Regulation
The Cato Review of Business & Government


Delegation Running Riot

Power Without Responsibility:
How Congress Abuses the People through Delegation
by David Schoenbrod
(Yale University Press, 1993), 260 pp.

Reviewed by Douglas H. Ginsburg

 

After a single preambulatory sentence, the Constitution of the United States begins with this simple proposition: "All legislative powers herein granted shall be vested in a Congress of the United States." What does it mean? According to David Schoenbrod, a professor of law at New York Law School and previously a litigator for the National Resources Defense Council, it means that Congress cannot delegate to any other body the power to make law, by which he means the power to prescribe a rule of conduct binding upon a private citizen or a state or local government. This leaves it open to Congress to delegate the power to make rules binding upon the federal bureaucracy, but little else. If he is correct, then most of the regulatory state was brought into being unconstitutionally; and although it could be legitimated if Congress were itself to adopt the Code of Federal Regulations into law, it is Schoenbrod's confident expectation that Congress would not codify much of what its delegates in the bureaucracy have done. Whatever the outcome, however, if agency-made laws were put to a vote, our democracy would be the stronger for the debate, and Congress would be politically accountable for the results.

Schoenbrod's reading of the first clause of our Constitution has long, but not strong, bloodlines. As he acknowledges, while the nondelegation doctrine seems to have been the unarticulated premise of a few Supreme Court decisions rendered in the 19th century, in none of them did the Court actually find an unconstitutional delegation. In United States v. L. Cohen Grocery Co. (1921), when the Court did strike down a statute that made it a crime to charge "unjust or unreasonable" prices for "any necessaries," it did so on the somewhat narrower ground that "Congress alone has power to define crimes against the United States. This power cannot be delegated either to the courts or to the juries of this country." In the next case clearly on point, J.W. Hampton, Jr. & Co. v. United States (1928), the Court upheld Congress's delegation to the president of the authority to impose tariffs to the extent necessary to offset differences in the cost of production between domestic and foreign producers. The Court thought it enough that Congress had stated an "intelligible principle" to guide, and presumably to confine, the discretion of the president in accord with the intent of the legislature.

The strong form of the nondelegation doctrine reappeared in two 1935 cases challenging the National Industrial Recovery Act (NIRA), A.L.A. Schechter Poultry Corp. v. United States and Panama Refining Co. v. Ryan. In the leading case, Schechter Poultry, the Court struck down the NIRA, an exercise in corporatism that involved tripartite boards prescribing so-called codes of fair practice binding upon industry. In a memorable concurrence, Justice Benjamin Cardozo characterized this scheme as "delegation running riot."

After Schechter Poultry, however, the Court retreated to the "intelligible principle" line it bad drawn earlier and upheld a series of industry-specific New Deal regulatory schemes delegating wholesale lawmaking authority to administrative agencies. In the end, the nondelegation doctrine was drained of all substance: Congress could delegate open-ended authority to regulate the private sector "in the public interest" without fear of judicial interference. (See, e.g., United States v. Rock Royal Co-op, Inc. (1939)) Indeed, Congress has delegated to the president comprehensive emergency authority to regulate wages and prices throughout the economy-authority that President Nixon used in 1971-constrained only by the "intelligible principle" that it is to be used to combat various economic problems, such as inflation. (See Amalgamated Meat Cutters and Butcher Workmen of North America, AFL-CIO v. Connally, (D.D.C. 1971))

So for 60 years the nondelegation doctrine has existed only as part of the Constitution-in-exile, along with the doctrines of enumerated powers, unconstitutional conditions, and substantive due process, and their textual cousins, the Necessary and Proper, Contracts, Takings, and Commerce Clauses. The memory of these ancient exiles, banished for standing in opposition to unlimited government, is kept alive by a few scholars who labor on in the hope of a restoration, a second coming of the Constitution of liberty-even if perhaps not in their own lifetimes.

Bad Economic Policy

Schoenbrod's brief-for that is what he has written-for reviving the nondelegation doctrine argues that delegation is not only bad constitutional law but also bad policy. He interweaves throughout his text two case studies, suitably chosen to give the naive reader a sense of the outrages to which federal regulation is prone. One concerns the "marketing orders" by which the secretary of agriculture has since 1933 exercised delegated authority to limit the sale of various commodities-he focuses upon oranges-lest the abundance produced by American farmers result in low prices for consumers. This bootless effort has done little to benefit citrus growers, for the usual reasons when there are rents to be distributed, but it has comfortably supported the management of the Sunkist Growers' Cooperative.

I was first introduced to the glories of orderly marketing in 1984, when I became the head of the Office of Information and Regulatory Affairs (OIRA) at the Office of Management and Budget. OIRA is the office that President Reagan created for the purpose of reviewing agency regulations for consistency with administration policy, principally the seemingly modest policy that the benefits of a regulation should exceed its costs insofar as the authorizing statute allows. My predecessor had had a dust-up with the secretary of agriculture over citrus marketing orders, the only result of which was an appropriations rider prohibiting our office from subjecting any further marketing orders to the discipline of regulatory review. I was aware, therefore, of the major coops' political clout, but I did not fully appreciate their gall until I received in the mail a shrink-wrapped lemon and a note from an insurgent grower. He reported that because shrink-wrapped lemons have a much longer shelf life than unwrapped lemons, the U.S. Department of Agriculture had made it unlawful to market a shrink-wrapped lemon.

Schoenbrod's other case study concerns the Clean Air Act of 1970 and its periodic amendments dealing with industrial and automotive emissions. His analysis of the act and of the difficulties it has both created and encountered is masterful; of course, he was involved intimately in some of the litigation it spawned. His thesis here is that the Clean Air Act of 1970 was "the first instance of narrow delegation." Congress, he maintains, felt the sting of critics who blamed its prior broad delegations to the Department of Health, Education, and Welfare (HEW) for "stymieing efforts to clean up the air." Congress reacted, not by making the hard decisions it had earlier ducked by means of broad delegations, but rather, by enacting detailed instructions to the implementing agency without "facing the conflicts between its twin promises of protecting health and the economy.

Schoenbrod's bill of particulars against the Congress of 1970 and all of its successors that amended the act before 1990 is long, specific, and persuasive. The 1970 Congress pretended that "technology-forcing" would resolve the conflict between achieving the national ambient air quality standards it prescribed and at the same time keeping factories running: "The states were left regulating those sources whose control would spark the fiercest opposition-existing sources." States were required to submit plans to deal with automobile pollution, but when they failed to do so, the act provided that the Environmental Protection Agency (EPA) would prescribe plans for them, although "Members of Congress, in debating the Act, had not spoken of national takeover of state and local land use and transportation policies, and had not provided the administrator political support for such actions." Again, Congress, knowing that there was no threshold "above which the air was risky and below which the air was healthy," but rather a continuum of degrees of risk, simply pretended otherwise and instructed the agency to set standards "allowing an adequate margin of safety [to] protect the public health." This left it to the EPA to reap the whirlwind.

Political Motives

There is much more in Schoenbrod's account of how the Clean Air Act has grown more than ninefold to its present length of 718 pages, not counting the thousands of pages of implementing regulations. Here is his conclusion: "It would be comforting, but wrong, to think that the delay, complexity, and confusion produced by the Clean Air Act is the result of some unforeseen mistake in its design rather than the inevitable consequence of delegation. Delegation inevitably produces these results because an agency is left to make the law and rationalize it in detail, and even then the law is not settled. Narrowing the delegation compounds the problem by making the agency jump through additional hoops. But the chief source of the difficulty is more insidious: Legislators seem to be unconcerned about imposing delay, complexity, and confusion on their constituents when they delegate." Why? Because legislators are themselves essentially an interest group, interested in showing the voters that they have acted in favor of the good and the true and in distancing themselves from the consequences, conflicts, and hardships that inevitably arise in the real world where government affects people. Indeed, by Schoenbrod's account, the legislators stand to gain from, and therefore favor ways of creating, regulatory nightmares for the citizenry; that puts them in a position to do "casework" on behalf of constituents beleaguered by the federal bureaucracy to which the legislators have delegated the hard choices.

Schoenbrod's analysis of legislators' motivation draws upon and extends the work of Harvard's Morris Fiorina. Like Fiorina, he assumes that legislators vote "partly on the basis of how their vote will look to the public rather than whether it helps the public." From the legislators' perspective, a publicly perceived benefit that outweighs a publicly perceived cost is a bargain-regardless of whether the real cost outweighs the real benefit. This model does not depend upon public stupidity in order to explain legislative action, however. Rather, it distinguishes between sophisticated, concentrated interests, which are unlikely to be fooled, and the general public, which is rationally ignorant-because the stakes are small for any one individual-of the details of what their representatives do. Thus, Schoenbrod suggests that Congress set auto emission standards itself because delegation would not have fooled the auto industry about the costs being imposed upon it or who was responsible, whereas the legislators could take credit for their vote with the general public, which would be ignorant of the costs.

Schoenbrod claims that his Clean Air Act example not only conforms to the predictions of Fiorina's model, it suggests additional ways in which legislators benefit from delegation. For example, delegation allows a legislator "simultaneously to support the benefits of an action and oppose its costs, which is political heaven. The 1970 Clean Air Act passed all but unanimously, but some legislators who had been its strongest advocates opposed EPA when it was compelled by the statute to impose costs on their constituents."

In addition, Schoenbrod argues that delegation is not unattractive to presidents, notwithstanding legislators' common practice of blaming presidential appointees when congressional policies are implemented at the expense of constituents. Schoenbrod suggests that because of the long delays built into the regulatory process, hardships will not generally be felt by the public until after one and often two intervening presidential elections, and the president can sign the delegating legislation knowing that he will not have to deal with the problems it will create. Moreover, to the extent that the implementing agency is somewhat independent of the president, he can distance himself from its regulations. Indeed, Schoenbrod points out that "even incumbent presidents try to 'run against the government,"' referring here to President Bush's declaration, shortly before the 1992 elections, of a 90-day moratorium on new regulations.

Schoenbrod's point about presidents is not very satisfying. While it is true that there is often a long lag between the delegation of lawmaking authority and the bite of regulation-especially environmental regulation, which is typically aimed at very complex phenomena-this is far from always the case. Increasingly, Congress even specifies that an agency is to issue regulations within 90 or 180 days. Moreover, the 1992 example would seem to cut the other way, in light of the outcome of the election. Indeed, in 1980 candidate Reagan made control of federal regulation one of the four elements of the economic platform on which he defeated an incumbent president. In sum, while one can understand why Congress would want to toss hot potatoes to the president, Schoenbrod has not explained why presidents seem willing to catch them.

Simple Solutions?

But to return to Schoenbrod's main concern with delegation: what is to be done? The author's breathtakingly simple conclusion is that the Supreme Court should declare unconstitutional any statute in which Congress delegates power to another body, whether agency or court, "to decide what conduct [is] prohibited." Anticipating his critics, Schoenbrod acknowledges that statutes will always require interpretation-both because texts are ambiguous and because they must be read in a context that is not self-defining-but maintains that, even so, not every statute delegates lawmaking power. The distinction comes down to this: "A law interpreter looks backward to what a past legislature thought, while a lawmaker looks forward to how a proposed law would affect society."

Unfortunately, things are not quite that simple-except in cases too simple actually to be litigated. In most real cases, the court or agency interpreting a statute cannot literally determine "what a past legislature thought" because the legislature did not address the precise issue in dispute. The court or agency can at best say with some degree of confidence what it thinks the legislature would have wanted had it considered the question-based upon the purpose of the legislation and upon the clues that can be gleaned from what the legislation says about closely related questions. That exercise will often, however, require the court or agency to look forward to how a [possible interpretation] would affect society"-in order to test that interpretation for consistency with the legislature's purpose-and that is the very thing Schoenbrod identifies as the hallmark of lawmaking.

Thus, Schoenbrod errs in thinking that he has answered the objection of Justice Scalia, who in dissenting from Mistretta v. United States (1989)-in which the Supreme Court upheld the constitutionality of the U.S. Sentencing Commission-reasoned as follows: (1) because no statute can be entirely precise, and . . . even some judgments involving policy considerations must be left to the officers executing the law and to the judges applying it, the debate over unconstitutional delegation becomes a debate not over a point of principle but over a question of degree"; (2) the degree of permissible delegation (here quoting Chief Justice Taft in the 1928 case) "must be fixed according to common sense and the inherent necessities of governmental coordination"; and (3) that involves "multifarious and (in the nonpartisan sense) highly political" considerations as to which Congress and not the Court has the institutional competence. Among those considerations Justice Scalia included "whether the Nation is at war" or faces some inchoate emergency, here referring to the lower court case upholding wage and price controls in 1971.

While claimed emergencies are indeed matters about which the Court is properly hesitant to second-guess the Congress, no meaningful concept even of necessity, much less emergency, could justify routine delegation of the lawmaking function to agencies and courts if there is a judicially manageable standard for determining excessive or unjustified delegations from those meeting "the inherent necessities of governmental coordination." If there is, Schoenbrod has not articulated it in the few pages he has devoted to the subject. Perhaps interest spurred by Schoenbrod's book will elicit further effort in that direction.

Meanwhile, increased sensitivity to the problems created by wholesale delegation of lawmaking authority, whether of the broad or narrow variety, at the very least invites attention to any other proposals that might mitigate the problem, even if they are not as powerful as the constitutional solution to which Schoenbrod would take us. There have been any number of attempts in the last 20 years or so to rein in the tendency of Congress to pass ever more regulatory laws, facilitated by its ability to delegate the hard choices to others. Thus, we have seen serious proposals for a regulatory budget, analogous to the fiscal budget, designed to constrain the burden that Congress may impose upon the private sector without actually raising taxes and spending public revenues. We have heard about so-called sunset laws that would force a periodic reexamination of regulatory legislation. Each successive president has devised his own form of regulatory review in order to put the stamp of his own administration on the output of the regulatory state. And Congress actually passed nearly 200 laws with a one-house veto provision, pursuant to which either house could disapprove a particular exercise of delegated authority, but all of which were declared unconstitutional by the Supreme Court in INS v. Chadha (1983).

My own favorite solution to the problem of excessive and irresponsible regulation, first proposed by then-Judge Stephen Breyer in a 1984 article in the Georgetown Law Journal, is also highly responsive to Schoenbrod's well-taken concern with delegation. Proposed regulations, or at least those that would impose a burden in excess of a specified amount, say $100 million, would not take effect unless affirmatively approved by both houses of Congress. In other words, Congress's delegation of legislative authority would not be unlimited; to the extent that its agent proposed a particularly expensive regulation, it would have to check back with its principal for authority. The proposed regulation would have to be either accepted or rejected, preferably by a roll call vote. This would restore political accountability for major regulations, while leaving smaller matters to the agencies. Only in Washington can a $100 million burden be regarded as insufficient to warrant the attention of real legislators; of course, the lower the threshold, the better.

Under this arrangement, one can readily anticipate that agency heads with even a modest concern for self-preservation would think twice before making expensive regulatory proposals that would subject their congressional overseers to intense scrutiny. When an agency did put forth such a proposal, the administrators presumably would make an effort to explain to the Congress, and hence to the public, how the agency reasoned from its congressional mandate to its proposed solution and what alternatives Congress might want to consider now that it sees the results of its mandate.

Of course, unlike a Supreme Court decision interpreting the Constitution, a proposed change in congressional procedures can succeed only to the extent that Congress can be persuaded to adopt and to stick with it. But the odds on selling regulatory reform to Congress are at this moment a good deal better than the odds on selling the nondelegation doctrine to the Court.


Regulation is published four times a year by the Cato Institute. Editorial and business offices are located at 1000 Massachusetts Avenue, N.W., Washington, D.C., 20001. For subscription information, please write to Circulation Department, Cato Institute, same address, or call (202) 842-0200.

| Regulation | Home | Order Regulation | Publications | Search