Judicial Non-Delegation, Part 2

|The Volokh Conspiracy |

The other day, I introduced my new article, The Inherent-Powers Corollary: Judicial Non-Delegation and Federal Common Law, which I've posted to SSRN. I'd appreciate any comments on how to make it better—thanks in advance!

In today's post, I explain that the non-delegation doctrine does apply to the judiciary. And constitutional structure, Supreme Court precedent, and policy considerations all suggest that it should-and even that it should be enforced more strongly against courts than it usually is against the executive branch.


As a matter of structure, consider the roots of the non-delegation doctrine in the Vesting Clause. Legislative power has been vested in Congress; this means Congress can't give it up; and this in turn has been interpreted to mean that Congress must limit any power it yields by an intelligible principle. The focus is on how much power Congress has given up, not who is the recipient of the power.

(Thus, as I've written elsewhere, the non-delegation doctrine applies to delegations to private parties in exactly the same way that it applies to delegations to executive agencies-though private delegations might end up raising concerns under other constitutional provisions.)


One of the very early non-delegation doctrine cases, Wayman v. Southard, arose precisely in the context of a congressional delegation to the judiciary. The same is true of one of the most recent non-delegation doctrine cases, Mistretta v. United States. These are cases about judicial rulemaking (procedural rules in Wayman, the Sentencing Guidelines in Mistretta). As for judicial adjudication, the Supreme Court also rejected a non-delegation challenge to the Sherman Act in Standard Oil v. United States. So there's no question that precedent supports the application of the non-delegation doctrine to courts.

Wayman v. Southard concerned federal procedural statutes enacted in 1789 and 1792, which provided that federal courts in any state would use state procedures as of 1789, "subject however to such alterations and additions as the said courts respectively shall in their discretion deem expedient, or to such regulations as the supreme court of the United States shall think proper from time to time by rule to prescribe to any circuit or district court concerning the same." In response to a challenge to such judicial rulemaking authority, Chief Justice Marshall wrote:

It will not be contended that Congress can delegate to the Courts, or to any other tribunals, powers which are strictly and exclusively legislative. But Congress may certainly delegate to others, powers which the legislature may rightfully exercise itself. . . .

The line has not been exactly drawn which separates those important subjects, which must be entirely regulated by the legislature itself, from those of less interest, in which a general provision may be made, and power given to those who are to act under such general provisions to fill up the details.

Chief Justice Marshall ultimately concluded that this statute was just about detail-filling-not about "strictly and exclusively legislative" powers-so the delegation was unobjectionable here. So a non-delegation analysis does apply to the judiciary. And indeed, the Supreme Court has repeatedly cited Wayman for general non-delegation propositions (including involving the executive branch) all the way to the present day.

Similarly, in Mistretta v. United States, the Supreme Court had no trouble applying the non-delegation doctrine to a delegation of power to the Sentencing Commission-"an independent commission in the judicial branch of the United States"-to write Sentencing Guidelines. The Court cited the "intelligible principle" test and all the classic (mostly executive) non-delegation cases, with no intimation that the test depended on the identity of the delegate, and upheld the delegation because it found plenty of intelligible principle in the statute.

Wayman and Mistretta are about judicial rulemaking, not the more familiar judicial activity of adjudication. The Court also applied the non-delegation doctrine to judicial adjudication in Standard Oil v. United States, where it rebuffed a challenge to the Sherman Act. The Court wrote:

So far as the arguments proceed upon the conception that, in view of the generality of the statute, it is not susceptible of being enforced by the courts because it cannot be carried out without a judicial exertion of legislative power, they are clearly unsound. The statute certainly generically enumerates the character of acts which it prohibits and the wrong which it was intended to prevent. The propositions therefore insist that, consistently with the fundamental principles of due process of law, it never can be left to the judiciary to decide whether, in a given case, particular acts come within a generic statutory provision. But to reduce the propositions, however, to this, their final meaning, makes it clear that in substance they deny the existence of essential legislative authority, and challenge the right of the judiciary to perform duties which that department of the government has exerted from the beginning.

Here, the Court mentions the prohibition on judges exerting legislative power, which is the definition of the non-delegation doctrine. And it rejects the non-delegation challenge in the usual way, which is to explain that the subject matter of the statute is adequately specified-rather than explain that the non-delegation doctrine doesn't apply to judicial adjudication. (This was before the "intelligible principle" formulation had been developed, but recall the language in Wayman v. Southard about the power to "fill up the details.")

Thus, we have Wayman and Mistretta to show that the non-delegation doctrine applies to courts in their rulemaking capacities, and Standard Oil to show that it applies to courts in their adjudicative capacities. This makes sense, since antitrust and various other broad statutory schemes are often referred to as delegations to the judiciary.


So far, delegations to the judicial branch seem to follow the same "intelligible principle" doctrine as delegations to anyone else. The non-delegation cases cite each other without caring who the delegate was in any particular case; nowhere is there any intimation that a different standard was involved for courts than for executive officials.

Still, do delegations to the executive and to the judiciary have to be judged by the same standard? A few commentators suggest that delegations to the judiciary are less problematic than delegations to the executive: Schoenbrod, for instance, suggests that this is because judges are more insulated from the political process and so pose a lesser danger to liberty than do agencies.

On the other hand, far from exempting judicial delegations from the non-delegation doctrine, we might want to be, if anything, more concerned about them than we are about delegations to the executive branch.

In recent years, some scholars have argued that delegations to the executive branch aren't truly problematic-mainly because of the executive branch's dual advantages of expertise and (through the President) accountability. And the Supreme Court has cited both of these advantages in support of the implicit delegation rationale of the Chevron rule.

Federal judges, by contrast, have neither (as generalists) much subject-matter expertise nor (as life appointees) much political accountability. Courts, unlike agencies, encounter issues in limited contexts involving specific disputes-often preventing them from taking as broad a policy-based view as agencies can (even if agencies, too, may sometimes choose to proceed by adjudication). Achieving policy uniformity among courts is harder than doing the same within an agency. Courts are, in some ways, less flexible in the actions they can take and less accessible to the public-with no oversight hearings or notice-and-comment proceedings. And while agencies' exercise of broad discretion is at least subject to judicial review, there is no other branch to police the delegation when courts themselves are the delegates.

It's true that judges acquire some subject-matter expertise by dealing with particular statutes over and over and can use the advice of experts, and some areas of judicial delegation are procedural issues that judges might understand better than Congress. But other areas deal with contested value judgments that courts are no better suited to resolve than legislatures or agencies.

And while political scientists point out that "the Supreme Court follows the election returns," judges on average seem less responsive than actual elected officials.

Judicial Non-Delegation as a Canon?

It has been suggested that the norm against delegation of legislative power is not loose so much as underenforced. The question of when delegation crosses the line into the forbidden delegation of legislative power is a difficult question of degree, and thus "not . . . readily enforceable by the courts." As Scalia wrote in his Mistretta dissent:

Since Congress is no less endowed with common sense than we are, and better equipped to inform itself of the "necessities" of government; and since the factors bearing upon those necessities are both multifarious and (in the nonpartisan sense) highly political . . . it is small wonder that we have almost never felt qualified to second-guess Congress regarding the permissible degree of policy judgment that can be left to those executing or applying the law.

But courts might be better able to determine when a congressional delegation to themselves is excessive and not justified by the necessities of the situation. So when the delegate is the judiciary, strong judicial enforcement of the rule against delegation of legislative power might be able to more closely approach the true constitutional norm.

Indeed, just as Cass Sunstein has argued that the executive non-delegation doctrine is not dead but merely operating in hiding through interpretive canons and clear-statement rules, perhaps the judicial non-delegation doctrine is already in operation.

In the context of delegation to the executive branch, if the scope of the delegation is ambiguous, courts occasionally use an avoidance canon and interpret the statute narrowly to sidestep the potential non-delegation problem. Sometimes courts act even more subtly and implicitly to avoid non-delegation problems-perhaps in doctrines like the Chenery I rule of administrative law that agency action stands or falls based on an agency's stated rationale; or in the modern-day resistance to Chevron maximalism, reflecting unease with giving agencies the power to authoritatively interpret ambiguous statutes.

Similarly, if a court finds that a statute potentially delegates too much power to the judiciary, it can unilaterally interpret the statute narrowly to avoid the problem. Consider the Alien Tort Statute, which grants federal jurisdiction over actions by aliens for torts that violate the "law of nations." In Sosa v. Alvarez-Machain, the Supreme Court decided to interpret "law of nations" narrowly to preclude recognizing offenses beyond the historical ones of "violation of safe conducts, infringement of the rights of ambassadors, and piracy." It did so in part because it wanted to "look for legislative guidance before exercising innovative authority over substantive law"-a concern that, at least on its face, looks like delegation avoidance.

Likewise, consider Armstrong v. Exceptional Child Center, Inc., where healthcare providers sued Idaho officials for setting insufficient reimbursement rates for the services they provided, in violation of § 30(a) of the Medicaid Act. The Supreme Court held, as a matter of statutory construction, that Congress didn't want to allow the providers to sue the Idaho plaintiffs in federal court for equitable relief. This was in part because the statute already provided a single remedy for state non-compliance-withholding of federal funds-and in part because the adequate-funding provision was "judicially unadministrable." The Court wrote, in language that smacks of delegation avoidance based on a distaste for free-floating judicial policymaking:

It is difficult to imagine a requirement broader and less specific than § 30(A)'s mandate that state plans provide for payments that are "consistent with efficiency, economy, and quality of care," all the while "safeguard[ing] against unnecessary utilization of . . . care and services." Explicitly conferring enforcement of this judgment-laden standard upon the Secretary alone establishes, we think, that Congress "wanted to make the agency remedy that it provided exclusive," thereby achieving "the expertise, uniformity, widespread consultation, and resulting administrative guidance that can accompany agency decisionmaking," and avoiding "the comparative risk of inconsistent interpretations and misincentives that can arise out of an occasional inappropriate application of the statute in a private action."

Also, the Supreme Court has recognized that courts aren't ideal at making antitrust policy, which is why it's often proceeded by making per se rules, trading off abstract optimality for administrability. And invalidating criminal laws under the Due Process Clause because they're void for vagueness also has a delegation-avoiding effect, even if the theory is rights-based (and thus applies to state criminal laws as well, which aren't subject to the non-delegation doctrine).

None of these examples explicitly grounds delegation avoidance in constitutional separation-of-powers considerations-it could just be courts' policy judgment about their own limitations relative to Congress (or, in the case of the void-for-vagueness doctrine, some other constitutional doctrine). Clearly there are many reasons to interpret judicial power narrowly-as I argue below, the appropriate scope of federal common-lawmaking is grounded in different concerns than non-delegation. Still, the effect of such narrow construction is delegation avoidance, and I can't rule out that it's partly motivated, below the surface, with some discomfort with broad delegations to the judiciary.

Delegation avoidance can go further than merely interpreting "narrowly": one can choose one's overarching theory of statutory interpretation based on these same avoidance concerns. Perhaps, if one interprets the Sherman Act purposively, as a free-floating license to make sensible antitrust policy, the Act might be an excessive delegation-but maybe not if one chooses textualism and interprets the Act as incorporating the preexisting law on unreasonable restraints of trade (though one still has to decide which strand of the common law to choose!), or if one focuses specifically on the evils that Congress was trying to eliminate at the time.

Judges who choose a theory that relies heavily on the state of the law at the time of enactment (i.e., static rather than dynamic) may be doing so to avoid delegation problems-and indeed, proponents of textualism often justify their choice as motivated, at least in part, by the need to limit judicial discretion.

So perhaps there's a lot of implicit delegation avoidance going on when courts confront possibly excessive delegations to themselves.

Is this valid? When an agency confronts an excessive delegation, it can't fix the problem by adopting a narrowing construction: the fix itself would be a forbidden exercise of legislative power! In such cases of delegation to agencies, the power to adopt a narrowing construction belongs only to courts. Perhaps this is just because, when the delegation runs to the agency, courts are the ones who can narrow the statute without exceeding their power. And this suggests that courts can't use narrowing constructions to save delegations to themselves from unconstitutionality. They could only invalidate the delegation entirely.

But perhaps this is too extreme: not all unconstitutional statutes have to be struck down in their entirety. If the unconstitutional part is severable, a court can just strike down part and leave the rest standing-and the result might look like adopting a narrowing construction. Still, it would be good for courts to be more explicit about the process and actually conduct the severability analysis, because not all narrowing is necessarily reducible to severability.