Part One - The Structure of Government · Module 4
The Taxing and Spending Power
Congress's checkbook powers and their conditions — from Butler through Dole to the NFIB spending holding.
Topics
The purposes for which Congress may tax and spend. Conditions on grants to State governments. The taxing power after NFIB.
Reading list
7 of 7 readings published
- Opinion United States v. Butler 297 U.S. 1 (1936)
- Note Bailey v. Drexel Furniture Co. (The Child Labor Tax Case) 259 U.S. 20 (1922)
- Note Chas. C. Steward Machine Co. v. Davis 301 U.S. 548 (1937)
- Note Helvering v. Davis 301 U.S. 619 (1937)
- Note Sabri v. United States 541 U.S. 600 (2004)
- Opinion South Dakota v. Dole 483 U.S. 203 (1987)
- Opinion (edited) National Federation of Independent Business v. Sebelius (redux) 567 U.S. 519 (2012)
- Current case Landor v. Louisiana No. 23-1197 (U.S. June 23, 2026) (what a private plaintiff may recover when a State accepts federal funds subject to a statutory condition).
Classroom visuals
Discussion
Congress Cannot Order the States. Can It Pay Them Instead? on Canvas — sign in required
This module in Canvas
Modules · Quizzes · Discussion boards · Files and recordings
How to read this list
A case named without a marker is assigned as an opinion and is to be read in its entirety. Opinion (edited) means the Professor has cut it for length; the text on this site is the assigned text. Note means the case is not assigned as an opinion at all — its holding, its facts to the extent they matter, and its place in the doctrinal line are given in a note written for this course, and you are responsible for that note exactly as you are responsible for an assigned opinion. Most modules close with a Current case, a recent decision chosen because it shows the doctrine you have just read being applied at its live edge; current cases are assigned as opinions. You are responsible for all assigned required materials, including the notes, questions, problems and commentary that accompany the cases, whether or not we discuss them in class.
Notes
Bailey v. Drexel Furniture Co. (The Child Labor Tax Case)
259 U.S. 20 (1922) · WestlawFour years after Hammer v. Dagenhart held that Congress could not bar the products of child labor from interstate commerce, Congress tried again through the taxing power: a ten per cent excise on the entire net profits of any business that employed children below specified ages. Chief Justice Taft, for eight Justices, struck it down.
The reasoning is the origin of a distinction you will use for the rest of this module and again in NFIB. A tax may incidentally discourage the thing taxed — every tax does — but this statute did more. “It provides a heavy exaction for a departure from a detailed and specified course of conduct in business.” 259 U.S. at 36. And so: “there comes a time in the extension of the penalizing features of the so-called tax when it loses its character as such and becomes a mere penalty with the characteristics of regulation and punishment.” Id. at 38.
Three features did the work, and they are worth naming because NFIB will run through the same list and come out the other way: the exaction was heavy rather than modest; it was triggered by knowing employment, so it carried a scienter requirement of the sort penalties have and taxes do not; and it was enforced in part by the Department of Labor rather than purely as a revenue measure.
Taft also states the era’s theory of the judicial role in a sentence worth putting beside Justice Roberts’s in Butler: “It is the high duty and function of this court in cases regularly brought to its bar to decline to recognize or enforce seeming laws of Congress, dealing with subjects not entrusted to Congress but left or committed by the supreme law of the land to the control of the States.” Id. at 37. The word “seeming” is carrying an argument.
Two questions to bring. First, if a tax and a penalty are distinguished by how hard they push, then every tax is a penalty at a high enough rate and every penalty is a tax at a low enough one. Is there any principled line here, or only a threshold the Court sets case by case? Second — and hold this until the end of the module — NFIB applies these very factors to the shared responsibility payment and concludes it is a tax: the burden is light, there is no scienter requirement, and the Internal Revenue Service collects it. Nothing about Bailey’s framework changed. Only the answers did. Ask yourself whether that is doctrine working as designed or a hundred-year-old test being run to a predetermined result.
Chas. C. Steward Machine Co. v. Davis
301 U.S. 548 (1937) · WestlawThe Social Security Act of 1935 imposed a federal payroll tax on employers and then allowed a credit of up to ninety per cent against it for contributions paid into a state unemployment compensation fund that met federal standards. A State that declined to set up such a fund did not lose federal money — its employers simply paid the full federal tax and got nothing back. Steward Machine paid under protest and argued that the credit scheme coerced the States into adopting a federal program.
Justice Cardozo, for five Justices, upheld it, and in doing so wrote the sentence that governs every coercion argument you will meet in this course: “But to hold that motive or temptation is equivalent to coercion is to plunge the law in endless difficulties.” 301 U.S. at 589–590. Note the verb — “in,” not “into”; the Court’s text is often misquoted.
The concession in the next breath is as important as the holding. Cardozo does not say coercion is never a limit. He says that if it is one, “the location of the point at which pressure turns into compulsion, and ceases to be inducement, would be a question of degree,—at times, perhaps, of fact.” Id. at 590. That sentence is the seed of everything. South Dakota v. Dole quotes the pressure-into-compulsion phrase fifty years later while finding the pressure mild. NFIB quotes it seventy-five years later and, for the first time, finds the point crossed.
So the doctrinal history is unusually tidy: a 1937 opinion that declines to draw a line, a 1987 opinion that admits the line exists and holds we are nowhere near it, and a 2012 opinion that says we have finally crossed it. Nobody has ever said where it is.
Two questions. First, Cardozo’s worry is administrability — that judges cannot measure temptation. Is that a reason to have no doctrine, or a reason to have a crude one? Compare his instinct here with Justice Marshall’s in Gibbons, that the restraints on a federal power are political rather than judicial; they are the same move made about different clauses a century apart. Second, notice the structure of the credit device. It does not hand a State money on conditions; it taxes the State’s employers and offers relief if the State cooperates. Is that meaningfully different from a conditional grant, or is it the same leverage arriving by a different route? Hold that question for Dole, where the leverage is a highway fund, and for NFIB, where it is the entire Medicaid program.
Helvering v. Davis
301 U.S. 619 (1937) · WestlawDecided the same day as Steward Machine, and by the same author, this case took up the other half of the Social Security Act — the old-age benefits, funded by taxes on employers and employees and paid out of the Treasury. A shareholder sued to stop his corporation from paying, arguing that providing pensions is not spending for the general welfare but a raid on the Treasury for a local and private purpose.
Justice Cardozo upheld the program, and the sentence to know is about who decides what the general welfare is: “The discretion belongs to Congress, unless the choice is clearly wrong, a display of arbitrary power, not an exercise of judgment.” 301 U.S. at 640.
Put that beside United States v. Butler, decided sixteen months earlier. Butler is the case that settled the two-century argument in Hamilton’s favour — the spending power is a distinct substantive power, not limited to the other enumerated powers. But Butler then struck the statute down anyway. Helvering is where the Hamiltonian reading gets its teeth, because it supplies the standard of review: near-total deference. After 1937 the general welfare requirement has essentially never invalidated a federal spending program. A limit that is announced in one case and rendered unreviewable in the next is a particular kind of constitutional object, and you should be able to say what kind.
Cardozo also gives the substantive reason, which is an argument about federalism rather than about economics: “The problem is plainly national in area and dimensions.” Id. at 644. States that provided generously for the aged would attract the aged and drive away business; a State acting alone is in a race it cannot win. That is a collective action argument, and it is the same structure you met in the Commerce Clause module — the reason for national power is that the States are individually incapable, not that the subject is inherently federal.
Two questions to bring. First, is deference here the same deference the Court gives under rational basis review elsewhere, or is it more? Try to state a spending program that would be “clearly wrong” in Cardozo’s sense, and notice how hard that is. Second, if the general welfare requirement does no work, then every real limit on the spending power has to come from somewhere else — from the conditions attached to the money rather than from the purpose of the spending. That is precisely where the doctrine goes next, in Dole, and it is why the restrictions there matter more than the clause itself.
Sabri v. United States
541 U.S. 600 (2004) · WestlawBasim Sabri, a Minneapolis developer, offered cash to a city councilman. He was charged under a federal statute making it a crime to bribe an official of any State or local government that receives more than ten thousand dollars in federal funds in a year. The statute requires no proof of any connection at all between the bribe and the federal money. Sabri argued that without such a link, Congress had no power to reach a local bribe.
Justice Souter, for the Court, disagreed; the judgment was unanimous, but Justice Thomas concurred only in the judgment and would have rested it on the Commerce Clause. The reasoning is why this note sits in a module about spending rather than one about crime. The spending power comes with a companion: “Congress has authority under the Spending Clause to appropriate federal moneys to promote the general welfare, Art. I, § 8, cl. 1, and it has corresponding authority under the Necessary and Proper Clause, Art. I, § 8, cl. 18, to see to it that taxpayer dollars appropriated under that power are in fact spent for the general welfare, and not frittered away in graft or on projects undermined when funds are siphoned off or corrupt public officers are derelict about demanding value for dollars.” 541 U.S. at 605.
On the missing link, the answer is that money is fungible and corruption is not compartmentalized: “this possibility portends no enforcement beyond the scope of federal interest, for the reason that corruption does not have to be that limited to affect the federal interest.” Id. at 606. A corrupt official handling ten thousand federal dollars is a bad bet for every federal dollar that follows.
Notice the doctrinal shape, because it recurs. The federal interest is not in the particular transaction but in the integrity of the institution receiving the funds — which is an aggregation argument, and you have seen aggregation carry enormous weight already, in Wickard and in Raich. In 2026, Landor read Sabri narrowly: it protects federal money from graft, and it is not a power over everyone the money reaches.
Two questions. First, compare this to Lopez. There the Court refused to accept a chain of inferences from guns near schools to the national economy. Here it accepts that a bribe with no proven relation to federal funds nonetheless threatens them. What distinguishes the two chains — the number of links, the plausibility of each, or the clause each runs through? Second, ask what limit survives. If Congress may criminalize conduct affecting any entity that receives federal money, and nearly every government does, is that a spending doctrine or a police power by another name?