An Administrative Fairy Tale

Most everyone is familiar with the Hans Christian Andersen tale in which only a child is willing to pronounce what everyone knows: The emperor’s clothes are no clothes at all. As the emperor marches through town, he is as naked as the day he was born. His magnificent new outfit is a fiction, but a fiction that all the adults accept lest they be condemned as traitors or fools. From the mouth of an innocent child comes the honest truth.

Ilan Wurman believes an equivalent unspoken truth about the administrative state must be laid bare. Although the Supreme Court long ago declared that “Congress cannot delegate legislative power to the President”[1], such delegations are commonplace. Indeed, the modern administrative state rests upon the wholesale delegation of vast legislative (and judicial) power to agencies within the executive branch.

The fiction to which Wurman objects is the insistence that no such delegation actually occurs. Existing doctrine maintains that it would be unconstitutional to delegate true legislative authority. Yet so long as Congress provides an “intelligible principle” to guide an agency’s exercise of delegated authority, what is delegated is not truly legislative power at all. In this way, the demands of the modern administrative state are reconciled with the Constitution’s command that “all legislative powers” be vested in the Congress.

Wurman’s Liberty Forum essay argues that if this reality were laid bare—were the pretense of non-delegation dispelled like the pretense of the emperor’s fine clothes—it would be easier to restrain the administrative power. In a sense, the truth will set us free. A more candid doctrine, he suggests, is the first step toward restoring the constitutional separation of powers and mitigating the harms to constitutional republican government wrought by the modern administrative state.

This is not the first call to recognize the reality of legislative delegation. Justice John Paul Stevens, writing separately in Whitman v. American Trucking Associations (2001), called for “frankly acknowledging” that federal statutes delegate “legislative” authority to federal agencies. According to Justice Stevens, “it would be both wiser and more faithful to what we have actually done in delegation cases to admit that agency rulemaking authority is ‘legislative power.’”[2] Justice Stevens’ call for candor was just that. Nothing in his opinion (or the rest of his jurisprudence) suggests that such a revelation would alter the underlying operation of modern government.

Wurman, on the other hand, maintains that a more candid judicial doctrine would effectively overrule other Supreme Court decisions (such as that reached in 1983 in Immigration and Naturalization Service v. Chadha, discussed below) and thereby facilitate other needed reforms. Perhaps; but he offers little beyond ipse dixit to suggest why this is so. Expressly accepting the reality of broad delegations of legislative power is not self-evidently necessary or sufficient for much of what he proposes. Calling the delegation of legislative power by name may be more satisfying and sincere, but it is not the sort of thing from which legal revolutions are made.

It is certainly true that delegation lies at the heart of the modern administrative state. Ever since the New Deal, the Supreme Court has green-lighted ever larger delegations of legislative authority. Under existing doctrine, Congress has virtually unlimited discretion to delegate policymaking authority to federal agencies, and uses this authority regularly. Through such delegation, Congress evades accountability by shifting responsibility for the resulting policies to administrative agencies, undermining political control of regulatory policy.

The Supreme Court allows this state of affairs to persist not so much because the justices refuse to recognize the constitutional problem, but because it is not entirely clear (at least to them) what they could do about it. As Justice Antonin Scalia explained in Mistretta v. United States (1989), “while the doctrine of unconstitutional delegation is unquestionably a fundamental element of our constitutional system, it is not an element readily enforceable by the courts.”[3] The dividing line between permissible discretion and unconstitutional delegation is not readily amenable to judicial resolution, and so the courts leave it to the political branches (and the political process) to police the contours of legislative power.

Unlike others, Wurman does not contest Justice Scalia’s formulation. He does not propose a more meaningful non-delegation doctrine, such as has been proposed by Gary Lawson or David Schoenbrod. Rather than restrain such delegations, he wishes to come to terms with them. Just as an alcoholic must admit to the problem as the first step toward recovery, we must, Wurman suggests, acknowledge the pervasiveness of unconstitutional delegation before we can do anything about it.

As with alcoholism, there is only treatment, not a cure. Wurman wants us to come to terms with our addiction to legislative delegation; he does not call for the courts to force Congress to sober up. In this sense, his approach may be practical—more incremental than revolutionary. Yet it relies upon a doctrinal reversal for which no sitting justice has called, and for which there appears to be no real demand, least of all among those who are concerned about the state of administrative law.

Wurman believes that if we acknowledged the reality of constitutional delegation, this would automatically empower Congress to discipline regulatory agencies in heretofore forbidden ways, such as through unicameral legislative vetoes of agency actions. It would, in other words, overturn INS v. Chadha, in which the Supreme Court held that such vetoes were unconstitutional.[4] Yet why he believes this to be the case is anything but clear.

The argument seems to be that if one accepts that Congress is delegating legislative power, then Congress must necessarily have the authority to reconfigure how such power is delegated. But this does not necessarily follow. The power to withhold a delegation is not the same as the power to slice it, dice it, and reassemble it in accordance with legislative whim.

Whether agency power is deemed legislative or not, it still comes from Congress, and the Constitution prescribes how Congress may act. The majority in Chadha accepted that the authority in question had been delegated by Congress, but still rejected the proposition that this enabled Congress to subject administrative action to veto by a single house of Congress acting alone.

The Court rejected the unicameral veto because such action is best understood as a legislative act, whether or not the delegated power at issue is also best characterized as legislative in nature. A legislative veto of an administrative determination or rulemaking alters the legal rules to which private actors can expect to be subjected. As such, it is a legislative act for which bicameralism and presentment are required under the Constitution.

In Chadha, a single house of Congress (the House of Representatives) sought to alter what the law required (whether an immigrant would be subject to removal). That is, the veto purported to have the same legal effect as would a legislative act. It was, in all essential respects, an exercise of legislative power and, as the Constitution makes clear, Congress may act legislatively only when it follows the constitutionally enumerated process. And if this were true in the case of whether the removal of one alien could be suspended, it is unquestionably true of most regulatory actions. How one characterizes the initial delegation does not change this fact.

Authorizing unicameral legislative vetoes is not all Wurman believes greater doctrinal candor could accomplish. He also believes it would enable Congress to exercise direct control over what we now call “independent” agencies, at least with regard to their legislative-like functions, and would enable Congress to subdivide agency functions. Yet the former is not a necessary implication of Wurman’s proposal, and the latter could be accomplished without any doctrinal change at all—Congress just has to enact it. It is also not clear why either reform is particularly desirable.

Greater (and more searching) judicial review of agency adjudications, on the other hand, would be most welcome, but here again Wurman fails to establish how recognizing the nature of delegation is either necessary or sufficient to achieve this goal. If Congress wanted de novo review of agency adjudications, it could provide for that now simply by revising the Administrative Procedure Act. A reformulation of the delegation doctrine would hardly be necessary.

A key selling point for Wurman’s proposal is that it is supposed to facilitate greater legislative control of regulatory agencies. Even if this were true—and I have already detailed some of my doubts—why does Wurman believe that majorities in Congress have any interest in exercising such control? After all, with the exercise of such power comes accountability, and accountability seems to be the one thing Congress seeks to avoid. Indeed, this is why there is so much delegation in the first place.

Congress has ample ability within the constraints of existing doctrine to limit federal agency overreach, and has largely neglected to take substantive steps in this direction. Congress may lack the authority to enact a legislative veto, but it could limit agency regulatory authority, such as through the Regulations from the Executive in Need of Scrutiny (REINS) Act (which would replicate the functional operation of unicameral vetoes by requiring congressional approval of major regulations), or by requiring agencies to submit specific types of rules to Congress for formal approval.[5] If these sorts of reforms have no political future—and Wurman says they do not—why would Congress show any greater interest in the additional reforms his proposal is supposed to allow? (And, perhaps more fundamentally, why do we think a single justice, let alone a majority of the Court, is interested in reformulating existing doctrine as Wurman proposes?)

Express recognition of the reality of delegation does not seem necessary to restrain the administrative state. It may not even be particularly helpful. To explicitly admit that Congress delegates legislative power, while allowing such delegations to continue, would effectively validate them constitutionally. It would amount to a concession that delegation is not only necessary, but perfectly acceptable.

Non-delegation concerns permeate much of administrative law. While the Supreme Court is not willing to enforce a meaningful non-delegation doctrine, the principle of non-delegation informs other inquiries. The shadow of the non-delegation doctrine arguably induces the Court to pay greater attention to the scope and nature of legislative delegations. Such concerns find expression in what Cass Sunstein termed “non-delegation canons”—interpretive canons that serve to constrain the scope of delegated agency authority and curb the administrative state, even if only at the margins.[6]

A prominent example of such a non-delegation canon at work is the so-called “major questions” doctrine, under which courts presume that Congress has not delegated questions of “deep economic and political significance” to regulatory agencies.[7] Whereas the Chevron doctrine instructs courts to defer to agency interpretations of ambiguous statutory provisions, the “major questions” doctrine takes weighty and substantial matters out of agency hands. This is because Chevron deference is premised on a theory of delegation: that Congress has delegated to the agency the authority to resolve the ambiguity at hand. Where much is at stake, however, the Court has held “there may be reason to hesitate before concluding that Congress has intended such an implicit delegation.”[8] As the Court has explained, if Congress had intended to delegate such authority to an agency, it would have done so explicitly.

The “major questions” doctrine rests on the assumption that there is something troubling about unmoored delegations. Because of this, implicit delegations are to be avoided; and the greater the delegation, the more explicit Congress must be. Congress may still delegate as much as it wants, but the Court will not lend a helping hand. Yet if pure legislative delegations are to be accepted, as Wurman recommends, a key reason for this doctrine would be removed.

Delegation concerns only restrain agencies on the margins, to be sure, and the Supreme Court has not consistently enforced even this modest brake on agency action. Wurman’s proposal would not fix this. To the contrary, it could induce the Court to abandon the enterprise altogether. Insofar as non-delegation concerns underlie other doctrines and influence judicial decisions in subtle but important ways, abandoning the fiction that Congress does not delegate true legislative power could further loose the administrative state in unintended ways.

In Andersen’s famous story, the truth-sayer breaks a taboo. The benefit is that this “speaking truth to power” allows others to speak. But eliminating the false pretense of the emperor’s procession was one thing—recall that it did nothing to loosen the emperor’s hold on power.

“But he hasn’t got anything on!” the whole town cried out at last. The Emperor shivered, for he suspected they were right. But he thought, “This procession has got to go on.” So he walked more proudly than ever, as his noblemen held high the train that wasn’t there at all.[9]

[1] Marshall Field and Co. v. Clark, 143 U.S. 649, 692 (1892).

[2] Whitman v. American Trucking Associations, 531 U.S. 457, 488 (2001) (Stevens, J., concurring-in-part and concurring-in-the-judgment).

[3] Mistretta v. United States, 488 U.S. 361, 415 (1989) (Scalia, J., concurring).

[4] 462 U.S. 919 (1983).

[5] On the REINS Act, see Jonathan H. Adler, “Placing ‘Reins’ on Regulations: Assessing the Proposed REINS Act,” NYU Journal of Legislation and Public Policy 16:1 (2013), available at http://www.nyujlpp.org/wp-content/uploads/2013/04/Adler16.1.pdf.

[6] Cass R. Sunstein, “Nondelegation Canons,” University of Chicago Law Review 67 (2000), 315-37.

[7] King v. Burwell, 135 S. Ct. 2480 (2015).

[8] FDA v. Brown and Williamson Tobacco Corp., 529 U. S. 120, 159 (2000).

[9] Quotation is from the Hans Christian Andersen Centre’s version of the tale, translated by Jean Hersholt, http://www.andersen.sdu.dk/vaerk/hersholt/TheEmperorsNewClothes_e.html.