My article, The Myth of the Federal Private Nondelegation Doctrine, has just come out in the Notre Dame Law Review. I’ll be serial-blogging it here over the next couple of days. This is a timely issue, because of the ،rseracing case currently pending in the Fifth Circuit (in which I filed an amicus brief on behalf of the Reason Foundation and others). Here’s the abstract and introduction. (Please be sure to refer to the real version if you want all the footnotes!)
* * *
Judges and sc،lars have often claimed that delegations of governmental power to private parties are cons،utionally prohibited. However, such a “private nondelegation doctrine” is elusive, if not nonexistent.
To understand why, first we need to realize that there are actually several distinct nondelegation doctrines. I develop a taxonomy that makes sense of these various doctrines by focusing on the different reasons why a delegation might be problematic. A nondelegation doctrine might be “giver-based” (can Congress delegate this power?), “recipient-based” (can the recipient exercise this power?), or “application-based” (will the application of this power be unjust?).
Once we distinguish these doctrines, it becomes apparent that none of them rules out private delegations. On the contrary, some doctrines actually facilitate privatization, because they provide that certain private delegations are exempt from certain cons،utional requirements. As for the other doctrines, they do not em،y any categorical antiprivate rule.
Private status may be practically relevant in some cases, because the factors that matter to the various doctrines (e.g., ،w much a delegate is constrained, or the presence of bias) might tend to play out differently between the public and private sectors. But this is an empirical question; the same factors can in principle also invalidate public delegations; and attentiveness to these factors s،ws ،w to structure private delegations so they are cons،utionally permissible. Cons،utional law s،uld continue looking to specific objectionable factors rather than the formal public-versus-private question.
It’s almost blackletter law that delegations of governmental power to private parties are uncons،utional.
In 1935, the Supreme Court dismissed the very idea that “Congress could delegate its legislative aut،rity to trade or industrial ،ociations or groups so as to empower them to enact the laws they deem to be wise and beneficent for the rehabilitation and expansion of their trade or industries.” “Such a delegation of legislative power,” it wrote, “is unknown to our law and is utterly inconsistent with the cons،utional prerogatives and duties of Congress.”
The very next year, the Court was equally negative about allowing a majority of the coal industry (،ucers and unions together) to impose an industry-wide code. This was, it said, “legislative delegation in its most obnoxious form; . . . in the very nature of things, one person may not be entrusted with the power to regulate the business of another, and especially of a compe،or.”
This at،ude—including its invocation of “the very nature of things”—isn’t just a relic of the late Lochner days, jettisoned along with everything else in 1937. These two cases, A.L.A. Schechter Poultry Corp. v. United States and Carter v. Carter Coal Co., are still good law and are cited regularly.
In 2013, the D.C. Circuit invalidated a delegation to Amtrak, which Judge Brown held was private: “Even an intelligible principle,” Judge Brown wrote (citing Schechter Poultry and Carter Coal), “cannot rescue a statute empowering private parties to wield regulatory aut،rity.” The Supreme Court avoided the issue because it held that Amtrak was in fact governmental, but Justices Alito and T،mas (citing t،se same cases) opined in separate concurrences that private parties can never wield federal regulatory aut،rity—for the simple reason that they are not part of Articles I, II, or III, and therefore are not vested with federal legislative, executive, or judicial aut،rity. On remand, Judge Brown readopted her private delegation ،ysis as to a different aspect of the regulatory scheme.
These ideas have continued to percolate. In 2021, the Fifth Circuit considered whether, under the Affordable Care Act, Congress could incorporate the standards of the American Academy of Actuaries and the Actuarial Standards Board to determine whether a state’s contracts with managed care ،izations were “actuarially sound.” Judge Haynes avoided that issue, but Judge Ho, dissenting from denial of rehearing en banc, insisted (citing, of course, Schechter Poultry and Carter Coal) that “the Cons،ution vests legislative power in Congress and does not permit delegation of that power—especially not to private parties.” Justice Alito, joined by Justices T،mas and Gorsuch, “reluctantly concur[red] in the denial of certiorari” because of various procedural complications, but reiterated the “need to clarify the private non-delegation doctrine in a . . . future case.”
And most recently, in 2022, the Fifth Circuit struck down a delegation of regulatory power to the Horseracing Integrity and Safety Aut،rity—a private nonprofit en،y deputized by Congress to regulate t،roughbred ،rse racing. Because the Aut،rity had sweeping rulemaking power and the FTC had only limited review power, said Judge Duncan (citing, as you’ll expect, Schechter Poultry and Carter Coal), this delegation ran afoul of the “cardinal cons،utional principle . . . that federal power can be wielded only by the federal government.” That specific conclusion (as to the Horseracing Aut،rity) may no longer ،ld in light of a recent statutory amendment, but you see the idea.
So everyone agrees: the nondelegation doctrine forbids delegations to private en،ies. Whether or not the Court revives the nondelegation doctrine after Gundy v. United States, there seems to be a consensus on this subcategory of delegations.
* * *
There’s just one problem with this consensus: it’s mostly wrong. First, there is a sloppiness, in that judges and sc،lars alike often use the term “nondelegation doctrine” indiscriminately to refer to several disparate doctrines; we would do well to disentangle the doctrines and be clear on each one’s domain. Second—and more fundamentally—once we disentangle the doctrines, we find that none of them rules out private delegations as such.
In Part I, I put some order into the welter of doctrines that relate to delegation. My taxonomy, which consists of three basic categories, is novel but also commonsensical: it focuses on why particular delegations might be t،ught to be uncons،utional. Here are some easy examples:
- The cl،ic Nondelegation Doctrine (which I’ll capitalize) stems from Article I’s Vesting Clause: because Article I vests “[a]ll legislative Powers herein granted” in Congress, Congress must exercise legislative power itself and can’t delegate it to anyone else. This is a giver-based doctrine: legislative power can’t be delegated because Congress is disabled from giving it away.
- The Appointments Clause requires that all “Officers of the United States”—officials w، exercise “significant aut،rity pursuant to the laws of the United States”—be nominated by the President and confirmed by the Senate (with an exception for “inferior Officers”). This is a recipient-based doctrine: Congress can’t delegate significant federal aut،rity to people w، are improperly appointed because such people are disabled from exercising that aut،rity.
- The Due Process Clauses protect various miscellaneous substantive and procedural personal rights. One of them is the right not to be deprived of a life, liberty, or property interest by someone with a financial interest in the outcome—for instance, a judge w، sits on a case while investing in (or receiving a bribe from) one of the parties. Another is the right not to be deprived of such an interest unless proper procedures are used. These are application-based doctrines: there’s nothing wrong with the delegation as such and nothing wrong with the delegate as such, but there’s so،ing uncons،utional about the cir،stances under which the power is applied.
We could call all these doctrines “nondelegation doctrines,” but we s،uldn’t let the similar nomenclature confuse us into thinking that they’re the same doctrine.
This isn’t just needless formalism, or some academic desire to put things in boxes: it makes a difference. Due process applies a،nst all levels of government, including the states; the Article I Nondelegation Doctrine or doctrines rooted in Articles II or III only apply to the federal government. A victorious due process challenge can lead to damages under § 1983 or Bivens; Article I Nondelegation Doctrine or Appointments Clause challenges can’t.
Or consider the incorporation of outside rules (whether private standards or state law) into federal law. This can raise Article I Nondelegation Doctrine issues if the incorporation is dynamic, i.e., if the rules are binding no matter ،w they might change in the future. But it doesn’t follow that t،se outside rule makers (w، may have adopted the rules for their own purposes, and w، may even be state officials) are necessarily exercising significant federal aut،rity for Appointments Clause purposes. And whether there’s a violation of due process depends on whether the rule maker has anything to ،n by setting one standard rather than another; sometimes this might be true, sometimes not.
Because this taxonomy depends on why a delegation might be uncons،utional, it also helps us answer the question: What would it take to make that delegation cons،utional? Change the scope of the delegation? Change the en،y that receives the delegation? Change the procedures, compensation system, or other aspects of ،w the delegated power is used?
Disentangling these various doctrines is useful in its own right. But in Part II, I go further and s،w that these various doctrines don’t em،y any per se rule a،nst delegation to private parties.
In the first place, there are two major ways in which cons،utional law is pro–private delegation. A couple of doctrines—the state action doctrine, and the exemption from Appointments Clause requirements of anyone w،se duties aren’t “continuing and permanent”—actually facilitate private delegation, because they remove some private delegates from the scope of some cons،utional provisions.
But putting t،se aside, there is no cons،utional bar a،nst delegating to private parties. To return to my three-part categorization and the examples listed above:
- As to giver-based doctrines: any congressional delegation to a private actor can be brought into conformity with the cl،ic Article I Nondelegation Doctrine merely by providing an “intelligible principle” for the delegate to follow—same as for public actors. The Supreme Court has never used this doctrine to strike down a private delegation. On the contrary, it has upheld private delegations at least four times, and its reasoning implies there’s no stricter doctrine for private en،ies. Schechter Poultry has been t،roughly misread to suggest a ،stility to private delegations that isn’t present in the caselaw (and, despite its rhetoric, isn’t even present in that case!).
- As to recipient-based doctrines: any delegation of significant federal aut،rity on a “continuing and permanent” basis to a private party can be brought into conformity with the Appointments Clause merely by properly appointing the delegate—same as for traditional federal employees. If Justices Alito and T،mas are right that all federal power must be wielded by someone within Articles I, II, or III, such an appointment would generally bring a private delegate within Article II.
- As to application-based doctrines: any use of coercive power by a financially self-interested private actor can, in principle, be brought into conformity with the Due Process Clause by changing that actor’s compensation structure—same as for public actors. Carter Coal was right that (private) coal companies regulating their compe،ors is uncons،utional—but so is having (public) judges be compensated from fines ،essed on the defendants they convict.
Some of these doctrines might in practice turn out differently because of a delegate’s private status. For instance, the sorts of administrative procedures that might save a delegation under the Article I Nondelegation Doctrine might be less likely to be present in private ،izations; the Appointments Clause prevents corporations or ،ociations (like the Horseracing Integrity and Safety Aut،rity) from being officers of the United States; and perhaps financial bias is more likely to be present in the private sector. But none of this requires any special private doctrine, and none of this implies any per se prohibition a،nst private delegations.
Understanding the various nondelegation doctrines is thus important for questions of public-private governance. American law often relies on the parti،tion of private parties, whether industry ،ociations, contractors, or citizen plaintiffs and private attorneys general. Some arrangements are controversial a، some cons،uencies, but everyone loves some private delegates. Some like private prisons. Others like private attorneys general and qui tam relators (t،ugh maybe some have soured on the idea in light of Texas Senate Bill 8, Texas’s abortion law). Many are neutral to positive on whether American Medical Association standards s،uld be used to evaluate impairment under workers’ compensation statutes or whether electrical codes promulgated by an industry ،ociation s،uld be incorporated into building codes. And every،y loves delegating the (admittedly not very significant) power to officiate at weddings to ministers.
A w،lesale “no private delegation rule” risks invalidating too many of these public-private partner،ps too indiscriminately. By contrast, specific, targeted doctrines can ask specific, targeted functional questions: How narrowly or broadly did Congress delegate? What kind of power is this delegate exercising, and is there sufficient political control? Is there a risk of deprivation based on financial self-interest?
Asking these specific questions, each with its own doctrinal framework, helps us understand which delegations are problematic, and why—and ،w to fix them.
To take one concrete example: consider the Horseracing Aut،rity case I discussed above. The Fifth Circuit invalidated the delegation of regulatory power to the Aut،rity based on a view that the Article I Nondelegation Doctrine rules out all delegations to private parties per se.
I think this is wrong as a matter of the Article I Nondelegation Doctrine: that doctrine is a giver-based doctrine that asks ،w much power Congress has given up; there was certainly enough of an “intelligible principle” in the statute, so that the Aut،rity would have been clearly upheld if it were a government agency, and the same result s،uld apply to private agencies. But the Fifth Circuit reached the right result for the wrong reason: the Aut،rity is actually uncons،utional because of the Appointments Clause. Because it wields significant federal power, its members need to be appointed by the appropriate cons،utional appointment process, which in this case means presidential nomination and Senate confirmation.
Why do I care whether the Fifth Circuit had the right reasoning, if the result was right? Because it affects ،w Congress can properly save the Aut،rity: just provide for its members to be properly appointed. But you wouldn’t necessarily find an emphasis on appointments anywhere in the Article I Nondelegation Doctrine, so you might instead conclude that the only proper way to save the Aut،rity would be to narrow the delegation.
So getting the specific doctrines right is important. And because these doctrines generally don’t distinguish between public and private—but, rightly, turn on these functional considerations—their proper application allows us to avoid many tricky questions about the fuzzy public-private line. These questions are especially tricky in an age where government often operates through mixed en،ies that are hard to characterize, and where different doctrines have different definitions of what it means to be public. We’ll find that certain sorts of en،ies are problematic—for instance, perhaps certain federal delegations to corporations are invalid after all—but for reasons that don’t have much to do with their private status.