Our [National] Federalism
abstract. “National Federalism” best describes the modern allocation of state and federal power, but it is a federalism without doctrine. Federalism today comes primarily from Congress—through its decisions to give states prominent roles in federal schemes and so to ensure the states’ continuing relevance in the statutory era. As a result, many of the most significant state sovereign acts now occur through state implementation of federal statutory law, but we have no law to effectuate this account of state authority. This is National Federalism: nationalism and federalism, simultaneous and in tension—and generated entirely by federal statutes. Unlike traditional federalism, it is neither a constant presence nor an entitlement: rather, it is a feature of federal statutory design. But nor does it have the usual trappings of nationalism, because it incorporates experimentation, variety and state historical expertise—the classic “federalism” values—into national law. State sovereignty remains, even if law does not yet recognize it as such. States pass state legislation, appoint new state officials and hear state-law cases in state courts, all as part of their work to implement federal statutory law, but in many ways autonomous from it. Yet, instead of having Chevron-like doctrines that give implementing states more policymaking discretion; or jurisdictional rules that keep more of these cases in state courts; or choice-of-law regimes requiring that state standards of review and state rules of administrative procedure should apply to the state laws enacted by states legislatures that shape the local implementation of federal law in ways unique to each state—instead of all of that, we have a doctrinal muddle and a Court that does not even see these questions as federalism questions in the first place. This essay develops the account of Congress as our primary source of federalism, and re-situates nationalism within that account. It then assembles a list of fifteen unresolved doctrinal questions that reveal the complexity and importance of federalism’s modern statutory domain.
author. Associate Professor of Law, Yale Law School. Special thanks to Heather Gerken for organizing this Feature and for her indefatigable support. Thanks also to Henry Monaghan, Judith Resnik, Ernie Young, Bruce Ackerman, Jon Michaels, the editors of the Yale Law Journal, my co-contributors to this Feature, and Katie Hall, Michael Sparer and the Columbia Mailman School of Public Health, for providing peaceful office space. For research assistance, I am indebted to Emily Barnet, Whitney Leonard, Cynthia Liao, Kathryn Madison, Kathryn Mammel, Kirstin McGuire, Nick McLean, Matthew Reed, Emily Rock, David Simins, and Josh Weiss.
There is a simultaneity of nationalist and state-centered impulses in almost every aspect of modern American federal law. But we do not have the theories to recognize it, or the legal doctrines to effectuate it. Federal law is now predominantly statutory law, and the reach of federal statutes into areas of historic state control continues to expand. But this “federal” law has an unmistakably state-centered component: With almost every national statutory step, Congress gives states new governing opportunities or incorporates aspects of state law—displacing state authority with one hand and giving it back with the other.
Federalists should pay attention: In the post-New Deal Era, this role for the states within federal legislation is a primary vehicle through which states have influence on major questions of policy, and through which state sovereign powers retain their relevance, albeit in ways different from those contemplated by the traditional account. Current doctrine is not at all keyed in to the ways in which a very great deal of state sovereign power—including state lawmaking and state-court jurisdiction—is exercised as part of federal statutory implementation, and so current doctrine does nothing to protect or effectuate that state authority. It is not that states do not retain relevance at the local level. But when it comes to most major policy questions, Our Nationalism has become a critical generator of Our Federalism.1
Federalism also is a key ingredient in Our Nationalism. The modern federal regulatory apparatus is increasingly attendant to questions of the state-federal allocation of responsibility, and also is dependent on state actors, in ways both practical and political. State implementation of federal statutory law and the incorporation of state law within federal statutory schemes are allocation-of-power strategies used by Congress to make federal legislation more effective; but they also restrain the breadth of national control and make legislation more politically palatable. There is something different about national statutory schemes when states have the primary policy and lawmaking roles—something this Essay argues is often, indeed, “federalism.”
This push-pull of nation and state—both from inside the landscape of federal statutes—is more than just an interesting theoretical observation. It is a “law” problem. When it comes to legal doctrines to deal with this new world of statutory federalism, ours is a sorry state of affairs. Modern state-federal relationships have given rise to many new and difficult legal questions— ranging from those of state-versus-federal-court jurisdiction to matters of administrative deference, statutory enforcement, and standards of review. Such questions have split the lower courts, have yet to be resolved by the U.S. Supreme Court, and are affecting how major federal laws are being carried out across the country. Half the time, the courts do not even recognize these questions as federalism questions, even though they unquestionably concern the discretion, influence, and sovereignty of states in a national legal landscape. Robert Schapiro wrote a decade ago that modern federalism lacks “rules of engagement.”2 We are still muddling through.
This essay makes two principal claims, both intended to provoke discussion. The first is about modern federalism’s primary domain and its source: federalism now comes from federal statutes. It is “National Federalism”—statutory federalism, or “intrastatutory” federalism, as I have called it in the past.3 One reason for the lack of developed doctrines is the resistance to recognizing that this is where modern federalism comes from and where its primary battlegrounds lie. Courts and scholars for decades have acknowledged the prevalence of “cooperative federalism,” which of course is often generated by overarching federal statutory schemes.4 Even some traditional federalists have come to recognize the state power to be gained from this interactive, rather than “separate spheres,” model of state-federal relations.5 But even these expansive inquires have not grappled with the perhaps startling conclusion that follows from recognizing that states today may exert their greatest powers from within these federal statutory endeavors: namely, that this federalism’s primary source is Congress.
Federalism today is something that mostly comes—and goes—at Congress’s pleasure. It is a question, and feature, of federal statutory design. Distinct from the dominant conceptions of federalism and state power, this federalism is neither a constant presence nor an entitlement. It looks different and has various levels of strength across a wide continuum of statutory schemes. But it has important parallels to the federalism of the past, particularly judicial federalism. Just as federal judges once reached for the state common law to fill the interstices of federal lawand to prevent the aggrandizement of lawmaking by the federal judiciary,6 Congress today reaches for the states to restrain the breadth of federal law and to bring the states’ expertise, variety, traditional authority, and sovereign lawmaking apparatus into federal statutes. Similarly parallel, the Erie questions of our time are not, as they once were, about the choice between state and federal common law but, rather, about how to choose between aspects of state and federal regulatory regimes. The critical choices between state and federal law today concern what rules of statutory interpretation, what standards of review, what administrative-law doctrines, and what other doctrines of statutory law federal courts should apply when they are interpreting state statutes, regardless of whether those state statutes stand alone or are the product of state efforts to implement federal legislation.
Is this federalism? Is this nationalism? It is both. The motivations are simultaneous and in tension. It is a nationalism that often lacks nationalism’s defining theoretical feature—uniformity—and so presses us to ask what “Our Nationalism” is all about, a question that has received scant theoretical attention. It is also a nationalism that incorporates values, like experimentation and local variation, that are traditionally associated with federalism. We have seen this before, in a different form: Paul Mishkin famously described the “variousness” of judge-made federal law.7 National Federalism recognizes that kind of state-oriented legal diversity in the federal statutes of the modern era.
Similarly, this federalism lacks the traditional appearances of federalism’s defining feature: sovereignty. And it will discomfit some, because this federalism leaves state power to the grace of Congress. Indeed, in some ways, this is the ultimate instantiation of Herbert Wechsler’s classic theory of the “political safeguards of federalism.”8 Wechsler argued that courts need not police federalism doctrine because the states are adequately represented in Congress.9 National Federalism goes further, embracing Congress as federalism’s primary source and viewing Congress as having as much, if not more, of a role to play in shaping federalism as do the courts.
But, to be clear, National Federalism is not a federalism shorn of state sovereignty. It is true that National Federalism emerges through congressional displacement of state law with a new, overarching federal statutory scheme. But this federalism depends on, and strengthens, the states’ continuing sovereign status in important ways that have yet to be recognized.10 When Congress calls on states to implement federal law, states act in their sovereign capacities to do so: They pass new state laws and regulations, create new state institutions, appoint state officials, disburse state funds, and hear cases in state courts—some cases, as I shall illustrate, that have been determined to be hearable only in state courts. It is true that this state action is not wholly separate from federal law; it is shaped by the federal statutes and states often need permission from the federal government to begin a course of federal statutory implementation. But that does not change the fact that, after such approval, the states’ sovereign apparatus acts in ways that are often indistinguishable from the kind of autonomy we see in exclusively state- law domains.
My second claim is about National Federalism’s lack of doctrine. This is a world of federalism-meets-statutory-law, but the doctrines of both federalism and federal legislation, as currently conceived, are unequipped for it. Non-dualist models of federalism have always suffered from a “wishy-washiness” problem when it comes to law—a problem that separate-spheres federalism, which does have some well-defined doctrines, like Commerce Clause doctrine, has not faced nearly to the same extent.11 Part of the reason is that the vast expanse of writing about interactive federalism mostly has been devoted to functional inquiries about the merits of state-federal interconnectedness, or descriptive efforts illustrating those connections in particular subject-matter areas.12 But alongside this important work, little attempt has been made to generate “law” effectuating the relationships being described.13
This is a problem that goes much deeper than the most recent example of it—the Supreme Court’s disappointing declination, in the 2012 health reform case, to devise a real law of federal-state coercion for Spending Clause legislation.14 It extends, for instance, into the important terrain of federal-state administrative relationships, where we have no doctrines that address whether state implementers of federal law receive any interpretive deference or any “process” when it comes to their interactions with federal agencies. It extends to judicial-power doctrines too, revealing gaping omissions in our laws of federal court jurisdiction and choice of law. One prominent puzzle is the lack of doctrines to determine whether the state laws and regulations that states enact to implement federal statutory schemes (for example, a state’s Medicaid Program, or a Clean Air Act State Implementation Plan) have the status of “federal” or “state” law. Resolving that puzzle is essential to answering other questions about whether state or federal courts have jurisdiction to review cases involving those programs, and if so, whether state or federal substantive law, such as their respective standards of review, is to be applied to them.
Federalism and statutory-law doctrines are equally to blame. On the federalism side, many judges and theorists do not even see most of these as questions to be answered, much less see them as questions of federalism. On the statutory-law side, legislation theory and doctrine have long suffered from a federal-law myopia—a resistance to seeing anyone other than federal actors as the creators and interpreters of federal statutory law—that has left us bereft of interpretive rules that could address these National Federalism relationships.15 The few federalism-related statutory interpretation presumptions that we do have, moreover, assume state-federal separation, not integration. They also act as one-way ratchets and so are unable to accommodate the idea of a continuum of federalist and nationalist relationships that differ across statutory schemes.
Perhaps most importantly, even if statutory interpretation doctrines could be tailored to recognize these interactive relationships, the legislation doctrines currently lack the teeth to bear the primary burden of effectuating our modern federalism. Congress barely knows the statutory interpretation rules,16 and so any idea that interpretive presumptions can do the work of protecting federalism in the legislative process—an idea embraced by some of the so-called “process federalists”—is a fantasy. The courts, for their part, do not apply the interpretive rules consistently or even treat them as precedential “law” in the same way that they treat other decision-making regimes (including the implementation doctrines of the Constitution17)—a practice that has left the legal status of the statutory interpretation doctrines remarkably unclear, as I have previously detailed.18 But a federalism that depends on federal statutory design is a federalism that turns on questions of congressional intent. It is a federalism with a key interpretive dimension, and requires real doctrines of statutory law.
The discussion that follows has three main lines of argument. Parts I and II develop the account of Congress as our primary source of federalism, using recent cases and other examples to substantiate the centrality of federalism’s statutory domain. Part III examines nationalism through this lens. Part IV assembles, and begins to frame answers to, fifteen unresolved doctrinal questions at this intersection of federalism and statutory law.19 Taken together, thequestions reveal the complexity of this domain and the want of coherent legal doctrines to guide this modern expression of our state-federal relationship.
This Part begins with the story of an old statute that tells us a great deal about modern federalism (and its pedigree) but receives virtually no attention from federalism scholars. Insurance law was once considered exclusively local law. That changed in 1944, when the Supreme Court held that insurance had become interstate commerce and so Congress could regulate it.20 Congress, however, turned around and immediately gave that power back to the states by federal statute, in the McCarran-Ferguson Act of 1945.21 That statute declared that the “continued regulation and taxation by the several States of the business of insurance is in the public interest” and announced a new default rule of statutory construction that “silence on the part of the Congress shall not be construed to impose any barrier to the regulation or taxation of such business by the several States.”22 Its enactment rested on all of the traditional federalism reasons, including the historic state control over insurance and the value of local variation.
Since its enactment, McCarran-Ferguson has been consistently invoked by the Court to allow states to do things they could not normally do, like violate the Dormant Commerce Clause, when it comes to insurance regulation. McCarran-Ferguson also has enabled the development of a pervasive and varied web of state insurance law that everyone conceptualizes as state insurance law. That’s federalism by the grace of Congress.
Fast forward to today. Members of the Supreme Court used the word “federalism” in twenty-six cases over the past three completed Terms.23 Twenty-one of those cases were statutory federalism cases. This quick exercise in counting should be evidence aplenty that National Federalism provides the terrain on which modern federalism’s most salient issues are playing out. Those twenty-one cases involved either state implementation of federal law; or a federal statutory scheme that incorporated reference to, or deferred to, state law or procedures; or other cases in which there was no question about congressional power to regulate in a field of traditional state control and the only issue was the interpretive question of how far Congress intended a particular federal statutory provision to go. Two additional cases concerned the Spending Clause—the primary legislative power Congress has used, after Printz v. United States,24 to offer states the option to implement federal law— and those cases presented quintessential questions about state-federal relations inside federal statutes. Eight more cases mentioned the word “sovereignty,” but only one of those cases involved anything like “separate spheres” sovereignty; the rest were about the states’ role within federal statutory schemes, or the clarity with which Congress used its acknowledged power to displace state law.25 Twelve more were statutory interpretation cases about preemption (that did not also mention federalism).26The word “nationalism” never came up.27
Qualitative examples paint the same picture of where we now see “federalism” and how it is that states continue to have national-policy influence in a federal statutory age.For instance, I have previously detailed how state experimentation—the most commonly-touted benefit of federalism—has arguably been better effectuated from states implementing federal statutory schemes than from them acting alone.28 The state of Massachusetts was the template for health reform not because Massachusetts acted as a sovereign “state as laboratory,”29 but rather because Massachusetts’s health reform experiment was made possible by a Bush Administration waiver that allowed Massachusetts to be creative in implementing the federal Medicaid program.30 Many other key policy experiments of the modern era, including the state air-quality innovations accomplished through the federal Clean Air Act,31 have been conducted in this fashion.
And with respect to state leverage, as another example, the states today that have extracted long-wanted concessions from the federal government to privatize Medicaid are getting their way not by insisting on separation of state and federal but, rather, by exerting their powers fromthe inside, as Medicaid administrators. There is a reason that states do not influence Medicare policy, but do influence Medicaid policy: Congress choseto design only one of those two parallel programs with states at the forefront.
This is not an argument with political priors. Prominent federalism scholar Ernest Young argued years ago that the states’ real power flows more from within these federal schemes than from enforcing areas of exclusive state authority.32 The conservative economist Douglas Holtz-Eakin recently made the same argument in the health reform context.33 The reason, as Young put it, is that federal statutory law has gone so far into the terrain of regulating the everyday affairs of the citizenry—from health, to telecommunications, to the environment—that the only way to “ensur[e] that states retain something meaningful to do”—is to empower them from within national law.34
The Court, however, resists this account. This resistance was on prominent display in the health reform case, National Federation of Independent Business (NFIB) v. Sebelius,35 in which seven Justices refused to acknowledge the federalism inside federal statutes. The four joint dissenters expressly disputed the proposition that Congress’s decision to allow “state employees to implement a federal program is more respectful of federalism than using federal workers” alone, and asserted that “[t]his argument reflects a view of federalism that our cases have rejected.”36 Three other Justices, speaking through Chief Justice Roberts, likewise focused only on the “independent power of the states.”37 But the Court’s arguments evinced a wish more than reality. The Court unrealistically assumed that erecting barriers to state implementation of federal law will stop Congress from enacting major federal legislation altogether. The New Deal, however, is here to stay. The question is not whether we will have major federal statutes but what the continuing relevance of the states in this landscape will be.
Even more importantly, there are voices beyond the Court’s that are shaping modern federalism. This is a significant development because the Court traditionally has been viewed as the primary arbiter of what federalism is and how it is protected. But Congress is now in the game and recognizes that the modern regulatory state has changed how federalism is generated. Often with explicit references to “federalism,” Congress has dealt with the massive expansion of federal power in part by giving some substantive lawmaking power back to the states through federal legislation. Long before health reform—from the family and old-age assistance programs of the New Deal era, to the environmental statutes of the 1970s, to the recent financial reform legislation38—Congress has incorporated state law into federal statutes and asked state actors to serve as frontline federal-law implementers.39
What is more, as part of Congress’s efforts to give the states substantive lawmaking roles in national schemes, Congress has asked the states to enact their own state laws, create new state institutions, and pass new state administrative regulations—in other words, to exercise their sovereign powers in service of the national statutory project.40 These congressionally generated opportunities have played a central but unappreciated part in both state identity and how state power is understood on the ground.41
Republican governors, for example, have argued for state administration of federal health reform as a mechanism for retaining state power to regulate insurance markets.42 A number of lower courts have held that the state laws implementing the federal Clean Air and Water Acts are just that—state law, not federal in nature.43 State laws implementing the federal Medicaid statute typically are not called “Medicaid,” but rather “TennCare” in Tennessee and “Husky Health” in Connecticut—evidence of their state-centered identities. This expressive dimension of federalism should not be overlooked; it matters how people experience these laws.44
Congress also goes beyond state administration of federal law. Congress’s incorporation of state law into federal statutes—moves likewise often made while invoking “federalism” or a desire to restrain the reach of national law—is another species of National Federalism that has gone almost entirely unrecognized.45 Examples include the provision of the Social Security Act that defines “child” by reference to the state-law definition of it,46 the provision of the Fair Labor Standards Act that defines “school” in accordance with the state-law definition,47 the provision of the Travel Act that defines “unlawful activity” to include any prostitution, extortion, bribery or arson offenses “in violation of the laws of the State in which they are committed,”48 and the Assimilative Crimes Act, which authorizes the use of state law in federal enclaves when federal law is silent.49
Seeing these incorporative statutes through the eyes of federalism unmasks some obvious linkages between these efforts and those in which Congress puts the states on the frontlines of federal statutory implementation. In particular, Congress may have the same motivations for state-law incorporation as it does for utilizing the states as implementers of federal law: Congress can draw on state expertise by taking well-developed bodies of state statutory or common law on the subject and incorporating them by reference into the new federal statute. Similarly, preserving this state role—whether though state implementation or state-law incorporation—allows for local policy variation within the confines of a new federal statute and in some ways limits the national reach.
I belabor this point about federalism’s statutory domain because some continue to resist it.50 The needed doctrines will not emerge, however, until we can establish the terrain on which federalism’s main questions are playing out. Very occasionally, there are glimmers in Supreme Court opinions of this recognition. Justice Breyer, for example, in an uncontroversial ERISA case that few federalism aficionados are likely to have noticed,51 observed that, “in today’s world, . . . the true test of federalist principle may lie, not in the occasional constitutional effort to trim Congress’ commerce power at its edges, . . . but rather in those many statutory cases where courts interpret the mass of technical detail that is the ordinary diet of the law.”52 Justice Kennedy has offered similar hints. In a 2004 case involving a power struggle between state and federal agencies over implementation of the Clean Air Act, Kennedy dissented to propose a new doctrine that would give states more discretion in implementing federal law, even at the expense of the traditionally privileged discretion of federal agencies, calling such a doctrine essential to “cooperative federalism.”53
Justice Scalia likewise has noticed the shift, but resists the idea that federalism is effectuated by it. In a well-known state-federal dispute over implementation of the Telecommunications Act of 1996, Justice Scalia’s opinion for the Court argued that “preemption” was not the issue in ruling for the federal agency. “[T]he question in these cases,” he wrote, “is not whether the Federal Government has taken the regulation of local telecommunications competition away from the States. . . . [I]t unquestionably has. The question is whether the state commissions’ participation in the administration of the new federal regime is to be guided by federal-agency regulations.”54 In a case decided last year, however, he called those matters—questions about the states’ discretion in implementing federal statutes and state implementers’ relationships to federal agencies—“faux-federalism.”55
To be sure, this federalism seems shorn of federalism’s defining features—state sovereignty and clear constitutionally protected domains. It is a federalism whose subject-matter areas are for Congress’s choosing, and from which Congress does not evenly choose. It is a federalism that puts enormous pressure on the famous “political safeguards” concept, because all the power rests in Congress’s hands. The increasing problem of congressional gridlock offers another twist: What does it mean to have a federalism entrusted to congressional action if Congress itself acts rarely and with growing difficulty and partisanship?56
The idea is not to take a rose-colored view. I do not dispute—indeed, I emphatically agree—that National Federalism does not always empower states (an inquiry complicated by the fact that “state interests” are not uniform57), and that it usually serves to aggrandize national power. I return to those concerns below, but for now the point is that the national impulses in these federal statutes do not mean that the state-centered impulses are not also present.
As it stands, our doctrines of federalism and statutory law are not tuned in to the ways in which states exercise their sovereign powers in the modern federal statutory era. One can envision doctrines designed to elevate this aspect of the states’ role. For example, we might have doctrines that give implementing states more policymaking discretion; or exclusive state-court jurisdiction;58 or provide that state-law standards of review and administrative procedure, rather than federal standards, should apply to disputes over state laws and regulations that implement federal statutes.
The final Part of this essay is devoted to these gaps in the doctrine. Part IV sets out a list of fifteen unresolved doctrinal questions related to National Federalism that are percolating in the courts, and that might be answered in ways that emphasize the states’ ownership of much of this federal statutory domain. It is a separate question whether we should empower states in this way, or whether Congress would want to, if asked. This essay does not fully take on those questions, although I will offer some evidence of Congress’s preferences. My goal is different: It is to illustrate, for those focused on state sovereignty or those already convinced of the benefits of federalism, the strong case that can be made for recognizing the state power within these national schemes and exploring the kinds of doctrines that would better effectuate it.
The current federalism doctrines, it is true, have not entirely ignored these questions. But the Court has put almost all of its energy into one particular exercise of state power within federal statutory schemes; the choice by states whether to participate in Congress’s conditional spending programs in the first place. As elaborated below, that set of doctrines is a mess. The Court has refused to draw clear lines to demarcate when such programs are unconstitutionally coercive.59 It also has used as its overarching theoretical framework for these questions the idea that states and the federal government are in a “contractual” relationship. But the Court has inconsistently deployed that framework and not deeply considered its implications for broader questions about the relationship.60
Even if that set of doctrines were clear, placing all of the doctrinal emphasis on the question of the states’ free choice to participate in federal programs gives short shrift to the intense political and sovereign dynamics that play out when states deliberate over federal implementation. For example, will a state cabinet position need to be created? Will state officials need to be authorized with new powers? Such an emphasis also underplays the importance of the many kinds of state-federal interactions that occur even before the federal statute is enacted, as part of the federal statutory-design process. As others have detailed, during that process, states—often acting in groups, through horizontal federalist entities such as the National Governors Association—use their leverage as would-be implementers to put their imprint on how the ultimate federal policy will look.61
The usual alternative to state participation in these federal programs, moreover, is not “sovereignty.” The federal government will step in to operate the programs for the states. If states decline to implement the Clean Air Act, states do not retain authority to shape that aspect of environmental law. The federal government will come in and implement that same federal law—but in accordance with federal, not state, policy preferences.
Finally, it should be stated that our doctrines of statutory interpretation have not completely ignored federalism either. But the few statutory-law doctrines that do train on federalism—the federalism “clear statement rules” and presumptions about federalism and preemption—are dinosaurs from the age of the separate spheres. Those presumptions, as Justice Scalia’s opinion in the Telecommunications Act case observed, are understood as obstacles to the initial federal displacement of state law—default rules requiring ambiguous statutes to be construed not to preempt state law.62 They are not generally used (although they may potentially be) to manage state-federal relationships once state law has unquestionably been displaced and both state and federal actors have interpretive authority within the same federal statute. The other relevant doctrines of statutory interpretation—those concerning agency implementation of federal law—apply only to federal actors, a weakness that I have detailed elsewhere.63 Those federal administrative law doctrines treat state implementers as if they do not exist.
A brief excursion into the cases in which the Court explicitly invoked “federalism” or state “sovereignty” during the last three Terms establishes the centrality of federalism’s statutory domain and some of its unresolved doctrinal questions. The end of this Part introduces an additional form of National Federalism not reflected in these cases and that the Court does not recognize. I refer here to federal judicial interpretation of state statutory schemes, the modern-day instantiation of the Erie doctrine and a central aspect of federal judicial review of state implementation of federal law.
Of the twenty-six times that members of the Court invoked “federalism” over the last three Terms, six occurred in ordinary statutory interpretation/preemption cases. Preemption cases do not raise questions about Congress’s power to legislate over state terrain, or even about its power to legislate on the particular subject at hand; they merely raise questions about how clearly Congress speaks to the particular issue in question.64 In other words, they are questions of federal statutory design. This category of six cases includes Arizona v. United States, the high-salience “federalism” challenge to Arizona’s immigration law.65
Other cases invoked federalism in less familiar ways. For example, in United States v. Windsor,66 the challenge to the federal Defense of Marriage Act, the Court invoked “federalism” not as an impermeable barrier to congressional legislation in the historic state terrain of domestic relations but as something of a resistance norm—a feature requiring special consideration when judging Congress’s intervention.67 (Justice Scalia called this “amorphous federalism” in his dissent.68)Another was City of Arlington v. FCC, the telecommunications case discussed above in which Justice Scalia’s opinion for the Court, crying “faux federalism,” rejected the idea that federalism comes into play in questions about the division of labor between state and federal agencies implementing the same federal statute.69
Another high-salience case, Shelby County v. Holder, concerned the special federal preclearance requirements applicable to only certain states under the Voting Rights Act.70 The Court repeatedly used the term “sovereignty,” but not to dispute the power of the federal government to interfere with the states’ control over their own elections. Rather, the Court used the term to emphasize that “all States enjoy equal sovereignty”—apparently within the confines of federal law.71 Federal intrusion was not the main problem; the fact that it applied unequally (without justification, in the Court’s view) to various states was. Another case, J. McIntyre Machinery, Ltd. v. Nicastro, was the first major case in decades to consider a state’s personal jurisdiction over an international defendant.72 The plurality went out of its way to surmise that Congress could authorize nationwide jurisdiction to modernize the current state-power-oriented landscape of personal jurisdiction.73
AT&T Mobility LLC v. Concepcion was a statutory preemption case, too, but is worth singling out because that case involved the construction of the Federal Arbitration Act’s “savings clause.”74Savings clauses are explicit, exceedingly common, and typically overlooked examples of National Federalism: Congress singles out particular aspects of state law to survive or interact with federal law within a new overarching federal statutory scheme.75 In Concepcion, Justice Breyer’s dissent seized on the FAA’s savings clause, invoked the idea that “states are sovereigns,” and argued that by using the clause, Congress embraced a “federalist ideal” and “reiterated a basic federal idea that has long informed the nature of this Nation’s laws.”76
Another nine of the twenty-six “federalism” invocations, plus two more mentions only of “sovereignty,” were habeas cases.77 This is worth pausing over, because even though habeas was the primary example used in Robert Cover and Alexander Aleinikoff’s seminal work on “dialectical federalism,”78 one rarely sees habeas integrated into modern-day discussions of cooperative federalism.79 Particularly relevant is that almost all of these recent habeas cases have a strong National Federalism component: the interplay between state and federal law in many of these cases is a matter of federal statutory design, through Congress’s choice in the Antiterrorism and Effective Death Penalty Act of 1996 (AEDPA) to, in the Court’s words, “promot[e] comity, finality, and federalism” by building deference to state procedures into the federal statute.80
Other cases raised, but did not resolve, new federalism questions. Wos v. E.M.A.,81 for example, raised the question of whose job it is—the state’s, the federal agency’s, or the Court’s—to fill gaps in cooperative statutory schemes when the statute (there, Medicaid) is silent. The majority, through Justice Alito, decided the statutory question itself, and did not defer to either state or federal implementers.82 Justice Breyer concurred specially to emphasize that the federal agency should make that decision, even though the federal agency actions in the case were informal and the state itself had relied on previous, contradictory, informal federal guidance.83 Chief Justice Roberts dissented for three Justices, called Medicaid a “state program,” and concluded that “the whole point of our federal system is that different States may reach different judgments about how to run their own different programs.”84
In another case, Virginia Office of Protection & Advocacy v. Stewart,85 the Court raised but did not answer the question of Congress’s power to “affect the internal operations of a State,”86 or to give state actors power they would not otherwise have under state law.87 This same question has been raised—but likewise not yet answered—in the context of the health reform legislation, which directs the states to enforce new insurance provisions, even though some state laws do not already give that power to state officials.88 It remains unresolved whether federal law alone can give state-created entities authority that they do not possess under state law.89
The enforcement of National Federalism programs posed different problems for the Court. Douglas v. Independent Living Center raised the question whether California citizens could challenge their state’s implementation of the federal Medicaid statute when the federal agency itself had not chosen to challenge the state’s action.90 The Court focused on the federal agency’s actions, and implied (but did not definitively conclude) that federal agency approval (or inaction) with respect to state implementation would typically shield states from these kinds of challenges by their own citizens, even where the federal agency does not exercise robust oversight.
Indulge a final pair of examples: the two Spending Clause cases that the Court decided during this period—the health reform case, NFIB v. Sebelius,91 and a sovereign immunity case, Sossamon v. Texas.92 Much could be said about these cases, but for present purposes, the point is to highlight the inadequacy of the doctrines invoked by the Court, particularly its use of the “contract” metaphor to describe the federalism relationships created by Spending Clause statutes. The question in the health reform case was the typical National Federalism question of whether the way in which Congress exercised its undisputed power to expand a federal program (there, Medicaid) was respectful of state “sovereignty.” Stating that Spending Clause legislation “was in the nature of a contract”93 between two sovereigns, the Court held that Congress’s Medicaid expansion was too “dramatic” to have been anticipated by the states.94 But the contract analogy in the end was of little help to devising real doctrinal rules. Instead, the Court adopted no test at all, holding: “We have no need to fix a line . . . . It is enough for today that wherever that line may be, this statute is surely beyond it.”95 Justice Ginsburg’s dissent chided the Court for failing to come up with a doctrine that could be applied in the future.96
The second case, Sossamon, presented the question whether a state’s receipt of federal funds sufficed to waive sovereign immunity for suits for money damages (there, under the Religious Land Use and Institutionalized Persons Act of 2000). In holding that Texas had not waived its immunity, the Sossamon Court backed off the contract analogy, but commented only briefly on the use of the contract framework as a linchpin in the opinions from the Fifth and Eleventh Circuits being reviewed—and the confounding way in which the federal programs there had been described. The Fifth Circuit had held that “Spending Clause legislation is not legislation in its operation; instead, it operates like a contract.” The Eleventh Circuit likewise used this distinction to treat the question of the private right of action differently from the case of “ordinary” legislation.97
There is another type of unrecognized statutory federalism—this one not generated by Congress—that comes into play when federal courts adjudicate state statutory-law questions. This set of cases is arguably the heir of Erie, the case that forms one of the cornerstones of traditional federalism doctrine. Notably, Erie itself stems from an instance of National Federalism. The Erie case involved the interpretation of the Rules of Decision Act (RDA), a federal statute that directs courts to use state law as the rule of decision unless federal law expressly directs otherwise.98 Coming at the dawn of the New Deal, however, Erie was about the choice between judge-made federal common law and the repository of state law, which at the time also was largely judge-made. Today, because statutory law dominates the state legal landscape just as it dominates the federal landscape, the Erie question has also necessarily evolved.
Today, when federal courts review state law questions (whether under their diversity jurisdiction, or as related to matters implicating their federal-question jurisdiction—including as part of reviewing state implementation of federal law) the difficulty is less in “finding” the state law than in interpreting it. As I have previously detailed, state courts have their own, unique approaches to statutory interpretation, administrative review, and other matters—some of which are different from the federal approach and often bring state-law-oriented values to bear.99
Federal courts, however, do not see in their Erie obligation an obligation also to apply the same interpretive principles to state statutes as the state courts would. When it comes to state statutes that implement federal law, federal courts likewise overlook other state decision-making regimes, such as standards of review. Nor do they see in those cases a federalism opportunity—a chance to give effect to the sovereign choices made by the states in the design of their own statutory terrain. I have made the case before why Erie should indeed be understood to apply to those interpretive questions (just as the Erie doctrine already is understood to require federal courts to apply other state-law decision-making regimes).100 Instead, the federal approach to interpreting state law has effectively been a massive exercise in federal judicial preemption: Each case results in (a likely unintended) displacement of state statutory and interpretive norms by the federal normative preferences that come from the application of the federal statutory doctrines.
On the reverse side, too, state courts bring their diverse perspectives to the task of interpreting federal statutory law. Congress has assigned a few limited areas to the federal courts’ exclusive jurisdiction but, otherwise, state courts sit with equal authority and duty as any federal appellate court to hear any federal statutory claim.101Simply because state courts hear so many more cases than federal courts, they play a key role in filling the interstices of federal statutory law. Scholars for decades have argued that, when federal courts adjudicate state-law cases, they cannot help but bring their federal-law sensibilities to that task—a fact that leads to some harmonization and nationalization of state statutory law.102 Similarly, and in reverse, it must be the case—even though it remains under the radar—that state-court federal statutory interpretation leads to a more diverse and local set of meanings about federal statutory law than would a system of exclusive federal court jurisdiction.103
Skeptics may be thinking that National Federalism is really just an attempt at national-law aggrandizement, and a deceptive one at that. The Court has repeatedly worried that these statutory schemes improperly diffuse accountability away from the federal government,104 and I myself have previously argued that these state-based schemes are powerful vehicles of subtle federal statutory entrenchment.105
But it also should be obvious that even such nationally oriented motivations have federalism within them. The idea that some members of Congress trust their home-state counterparts to administer federal law more than they trust the executive branch (particularly the executive branch of the opposing party)—an idea that has some empirical support106—depends on the notion that these administrators are not all equal. It is true that the localness of the way in which these programs are encountered complicates the concerns about accountability to which National Federalism schemes give rise. Critics may be correct that National Federalism diffuses accountability, but maybe exclusive federal accountability is undeserved; that is, perhaps the states should be held at least partially responsible for those aspects of the implementation that are, in fact, state programs. If California chooses to pay doctors less in its version of Medicaid than does New York, why not hold California accountable? With sovereignty comes responsibility.
What does all of this mean for theories of nationalism? More aptly, what is Our Nationalism? The word “federalism” comes up 610 times in the Westlaw Supreme Court case database. The world “nationalism” comes up only thirty- three times, and not once in ways that concern federal statutory law or that otherwise have any relevance to the kinds of questions posed by this essay.107 While these statistics may be unsurprising—talk of “nationalism” makes courts uncomfortable because of its strong connotations of centralization—academics rarely talk about it either.108 Whereas “federalism values” like variation and experimentation are heavily utilized concepts, we have no common theory of nationalism.
On the other hand, the word “uniformity” comes up 1,407 times in the Westlaw Supreme Court database as a justification for congressional policies or for certain types of judicial decisions. It may well be that uniformity is the value most often associated with nationalism, particularly in the context of congressional legislation. But uniformity no longer seems a useful concept to anchor theories of nationalism when many major federal statutes give states frontline roles precisely because Congress desires disuniform implementation of national law. Values like experimentation, variation, and tailoring to local circumstances are also now integral components of nationalist policy making.
There is a noteworthy parallel to draw between the way in which Congress has thus expanded national power and the way in which the federal courts did the same in an earlier era. Paul Mishkin’s famous work on the “variousness of ‘federal law’” made the case that, in filling gaps in federal statutes, the federal common law work of federal courts need not be, and in fact should not be, completely “federal” in nature.109 Drawing instead on the traditional federalism values, including local variation and the background norm of federal restraint, Mishkin argued that consideration of those values should drive federal judicial decisions about when to take state law as the rule of decision—for example, applying a state-law definition for an undefined federal statutory term.Voluntary federal judicial incorporation of state law, Mishkin argued, helped to avoid an “unwarranted intrusion into areas traditionally and properly regarded as state domain.”110 Following Mishkin, Carol Goldberg-Ambrose took this point into the realm of federal-court jurisdiction, suggesting “nationalism” reasons why Congress might wish to create federal-court jurisdiction over certain questions, but still use state law as the substantive rule of decision.111
The link to National Federalism should be clear. When Congress incorporates state law rather than creating new federal categories, or when Congress offers the states a primary federal implementation role, it is making federal law with some self-conscious restraint and building diversity into it. That restraint may be motivated by instrumental reasons—including a desire to push federal law into areas of historic state dominance—or by “federalism” reasons. Most likely it is both. Work like Mishkin’s has shown the internal state-centered diversity of federal law for some time, and National Federalism continues in that tradition. Likewise, and in connection with Goldberg-Ambrose’s work, one can see in National Federalism a motivation on the part of Congress, too, to assert some federal control over the system but, at least sometimes, to build the states into it.
It also seems evident that we sometimes have nationalism in lawmaking without Congress or federal judge-made law at all. This is a point that goes beyond the way that the states, as centers of political activity, influence public debate through their positions on federal statutes in which they have no formal role.112 John Nugent and Judith Resnik have each written about how groups of state and even translocal actors together play central roles in federal statutory politics.113 States also do sometimes still act as first-movers, performing their traditional “states as laboratories” role, in trying out controversial policies.114 Sometimes, such state innovation even creates what might be understood as a different kind of “national law”—what William Eskridge and John Ferejohn have described as an informal fifty-state convergence that makes federal legislation unnecessary.115
Other times, those state convergences take on a more formal character, for instance when one state models its laws on those of another. A striking example can be found in a slew of recent state food safety laws, which condition the effective date of the state law on the adoption of a similar law by a number of other states.116 State courts also must sometimes create federal common law.117 Another example of more formal action exists in the adoption by many states of Uniform Laws. The Uniform Commercial Code is the most prominent but only one of many such laws. These Uniform Laws exemplify how “national law”—law sometimes even more uniform than federal statutory law that depends on varied state implementation—can be created by states, without Congress.
The point is not to undersell the other ways in which states contribute to the national landscape or to minimize the continuing benefits of local governance in areas that Congress has not entered. My argument is also in some ways the opposite of arguments by scholars like Heather Gerken, whose important work views “federalism” as a means to a national end—a way of churning the system to reach an “ideal” national policy solution.118 This essay, instead, takes continuing variety and state power as the end worth preserving and aims to convince states-rights theorists that nationalism is one important means to it. Of course, Congress will sometimes shut off that state variety—straight preemption is always an option. But the alternative to National Federalism is not state autonomy; it is more Washington-controlled federal legislation.
The point is that nationalism, like federalism, now takes different forms. How “national” any federal statute is, in the uniformity/preemption sense, will vary across the U.S. Code. It is for that reason that the details of the federal statutory design—from which we can infer where on the spectrum Congress intends a particular statutory scheme to lie—must now take on greater significance.
This Part sets out fifteen unresolved doctrinal questions to which National Federalism has given rise, and which, in many cases, already are dividing the lower courts. Undoubtedly, there are other questions that could be added to the list, and this introductory exposition shortchanges the depth of analysis that any one of these questions is due. But there is value to assembling the questions in one place, as the start of a doctrinal agenda that needs to be tackled. The discussion divides the questions into four groups—focused on, respectively, state-federal regulatory interaction; the utility of the contract framework; state sovereignty; and the challenges of using statutory interpretation doctrine as the primary legal regime for this domain.
There are also normative matters that cannot be addressed here and on which lawyers will disagree—most importantly, the extent to which legal doctrine should actually try to enable state power within these statutory schemes, even if Congress so intends. This list of questions, instead, is based on three potentially controversial premises: first, that Congress sometimes does intend for states to have discretion when implementing federal statutory law—an assumption that has an empirical basis, but merits more verification;119 second, that Congress is entitled to, and should, play this role in generating modern federalism; and third, that if federalists recognize federalism’s statutory domain, they should be interested in doctrinal solutions that give greater effect to state power within federal statutory schemes.
A. Questions Related to the Relationship Between State and Federal Implementers and Congressional Intent to Delegate
It should not be necessary to make the case for how important the doctrines of administrative deference are to questions of statutory interpretation and implementation. Chevron, the Court’s flagship deference doctrine, is one of the most cited cases in history, and the significance of the interpretive authority that the Court has given to federal agencies, as Congress’s purported delegates, has been detailed by hundreds of commentators.120 But the Court has never resolved the question whether deference is available when multiple agencies are involved—even when the question involves only multiple federal agencies.121 The Court also has steadfastly refused to answer the question whether federal agencies may receive Chevron deference for federal-agency actions that would preempt state law.122 And the Court has never considered anything like deference to state (or private) implementers of federal law, even though some lower courts have granted such deference. Nor has the Court addressed the question of what, if any, process might be due to the states when they are negotiating with federal agencies, whether informally or through the administrative waiver process, about their joint role in implementation.123
1. Does National Federalism Suggest There Should Be a Chevron Deference Regime for State Implementers of Federal Law?
The Court does not recognize any kind of interpretive deference for state implementers of federal law, despite indications that Congress sometimes does intend for states to have discretion. Particularly puzzling about the Court’s federal-law myopia in this context is that, when it comes to federal agencies, the Court does take a more congressionally-focused and varied approach that would map well onto an account that includes state implementers.
I refer to United States v. Mead Corp.,124 in which the Court narrowed its broad reading of Chevron—which previously had operated as an across-the-board presumption of interpretive deference whenever statutes were ambiguous—and instead adopted a more nuanced understanding of deference as a varying feature of congressional practice; very much as I have described National Federalism as a feature of federal statutory design. With explicit recognition of the complexities of the modern administrative state, the Court in Mead moved to “tailor deference to [the] variety” of ways in which Congress delegates.125Chevron, as modified by Mead, however (despite the emphasis on legislative reality), suffers from the same federal-law bias as the other statutory interpretation doctrines and does not include nonfederal implementers.
If the Court is serious about linking deference to congressional intent, there is evidence that Congress does sometimes intend to defer to state implementers. Congress writes state implementation flexibility directly into some provisions of federal law. The health reform statute mentions state “flexibility” six times;126 its state-administrative waiver provision is expressly titled “Waiver for State Innovation,”127 and similar waiver provisions are scattered across the U.S. Code. Congress also makes direct delegations to the states and offers broad federal grants to states in which the given federal agency’s role is limited mostly to administering the federal-to-state financial flows.128
My recent study of congressional drafting, with Lisa Bressman, offers the first empirical evidence that drafters of federal legislation sometimes do intend for states to have implementation flexibility and sometimes explicitly intend for them to have interpretive deference on a level with Chevron.129 The congressional staffers surveyed also emphasized that the extent of the intended state roles varies across statutes. Seen through the lens of National Federalism, it should come as no surprise that some staffers reported that states are intended to have more salient roles in federal statutes operating in areas of historic state authority.130
Some lower courts have grappled explicitly with the idea of a Chevron for the states.131 Some courts have rejected deference, arguing that only federal agency approval matters.132 Other courts have argued that “Chevron’s policy underpinnings emphasize . . . the need for coherent and uniform construction of federal law nationwide. Those considerations are not apt to a state agency.”133Still other courts have argued that Chevron’s expertise rationale does not apply because “[s]tate agencies have no expertise in interpreting federal law.”134
Putting aside the fact that these kinds of arguments lack the kind of focus on congressional intent to delegate that the Court embraced in Mead, these cases also overlook the reason that Congress looks to the states in the first place. Unlike in the case of simple federal agency administration, uniformity is often the opposite of the goal when states have primary roles. So too, although state agencies may lack “federal law” expertise, Congress often relies on states because the law being implemented covers an area of historic state expertise, making states qualified to fill in policy gaps.
The possibility of deference for state implementers is not an easy question. States actors are not accountable to Congress or the President as federal agencies are and that alone might be a reason for eschewing Chevron-like deference for them. Congress also does not intend to give states the same kind of policy making discretion across all statutes, so there cannot be a single, consistent answer to this question even if courts did wish to effectuate congressional intent. But if the doctrinal focus is going to be on congressional intent, some level of deference (even if something less than Chevron) is worth exploring.
2. Does National Federalism Help to Resolve Questions About Whether There Should Be Deference to Multiple Implementers of Federal Law?
Congress often simultaneously charges federal agencies with implementation duties alongside the states. This question of multiple delegations obviously complicates a “Chevron for the states” analysis, because when a federal agency is also involved—particularly when the areas of state and federal responsibility overlap—Congress’s preferences on questions of uniformity, accountability and expertise may be less clear.
The bigger baseline problem raised by this question is that the Court does not have any kind of framework to evaluate questions of multiple implementers, even when only federal agencies are involved. Some courts that have considered the question have held that the presence of multiple federal implementers means Chevron deference for none. Chief Justice Roberts himself highlighted this doctrinal gap in a dissent last year.135
Here, too, Mead’s emphasis on congressional intent and statutory variety helps to chart a course. The Gluck-Bressman drafting study, for instance, found empirical evidence that drafters of federal legislation sometimes do intend to delegate to more than one implementer simultaneously.136 Sometimes the multiple implementers are all federal; other times they are mixed.137
Even in the federal-only context, multiple delegations still raise questions (indeed many of the same questions) concerning what kind of accountability, uniformity, and expertise deference doctrine is supposed to further. These are questions that require exploration and the difficulty of which I do not minimize. It may be the case that when it comes to multiple implementation (whether all federal or mixed), matters like accountability may trump Mead’semphasis on congressional intent; the point is that we have yet to see any such conversation at the Court.
3. Does National Federalism Require Some Restraint on Federal Agencies When They Attempt to Preempt or Otherwise Direct State Implementation of Federal Law?
What all of these unresolved matters have in common is that they go to the ability of states to flex their muscles in the implementation of federal statutory law and to the relationship between state actors and federal agencies in that endeavor. Others have previously illustrated that, even if one buys into the idea of the political safeguards of federalism, those political safeguards are lacking when it comes to federal agencies, where the staff is usually federal-law and uniformity focused and states are not represented.138
As the doctrine currently stands, federal agencies have almost unrestrained power to make all of the critical allocation decisions. The Court’s most recent statement at the intersection of Chevron and federalism, the City of Arlington case discussed in Part II, extends the deference accorded federal agencies even further, to include questions of the agency’s jurisdiction, even when state law would be affected by that decision.139 The federal statute in that case also contained an express “savings clause” for the preservation of state law, like those discussed in Part II.140 It is not a great leap from Arlington to the idea that federal agencies, armed with Chevron deference, could interpret statutes in ways that would constrain state flexibility in statutory schemes that Congress intended as internally federalist (or vice versa, making statutes more internally federalist than Congress may have intended141). Cases like Arlington reveal the deep tension between two of the Court’s favorite interpretive rules: Chevron deference for federal agencies and the presumptions that favor federalism for the states.142
To be sure, it would be difficult for courts to make these allocation choices themselves, particularly because they vary across statutes. More onus might be put on Congress to make its intentions clear. As one promising example, in the recent Dodd-Frank financial reform legislation, Congress took the rare step of expressly prescribing the deference level it desired for federal agencies to have on questions of preemption, and chose the lesser, Skidmore level of deference.143
4. Does National Federalism Suggest There Should Be Procedural Protections for State-Federal Regulatory Negotiations, Including but Not Limited to the Statutory Waiver Process?
Relatedly, there are “process” issues. Returning to the example of the ongoing health reform implementation, many state officials have complained about making time-sensitive implementation decisions without formal guidance from federal agencies about what the agencies ultimately would require.144 States hesitated to implement the law for fear that work undertaken would later be displaced by conflicting federal regulations, and HHS has often used informal processes in dealing with the states, even on key matters of implementation.145
Even in the context of administrative waivers—an important vehicle of state flexibility written explicitly into many federal statutes—there is a mysterious informality to the process, with most of the critical interactions happening as behind-the-scenes negotiations rather than through any formal, transparent procedure. No scholarly treatment appears to have considered, through the lens of process, precisely how the state-federal waiver practice works, even though scholars recently have begun to see waivers as significant vehicles of federalism.146 Erin Ryan’s important work, which argues for the application of bargaining theory to all aspects of the state-federal negotiations, comes closest.147
5. How Do Federal Agencies Interact with State Sovereign Acts in Service of National Federalism Schemes? Could Federal Agency Action Displace State Legislation Implementing Federal Law?
A related question concerns the balance-of-power between federal agencies and state governments. The Court recently held, in National Cable & Telecommunications Ass’n v. Brand X Internet Services, that a federal agency interpretation of an ambiguous federal law could displace a prior judicial interpretation of that same law.148 The Court maintained its stylized federal-actor-only perspective when deciding Brand X, but the case raises potentially explosive questions about how federal agency implementation intersects with the exercise of state sovereignty in federal statutory implementation—because state governments implement federal law, too. Does Brand X mean that, if a state legislature has passed a state law to implement a federal statute, a later federal agency interpretation could effectively nullify it?
No one would assume that a federal agency action could displace a congressional enactment. But this same issue, on the state legislative enactment side, has raised questions. Outside the health law context, for instance, a bill has been repeatedly introduced in Congress to “prevent unilateral actions by the EPA that second-guess the decisions of the state regulatory agency” and remedy the “atmosphere of regulatory uncertainty.”149 National Federalism may cast more light on this Brand X question, by highlighting the extent to which it is state sovereign activity—like state legislation—that is in danger of being displaced. A few lower courts have acknowledged this potential distinction, and held that statutory ambiguity is not enough to sustain later federal agency action to displace state law in cooperative schemes.150
B. Questions About Using the Contract Framework to Analyze State-Federal Intrastatutory Relationships
Of all of the questions on this list, the ambiguity of the “contract” metaphor may be most surprising. Congress’s power to spend for the general welfare has been its primary vehicle in the post-Printz era to entice states to enlist as implementers of federal programs, and the idea that this kind of legislation is essentially a “contract” has provided at least the rhetorical grounding for some of the highest profile federalism decisions from the Court in recent years, including the cases discussed in Part II.151Indeed, it is really the only theoretical framework, apart from preemption, that the Court has utilized to describe this modern terrain.
Part II already set out the ambiguity: courts have vacillated between whether the “contract” metaphor is a metaphor, or is intended to describe the actual status of the state-federal agreement, or is only intended to describe the context of the states’ choice to participate (the last option, in my view, is probably the correct limitation). In the end, it has served only to confuse.
The suggestion that these statutes are not “law” on the same level as other pieces of legislation makes little sense. And it is not how these statutes are actually treated with respect to other legal questions. For example, contracts have their own principles of interpretation, and those principles are not the same as the principles of statutory interpretation that courts apply to federal legislation, including to Spending Clause legislation.152 The Court also still applies Chevron deference to federal agency interpretations of Spending Clause legislation, but the courts do not otherwise allow federal agencies to modify the terms of federal contracts based on their own interpretive or policy preferences. Samuel Bagenstos offers other examples, including that Spending Clause legislation goes through the same, constitutional bicameralism and presentment process as any other federal law,153 and that “[if] spending conditions are not ‘law,’ . . . those conditions [could not] preempt inconsistent state laws under the Supremacy Clause.”154
7. Does the Contract Framework Tell Us Whether States Are More “Special” than Private Implementers of Federal Law?
The second big question for the contract framework is whether it really serves to highlight the sovereignty of the states in the ways that the courts seem to intend. Congress increasingly does “contract” with private entities to implement aspects of federal law. It does not appear that the Court has deeply engaged the question of how those private contracts might differ from the “contracts” between states and the federal government, but it seems doubtful that courts would hold them on the same level. Private delegations do not result in the kind of law- and regulation-making processes that state delegations spur. Nor are contracts with private implementers eligible for agency deference. The contract metaphor is deployed to elevate the position of the states in the legal analysis. But understanding these cooperative schemes as mere contracts may actually cheapen the contributions of the states—particularly insofar as it fails to differentiate them from the efforts of private entities.
As the foregoing pages have detailed, there is a great deal of what normally would be considered state sovereign activity that occurs when states implement federal statutory law. In the context of health reform, for example, at least 32 states have already passed state laws or taken regulatory action to implement the Medicaid and insurance marketplace provisions of the federal Act.155 Nearly every state has created or empowered a state entity or commission to evaluate the state’s options or begin implementation.156 All of these state actions were incentivized by federal law, but the precise forms they took—for example, whether Medicaid was expanded by state law or state regulation or how many insurers have been allowed to sell plans—varied across the states and depended on individual state constitutional and statutory authorizations. State courts will hear challenges to the operation of these regimes for years to come, and many of those challenges will implicate the details of state law on matters ranging from rules about procurement to state constitutional rights.
Law does not currently recognize these state sovereign actions as something that “federalism” doctrine might protect. A more state-centered approach might elevate the state character of these actions—privileging state courts, state substantive law, even mechanisms to further state accountability—than an approach that would emphasize the federal counterparts to these options. My aim is not to choose among the approaches here but, rather, to frame the kinds of inquiries that National-Federalism-oriented doctrines might undertake.
8. What is the Legal Status of the New Institutions and Laws Created by National Federalism? Are They Federal, State, or Both?
When it comes to what these state actions “are,” the case law is a muddle. As Young has pointed out, there remains deep judicial confusion about whether the “state implementation plans” that states must pass to implement the federal Clean Air Act have state-law or federal-law status for purposes of matters such as the subject-matter jurisdiction of the federal courts.157 Similar questions arise with respect to state water quality standards, passed as state legislation, but to implement the federal Clean Water Act;158 and to whether telecommunications agreements approved by states pursuant to their regulatory authority under the federal Telecommunications Act of 1996 give rise to federal questions.159 The Second Circuit has held that “there is no authority anywhere supporting the proposition that a state Medicaid regulation becomes a federal law merely by virtue of its inclusion in a state plan required by federal law.”160 And, with respect to federal statutes that incorporate state law by reference, others have detailed the state of indecision among lower courts about the related question of whether challenges to the state-law components of those statutes give rise to state or federal jurisdiction.161
This question about the “identity” of the state actions and institutions generated by National Federalism is the first-order inquiry on which the answers to countless other doctrinal questions depend. As the questions that follow illustrate, courts cannot draw lines between state and federal court jurisdiction, or choose whose law applies, or determine how the statutes should be enforced without first making a determination about the state or federal character of these actions.
9. Do Questions Involving the State Laws and Institutions of National Federalism Give Rise to Federal or State Court Jurisdiction, or Both?
As already noted, there is divergence among the courts, and even within courts, about when questions involving state implementation of federal law invoke the subject matter jurisdiction of the federal courts. Part of the problem is that the Supreme Court’s own basic federal-question-jurisdiction jurisprudence is so mushy. In its most recent sustained treatment, Grable & Sons Metal Products, Inc. v. Darue Engineering & Manufacturing, the Court declined to adopt a “single, precise all-embracing test”162 for state law claims implicating federal questions. Instead the Court articulated several other factors, including whether the federal issue is necessarily arising and substantial, and whether the question is of the sort that “a federal forum may entertain without disturbing any congressionally approved balance of federal and state judicial responsibilities.”163 Of particular relevance, the Court also put some emphasis on congressional intent, holding thatasserting federal jurisdiction must be “consistent with congressional judgment about the sound division of labor between state and federal courts governing the application of § 1331,” the federal question statute.164
Under Grable, were the courts focused on National Federalism, one might predict a massive shuttling of these statutory federalism cases into state courts. It is true, as Resnik has argued, that the current “language of federal jurisdictional law—describing cases as having ‘federal ingredients’”—in some ways “captures the many instances in which state and federal laws overlap,”165 but the Court has declined to hold that every case with substantial federal elements gives rise to federal jurisdiction.166 Moreover, the federal courts are unlikely to want the hundreds of state-law cases implementing federal statutes—cases that often involve run-of-the-mill state-law issues such as matters of contracting and personnel.167 To that end, consider this statement from the Second Circuit, which held, in declining jurisdiction over a challenge to New York’s Medicaid law, that to decide otherwise “would provide a jurisdictional basis for federal judicial review of every disputed state administrative ruling relating to Medicaid.”168 Recognizing the state’s ownership over these aspects of federal law implementation would keep such cases out of the federal courts—a result that most federal judges likely would prefer; that arguably would reinforce state autonomy within these schemes;169 and to which Grable and the Court’s cases that have followed seem to point.170
The counterargument is that if one has concerns about state-centered rebellion171 or too much variety in implementation, recognizing the federal-law underpinnings of these statutory schemes might allow a federal, harmonizing, “protective” influence over their on-the-ground implementation.172 In the end, the recent balancing-test approach to federal jurisdiction that Grable adopts may, indeed, be better suited to the kind of continuum of federalism and nationalism that I have described than a set of bright-line rules. The current test, however, is too imprecise and inconsistently applied. More importantly, it is not well aimed at the kinds of questions to which National Federalism gives rise. For instance, courts generally do not ask whether preserving a (perhaps exclusive) role for state courts to decide these cases furthers state autonomy within national statutory schemes, or how the tension between nationalist and state-centered impulses within a particular statute might affect the answer to that question.173
10. Do State or Federal Legal Doctrines, Such as Standards of Review, Apply to State Laws and Institutions that Are Part of National Federalist Statutory Schemes?
Then there are the choice-of-law questions detailed in Part II, questions that courts do not appear to recognize as federalism questions in the first place. These are different questions from questions about whether federal courts should take jurisdiction; these questions are about what decision-making regime—state or federal—applies, regardless of which court is deciding the case.
I have previously documented how federal courts fail to apply state interpretive principles to state statutory questions, and how that practice is inconsistent with Erie.174 A recent student note likewise chronicled how federal courts apply federal administrative law principles, including requirements of the federal Administrative Procedure Act, even when they are reviewing implementation by state agencies of cooperative federalism schemes.175 When it comes to deference, many states have their own, different principles of agency deference from the federal regime—including some states that prohibit any deference at all176—but some lower federal courts have applied federal deference doctrines to state agency implementation of federal law.177
This doctrinal disarray presents at least two different kinds of “federalism” issues. As an initial matter, identical forms of state action in service of federal law are being reviewed in different ways in different cases, depending on which court is hearing the case. This is precisely the kind of cross-court inequity that the Erie doctrine aims to avoid. Second, from the perspective of developing doctrines to reinforce the sovereign actions of states within national schemes, many of these choice-of-law decisions are missed federalism opportunities. Courts that apply federal law to matters concerning the state regulatory apparatus miss the chance to build into federal statutory law more of the kind of diversity that Congress may have intended by looking to the states in the first place. Those courts also fail to recognize, and so undermine the autonomy of, what may be some fundamentally “state” actions—actions that merit the application of state legal standards—within the national implementation process.
When it comes to how these federal schemes can be enforced against state implementers, that case law, too, is unsettled. The Court seems torn, or perhaps not focused, on what kind of accountability the doctrines should further.
Enforcement is a complex area that implicates many other strands of the Court’s case law (for example, the Court’s precedents concerning when it will imply a right of action) and so complete treatment cannot be given here. Suffice it to offer two cases that illustrate different approaches to accountability in this context. In the Douglas case, discussed in Part II, the Court effectively shielded California from a challenge to its Medicaid program because the federal agency had approved the program. Compare Douglas to U.S. Department of Energy v. Ohio, an older case in which the Court held that state-law-created fines, enacted as part of state implementation of the Clean Water Act and approved by the federal agency, did not “arise under federal law” for purposes of triggering enforcement of the fines against the United States.178 In Douglas, the state action was effectively federalized; in the other it was not. As I already have suggested, a state-sovereignty-focused perspective on National Federalism also might point toward making states more accountable than the courts often do for their roles in these schemes, because at least some of those actions would be understood as the states’ own.
12. Can Congress, Through National Federalism Statutes, Give State Actors Powers They Do Not Have Under State Law?
Finally, there are questions that implicate this state/federal ambiguity but that are trained on the power of Congress, not the courts. One important illustration was discussed in Part II—the question of whether Congress can give state actors powers that they do not have under their own state laws.179 At the broadest level, as Jim Rossi has pointed out, when Congress delegates to state agencies, it does not seem to consider the fact that some state constitutions prohibit agency delegations entirely.180 But also with respect to specific delegations, it remains unanswered whether Congress can imbue state actors with administrative authority that they otherwise do not have under state law and what the precise nature of that authority would be. In the federal housing law context, for instance, the question has arisen whether federal housing laws can empower local housing agencies to use procedures that state laws prohibit.181 Answering these questions likewise requires a developed account of what exactly the states are doing within these federal schemes, including whether and when they are exercising state powers.
The last category of questions arises from conceptualizing federalism as a phenomenon over which Congress has primary control. This theoretical reorientation puts enormous pressure on the rules of statutory interpretation to serve as the key doctrines of the state-federal relationship. But those doctrines fall short in multiple ways, ranging from their inattention to congressional intent and statutory variety to their floppiness as legal rules.
13. Do Current Federalism Statutory Interpretation Doctrines Really Aim to Effectuate Congressional Intent?
Our current federalism-focused rules of statutory interpretation are black-and-white rules that have no empirical grounding in congressional intent, and in any event work only to separate state and federal law, not to negotiate their interaction. They rarely see federalism as existing in varied, uneven fashion across the federal statutory landscape, much less see that variety as tethered to questions of congressional intent and federal statutory design. As such, these doctrines stand in stark contrast to the Court’s efforts, in the administrative law context, to “tailor deference to [the] variety” of ways in which Congress legislates.182 National Federalism posits multiple forms of “federalisms”183 (and nationalisms) that a set of “on/off” interpretive doctrines cannot capture.
The presumption against preemption and the other federalism canons, for instance, have no apparent empirical grounding in congressional preferences. The federalism-reviving Rehnquist Court of the 1980s devised and utilized these presumptions to bring judicially-preferred federalism values to bear on its interpretive process, a strategy that some have called quasi-constitutional law, and others have called under-the-radar judicial lawmaking.184 Since then, the Court has deployed the convenient fiction that these canons not only reflect important constitutional norms, but are background norms that Congress shares and against which it is reasonable to assume that Congress legislates.
Some “process federalists” have seized on this account. These process theorists are the closest we have to theorists of National Federalism—because they understand Congress’s central role in establishing the state-federal allocation, rely on “the states’ representation in Congress as the primary means of protecting state sovereignty, and envision judicial intervention only to ensure that this process is functioning properly.”185 Some process theorists thus embrace the canons of statutory interpretation as rules that should govern the game.186But Congress is generally uninterested in and ignorant of these interpretive rules, in part because the courts apply them too inconsistently for them to serve as the basis of coordinating interbranch behavior. The canons cannot serve as rules of the game if Congress does not know them and if the Court does not give them the legal teeth to incentivize Congress to learn them. Perhaps the best we can say is that they are judicially imposed policies, but that court-centric perspective is not well suited to a federalism that is so linked to questions of congressional federal statutory design.
My recent congressional drafting study offers evidence confirming the suspicion that congressional staffers know few of the canons of statutory interpretation. The Court’s so-called “clear statement rules”—rules that require Congress to use special “magic words” to make known its intentions on high salience issues like federalism—fared particularly poorly under empirical scrutiny. The study found near complete ignorance of those rules,187 an enormous problem because the ostensible goal of clear statement rules is to provide the Court and Congress with a shared language so that Congress can make its intentions with respect to federalism known.
Regardless, these doctrines operate as one-way ratchets; each assumes that the scale should tip in the same direction every time (almost always toward states), an assumption belied by both the variety of forms that National Federalism takes and the often-simultaneity of both federal and national impulses within a single statutory scheme. In recent years, the Court has made some small moves that show the promise of expanding the doctrines’ capacity. In one of the Voting Rights Act cases decided last term, for instance, the Court reversed the presumption against preemption for that particular statute, a move it has likewise effectively made for the Federal Arbitration Act188 (but again the presumption still operates in one direction; now, toward nationalism). Outside the federalism context, as I have detailed elsewhere, the Court deploys more than a hundred subject- or statute-specific rules of statutory interpretation, such as the presumption that ambiguities in the Bankruptcy Code be construed in favor of the debtor, or that exceptions to the Tax Code be narrowly construed.189
These small efforts to disaggregate what have been mostly sweeping interpretive presumptions have not been widely noticed, and it is too soon to know if it is a trend that will continue. The greater the number of these rules, moreover, the more the risk of unbearable legal complexity. But the Court openly utilizes a variety of interpretive doctrines for the common law, and even the Constitution. What isremarkable is that a single set of generally applicable presumptionshasdominated the statutorylandscape for so long. It seems possible that these subtle recent efforts by the Court to differentiate among the doctrines evince a maturation of the Court’s understanding of the kinds of law needed for a statutory age: an evolution toward a set of rules that, like National Federalism, recognize the diversity across statutes that one would naturally expect in a legal landscape dominated by them.190
14 Would National Federalism Doctrines Of Statutory Interpretation Be Real “Law”? If So, What Kind Of Law, And Could Congress Overrule It?
The biggest problem, however, may be that these interpretive doctrines are not much of “doctrines” at all. Even if the doctrines of statutory interpretation could be better tailored to individual federal statutory designs, the entire regime of statutory interpretation suffers from an overarching jurisprudential ambiguity that could be fatal to any theory of federalism that rests on it. The rules of statutory interpretation do not have a clear legal status. It is a puzzle whether they are law, judicial philosophy or something in between.191 Even when a majority of Justices agrees on an interpretive principle in a particular case, that principle is not viewed as “law” for the next case, even when the same statute is being construed. Instead, courts and scholars routinely refer to the canons as “universal” principles” or “rules of thumb”—a sharp divergence from the way in which they treat analogous decision-making principles, including those of constitutional law (where doctrines like the tiers of scrutiny and also federalism doctrines, like the Commerce Clause test, are treated as precedential law).
I have previously criticized in detail the resistance to treating statutory interpretation methodology as “law.”192 For present purposes, the point is not to resolve the question of the legal status of these interpretive presumptions or whence they derive. Rather, the point is to establish that the legal status of these rules is indeed in question and that, regardless of their status, application of the rules does not receive precedential effect. Another question that remains unresolved is the very big question of what role exactly it is that these rules—and by extension, the courts—are supposed to perform in the interpretive endeavor. It remains uncertain, for instance, whether the federalism presumptions are supposed to reflect how Congress drafts, affect how Congress drafts, or simply layer judicial/constitutional values atop Congress’s work product. Each is a very different aim, from the standpoint of how much courts should interfere with the legislative process or of which branch controls questions of state-federal intrastatutory allocation. The legal status of the doctrines also has profound implications for the question of who can change them. Understood as common law, or even as “constitutional common law,” Congress could override the doctrines by statute;193 understood as constitutional law or as something internal to the individual judge (and so not law at all), Congress could not.
However one comes down on these questions, it would be odd to give the great weight of federalism’s doctrinal regime to a set of presumptions that most judges currently view as mere “rules of thumb.” This is not to say that statutory interpretation must go it alone—other efforts, such as finding ways to add more state-centered voices to the legislative or administrative process would offer a political response, rather than a doctrinal one. But from the perspective of one who believes that a role for courts should persist even in this political context, for the doctrines of legislation to play a central role of effectuating National Federalism, they need to be doctrines.
15. How Might National Federalism Be Affected by Ongoing Methodological Disputes, Such as the Dispute Between Textualist and Purposivist Judicial Interpreters?
Finally, at a more granular level, the intersection of congressional intent and federalism also will make current methodological disputes about statutory interpretation more relevant to these critical allocation-of-power questions. As a parallel example, in the context of preemption, scholars have described that the Court deploys that doctrine in ways more attendant to statutory purposes than to text.194 So too, in the National Federalism context, the foregoing pages have demonstrated how the different Justices have approached the interpretive questions. Justices Stevens, Ginsburg, Breyer, and even Chief Justice Roberts have been more willing to see the federalism inside federal statutes than have more textualist Justices like Justice Scalia, even though textualist judges have often been more federalist. Justice Scalia wants clear direction and bright lines—a desire incompatible with a Congress that legislates with gaps for administrative flexibility195 and that utilizes states in varied ways across different statutes.
Federalism has come into the statutory age, but without the doctrines it requires. Today, it is Congress that decides the major questions of state-federal allocation and it is the role of the states within federal statutory schemes that ensures the states’ enduring relevance. The states in turn have evolved their own identities, continuing to pass state laws, appoint state regulators, and hear cases in state courts—all sovereign acts, but ones put into motion by national law rather than as alternatives to it. This National Federalism is not easy. It is dynamic, uneven, and dependent on questions of federal statutory design and the whims of politics. But the difficulties are no excuse for the lack of law to answer National Federalism’s many open questions. Our Federalism (and Our Nationalism) depends on it.