The 1935 Invalidations

The 1935 Invalidations

The Supreme Court has found the required intelligible principle missing in only two statutes in its history, and Whitman v. American Trucking Associations, Inc. says so in terms.¹ According to Whitman, one statute “provided literally no guidance for the exercise of discretion,” and the other “conferred authority to regulate the entire economy on the basis of no more precise a standard than stimulating the economy by assuring ‘fair competition.’”² The first is Panama Refining Co. v. Ryan; the second is A.L.A. Schechter Poultry Corp. v. United States. Both came down in 1935.³

The National Recovery Administration, one of the most visible and controversial of the New Deal’s regulatory agencies, gave the Court its chance to show that the nondelegation doctrine could become a real constraint on administrative power.⁴ Panama Refining was the first time in the Court’s history that it struck down an Act of Congress as an overly broad delegation of legislative power.⁵ By then the Court had already moved past its early formulas, which proved too restrictive for a modern industrial economy, and had begun asking whether the legislature supplied sufficient standards to limit agency discretion.⁶ The 1935 cases applied that question and found the answer missing.

Whitman cites both decisions as precedent while holding that the discretion § 109(b)(1) allows is “well within the outer limits of our nondelegation precedents.”⁷ The Court has not repeated its 1935 result since. The two statutes also failed in different ways. One set no standard at all for when the President should act. The other offered purposes so general, and so much at odds with one another, that they could not limit what the President approved. Whitman supplies a principle that explains why the second statute needed more: the degree of agency discretion that is acceptable varies with the scope of the power conferred, and the NRA codes reached the entire economy.⁸ Panama Refining illustrates the first failure and Schechter the second.

Panama Refining and Hot Oil

Panama Refining Co. v. Ryan (1935) struck down a statute that let the President decide whether and how to prohibit the interstate transportation of “hot oil,” petroleum produced or withdrawn from storage in excess of state-set quotas.⁹ Congress wanted to reduce disruption in an oil industry facing falling demand and a growing supply from newly discovered fields, but the statute gave the President no guidance on the circumstances in which to impose the prohibition.¹⁰ Chief Justice Hughes, writing for an 8–1 majority, held the provision an improper delegation of legislative power because it set forth “no requirement, no definition of circumstances and conditions in which the transportation is to be allowed or prohibited.”¹¹

The limit comes from Article I. The Vesting Clause forbids Congress to hand its “legislative Powers” to anyone else, including agencies within the executive branch.¹² Hughes put the point this way: Congress “manifestly is not permitted to abdicate or to transfer to others the essential legislative functions with which it is thus vested.”¹³ Yet the same opinion left plenty of room for delegation. Legislation must often be adapted to complex conditions and a host of details that Congress cannot handle directly, and the Constitution has never been read to deny Congress the flexibility to lay down policies and establish standards while leaving to selected instrumentalities “the making of subordinate rules within prescribed limits and the determination of facts to which the policy as declared by the Legislature is to apply.”¹⁴ Those two permitted tasks are the early formulas under other names: filling up the details and contingent fact-finding. The hot oil provision fit neither. Justice Gorsuch, describing the decision in his Gundy dissent, makes the point directly. The statute did not call for the executive to “ascertai[n] the existence of facts to which legislation is directed,” and it did not ask the executive to “fill up the details” “within the framework of the policy which the legislature has sufficiently defined.”¹⁵

No finding of fact stood between the statute and the ban, and no declared policy gave the President details to fill. Whether to prohibit shipments at all, and in what circumstances, was the open question, and Congress “left the matter to the President without standard or rule, to be dealt with as he pleased.”¹⁶ The Court looked to the statute to see whether Congress had declared a policy and set up a standard, and it concluded that nowhere did Congress declare or indicate any policy or standard to guide or limit the President.¹⁷ The Gundy dissent compresses the holding into the Court’s own formula: Congress “ha[d] declared no policy, ha[d] established no standard, ha[d] laid down no rule.”¹⁸

A tempting objection says Congress did declare a policy, because the Act opened with a statement of purposes. It did list objectives, among them removing obstructions to commerce and reducing unemployment.¹⁹ The Court treated these diverse, broadly stated objectives as a recitation of the interests Congress faced, set down without any resolution of them.²⁰ A goal limits executive discretion only if it narrows the choices. These goals pointed in several directions and said nothing about whether banning a particular shipment advanced any of them, so the President remained free to ban or allow shipments for any reason the President could tie to the Act. A statement of purposes supplies a policy only when the objectives, taken together, let a reviewer decide whether a given action serves them.

Three things go wrong when a statute supplies no standard. Regulated parties cannot learn in advance what the President may prohibit, because nothing in the statute hints at the conditions that trigger a ban.²¹ A reviewing court has nothing to measure a particular prohibition against, since no declared policy shows whether the President stayed inside it. And the President decides case by case what the law is, so the line between choosing a policy and carrying it out disappears. The Court put the structural stake bluntly: if the statute were held valid, “it would be idle to pretend that anything would be left of limitations upon the power of the Congress to delegate its law-making function.”²² The question, it added, “is not of the intrinsic importance of the particular statute before us, but of the constitutional processes of legislation which are an essential part of our system of government.”²³ The sympathetic purpose of the oil regulation therefore decides nothing. What matters is whether Congress itself supplied the policy.

Statutes that give an executive officer comparable power survive when Congress states the conditions for using it. Mistretta described Panama Refining and Schechter as decisions in which Congress had failed to articulate any policy or standard to confine the discretion of the authorities receiving power, and it distinguished them on the ground that the Sentencing Reform Act does not make crimes of acts never before criminalized.²⁴ In Touby v. United States, the Controlled Substances Act let the Attorney General temporarily schedule a new drug only if necessary to avoid an imminent hazard to public safety, and only after considering the drug’s history and pattern of abuse, the scope and duration of the abuse, and the risk to public health; those limits were specific enough to survive. The Attorney General in Touby, like the President under the hot oil provision, could create binding restrictions by executive action, so what separated the two statutes was whether Congress stated when the power could be used.

The Panama Refining litigation also exposed a procedural defect. The executive regulation at issue had never been officially published, and shortly before the Court heard argument it was discovered to have been accidentally amended out of existence.²⁶ A rule that regulated parties cannot locate is one they cannot obey or challenge, so the publication problem echoes the notice concern behind the standards requirement. One beneficial side effect of the decision was legislation requiring federal agencies to publish the official texts of their regulations in the Federal Register, now codified at 5 U.S.C. § 552(a)(1).²⁷ Procedure returns in Schechter and again in the later discussion of procedures and judicial review.

On an exam, a stem that lets an official prohibit conduct calls first for one question: what did Congress say about when the power may be used? The hot oil provision is the model of a binding prohibition the President could impose or lift at will, with no declared policy and no standard against which to measure the choice. A statute that names the triggering conditions, or declares a policy within which the delegate works out particulars, points toward validity.

Schechter Poultry and the NRA Codes

In A.L.A. Schechter Poultry Corp. v. United States (1935), the Court held that section 3 of the National Industrial Recovery Act unconstitutionally delegated legislative power, because Congress cannot give the President an unfettered discretion to make whatever laws the President thinks may be needed or advisable for the rehabilitation and expansion of trade or industry.²⁹ The Court framed the question as one about the statute’s text: in authorizing “codes of fair competition,” had Congress itself established the standards of legal obligation, or had it tried to transfer that function to others by failing to enact such standards?³⁰ The hot oil provision had given the President no guidance at all. Section 3 came with a declaration of policy, so Schechter shows what happens when a statute has words of purpose and nothing that limits choice.³¹

The prosecution began in Brooklyn, where the Schechter brothers ran wholesale poultry slaughterhouse markets. They bought live birds, most shipped in from other states, slaughtered them locally, and sold them to retail dealers and butchers in the city.³² The Live Poultry Code they were charged with violating had been proposed by the New York Live Poultry Chamber of Commerce and approved by President Roosevelt on April 13, 1934, under section 3. It capped work at forty hours a week, set a minimum wage of fifty cents an hour, and contained a “straight killing” rule that barred customers from choosing individual birds from the coops.³³ Justice Gorsuch’s account in his Gundy dissent adds that the rule often made customer selection a federal crime and that kosher butchers such as the Schechters had a hard time following it.³⁴ A jury convicted on every count of an indictment charging one conspiracy and twenty-nine Code violations.³⁵ Justice Cardozo observed that ten of those counts rested on a rule aimed at selective buying, an established practice that was neither unethical nor oppressive.³⁶

Before reaching the text of section 3, the Court answered two arguments about how to approach the case. The first was that the statute had to be viewed against the grave national crisis Congress faced. The Court replied that extraordinary conditions do not create or enlarge constitutional power.³⁷ The second was that the crisis demanded cooperation from industry, which the codes fostered by letting trade groups initiate them. The Court answered that the codes were codes of laws, binding equally those who assented and those who did not, and that violations were punishable as crimes.³⁸ Both answers keep the inquiry on the statute’s text. If an emergency could supply what a statute omitted, the standards requirement would loosen in exactly the periods when Congress is most tempted to hand an overwhelming problem to the executive.

Turning to section 3, the Court looked for limits on the President’s discretion and found two conditions that did little work. The first required the President to find that the groups proposing a code were truly representative and imposed no inequitable restrictions on membership, but that condition concerned who initiated the new laws and said nothing about what the laws could contain.³⁹ The second required a finding that the code was not designed to promote monopolies or oppress small enterprises, yet that restriction left virtually untouched the wide field of policy envisaged by section 1, where a code’s proponents could “roam at will” and the President could approve or reject their proposals as he saw fit.⁴⁰

The last condition was a finding that the code “will tend to effectuate the policy of this title.” The Court gave it little weight: although the statute called it a finding, it was really only a statement of an opinion about the general effect of a scheme of laws on the promotion of trade or industry.⁴¹ That distinction between finding and opinion carries the argument, and it connects to the contingent fact-finding formula from the early cases. A real finding, such as whether shipments exceeded a state quota, has an answer a court can test against evidence, which is why a statute built on one leaves the policy choice with Congress. An official’s belief that a whole scheme of rules will be good for an industry has no such answer, because it cannot be wrong in any way a court could demonstrate. Requiring that belief gave courts nothing to review and the President nothing to find, so the condition restated the President’s approval instead of limiting it.

That left the declared policy as the only possible source of a standard, and it supplied none. The Secretary of Agriculture’s report recommending approval of the Live Poultry Code listed the objectives the Code would serve: removing obstructions to the flow of commerce, promoting organization of industry for cooperative action among trade groups, eliminating unfair competitive practices, using productive capacity fully, avoiding undue restriction of production, increasing purchasing power, and otherwise rehabilitating industry.⁴² The government argued that all of these aims pointed toward a single goal, the rehabilitation of industry.⁴³ Read as instructions, though, they pull apart. The Act adopted a policy of preventing monopolies while promoting cooperation among trade groups, and a policy of encouraging production while improving wages and labor conditions, and it said nothing about how to weigh these values against each other.⁴⁴

When aims conflict, deciding which one prevails is the legislative decision, and section 3 passed that decision to whoever wrote the code. Here that meant a trade group proposing and the President approving, with no vote in both Houses on the choice and no standard a court could use afterward to ask whether a code had gone too far. Any provision could be defended by pointing to whichever aim it served. The Court asked pointedly whether an effort to delegate lawmaking power to trade groups could be made valid by “such a preface of generalities as to permissible aims as we find in section 1 of title I.”⁴⁵ Nor did Congress borrow a yardstick that already existed. According to the Gundy dissent, Congress referred to no pre-existing body of common law on fair competition, as it arguably had with the Sherman Act, and it announced no rules contingent on executive fact-finding.⁴⁶ The majority drew the same line when it found that section 3 was not confined to unfair practices that already offended existing law and could be condemned by courts without new legislation. Its purpose was to authorize new and controlling prohibitions that the code’s formulators would propose and the President would approve as wise.⁴⁷

Justice Cardozo’s concurrence marks where valid delegation ends and invalid delegation begins. If codes of fair competition were codes eliminating unfair methods of competition ascertained upon inquiry to prevail in one industry or another, he said, directing the President to inquire into those practices and denounce them would be no unlawful delegation, since a like power had long been committed to the Federal Trade Commission with the Court’s approval.⁴⁸ Delegation of that kind is born of necessity, because the country’s industries are too many and diverse for Congress to legislate directly with adequate appreciation of varying conditions, and advice from trade associations leaves the substance of the power unchanged, since the President’s approval is what gives the rules the quality of law.⁴⁹ The other conception of a code reaches whatever rules might be desirable or helpful to an industry’s prosperity, which Cardozo described as a roving commission to inquire into evils and upon discovery correct them.⁵⁰ Under that conception anything Congress could do under the commerce clause for the betterment of business could be done by the President on a trade association’s recommendation by calling it a code, and he called the result “delegation running riot.”⁵¹

A tempting shortcut treats the phrase “fair competition” as simply too vague and stops there. Vocabulary cannot be the dividing line, because the Federal Trade Commission’s power over unfair methods of competition survived under a statute built on the same idea.⁵² The dividing line is whether the statute points to a yardstick that exists apart from the delegate’s own preferences. Practices that prevailing business standards already treat as unfair can be identified in advance, discovered on inquiry, and tested in court. A mandate to adopt whatever rules might help an industry prosper has no yardstick, and the Live Poultry Code showed the consequence by setting up a comprehensive body of rules without reference to standards, ethical or commercial, that could be known or predicted in advance of its adoption.⁵³

Two other features of the scheme drew the Court’s attention and return in the later discussion of factors beyond the standard. The codes were drafted by the groups they regulated, which led the Court to ask whether Congress could make trade associations legislative bodies because such groups are familiar with the problems of their enterprises.⁵⁴ And the statute dispensed with trial-type hearings, notice to interested persons, and judicial review of the kind that accompanied earlier delegations to the Interstate Commerce Commission and the Federal Trade Commission.⁵⁵

The Court also held that the Code’s wage and hour terms, as applied to the Schechters’ local slaughterhouse sales, exceeded the commerce power, because the interstate transactions in poultry had ended when the shipments reached Brooklyn. The law is “not indifferent to considerations of degree,” and activities local in their immediacy do not become interstate and national because of distant repercussions.⁵⁶

Schechter and Panama Refining fail in different ways. The oil provision gave the President no standard on when to act, while section 3 declared purposes that pulled against one another and attached them to the regulation of entire industries. When an exam stem describes an official approving industry-proposed rules, backed by criminal penalties, on a finding that they will serve a list of broad aims, three questions decide the case. Do the declared aims conflict with no ranking among them? Does the required finding have an answer a court could test, or does it only express approval of the scheme? Does the statute borrow a yardstick such as practices already recognized as unfair? Conflicting aims, a finding that is only approval, and no yardstick describe Schechter. A direction to inquire into recognized unfair practices describes the valid side that Cardozo’s concurrence marks.