Concept
Limits on the Anticommandeering Rule
State immunity from federal law, including the 10th Amendment
Limits on the Anticommandeering Rule
Justice O’Connor stated the anticommandeering rule in a single sentence: “[t]he Federal Government may not compel the States to enact or administer a federal regulatory program.”1 That sentence forbids commands, and the rule stops where commands stop. Each limit in this section answers one question: whom does the federal provision direct, and by what means? A provision may reach states and private actors alike because both engage in the regulated activity. It may reach a state through an offer the state can decline at the price of the benefit. Or it may reach a state by command, ordering its legislature or officers to enact, administer, or refrain from enacting a federal program. Only the third is commandeering.
New York itself described two offers of the second kind. Congress could occupy a field with its own regulatory program unless and until the state legislature passed a law that Congress wanted on the books, and it could condition access to federal funds on a state’s adopting a favored program.2 The Court called these tools “conditional preemption” and “conditional spending” and found them fundamentally different from an outright mandate, because “[w]here Congress encourages state regulation rather than compelling it, state governments remain responsive to the local electorate’s preferences.”3
A choice between two commands is another matter. An order to take title to waste, standing alone, and a direct order to regulate, standing alone, were each beyond Congress’s authority, so New York held that Congress lacked the power to offer the states a choice between them: “A choice between two unconstitutionally coercive regulatory techniques is no choice at all.”4
The reasons behind the rule explain why only the third situation is commandeering. When a state is regulated like everyone else, Congress conscripts none of its machinery and answers to the voters for the rule. When a state accepts or declines an offer, its own voters keep the decision. The accountability and cost concerns described in the last section therefore arise only when Congress commands. Garcia’s rejection of state immunity remains good law, and it makes clear that the federal government may regulate many state activities, just not through commandeering.5
The rule also depends on the power Congress invokes. The Tenth Amendment reserves to the states the powers that the Constitution neither delegates to the United States nor prohibits to the states,6 and the Constitution grants Congress no general police power.7 Every federal rule therefore rests on a particular delegated power, and the reach of the anticommandeering rule varies with the power. New York and Printz were commerce-power rulings, and there is little doubt that Congress is free to commandeer state action when it acts under its Fourteenth and Fifteenth Amendment enforcement powers.8 A problem built on one of those powers needs a different analysis, so identify the source of congressional power before applying the rule.
The usual mistake on this topic is to find commandeering because a federal law burdens a state, costs it money, or reaches its employees. The question is narrower. Does the provision dictate what a state legislature or its officers must do in governing, as the sports-betting ban in Murphy and the background-check duty in Printz did? Or does it regulate an activity in which states and private parties both engage, as the wage rule in Garcia did? A command to the state invalidates the provision. A rule of general application leaves the state where it leaves everyone else engaged in the activity.
Evenhanded Regulation of States
The Tenth Amendment reserves to the states the powers that the Constitution neither delegates to the United States nor prohibits to the states.9 Its text is essentially a tautology and supplies no limit of its own; it confirms instead that federal power is subject to limits that may, in a given instance, reserve power to the States.10 The Amendment therefore directs a court to ask whether an incident of state sovereignty is protected by a limitation on an Article I power.11 Murphy marked where that limitation stops: “The anticommandeering doctrine does not apply when Congress evenhandedly regulates an activity in which both States and private actors engage.”12
A rule that covers an activity reaches a state only because the state takes part in that activity, and it tells the state nothing about how to govern. Garcia, the model case, applied the wage and hour provisions of the Fair Labor Standards Act to state employees. The Court held that federalism is protected, if at all, through the political rather than the judicial process, because Congress is composed of representatives from the states and must make the judgments about the scope of any intrusion on state sovereignty.13 In Garcia’s words, “[t]he political process ensures that laws that unduly burden the States will not be promulgated.”14
Federalism can be enforced judicially against Congress in two ways. It may operate as a rule of construction that limits the scope of granted powers, as in Lopez, where the Court narrowed the commerce power by putting teeth into the “substantially affects” test.15 Or it may serve as an independent check on a granted power, much as the Bill of Rights does: Congress may not use its commerce power to transgress the First Amendment, and it may not use that power to violate the principle of federalism.16 The anticommandeering rule is a check of the second kind.
Two reasons behind the rule show why evenhanded regulation falls outside it. The first is cost. When Congress enacts a law and requires the Executive Branch to enforce it, Congress must appropriate the funds needed to administer the program, and it is pressured to weigh the program’s expected benefits against its costs; if Congress can compel the States to enact and enforce its program, it need not engage in any such analysis.17 Printz rests on the same structure: Congress must depend on the federal Executive, not the state executive, to enforce its laws.18 The second reason is accountability, since a federal rule that binds states and private parties in the same terms belongs visibly to Congress. Neither concern arises under such a rule. Private parties carry the same burden as the state and can press the same objections in the political process, and no state treasury pays for a federal program. A law that speaks only to the states lacks all of those features.
New York drew the line itself. Most of the Court’s recent Tenth Amendment cases, it observed, had concerned the authority of Congress to subject state governments to generally applicable laws.19 Garcia was distinguishable as a case in which Congress had subjected a State to the same legislation applicable to private parties, because the FLSA applied equally to public and private employers.20 The take-title law, by contrast, failed to qualify as generally applicable because it imposed legal duties only on state governments.21 Printz left the same line in place. It too did not question Garcia, which recognized the power of the federal government to subject states to commercial regulations equally applicable to private-sector and public-sector activities.22 A state can therefore be required to pay its employees the federal minimum wage under Garcia.23 It cannot be ordered, under New York and Printz, to have its officers monitor private employers’ compliance with the wage law.
South Carolina v. Baker shows evenhandedness at work in the tax code. Congress removed the federal tax exemption for interest on bearer bonds.24 Even if that removal were tantamount to an outright ban on issuing bearer bonds, Murphy explained, the law would simply treat state bonds the same as private bonds.25 Murphy also read Baker as a case in which the law ordered no state to enact or maintain any law; its effect on state behavior was indirect pressure, which Congress may create by imposing duties on private actors.
Reno v. Condon applied the same reasoning to a statute that reached closer to the daily work of state government: a federal restriction on disclosing personal information drawn from state motor vehicle records, which bound state agencies and private resellers alike. South Carolina objected that compliance would consume its employees’ time and resources. The Court accepted that compliance would take effort and still found no commandeering. Murphy later summarized the decision this way: the law applied equally to state and private actors and did not regulate the States’ sovereign authority to “regulate their own citizens.”26 The Court left open whether general applicability is a constitutional requirement for regulating states, because the Act satisfied it anyway.
Haaland v. Brackeen confirms that the face of a statute does much of the work. The Court rejected anticommandeering objections to several provisions of the Indian Child Welfare Act mainly because they applied to public and private actors alike, and it remarked that a commandeering argument is a heavy lift when a federal statute on its face applies to both private and state actors. The phrase marks the argument as difficult without ruling it out, and “on its face” tells you to begin with the statute’s text, which shows whether private actors are regulated alongside the state.
The provisions that failed share the opposite feature. New York’s take-title provision addressed only state governments, which is why the Court said its case was not one in which Congress had subjected a State to the same legislation applicable to private parties.27 The Brady Act’s background-check duty fell on state law enforcement officers, and Printz struck down even seemingly minor duties imposed only temporarily, until a federal system became operative.28 PASPA’s ban on state authorization of sports betting spoke to state legislatures alone and imposed no federal rule on private bettors. In each of the three, the provision addressed the state as a governing body, and no private party stood beside the state to share the burden or to object.
Exam problems usually describe a statute that mentions both a state and private parties, and the useful question is what the statute requires the state to do. A statute built on the Condon model passes, and one that borrows the state as an enforcer does not.
The second and third steps do the real work. The cost of compliance drops out of the analysis, because the state in Condon objected to exactly that cost and lost. A student who concludes that a law commandeers because it burdens state employees has asked a question the Court set aside. The opposite error treats the mention of private parties in the statute as a complete answer, when it is strong evidence that still leaves open what the provision demands of state officers.
Sanctuary policies supply a modern application. Federal law cannot compel state and local officials to enforce federal immigration law,30 though federal statutes may authorize willing state and local officials to carry out federal immigration laws in designated circumstances.31 In City of New York v. United States, the Second Circuit upheld federal statutes that bar local rules forbidding government employees to exchange information with federal immigration officials, because the statutes did not compel state and local governments to enact or administer any federal regulatory program. The two propositions fit together: the information provision orders no official to gather or report anything and displaces only a local rule that stood in the way of voluntary communication. That decision preceded Murphy, which held that a federal prohibition addressed to a state legislature is as much a command as an order to act, so a problem built on City of New York should ask whether the federal provision regulates private conduct or only dictates what governments may enact.
Coercive Funding Conditions
Because the federal government may not compel the States to enact or administer a federal regulatory program,33 Congress that wants state cooperation must offer something for it, and money is the usual offer. The Tenth Amendment reserves to the states the powers that the Constitution neither delegates to the United States nor prohibits to the states,34 yet that reservation has not stopped Congress from attaching conditions to grants. New York itself recognized that Congress may encourage a state to adopt a federally favored program by conditioning its access to federal funds.35 South Dakota v. Dole drew on Oklahoma v. Civil Service Commission, where the Court found that the United States “has no power to regulate, local political activities as such of state officials,” but “does have power to fix the terms upon which its money allotments to states shall be disbursed.”36 The state in that case could, and did, adopt “the ‘simple expedient’ of not yielding to what she urges is federal coercion.”37 A condition the state may disregard compels nothing, so funding conditions are the lawful counterpart of the commands that the commandeering cases forbid.
An offer leaves the state an exit, and the exit explains the rule. The state’s voters can weigh the money against the cost of the federal program and direct their officials to accept or refuse, so the officials remain answerable for the choice. The same logic sets the boundary, because an exit counts only if the state can take it without disaster. New York said that a choice between two commands is no choice at all,38 and a choice between compliance and ruin fares no better. Dole recognized the point when it quoted Steward Machine Co. v. Davis: in some circumstances a financial inducement might be so coercive as to pass the point at which “pressure turns into compulsion.”39
The loss in Dole was small. Congress directed only that a state desiring a minimum drinking age below 21 lose a relatively small percentage, five percent, of certain federal highway funds,40 and the Court called the argument that this was coercive “more rhetoric than fact.” South Dakota also argued that the program’s success in raising drinking ages proved it coercive. The Court answered that a conditional grant is not unconstitutional simply because it succeeds in achieving the congressional objective.41 It quoted the observation that every rebate from a tax conditioned upon conduct is in some measure a temptation, and that to hold temptation equivalent to coercion “is to plunge the law in endless difficulties.”42 Success measures how attractive the offer was; the question is whether the state could have said no. The condition was also directly related to one of the main purposes for which highway funds are spent, safe interstate travel.43 Congress stated the conditions as clearly as it could, and the state itself did not contend that the congressional action was unrelated to a national concern.44
The Medicaid expansion in National Federation of Independent Business v. Sebelius supplied the opposite pole. The Affordable Care Act required states to cover adults with incomes up to 133 percent of the poverty level, and a state that refused the new condition would lose not only the new federal money but all of its existing Medicaid funding.45 The states argued that Congress had “crossed the line distinguishing encouragement from coercion” by threatening their existing funds instead of simply refusing the new ones, and the Chief Justice agreed.46 Congress may condition funds on compliance with restrictions on their use, because that is how it ensures that the funds are spent according to its view of the general welfare; conditions that do not govern the use of the funds cannot be justified on that basis.47 Seven Justices agreed, through two separate plurality opinions: the Chief Justice’s, joined by Justices Breyer and Kagan, and a joint opinion by Justices Scalia, Kennedy, Thomas, and Alito. Both focused on the difference between inducement and compulsion.48
The Chief Justice began from the premise that Spending Clause legislation is “much in the nature of a contract,” so that its legitimacy rests on whether the state voluntarily and knowingly accepts the terms. Respecting that limit, he said, is critical to ensuring that spending legislation does not undermine the status of the states as independent sovereigns.49 Conditions that threaten to terminate other significant independent grants are properly viewed as a means of pressuring states to accept policy changes.50
He then compared the two threats directly. In Dole the loss was less than half of one percent of South Dakota’s budget, which left the state a prerogative to reject Congress’s policy “not merely in theory but in fact.”51 Medicaid, by contrast, accounted for over 20 percent of the average state’s budget, with federal funds covering 50 to 83 percent of those costs, so the threatened loss of over 10 percent of a state’s overall budget was “a gun to the head” and “economic dragooning” that left states “no real option but to acquiesce.”52 Size separates the two threats, and so do the kind of money at risk and the kind of program being changed. The highway condition in Dole took part of the very funds that serve safe interstate travel, the goal the condition addressed. The Medicaid threat put existing grants at risk to force a new program, which the Chief Justice described as a shift in kind, since Medicaid was no longer a program for the neediest and had become part of a comprehensive national plan for universal coverage.53 Justice Ginsburg, joined by Justice Sotomayor, answered that Medicaid as amended is a single program with a constant aim, that the statute warned of changes, and that states have no entitlement to Medicaid funds, only the opportunity to accept them on Congress’s terms.54
Neither opinion in NFIB fixed the line between persuasion and coercion. The Chief Justice observed that Steward Machine had not tried to fix the outermost line either, and he found it enough that, wherever the line lies, the Medicaid expansion was surely beyond it.55 The joint opinion would apply the anticoercion rule deferentially, finding a violation only where the coercive nature of the offer is unmistakably clear, and it still had no doubt that the expansion qualified.56 The dissenters objected that the inquiry is unworkable, asking which fraction of which state’s budget should count when twenty-six states with different fiscal situations had sued.57 Lower courts have filled the gap case by case. The Fifth Circuit in Gruver v. Louisiana Board of Supervisors treated the threatened loss of just under 10 percent of a state university’s funding as closer to the mild encouragement in Dole than to a gun to the head, in part because the condition was one addition to a longstanding set of Title IX funding rules and not a brand-new program. Courts rejecting challenges to a refugee-coverage rule reasoned that “the amount of the loss of funding is only one element of the coercion test.”58 The key point, they said, was that Congress, unlike in the ACA, was not “using [the] program’s funding as leverage to force the states to accept [a] new condition.”59
The doctrine matters for state immunity because it governs how Congress may obtain the cooperation it cannot command. Justice O’Connor, concurring in Printz, noted that Congress remained free to continue the interim background-check program on a contractual basis with the states, as it does with other federal programs, and she cited the conditioning of highway-safety funds on compliance with federal requirements.60 The same sentence marks the limit. A contract of that kind is an offer the state may refuse, and an offer that no state can afford to refuse works like the order the Constitution forbids. NFIB was the first time the Court held a condition on federal funds unconstitutionally coercive.
Funding threats against so-called sanctuary jurisdictions show the doctrine in current dispute. Federal law cannot compel state and local officials to enforce federal immigration law,61 but federal statutes may authorize willing officials to carry out federal immigration laws in designated circumstances.62 Together, the two propositions mean that Congress may enlist cooperation but may not conscript it.63 A 2017 executive order provided that jurisdictions identified as sanctuary cities would not receive federal grants, and the argument against the order is that Dole and NFIB bar the federal government from withdrawing funding wholesale to make local officers enforce federal immigration policy. The Ninth Circuit upheld an injunction against the order on separation-of-powers grounds, because the administration had not shown that Congress authorized the withdrawal of grant money. An exam answer should therefore present the coercion theory as an argument and not as the ground the court adopted.
Two errors are common on this topic. One is to say that the Tenth Amendment forbids conditions on funds, which Dole rejected. The other is to call a condition coercive merely because the state needs the money or because states complied, when Dole held that neither fact turns pressure into compulsion. This chapter covers only the part of spending doctrine that bears on state immunity. Dole’s other limits, that Congress spend for the general welfare, state its conditions unambiguously, tie them to the federal interest in the funded program, and avoid any independent constitutional bar, belong to the full treatment of the spending power alongside the coercion limit.
State Courts and Federal Law
The Tenth Amendment reserves to the states the powers that the Constitution neither delegates to the United States nor prohibits to the states,65 and the anticommandeering cases read the Constitution as delegating no power to command state legislatures or executive officers. State courts stand differently, because the Constitution speaks to their judges directly. The Supremacy Clause declares that the laws of the United States are “the supreme Law of the Land; and the Judges in every State shall be bound thereby.”66 A federal claim that a state court has jurisdiction to hear therefore arrives with a binding rule of decision, whatever the state’s own preferences about the subject.
The objection in New York was that the Act commandeered “the legislative processes of the States by directly compelling them to enact and enforce a federal regulatory program.”67 A court that hears a federal claim does neither. It applies law written by another sovereign to the case before it.
Printz explained why the Constitution treats judges differently from other state officers. Under the Madisonian Compromise, Article III created only a Supreme Court and made lower federal courts optional, even though the Supreme Court alone could not hear every federal case, so the plan assumed that state courts would hear federal claims. The Supremacy Clause then bound the judges in every state. And courts, unlike legislatures and executives, apply the law of other sovereigns all the time; the principle underlying so-called transitory causes of action was that laws operating elsewhere create obligations in justice that courts of the forum state will enforce.68 The reasoning has a functional core. A judge who applies a federal rule decides who wins a particular case under law that another sovereign wrote, which is what a court already does with a sister state’s law. A legislator or executive officer asked to administer a federal program instead makes and carries out state policy and, as the earlier sections showed, absorbs the political blame and the budget cost of a program Congress designed.
Testa v. Katt applied the principle to a federal statutory claim. Congress had given private litigants a claim under the Emergency Price Control Act of 1942, and the Court unanimously held that state courts of appropriate jurisdiction must adjudicate such claims however crowded their dockets are with state-law matters.69 A state court that hears analogous state-law claims cannot turn away the federal one because the state disagrees with the federal policy behind it. The Printz majority read Testa as standing for the proposition that state courts cannot refuse to apply federal law, a conclusion mandated by the terms of the Supremacy Clause, and it added that this says nothing about whether state executive officers must administer federal law.70 New York drew the same line when it explained that the text of the Supremacy Clause supports federal direction of state judges without permitting Congress to order state legislatures to comply with federal regulatory commands.71
The Printz dissent doubted that the distinction carries weight. It faulted the majority for giving an incomplete explanation of Testa and demeaning its importance, because requiring state courts to adjudicate federal claims however crowded their dockets are is a much greater imposition on state sovereignty than the majority’s description of the case as merely holding that state courts cannot refuse to apply federal law.72 It also called “perverse” the idea that the Framers would have had no reluctance to press state judges into federal service against their will but would have regarded a similar, indeed far lesser, burden on town constables as an intolerable affront to state sovereignty.73 The majority’s view controls. The settled ground is narrow, because both liberals and conservatives agree that state justices must enforce federal laws when adjudicating lawsuits.74 Whether Congress may also force a state court to open its doors to a claim the court would rather not entertain remains open, as Haywood left it.
Even where a state court must hear a federal claim, it may generally use its own procedures, provided it applies them without discrimination to state and federal cases and they do not impinge on the federal rights being asserted. The nondiscrimination half keeps a state from singling out federal claims for worse treatment. The nonimpairment half addresses a rule that treats every claim alike and still defeats the federal right. The converse problem, what law governs in a federal court sitting in a case not based on a specific federal law, most commonly in diversity cases, is the Erie problem.75 It goes to the heart of the relationship between the federal and state governments, and advanced courses in conflicts or federal jurisdiction study it intensively.76
Haywood v. Drown shows the second half at work. New York withdrew its courts’ jurisdiction over damages claims against corrections officers, whether based on state or federal law, and left claimants to a court of claims that could not award injunctive relief, attorney’s fees, or punitive damages, all familiar parts of section 1983 litigation. The state’s highest court upheld the scheme as a neutral rule. The Supreme Court reversed, treating the rule as an attempt to nullify federal law in pursuit of a local policy preference and observing that the absence of discrimination cannot alone save a state law.77 The Court did not decide whether Congress may compel a state court to provide a forum at all, since New York’s courts of general jurisdiction routinely hear section 1983 suits against state officials outside the prison setting.78 Whether the state itself may be made a defendant in its own courts is a different question, one of sovereign immunity from suit, and it lies outside this chapter.
An exam answer that carries the Printz rule over to a federal claim brought in state court has asked about the wrong actor. The question is whether the federal provision addresses a court deciding a case or an officer or legislature making and carrying out state policy. A direction to the court is analyzed under the Supremacy Clause and the nondiscrimination principle, and a direction to the officer or legislature is a commandeering problem.