Join us.

We’re working to create a just society and preserve a healthy environment for future generations. Donate today to help.


This post was originally published by the Yale Journal on Regulation's Notice & Comment blog. Reprinted with permission.

Every President since Jimmy Carter has called on agencies to make retrospective reviews of their regulations. President Clinton’s Executive Order 12866 required agencies to create a program of periodic review of existing significant regulations. More recently both Presidents Obama in E.O. 13563 and Trump in E.O. 13771 likewise have required agencies to engage in retrospective reviews. Numerous commentators, not the least of which is Professor and former OIRA director Cass Sunstein, have extolled the potential value of retrospective reviews. And the Administrative Conference of the United States has issued recommendations providing support for agencies to review their existing regulations. Indeed, the Regulatory Flexibility Act (RFA) requires agencies to make a retrospective review of 10-year-old regulations that “have a significant economic impact upon a substantial number of small entities.” 5 U.S.C. § 610(a). Moreover, the RFA even provides for judicial review of an agency’s failure to comply with this requirement. 5 U.S.C. § 611.

Nevertheless, everyone who has studied what agencies in fact have done have concluded that agencies have largely failed in complying with these varying requirements. What is to be done? The Department of Health and Human Services (HHS) has a proposed answer: absent a retrospective review within a designated period sunset the regulation. Securing Updated and Necessary Statutory Evaluations Timely, 85 Fed. Reg. 70096 (2020).

On November 4, 2020, HHS issued a notice of proposed rulemaking to establish a schedule and standard for reviewing all existing regulations to assess first if they have a significant impact on a substantial number of small entities and, if so, to review those regulations to determine whether they should be retained, amended, or rescinded. The Department states that the proposal is responsive to the requirement in the RFA, but it seeks comments on whether its retrospective review should be expanded to include all regulations deemed significant under EO 12866. If, for any reason, the agency fails to make the assessment or (if an assessment finds the regulation has a significant impact on a substantial number of small entities) it fails to make the review, then the regulation would sunset and cease to exist.

The Department notes that it has in the past failed to make the reviews required by EOs and the RFA. It suggests that there are not sufficient incentives for the Department to comply with the literal requirements of the EOs and the judicially enforceable statutory requirement of the RFA. Thus, the threat of a regulation sunsetting is intended to provide the necessary incentive to making the required reviews.

There are several responses to the claim that sunsetting unassessed or unreviewed regulations is a necessary incentive to making required assessments and reviews. First, one might question what the real purpose of the proposal is. Second, one might question the legality of such a system. Third, one might question the wisdom of such a regulatory system.

Coming in the last months of an administration that has stressed the importance of rescinding regulations, such as the requirement in EO 13771 that agencies rescind two regulations for every new regulation adopted, and which has often been stymied in rescinding regulations because it did not follow the required legal procedures, one might guess that this proposal is intended to make rescission exceptionally easy. Under the proposal, all regulations that were adopted ten or more years ago, whether or not they have been amended since, must be assessed and reviewed within two years of adoption of the proposal or they cease to exist. Thus, an affirmative decision not to assess a regulation would result in the death of that regulation without further process. Why bother to assess or review an unwanted regulation for possible rescission when you can effectively rescind it simply by not assessing and reviewing it?

As to legality, imagine that a regulation requiring the Food and Drug Administration (whose regulations are subject to this proposal) to approve a drug before it may be sold is not assessed or reviewed by the required deadline, resulting in its immediate rescission. Someone challenging that rescission might argue both procedural and substantive claims. First, they might argue that a rescission may not take place without prior notice and comment. HHS answers this argument in its proposal by saying that this proposed rule, which would be the basis for the effective rescission of the FDA regulation, provides the required notice and comment. The substantive argument would be that it is arbitrary and capricious to rescind a regulation designed to protect the health and safety of the public simply because, for whatever reason, the Department failed to conduct a required assessment or review. HHS answers this argument in the proposal by saying that the benefits of providing the necessary incentive to review regulations outweighs whatever costs might be incurred by the sunsetting of whatever regulations the Department’s failed to assess or review.

The Department is probably right on the procedural argument. Certainly, an agency can provide for the termination of a rule or rules in the future in a present rulemaking, at least if in the present rulemaking it provides a reasoned basis for why the rule or rules should terminate on a certain date or on the existence of a particular condition. The problem for HHS is whether the present rulemaking does provide such a reasoned basis. Is it reasonable to have a rule terminate simply because the agency does not review it within the specified time? It is hard to imagine that a court would uphold the termination of a particular regulation simply because the agency failed (or decided not) to review that regulation within the regulatory time limit. For example, the Affordable Care Act regulations would all be required to be reviewed within two years of the proposal being adopted. Had President Trump been reelected, his promise to eliminate that Act, stymied in both Congress and the Supreme Court, could be fulfilled simply by HHS not reviewing those regulations.

Of course, the Department’s justification for its sunset provision does not rely on the reasonableness of the rescission of any particular regulation; instead, it argues that the sunset provision is reasonable because it is a necessary incentive for reviewing its regulations generally. With this incentive, the Department implies, all regulations will be assessed and reviewed within the time limits, so no regulation will be rescinded by a failure to be assessed and reviewed in time. This at least creates a ripeness issue with respect to a judicial challenge to the proposal, were it adopted. It could be argued that only when a particular regulation is made ineffective because it was not assessed or reviewed within the time limits would a judicial challenge be ripe. As suggested above, such a challenge then would presumably be successful, negating the sunset of the regulation.

Third, there is the question of the wisdom of using a sunset provision as an incentive to force review of existing regulations. Why is it that HHS needs such an incentive? Why is it that despite Executive Order and statutory requirements for retrospective reviews, HHS has largely failed to meet those requirements? The preamble to the proposed rule is strangely silent on that question. Instead, it quotes at length from a law review article but omits that article’s statement that the “chief reason agencies cite [for failing to do retrospective reviews] is that they lack time and resources.” Yoon-Ho Alex Lee, An Options Approach to Agency Rulemaking, 65 Admin. L. Rev. 881, 895 (2013). The Department plays down the time and resources necessary to do retrospective review, saying “[o]nly a handful of Department employees are needed to perform the periodic reviews.” 85 Fed. Reg. at 70099.

But, if that is so, why has it not complied with the retrospective requirements in the past? The answer is simple, no matter how you spin it: other matters were more important. Those more important matters would continue notwithstanding any impending sunset deadlines. The result would be either that the deadlines will be missed, and regulations will be made ineffective, or that the retrospective reviews will be perfunctory and superficial in order to meet the deadline. This has been the near universal conclusion of scholars involved in the field. As one particularly exhaustive study states: “most scholarship suggests the benefits of sunset laws are largely intangible and likely insignificant compared to the costs. Moreover, the burdens of such mandatory reviews can draw staff away from performing other vital oversight duties. Generally, sunset requirements produce perfunctory reviews and waste resources.” Jason A. Schwartz, 52 Experiments with Regulatory Review 24 (Report for Institute for Policy, 2010).

Moreover, the assumption that only a handful of employees could perform the assessments and reviews within the time limits is questionable. The Department’s Regulatory Impact Analysis indicates that in the first two years somewhere between 2,480 and 12,400 regulations, depending on how you count the Department’s regulations, would need to be assessed to determine if they have a significant effect on a substantial number of small entities. The Department claims that only 10 percent of these regulations were originally found to have a significant impact on a substantial number of small entities and therefore would necessarily require a review. That claim is based on a random sample of 10 regulations (of 2480!) only one of which would have necessarily triggered a review. Of course, changed circumstances might require a review of some of the remaining regulations, and all would need to be assessed in any case. According to HHS these figures would require between 27 and 67 personnel at a cost of between $7.5 million and $19 million in the first two years. No mention is made of asking for additional appropriations to cover the costs of these personnel, so it is likely that, as in the past, they will be competing with other demands on the agency, leading to a failure to complete the assessments and reviews and a rescission of regulations for no good reason.

In short, HHS’s proposal is an unwise and legally questionable solution to the problem. Whether HHS will actually attempt or be able to finalize its proposal and make it effective before the new administration takes office is doubtful. Looking forward, there are alternatives, such as those proposed by the Administrative Conference of the United States in Recommendation 2014-5, 79 Fed. Reg. 75114 (2014). Ultimately, however, the success of any undertaking of retrospective review will not come from artificial incentives but from political commitment to the process by leaders of the agency.