Concept
The Test in Modern Practice
Nondelegation doctrine
The Test in Modern Practice
Since 1935 the Court has found an intelligible principle in every statute it has reviewed. That year it invalidated two statutes, in Panama Refining and Schechter Poultry, and in the decades since it has upheld Congress’s ability to delegate under broad standards “again without deviation.”1 Whitman v. American Trucking counted the same two: one conferred authority with literally no guidance, and the other authorized regulation of the entire economy on no more precise a standard than “fair competition.”2 The limit is stated firmly, then, and the bar for clearing it is low.
That record follows from what the test asks. It asks whether Congress laid down a principle, not whether it laid down enough detail, and the Court grounds that choice in a practical understanding: in an increasingly complex society, Congress “simply cannot do its job absent an ability to delegate power under broad general directives.”3 A ruling that a statute left too much to an agency would need a measure of how much detail Article I demands, and neither the text nor the cases supply one. The sections below take the subject in two steps. The first collects the broad standards the Court has accepted and explains why it declines to police degree. The second examines two decisions that apply the test: Whitman, which concerns air quality standards, and Mistretta, which concerns a commission in the judicial branch.
On an exam, a stem that attacks a statute as too open-ended should start from the expectation that the delegation will almost certainly stand if the statute names a policy and a recipient and limits what the recipient may do. Reserve the contrary conclusion for a statute that, like those of 1935, leaves the recipient with no workable standard at all.
H3 e majority described the Court’s record since 1935 as an unbroken line of decisions upholding Congress’s ability to delegate under broad standards. H4 The opinion collected the phrases that proved sufficient, and the collection reads as a catalog of open-ended value words. Each phrase below guided an agency in a delegation the Court sustained:
- “Public interest, convenience, or necessity” guided the Federal Communications Commission’s broadcast licensing in National Broadcasting Co. v. United States.4
- Commodity prices that would be “fair and equitable” and would effectuate the purposes of the Emergency Price Control Act of 1942 guided the Price Administrator in Yakus v. United States.5
- “Just and reasonable” rates guided the Federal Power Commission in FPC v. Hope Natural Gas Co.6
- A determination of “excessive profits” was the delegated task in Lichter v. United States.7
- “Unfair methods of competition” limited the Federal Trade Commission in FTC v. Gratz.8
The dissent in INS v. Chadha reached the same conclusion from a similar catalog, adding a bare “public interest” standard and the phrase “public convenience, interest, or necessity,” and it described these formulations as the “intelligible principle” through which agencies have gained enormous control over the nation’s economic affairs.9 Words like these name the values an agency must serve and leave the result to the agency. “Fair and equitable” does not name a price, and “public interest” does not name a licensee. The Court has treated that as enough, because a stated value attached to a named agency and a defined subject answers the question the test asks, even when the principle leaves wide room for judgment.
The same leniency governs a modern delegation to the President. Section 1182(f) of the immigration statute provides that whenever the President finds that the entry of any aliens or class of aliens would be detrimental to the interests of the United States, the President may by proclamation suspend their entry for as long as the President deems necessary, or impose whatever restrictions on entry the President deems appropriate.10 A literal reading of Article I, § 1 would treat that grant as a transfer of lawmaking authority over immigration. The Court has taken a functional approach to the vesting clause instead, so the question is whether the statute supplies an intelligible principle, and the phrase “detrimental to the interests of the United States,” despite its seeming breadth, easily satisfies that very generous standard.11
Breadth of language does not explain why “fair competition” failed in Schechter when “public interest” survives. The dissent in Clinton v. City of New York described the Schechter delegation as a broad standard (“fair competition”) combined with power in private parties to promulgate rules applying it to virtually all of American industry, and it called Panama Refining a special case because the code at issue turned out to contain no legally operative sentence.12 On that account, a phrase as open as “public interest” survives when it governs one agency’s work in one field, while an equally open phrase fails when it is joined to rule-writing by private parties across the economy. Later sections return to the private-recipient and subject-matter factors that this comparison previews.
Several reasons together explain the Court’s refusal to measure how much discretion is too much. Congress lacks the time and expertise to write intricate rules for every field it regulates, so it enlists the coordinate branches to do that work.13 Virtually every statute also involves some delegation, because however carefully a law is drafted, some policy judgments must be left to the officers who execute it and the judges who apply it.14 The Court’s early cases showed the same thing. They described the executive as merely ascertaining the event on which the legislative will was to take effect or filling up details, yet the executive in those cases was exercising a substantial measure of policy judgment.15 Once these facts are conceded, a court cannot treat discretion itself as the defect, and the constitutional question stops asking whether a statute delegates and asks how much it delegates.
A question of degree needs a measuring stick, and neither Article I nor the decided cases supply one. Justice Scalia, who dissented in Mistretta on other grounds but agreed that the delegation challenge failed, observed that the Court has almost never felt qualified to second-guess Congress about how much policy judgment may be left to those executing the law, and he asked what legislated standard could possibly be too vague to survive when the Court has repeatedly upheld a “public interest” standard.16 The practical problem is that a test of degree would have to be objective enough to apply. If it were too subjective, every new regulatory statute would face serious threat, because a party burdened by an agency rule has every reason to argue, with hindsight, that Congress should have spoken in more detail.17 Courts may also have held back because other safeguards keep broad grants of power from becoming instruments of administrative oppression, and later sections of this chapter examine them.18
A challenge that rests only on the breadth or vagueness of a statutory phrase is therefore a weak challenge. The stronger analysis names the standard, identifies the agency and subject it governs, and then looks for the features that sank Schechter: a bare standard paired with rule-writing power in private hands, or a power that reaches the whole economy with no more guidance than a slogan. Without those features the delegation stands.
Whitman v. American Trucking
Section 109(b)(1) of the Clean Air Act directs the Administrator of the Environmental Protection Agency (EPA) to set national ambient air quality standards, meaning uniform national limits on pollutants in outdoor air, at a level “requisite” to protect the public health with an adequate margin of safety.20 In Whitman v. American Trucking Associations, Inc. the Court tested that provision against the intelligible principle requirement and upheld it, which makes the decision the clearest modern illustration of how the test works.21 The dispute began in July 1997, when EPA issued final rules revising the standards for particulate matter and ozone, and American Trucking Associations and other industry and state groups challenged the rules in the D.C. Circuit.22
The court of appeals held that EPA had unlawfully considered factors other than public health. It also agreed with the challengers that § 109(b)(1), as the Administrator interpreted it, supplied no intelligible principle and so delegated legislative power in violation of Article I, § 1.23 Yet the court did not declare the section unconstitutional. It reasoned that EPA might avoid the problem by adopting a restrictive construction, so it remanded the standards to the agency, over Judge Tatel’s dissent that the statute was constitutional as written.24 That remand is the move an examinee must recognize as wrong.
The Supreme Court began with Article I, § 1, which vests “[a]ll legislative Powers herein granted” in Congress and so permits no delegation of those powers. When Congress gives decisionmaking authority to an agency, Congress itself must lay down the intelligible principle the agency is to follow, and the Court has never suggested that an agency can cure an unlawful delegation by adopting a limiting construction of its own.25 The reason lies in what makes a standardless statute defective. Congress left the standard unwritten, and an agency that fills the gap by choosing how much of its power to exercise has written the standard itself. Writing the standard is the legislative act that Article I withholds from the agency, so the supposed cure would repeat the violation it was meant to repair. The Court called the idea internally contradictory for that reason, and it added that whether a statute delegates legislative power is a question for the courts, to which an agency’s voluntary self-denial is irrelevant.26
Two older decisions that mention agency regulations in their delegation discussions do not undercut that rule. In Lichter a later Congress had written the regulations into a revised version of the statute, and in Fahey v. Mallonee the regulations reflected customary practices that the statute implicitly incorporated.27 In both, the limit counted because it already formed part of what Congress had enacted, and a limit the agency chooses for itself has no such source.
With the remand set aside, the question became whether the statute itself supplies a principle, and the Court held that it does. It accepted the Solicitor General’s reading that § 109(b)(1) at a minimum requires EPA to establish uniform national standards, for a discrete set of pollutants and based on published air quality criteria reflecting the latest scientific knowledge, at a level requisite to protect public health from the pollutant’s adverse effects, and it read “requisite” to mean “sufficient, but not more than necessary.”28 That reading settles the three things the test asks Congress to settle.29 The general policy is protection of public health from the pollutant’s adverse effects, the agency is EPA, and the boundaries are the discrete set of pollutants, the published criteria, and a level that may be neither lower nor higher than health protection requires. The two-sided limit matters most, because it gives a reviewing court something to measure: a standard set above what health requires, or below it, departs from what Congress directed.
The Court also placed the provision among its own precedents. The limits on EPA’s discretion are strikingly similar to those in Touby v. United States, which let the Attorney General designate a drug as a controlled substance for purposes of criminal drug enforcement if doing so was “necessary to avoid an imminent hazard to the public safety,” and they resemble the Occupational Safety and Health Act provision requiring the agency to set the standard that most adequately assures, to the extent feasible, that no employee will suffer impairment of health, which the Court upheld in Industrial Union Department.30 Measured against a record in which the Court has found the required principle missing in only two statutes, Panama Refining and Schechter, the discretion in § 109(b)(1) fell well within the outer limits of the precedents.31
The court of appeals had demanded more than a principle. It required a “determinate criterion” for saying how much of the regulated harm is too much, and the Supreme Court answered that it has never demanded one, even for sweeping regulatory schemes.32 The statute in Touby did not decree how imminent was too imminent, how necessary was necessary enough, or how hazardous was too hazardous, and the statute in Lichter let agencies recoup excess profits on wartime Government contracts without saying how much profit was too much.33 The challengers pointed out that ozone and particulate matter harm health at any concentration above zero, so EPA must always make judgments of degree. The Court held that this fact is not conclusive for delegation purposes, because a certain degree of discretion, and thus of lawmaking, inheres in most executive or judicial action.34 When harm rises along a continuum, every regulatory line, here a permitted concentration, is a judgment of degree. A rule requiring Congress to supply a formula for each such line would force Congress to write the technical standards itself, which is the work delegation exists to hand to agencies.
The last step in Whitman keeps the test from becoming a formality. The degree of agency discretion that is acceptable varies with the scope of the power Congress confers.35 Congress need not give EPA any direction on how to define “country elevators,” the grain elevators exempt from new stationary-source regulations, but it must provide substantial guidance on setting air standards that affect the entire national economy.36 The scale tracks how many policy choices a delegation carries with it. A definition of a minor exemption carries few. National air standards shape the costs borne by nearly every industry, so the choices Article I reserves to Congress are more numerous and weightier, and Congress must make more of them itself. Section 109(b)(1) met that raised demand, because it fits comfortably within the discretion the Court’s precedents permit.37
The comparison with 1935 shows the scale at work. Schechter involved authority to regulate the entire economy on no more precise a standard than stimulating it by assuring “fair competition.”38 Section 109(b)(1) also reaches the national economy, but it names one objective, public health, and bounds that objective from both directions. The two powers have a similar reach, so the difference lies in how much each statute tells the agency and a reviewing court.
Two concurring opinions in Whitman planted ideas that later sections of this chapter develop. Justice Thomas wrote that he was not convinced the intelligible principle doctrine prevents every transfer of legislative power, because some cases may involve a principle that is intelligible and a delegated decision whose significance is too great to call it anything but legislative.39 Justice Stevens, joined by Justice Souter, urged the Court to acknowledge frankly that rulemaking authority delegated to an agency is legislative power and to uphold such a delegation when the statute adequately limits it, rather than pretend that the delegated authority is something else.40 Thomas asks whether importance can defeat a delegation even when the standard is clear, and Stevens asks how the Court should describe the power it approves. Neither point changed the majority’s holding.
An answer to a stem that reports a court finding a statute standardless but sending the matter back so the agency can narrow its own reading should say first that the remand cures nothing. An agency’s self-imposed limit cannot supply the omitted standard, so the analysis returns to the statutory text and asks whether Congress itself stated the policy, the agency, and boundaries that give a court a measure. A statute that sets a health target and limits the agency to the level necessary to reach it, as § 109(b)(1) does, passes even though it leaves judgments of degree to the agency. The sliding scale then asks only that a power reaching the whole economy carry correspondingly specific guidance.
Mistretta and the Sentencing Commission
In Mistretta v. United States (1989) the Court applied the intelligible principle test to a recipient of lawmaking power that was neither an executive department nor part of Congress.44 John Mistretta, who was sentenced under the federal Sentencing Guidelines, argued that Congress had given an independent Sentencing Commission excessive legislative discretion when it empowered the Commission to promulgate sentencing guidelines for every federal criminal offense, and the Court rejected the argument.45
The Act placed the Commission within the judicial branch, so the case shows that lawmaking authority can be conferred on the judiciary as well as on executive agencies.46 The district court had looked past the statute’s language and characterized the Commission as having executive branch status, likening the Guidelines to substantive rules issued by other agencies.47 The Supreme Court needed no such relabeling. The test asks whether Congress directed “the person or body authorized” to act to conform to an intelligible principle, and it says nothing about which branch the recipient belongs to.48 The Court rested its delegation holding on the statute’s standards, and only after finding them sufficient did it turn to Mistretta’s separate claim that the Act violates the separation of powers.49 That second claim, which concerns the Commission’s placement and the service of federal judges on it, belongs to another reading.
The majority began where the doctrine begins. Article I, § 1 generally forbids Congress to delegate legislative power, yet the Court’s decisions rest on the practical understanding that, in an increasingly complex society, Congress cannot do its job without the ability to delegate power under broad general directives.50 It is constitutionally sufficient, the Court said, if Congress clearly delineates the general policy, the public agency that is to apply it, and the boundaries of the delegated authority.51 The Sentencing Reform Act supplied each of the three:52
- The general policy was the set of purposes of sentencing that Congress stated, which told the Commission what punishment was meant to accomplish.
- The public agency was the Commission itself, a standing body created to carry that policy out.
- The boundaries were the factors Congress told the Commission to consider, the categories of offense and offender it had to recognize, and the ranges within which it had to work.
The Court adopted a lower court’s summary of the statute, which “outlines the policies which prompted establishment of the Commission, explains what the Commission should do and how it should do it, and sets out specific directives to govern particular situations.”53 On that description the Court concluded that the Act “sets forth more than merely an ‘intelligible principle’ or minimal standards,” and it found the criteria Congress supplied wholly adequate for carrying out the Act’s general policy and purpose.54
That finding separates this delegation from the one in Schechter. Both left the recipient wide room for judgment, but Schechter paired a broad standard, “fair competition,” with power in private parties to write rules for virtually all of American industry.55 The Sentencing Reform Act set the purposes, the factors, and the ranges itself and left the Commission to apply them across the criminal code. Congress made the policy choices and sought help only in carrying them out, so a reviewer can compare any guideline with the statute to see whether the Commission followed Congress’s direction. That comparison is the job an intelligible principle exists to perform for the courts.
The Court gave a second reason, tied to the kind of work involved. Developing proportionate penalties for hundreds of different crimes, it said, is “precisely the sort of intricate, labor-intensive task for which delegation to an expert body is especially appropriate.”56 Proportion can be set only by comparing offenses with one another and with the offenders who commit them, studying how courts have actually sentenced, weighing alternative sentencing models, and revising the ranges as experience accumulates. A standing commission can do that continuously, whereas relying on Congress to amend its own law carries a high decisional cost, and agencies with lower decision costs are often better at working through a labyrinth of detail.57 The Court acknowledged that Congress had delegated significant discretion to the Commission to draw judgments from existing sentencing practice and alternative models, and it answered that Congress “is not confined to that method of executing its policy which involves the least possible delegation of discretion to administrative officers.”58 Discretion alone is therefore never the defect. The question stays whether Congress supplied the principle and the limits.
An exam stem can test whether the expert-body rationale carries a delegation by itself, simply by keeping the commission and taking the standards away.
Expertise explains why the particulars of sentencing go to a commission, and it says nothing about who sets the purposes those particulars serve. In the Sentencing Reform Act Congress set them, which is why the real statute stood and the invented one would not.
Justice Scalia, who dissented from the judgment, did not claim that Congress had written the standards too loosely. He agreed with the Court’s rejection of the delegation challenge and treated the limits of delegation as a question of degree that the Court has almost never felt able to police.59 His objection went to the kind of body Congress had built, since he could “find no place within our constitutional system for an agency created by Congress to exercise no governmental power other than the making of laws.”60 On that view a commission whose only function is to write law exercises legislative power in its purest form. His position previews the later debate over whether the lenient test should survive, and it shows that the majority’s holding on the standards did not depend on resolving his concern about the Commission’s character.
When an exam stem reports a delegation to a commission, board, or other body outside the executive branch, the same test as in any other delegation answers the question: did Congress state the policy, name the body, and set boundaries a court can check? A statute that names the purposes of the power, lists the factors, and fixes the categories and ranges, as the Sentencing Reform Act did, stands even though the recipient has broad judgment to exercise and a heavy technical task to perform. The recipient’s placement and the service of judges on it raise a different question, which a stem will signal by asking about the body’s location or membership.