The intelligible principle standard lives to see another day—but for how long remains unclear.
Here we go again. Will the nondelegation doctrine, long thought dead, be resurrected? Court watchers are buzzing over the U.S. Supreme Court’s recent decision in Gundy v. United States, the latest in the Court’s nondelegation jurisprudence. Contrary to the wishes of some, the Court—yet again—upheld a statute against a nondelegation challenge. Yet again, however, the justices offered hope to those who wish the Court would apply the doctrine with more (or really any) rigor.
The Supreme Court has been quite consistent over time in asserting that Congress cannot delegate legislative power without running afoul of the U.S. Constitution’s Article I vesting clause. For about a century, the Court has evaluated whether Congress has violated this principle by asking whether the statute in question contains an “intelligible principle” to direct and constrain executive action. Yet in the history of the Republic, the Court has employed the nondelegation doctrine to invalidate only two provisions of a single statute, the National Industrial Recovery Act, in Panama Refining Co. v. Ryan and ALA Schechter Poultry Corp. v. United States, in 1935.
Since those cases, the Court has repeatedly upheld statutory provisions calling for agencies to adopt such regulations as are “reasonable,” “feasible,” “in the public interest,” or some equally vague term—leading many commentators to declare that the nondelegation doctrine is dead. Yet the nondelegation doctrine, like hope, springs eternal.
Forty years ago, prominent judges and scholars such as Carl McGowan, John Hart Ely, and Ernest Gellhorn made the case for reinvigorating the nondelegation doctrine and curtailing agency discretionary power. In Industrial Union Department, AFL-CIO v. American Petroleum Institute, the Supreme Court in 1980 considered whether the Occupational Safety and Health Act violated the nondelegation doctrine by allowing the Occupational Safety and Health Administration (OSHA) to adopt limitations for workplace exposure to harmful chemicals. OSHA adopted those limitations by using an open-ended “feasibility” standard that itself was inferred from, rather than explicitly provided by, a statutory requirement that such limitations be “reasonably necessary and appropriate.” The Court found the statute constitutional based on an interpretive inference that the statute required a finding of “significant risk,” although it also invalidated OSHA’s regulations for failing to provide that finding. Justice Rehnquist concurred in the judgment but, citing Ely, argued that the statute was limitless and violated the nondelegation doctrine.
Twenty years ago, the U.S. Court of Appeals for the D.C. Circuit surprised many lawyers by declaring in American Trucking Associations, Inc. v. EPA that the Environmental Protection Agency violated the Clean Air Act by construing too broadly the Act’s authorization of national ambient air quality standards limited solely by fuzzy terms like “requisite to protect the public health” with “an adequate margin of safety.” The Supreme Court reversed, with Justice Scalia for the Court rejecting the notion that an agency can “cure an unconstitutionally standardless delegation of power” and also finding the statute’s broad terms fell “well within the outer limits of nondelegation precedents,” and suggesting that the Court would only rarely be qualified to second guess Congress’s decision to give agencies policymaking discretion. Justice Thomas concurred but expressed “willingness to consider whether delegation jurisprudence has strayed too far from our Founders’ understanding of separation of powers.”
And now, this most recent term, Gundy. Philip Hamburger’s critique of the contemporary administrative state has captured the imagination of many scholars and commentators. Hamburger rejects as unconstitutional virtually any statutory grant of authority to adopt legally-binding regulations—irrespective of whether the statute contains an intelligible principle. Meanwhile, several of the justices in recent terms have expressed dissatisfaction with and willingness to reconsider aspects of administrative law doctrine to curtail agency discretion. The Court’s taking up of Gundy prompted many to wonder whether this time, just maybe, the Court would breathe new life into the nondelegation doctrine.
Gundy concerned the constitutionality of section 20913(d) of the Sex Offender Registration and Notification Act (SORNA), which authorizes the Attorney General to determine the applicability of SORNA’s registration requirements to sex offenders convicted prior to SORNA’s enactment. The provision, taken on its own, is quite open ended. It directs the Attorney General “specify the applicability of” SORNA’s requirements and “prescribe rules for the registration of” pre-Act sex offenders, without further direction. Of course, that provision is part of a larger statute that is reasonably detailed regarding the mechanics of registration for sex offenders more explicitly covered by the statute’s terms. Suffice it to say, the Attorney General relied on this delegation of authority to adopt regulations that required pre-Act offenders to register.
Justice Kagan wrote the opinion of the Court, representing a plurality of four justices. She read the statute as requiring the Attorney General to “apply SORNA’s registration requirements as soon as feasible to offenders convicted before the statute’s enactment.” Of course, the relevant statutory provision does not actually include the word “feasible.” Rather, Justice Kagan inferred this feasibility from SORNA’s statutory declaration of purpose and legislative history, as well as the Court’s 2012 decision in Reynolds v. United States, which she said had held that Congress clearly intended SORNA’s registration requirements to apply to pre-Act offenders, albeit only after the Attorney General adopted regulations. In other words, SORNA obligated the Attorney General to adopt regulations adapting and applying the rest of SORNA’s registration scheme to pre-Act offenders, with only a little latitude for feasibility.
Justice Alito concurred in the judgment purely because of stare decisis. A majority of participating justices were unwilling to reconsider the intelligible principle standard, and given the Court’s history of applying that doctrine, “it would be freakish to single out for special treatment” the provision in SORNA.
Justice Gorsuch wrote a blistering dissent, joined by Chief Justice Roberts and Justice Thomas. Describing SORNA’s section 20913(d) as permitting “unbounded policy choices” with “profound consequences,” and noting the U.S. Department of Justice’s admission that the provision “does not require” the Attorney General “to act at all,” Justice Gorsuch observed further that different Attorneys General have adopted different rules governing which pre-Act offenders might be covered.
According to Justice Gorsuch, under a different Attorney General, Gundy might not have “faced an additional 10-year prison term.” Gorsuch also took particular aim at the lack of guidance provided by mushy words such as “feasible,” noting that this word “might refer to ‘technological’ feasibility, ‘economic’ feasibility, ‘administrative’ feasibility, or even ‘political’ feasibility.” He also noted that SORNA’s section 20913(d) did not actually impose a feasibility requirement, but rather that the plurality inferred such from other sources.
SORNA aside, of greater significance is Justice Gorsuch’s discussion of delegation more generally. Surveying the history of the Court’s nondelegation doctrine, Justice Gorsuch contended that the intelligible principle standard had gained “a life of its own” unmoored from traditional understandings of the nondelegation doctrine. Yet, admitting the need for some test for evaluating congressional delegations, Justice Gorsuch outlined what he described as three “important guiding principles” provided by the framers.
First, although Congress must “make the policy decisions when regulating private conduct, it may authorize another branch to ‘fill up the details.’” Second, although Congress must “prescribe the rule governing private conduct, it may make the application of that rule depend on executive fact finding.” Finally, “Congress may assign the executive and judicial branches certain non-legislative responsibilities.”
By one reading, SORNA’s section 20913(d) was fortunate to survive at all, and Gundy was likely the intelligible principle standard’s last gasp. Four justices declared themselves willing to replace the intelligible principle standard. Assuming that Justice Kavanaugh agrees with them, SORNA’s section 20913(d) was saved purely by the fact that Kavanaugh was confirmed shortly after the Court heard arguments in Gundy, and a reinvigorated nondelegation doctrine is tantalizingly close to becoming a reality.
On the other hand, key obstacles remain in the way of a more rigorous nondelegation doctrine. The first is the difficulty of finding a meaningful substitute for the intelligible principle standard. Justice Gorsuch is correct that the intelligible principle standard is worthless as a meaningful limitation on congressional delegations of legislative power. If statutes that rely on content-less terms like “reasonable,” “feasible,” or “in the public interest” reflect intelligible principles, then no real-world piece of legislation will ever violate the standard, and Panama Refining and Schechter Poultry are best viewed as dusty relics of their political climate.
But finding a better and more rigorous standard for discerning between acceptable from unacceptable grants of rulemaking authority is very, very hard. Contrary to Philip Hamburger, none of the justices has suggested eliminating agency rulemaking altogether. Justice Gorsuch even seemed to find acceptable “statutes that allow federal agencies to resolve even highly consequential details so long as Congress prescribes the rule governing private conduct.” His primary objection seems to be with mushy terms like “feasible,” and perhaps even then not all the time. Consequently, any more rigorous replacement for the intelligible principle standard will need to facilitate such line drawing. Justice Gorsuch’s first effort, contrasting “mere ‘details’” with rules governing final conduct, seems too susceptible to the whim of the moment.
Then there is the problem of stare decisis. Justice Kagan’s assertion that “if SORNA’s delegation is unconstitutional, then most of government is unconstitutional” is clearly overblown. Yet clearly, any more rigorous nondelegation standard that the Court might adopt would immediately call into question the ongoing validity of more than a few existing statutes, some quite longstanding, as well as regulations implementing those statutes and an even larger number of enforcement actions based on those regulations. Opinions in other cases this term have suggested that some number of justices may be willing to approach stare decisis more aggressively and overturn past precedents. But the justices have not yet pulled the trigger.
This is not to say that those justices who think administrative law has tilted too far in agencies’ favor should do nothing to rein in delegations they find objectionable. Justice Gorsuch went on to note in his dissent that the Court “still regularly reins in Congress’s efforts to delegate legislative power,” just using “different names” such as the “major questions” doctrine or “void for vagueness.” And the Court uses other doctrines beyond those to limit agency discretion.
In just this past term, for example, the Court curtailed Auer deference in Kisor v. Wilkie and applied the State Farm standard to reverse and remand a decision to include a citizenship question on census forms in Department of Commerce v. New York. As statements about separation of powers and the roles and responsibilities of the three traditional branches, these alternatives lack the symbolic value of replacing the intelligible principle standard. But they also lack the baggage.
In short, the intelligible principle standard survives another round, while its opponents are given ample reason to hope and predict that, surely, the next case will see the standard’s end. At the very least, perhaps we will not need to wait another twenty years for that next case raising the nondelegation doctrine.
This essay is part of a series, entitled The Supreme Court’s 2018–2019 Regulatory Term.