The Yale Law Journal

April 2014

From Sovereignty and Process to Administration and Politics: The Afterlife of American Federalism

abstract. Announcing the death of dual federalism, Edward Corwin asked whether the states could be “saved as the vital cells that they have been heretofore of democratic sentiment, impulse, and action.” The federalism literature has largely answered in the affirmative. Unwilling to abandon dual federalism’s commitment to state autonomy and distinctive interests, scholars have proposed new channels for protecting these forms of state-federal separation. Yet today state and federal governance are more integrated than separate. States act as co-administrators and co-legislatures in federal statutory schemes; they carry out federal law alongside the executive branch and draft the law together with Congress. Lacking an autonomous realm of action, states infuse federal law with diversity and competition, aligning themselves with certain federal actors to oppose others. States also participate in national political contests on behalf of Americans both inside and outside their borders. They facilitate competition between the Democratic and Republican parties and offer staging grounds for national networks seeking to advance their agendas through direct democracy. Instead of focusing on state autonomy and distinctive interests, we should accordingly recognize contemporary American federalism as an expression of our multifarious nationalism. This need not lead us to answer Corwin’s question in the negative: precisely because states are disaggregated sites of national governance, not separate sovereigns, they continue to serve as vital cells of “democratic sentiment, impulse, and action.”

author. Associate Professor, Columbia Law School. I am grateful to Robert Ferguson, Heather Gerken, Abbe Gluck, Kent Greenawalt, Bernard Harcourt, Jeremy Kessler, Alison LaCroix, Tom Merrill, Gillian Metzger, Henry Monaghan, Jeff Powell, David Pozen, Judith Resnik, Dan Richman, Cristina Rodríguez, Chuck Sabel, and participants in Columbia’s Faculty Workshop for stimulating and clarifying conversations. I also thank Bridget Fahey and the editors of the Yale Law Journal for wonderful editorial suggestions and Jeremy Girton for excellent research assistance.


Edward Corwin famously announced the passing of dual federalism in February of 1950. In the wake of the Great Depression, the New Deal, and two world wars, he argued that the federal system had “shifted base in the direction of a consolidated national power” and wondered whether “the constituent States of the System [could] be saved for any useful purpose.”1

Federalism scholars have offered a variety of answers to his question. For some, the imperative is to revive dual federalism, a commitment to judicially enforced separate spheres of state and federal authority.2 But for many more, Corwin’s words have been read not as a eulogy but as a challenge, a call to consider how federalism might survive without clearly delineated domains of state and federal authority. If the “standard view held by students of law and politics is that the history of American federalism has been an inexorable series of moves in which national power has displaced state and local power,”3 many of the contrarians who compose the federalism academy have found within the very moves taken to aggrandize national power—in particular, the rise of the administrative state and the rise of national political parties—the keys to federalism’s longevity. Dual federalism made way for process federalism,4 which has in turn recast administration and party politics as channels through which state power is preserved. The administrative state and the two major political parties have been reimagined as guardians, rather than slayers, of American federalism.

This embrace of administration and politics has been productive but incomplete. Even as the literature has ceased to insist on judicially enforced, clearly defined spheres of state and federal authority, it has clung to a dualist vision of state-federal separation. Dual federalism may be dead, but its focus on autonomous state governance and distinctive state interests—on separation—lives on. What has changed is simply the means of protecting these ends. In our leading contemporary federalism narratives, integration paradoxically yields separation: state actors use their connections to federal politicians and administrators to safeguard state autonomy and to advance particularistic state interests.

But integration yields integration, not separation. Taking administration and partisanship seriously means we must acknowledge a certain incoherence to invoking state governance and interests in opposition to federal governance and interests. Instead, state and federal governance and interests are deeply intertwined and, in many cases, indistinguishable. The state and federal governments together produce national governance in the service of various national interests.

First, in our most substantial federal statutory schemes, states act as co-administrators and even co-legislatures, complementing and competing with the President and Congress. Insisting on an autonomous realm of state action neglects this important aspect of today’s federalism and also leads us to overlook how states affect the federal government. As administrators and authors of federal law, states alter the roles and relationships of federal political and bureaucratic actors and amplify competition across and within the branches of the federal government. We cannot understand today’s federalism without considering the separation of powers, and we likewise cannot understand the separation of powers without considering federalism. Our federal government is really a federalist government.

Second, and extending beyond state administration of federal law, states participate in political controversies that are national in scope. They serve as staging grounds for competition between our two major parties and for a variety of networked interests seeking to translate political commitments into law. Insisting on distinctive state interests neglects this important aspect of today’s federalism and also leads us to overlook how federalism facilitates national political conflict. By advancing a Republican or Democratic agenda, or by offering a toehold for networked interests to further their causes through direct democracy, states engage in political struggles on behalf of people both inside and outside their borders, and they are watched by Americans nationwide. National political conflict is drawn on the canvas of the states.

To some, this may sound like nationalism, and not federalism. But it is not nationalism as centralization and homogenization. Focusing on the states, federalism scholarship tends to conceptualize the federal government as unified and to treat the dominant voice in the federal government as the voice of a single national interest. Our federal government and our nation are irreducibly multifarious, however, and today’s federalism instantiates this national diversity.

Understanding “federalism as the new nationalism”5 thus complicates both the “federalism” and the “nationalism” sides of the equation. The phrase should be taken to refer to a dynamic process of coevolution rather than the colonization of states by the federal government, on the one hand, or the preservation of federalism as state-federal separation, on the other. While existing accounts locate federalism in state autonomy and distinctive interests and proclaim it either dead or alive, this Essay instead locates federalism in the legally and politically generative interaction among the state and federal governments and the American people. Regarding the state and federal governments as interdependent sites of national governance, it tells a story not about the life or death of federalism as a rival to nationalism, but about federalism’s afterlife as a form of nationalism.

Part I critiques the federalism literature’s dominant account of integration-as-separation. Part II considers state-federal integration in the administrative realm, while Part III considers such integration with respect to partisan politics and direct democracy. The Essay calls for an embrace of administration and politics as transformative, rather than preservative, of American federalism. Attending to how states pluralize our national life offers the most plausible account of their contemporary vitality.

I. the death and life of american federalism

Federalism scholarship in the latter half of the twentieth century and the beginning of the twenty-first has largely resisted the easy narrative of federal aggrandizement at the expense of the states. Acknowledging that the federal government has assumed regulatory authority over ever more domains, the literature has argued that this signifies not federalism’s demise but rather a change in the mechanisms that safeguard the place of states in our system. A variety of process federalisms have recast forces traditionally envisioned as threats to a robust federalism as its guardians: Congress and the President, the Democratic and Republican parties, and the administrative state have each, in turn, been given a new role. But even as these recastings shed dual federalism’s insistence on judicial review and clearly delineated spheres of state and federal authority, they retain its core commitment to state autonomy and distinctive interests. Novel forms of state-federal integration are thus treated as means of preserving state-federal separation: the integration of state and federal actors safeguards the separation of state and federal action.

A. Congress and the President, Political Parties, and the Administrative State: Recasting Threats

Just as Corwin announced dual federalism’s passing, a new wave of scholarship began to proclaim federalism alive and well. The key, adherents of process federalism insisted, was to look beyond a judicially enforced model of state sovereignty to the protections accorded states through their role in the composition and selection of the federal government.6 The political safeguards that Herbert Wechsler described in 1954 were not newly discovered—as Wechsler acknowledged, James Madison had sketched them nearly two centuries earlier7—but they assumed new urgency as critics worried about the vitality of American federalism.

Wechsler’s main conceptual move was to re-envision a looming threat to state power as a safeguard. On the dual federalist account, Congress and the President were precisely what the states had to fear: these federal actors would overreach and impinge on state sovereignty absent a judicial check; with the demise of that check, Americans were destined to inhabit a world of unconstrained federal power. Process federalism first neutralized the threat. In Wechsler’s view, state law was the default and those who favored federal intervention bore the burden of persuasion, ensuring federal law remained interstitial. He then went further: the branches of the federal government could be counted on affirmatively to preserve state autonomy because the states selected their members. The formal state role in choosing Senators and, to a lesser extent, members of the House through districting and the President through the Electoral College, he argued, gave federal politicians incentives to avoid encroaching “on the domain of the states.”8 While for Corwin, the aggrandizement of congressional and presidential power threatened federalism’s survival, for Wechsler and his followers, congressional and presidential power would themselves be marshaled in support of state autonomy.9

By the end of the twentieth century, notwithstanding its star turn before the Supreme Court,10 Wechsler’s formal process federalism had lost some of its luster. One problem, critics pointed out, was that the constitutional structures he emphasized were already on their last legs when he wrote. Senators had long been chosen directly by the people; state legislative control over districting for the House was curtailed by the reapportionment cases of the 1960s; and even Wechsler couldn’t muster much of an argument for the Electoral College.11 But a deeper problem, critics maintained, was that these structural safeguards at best protected geographically concentrated interests, not the autonomy of state institutions. In other words, Wechsler’s process federalism failed to protect federalism.12 The hunt was on for new safeguards.

Two more practical contenders quickly emerged. Kicking off process federalism’s second wave, Larry Kramer argued that the political party was the most important institution of American federalism. The formal structures Wechsler posited were not up to the task; instead, the nitty-gritty politics that animated them were the real political safeguards of federalism. The Democratic and Republican parties, Kramer argued, are loose confederations of national, state, and local organizations, lacking ideological and programmatic discipline. Candidates at each level depend on assistance from party organizations and officials at the other levels, and this mutual dependence leads them to respect one another’s concerns. In particular, federal politicians’ reliance on their state co-partisans to get and remain elected makes them respect the desires of state officials, desires that include the protection of state autonomy.13

Like Wechsler’s formal variant of process federalism, Kramer’s informal one recast a threat to the state-federal balance as, in fact, its redeemer. The framing vision, Kramer maintained, assumed a natural antagonism between state and federal politicians, and because the former would enjoy more popular support, they would have an advantage in turf wars with Congress. But this natural antagonism was undermined by the rise of national political parties, which the Framers had not foreseen: “The ‘natural’ fault line between state and federal officials was soon bridged by cross-cutting attachments based on ideology and party affiliation, and the most important anticipated source of protection for states was promptly rendered ineffective.”14 If parties posed a threat to state autonomy in this respect, however, Kramer suggested that they were ultimately a boon for it. Properly understood, political parties safeguarded federalism by “link[ing] the fortunes of federal officeholders to state politicians and parties” in a way that “assured respect for state sovereignty.”15 Much as Wechsler had reimagined Congress and the President as protectors of state autonomy, so Kramer reimagined the national political parties as protectors of state autonomy.

Perhaps the most pronounced inversion of menace into safeguard has emerged in the administrative realm. The rise of the administrative state has long been viewed as the principal threat to American federalism. The growth of federal power that Corwin and his contemporaries emphasized, for instance, was largely the growth of administrative power; it was the executive branch, including a host of new agencies, that assumed regulatory authority over traditional state domains, from welfare and health, to education and safety, to housing and the environment.16 But federalism scholars have in recent years recast the administrative state as not only consistent with but also a protector of federalism by extending the process literature’s logic to administration.

While the classic political safeguards account posits that Congress protects state autonomy, the administrative federalism account argues that federal agencies do so. Taking seriously Corwin’s argument that the centralization of power in the federal system was accompanied by a transfer of power within the federal government from the legislative to the executive branch, scholars including Gillian Metzger, Catherine Sharkey, and Brian Galle and Mark Seidenfeld have suggested it is better for the states to have administrative agencies on their side than to rely on Congress alone. Indeed, they maintain, agencies are not only necessary adjuncts to Congress but the institutions best positioned to protect the states.17 Focusing on notice-and-comment rulemaking in particular, these scholars argue that administrative law requirements including solicitation of and response to public comments make agencies more transparent, deliberative, and accountable than Congress and better able to engage in dialogue with state governmental entities about the impact of federal regulation on the states.18 On this view, the expansion of federal administrative authority need not be a substitute for state autonomy; it may instead be the vehicle through which state autonomy is protected.

B. Integration as Separation

With the shift from sovereignty to process, the three forces deemed most threatening to American federalism—Congress and the President, national political parties, and the administrative state—have thus been neutralized and recast as safeguards. Perhaps unsurprisingly, the “federalism” secured by each of these moves has required some reconceptualization. Most formal definitions of federalism, for instance, provide that each level of government must be sovereign in at least one policy realm and that such sovereignty must be constitutionally entrenched.19 But process scholars have unmoored federalism from constitutionally fixed spheres of state and federal action. A successful federalism, the literature maintains, does not require ex ante specification of a domain of exclusive state authority.20 Closely related, process theories argue that judicial enforcement of federalism is, at least largely, unnecessary. In contrast to the dualist suggestion that judicial review is a life or death matter for the states,21 process scholars contend that political actors may be relied upon to mediate the federal-state bargain.22

Even as it has conceded much to reconcile federalism with the substantial powers of Congress and the President, the nationwide operation of two major political parties, and the rise of the administrative state, however, the process federalism literature has clung to a dualist core of state-federal separation. As this Section explores, it continues to insist on such separation in two closely related ways. First, as a matter of state autonomy: federalism is considered secure as long as states enjoy a space in which to set their own policies without the federal government’s interference. While this space is not the fixed sovereign sphere of dual federalism, scholars nonetheless demand some domain in which states have exclusive regulatory authority. The departure from dualism thus inheres not in the rejection of an exclusive state domain, but only in the extent to which this domain is fixed versus fluid and judicially protected rather than secured through more informal means.23 Second, process federalism venerates distinctive state interests: we need federalism because state residents have different preferences from the national populace and residents of other states. These two concepts are closely related and often conflated: autonomous state governance is valued because it allows state populations to advance their particularistic interests.

Understanding federalism in terms of state autonomy and distinctive interests, various process narratives posit political representation, partisanship, and administration as guarantors of such state-federal separation. If scholars have saved federalism from the looming danger of the national by recasting threats as safeguards, they have largely done so by treating integration as a means of separation. Radical as Wechsler’s account was in certain respects, for example, he focused on the preservation of state autonomy.24 To be sure, such autonomy would not be guaranteed by the courts; it would have to be fought for in Congress. And its sphere would not be neatly defined; it would be contested and shifting. But, at bottom, the political safeguards of federalism protected a type of state power similar to that championed by dual federalism.25 What was different was the means by which it would be ensured. Instead of relying on separation to yield separation, Wechsler’s process federalism was a theory of integration yielding separation: the relationships of state and federal actors would guarantee autonomous arenas for state and federal action.26

Kramer’s party-based account likewise defined the success of process federalism with reference to dual federalism. Just as Wechsler’s critique of dualism did not disturb its animating vision of state autonomy, Kramer’s critique of Wechsler’s formal process federalism preserved this vision; Kramer simply insisted that Wechsler’s safeguards were insufficient to protect state autonomy. Political parties, he argued, were the true safeguard of federalism. His claim was not that partisan ties lead federal officials to respect the substantive interests of state politicians in, say, regulating pollution, lowering taxes, or providing early childhood education; it was, rather, that they lead federal officials to respect the authority of state institutions to advance these particular interests themselves.27 Indeed, Kramer mocked the idea that state interests could be furthered by the federal government: “So far as I am aware, no one defends federalism on the ground that it makes national representatives sensitive to private interests organized along state or local lines. Rather, federalism is meant to preserve the regulatory authority of state and local institutions to legislate policy choices.”28 Like Wechsler, Kramer stripped away sovereignty-through-judicial-review but retained the aspirational core of state autonomy.29 States would further their distinctive interests through their own governance, not through their participation in national politics. And, also like Wechsler, Kramer insisted that integration of one sort yields separation of another: the integration of state and federal officials through confederated parties leads them to respect each other’s separate domains.

The administrative variant of process federalism similarly understands integration as a basis for ensuring state-federal separation. Scholars who argue that state regulatory authority may be protected within the administrative realm generally do not claim that states can ensure the substantive goals favored by their residents are realized through federal administration. They reason, rather, that federal agencies may carve out space for states to autonomously further their particular interests.30 This is a closely contained state domain—one nestled within the administrative state and coming largely at the mercy of federal bureaucrats and politicians—but it is a separate domain all the same. The argument, that is, continues to define federalism in terms of state autonomy and distinctive interests. And like Wechsler’s and Kramer’s versions of process federalism, administrative federalism, too, regards state-federal integration as the basis for state-federal separation: the close connections between federal and state administrators are what incline the federal bureaucracy to respect state autonomy.31

C. Integration as Integration

Process scholarship has offered tremendously valuable insights into contemporary federalism, but it is a mistake to believe that national political parties and the administrative state preserve autonomous state governance and distinctive state interests. Administrative and political integration do not yield state-federal separation. They yield ever-more thoroughgoing state-federal integration as states become sites of national policymaking and partisan conflict. In process scholarship’s insistence that new conditions have preserved American federalism, we have largely missed both how our federalism has changed and how our nationalism has changed along with it.

States today serve as disaggregated sites of national governance in two main respects. First, they are component parts of the national administrative apparatus. In our major statutory schemes, states carry out federal law alongside the executive branch and write the law alongside Congress. This state role challenges depictions of states as autonomous governments. In critical areas, states do not enjoy a realm in which to set their own policies; instead, they set national policy together with federal politicians and bureaucrats.

Second, states facilitate competition between the Democratic and Republican parties and offer staging grounds for national networks seeking to further their agendas. This state role challenges depictions of state interests as distinct from national interests. Instead of advancing particularistic commitments, states often give concrete form to interests that exist throughout the nation; functioning as a single platform on which political disputes play out for the American people, they enable a variety of national publics to emerge.32

The remainder of this Essay explores ways in which contemporary federalism is best understood in terms of integration, not separation.33 States’ role in administering federal law and framing national political competition calls into question the federalism literature’s many arguments that integration preserves separation while offering new opportunities to consider how states produce national governance. Taking administration and politics seriously reveals how states shape the balance and exercise of power within the federal government, as well as how they facilitate national partisan competition and nationwide political change.

II. administration: federalism as the separation of powers

The federal government’s move, particularly with the social revolutions of the New Deal and Great Society, to regulate in areas traditionally occupied by the states has, in turn, bound up states in the administration of federal law. The dynamic is one of mutual empowerment more than federal aggrandizement: as the fourth branch grows, so too does the states’ role grow within it.34 States furnish administrative capacity and democratic legitimacy, bolstering the ability of the federal government to achieve its objectives. But they also inject diversity, contestation, and a degree of chaos, mild or otherwise,35 into national governance. Rather than a federal executive branch solely responsible for carrying out national programs, we see states serving as co-administrators, frequently challenging the federal executive’s exercise of its statutory authority. States also increasingly serve as a shadow federal legislature, rewriting substantial portions of statutory schemes subject to waivers from the federal executive. Here too, our federal government becomes federalist as states vary the content of national programs and push back against congressional decisions.

The state role in administering and authoring federal law complicates conceptions of federalism grounded in autonomy. As the cooperative federalism literature has long noted, a separate space for state action is nowhere to be found in the many programs that intertwine state and federal governance.36 In thinking about states’ role in federal statutory schemes, however, we have not paid sufficient attention to divisions within the federal government—whether across the branches, as this Part contemplates, or along other dimensions, such as partisan politics, as the next Part explores.

Breaking open the national side of cooperative federalism illuminates two important dynamics. First, it reveals that the diversity and competition generated by state administration of federal law do not follow from state-federal separateness. Instead, states ally themselves with certain federal actors and interests in order to oppose others. Diversity that also exists within the federal government assumes concrete form in the fifty states. Second, taking states seriously as national actors reveals that the separation of powers and checks and balances at the federal level are shaped by an actor outside the branches of the federal government. If we cannot fully understand today’s federalism without parsing the federal government, neither can we fully understand today’s separation of powers without considering the states.

A. States as Another Executive Branch

The main role states play in federal statutory schemes is as administrators of national programs, a sort of second executive branch operating alongside the President and the D.C. bureaucracy. Through its conditional spending and conditional preemption powers, and often a combination of the two, Congress has brought states into the administration of the United States’ most substantial statutory schemes. States exercise concurrent authority with the federal executive in social welfare programs like Medicaid and the Patient Protection and Affordable Care Act; environmental programs like the Clean Air Act and the Clean Water Act; and a host of other schemes from immigration to telecommunications to financial regulation.37

While these programs often travel under the label “cooperative federalism,” state actors do not always cooperate with their federal counterparts; their actions are often decidedly uncooperative.38 But such uncooperative behavior is rarely framed in terms of state versus federal authority as such. Instead, states tend to fight with the federal executive branch about the meaning of the statute Congress has designed and the allocation of federal authority. We witness debates about federalism that are really debates about the separation of powers.39

Consider two examples from recent years: state opposition to the Bush Administration’s failure to regulate greenhouse gas emissions and to the Obama Administration’s immigration policies. The Clean Air Act authorizes both the Environmental Protection Agency and California to adopt vehicle emissions standards.40 Because the state’s power to regulate is secured by a federal statute, this scheme frames regulatory disputes between the state and federal executive in terms of the congressional grant of authority, not state autonomy. When the Bush EPA declined to regulate greenhouse gas emissions and also denied California the waiver that would have allowed the state to regulate such emissions, California insisted that it was attempting to faithfully implement the Clean Air Act while the federal executive branch disregarded the statute. Its argument had less to do with federalism than with the distribution of federal authority: state officials argued that the Bush Administration was abdicating its statutory duty to regulate such admissions and that the state was seeking to vindicate congressional intent.41

Arizona’s recent fight with the Obama Administration concerning immigration policy has had much the same character. Arizona’s SB 1070 would have, among other things, made it a crime to be in Arizona without carrying registration papers, required police to determine a person’s immigration status during a stop upon reasonable suspicion that the person was unlawfully present, and allowed police to make warrantless arrests of persons they believed to have committed certain crimes.42 In defending the state law, Arizona did not challenge federal authority over immigration or insist on state autonomy. Just the opposite: it claimed the mantle of Congress and argued that it, and not the federal executive branch, was seeking to execute federal immigration law as Congress intended. The problem, Arizona argued, was that the federal executive was not carrying out federal immigration law to its fullest extent. And, it continued, Congress had lent states authority to cooperate in immigration enforcement so as to avoid such laggardness.43

As both the Clean Air Act and immigration examples suggest, when states want to carry out federal law differently from the federal executive, their most powerful objection sounds not in federalism, but rather in the separation of powers: they try to tar the federal executive’s choices as inconsistent with the statute that governs state and federal action alike. Rather than challenge the raw exercise of federal power, states instead challenge the faithfulness of the executive to the statutory scheme. They rely on power granted to them by one part of the federal government to contest the decisions of another part, and, whether rightly or wrongly, they cast themselves as Congress’s faithful agents, in contrast to a wayward executive branch.44

This state role is thus quite different from a traditional federalism conception pitting autonomous states against an autonomous federal government; integration, not separation, is key. Because a state’s strongest claim of right when it disagrees with the federal executive about how to carry out federal law comes from an appeal to the underlying statute, the federal law—and not arguments about state power as such—is where states turn regardless of what is motivating their challenges. Whether in private conversations between state and federal administrators, discussions mediated by legislators or the public, or filings in lawsuits, states trying to affect the federal executive’s decisions or chart a different course force the federal executive to defend itself with reference to the congressional grant of authority. Parsing the federal government proves critical to understanding state challenges.

Recognizing this intersection between federalism and the separation of powers does not only reorient our thinking about federalism; it also helps us to think differently about the separation of powers.45 The rise of the administrative state has long fueled concerns about the aggrandizement of executive power and the attendant demise of the separation of powers and checks and balances within the federal government.46 As Corwin himself observed, the passing of dual federalism was also the passing of a particular vision of the separation of powers: “the Federal system has shifted base in the direction of a consolidated national power, while within the National Government itself an increased flow of power in the direction of the President has ensued.”47 But these two concerns may answer more than exacerbate one another. Because states have been folded into the federal executive branch, state administration of federal law reproduces checks and balances within the enlarged executive domain.

Sometimes these checks and balances revolve directly around Congress. State administration of federal law may substitute for congressional monitoring of its delegates, forcing both state and federal administrators to defend their decisions in terms of the underlying statute. Often, however, a focus on Congress will not yield much. Especially with respect to the broad delegations that exercise critics of the administrative state, Congress’s intent may simply be that its delegates make the hard decisions. The very feature that leads to executive aggrandizement in the first instance also complicates efforts to cabin executive power with reference to congressional intent. The Clean Air Act, for example, confers substantial discretion on the executive branch, providing that regulation should ensue when in the EPA Administrator’s “judgment” emissions “cause or contribute to air pollution which may reasonably be anticipated to endanger public health or welfare.”48 Executive discretion is even more pronounced in the realm of immigration given that enforcement questions are by default subject to the executive’s policy choices, resource allocation, and particular judgments. Indeed, the Supreme Court’s decision in Arizona v. United States rests on the view that Congress intended the federal executive to set enforcement priorities and thus that federal legislative and executive choices cannot be parsed in the way the state argued.49

Coming up against the limits of traditional separation of powers analysis, state administration in such cases does not so much influence the relationship of the federal executive and legislative branches as generate, at one remove, the competition that is lacking between these branches. Rather than checks and balances between Congress and the federal executive, we see checks and balances between the states and the federal executive. State administration tees up both disputes about the powers and intentions of the branches of the federal government and nationwide public debates about federal law. It forces the executive branch to have conversations it wants to avoid. As these conversations are broadcast nationwide, the lines between state and federal become further blurred. The challenger state and federal executive each find supporters and detractors throughout the country, embedded in the federal government and state governments alike.50

When California sought to regulate greenhouse gas emissions pursuant to federal statutory authority, for instance, it forced the Bush Administration to defend its actions before a broad public. The EPA had to offer detailed reasons for its decisions and justify its preferred policies in ways it might have sidestepped were it the sole administrator of the law. And it found itself a target not only of angry state politicians, but also of angry federal politicians who agreed with the state’s substantive commitments and hauled agency officials in to testify. The controversy also generated substantial public debate, as the state-federal tussle crystallized the issues. This was no mere academic dispute: a state was ready to regulate greenhouse gas emissions under the Clean Air Act and, for better or worse depending on one’s view, the EPA was standing in its way.

When Arizona sought to ramp up enforcement of federal immigration law, it similarly forced a conversation that the Obama Administration was not sure it wanted to have.51 To invalidate Arizona’s law, the federal executive branch had to spell out its own decisions about immigration enforcement. While the President invoked statutory grants of authority and congressional priorities, he also suggested that with a gridlocked Congress, it had fallen to his Administration to shape immigration policy. But with states like Arizona contesting federal policy and seeking to enforce their own schemes, the Administration could not simply cite executive discretion. It had to defend its policy commitments and priorities before both the Supreme Court and the nation.

B. States as Another Congress

States do not only reproduce checks and balances within the executive’s domain. They also stand in more directly for Congress as authors of federal law, drafting new versions of federal provisions when Congress permits them to do so. This state role is particularly apparent with respect to what David Barron and Todd Rakoff have recently called “big waiver.” Instead of empowering agencies to set policy when Congress has left a gap, as traditional delegation would have it, big waiver grants agencies the power to decide whether the rules Congress has adopted should be dispensed with or replaced. It is the delegation of the power to unmake and change law Congress has made.52 Barron and Rakoff cite the signature regulatory initiatives of the last two Presidents—President Bush’s No Child Left Behind Act, and President Obama’s Patient Protection and Affordable Care Act—as evidence of big waiver’s rise: “Each of these laws sets forth a fully reticulated, legislatively defined regulatory framework. And yet each has been targeted for nearly wholesale administrative revision pursuant to seemingly broad waiver provisions Congress included in the very same statutes.”53

While Barron and Rakoff assess big waiver in traditional constitutional and administrative law terms—that is, with respect to the branches of the federal government—these and many other big waiver schemes involve a critical third party: the states. When the federal executive waives a statutory requirement and opens the way for rewriting federal law, the states then step in as drafters. In a variety of significant statutory schemes, the federal executive may unmake the law Congress has made, but it is the states that remake it.

The No Child Left Behind Act of 2001 (NCLB), for example, imposes stringent conditions on state educational systems in return for billions of dollars of aid. The core of the program requires states to adopt academic standards, test students based on those standards, and demonstrate “adequate yearly progress,” culminating in 100% proficiency by the 2013-14 school year.54 Even as it imposes detailed requirements on the states, however, NCLB grants the Secretary of Education the authority to “waive any statutory or regulatory requirement of this chapter,” with limited exceptions.55 To receive a waiver, a state must show, among other things, that the waiving of particular requirements will improve academic achievement and enable the state to measure students’ annual progress toward specified educational goals.56 Subject to the Secretary’s approval, then, states may significantly alter the federal law, and a vast majority of states are doing so. From Maine to Nevada, Alaska to Florida, forty-two states, as well as Washington, D.C., and Puerto Rico, are redrafting the federal requirements pursuant to waivers.57 As Congress has failed to amend NCLB, states have become a disaggregated national legislature.

The Patient Protection and Affordable Care Act of 2010 (ACA) similarly authorizes the Secretary of Health and Human Services to waive the main provisions of the Act to allow state innovation that provides health care coverage at least as comprehensive and affordable as that required by the Act itself.58 The Act also puts a thumb on the scale in favor of waiver: the Secretary must notify Congress when she denies a waiver and provide reasons for the denial.59 While the provision is not scheduled to take effect until 2017, early signs suggest states may use waivers to substantially revise the statute, including, in some cases, creating the single-payer health care system that lacked majority support in Congress.60 Through a series of smaller waivers, moreover, states are already changing portions of the statutory scheme.61

NCLB and ACA are prominent recent examples, but they are far from alone. Many other conditional spending statutes allow the states to depart from federal prescriptions upon receipt of a waiver. For instance, states have received thousands of waivers under the Medicaid statute and various welfare statutes, including some that have significantly modified statutory commitments.62 In each case, states stand in for the federal legislature.

Although what we might call “big waiver to states” contains a nod to dual federalism—if states were once the exclusive lawmakers in areas like health and education, today they are critical lawmakers within national programs—it is also an inversion. The power states have to write the law is not based on their sovereign status. Instead, it follows from a congressional grant of authority, subject to federal executive approval and superintendence.63 States serve as parts of an expanded national government, not separate polities. When they act pursuant to a waiver, moreover, they instantiate a variety of different versions of a single national program, not many separate state programs. As is true of state administration of federal law, then, state authorship of federal law poses a challenge to conceptions of federalism grounded in autonomy.

It also poses a challenge to conceptions of federalism grounded in distinctive state interests. States’ various versions of national programs tend to find supporters and detractors across the country, including within the federal government. Sometimes federal support and opposition can be mapped onto the executive and legislative branches; more often, as I will elaborate in Part III, it is necessary to draw finer-grained distinctions. Modifications of the ACA such as a single-payer system, for instance, were advanced, though defeated, during debate on the federal bill.64 State waivers give the losers in that fight a second chance: indeed, supporters introduced the expansive state waiver provision at least in part to facilitate state-based single-payer systems.65 Similarly, most states opting out of certain NCLB requirements have adopted the Common Core standards, a curriculum embraced by President Obama and his Secretary of Education but opposed by other states and a number of congresspersons.66

These examples, particularly the latter, might raise a distinct concern about the effects of state waivers on the separation of powers. I have suggested, in contemplating states as co-administrators of federal law, that states may check the federal executive. Big waiver seems to present just the opposite possibility: that states may collude with the executive branch to sidestep Congress. Are the states and the Obama Administration, for example, teaming up to get around NCLB through initiatives like the Common Core curriculum and Race to the Top? Rather than save the administrative state from concerns about executive aggrandizement at the expense of Congress, perhaps states’ role in national governance exacerbates such concerns.

While further study of these waiver regimes is warranted, early signs seem to point the other way. For one thing, it bears repeating that Congress drafts the waiver provisions in the first instance. When the federal executive authorizes states to amend federal law, it does so pursuant to statutory authority. And while many fears about executive aggrandizement concern Congress’s giving away its powers freely rather than the executive’s seizing them—that is, among other things, the worry about legislative delegation—states here introduce the sort of multiplicity that is feared lost from unilateral exercises of executive power. If a leading concern about Congress’s giving its powers to the executive is that a single actor (or, at least, a single branch) will have too much authority, including states in these schemes answers the concern by parceling out authority among multiple actors. As I have suggested with respect to administration, it is the recasting of checks and balances that is more significant than the bolstering of preexisting executive-legislative checks. States that receive waivers not only frame disputes about the powers and intentions of the branches of the federal government, but also generate nationwide discussion about federal law.

There is, for instance, more likely to be public visibility and debate when both the states and the federal executive are cohabiting a statutory scheme. When they act pursuant to waivers, state officials tend to call attention to the scheme to trumpet their policy innovation and leadership. So too, even though waiver schemes are premised on state-federal cooperation—the states have to receive permission from the federal executive and thus have incentives to play nice—that does not foreclose antagonism. If the federal executive grants waivers to certain states and not to others, we can be confident that disfavored states will complain, and even states that start off cooperative may not remain so. Recently, for instance, Utah and Nevada sought flexibility in administering their welfare programs, and the Obama Administration offered to waive the work participation requirements of the Personal Responsibility and Work Opportunity Reconciliation Act under the condition that more recipients find jobs.67 But almost as soon as these waivers became a real prospect, the Governors of Utah and Nevada criticized the executive for running around Congress and gutting the statute.68 While states’ ability to place executive decisionmaking in the spotlight does not fully answer questions about state-executive collusion, it reveals that these questions are more complicated than they first appear. Especially as compared to unilateral executive action, big waiver to states fosters transparency, accountability, and political debate that may check rather than enhance federal executive power.69

A second separation of powers point is also noteworthy: big waiver to states reintroduces legislators as policymakers. On one account of the separation of powers, Congress is responsible for making the law because it is a deliberative body whose members are subject to frequent and staggered elections that make them responsive to the people.70 This type of normative matchmaking gets lost in the administrative state when members of the executive branch are setting policy. But big waiver to states tends to rematch the policymaking function with the legislative branch. Of course, it is the state legislative branch, not Congress, but state legislatures are also deliberative bodies subject to frequent elections. The administrative state’s conflation of legislative and executive authority, coupled with big waiver’s conflation of state and federal lawmaking, thus presents an opportunity to restore legislatures as lawmakers within a national domain of administration.

III. politics: federalism as staging ground

Taking administration seriously poses a challenge to one aspect of the federalism literature’s insistence on state-federal separation: the focus on state autonomy. But integration runs deeper still. Not only do states participate in federal statutory schemes as administrators or drafters; in so doing, they also tend to further interests that cannot neatly be said to be state interests. Instead of opposing the federal government writ large or advancing distinctive state-based interests against national interests, states’ role in federal statutory schemes empowers them to instantiate competing views of national policy—in particular, competing Democratic and Republican views—that exist at both the state and federal level.

This merging of state and national interests is not limited to cooperative federalism schemes. Even when states are not carrying out federal law, they stage competition between the Democratic and Republican parties and enable national networks to translate discrete political agendas into law by harnessing the tools of direct democracy. Partisanship and political networks do not so much offer states channels to vindicate their particularistic interests as enable them to flesh out nationwide controversies. As with administration, attending to how states pluralize, rather than stand apart from, national politics best captures federalism’s contemporary vitality.

A. Partisan Federalism

If state administration of federal law injects diversity and competition into federal schemes, what generates this diversity and competition? Once we accept that state challenges to the federal government find supporters and detractors at both the state and federal level, it no longer makes sense to focus on distinctive state and national interests. More often than not, it is competition between the Democratic and Republican parties that motivates state challenges to particular federal policies. Such challenges advance one set of national interests against another set of national interests, not state interests against national interests.71

Return for a moment to some examples from Part II: greenhouse gas emissions, immigration, education, and healthcare. In each case, fights that wear federalism’s garb are revealed, upon even a cursory inspection, to be partisan warfare. When California fought with the Bush Administration about the regulation of greenhouse gas emissions under the Clean Air Act, the state advanced approaches to regulation that had previously been broached, without success, by Democrats in Congress; Senate and House Democrats, in turn, supported California’s waiver application and investigated the EPA’s denial. The cleavage was primarily partisan—Democrats wanted to regulate greenhouse gas emissions and Republicans did not. But this partisan fight played out in federalist terms because California’s role in the statutory scheme gave it a unique set of tools.72

So too, the fight between Arizona and the Obama Administration has been a partisan one.73 Republicans at both the state and federal level have sided with Arizona, while Democrats at both the state and federal level have sided with the Administration. Again, states are critical players because they can take actions that federal politicians cannot. Arizona passed its own law and forced a confrontation with the federal executive branch instead of simply stating an opposing view on the floor of Congress.

While the No Child Left Behind Act initially seemed a hallmark of bipartisanship—the rarest of bills championed by both President George W. Bush and Senator Edward Kennedy—recent months have seen partisan fights over its implementation. A growing number of red states are opting out of the Common Core, arguing that this National Governors Association-spearheaded curriculum amounts to a federal takeover of education.74 Once again, a national partisan debate looks like a debate about state versus federal authority. Because they have become the relevant legislators and administrators, state politicians who dislike NCLB exercise powers that their federal counterparts lack, even as they are supported and goaded by federal actors.75

Perhaps the clearest example of partisan competition in recent years involves the Patient Protection and Affordable Care Act. Almost all of the wrangling has assumed a state-federal cast: it was a group of states that sued to invalidate the Act as overstepping federal authority, and it is today states that are refusing to expand Medicaid and to set up insurance exchanges. Yet these resistant states are united by one thing—the Republican Party affiliation of their leaders—and they have been supported by federal Republican politicians, no surprise given that the Act was a Democratic President’s signature achievement and not a single Republican member of Congress voted for the bill. Meanwhile, Democratic states have aligned themselves with the Obama Administration.76 The arguments we are hearing for state versus federal authority are real enough, but they are motivated by individuals’ partisan commitments, not which level of government they inhabit.

In these and many other instances, we see political actors using state and federal governments to articulate, stage, and amplify competition between the political parties. This phenomenon—which I call partisan federalism77—follows from the integration of state and federal actors in two key respects. First, the state and federal governments operate in overlapping rather than separate domains; from emissions regulation to immigration, from healthcare to education, no regulatory domain is either the federal government’s or the states’ alone.78 Second, our political parties have become ideologically cohesive and polarized at both the federal and state levels. Democratic state officeholders in Missouri, Mississippi, and Montana today have much more in common with each other and Democratic federal officials than they would have in the middle of the twentieth century. Occupying the same regulatory space as the federal government while being governed by representatives of polarized political parties, states become critical platforms for the party out of power to fight the party in power in Washington.79

In contrast to existing accounts’ suggestion that the partisan integration of state and federal actors preserves a space for states to advance their separate interests, today’s partisan integration is integration all the way down. More often than states seizing on their connections to federal politicians to further distinctive state ends, we see state and federal actors alike seizing on states’ powers to further shared partisan ends.

Partisan federalism thus reorients our understanding of contemporary federalism. Rather than posit integration as a basis of separation—treating regulatory overlap and partisan cohesiveness as means by which states may advance particularistic state interests—it leads us to recognize that integration yields integration, that regulatory overlap and partisan cohesiveness render states and the federal government a single stage for national partisan competition. The challenge posed to federalism scholarship by taking politics seriously is perhaps even more substantial than that posed by taking administration seriously. Not only do states participate in the project of national governance, but state and national interests are one and the same.

Again, this challenge underscores the need to think in more nuanced terms about the national side of federalism. States’ critical role in staging partisan conflict concretizes the diversity of interests that may be properly understood as “national.”80 When states challenge federal policies on partisan lines, they flesh out competing partisan positions that cannot simultaneously be instantiated by the federal government. While the minority party in Congress makes speeches or engages in obstructionism, states rely on a distinct set of powers. They may, for instance, enact their own legislation to dissent from federal policy and chart a different course. Or they may administer federal laws in ways not intended or welcomed by the federal administration. Because party politics, and not something about state authority as such, is the motivating force, states undertaking these challenges find supporters and detractors both in the federal government and among the fifty states. Furnishing a variegated institutional terrain for substantive political conflict, states make real and visible for all Americans various partisan alternatives.81

Focusing on partisanship also suggests that states multiply opportunities for Americans’ national identification.82 As much social science literature teaches us, partisanship is a powerful sociopolitical identity that informs not only what we believe about particular issues, but also how we understand ourselves as Americans and how we form communal attachments.83 If partisanship binds Americans to the nation and to each other, so too does it create the conditions for disaffection and estrangement when one’s party is out of power in Washington, and in particular when one’s party does not hold the presidency.84 In such instances, states become important sites of partisan identification: Americans who feel alienated from the federal government can turn to the states and know there are government institutions controlled by their “team.”85 This form of state identification inheres not in something categorically distinctive about the states vis-à-vis the federal government, but rather in their shared ability to represent the political parties yet the divergence in how they do so at any given time.86 Partisan identification with states may therefore be a way of affiliating with the nation while distancing oneself from the current federal government.87 Appreciating states as national actors underscores that state identification need not be opposed to national identification but may be a means of preserving it.

B. Networked Interests and Direct Democracy

While competition between the Democratic and Republican parties animates much state-federal divergence, there are many political issues that don’t inform national partisan competition, whether because they are subject to ample intrapartisan dispute, too small and seemingly insignificant to consume the parties’ attention, or just too new to have yet established a partisan valence. For individuals invested in these issues—ranging from the labeling of genetically modified foods, to the prevention of animal mistreatment, to the recognition of assisted suicide—states are the governments that translate political commitments into practice. Yet these issues are not “state” issues in any necessary sense. Not in a legal sense: the federal government could set policy or alter existing policy. And not in a communal or identitarian sense: there are individuals throughout the country pushing for reform; these are not local issues of interest only to particular states’ residents. When states engage these political struggles, they do so on behalf of people both inside and outside their borders, and they are supported and opposed by Americans nationwide. These are, then, state issues in a much thinner instrumental sense. Because it is easier to pass new state laws than new federal laws, time and again states prove more accessible fora for nationwide movements to promote their ultimately national agendas.88

Indeed, a critical difference between the forms of state and federal governance underlies a good deal of this activity: nearly half of the states recognize some version of direct democracy. While the Democratic and Republican parties are increasingly shaping such contests,89 ballot initiatives and referenda create a space for lawmaking outside the usual partisan processes, a popular nonpartisan (rather than bipartisan) space that does not exist at the federal level. Notably, however, participation in state referenda and initiatives is not limited to state residents. Only state residents may vote at the end of the day, but the gathering of petition signatures, the get-out-the-vote efforts, the staffing of campaigns, the advertising, and, underlying most of this, the bankrolling tend to come from outside the state as much as inside of it. State-level direct democracy thus provides a forum for Americans nationwide to participate in political contests that may fall outside of national party politics.

Consider, for instance, the movement to legalize marijuana. This is a decidedly national movement. Yet individuals and groups across the country advocating legalization are not focused on the federal government. Instead, they are looking to the states, relying on direct democracy by bringing nationwide resources to bear on particular state initiative contests. Recognizing that the public was more willing to support medical marijuana than were either Democratic or Republican politicians, for instance, various national groups, such as Americans for Medical Rights, NORML, and the Marijuana Policy Project, began in the 1990s to seize on state ballot initiatives to push for the legalization of medical marijuana.90 Since 1996, twenty states have legalized medical marijuana, more than half through ballot initiatives.91 And in these contests, out-of-state individuals and groups have provided critical support.92 Americans across the country have appreciated state initiatives as a way to lay the groundwork for national change, whether through the sort of cascade of state laws that has in fact occurred93 or perhaps by ultimately triggering a change in federal law.

More recently, the legalization movement has focused on a bigger target: legalization of all personal use of marijuana. Again, direct democracy has been the critical means for national groups to advance this ultimately national goal. In 2012, voters in Washington and Colorado approved measures legalizing personal use of marijuana, the first successes for the movement following defeats in several other states. The official sponsor of Washington’s Initiative 502, New Approach Washington, included staff members who also worked for, or had recently left positions with, the ACLU and NORML,94 and nearly two-thirds of the funds for the campaign were raised outside of Washington.95 Colorado’s campaign was similarly financed primarily by out-of-state funds and enjoyed the support of staff from national organizations.96 As with the ballot initiatives to legalize medical marijuana, Americans nationwide recognized the contests in Washington and Colorado as national, not purely local, political contests.97

The movement to legalize marijuana is far from the only political movement that has turned to state ballot initiatives to achieve national ends. The term limits movement that swept the country in the 1990s was a national one that progressed through state initiatives.98 The movement to label genetically modified foods is currently focusing on direct democracy, as out-of-state donors provide funds both for and against labeling.99 From ballot initiatives that would either recognize or prohibit same-sex marriage, to those that would ban fracking, to those that would protect animals from mistreatment, recent years have witnessed a host of other direct democracy contests fueled by networked interests.100

National networks pushing national agendas through the states pose a serious challenge to conceptions of federalism grounded in distinctive state interests. One might try to resist this framing by telling stories linking particular initiatives to particular state, or at least regional, cultures. Perhaps the fact that Colorado and Washington legalized marijuana, for instance, indicates there was something peculiarly western about these initiatives. Yet multiple stories can always be told: if the legalization of marijuana may be tied to a libertarian, western identity, so too may the fact that some police officers’ groups want to legalize marijuana tie the cause to a law-and-order rationale. The way in which initiatives spread from state to state complicates efforts to ground them in particular state cultures and to limit the stories we tell about them. For example, states that have legalized medical marijuana by initiative include Alaska, Arizona, Massachusetts, and Michigan, hardly a unified regional group.101 Narratives linking state ballot initiatives to distinctive state cultures are likely to be opportunistic. And there is no getting around the fact that what differences do exist across the states tend to be grounds not for truly local grassroots organizing, but rather for targeting by national movements looking for the most fertile ground to plant their policy ideas. Ultimately, these contests can only be fully understood by accepting their national dimensions: they are generally spearheaded by national groups, they attract substantial out-of-state support and opposition, and the ultimate goal is nationwide change. Americans throughout the country are seizing on the states to advance national agendas.

To frame these contests as national is not to deny the significance of the states; it is to reconceptualize their significance, to stop focusing on state-federal separation and to instead embrace states as important national actors. States are critical sites for national contests insofar as they allow national networks to find a toehold and translate their political commitments into law. And states help shape nationwide policy, as other states follow suit, as support builds so that federal politicians take up the issue, or even as organized interests that have opposed federal action come to realize that disparate, and perhaps more stringent, state laws are worse than a single national regime and thus begin to push for federal regulation.102 States extend the arena for national governance beyond Washington, D.C., generating more opportunities for all Americans to advance their political commitments. Indeed, the same cultural, political, and technological forces that undermine the distinctiveness of states qua states make it easier for Americans nationwide to know what is happening in a variety of states, to push for changes in different states, and to understand other states’ actions as bearing on their lives.

Conclusion: Federalism’s Afterlife

Announcing dual federalism’s passing, Corwin wondered whether the states could be “saved for any useful purpose, and thereby saved as the vital cells that they have been heretofore of democratic sentiment, impulse, and action.”103 Our existing answers to this question seek to preserve state autonomy and distinctive interests while proposing new channels for the protection of these forms of state-federal separation. Yet today state and federal governance and interests are more integrated than separate. Instead of focusing on the life or death of American states as autonomous, independent actors, we should think more seriously about federalism’s afterlife as a form of nationalism.

Acknowledging the conjunction of federalism and nationalism gives us new purchase on states’ contemporary role. As this Essay has explored, states frequently act as co-administrators and co-legislatures in our federal statutory schemes; they carry out federal law alongside the executive branch and write the law alongside Congress. Lacking an autonomous realm of action in critical areas, states nonetheless imbue federal law with diversity and competition, aligning themselves with certain federal actors to oppose others. As differences that exist in Washington, D.C., assume concrete form in the states, states also affect checks and balances across the branches of the federal government. We must consider the separation of powers to understand federalism, and so too must we consider federalism to understand the separation of powers.

States do not only set national policy as key players in federal statutory schemes; they also advance a variety of national interests, rather than particularistic state interests. States play out competition between the Democratic and Republican parties, expanding the governance tools of the party out of power in Washington. And they offer staging grounds for national networks seeking to advance their agendas in discrete state fora by, for instance, harnessing direct democracy. In these various respects, states participate in national political competition on behalf of all Americans, and individuals throughout the country look to the states as well as the federal government to represent their interests.

Taking administration and politics seriously ultimately points not to preservation of traditional conceptions of federalism, but instead to a new anti-essentialist view of states and their role in our compound republic. States need not enjoy an autonomous sphere of action nor advance distinctive state interests. Precisely insofar as they are disaggregated sites of national governance, not separate sovereigns, they continue to serve as vital cells of “democratic sentiment, impulse, and action.”104