The Volokh Conspiracy
Mostly law professors | Sometimes contrarian | Often libertarian | Always independent
On February 27, 2001, the Supreme Court handed down Whitman v. American Trucking Ass'ns, where the Supreme Court considered whether the EPA could validly promulgate National Ambient Air Quality Standards. The challengers had argued that the Clean Air Act violated the non-delegation doctrine by delegating too much power, in terms that were too vague, to the EPA. The Supreme Court upheld the delegation 9-0: the prevailing test is whether Congress has stated an "intelligible principle" to guide the EPA, and given the sorts of broad delegations that had been upheld in the past (including a law requiring an agency to act in the "public interest"!), the delegation in the Clean Air Act easily passed muster.
Justice Thomas joined the main opinion, but he wrote the following in his brief concurrence:
The parties to these cases who briefed the constitutional issue wrangled over constitutional doctrine with barely a nod to the text of the Constitution. Although this Court since 1928 has treated the "intelligible principle" requirement as the only constitutional limit on congressional grants of power to administrative agencies, [citing J.W. Hampton (1928), the case that first used that test,] the Constitution does not speak of "intelligible principles." Rather, it speaks in much simpler terms: "All legislative Powers herein granted shall be vested in a Congress." U. S. Const., Art. 1, § 1 (emphasis added). I am not convinced that the intelligible principle doctrine serves to prevent all cessions of legislative power. I believe that there are cases in which the principle is intelligible and yet the significance of the delegated decision is simply too great for the decision to be called anything other than "legislative."
As it is, none of the parties to these cases has examined the text of the Constitution or asked us to reconsider our precedents on cessions of legislative power. On a future day, however, I would be willing to address the question whether our delegation jurisprudence has strayed too far from our Founders' understanding of separation of powers.
That future day was Monday, the day before yesterday, when the Court handed down Dep't of Transportation v. Ass'n of American Railroads. I wrote about Justice Kennedy's opinion on Monday, and wrote about Justice Alito's concurring opinion on Tuesday. Today, I'll summarize Justice Thomas's opinion (which Eugene also excerpted earlier), where he gives his originalist view of the non-delegation doctrine at length (his opinion, at 27 slip-opinion pages, is longer than the main opinion and Alito's opinion combined).
Traditional non-delegation doctrine as a species of a more general non-delegation idea
First, Justice Thomas sees the traditional non-delegation doctrine as part of a broader separation-of-powers doctrine:
When the Court speaks of Congress improperly delegating power, what it means is Congress' authorizing an entity to exercise power in a manner inconsistent with the Constitution. For example, Congress improperly "delegates" legislative power when it authorizes an entity other than itself to make a determination that requires an exercise of legislative power. [See Whitman v. Am. Trucking Ass'ns.] It also improperly "delegates" legislative power to itself when it authorizes itself to act without bicameralism and presentment. [See, e.g., INS v. Chadha (1983).] And Congress improperly "delegates"-or, more precisely, authorizes the exercise of, [see Thomas's concurrence in the judgment in Monday's other case, Perez v. Mortgage Bankers Ass'n,] (noting that Congress may not "delegate" power it does not possess)-executive power when it authorizes individuals or groups outside of the President's control to perform a function that requires the exercise of that power. [See, e.g., Free Enterprise Fund v. PCAOB].
Conventionally, these doctrines are thought of as somewhat distinct: the non-delegation doctrine guards against Congress giving up too power; Chadha guards against Congress acting legislatively without bicameralism and presentment; and Free Enterprise Fund has to do with presidential control of the administration via appointment and removal. But Justice Thomas conceptualizes these as all part of a common problem of "delegation."
Anyway, under Thomas's view, you have to rigorously establish what sort of act something is—whether it's legislative, executive, or judicial. (The same act can sometimes be of more than one type, so it can be performed by more than one branch.) The bottom line of his opinion is: "The function at issue here is the formulation of generally applicable rules of private conduct. Under the original understanding of the Constitution, that function requires the exercise of legislative power. By corollary, the discretion inherent in executive power does not comprehend the discretion to formulate generally applicable rules of private conduct."
History of non-delegation in England and the Founding
What follows is an extended discussion of the history of the idea of the separation of powers and how it was thought of as being an aspect of the "rule of law." The historical discussion seems deeply indebted to Philip Hamburger's recent book, Is Administrative Law Unlawful? (Hamburger has guest-blogged right here: see this post, this one, and this one. To the question posed in Hamburger's title, Adrian Vermeule answers "No"; see Gary Lawson's review for a favorable take.)
Thus, Justice Thomas discusses Henry VIII and the Act of Proclamations, where Parliament (for a while) delegated to the king a limited power to legislate by proclamation. He next discusses the Case of Proclamations, where Chief Justice Coke repudiated James I's effort to make rules governing private conduct by proclamation. And he quotes Locke and Blackstone arguing that rules governing private conduct should be made by the legislature, not by the king.
These experiences "profoundly influenced" our own Founders. "This devotion to the separation of powers is, in part, what supports our enduring conviction that the Vesting Clauses are exclusive and that the branch in which a power is vested may not give it up or otherwise reallocate it. (After all, the text of the Vesting Clauses alone doesn't establish any non-delegation principle, only a "starting allocation" for all federal power—and Justice Stevens took precisely that "starting allocation" view in American Trucking.) "The Framers were concerned not just with the starting allocation, but with the 'gradual concentration of the several powers in the same department'" (quoting Federalist 51). And—not believing, as Blackstone did, in legislative supremacy—they thought that a delegation of legislative power passed by the legislature (like Henry VIII's Act of Proclamations) should be invalid (citing James Wilson at the Pennsylvania ratifying convention). There are more citations, to the Federalist Papers, Montesquieu, and Locke.
Justice Thomas closes his historical discussion by saying: "This history confirms that the core of the legislative power that the Framers sought to protect from consolidation with the executive is the power to make 'law' in the Blackstonian sense of generally applicable rules of private conduct."
Development of non-delegation doctrine since the early 1800s
Now he turns to the doctrine, as it's evolved since Brig Aurora (1813), through Wayman v. Southard (1825) and Field v. Clark (1892), to the "intelligible principle" doctrine of J.W. Hampton (1928). The early delegations that had been upheld "had taken the form of conditional legislation": Congress had passed the law (e.g., an embargo against France), and it was to go into effect if the President found a fact (e.g., has France violated the neutral commerce against the United States?). In Justice Thomas's view, the practice of conditional legislation, with its "factual determination[s]," "does not seem to call on the President to exercise a core function that demands an exercise of legislative power." This, he says, is like "the type of factual determination on which an enforcement action is conditioned: Neither involves an exercise of policy discretion."
Whoa! Factual determinations don't involve policy discretion??? Well, Justice Thomas grants immediately afterwards that sometimes there are at least implicit policy determinations. (One might object: Always.) When this is the case, there's a "constitutional problem" because the statute would "effectively permit the President to define some or all of the content of that rule of conduct." But when the Court upheld such statutes in Field v. Clark and J.W. Hampton, it did so on the (perhaps mistaken) theory that there was no great discretion involved. The analysis in those cases "may have been premised on an incorrect assessment of the statutes before the Court, but neither purported to define executive power as including the discretion to make generally applicable rules governing private conduct." So it's a mistake to see them as blessing broader delegations—even of the conditional kind—that really do involve executive discretion. (For those who think there's always executive discretion, then perhaps even conditional legislation is generally unconstitutional under Justice Thomas's view.)
Justice Thomas also interprets Wayman v. Southard, discussing a statute giving rulemaking authority to the courts, to hold that courts can't be the recipients of delegated power to make rules governing private conduct—only to make rules governing the conduct of judicial officials themselves. This, Justice Thomas says, is the right reasoning: the power to make rules affecting substantive private rights can't be delegated.
Obviously, the modern "intelligible principle" doctrine can't be squared with Justice Thomas's conception: statutes have been upheld that delegate power to agencies to make all sorts of binding rules, with guidance no more specific than "fair," "not unduly complicated," "public interest," etc. Justice Thomas writes: "Our reluctance to second-guess Congress on the degree of policy judgment is understandable: our mistake lies in assuming that any degree of policy judgment is permissible when it comes to establishing generally applicable rules governing private conduct."
(Note: but if, as I suggested above, any conditional legislation involving factfinding necessarily involves policy discretion, and if this is basically similar to the factfinding that occurs when the executive decides whether to initiate a prosecution, what does that do to the validity of core executive activities like initiating prosecutions? Conversely, if we carve out conditional legislation involving factfinding as being per se O.K., lots of the administrative state might be re-created in this conditional form, with the factfinding involving exceptionally much policy discretion and taking the form of a lot of filling-in of details.)
Anyway: "We should return to the original meaning of the Constitution: The Government may create generally applicable rules of private conduct only through the proper exercise of legislative power." Justice Thomas accepts that this would cramp the federal government's style, but this is a feature, not a bug.
Moving closer to the Amtrak case: regarding private delegations
Now comes the application of these principles to the case at hand. Justice Thomas accepts the D.C. Circuit's conclusion that you can't have delegation to private parties:
Although no provision of the Constitution expressly forbids the exercise of governmental power by a private entity, our so-called "private nondelegation doctrine" flows logically from the three Vesting Clauses. Because a private entity is neither Congress, nor the President or one of his agents, nor the Supreme Court or an inferior court established by Congress, the Vesting Clauses would categorically preclude it from exercising the legislative, executive, or judicial powers of the Federal Government. In short, the "private nondelegation doctrine" is merely one application of the provisions of the Constitution that forbid Congress to allocate power to an ineligible entity, whether governmental or private.
For this reason, a conclusion that Amtrak is private-that is, not part of the Government at all-would necessarily mean that it cannot exercise these three categories of governmental power.
As you'll know if you've followed my writings on this issue (for example, yesterday's Alito post), I don't think there is any principle that you can't delegate to private parties, and the case that the D.C. Circuit cited for the proposition (and which both Justices Alito and Thomas repeat), Carter v. Carter Coal, is properly seen as a Due Process case, not a non-delegation case.
Not that I disagree with the broad principle: I do think that Amtrak might be unconstitutional because (as Justice Alito writes) its president isn't appointed by the proper Appointments Clause process. That's an Appointments Clause issue, which applies to anyone who's classified as an Officer of the United States (which the Amtrak board members and president may well be).
But what's an Officer? Buckley v. Valeo holds that anyone who exercises significant authority under the laws of the United States is an Officer, though an OLC opinion suggests that "Officer" applies only to people who exercise significant authority in a continuous way. Under OLC's view, there would seem to be no Appointments Clause problem with (private) qui tam plaintiffs. Under either view, there would seem to be no Appointments Clause problem with delegating some power to private people who exercise some, but not significant, authority.
But the non-delegation doctrine of Article I, § 1 protects a distinct principle: that whoever the delegate, Congress can't give up so much power that it's legislative. The non-delegation doctrine focuses on how much (or what kind of) power Congress has given up; other clauses, like the Articles II and III Vesting Clauses and the Appointments Clause, focus on who may exercise the power. (Other clauses, like the Due Process Clause at issue in Carter Coal, focus on whether the power granted is exercised fairly.)
(For instance, what about the power to make these factual determinations under conditional legislation? In an elusive case where no discretion is involved, perhaps it might be consistent with the Appointments Clause to delegate this power to a private (non-Officer) person; and it would also be a non-legislative judgment that could be delegated by Congress under the non-delegation doctrine. Perhaps the on-off switch held by private industry in Currin v. Wallace (1939) might qualify… but let's wait a bit to discuss Currin.)
Admittedly, under Justice Thomas's view of what doctrine should be, perhaps he's right that a private delegate would never wield any governmental power—perhaps even temporary officers (contrary to the OLC opinion) are still Officers, so qui tam plaintiffs are unconstitutional, and perhaps even wielding insignificant authority under the laws of the United States is enough to make you an Officer, or at least to require that you be in the Executive Branch somewhere. If that's so, then the three Vesting Clauses together cover the entire range of permissible government power, which would imply that there can be no delegation to a private actor—though I would locate this doctrine in the Article II Vesting Clause, not in the Article I Vesting Clause.
(But if all implementation of federal law, even where one wields insignificant authority, must be done by non-private members of the executive branch, then goodbye federal prison privatization, and privatization of pretty much everything else? It seems like there has to be room for implementation by non-Executive Branch actors; even if qui tam plaintiffs are no good, and even if no federal functions can be privatized, what about just any old private plaintiff implementing federal law by suing with a private right of action?)
Anyway, just note that Justice Thomas is talking about his preferred view, not about current doctrine: this is not something that the D.C. Circuit can take into account on remand. Under current doctrine, not all exercises of authority are significant, they don't all require Officers, and they don't have to all be in the Executive Branch. So the exclusive vesting idea needn't prevent all grants of authority to private parties. The reasoning above thus doesn't establish that the non-delegation doctrine prohibits any private delegations, and so there's room for the non-delegation doctrine to apply to private delegates just as to public ones.
Finally, the Amtrak case itself: regarding public delegations
Let's move past private delegations now, because after all, the Court has just held that Amtrak is public. Justice Thomas's previous analysis says that if rules governing private conduct are involved, they have to be made by Congress. In the Amtrak case, this is legislative power: the metrics and standards that Amtrak participates in formulating "shape the types of contracts that satisfy the common-carrier obligations because § 207 [of PRIIA] provides that 'Amtrak and its host rail carriers shall' include the metrics and standards in their contracts '[t]o the extent practicable.'"
But what about Currin v. Wallace (1939) (and a related case, United States v. Rock Royal Co-operative (1939)), which said that private industry could, consistent with the non-delegation doctrine, hold an on-off switch to determine whether previously determined regulations would go into effect? The Court said there that deciding whether and when regulations go into effect isn't an exercise of legislative power. Justice Thomas's previous analysis clearly implies that these cases are wrong. But Justice Thomas says more than that: "these precedents are directly contrary to our more recent holding that a discretionary 'veto' necessarily involves an exercise of legislative power" (citing INS v. Chadha). Currin and Rock Royal thus "have been discredited and lack any force as precedents."
Now this may be fine under Justice Thomas's view of what doctrine should be, since the industry veto did change the otherwise applicable regulations governing private conduct. (But if we're going to be totally remaking all of doctrine since the New Deal revolution, why worry about whether a 1939 precedent was overruled by a 1983 precedent?) But if this is a statement about the validity of Currin under current doctrine, I think this is going too far.
Chadha did say that a legislative veto was an exercise of legislative power because it was Congress changing the legal status quo. (If Congress hadn't acted, Chadha would have stayed in the U.S.; after Congress acted, Chadha had to go leave the U.S.) But that doesn't mean that the same veto, if wielded by the President under a valid grant of authority, would also have been an exercise of legislative power: the Immigration and Nationality Act authorized the President (through the Attorney General) to decide whether to let Chadha stay in the U.S. even though he was deportable; this was a change in the legal status quo, and this may well have coincided with an inherent Article II power of the President (if you buy the Curtiss-Wright theory that foreign-affairs-like stuff, perhaps including immigration, is a more presidential area).
So the same act—determining whether or not Chadha gets deported—can be legislative when done by Congress and executive when done by the President. I read Chadha as merely saying that when Congress does such an act, it becomes a legislative act (because Congress can only act legislatively, see also Bowsher v. Synar), and must therefore comply with bicameralism and presentment. I don't read Chadha as standing for the broader proposition that any veto must be a legislative power. While Currin may be bad law under Justice Thomas's new and improved view of what doctrine should be, it's not bad law under current doctrine, because Chadha hasn't implicitly overruled it.
(Finally, and least interestingly, Justice Thomas endorses Justice Alito's view of the remaining separation-of-powers issues present in the case.)