Prof. John Harrison: More Reasons Unlawful Regulations Are Void Ab Initio

This is the third in a series of posts summarizing an article titled Remand Without Vacatur and the Ab Initio Invalidity of Unlawful Regulations in Administrative Law, which is forthcoming in the BYU Law Review. The current draft is available on SSRN.

Regulations that are unlawful as that term is used in section 706(2) of the APA are void when adopted. The previous post reviewed the criteria that make a regulation unlawful under that provision, showing why satisfying any of the criteria makes a regulation void when adopted. This post sets out two more reasons for the conclusion that unlawful regulations are never binding.

First, review of regulations in enforcement proceedings, which the APA specifically contemplates, rests on the assumption that unlawful regulations are void before any court evaluates them.

"Except to the extent that prior, adequate, and exclusive opportunity for judicial review is provided by law, agency action is subject to judicial review in civil or criminal proceedings for judicial enforcement." 5 U.S.C. § 703. Section 706 governs the scope of review, so its criteria apply when review takes place in an enforcement proceeding. Enforcement proceedings often involve conduct that has already taken place; criminal enforcement proceedings always do so. If the enforcement court finds that the regulation being enforced is unlawful, the defendant prevails. That familiar result reflects the assumption that unlawful regulations are invalid before any court has the opportunity to find them so. If unlawful regulations were binding until displaced by a court, a violation of a regulation enforced by criminal punishment that took place before judicial displacement would be a crime. Violations of unlawful regulations are not crimes, because unlawful regulations are never valid.

Next, although the Supreme Court has not addressed the doctrine of remand without vacatur, it has had occasion to identify the features that make a regulation binding law. Chrysler Corp. v. Brown, 441 U.S. 281 (1979), concerned proprietary business information that Chrysler had submitted to the Defense Department. Chrysler sued Secretary Brown, seeking an order barring public disclosure of that information. Chrysler relied on the Trade Secrets Act, which forbade disclosure of specified business information "'to any extent not authorized by law.'" Id. at 295. As authorization for disclosure, the government pointed to a regulation issued by the Office of Federal Contractor Compliance Programs (OFCCP) of the Department of Labor.

To satisfy the Trade Secrets Act, the Court found, a regulation had to have the force and effect of law, and the OFCCP regulations lacked binding legal force. The Court concluded that the statutes on which the government relied as authorizing the regulations did not do so. Nor had the regulations been adopted through the procedure required to make binding rules, because the Department of Labor had characterized them as interpretative and had not conducted notice and comment rulemaking. Lacking statutory authorization and the correct procedure, the regulations were not legally binding.

The Court's conclusion in Chrysler Corp. is unsurprising. Of course statutory authorization and proper adoption procedures are necessary conditions for the promulgation of a legally binding regulation. The anomaly is not Chrysler Corp., but remand without vacatur. According to that doctrine, regulations have the force of law even if they rest on no statute, and even if they were not adopted through the process Congress requires.

As the preceding post and this post have shown, the principles that govern the ab initio validity of unlawful agency regulations are different from the principles that govern the binding force of lower-court decrees that rest on error. The latter are binding when issued, but the former generally are not. I say generally, because within constitutional limits, Congress has power to give an agency authority to act with provisionally binding effect, even when the agency acts contrary to a statutory directive. When Congress does that, it often can enable courts to take a step that undoes the agency's action and functionally resembles vacatur by an appellate court. With respect to regulations that impose duties of conduct on private parties, however, Congress rarely does so. The analogy between agencies and lower courts on which remand without vacatur rests is unsound.

The next post considers remand without vacatur in two additional contexts. First, it deals with statutes regulating private conduct that eliminate enforcement-stage judicial review, replacing it with a form of pre-enforcement review. Although systems of review of that kind might be thought to depart from the principle of ab initio invalidity of unlawful regulations, they do not. Second, the next post briefly discusses agency activities other than regulation of private conduct. I explain that the analogy between agencies and lower courts on which remand without vacatur rests hinders proper understanding of the courts' responses to the wide variety of agency activities.

Comments

[ReaderY]

Would like to follow up on the reliance issue.

As I understand it, the basis for lower court orders having binding effect is essentially equitable, because of the unfairness that would occur if a reversed lower court order was considered void ab initio, so that parties and other people were bound to act as if the order hadn’t existed. Problems similar to ones addressed by the ex post facto clause would ensure.

As I understand it, the argument you are goving is statutory, basically that Congress has to authorize everything an agency does, so if its regulations are to be provisionally binding, Congress has to say so.

My first question is, why can’t courts use equity to fill in gaps? Why can’t it consider things like reliance consequences, particularly if a regulation had a minor and easily correctable technical issue but voiding it would have devastating and disruptive consequences?

My second question is, are there arguments that this situation is really different from an equitable point of view from when a lower court’s orders are reversed?

Let me try to come up with as strong a hypothetical where equitable reliance interests might apply that I can think of off the top of my head. The FDA makes a procedural error in approving a drug, which everyone agrees is safe and effective and should have been approved if only the FDA had crossed the right T’s and dotted tbe right I’s. Suppose the drug is lifesaving. Does the drug really have to be pulled from the market immediately with all patients left in the lurch? Does a court really have no ability to apply equitable principles to tell the FDA to go back and follow the correct procedure without voiding its action entirely?