Concept
Other Judicial Considerations
Nondelegation doctrine
Other Judicial Considerations
Because the intelligible principle test is remarkably easy to satisfy, a court weighing a delegation challenge rarely stops at the words of the statutory standard.1 Modern decisions examine the total system of controls, substantive and procedural, that limits the delegate’s power. They ask whether the statutory scheme, taken as a whole, assures adequate control and accountability in the exercise of official power.2
Three considerations recur in that examination. The constitutionality of a delegation may turn on its subject matter, on the identity of the recipient, and on the presence or absence of accountability mechanisms such as mandatory procedures or judicial review.3 Each becomes a question. How much power does the statute confer, and over what? Who receives that power? Does the statute surround the recipient with procedures and review that let a court hold it to the will of Congress?
These considerations adjust the intelligible principle test; they do not replace it. Each raises or lowers the amount of guidance a court will find sufficient. A thin standard may survive when the power is narrow, the recipient is a public body, and procedures give a reviewing court something to measure against. The same standard is in serious trouble when the power is sweeping, the recipient is private, and no procedures exist. The three questions therefore help predict how a court will treat a delegation it has not yet seen.
Subject Matter and Scope
Chief Justice Marshall identified the problem in 1825: “The line has not been exactly drawn which separates those important subjects, which must be entirely regulated by the legislature itself, from those of less interest, in which a general provision may be made, and power given to those who are to act under such general provisions to fill up the details.”4 The scope consideration is the modern effort to locate that line one statute at a time. The more important the subject and the wider the power, the more of the policy Congress must supply itself.
Whitman v. American Trucking Associations made the scaling explicit: the degree of agency discretion that is acceptable varies according to the scope of the power Congress confers.5 Congress need not give the EPA any direction about how to define “country elevators,” the grain facilities exempt from the new-source rules for grain elevators, but it must provide substantial guidance on setting air standards that affect the entire national economy.6 A loose standard that suffices for the first task may therefore fail the second.
Two reasons support the scaling rule, and they work on different sides of the test. The first is the stakes of the choice. A narrow definitional power moves few interests. Whoever exercises it settles a detail within a policy Congress already chose, and an error is easy to notice and correct. A national air standard decides how much cost and risk the whole economy will carry, which is the kind of policy choice that Article I’s filters of two Houses, presentment, and accountability to voters exist to control. The second is review. A court can hold a delegate to the will of Congress only if the statute supplies something to measure the delegate’s action against, and the wider the field, the more room the delegate has to substitute its own policy unless the statute marks the boundaries precisely.
Greater scrutiny of a broad power is not a demand for precision. Even in sweeping regulatory schemes the Court has never required a statute to provide a “determinate criterion” for saying “how much [of the regulated harm] is too much.”7 In Touby the Court did not require Congress to decree how “imminent” was too imminent, how “necessary” was necessary enough, or how “hazardous” was too hazardous. In Lichter the statute authorized agencies to recoup “excess profits” paid under wartime Government contracts, and the Court did not insist that Congress specify how much profit was too much.8 Scope raises the direction a statute must give about its policy and boundaries, but it never requires a number.
At the far end of the scale, some congressional powers, such as the power to declare war, are so central to legislative authority that they may not be delegated, or may be delegated only under a narrowly defined intelligible principle.9 The reason follows from the scaling logic. A decision to commit the nation to war settles the largest policy question a government faces, so a statute that handed the decision to a delegate would transfer the entire choice and leave nothing for the delegate to fill in. Subject matter can also cut the other way when the recipient already holds constitutional authority in the field. The President’s powers in military matters and foreign affairs justified delegating the power to define court-martial offenses and to activate a statutory arms embargo.10 A declaration of war lies at the core of Congress’s own authority, while an embargo lies where the President’s sphere overlaps with Congress’s. Both concern foreign or military affairs, yet the answers differ.
Delegated power over fundamental liberties forms a third setting, and courts handle it by construing the delegation narrowly. In Kent v. Dulles the Court held that the power delegated to the Department of State to issue passports did not include denying a passport because of the applicant’s political affiliation, since Congress had not clearly authorized that action.11 Where the activity is one, like travel, that is natural and often necessary to a citizen’s well-being, the Court said it would construe narrowly all delegated powers that curtail or dilute it.12 Hampton v. Mow Sun Wong applied the same technique. The Civil Service Commission had issued a rule denying federal jobs to noncitizens, and the Court, finding that the rule restricted an important personal liberty, declined to presume that a body charged with promoting the efficiency of the civil service had been given that power.13 Narrow construction does not strike the statute down. It leaves the liberty-restricting choice to Congress, in terms clear enough to pass through the Article I process, and an agency cannot supply that choice by reading its grant expansively. A later section takes up the technique as an application of constitutional avoidance.
The two concerns reinforce each other when a grant is both broad and aimed at liberty. In United States v. Robel, Congress had authorized the Secretary of Defense to deny employment in any defense facility to members of the Communist Party. Justice Brennan, concurring, relied on the nondelegation cases and observed that formulating policy is a legislature’s primary responsibility. The statute never defined “defense facility,” he noted, so the phrase reached “virtually every place of employment in the United States.”14 A term that open could not mark the field in which the Secretary was to act, and no one could tell whether the Secretary had stayed within the limits Congress intended.15 His reasoning fits the scaling rule: the reach of the power was nearly unlimited, and the liberty at stake was substantial.
Criminal sanctions raise a further question that the Court has left open. In Touby the Court acknowledged that it is unclear whether something more than an intelligible principle is required when Congress authorizes another branch to promulgate regulations contemplating criminal sanctions.16 The concern comes from the structure of a prosecution. When an agency writes the rule and the executive enforces it, one branch defines the prohibited conduct and charges those who violate it, and the argument runs that a person facing punishment should be able to learn from Congress’s own text what conduct can become a crime. A familiar arrangement pairs a legislative penalty with agency-defined conduct. The national forest statute authorized the Secretary of Agriculture to make regulations on the use and occupancy of the forests and itself provided the punishment for violating them, and the Secretary’s grazing regulations required sheep grazers to obtain a permit.17 These pressures support an argument for a stricter test, although the Court has not adopted one. The lower courts have divided. The Eleventh Circuit in United States v. Brown answered that no heightened standard is required. The Second Circuit in United States v. Dhafir treated the issue as open and upheld a delegation of presidential authority that could be exercised only to deal with an unusual and extraordinary threat for which a national emergency had been declared, constraints similar to those found sufficient in Touby.18 An exam answer should call the heightened-standard question unresolved, argue that criminal enforcement raises the stakes, and then test whether the statute carries Touby-like constraints, since such constraints help the statute under either view.
The first step does the main work in most problems. The other three adjust the result when the subject matter or the consequences make the delegation more or less defensible.
Private Recipients
Carter v. Carter Coal Co. is the leading decision on private recipients. It held invalid the delegation of legislative power to the Bituminous Coal Commission and to the code-making authorities created by the Bituminous Coal Conservation Act of 1935.20 The Act tried to stabilize the coal industry through price fixing, wage and hour rules, and collective bargaining provisions enforced through a code system.21 What made the arrangement especially suspect was who held the power: decisionmaking authority had effectively passed to industry representatives instead of government officials.22 The Court called it “delegation in its most obnoxious form; for it is not even delegation to an official or an official body, presumptively disinterested, but to private persons whose interests may be and often are adverse to the interests of others in the same business.”23
The Court had voiced the same worry in Schechter Poultry, where it disfavored the authority given to private trade associations to write their own “codes of fair competition” because the associations might protect themselves at the expense of competitors and consumers.24 When governmental power passes to a private group, some members of the industry may use it to harm or exclude their competitors, and regulation by a biased decisionmaker may be fatal to a delegation under the due process clause.25 Many scholars and at least three Justices have since explained the NIRA decisions as resting not on a bar against delegating policymaking power to a government agency, but on a bar against delegating broad regulatory power to private market participants with clear conflicts of interest.26
Two reasons support the suspicion, and each fills a different gap in Article I’s design. The first is accountability. Only Congress holds the federal legislative power because only Congress is politically accountable for its policy choices.27 An agency is unelected, but it sits beneath a President who answers to the voters and oversees its choices, while a private body escapes that oversight.28 The second is motive, and it does the more distinctive work. Government officials are presumptively disinterested, but a trade association or industry committee has a stake in the rules it writes. If market participants draft a rule that burdens their competitors, they can raise rivals’ costs while sparing themselves, and the consumers who pay the resulting prices have no voice in the drafting. Requiring a disinterested person to make the binding choice removes that temptation at the source. By the same logic, delegations to the President, the states, or tribal governments, which hold independent constitutional or sovereign authority, are less likely to be problematic.29
The prohibition does not bar every arrangement in which private people take part, and two decisions shortly after Carter Coal show where the line runs. In Currin v. Wallace the Court upheld the authority of the Secretary of Agriculture to set binding tobacco standards that took effect upon a vote of two-thirds of the growers.30 Public officials determined what the standards said, and the growers decided only whether those standards would go into effect, so no private party chose the content of the law. In Sunshine Anthracite Coal Co. v. Adkins the Court allowed local coal producers to sit on boards regulating coal sales, because the governmental Bituminous Coal Commission oversaw the boards and the statute required public officials to review the privately devised regulations and place an official imprimatur upon them.31 Neither case overruled Carter Coal. What separates them from it is who holds the last word on content. In Carter Coal the interested committees’ decisions operated as law on their own authority; in Currin government wrote the rule, and in Sunshine government approved it before it bound anyone.
The Court has continued to emphasize government review of private decisionmaking when it upholds private delegations.32 The principle itself survives, for the Fifth Circuit has stated that the “private nondelegation doctrine” has “continuing force.”33 Litigation over Amtrak shows the question arising in a modern setting. Congress gave Amtrak a role alongside the Department of Transportation in developing performance standards for trains running on tracks Amtrak uses. In Association of American Railroads v. Department of Transportation, the D.C. Circuit treated Amtrak as a private corporation and struck the provision down, because Congress could not hand such a body power to set standards by which it might itself benefit.34 The Supreme Court reversed, treating Amtrak as a public company. That treatment did not dispose of the case, because further questions remained, including one involving the Surface Transportation Board.35 The lesson is that classifying the recipient as public or private can change the nondelegation analysis.
Private parties can also gain power at one remove. An agency may outsource functions by contract, so that Congress delegates to the agency and the agency, through contract, delegates to private providers, which is delegation twice removed. Federal procurement laws specify the procedures and rules that govern how public contracts may be let.36
The third and fourth steps decide most problems. A statute survives when public officials set the content and private parties only decide whether it takes effect, as in Currin, or when public officials review and approve what private parties devise, as in Sunshine. A statute fails when interested private persons write rules that bind their competitors and the rules operate without any public check, as in Carter Coal.
Procedures and Judicial Review
In INS v. Chadha, which a later section takes up, the Court explained why an executive officer’s action under delegated authority needs no approval from both Houses of Congress. When the Attorney General acts under the statute authorizing suspension of deportation, the Court said, the action is always subject to the terms of the authorizing legislation, is open to judicial review if it exceeds that authority, and remains exposed to Congress’s power to modify or revoke the authority entirely.38 Once a case or controversy arises, courts can always “ascertain whether the will of Congress has been obeyed” and can enforce adherence to statutory standards.39 That description treats a delegation as one part of a system. The statutory standard is the first control, and the procedures surrounding the delegate and the courts reviewing its action are the others. A broad standard therefore looks different once the whole system comes into view.
Procedure and review matter for a reason that a standard cannot supply on its own. A standard is only language until something requires the delegate to show how it applied that language. Notice and a hearing let those who will be bound put facts and objections before the delegate. A written explanation exposes the basis for the choice. Judicial review lets a court compare that basis with the statute and set aside action that exceeds it. Where those pieces exist, a broad standard carries less risk that discretion will be used arbitrarily. Where they are missing, the standard carries the burden alone.
The 1935 decisions show what the missing pieces look like. In Panama Refining and Schechter the Court placed considerable emphasis on the fact that the statutes did not require the President to use fair and open administrative procedures or to explain his decisions clearly.40 Schechter made the comparison explicit. Other broad delegations, to the Federal Trade Commission, the Interstate Commerce Commission, and the Federal Radio Commission, had come with at least a minimal set of procedural controls, while the National Industrial Recovery Act, in the Court’s words, “dispenses with this administrative procedure and with any administrative procedure of an analogous character.”41 The Act required no trial-type hearing and no notice or right to participate for interested persons, and it gave no opportunity for judicial review to anyone a code harmed.42 Each omission removed a different check. Without notice, the competitors and workers a code would bind could not object. Without a hearing, no record showed what the proposing industry had told the government. Without review, no court could hold the code to the statute. The private-recipient discussion above showed the same gap in Carter Coal, where interested industry bodies made rules with no government procedure between their interests and the law.43
The favorable side of the ledger rests on the same logic. American Power & Light Co. v. SEC treated a delegation as sufficient when Congress delineates the general policy, the agency, and the boundaries of the authority, and it added that “[p]rivate rights are protected by access to the courts to test the application of the policy in the light of these legislative declarations.”44 Yakus v. United States turned that idea into a working rule, holding that the requirement of statutory standards made judicial review of agency action feasible because the standards gave courts a baseline for measuring what the agency did.45 The dependence runs in both directions. Standards give review something to apply, and review gives standards force, since a limit that no court can enforce restrains the delegate only as far as the delegate chooses to be restrained.
Touby v. United States shows the idea at work where a statute seemed to remove review. Under a 1984 amendment to the Controlled Substances Act, the Attorney General could issue a temporary scheduling order imposing strict controls on newly invented designer drugs, and the Act said such an order was not subject to judicial review.46 A manufacturer prosecuted for violating an order argued that the statute was an invalid delegation, in part because of that bar.47 The Court held that the bar referred only to preenforcement challenges, so a person facing criminal charges could still attack the order’s validity as a defense to prosecution.48 The reading was somewhat strained, and by adopting it the Court seemed to acknowledge that a more literal reading might have raised serious doubts about the constitutionality of the Attorney General’s powers.49 Preserving a route to a court kept the check in place and let the statute stand.
Justice Breyer’s dissent in Clinton v. City of New York presents the same considerations from the other side. Defending the delegation in the Line Item Veto Act, he conceded that it differed from ordinary agency delegations in ways that argued against his conclusion. Agencies often develop subsidiary rules that explain general language such as the public interest, which diminishes the risk that an agency will use the breadth of its grant as a cloak for unreasonable or unfair implementation. Agencies are also typically subject to judicial review, an additional check against arbitrary implementation.50 The President had not narrowed his discretion through rules, and his decisions were not reviewable under the Administrative Procedure Act.51 Breyer found those differences important but not determinative. The President, unlike most agency decisionmakers, is an elected official responsible to the voters, and whether his expenditure decisions are arbitrary has been left primarily to those voters to consider.52 The opinion is a dissent, and the Court struck the Act down on bicameralism and presentment grounds that a later section takes up.53 It nonetheless names the checks that ordinarily accompany a broad grant, and it shows political accountability occupying the place of procedural ones.
Procedures and review do not form a freestanding requirement; they are considerations that move outcomes. Whether the statute conditions the delegated power on procedural constraints such as mandatory public participation, or provides judicial review that checks the delegate through an independent judiciary, may be relevant to how a court treats an otherwise thin standard.54
The same idea explains a technique that a later section develops. Courts rarely invalidate statutes on delegation grounds, and the more common course is to interpret a statute so that the delegation is acceptable.55 Courts sometimes adopt relatively restrictive interpretations of regulatory legislation, stating or suggesting that a more permissive reading could raise nondelegation problems, and they have sometimes implied procedural safeguards in this way.56 When the choice is between narrowing a statute and invalidating it, narrowing has evident advantages. A holding that a delegation is unconstitutional invites a major confrontation between the branches, because it may require radical restructuring or even abandonment of an entire program. Interpretations that specify limits or require procedural fairness are less disruptive and more easily corrected if the court has misread the will of Congress.57 Narrowing has a cost too: substituting judicial for executive lawmaking replaces one unelected decisionmaker with another and does little to encourage Congress to take responsibility for major policy choices.58
The second step matters because a literal bar on review may be read narrowly, as in Touby, and the fifth reflects that these factors adjust the intelligible principle test without replacing it. An answer arguing that a delegation is too broad can point to the missing hearings, notice, and review in Schechter. An answer defending a thin standard can point to the availability of procedure and review as the controls that make it workable.