Concept
Modern Leniency
Nondelegation doctrine
Modern Leniency
Since Panama Refining and Schechter, nearly ninety years ago, the Supreme Court has not struck down a single federal statute as an impermissible delegation. Those decisions have never been expressly overruled, but the Court has never followed them either, and every delegation it has examined since, however broad, has stood.1
The intelligible principle test makes that record unsurprising. So long as Congress “lay[s] down by legislative act an intelligible principle to which the person or body authorized to [act] is directed to conform, such legislative action is not a forbidden delegation of legislative power.”2 The rise of the modern state has made broad delegations well-nigh unavoidable, and a test that asks only whether Congress supplied a guiding principle leaves Congress wide room to write them.3
The Court’s own tally shows how seldom the test has bitten. In its entire history the Court has found the required principle missing in only two statutes: one provided literally no guidance for the exercise of discretion, and the other conferred authority to regulate the entire economy on a standard no more precise than assuring “fair competition.”4 A statute that names a purpose and a delegate has survived every delegation challenge since, even in terms as open as “generally fair and equitable.”5
Three subsections show the leniency at work. The first collects the broad standards the Court has sustained. The second examines Whitman v. American Trucking Associations, the leading modern application of the test to a major regulatory program. The third examines Mistretta v. United States, which tested the doctrine against an unusual delegate, a commission that writes sentencing guidelines.
Broad Standards Upheld
The Court’s own survey in Whitman shows how little the intelligible principle test has demanded. The Court found the principle lacking in only two statutes, the two struck down in 1935, and it has approved delegations written in language that looks hopelessly open-ended.6 It has sustained delegations of each of the following kinds of authority:
- Broadcast licensing as “public interest, convenience, or necessity” require, in National Broadcasting Co. v. United States.7
- Approval of railroad consolidations by the Interstate Commerce Commission, in New York Central Securities Corp. v. United States.8
- Price fixing at levels that “will be generally fair and equitable and will effectuate the purposes of th[e] Act,” in Yakus v. United States.9
- Determination of “just and reasonable” rates, in FPC v. Hope Natural Gas Co.10
- Recovery of “excess profits” paid under wartime government contracts, in Lichter v. United States.11
- Modification of holding company structures so that they are not “unduly or unnecessarily complicate[d]” and do not “unfairly or inequitably distribute voting power among security holders,” in American Power & Light Co. v. SEC.12
- Regulation of “unfair methods of competition,” in FTC v. Gratz.13
- Designation of a drug as a controlled substance when “necessary to avoid an imminent hazard to the public safety,” in Touby v. United States.14
- Air quality standards “requisite to protect the public health,” in Whitman itself.15
Each formula states a purpose and leaves a question of degree to the delegate. Whitman made the point directly when it rejected the lower court’s demand for a “determinate criterion” saying “how much [of the regulated harm] is too much.” Even in sweeping regulatory schemes, the Court said, it has never required one, and in Lichter it did not insist that Congress specify how much profit was too much.16 No statute can be entirely precise, and some judgments, even some involving policy, must be left to the officers who execute the law and the judges who apply it. Once that is conceded, Justice Scalia observed in his Mistretta dissent, the delegation debate is “not over a point of principle but over a question of degree.”17 Chief Justice Marshall had said much the same in 1825: “The line has not been exactly drawn” between the subjects the legislature must regulate itself and those on which it may give others power “to fill up the details.”18
Yakus shows the leniency at its height, because the statute’s own purposes pulled against one another. Congress told a wartime administrator to fix prices at levels that would be generally fair and equitable and would effectuate the Act’s purposes, which in some respects conflicted, and the Court upheld that direction.19 As the Court later recalled in Mistretta, Yakus itself put the principle this way: “Congress is not confined to that method of executing its policy which involves the least possible delegation of discretion to administrative officers.”20 A challenged price can still be measured against the words “generally fair and equitable.” A price that gouges buyers or ruins sellers falls outside them, and a court can say so even though the administrator keeps wide latitude inside them.
The same open texture appears in current statutes. Section 1182(f) of the immigration code, at issue in Trump v. Hawaii, lets the President suspend the entry of aliens whenever the President finds that their entry “would be detrimental to the interests of the United States.” Like the older formulas, it names the delegate, the finding that triggers the power, and the interest the finding must serve.
Two further decisions show how little precision the Court demands when the delegate is the President. Curtiss-Wright involved a 1934 Joint Resolution authorizing the President to prohibit arms sales to Bolivia and Paraguay, which were then in armed conflict, and Justice Sutherland closed his opinion by stressing “the unwisdom of requiring Congress in this field of governmental power to lay down narrowly definite standards by which the President is to be governed.”21 In Loving v. United States the Court found it “hard to deem lawless a delegation giving the President broad discretion to prescribe rules” governing the factors to consider in imposing the death penalty in courts-martial, because “[t]he delegated duty [is] interlinked with duties already assigned to the President by express terms of the Constitution.”22 A later section returns to how subject matter and the identity of the recipient affect the analysis.
Why did “fair competition” fail when “generally fair and equitable” survived? Vagueness cannot be the whole answer. Whitman described the failed standard as no more precise than assuring “fair competition,” but it also stressed what rode on it: authority to regulate the entire economy.23 The other invalidated statute, the oil-industry provision in Panama Refining, gave the President “absolutely no guidance” about the circumstances in which to impose its prohibition.24 (Panama Refining has also been called a special case, because the petroleum code at issue turned out to contain no legally operative sentence.25) The pattern fits Whitman’s statement that the acceptable degree of agency discretion varies with the scope of the power conferred, which the next subsection develops.26 The invalidated provisions thus combined no guidance, or a bare standard, with sweeping scope, and the Court has not followed those decisions since.27
Three considerations support the Court’s leniency. Broad delegations first serve real needs. For many regulatory problems the legislature can neither foresee what an agency should do nor revise the statute whenever conditions change, and an effort to write highly detailed standards can delay a bill or put its enactment at risk.28 Vague delegations can also be defended as shifting policymaking discretion to appointees of a President who has his or her own electoral base, which gives an incumbent administration the latitude to carry out its mandate.29
Courts also lack a workable test. If the standards for applying the doctrine were too subjective, every new regulatory statute would be under threat, because with hindsight a private party can almost always argue that Congress should have said more.30 The pattern of decisions reflects a judgment that broad delegations are necessary in a complex modern economy and that the judiciary is ill equipped to draw meaningful lines.31
Finally, other controls carry part of the load. Courts have found alternative ways to keep broad grants of power from becoming instruments of administrative oppression, such as reading a statute to specify limits on regulation or to require procedural fairness.32 Those techniques are less disruptive than invalidation, which can force the radical restructuring or abandonment of an entire program, and a narrowing construction is easier to correct if the court has misread Congress.33
The permissive approach may not last. The Court’s growing skepticism about unconstrained administrative power suggests that it may not persist indefinitely,34 and a later section takes up the revival of the doctrine, beginning with Gundy v. United States. Even there the Court held that the Attorney General’s authority under SORNA “easily passes constitutional muster.”35
To defend a delegation, point to the general policy, the delegate, and the boundaries Congress supplied, however broad, and note that breadth alone has never led the Court to invalidate a statute. To attack one, look for what the 1935 statutes shared: no guidance at all, or thin guidance joined to sweeping scope. Then acknowledge that the Court has drawn that line only twice. The next subsection turns to Whitman, where the Court rejected a nondelegation attack on a standard as general as any listed above.
Whitman v. American Trucking
Section 109(b)(1) of the Clean Air Act directs the EPA Administrator to set national ambient air quality standards, the nationwide limits on how much of a pollutant the outdoor air may contain. Those standards must be ones “the attainment and maintenance of which in the judgment of the Administrator, based on [the] criteria [documents of § 108] and allowing an adequate margin of safety, are requisite to protect the public health.”37 In Whitman v. American Trucking Associations, Inc. the Court faced four questions, and the first asked whether § 109(b)(1) delegates legislative power to the Administrator.38 The other three, which concerned cost considerations and the EPA’s implementation of the revised ozone standards, fall outside this chapter.39
The court of appeals held that the EPA’s interpretation of § 109(b)(1) supplied no intelligible principle: the agency “lack[ed] any determinate criteria for drawing lines” and “failed to state intelligibly how much is too much.” It faulted the EPA’s interpretation, not the statute itself.40 That placement of the defect implied that the agency could cure it by narrowing its own reading. The Supreme Court reversed in two moves: it held that an agency cannot supply a missing limit, and then that the statute, read by the Court, already contained one.
The first move rests on Article I, § 1, which vests “[a]ll legislative Powers herein granted” in Congress.41 Congress may seek help from the other branches, but only if it lays down by legislative act an intelligible principle to which the delegate must conform.42 The Court called the idea that an agency can cure an unconstitutionally standardless delegation by declining to exercise some of its power “internally contradictory.” The very choice of which portion of the power to exercise, meaning the prescription of the standard Congress had omitted, would itself be an exercise of the forbidden legislative authority.43 A limit the agency chooses is also one the agency can drop, so it never operates as a command from Congress. Whether a statute delegates legislative power is a question for the courts, and an agency’s voluntary self-denial has no bearing on the answer.44
Two earlier decisions mention agency regulations in their nondelegation discussions, and neither points the other way. In Lichter v. United States a later Congress had incorporated the regulations into a revised version of the statute, and in Fahey v. Mallonee the customary practices implicitly incorporated into the statute were reflected in the regulations.45 In each case the limit rested on something Congress had enacted or incorporated, not on the agency’s own restraint.
The second move holds that the statute itself supplies a principle. The Court accepted the reading that, “[f]or a discrete set of pollutants and based on published air quality criteria that reflect the latest scientific knowledge, [the] EPA must establish uniform national standards at a level that is requisite to protect public health from the adverse effects of the pollutant in the ambient air,” and that “requisite” means “sufficient, but not more than necessary.”46 That word draws a boundary on both sides. A standard below what is necessary is not sufficient to protect health, and a standard above it exceeds what Congress authorized, so the agency has a target and a reviewing court has something to measure the agency’s choice against. The Court found these limits “strikingly similar” to those it had approved in Touby, where the Attorney General could designate a drug as a controlled substance if doing so was “necessary to avoid an imminent hazard to the public safety.”47
The Court’s reading of “requisite” may look like the limiting construction it refused to let the agency supply, but the two differ in source. The limit comes from a word Congress enacted, read by a court that decides what the statute means, whereas an agency’s self-restraint is a choice Congress never made.
The court of appeals had also demanded more than a boundary. It wanted a “determinate criterion” for saying “how much [of the regulated harm] is too much,” and the Court answered that even in sweeping regulatory schemes it has never required one.48 Judgments of degree cannot be removed from a regulatory statute,49 so the question is whether Congress marked the boundaries within which the agency makes them. A directive to set standards at the level that is “requisite,” meaning not lower or higher than necessary, to protect the public health with an adequate margin of safety “fits comfortably within the scope of discretion permitted by our precedent.”50
Whitman also teaches that the amount of guidance a statute must supply is not constant. The Court acknowledged that the acceptable degree of agency discretion varies with the scope of the power Congress confers. Congress need not direct the EPA on how to define “country elevators,” which are exempt from new regulations for stationary sources governing grain elevators, but it must provide “substantial guidance on setting air standards that affect the entire national economy.”51 The reason lies in how much policy choice rides on the delegate’s decision. Defining a narrow category of grain facilities moves few interests, while a national air standard decides how much cost and risk the whole economy will bear. Some powers, such as the power to declare war, are so central that they may be delegated only under a narrowly defined principle, if at all.52 Section 109(b)(1) cleared the higher bar for a national program because it paired “requisite” with a defined set of pollutants and published scientific criteria.53 A later subsection returns to scope as a factor of its own.
The first step disposes of any argument built on an agency’s promise to regulate narrowly, and the second and third do the work when a broad but real standard is attacked. A challenger to a delegation of national reach should concentrate on the fourth step, because it is the variable the Court said can raise the bar.
Two Justices wrote separately, and each opinion looks beyond this case. Justice Stevens agreed that the delegation was constitutional but would have explained why differently. The Court could frankly acknowledge that the power delegated to the EPA is “legislative” and still find the delegation valid because the statute adequately limits it. Instead, he said, the majority pretended that the authority is not legislative power at all, and he thought it “wiser and more faithful to what we have actually done in delegation cases” to admit that agency rulemaking authority is legislative power.54 He added that Article I’s grant of “All legislative Powers” to Congress, like Article II’s grant of the executive power to the President, does not purport to limit its recipient’s authority to delegate to others.55 On his approach the question remains whether Congress limited the delegate, so the label would not have changed the outcome.
Justice Thomas wrote that no party had examined the text of the Constitution or asked the Court to reconsider its precedents on cessions of legislative power, but that on a future day he would be willing to address “whether our delegation jurisprudence has strayed too far from our Founders’ understanding of separation of powers.”56 That open question returns near the end of this chapter, where the major questions doctrine already limits the scope of existing statutes in ways that function like nondelegation.57 For exam purposes, the majority opinion states the governing law, and the separate opinions show where a challenger can push.
Mistretta and the Sentencing Commission
Mistretta v. United States put the intelligible principle test to an unusual delegate: a sentencing commission Congress created to write guidelines for federal criminal sentences, drawing its judgments from analysis of existing sentencing practice and alternative sentencing models.59 The case reached the Supreme Court before the court of appeals had ruled, because both Mistretta and the United States petitioned for certiorari before judgment. The Court granted review because of the “imperative public importance” of the issue and the disarray among the federal district courts.60
The Court began from familiar premises. Article I vests “[a]ll legislative Powers herein granted” in Congress, so Congress generally cannot delegate legislative power to another branch. It may, however, obtain the assistance of its coordinate branches, and the extent of that assistance “must be fixed according to common sense and the inherent necessities of the government co-ordination.”61 A delegation is “constitutionally sufficient if Congress clearly delineates the general policy, the public agency which is to apply it, and the boundaries of this delegated authority.”62
The Act met that formula. Congress set forth the purposes of sentencing, the factors to consider in determining a sentence, and the kinds of penalties that may be imposed, and it directed the Commission to promulgate guidelines consistent with those directives.63 The purposes supply the general policy, the Commission is the agency, and the factors and permitted penalties mark the boundaries. The Court added that the Act sets forth “more than merely an ‘intelligible principle’ or minimal standards.” It quoted with approval another court’s description of a statute that “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.”64
Significant discretion did not defeat the delegation. Congress had left the Commission to draw judgments from its analysis of existing practice and alternative models, but “Congress is not confined to that method of executing its policy which involves the least possible delegation of discretion to administrative officers,” and the criteria Congress supplied were “wholly adequate for carrying out the general policy and purpose” of the Act.65 The question is the one Whitman asked of the Clean Air Act: whether Congress marked the boundaries within which the judgment is made.
Justice Scalia dissented, and his objection was structural. Once one concedes that no statute can be entirely precise and that some judgments, even some involving policy, must be left to the officers who execute the law and the judges who apply it, he wrote, the delegation debate becomes “a debate not over a point of principle but over a question of degree.”66 He therefore looked past the sufficiency of the standards to a limit of principle: “the power to make law cannot be exercised by anyone other than Congress, except in conjunction with the lawful exercise of executive or judicial power.”67 By that test, trial judges could be given the power to decide which factors justify a greater or lesser sentence within statutory limits, because that power is ancillary to their power to pronounce sentence. The President, through a Parole Commission subject to his appointment and removal, could be given the power to issue guidelines on when parole is available, because that power is ancillary to the executive power to hold and release prisoners.68 On his account, a commission whose work consists of writing sentencing guidelines fit neither category, because its lawmaking was not tied to any executive or judicial function.
The two opinions therefore ask different questions of the same statute. The majority asked how much guidance Congress supplied, found purposes, factors, and permitted penalties, and sustained the delegation. Scalia asked what else the lawmaker does besides make law. The parole example shows the line inside his theory: guidelines about when prisoners are released are ancillary to the executive’s job of holding and releasing them, while guidelines about sentences, written by a body that neither adjudicates nor executes, are not. His test did not carry the day, and the majority’s sufficiency analysis is the one to apply.
The Court took up a separate claim only after it had finished with delegation. Having found the standards sufficient, it turned to Mistretta’s claim that the Act violates the constitutional principle of separation of powers.69 Separation-of-powers disputes arise among all three branches of the federal government.70 The Court has relied on the doctrine to prevent congressional and executive encroachment upon the judicial branch.71 On Mistretta’s claim the Court followed the flexible, functional approach of Morrison v. Olson rather than the more formal analysis of Bowsher v. Synar and INS v. Chadha, which a later section examines. That holding concerns the structure of the judiciary, not the adequacy of the standards, so a delegation answer should rest on the first holding alone.