The Inherent-Powers Corollary: Judicial Non-Delegation and Federal Common Law

|The Volokh Conspiracy |

I've posted my new paper, The Inherent-Powers Corollary: Judicial Non-Delegation and Federal Common Law, to SSRN. I'll be blogging about it over the next several days. (This is the same paper that used to be called The Interlinking Doctrine (same subtitle), in case you saw that link on Larry Solum's blog a couple of weeks ago.) Here's the abstract:

Congress often explicitly delegates lawmaking power to courts-or does so implicitly by giving courts ambiguous terms to interpret. Do these laws violate the non-delegation doctrine?

I argue that there's nothing about the doctrine that exempts delegation to judges. But not all such delegations are unconstitutional. The traditional non-delegation doctrine is relaxed when Congress delegates into an area that's within, or "interlinked with," power that the delegate already has. This doctrine, which I call "the Inherent Powers Corollary," is central to understanding non-delegation.

Applying the Inherent-Powers Corollary to the judiciary comes down to determining what exact powers federal courts have. The inherent-powers perspective provides interesting insights into the constitutional foundations of the Erie doctrine. It also connects the non-delegation doctrine with the debate over the scope of federal common law. Federal courts' lawmaking powers include the power to make procedural rules, to make law in areas involving "uniquely federal interests," to create statutory defenses and remedies, and to engage in ordinary statutory interpretation. (Non-delegation arguments in the statutory context thus overlap with arguments about the permissibility of certain kinds of statutory interpretation.) It turns out that even under fairly restrictive views of federal common law, federal courts' lawmaking powers are extensive enough to save a great many-but not all-delegations.

* * *

Here's the Introduction to the paper:

Consider the following statutes, which seem to grant courts wide discretion and invite or require them to engage in substantial lawmaking:

  • The Rules Enabling Act: "The Supreme Court shall have the power to prescribe general rules of practice and procedure and rules of evidence for cases in the United States district courts . . . and courts of appeals."
  • Federal Rule of Evidence 501: "The common law-as interpreted by United States courts in the light of reason and experience-governs a claim of privilege unless [the Constitution, a federal statute, or Supreme Court rules] provide[] otherwise . . . ."
  • The Religious Freedom Restoration Act: "[The federal g]overnment may substantially burden a person's exercise of religion only if it demonstrates that application of the burden to the person (1) is in furtherance of a compelling governmental interest; and (2) is the least restrictive means of furthering that compelling governmental interest." (Similar language applies to state governments in a few limited contexts, under the Religious Land Use and Institutionalized Persons Act.)

Or consider the following statutes, which might direct courts to engage in substantial lawmaking, depending on how one chooses to interpret them:

  • Section 1 of the Sherman Act: "Every contract, combination in the form of trust or otherwise, or conspiracy, in restraint of trade or commerce . . . is declared to be illegal." (This grants the judiciary less discretion if one interprets it to refer only to the preexisting cause of action for unreasonable restraint of trade, more discretion if it invites the judiciary to develop a sensible antitrust policy.)
  • The Alien Tort Statute: "The district courts shall have original jurisdiction of any civil action by an alien for a tort only, committed in violation of the law of nations or a treaty of the United States." (Less discretion if it refers only to torts recognized under the Founding-era law of nations, more discretion if it invites the judiciary to create new doctrines.)
  • The Rules of Decision Act: "The laws of the several states . . . shall be regarded as rules of decision in civil actions in the courts of the United States . . . ." (Less discretion if it requires federal courts to use state decisional law as well as state statutory law, more discretion if it allows federal courts to create a "general law" when state statutes are silent.)

Is there anything unconstitutional about granting judges such discretion? One way to approach the question would be to ask whether these statutes would be constitutional if rewritten to provide comparable discretion to agencies. Here are two actual statutes, one which may be taken to give the executive branch (through its military tribunals) authority to interpret the "law of war" (much like the ATS may be taken to allow the courts to interpret the "law of nations"), and another which gives the executive branch (through the Secretary of Labor) authority to develop evidentiary privileges (much like Federal Rule of Evidence 501 does for courts):

  • Article 21 of the Uniform Code of Military Justice: "[M]ilitary tribunals [shall have] concurrent jurisdiction with respect to offenders or offenses that by statute or by the law of war may be tried by . . . military tribunals."
  • Section 6607 of the Patient Protection and Affordable Care Act, amending § 504 of ERISA: "The Secretary [of Labor] may promulgate a regulation that provides an evidentiary privilege for, and provides for the confidentiality of communications between or among, any of the following entities or their agents, consultants, or employees: [various state and federal agencies or a]ny other Federal or State authority that the Secretary determines is appropriate for the purposes of enforcing the provisions of this subchapter."

It would be clear that such statutes would be subject to analysis under the non-delegation doctrine. The Vesting Clause of Article I, § 1 says that "[a]ll legislative powers herein granted shall be vested in a Congress of the United States," and this has been taken to mean Congress can't divest itself of its legislative power (though this reading isn't obvious). In turn, this principle has been implemented by requiring that Congress, in delegating power, always provide an "intelligible principle" to guide the delegation and prevent it from becoming a forbidden delegation of legislative power (though this, too, isn't an obvious construction). In other words, Congress must make at least certain hard choices rather than entirely passing responsibility to someone else.

Not that such statutes would necessarily fail-the non-delegation doctrine is notoriously lax. Maybe lax is putting it mildly. Maybe it's kind of fictitious. The non-delegation doctrine has been capacious enough to uphold virtually every statute, including ones involving concepts like "unduly or unnecessarily complicate[d]" corporate structures and "unfair[] or inequitabl[e] distribut[ions]" of voting power, "generally fair and equitable" price controls, and the "public interest." In light of that caselaw, perhaps most statutes delegating to agencies would survive.

But what's clear is that some non-delegation inquiry would be appropriate for statutes delegating to agencies. And if the agency statutes clearly delegate authority to agencies, then it seems plausible that the statutes I began with should be thought of as delegating to courts. One might then wonder whether the non-delegation doctrine has anything to say about these congressional delegations to the judiciary.

This matters, because some of the judicial examples look iffy-the procedural statute has no guidance at all, and the evidentiary statute merely refers to "the light of reason and experience," which is about the same. Federal Rule of Evidence 501 isn't a fluke: § 3008 of the Resource Conservation and Recovery Act and § 113 of the Clean Air Act likewise provide that "[a]ll general defenses, affirmative defenses, and bars to prosecution that may apply with respect to other Federal criminal offenses may apply . . . and shall be determined by the courts of the United States according to the principles of common law as they may be interpreted in the light of reason and experience." A statute telling agencies to develop substantive rules "in the light of reason and experience" would seem vulnerable, even under the current loose regime.

It isn't obvious to everyone that the non-delegation doctrine even applies to the judiciary; some authors even state flat out that it doesn't apply, and others argue that it wouldn't make sense if it did, given the exceedingly general standards that courts often implement. I argue in Part II that the non-delegation doctrine does indeed apply to any delegate, whether the executive branch, the judiciary, or anyone else. The Supreme Court has on a few occasions held that it applies to courts in both their rulemaking and adjudicative capacities; and in fact there are sound policy grounds for applying the doctrine more strongly when the delegation goes to the judiciary than when it goes to the executive.

But that's fully consistent with our intuitions about the judiciary as Doer of Vague Things. The Supreme Court has also consistently held that the legitimacy of any delegation must be assessed in the context of the powers that the delegate already has: the ordinary "intelligible principle" doesn't apply when Congress delegates in an area that's within the delegate's existing powers. I lay out this doctrine, which I call the "Inherent-Powers Corollary," in Part III. In fact, merely talking about the delegate's existing powers is a bit too narrow: this relaxed non-delegation doctrine even applies when Congress delegates in an area that's slightly outside of the delegate's existing powers, but interlinked with them. That's the "Interlinking Curlicue" to the Inherent-Powers Corollary.

The non-delegation doctrine turns out, then, to be more interesting than we thought. Level 1 is the standard non-delegation doctrine with its "intelligible principle" test. Level 2 is the Inherent-Powers Corollary that relaxes the doctrine when the delegate could have done the congressionally mandated action anyway. Level 3 is the Interlinking Curlicue, which extends the Inherent-Powers Corollary to even some "interlinked" areas that are beyond the delegate's own powers.

For delegations to the executive branch, these core powers involve the military or foreign affairs. For delegations to federal courts, they're whatever is included in (or interlinked with) the judicial power. We can't argue about the constitutionality of congressional delegations to the judiciary without first having a view on the permissible scope of the federal courts' common-lawmaking powers, since the domain of federal common law (as well as other judicial powers) will tell us where the Inherent-Powers Corollary applies.

The idea that the non-delegation doctrine applies to delegations to the judiciary is a powerful one. It makes many ordinary statutes appear potentially suspect, and offers the possibility of wide new vistas of unconstitutionality. Thus, Andrew Oldham, writing about antitrust but here speaking more generally, writes that "Congress cannot deputize the federal courts-and federal judges cannot accept such congressional delegation-to make standardless policy judgments." Margaret Lemos argues that "the Sherman Act would be a likely candidate for constitutional invalidation" under this view; Eugene Kontorovich argues that a broad reading of the Alien Tort Statute would violate the non-delegation doctrine (as well as the constitutional reservation to Congress of the power to define offenses against the law of nations); Aaron Nielson argues that that the pre-Erie understanding of the Rules of Decision Act likewise violated the doctrine.

And yet, the Inherent-Powers Corollary is almost equally powerful in the other direction. While students casually understand Erie as having heralded the death of federal common law, students of Federal Courts learn that a reformed federal common law-quite unlike the pre-Erie "general law"-is alive and well, and applies pervasively, if usually interstitially, in a variety of miscellaneous contexts. Many of the statutes one might think would be suspect under the non-delegation doctrine thus turn out not to be.

But-perhaps because academic commentary on the applicability of the non-delegation doctrine to the judiciary is still in its infancy, the relevance of the Inherent-Powers Corollary has been either ignored or passed over too quickly. Some authors don't mention the Corollary at all; others mention it but treat it as though it only applied in a narrow area, like procedural rulemaking or remedies or common lawmaking associated with statutes. Justice Scalia suggests that statutory interpretation and procedural rulemaking don't violate the non-delegation doctrine because they're "ancillary" to courts' exercise of judicial power; this is on the right track because it implicitly recognizes an Inherent-Powers Corollary, but he doesn't develop that idea any further or explain what makes something "ancillary."

The Inherent-Powers Corollary fills the gaps in these previous accounts of non-delegation-and helps to clarify our thinking. The proper scope of federal common law is hotly disputed, and the suggestion that non-delegation imposes limits on federal courts' common-lawmaking power is an interesting contribution to that debate. But this suggestion has serious flaws, and the Inherent-Powers Corollary shows us why. The Corollary requires us, when doing a non-delegation analysis, to first determine whether the delegation to courts falls within an area where courts have an inherent (i.e., statute-independent) federal common-lawmaking power (or, under the Interlinking Curlicue, in an area "interlinked" with such power). If it does fall within that area, the Inherent-Powers Corollary applies, and so the non-delegation doctrine has virtually no force.

Only if the delegation falls outside the area of inherent judicial power do we go to the next step: to determine whether Congress has sufficiently limited the delegation with an intelligible principle. At that point, the power that courts are exercising comes to them by delegation from Congress, rather than being inherent. Only then can one argue that non-delegation principles might constrain this statute-dependent federal common-lawmaking power. But to skip that first step is to miss a fundamental aspect of the problem and deny courts their inherent, pre-statutory flexibility.

Making the Inherent-Powers Corollary central to non-delegation analysis shows how one can save a number of statutes that some thought were doomed (and that others had never even realized were threatened). Perhaps some statutes really are doomed, but determining this will require some moderately fancy footwork. This is the job of Part IV, where I go through a number of areas of judicial power.

First, I analyze federal courts' procedural rulemaking power. Commentators have been right to suggest that such delegations are valid, but they've generally ignored a potentially problematic aspect of such delegations: the Rules Enabling Act gives courts the highly unusual power to repeal existing statutes that conflict with the procedural rules that the courts end up promulgating. Is this constitutional? At first glance, obviously not! At second glance, maybe, if we squint hard enough; but at third glance, no, after all.

Second, having disposed of courts' procedural rulemaking power, I turn to courts' adjudicative powers-which immediately invites thinking about the constitutional foundations of Erie. Nielson's view-that the Rules of Decision Act, if interpreted in the Swift v. Tyson way, was an invalid delegation-is problematic, because the Supreme Court has held that the Rules of Decision Act merely restated existing law. Thus, current doctrine states that Erie not only is now, but has always been, the proper rule, even if there had been no congressional action. But with no congressional action, there can be no non-delegation problem. Moreover, as I've stated above, the Inherent-Powers Corollary teaches us that evaluating the Rules of Decision Act under the non-delegation doctrine requires us to first determine whether federal courts have a general lawmaking power-which is the very holding of Erie. If that's so, Erie itself can't be justified by the non-delegation doctrine; to do so is to invert the correct order of analysis.

Third, I look at federal courts' power to make common law in areas involving "uniquely federal interests." There's disagreement over how far this category extends, so what statutes are constitutional under the Inherent-Powers Corollary depends on one's prior theory of the scope of these uniquely federal interests. But it's clear that at the very least, these interests include particular subject-matter enclaves like interstate controversies, foreign affairs, admiralty, and the proprietary interests of the federal government. Because federal courts have power to make common law related to foreign affairs, the Alien Tort Statute is constitutional even if it lacks an intelligible principle.

Fourth, I discuss federal courts' power to create statutory defenses. This theory could support Federal Rule of Evidence 501 and comparable delegations of the power to create criminal defenses in RCRA and the Clean Air Act-but would have greater difficulty supporting RFRA.

Fifth, I look at federal courts' power to create remedies. The sort of unconstrained discretion that judges exercise in choosing a sentence from a statutory range typically isn't thought to be unconstitutional, and thus seems to be an inherent judicial power. If that's so, this power is an alternative basis for upholding the Sentencing Guidelines.

Finally, I address statutory interpretation. The idea that passing statutes with ambiguous terms is a delegation to the judiciary-and that statutory interpretation is thus delegated lawmaking-may not be obvious to everyone, but in this section I defend the idea that it is.

At least some interpretive methods are within courts' inherent power; perhaps some aren't. Usually, when a judge chooses an interpretive method that's beyond his power, we simply say he's acting ultra vires: Congress has nothing to do with it. But we could, seeking congressional intent, interpret the statute as commanding the judge to use a particular method. If that method is within the judge's power, we have a delegation that's valid under the Inherent-Powers Corollary. If not, Congress might still be able to mandate that the judge use it, but because the Inherent-Powers Corollary doesn't apply, we require an intelligible principle.

Much talk of impermissible interpretive methods can actually be recast in non-delegation terms, to the extent the method is both too vague and exceeds judges' inherent powers. But if you think all interpretive methods are fair game for judges, then all statutory interpretation is valid under the Inherent-Powers Corollary.

I apply this reasoning to the residual clause of the Armed Career Criminal Act, RFRA, the Alien Tort Statute, and § 1 of the Sherman Act.

I conclude in Part V. I suggest that on the one hand, the Inherent-Powers Corollary should be central to how we talk about delegation. But on the other hand, the Interlinking Curlicue-which relaxes the non-delegation doctrine even when the delegate lacks an inherent power-is dubious and should be reconsidered.

* * *

P.S. I now see a blog post by Josh Blackman on one topic that I cover in my paper, which links to an article by Leslie Kelleher that it looks like I should read now.