Can Congress delegate procedural rulemaking power to courts?

I've recently been blogging about my new article, The Inherent-Powers Corollary: Judicial Non-Delegation and Federal Common Law, which I've posted to SSRN. Here's a link to my initial post; here's a link to my second post, explaining how the non-delegation doctrine applies to delegations to the judiciary and why this makes sense.

In my third post, I discussed my basic thesis, which gives my article its name—the Inherent-Powers Corollary. The non-delegation doctrine is relaxed when Congress delegates to anyone (whether the President, courts, or anyone else) in an area where they already have inherent power. Thus, for instance, a foreign-affairs delegation to the President doesn't need the "intelligible principle" that's required for delegations in other areas. This Corollary also comes with an "Interlinking Curlicue"—actually, all that's necessary is that the delegated power be interlinked with powers that the delegate already has; it's not necessary that the delegated power be something that the delegate could already do. I discussed how the Inherent-Powers Corollary makes good sense, but the Interlinking Curlicue is much more dubious.

Today, I'll start applying this theory to congressional delegations to the judiciary. I'll start out by talking generally about the application of the non-delegation doctrine and the Inherent-Powers Corollary to courts; then I'll launch in on my first example, which is the delegation of procedural rulemaking power. I think that, generally speaking, courts have an inherent power (though concurrent with Congress) to regulate their own procedures, so delegating that power to the courts doesn't require any specificity. But I think the power to repeal contrary statutes is probably unconstitutional.

* * *

The Inherent-Powers Corollary has also been applied to courts, in the context of their quasi-legislative power to write Sentencing Guidelines. The Supreme Court held in Mistretta v. United States that Congress may delegate to the judiciary various "nonadjudicatory functions that do not trench upon the prerogatives of another Branch and that are appropriate to the central mission of the Judiciary." This includes the power, vested in judicial councils, to "make 'all necessary orders for the effective and expeditious administration of the business of the courts.'" Moreover, the Court said:

Though not the subject of constitutional challenge, by established practice we have recognized Congress' power to create the Judicial Conference of the United States, the Rules Advisory Committees that it oversees, and the Administrative Office of the United States Courts whose myriad responsibilities include the administration of the entire probation service. These entities, some of which are comprised of judges, others of judges and nonjudges, still others of nonjudges only, do not exercise judicial power in the constitutional sense of deciding cases and controversies, but they share the common purpose of providing for the fair and efficient fulfillment of responsibilities that are properly the province of the Judiciary.

Thus, although the judicial power of the United States is limited by express provision of Article III to "Cases" and "Controversies," we have never held, and have clearly disavowed in practice, that the Constitution prohibits Congress from assigning to courts or auxiliary bodies within the Judicial Branch administrative or rulemaking duties that, in the words of Chief Justice Marshall, are "necessary and proper . . . for carrying into execution all the judgments which the judicial department has power to pronounce." Because of their close relation to the central mission of the Judicial Branch, such extrajudicial activities are consonant with the integrity of the Branch and are not more appropriate for another Branch.

In Mistretta, the non-adjudicatory function closely related to the central mission of the judiciary was the development of sentencing guidelines-not much different, in the Court's view, from the development of rules of procedure binding on other courts, especially if one rejects the idea that the substantive/procedural dichotomy is relevant for the separation-of-powers inquiry.

Taking the Inherent-Powers Corollary as given, the fact that the judiciary is accustomed to dealing with vague tests like "reasonableness" and virtually unconstrained common law-style adjudication can be perfectly consistent with the non-delegation doctrine. Just as with executive agencies, the non-delegation doctrine applies with extra leeway in areas where judges have independent authority, or in areas interlinked with those. It turns out there are many such areas.

The task is straightforward: just look for areas where judges have power independent of statute. Then, delegation into (or near) those areas is acceptable even with no (or few) standards. We know (from Mistretta and history) that judges have a certain power to make their own procedures. And judges obviously have adjudicative powers; in the context of adjudication, looking for judges' independent power requires asking about the proper scope of federal common law.

What is federal common law-and didn't we learn in Erie that there's no such thing? In fact, there is such a thing, but it's perhaps easier to know when you see it than to define. Martha Field defines it broadly, as "any rule of federal law created by a court (usually but not invariably a federal court)-when the substance of that rule is not clearly suggested by federal enactments-constitutional or congressional." Thomas Merrill likewise defines it broadly, as "any federal rule of decision that is not mandated on the face of some authoritative federal text-whether or not that rule can be described as the product of 'interpretation' in either a conventional or an unconventional sense." Others, like Alfred Hill, define federal common law more narrowly, to include lawmaking and exclude interpretation.

I prefer the broader definitions, which encompass everything that courts can do on their own, and in particular include interpretation. In the first place, it's hard to draw a clear line between lawmaking and interpretation; and in the second place, this broad usage is more useful in talking about the Inherent-Powers Corollary as it applies to delegations to the judiciary. (Not that it makes much of a difference whether we call statutory interpretation federal common-lawmaking or something else-all that matters is that courts also have an inherent power to interpret words, however far that extends. Talking about it under the common rubric of "federal common law" is merely convenient here to avoid superfluous categories.)

There is a voluminous literature debating how far federal common-lawmaking power should extend; here, I'll mostly take its current scope as given, while flagging where disagreements over federal common law would make a difference to the non-delegation inquiry. The rest of this Article is devoted to tying particular powers of courts to the legitimacy of particular congressional delegations.

First, I discuss judicial rulemaking-apparently an easy case, though the statute providing for judicial repeal of congressionally enacted procedural statutes seems unconstitutional. Next, I turn to judicial adjudication, beginning with showing how the Inherent-Powers Corollary allows us to reject the non-delegation interpretation of Erie. Within judicial adjudication, I look at the judicial power to make law in areas involving "uniquely federal interests," create statutory defenses and remedies, and-most importantly-interpret statutory terms.

Procedural Rulemaking: An Easy Case?

1. The Power to Make Rules

As the Mistretta blockquote above states, Congress may delegate procedural rulemaking power to courts.

This is as old as the Republic itself: the Judiciary Act of 1789 gave federal courts procedural rulemaking power. In 1825, Wayman v. Southard at least implicitly blessed the judicial rulemaking provided for in the Process Act of 1792. An explicit statement that courts' regulating their own procedure was a proper judicial function came a few days later, in Bank of the United States v. Halstead. And ten years later, Justice Story treated the propriety of delegating judicial rulemaking to the judicial branch as well established by Wayman and Halstead.

Presumably the same analysis would validate the many subsequent statutes reaffirming the Supreme Court's rulemaking power over common law, equity, admiralty, bankruptcy, criminal, and habeas procedure. Some of these exercises of delegated authority subdelegated further: In the 1822 and 1842 equity rules, the Supreme Court even subdelegated rulemaking power still further-to the Circuit Courts and the English High Court of Chancery! Delegating power to the English High Court of Chancery would seem dubious if it were a dynamic delegation of binding power-though at least the 1842 rules made clear that English Chancery practice was merely incorporated as it existed in 1842, and only as persuasive authority.

The 1912 equity rules likewise subdelegated rulemaking power-to district courts, subject to the approval of the relevant circuit judges. This subdelegation was almost totally unconstrained: "With the concurrence of a majority of the circuit judges for the circuit, the district courts may make any other and further rules and regulations for the practice, proceedings and process, mesne and final, in their respective districts, not inconsistent with the rules hereby prescribed, and from time to time alter and amend the same."

The Rules Enabling Act of 1934 gave the Supreme Court the power to make federal rules of civil procedure, and Congress's power to do so-both to regulate procedure and to delegate the procedural rulemaking power to courts-was upheld without deep discussion in Sibbach v. Wilson & Co.

Generally speaking, Congress's procedural rulemaking has been authorized by Congress, though Congress has occasionally fiddled with procedure itself. But it seems clear that such authorization isn't necessary. Courts must have some power of their own over their internal procedures-indeed, if Congress had established federal courts but remained completely silent about their procedure, courts would necessarily have had to adopt procedures of their own.

Courts may thus adopt procedural rules-at least as long as the rules don't "abridge, enlarge, [or] modify the substantive rights of any litigant," and possibly as long as they don't substantially affect the plaintiff's choice of forum. To be sure, Congress generally has concurrent power in this area, so it can authorize the rules that courts would already be making, or partly mandate the content of future rules, or override some past rules, or allocate rulemaking power to particular actors in the judicial branch (such as the Supreme Court). But broad congressional delegations with no intelligible principle would be acceptable here.

This idea applies more broadly than just internal procedural rules. Though federal courts can't create common-law crimes, they do have an inherent power to impose sanctions for bad faith and fraudulent litigation behavior; this power has been justified by the need to "vindicat[e] judicial authority." United States v. Hudson & Goodwin calls this an "implied power[]," one "which cannot be dispensed with in a Court, because [it is] necessary to the exercise of all others." Courts' "supervisory power to formulate and apply proper standards for enforcement of the criminal law in the federal courts" even extends into policing some aspects of grand-jury-related conduct. A standardless delegation into these areas of judicial power would likewise be acceptable.

Finally, when the Supreme Court suggested Federal Rules of Evidence, questions over whether this power fell within the rulemaking power originally granted by the Rules Enabling Act led to the Rules' being adopted as a statute by Congress. It's likely, though, that courts could have adopted rules of evidence if Congress had remained silent-and indeed they did make rules of evidence for a long time, in common-law fashion, without any statutory authorization. Federal Rule of Evidence 501 is therefore valid, as are the later enabling acts for rules of evidence.

1. The Power to Repeal Procedural Statutes

But even in procedural rulemaking-supposedly the easiest case for congressional delegation to courts-there can be unconstitutional delegations. In the Rules Enabling Act, right after giving the Supreme Court the power to prescribe procedural rules, Congress provided that "[a]ll laws in conflict with such rules shall be of no further force or effect after such rules have taken effect." The legislative history of this provision is obscure, but Robert Clinton suggests a rationale for this provision: The Conformity Act of 1872 had required the federal judiciary to comply with state procedure. The Rules Enabling Act aimed to undo this regime and allow the Supreme Court to enact rules of federal procedure. This provision may have been added so that the Conformity Act would be implicitly repealed automatically if and when the new federal rules went into effect. "So viewed, the power to abrogate and alter acts of Congress has long since ceased to serve a useful purpose. Nevertheless, this power has been reenacted in most of the statutes authorizing Supreme Court rulemaking authority since 1934."

Can Congress delegate to the Supreme Court the power to repeal statutes? It seems doubtful that Congress could delegate such a power to, say, the EPA or the IRS-and the unconstitutionality of the Line Item Veto Act suggests that "doubtful" is putting it mildly.

Thus, imagine that you're suing the U.S. government in an admiralty lawsuit in federal court. You serve the U.S. Attorney as required, but you do so fairly late: within the extendable 120-day deadline contemplated by the Federal Rules of Civil Procedure, but contrary to the older statutory rule that, in admiralty cases, service must occur "forthwith." One might think that the statute governs, and yet the Rules Enabling Act says the rule governs. In Henderson v. United States, the Supreme Court didn't need to resolve the question, since Congress had later enacted the rule into law. Clearly, then, the rule controls. But the Court suggested that this wasn't necessary: "As the United States acknowledges . . . , a Rule made law by Congress supersedes conflicting laws no less than a Rule this Court prescribes."

Justice Rehnquist had said, similarly, in dictum in Davis v. United States two decades earlier, that explicit conflicts between a statute and a rule would be governed by the rule and that implicit conflicts would present "a difficult question of repeal by implication."

(In Sibbach v. Wilson & Co., the Supreme Court seemed to implicitly suggest that the repealing power was of doubtful constitutionality, since-some years after the Rules Enabling Act-it wrote that "Congress has undoubted power to regulate the practice and procedure of the federal courts, and may exercise that power by delegating to this or other federal courts authority to make rules not inconsistent with the statutes or Constitution of the United States." But any such implicit doubts seem to have evaporated by the time of Davis and Henderson.)

Even if federal courts have a natural power (mostly concurrent with Congress) to make federal procedural rules, this is a far cry from having a power to repeal statutes. Courts certainly wouldn't have had that power in the absence of the Rules Enabling Act, so the Inherent-Powers Corollary doesn't seem to help; and the power to repeal statutes seems so distant from the judicial power that the Interlinking Curlicue is unlike to help either. A free-floating repeal power thus seems unconstitutional-either because it must be hemmed in by some intelligible principle to pass non-delegation scrutiny, or because Congress simply can't delegate a repeal power at all, intelligible principle or no.

Justices Black and Douglas suggested that it was the latter: the repeal power belongs exclusively to Congress. As they wrote in objecting to the 1963 amendments to the Federal Rules of Civil Procedure, rules substantially affecting litigants' rights are:

in practical effect . . . the equivalent of new legislation which . . . the Constitution requires to be initiated in and enacted by the Congress and approved by the President. The Constitution . . . provides that all laws shall be enacted by the House, the Senate, and the President, not by the mere failure of Congress to reject proposals of an outside agency.

In modern terms, this would be like a violation of INS v. Chadha; giving away statute-repealing power in any way is beyond Congress's power.

The Chadha analogy doesn't quite hold up, though. Many statutes provide a rule that's in force until an agency promulgates a regulation. For instance, a section of RCRA prohibits land disposal of certain hazardous wastes if the EPA fails to promulgate appropriate regulations or make appropriate determinations by a certain date; the EPA can also vary effective dates that the statute says would otherwise apply. Similarly, various statutes provide monetary thresholds that can be adjusted based on an agency's calculation of inflation or percentage change in GNP.

These don't violate Chadha, since they don't really delegate an amending or repealing power; they just incorporate a flexible term to be supplied by an agency (not Congress itself, as in Chadha). That delegation is governed by the non-delegation doctrine, and survives if there's an intelligible principle.

The validity of these statutes doesn't yet establish the validity of the Supreme Court's repeal power. Yes, I can think up an approximately equivalent statute that would "simply" amend every existing procedural statute in the Code to be a mere default rule, which would apply as long as there isn't a contrary Rule. But the Supreme Court's repeal power goes further, since it also applies to later-enacted statutes. My hypothetical equivalent statute would have to not only amend existing statutes to be default rules, but also enact a rule of interpretation for future statutes that don't explicitly exempt themselves.

Now, this isn't that crazy. RFRA implicitly amends every present and future statute to provide for a religious exemption regime. And § 3 of the Defense of Marriage Act (in its day), like every other definitional statute, implicitly amended every present and future statute that referred to state definitions of marriage to exclude gay couples.

But both RFRA and DOMA can be overcome by strong evidence in a future statute that RFRA or DOMA wouldn't apply. If a statute can explicitly exempt itself from RFRA or DOMA, surely the textual, intentional, or purposive evidence can be such that it also implicitly exempts itself (or, which is about the same, implicitly effect a partial repeal of RFRA or DOMA as to that statute), despite the presumption against implicit repeals.

The Supreme Court's repeal power should thus be understood in this slightly restricted way. One might imagine a procedural statute, enacted after 1988 (when the repeal power was last codified) but at the time not in conflict with any Rule. This procedural statute might play a crucial role in its statutory scheme, such that there are extremely strong reasons to interpret it as non-severable. And these reasons might overcome the implicit rule of interpretation that all procedural statutes enacted after 1988 are merely default rules in the absence of a conflicting Rule promulgated by the Supreme Court.

If we thus understand the repeal power as being merely a defeasible presumption as to post-1988 statutes, it might be constitutional; otherwise, it's an unconstitutional delegation to the judiciary of the exclusive congressional power to repeal statutes.

But there's still one final roadblock. Compare this scheme to the one invalidated in Clinton v. City of New York. Congress had allowed the President to cancel particular items of expenditure in the budget-a process that was colloquially called a "line-item veto." The majority's view was that the process, "[i]n both legal and practical effect," allowed the President to amend acts of Congress-an end-run around the finely tuned bicameralism and presentment process.

Modern separation-of-powers doctrine lurches between formalism and functionalism; this was the Supreme Court having a formalist day. But it was formalism done wrong. As Justice Scalia pointed out, this wasn't really an amendment to the constitutional veto process. Rather, it was Congress giving the President power to choose not to spend appropriated funds-a power that Congress had exercised on many occasions and that was valid under the non-delegation doctrine. I think Justice Scalia was right, but he and I are both in dissent.

So given Clinton v. City of New York, even the restricted understand of this statute-as an implicit amendment of earlier statutes plus an implicit presumptive rule of construction for later statutes-seems invalid. The power being implied-the power to, in effect, cancel out the effect of statutes altogether-doesn't seem like something that can be implied, because the overall effect would be, in substance, to give repeal authority to a branch-the judiciary-that doesn't even have any formal role in legislation.