groundswell of comments with strongly worded opposition if it wants to avoid congressional pressure to change the rule or even outright congressional reversal of the rule.
In sum, notice-and-comment procedures are best viewed not as providing information directly to persuade an agency whether a particular final rule is best but rather as part of a mechanism to improve agency accountability by forcing them to accurately reveal the effects of any final rule they adopt and thereby make transparent value judgments inherent in the rule. Hence, it is important that agencies know that ultimately they will have to allow notice-and-comment and convince a reviewing court that the rule adopted after such procedures satisfies the judicial standards for arbitrary and capricious review.
The most significant asset of the good cause exception is that it allows an agency to adopt and put into effect rules more quickly than if the agency must go through the notice-and-comment process. This is a benefit to the extent that the rule adopted is better than the status quo ante. The greater the impact of the rule, the greater the benefit of reducing delay.
When Congress enacted the APA, notice-and-comment rulemaking and judicial review of rules adopted using this procedure were envisioned as efficient and flexible. The point of allowing comments was to allow those affected by the rule to inform the agency of the rule’s impact on them. The agency was not required to justify its decision in light of the record of all the information before the agency, including the filed comments. Rather, comments were meant to provide the agency with information that it could consider before using its expert judgment to decide what rule best served the goals of the statute. Courts were to review rules under a highly deferential arbitrary and capricious standard of review.
The speed and efficiency of notice-and-comment rulemaking, however, were never realized. Before Congress enacted the APA, agencies established most policies via adjudication, which required formal trial-type procedures and review of findings of fact under the substantial evidence standard applied to the record as a whole, which is less deferential than arbitrary and capricious review. Following the APA’s enactment, agencies did not immediately change to using rulemaking to make policy: agencies continued to rely primarily on adjudication to do so. Those agencies that did use rulemaking often followed the formal rulemaking requirements of the APA when their authorizing statute called for rules to be adopted after a hearing.
Agencies’ undue aversion to the use of informal rulemaking prompted the Supreme Court to send a clear message to agencies and lower courts that when a statute calls for rulemaking after a hearing, only informal notice-and-comment procedures are required. By the time the Supreme Court clarified that informal rulemaking was generally all that was required, however, many scholars had become concerned about agency abuses of discretion and biases against the public interest in the regulatory process. “Sometime in the middle of last century, ‘capture theory’ became the dominant paradigm of bureaucratic behavior.” Public choice theorists explained how the rulemaking process favored those with focused interests over those with diffuse interests. This scholarly unease in turn led to judicial expressions of concern by judges and Congress.
This concern prompted courts to impose requirements that burdened the informal rulemaking process. In response to the increased use of notice-and-comment rulemaking, initially, courts reserved the authority to order agencies to add procedures in addition to those required by their authorizing statutes and the APA to create a sufficient record for public and judicial scrutiny. Courts also viewed notice-and-comment procedures as creating a rulemaking record on which the agency had to justify its decision. In addition, they read the APA requirement that agencies include a statement of the basis and purpose for a rule when adopting it along with the APA’s authorization of judicial review to ensure that an agency rule is not “arbitrary, capricious, [or] an abuse of discretion” as justifying rigorous judicial inquiry into the agency’s factfinding, predictions, and reasoning when adopting a rule. Although the Supreme Court ultimately held that courts have no authority to order an agency to add procedures not required by the APA or other statutes, the Court affirmed that judges are to review rules to ensure that the agency adequately considered all relevant facts and arguments and provided a meaningful opportunity for interested persons to comment on the proposed rule.
These additional requirements for adopting a rule by notice-and-comment procedures greatly contributed to bogging down the process of informal rulemaking. Today, it is not unusual for an agency to take several years to develop a proposed rule and several more after the issuance of the NOPR for the agency to adopt the rule. Given the length of time it takes an agency to adopt a rule, the benefit of adopting a rule more quickly by dispensing with notice-and-comment is significant.
Congress’s dysfunction is another reason that the good cause exception is more attractive today than it was when the APA was enacted. Even when faced with situations that both parties agree would benefit from a regulatory response, Congress usually cannot reach agreement on what that response should be. Neither party is willing to support legislation proposed by the other party because that would allow the proposing party to take credit for solving the problem. Even when one party controls both the Capitol and White House, the filibuster, more often than not, prevents Congress from using traditional legislative processes to enact statutes that provide needed reforms. That puts pressure on the administrative state to regulate to prevent significant, sometimes catastrophic, outcomes. Given the current political realities of congressional inability to act, the potential benefit of quick regulatory fixes is substantial.
The discussion above indicates that use of the good cause exception to adopt an ultimate final rule can deliver substantial benefits but also might generate substantial regulatory costs. Use of the good cause exception would serve the public interest for adopting any rule when the cost of delaying a rule outweighs the cost of adopting a rule inferior to one that would be adopted if notice-and-comment procedures are used. It is difficult for courts to evaluate the likelihood that skipping notice-and-comment procedures will lead to adoption of an inferior rule or how greatly inferior such a rule would be without a full rulemaking record and a justification for the rule in light of filed comments. But, if the courts require comments and a full agency justification, then they have essentially required notice-and-comment proceedings and forfeited the potential benefits from use of the exception. Hence, courts might seem to be in a catch-22 in that they can only decide whether the agency was justified in skipping notice-and-comment procedures by requiring notice-and-comment.
Luckily, there is an alternative to both allowing the agency to use the exception to generate the ultimate final rule and forbidding the agency from doing so. An agency can issue an IFR which allows the agency quickly to implement a rule that is superior to the regulatory status quo ante, thereby reducing the cost of regulatory delay while allowing notice-and-comment procedures to commence immediately upon adoption, ultimately promising that the agency can choose a better rule if the comments cause the agency to believe that there is one superior to the IFR. Thus, the use of an IFR seems like the perfect solution to the good cause exception conundrum.
One might object that expanding the good cause exception is contrary to the APA, which sets out preadoption notice-and-comment as the general means for agencies to enact rules. However, the APA text states that the section on rulemaking does not apply when the agency can show good cause that notice-and-comment would “be contrary to the public interest.” Thus, the text of APA is entirely consistent with my thesis that the agency should invoke good cause and use interim final rulemaking whenever that best serves the public interest.
Nonetheless, some object to the expansion of the good cause exception because they claim it is contrary to the intent of Congress. For example, Professor Kristin Hickman and Mark Thomson have advocated for a strong presumption against use of IFRs, which the agency would have the burden of rebutting because liberal use of IFRs would undermine the intent of Congress when it enacted the APA that all but those rules falling within narrow exceptions were to adopted using pre-enactment notice-and-comment proceedings. Reliance on legislative intent divorced from statutory text, however, is a precarious position to defend. Textualists disavow that legislative intent even exists, let alone that it can be determined reliably enough to give meaning to otherwise ambiguous statutory text. But even if one views legislative history as a valid source of legislative meaning, defining such intent as some subjective desire of the legislature is not a defensible understanding of legislative intent. To have any claim to legitimacy as a tool of statutory interpretation, legislative intent must mean the legislature’s understanding of the meaning of the text they enacted into law. So understood, Congress’s intent regarding the breadth of the good cause exception is far from clear.
Admittedly, when the APA was adopted in 1946, the good cause provision was seen as a narrow exception to the notice-and-comment requirement. But the expectation that the exception would have limited breadth developed against a backdrop understanding that notice-and-comment procedures would render rulemaking quick and flexible. That understanding was never realized. Thus, the delays that notice-and-comment procedures create for rulemaking increase the cost they impose. Whether members of the Congress that enacted the APA would have believed that those costs were so great that following those procedures would be contrary to the public interest is anyone’s guess. In essence, what has changed since 1946 is not the meaning of the APA but the understanding that informal rulemaking would demand significant time and agency resources.
One might also object that Congress may have required notice-and-comment procedures for reasons other than to specify the functional mechanism for adopting the best rules. As then-Professor Scalia pointed out in his article on Vermont Yankee Nuclear Power Corp. v. National Resource Defense Council, Inc., Congress might require greater or lesser procedural burdens on agency rulemaking to respectively discourage or encourage the amount of rulemaking. This view depends on the seemingly unlikely conclusion that the Congress that enacted the APA predicted that the notice-and-comment process would develop into the cumbersome process it is today. But even if Congress in 1946 foresaw the development of today’s protracted notice-and-comment process, issuance of an IFR would not undermine any intent that notice-and-comment procedures discourage rulemaking in general because such issuance does not exempt an agency from the burden of notice-and-comment rulemaking. It merely delays that burden so long as the agency commits itself to allow and consider comments on the IFR and ultimately to adopt an FFR. Moreover, the issuance of an IFR does not alter the nature of judicial review of the FFR. In other words, use of the IFR process commits the agency to the devote the same level of agency resources to consider comments and justify its decision in light of them as would notice-and-comment rulemaking that began with a prepromulgation NOPR.
Nonetheless, despite the rosy picture this Article paints about the use of IFRs, unfortunately, their use does open up the possibility of agency abuses and other undesirable effects. Hence, while IFRs are promising for solving the good cause conundrum, agencies’ use of IFRs should be restricted to avoid these effects.
At first blush, the IFR seems perfectly suited to resolving the conundrum of the good cause exception and generally should be considered procedurally proper when issued without prior notice-and-comment. Issuance of an IFR, however, still allows an agency leeway to abuse the good cause exception and may also cause the agency to adopt something other than the ideal FFR. Thus, there will be factors that courts should evaluate to determine when these potential problems with a particular IFR should preclude an agency issuing it under the good cause exception.
At the outset, the use of interim final rulemaking raises questions about how to structure judicial review to ensure that an agency has an incentive to adopt an IFR only when doing so would serve the public interest. The problem arises because IFRs may only be in effect for a relatively short time, but their issuance can cause harm that extends beyond the time they are in effect and, in fact, might be permanent. In the most problematic scenario, an agency may issue an IFR that is challenged in court, but the agency then issues an FFR before the court has time to fully review the challenge to the IFR. Certainly, a regulated entity that has been fined or otherwise adversely affected because it violated the IFR would still be able to claim the invalidity of the IFR as a defense to the order penalizing it. But, there is a question whether an entity could argue for reversal of the FFR on grounds that the IFR was improperly issued. And even if the impropriety of an IFR provided procedural grounds for reversing a resulting FFR, it is unlikely that any deleterious effect the IFR had on the FFR would be remedied by reversing the FFR on such grounds. It is thus crucial that allowing the use of IFRs when they likely serve the public interest requires some means by which courts can quickly review them to ensure that, given their impacts, including their effects on any subsequent FFR, IFRs do not cause more harm than good.
One way for courts to assure timely review of the propriety of using the interim final rulemaking process would be for them to expedite consideration of a petition to stay an IFR solely on grounds that the IFR is not justified under the good cause standards as set out in the remainder of this Article. While this might seem to be asking a lot of the courts, the issues raised by a petition for such a stay would be limited to whether the IFR is justified as good cause for pre-notice-and-comment promulgation of the rule and not any issues that address the legal authority of the agency to adopt the rule or whether the rule passes judicial standards of arbitrary and capricious review.
The potential pathologies of an IFR include the following: that the IFR itself is not better than the regulatory status quo ante; that the agency might abuse the IFR to adopt a rule that is preferred by the current administration but that judicial or congressional review would prevent an agency from adopting if it had to use notice-and-comment procedures; and that the IFR will bias the subsequent consideration of comments so that the FFR serves the public interest less well than would a rule that the agency would ultimately have adopted had it not issued the IFR. Although all of these represent potential problems with agency issuance of an IFR, regardless of the nature of the problem, the ultimate question is whether the benefit of an IFR in terms of decrease in regulatory delay exceeds the costs of the agency adopting a less-than-ideal FFR. The remainder of this part of the Article addresses each pathology and explains how it should limit judicial approval of IFRs under the good cause exception to notice-and-comment rulemaking.
The possibility that the lack of notice-and-comment might allow the agency to adopt a rule that is inferior to the status quo ante counsels that courts should restrict the bounds of IFRs to rules with objectively recognized benefits. To justify invoking the good cause exception, an agency should present credible evidence that the rule will better serve the statutory scheme that the agency is authorized to regulate, as well as the public interest. It may be that under a full hard look review opponents to the rule could demonstrate that the agency failed to consider whether there are rules that would better serve the public interest. But, on the limited question of whether to stay the rule because it was issued prior to the opportunity for comment, the need for a quick judicial decision warrants judges addressing simply whether the rule is likely better than the regulatory status quo ante. Even with respect to that question, the nature of an IFR as an invocation of the good cause standard warrants some deviation from the potentially demanding standard that courts would apply under arbitrary and capricious review.
In reviewing an IFR, the agency should still have the burden of convincing the court that the IFR likely will be objectively better than the regulatory status quo. Admittedly, there may be considerable judicial leeway in determining whether the agency has identified what all would agree are net benefits over the regulatory status quo. However, consistent with the usual standards of reasoned decisionmaking, this will require the agency to explain how the IFR will affect those intended to be protected by the agency’s regulatory program. Often, this will involve matters on which the agency has expertise. But experts should be able to explain why their choice of issuing an IFR is most likely to lead to objectively better outcomes. An agency should identify evidence from which it concludes that the IFR is likely better than the regulatory status quo. Clearly, given the fact the IFR is a temporary gap filler, the agency need not conduct new studies to justify its adoption of an IFR. By the same token, it should not ignore publicly available information that undermines its evaluation of the superiority of the IFR. And if the evidence does not resolve that question with certainty, the agency should explain why it believes that it is more likely than not that the IFR will provide objective benefits vis-à-vis the status quo. The court should not defer to the agency due to mere uncertainty, but if the agency can explain why the rule is likely to lead to objectively superior outcomes, the court should not reject the IFR simply because it is uncertain that the agency’s findings and predictions are correct. In that limited sense, the court should defer to the agency’s expertise.
Although my liberalization of the good cause exception would generally expand the potential for its use when an agency adopts an IFR, there are cases in which courts have upheld agency invocation of the exception but that, under my approach, would have been stayed on grounds that the agency failed to show that the IFR likely would be objectively better than the regulatory status quo. One such matter involved whether the Attorney General appropriately invoked the good cause exception when he issued an IFR making the Sex Offender Registration and Notification Act’s (SORNA’s) registration requirement applicable to sex offenders who committed their offenses prior to SORNA’s effective date. The Circuits split on that question, largely because they disagreed whether the rule addressed an emergency given that Congress had given the Attorney General three years to decide the question, and the Attorney General had taken more than seven months to issue the IFR. From my point of view, the fact that the issue may not have been an emergency is not particularly troubling because if the rule provided a clear benefit, then there was value in adopting the rule sooner rather than later. What does trouble me about the use of interim final rulemaking for this matter is the Attorney General’s justification that the IFR would provide a benefit. The Attorney General did not rely on any evidence that making SORNA’s notice requirement retroactive would have any effect on sexual offenses; he instead relied on the mere possibility that retroactivity would reduce such crimes.
A special case in which an agency essentially fails to claim objective superiority of the IFR is when the agency chooses the rule based on a subjective value judgment. In such a case, courts should stay an IFR and await the agency adoption of the FFR before deciding on whether the rule is justified. This is one way that my proposed standard for review of IFRs differs from traditional arbitrary and capricious review.
Courts generally have held that, in applying arbitrary and capricious review to agency rules, courts are not to second guess evaluations of such judgments an agency makes. Thus, rules which the agency determines are better than alternatives because of such value judgments may satisfy hard look review. The problem with the agency relying on value judgments to justify adopting an IFR is that such judgments are inherently political in nature. Rules that are justified only by an agency’s value judgments essentially would likely not have been adopted by an agency within an administration that holds different subjective values about the outcomes to which the rule leads. Accepting such subjective justifications for IFRs would thus allow agencies to undo rules adopted by prior administrations without delay, leading to potential regulatory instability and uncertainty. It would also undermine the use of notice-and-comment to allow for public input into what is essentially a political matter. In contrast, if an agency adopts a rule because it learns of dangers inherent in the regulatory status quo that all agree are truly dangers to be avoided, then the efficacy of the rule depends on the accuracy of the agency predictions of the factual outcomes that flow from the rule, rather than simply to a different subjective evaluation of those outcomes. One might contest the accuracy of the agency predictions of outcomes from the rule, but if those predictions are supportable, there would be consensus that the rule likely would be superior to the status quo ante and would be a proper candidate to be issued as an IFR.
In addition, rules that are justified by appeal to subjective preferences often address issues meant to please the White House or perhaps to energize the base of the administration’s party. As such, frequently, they are adopted because of political pressure on the agency. While such pressure is not illegitimate per se, it often will short-circuit the agency’s deliberative process, which can lead the agency essentially to fail to reveal honestly the objective trade-offs of adopting the rule. Thus, it is precisely for such rules that notice-and-comment plays an important role in ensuring transparency and fair-minded deliberation in the agency rulemaking process. Hence, an agency should not be able to invoke the good cause exception for such rules.
A quintessential example of an IFR that was predicated on the different value judgment of an administration from that of the prior administration was the Department of Health and Human Services’ rule allowing employers and insurance companies an exception for religious reasons even from the need simply to inform the agency that they objected to the Affordable Care Act requirement that they provide coverage for contraception. The comparative benefits of this rule versus the regulatory status quo depended on the weight one puts on one’s religious conviction that reducing the incidents of sex that does not lead to pregnancy and the value individuals put on not feeling compelled to facilitate in any manner a program that they feel compromises their religious values compared to the value one places on a person’s interest in choosing to have sex without the risk of pregnancy. The values the American public places on these alternatives vary greatly. Hence, there would not be objective consensus about whether a rule serves the public interest even if we knew with certainty the extent to which the rule contributes to sex engaged in for reasons other than procreation. Because there was no objective indication of the superiority of the IFR over the regulatory status quo, courts should have prohibited the agency from using the interim final rulemaking process to issue this rule.
Even if an IFR likely better serves the public interest than the regulatory status quo ante, it may not be the best rule the agency could adopt. The public interest would best be served by replacing an IFR if agency analysis of comments indicates that there is a superior FFR. This suggests that an agency should act to finalize an FFR as soon as practicable. In his comprehensive study of IFRs for the Administrative Conference of the United States (ACUS), Professor Michael Asimow analyzed the outcome of IFRs and found that agencies did not finalize between 42% and 53% of them. Professor Asimow noted that many of these IFRs were airworthiness directives issued by the Federal Aviation Administration (FAA), which generally do not draw any comments and which the FAA does not finalize. But even excluding these directives, based on the data reported by Professor Asimow, agencies failed to finalize about 27% of the IFRs within three years after they were issued. This indicates that many IFRs are left in place far longer than is warranted and, in many cases, indefinitely. Addressing this problem requires an evaluation of why agencies fail to finalize many IFRs and whether that sometimes is justifiable.
Agencies might have reasons to want the IFR to remain in place, even though comments might reveal that there is a better rule that the agency could adopt. An agency might have adopted an IFR in the first place to avoid notice-and-comment procedures because the agency is under political pressure to adopt a rule that might not survive judicial scrutiny if the agency had to defend it in light of comments generated by the notice-and-comment process. If this were the case, the agency would be using the IFR to adopt what is essentially a permanent rule that it might not be able to adopt using notice-and-comment procedures. Similarly, an agency might reason that having to justify the rule in light of comments might provide an alarm that the agency is regulating in a manner with which those in Congress or the public disagree. Allowing an IFR to remain in effect without considering the comments filed in response to it could thwart potential legislative oversight. Clearly, either of these rationales malevolently thwarts the structures intended to hold agencies accountable and, thus, does not provide a legitimate basis for invoking the good cause exception.
Alternatively, an agency might reason that the IFR, once issued, is good enough. In other words, an agency might reason that its best use of its resources would be to allow the IFR to remain in place rather than devoting staff time and labor to consider amending it and ultimately defending whatever FFR was adopted. While this rationale seems more innocent than agency intent to thwart oversight, it is contrary to the structure of the APA, which envisions that the agency must be willing to devote the resources to analyzing any permanent rule that it adopts. As noted above, statutes may require administrative procedures not only as instruments to improve the quality of agency decisionmaking but also as burdens on agency resources by which Congress can manipulate agency incentives to regulate. Thus, avoiding notice-and-comment procedures altogether may simply make it too easy for an agency to adopt rules, resulting in more rules promulgated than Congress envisioned.
Moreover, even if an agency skips notice-and-comment procedures to minimize the resources it has to expend to deliberate about and defend a rule—rather than malevolently to minimize judicial or legislative oversight—taking the good cause shortcut in adopting a permanent regulation is likely to lead to less deliberative rulemaking. According to the psychology of accountability, hard look review after notice-and-comment rulemaking improves agency rulemaking not so much because courts will find and correct agency errors but rather because such review is structured to encourage the agency carefully to deliberate and to ameliorate potential decisionmaking biases. Thus, whether malevolent or not, allowing an agency to adopt an IFR and then avoid consideration of comments and comprehensive justification for the rule generally will result in the promulgation of suboptimal rules.
Currently, once an IFR is in place, an agency generally is not obligated to reconsider the interim rule in light of the comments it receives. And there are reasons why agencies often do not seriously reconsider IFRs. If an IFR reflected a reaction to political pressure, the agency, by issuing the IFR, often has already triggered the political response by the administration’s party’s base or the significant campaign contributor who it was meant to benefit. And to the extent that the IFR addresses some underlying problem that the public felt needed to be addressed, the fact that it ameliorated the problem to some extent likely would relieve some of the public pressure on the agency to further address that problem. Thus, not infrequently, an agency will simply let an IFR remain in force without seriously considering the comments the IFR generated. In essence, agencies often have an incentive to abuse the use of IFRs to adopt a rule that they desire, but that may not best serve the public interest.
To assure that IFRs are subject to notice-and-comment procedures within some reasonable time after they are adopted and that the rule is justified in light of those comments, courts should require that an IFR sunset at a certain time after the IFR is issued. If courts are to require every IFR to be subject to a sunset provision, however, that requirement needs to be implemented in a manner that both allows the agency sufficient opportunity to finalize the IFR in a manner that best serves the public interest, while also constraining agencies’ abilities to abuse the use of IFRs and thereby forfeit the benefits described above.
First, there should not be a standard sunset period that applies universally to IFRs. Each rule involves issues that relate in unique ways to the information to which an agency is privy, either by its own investigation or from the comments it receives. Also, different rules raise issues of different levels of technicality and complexity that will require different investments of agency time and resources to analyze carefully. Hence, the period during which an IFR remains in effect should consider the nature of the regulatory issue and the burden that collection and analysis of relevant information will impose on the agency.
To minimize the cost of having a less-than-optimal rule in place while the IFR is operational, an agency should maintain an IFR only for the minimal time necessary for the agency to digest comments likely to be filed and consider and react to the concerns they might raise regarding the interim rule. The agency itself, which has the best knowledge of how much time would be reasonable for thorough consideration of the FFR, should be responsible for proposing the time at which the IFR sunsets. As part of the agency’s justification for invoking the good cause exception that the APA requires, the IFR should explain why it would be unreasonable for the agency to have to finalize the IFR prior to the sunset date. In this manner, the IFR can maximize the benefit of a rule that immediately takes effect without compromising the potential for greater benefits when the rule is reconsidered in light of comments.
In his ACUS report on IFRs, Professor Michael Asimow decided against suggesting that all IFRs contain a sunset provision “because it would be likely to cause serious practical problems.” But, I believe that Professor Asimow gave insufficient weight to the APA’s understanding that permanent rules are to be subject to notice-and-comment and failed to consider how courts might treat whatever action the agency takes when the rule is finalized. Professor Asimow’s objections can be lumped into two basic categories: concern regarding the allocation of agency resources and the potential for undue influence of the IFR on agencies’ ultimate rulemaking decisions.
The first expresses concern that a sunset provision might obligate an agency to commit resources to analyzing comments and amending an IFR when agency resources can be better employed on other regulatory matters. The APA, however, clearly envisions that an agency will invest such resources when it commits to creating a final legislative rule. As noted earlier, the burden imposed by the rulemaking requirements may reflect legislative intent to prevent an agency from engaging in regulation of marginal import. Once the agency decides to consider issuing a regulation, it is committing to invest such resources as are required by the notice-and-comment process, subsequent potential judicial review, and political oversight. Use of IFRs should be about the timing of when such consideration occurs, not whether it must occur at all.
Perhaps Professor Asimow was envisioning an agency faced with a dire emergency that found it had to choose between abandoning regulating the matter addressed by an IFR with a looming sunset deadline and a matter that demanded immediate attention by a large part of its rulemaking staff. For example, Congress might have enacted a new statute that demanded a large initial investment of agency resources in initial implementing regulations. Or, perhaps, some catastrophic event might have occurred, such as the COVID-19 pandemic, whose importance dwarfed that of other agency regulatory obligations at agencies like the Centers for Disease Control and Prevention. In such instances, the agency could then amend the IFR by extending the sunset deadline. Such action would be an amendment of an existing rule and hence would be final agency action that could be challenged in court. The only issue would be whether the alleged emergency truly required delaying finalizing the IFR. Such instances would be extremely rare, and I envision that courts would reverse any extension of an IFR sunset deadline unless the emergency the agency claimed it had to address first was something the agency was unaware of when it issued the IFR and that all would agree was of sufficient import to take precedence over the agency’s preexisting regulatory commitments.
The second category of problems created by requiring IFRs to contain sunset provisions is that the provisions will bias the ultimate agency decision adopting an FFR. On the one hand, at the end of the sunset period, the agency might rush its evaluation of the comments and adopt a non-optimal FFR. However, adopting the FFR would be an amendment of the IFR and hence would again be final agency action subject to judicial challenge. And the FFR will have to address comments filed in response to the IFR. If the agency cannot justify its adoption of the rule rather than a superior alternative, the FFR will be reversed as arbitrary and capricious. In fact, just knowing that the FFR will be subject to judicial review will encourage the agency to carefully deliberate even in the face of the sunset deadline.
On the other hand, at the end of the sunset period, the agency might decide that it no longer wishes to devote the resources needed to determine and justify the best FFR, in which case it might simply allow the IFR to lapse. The sunset provision then essentially changes the default outcome if the rulemaking lapses from the IFR to the regulatory status quo that preceded the IFR. This would seem to be an undesirable outcome given that, to have constituted a justifiable invocation of the good cause exception, the agency must have demonstrated that it was likely objectively superior to the status quo ante. The problem is that without the sunset provision, an agency might issue an IFR that it never intends to finalize as a means of undermining judicial review and legislative oversight. A court would have no reliable means to ensure that the use of IFR was not malevolent, which would encourage agencies to use interim final rulemaking for all the wrong reasons.
One way to reduce an agency’s incentive to change its ultimate decision due to the burden of having to defend that decision when finalizing a rule would be to consider whatever action or non-action the agency takes regarding finalizing the rule to be final agency action subject to judicial review. Obviously, were the agency to amend the IFR, that would essentially change the existing rule. Similarly, were the agency to adopt the IFR as the FFR instead of letting it sunset, that too would constitute action that changes the rule from what it otherwise would be if the agency simply did not act. Thus, both of these courses of action would almost certainly be final agency action subject to judicial review. This Article proposes, in addition, that even if the agency fails to do anything and lets the IFR sunset, that inaction should be considered reviewable agency action. Essentially, when finalizing an IFR the agency has three choices: adopt an amendment to the IFR as an FFR, adopt the IFR as the FFR, or allow the rule to revert back to the regulation in place before the IFR was adopted. By considering allowing an IFR to sunset, the agency would have to defend its ultimate decision regardless of which of these three outcomes it chooses. This would avoid the agency factoring in the costs of defense of its action regardless of the agency’s choice.
One might object that having courts require that an IFR contain a sunset provision would violate the edict in Vermont Yankee that courts may not add procedures to those required by the agency authorizing statute and the APA. Courts, however, have not read Vermont Yankee to preclude them from adding procedures that render those required by the APA meaningful. The suggestion that an IFR must include a sunset provision is necessary to ensure that an agency will not abuse the good cause exception to escape entirely the APA requirement of notice-and-comment when such procedures serve the public interest. Hence, requiring a sunset provision is consistent with Vermont Yankee to the extent a court finds it necessary to ensure that the agency invocation of the exception meets the criteria set out in the APA.
Because an IFR may be an imperfect substitute for a NOPR, it may be appropriate for a court to limit interim final rulemaking to those situations where it is a good substitute, as well as to reverse an agency FFR that bears the vestiges of the use of the imperfect IFR. Some courts have rejected any FFR that results from an IFR that the court finds procedurally invalid. Others have categorically affirmed the use of the IFR to start the FFR rulemaking. Still, others have focused on the likelihood that the use of the IFR to commence the FFR rulemaking altered the FFR from the rule that the agency would have adopted had it issued a NOPR rather than issuing an IFR. Below, I assess each of these general approaches for reviewing not only the IFR itself but also the FFR that results from the use of interim final rulemaking.
Courts that categorically reject any FFR that results from an IFR that the court has determined is an invalid exercise of the good cause exception often suggest that the use of the IFR to commence the rulemaking may affect the final FFR, but they do not rely on the likelihood of such an effect to justify such rejection. Instead, these courts seem to be concerned that use of interim final rulemaking will encourage agencies to substitute such rulemaking for almost all rules, thereby allowing the good cause exception to swallow the preadoption notice-and-comment rulemaking requirement in the APA.
The thesis of this Article should make it clear that potential replacement of the notice-and-comment paradigm for rulemaking does not bother me. If the use of the IFR for this purpose results in a rule that serves the public interest as well or better than a rule that would result from a separate issuance of a NOPR, I contend that courts should accept the move away from prepromulgation notice-and-comment as a good development. Even from a formalistic perspective that seeks to remain true to the notice-and-comment paradigm, however, such categorical reversal of FFRs fails to consider that the IFR plays two separate roles: first, as a rule that carries independent force of law, albeit for a temporary period; second, as providing the notice of the agency intention to commence notice-and-comment rulemaking. The fact that an IFR is invalid as a legislative rule under the good cause exception has no bearing on whether it meets the criteria for a valid NOPR under the APA. And if it does, unless the use of the IFR instead of a NOPR changes the ultimate FFR, there is no formal or practical reason to reject the ultimate FFR.
By the same token, courts that categorically accept FFRs as procedurally legitimate also are problematic. These courts rely on the APA provision allowing them to take into account the rule of prejudicial error. For example, in Little Sisters of the Poor Saints Peter & Paul Home v. Pennsylvania, Justice Thomas, writing for the Court, upheld the FFR because—even if technically the APA requires a NOPR rather than an IFR to trigger submission of comments—the use of the IFR was harmless error. To be harmless, however, an error must have “no bearing on the procedure used or substance of the [ultimate] decision reached.” Use of an IFR, however, might alter the extent and nature of the comments submitted or bias the agency toward maintaining the IFR when it issues the FFR. But Justice Thomas never considered whether the use of the IFR detrimentally affected the process by which the agency ending up adopting an FFR that was identical to the IFR that commenced the comment process.
Having rejected categorical rejection or acceptance of a permanent rule that comes about by interim final rulemaking, I turn to the task of identifying and analyzing how the informal rulemaking process might alter the FFR from the rule that would result had the agency issued a NOPR instead. IFRs can influence the FFR detrimentally because their issuance might generate a different set of comments than would be filed if the agency followed notice-and-comment procedures before issuing any rule with legal force. IFRs might also affect an FFR by creating path dependencies as well as causing an agency to succumb to the well-demonstrated cognitive phenomenon of confirmation bias.
Some courts and commentators have argued that issuance of an IFR will change the likelihood and perhaps the quality of comments that the agency received compared to those it would receive had it proceeded by using notice-and-comment procedures—that is, issuing a NOPR instead of an IFR. They reason that psychologically a stakeholder that has problems with an IFR would be less likely to believe that the agency will pay attention to their comments once it has already decided to issue the IFR, which has independent force of law. That is, by making the rule effective pre-comment, stakeholders are likely to find that the rule is already cemented in place and, hence, would not bother to comment. But these judges and commentators fail to recognize that the agency has already committed to a basic rule when it issues a NOPR, and that comments influence the agency mostly by threatening potential judicial reversal when courts apply hard look arbitrary and capricious review to the adopted rule. If influencing judicial review is the primary motivation for comments, it would seem that an entity that finds an IFR problematic is just as likely to file comments expressing its opposition as it is to file a such comment in response to a NOPR.
Contrary to the assumptions of these judges and commentators, one might even surmise that issuing an IFR instead of a NOPR is more likely to generate comments, as well as to improve the quality of those comments. Faced with a NOPR, those affected by the proposed regulation might discount the likelihood that the agency will adopt the proposed rule or at least believe that the rule might be altered in a way they find more acceptable. They might, therefore, think it advantageous to free ride on the expectation that others will comment. Issuance of an IFR might signal that the agency is more serious about changing the regulatory status quo ante, and that filing comments is more imperative. Perhaps more significantly, the experience of stakeholders under the IFR might focus them on concrete ways that the rule affects them that they might have failed to perceive or think serious until the rule goes into effect. In essence, having an IFR in effect might educate stakeholders about the rule, which could lead to more specific and higher quality comments than they might file in response to a NOPR. As a bottom line, there are no data or convincing arguments that indicate that IFRs discourage comments or that they affect the care taken by commenters and, hence, the quality of the comments filed.
In short, it seems unlikely that the failure of the agency to seek comments before adopting a NOPR will deleteriously affect filing of comments. Given the lack of any reliable basis for thinking that use of interim final rulemaking will result in fewer or less helpful comments, I would suggest that this concern not disqualify either an IFR or the FFR that ultimately comes out of the process generated by that IFR.
Path dependence, in its most general sense, refers to the concept that a past choice can constrain future choices. “Inefficient path dependence” occurs when a past choice influences the cost-benefit calculus of a future choice so that the best choice is not the one that would have been best if the initial choice had not been made. Choice of railway track gauges is a classic example of inefficient path dependence. More than half of the world’s railroad gauges are 1,435 millimeters, even though there are significant net benefits that would have accrued had this gauge not been chosen for the first railroads in England in the 1820s. But once chosen, the need to interconnect railroads and perhaps to make construction of engines and cars universal provided incentives for subsequent railroads to use 1,435 gauge tracks, and in fact, going forward, it was probably efficient to do so.
Essentially, inefficient path-dependent regulation will occur when an initial regulation requires or stimulates investment particular to that regulation. Once that investment is made, the benefits of future investment may be less than if the initial regulation had not been promulgated. This often occurs because the initial regulation addresses a regulatory problem in part but does not cure the problem entirely. Thus, the benefit of further regulation is decreased. If the investment stimulated by the initial regulation would be lost by amending the regulation to solve the regulatory problem more fully, then investment in an amended regulation might cost more than the resulting benefit from that regulation.
To make this more concrete, consider the following example. Suppose a rule would reduce the cost of a particular air pollutant. At the time when the agency first addresses this problem, the agency is aware of a technology that will reduce this cost by $3 billion per year and will require an investment of $2 billion per year. The agency, finding that implementation of this technology provides a significant net benefit of $1 billion per year, promulgates a regulation requiring those who emit the air pollutant to install the technology and entities subject to the rule to comply with it. Suppose that subsequently, after the first technology is installed, the agency learns of a different technology that would have reduced the cost of the particular air pollutant by $5 billion per year and investment in the new technology would have cost $3 billion per year. The second technology is entirely different from the one required by the initial regulation so that all investment in the initially required technology does not reduce the cost of implementing the new one, but the benefits of the second technology do not increase due to the implementation of the original one. That is, the total benefit remains at a reduction of $5 billion from the status quo that existed before the first regulation was adopted. Had the agency known of the second technology when it first addressed the problem, the best regulation would have mandated the second technology, providing a net benefit of $2 billion a year. But, once the initial regulation is implemented, the additional benefit of installing the second technology is only $2 billion per year, which does not justify the additional cost of $3 billion per year to implement that technology. This analysis suggests that, in general, it would be unwarranted for an agency to adopt an IFR that would impose significant investment costs that likely would be wasted if the FFR differs from the IFR.
Path dependence does not mean that the good cause exception should never be used when the rule the agency would promulgate would require significant upfront investment. The loss of benefits from adopting the non-ideal rule must be balanced against the delay in implementing the ideal rule. Recall that the major benefit from an agency imposing an IFR is the avoidance of delay in getting the rule in place. Suppose that notice-and-comment procedures would allow the agency to learn of the ideal rule but that using those procedures would delay the effective date of the rule by one year. In the numerical example above, the extra benefit of the ideal rule is $1 billion per year over that of the initial rule. The one-year acceleration in getting a rule in place, however, provides a one-time net benefit of $1 billion. Hence, it would take only a year before the later-adopted ideal rule would increase total net benefits over those provided by the earlier-adopted non-ideal rule. It seems reasonable to assume that the ideal rule would remain effective for more than one year and, hence, that use of an IFR would be unwarranted.
Suppose instead, however, that the initial rule would provide a benefit in pollution reduction of $50 billion per year, and the ideal rule would provide a benefit of $52 billion per year. Assume that the investments for each technology remain $2 billion a year for the non-ideal technology and $3 billion for the ideal technology. Now a one-year delay in implementing the ideal rule will forfeit $48 billion. It would take forty-eight years for the benefits of the ideal rule to compensate for the costs of delay. It is probably unreasonable to assume that the nature of pollution reduction technology will not change such that a better technology than that which is currently ideal would provide even significantly more benefits. Hence, in this latter numerical example, it would pay to skip the preadoption notice-and-comment procedures and adopt the non-ideal IFR quickly.
The bottom line is that courts should hesitate to allow agencies to issue an IFR if the interim rule requires substantial investment by those subject to it unless the agency can show that the interim rule also promises much greater benefits than the cost of the investments it requires. Thus, when a potential IFR promises overwhelming benefits over the status quo ante, an agency would be justified in promulgating the IFR even if it obligated regulated entities to make significant investments that would not be recouped if a better rule is later enacted.
To sum up my analysis of use of IFRs in light of path dependence, an agency should generally be free to adopt an IFR under a relaxed good cause exception standard if doing so provides significant regulatory benefits over the regulatory status quo ante, and the IFR would not require significant investments that might be lost if the agency later determines that another rule is better. But even if a rule that promises substantial benefits would require regulated entities to make such significant investments, the courts should evaluate agency use of an IFR under the traditional strict standard that would meet the traditional narrow good cause exception standard for an emergency.
If the agency issues an IFR that causes inefficient path dependence, which for that reason would be improper under my approach to the good cause exception and subsequently adopts the best FFR going forward, what should a court do in response to a challenge to that FFR? The court will then face the question of how to review the FFR which resulted from an invalid IFR. The answer is quite clear, even if surprising. Once the agency adopts and the industry implements the IFR, the path dependence has occurred. At that point, the most efficient rule going forward is not the one that an omniscient agency would have found most efficient prior to the IFR being adopted. Thus, if the agency simply focuses on developing the best rule after the IFR has been issued and implemented, then the court should ignore the invalidity of the IFR and review the FFR as if it had resulted from the agency using traditional prepromulgation notice-and-comment.
The very act of adopting an IFR could bias the agency consideration of what constitutes the best FFR toward the IFR. Psychologists have found that decisionmakers have a propensity to reaffirm a tentative initial preference or a choice they previously made—a phenomenon they label as confirmation bias. This bias can exist even at the subconscious level—that is, when decisionmakers do not intend to give and even are unaware that they are giving favored treatment to their initial choice. Confirmation bias is ubiquitous; cognitive and social psychologists have demonstrated that it occurs in political, economic, scientific, and even judicial decisionmaking. Confirmation bias manifests itself in two ways: decisionmakers tend to search for evidence confirming their prior choice more than disconfirming evidence; decisionmakers also tend to discount the value of disconfirming evidence compared to that of confirming evidence.
Confirmation bias can be triggered even by forming an initial hypothesis. Thus, when an agency simply proposes a rule, the rulemaking staff charged with collecting data to evaluate the costs and benefits of the rule are potentially subject to confirmation bias. It is not clear whether agency bias will increase if the agency takes additional action committing it to the rule, such as adopting it as an IFR. Confirmation bias might reflect cognitive dissonance between having generated the rule and then considering arguments that it is not valid, in which case it is unlikely to be increased merely by adopting the rule instead of just proposing it. But confirmation bias may also reflect a disfavor for having to spend additional time and attention doing an analysis in addition to the one that the staff completed with respect to the initial proposed rule or IFR. If this explains the bias, then whether the agency will be more biased depends on whether staff had to invest more time and resources in issuing an IFR than a NOPR.
In addition to confirmation bias, the fact that an agency issued an IFR might indicate that the agency has a conscious preference for that rule over alternatives before the agency considers comments. Some courts have addressed this concern by presuming that issuance of an IFR indicates that the agency had made up its mind about the ultimate FFR it would adopt, putting the burden of rebutting this presumption on the agency. In applying the open mind standard, the D.C. Circuit has looked at two factors as indicative that the agency had an open mind: whether the agency actually incorporates changes that reflect the comments into the FFR, and whether the agency explicitly engages in careful and searching consideration of comments, especially by discussing them in the preamble to the FFR.
Probing the mind of the regulator, however, is a difficult task that necessarily entails uncertainty. That task is made more difficult by the courts specifying the criteria they will use in determining whether the agency had an open mind, because the specification allows the agency to manipulate the criteria to suggest that it had an open mind when it didn’t. For example, an agency can make sure that the FFR includes some changes to the details of the IFR that might make little difference in the fundamental way the rule operates, about which the agency might have had a securely closed mind. And an agency might include in the rule’s preamble a discussion of each comment opposed to the IFR, even though it might have intended from the outset not to change the rule in light of those suggestions and arguments.
Finding both the categorical approach to validity of an FFR as well as the open mind test problematic, Professor Kristin Hickman and Mark Thomson suggested a “better middle ground” to resolve whether courts should reverse an FFR due to the potential for agency bias. Their approach, however, is also grounded in ensuring that the agency has an open mind about the ultimate FFR. They approve of the D.C. Circuit’s consideration of whether the FFR differs from the IFR and whether the agency explicitly addressed comments in opposition to the IFR when it adopted the FFR. They add that courts should explicitly look for an agency motive other than a belief that the agency used the IFR strategically to simply avoid the “hassle” of notice-and-comment or to “expedite the agency’s policy preferences,” and they argue that such considerations make it less likely that the agency would take “postpromulgation public input seriously.”
The problem with both the D.C. Circuit’s open mind inquiry as well as the Hickman and Thomson variation on it is that it focuses on agency state of mind and motivation for invoking the good cause exception. The best argument for this inquiry is that it will deter agencies from abusing the IFR process to push through rules that they otherwise could not justify. The problem is, however, that having an open mind or being motivated to adopt the IFR as the ultimate rule does not closely correlate with whether the FFR is, in fact, the best rule. In other words, the open mind and Hickman and Thomson standards will deter agencies from using the IFR process in many cases when issuing an IFR will best serve the public interest.
Perhaps more significantly, even if an agency adopts an FFR with a less than open mind—either consciously or subconsciously—than it would have if the rule that has gone through preadoption notice-and-comment, it is unclear that reversing the FFR is an appropriate or effective remedy. Although in a technical sense, such bias would not be harmless error because it is possible that issuance of the IFR detrimentally affected the FFR, nonetheless, were a court to rule that the issuance of an IFR was a procedural error that invalidated an FFR that otherwise is justifiable (i.e., would withstand arbitrary and capricious review), the remedy would require the court to remand the matter to the agency. The agency would then be free to readopt the rule using prepromulgation notice-and-comment procedures. But, given that it had initially adopted an IFR, whatever extra confirmation bias use of the IFR would generate would still affect the rulemaking process. And there is no reason to think that the agency would somehow abandon any conscious bias it had just because it restarted the rulemaking by issuing a NOPR. The result would be unnecessary delay and added administrative expense to adopt a rule subject to the same bias as the FFR. Alternatively, after remand, the agency could decide to abandon the rulemaking, returning regulation to the status quo ante before the IFR was issued, despite having justifiably concluded that the FFR was better than that status quo. Either choice is worse than the court affirming the FFR.
The determination that an agency’s motivation for its action differed from its reasons justifying the action might indicate an increased probability that the agency’s justification for the decision is inadequate. But the key to minimizing the risk of bias is recognition that courts can ameliorate it by ensuring use of appropriate procedures and judicial review of agency action. Notice-and-comment procedures, even when such procedures occur after the IFR has taken effect, coupled with the hard look standard of judicial review, provide a powerful constraint against biased agency decisionmaking. Commenters who disfavor the agency’s preferred rule can file comments that include disconfirming data and analyses, which ameliorates the problem of a biased agency exposing itself only to information and analyses that support its action. And hard look judicial review demands that the agency justify its action in light of relevant information and analyses included in the comments. Explicitly demanding a decisionmaker to consider disconfirming evidence that is provided in comments provides a strong palliative against biased agency action. Therefore, the structure of such review, which essentially asks the agency to analyze the disconfirming evidence in the comments, encourages an agency carefully and honestly to evaluate the true bearing disconfirming evidence has on the question of whether a proposed rule is best.
If the opportunity to file comments and judicial review are to counteract confirmation bias effectively, as part of such review courts must guard against an agency failing to justify the FFR in the face of disconfirming comments. Unfortunately, courts sometimes apply more deferential standards when evaluating whether an agency action is arbitrary and capricious. In the context of review of rules that reflect technical or scientific inquiries, courts generally apply what some have labeled “super deference.” Even in review of non-technical rules that are adopted using orthodox notice-and-comment procedures, courts often apply what others have called “thin rationality review.” In particular, in cases where an agency choice reflects uncertainty, courts have sometimes simply deferred to the agency choice without considering whether the agency resolution of the uncertainty is the most likely one. It is especially important, however, in light of the potential increased risk of bias from use of the interim rulemaking process, that a court not simply defer to an agency’s resolution of uncertainty, even for technical choices. Thus, when an agency issues an IFR to initiate the rulemaking that results in adopting its ultimate final rule, courts should demand that an agency justify why it believes that its factual determinations and predictions are the most likely resolution of uncertainty if judicial review is to provide incentives against bias.
One might question whether the relaxed standard for invoking the good cause exception if an agency issues an IFR that meets my criteria would actually prompt agencies to issue IFRs in order to gain the advantage of an objectively good rule taking effect immediately. The standard I propose seems to leave discretion to the courts to decide whether a rule is objectively better than the status quo, which might prompt agencies, fearing judicial reversal, to refrain from attempting to use the expanded scope of the good cause exception that I propose. Also, it may be difficult to identify any rule that does not involve some reliance by those subject to it, raising the question of whether any IFR would meet my criteria that would not be justified under the current strict standard for invoking the good cause exception. Therefore, I think it is helpful for me to identify a potential candidate for use of my standard.
Recently, the Federal Trade Commission (FTC) proposed a rule to prohibit employers from entering into, or attempting to enter into, a non-compete clause with a worker. The proposed rule broadly prohibits any person who “hires or contracts with a person to do work for them” from entering into a contract “that prevents the worker from seeking or accepting employment with a person, or operating a business, after the conclusion of the worker’s employment with the employer.” Although this proposed rule has been criticized by some as overly broad, such agreements may legitimately allow an employer to prevent the disclosure of valuable proprietary information or allow competitors to gain an advantage by poaching employees after they obtain industry specific skills that require significant investment by the employer in on the job training. Imagine, however, that the FTC had instead adopted as an IFR a prospective ban on non-compete agreements applied only to employees whose work does not give them access to an employer’s trade secrets and for whom the employer does not invest significantly in developing industry specific skills that could be captured by a competitor were the employee to change jobs. Non-compete agreements in these situations serve no function other than to restrict competition in the labor market. As such, an IFR restricting employee non-compete agreements for such employees would seem to provide an immediate objective benefit and would not involve any significant investment by employers in complying with the rule.
Such an IFR would allow the FTC to work out the details of determining which jobs are covered. Line operators in a manufacturing plant, and servers and cooks in the fast food industry would seem easily to be covered by the rule. For other workers, the FTC might use indicia such as whether an employee works at will and an employee’s wage as proxies for any employer claim of significant investment in teaching the employee skills that would be valued by a competitor in the industry. Ideally, the FTC might provide for the rule to take effect within sixty days of issuance, and for the comment period on the IFR to extend for one year from the time the IFR takes effect. The preamble to the rule should indicate that when issuing the IFR, the FTC will consider criteria that would allow it to apply to employee non-compete agreements more broadly than the criteria that limits the IFR. By doing so, the agency would gain a year’s experience with this limited rule that would help it craft an FFR that would be better grounded in predictions of the benefits and costs of a rule prohibiting employee non-compete agreements in contexts where such benefits and costs are less clear.
The problem of employee non-compete agreements is clearly not an emergency that would allow the FTC to issue an IFR addressing the problem under current standards for invoking the good cause exception. The depression of the wages that might result from such agreements does not threaten an individual’s life or health, nor does the allowance of such agreements threaten severe disruption of the national economy. But it would serve the public interest to make an initial limited ban on employee non-compete agreements where such agreements can serve no legitimate effective as soon as possible.
Since Congress enacted the APA in 1946, courts have consistently repeated the rhetoric that the good cause exception to notice-and-comment rulemaking applied in very limited circumstances. Recently, agencies have invoked the exception more broadly, and courts have not consistently applied it narrowly. Nonetheless, the universally acknowledged understanding of the exception is that it should not apply generally, lest it undermine notice-and-comment procedures as the paradigm for agencies to make rules.
This Article challenges the rhetoric that the exception should apply narrowly. It does so by arguing that an agency often can best serve statutory goals and the public interest by invoking the exception and issuing an IFR. It notes that an IFR allows an agency rule to go into effect without the long delay often required to complete notice-and-comment rulemaking. As long as the rule is better than the regulatory status quo, it is better for it to become effective sooner rather than later. It also rebuts arguments that issuing an IFR likely will result in a less desirable FFR because it will alter the extent and nature of comments filed or because it will lock the agency into a rule that has not reflected the deliberation associated with rules that are adopted using notice-and-comment.
The Article does concede that there are limitations on when agency issuance of an IFR is likely to better serve the public interest than rulemaking adopted via a NOPR and prepromulgation comments. It proceeds to identify those situations and counsel against use of an IFR when they exist, as well as highlighting certain ways that courts can ensure that the use of IFRs do serve the public interest. But it demonstrates that such situations are not so prevalent that they justify the restrictions on invocation of the good cause exception that traditional understanding of the breadth of the exception would demand.
I would like to thank Brenna Morecraft for her thorough and dedicated research assistance for this Article. I would also like to thank Bill Funk, Ron Levin, Nina Mendelson, Chip Murphy, Dick Pierce, and Chris Walker for helpful comments on previous drafts of this Article. Finally, I would like to thank the SMU Dedman School of Law, and Tom Mayo, its Acting Dean during the summer of 2022, for granting me faculty privileges so that I could continue my work on this Article during that summer when I had to be in Dallas for personal reasons.
