State Public-Law Litigation in an Age of Polarization

Article - Volume 97 - Issue 1


Public-law litigation by state governments plays an increasingly prominent role in American governance. Although public lawsuits by state governments designed to challenge the validity or shape the content of national policy are not new, such suits have increased in number and salience over the last few decades—especially since the tobacco litigation of the late 1990s. Under the Obama and Trump Administrations, such suits have taken on a particularly partisan cast; “red” states have challenged the Affordable Care Act and President Obama’s immigration orders, for example, and “blue” states have challenged President Trump’s travel bans and attempts to roll back prior environmental policies. As a result, longstanding concerns about state litigation as a form of national policymaking that circumvents ordinary lawmaking processes have been joined by new concerns that state litigation reflects and aggravates partisan polarization.

This Article explores the relationship between state litigation and the polarization of American politics. As we explain, our federal system can mitigate the effects of partisan polarization by taking some divisive issues off the national agenda, leaving them to be solved in state jurisdictions where consensus may be more attainable—both because polarization appears to be dampened at the state level and because political preferences are unevenly distributed geographically. State litigation can both help and hinder this dynamic. The available evidence suggests that state attorneys general (who handle the lion’s share of state litigation) are themselves fairly polarized, as are certain categories of state litigation. We map out the different ways states can use litigation to shape national policy, linking each to concerns about polarization. We thus distinguish between “vertical” conflicts, in which states sue to preserve their autonomy to go their own way on divisive issues, and “horizontal” conflicts, in which different groups of states vie for control of national policy. The latter, we think, will tend to aggravate polarization. But we concede—and illustrate—that it will often be difficult to separate out the vertical and horizontal aspects of particular disputes and that in some horizontal disputes the polarization costs of state litigation may be worth paying.

We argue, moreover, that state litigation cannot be understood in a vacuum but must be assessed as part of a broader phenomenon in American law: our reliance on entrepreneurial litigation to develop and enforce public norms. In this context, state attorneys general often play roles similar to “private attorneys general,” such as class action lawyers or public interest organizations. And states, with their built-in systems of democratic accountability and internal checks and balances, compare well with other entrepreneurial enforcement vehicles in a number of respects. Nevertheless, state litigation efforts may not always account well for divergent preferences and interests within the broad publics that the states represent, and this deficiency becomes particularly important in politically polarized times. Although our account of state litigation is, on the whole, a positive one, we caution that state attorneys general face a significant risk of backlash by other political actors, and by courts, if state litigation is (or is perceived to be) a bitterly partisan affair.


This Article explores two highly salient phenomena in American politics and seeks to better understand the relationship between them. The first is the advent, largely over the past few decades, of high profile public-law litigation by state attorneys general (AGs) acting on behalf of state governments and citizens—the sort of thing that Texas Attorney General Greg Abbott meant when he described a typical workday as, “I go into the office, I sue the federal government and I go home.”[3] Such cases include red-state challenges to the Affordable Care Act (ACA) and the Obama Administration’s Deferred Action for Parents of Americans program (DAPA), and “blue-state” challenges to the Bush Administration’s environmental policies and the Trump Administration’s travel bans. Yet state AGs’ influence over national policy extends beyond those well-known examples. It also includes significant increases in amicus curiae filings by state governments, multistate litigation by groups of AGs working together to combat questionable business practices,[4] as well as state efforts to enforce federal law in ways that may deviate from the national Executive’s priorities.[5] State AGs are playing a pivotal role in some of the most important national political debates of the day, and they are doing so largely through entrepreneurial litigation.

The second phenomenon is political polarization. Americans are more divided today along partisan and ideological lines than they have been for some time.[6] This polarization has important consequences, rendering national politics unusually contentious and often undermining our capacity for self-governance. It may cause legislative gridlock, prompting unilateral presidential action. At other times, polarization can lead to more extreme national legislation.

State public-law litigation, especially in its most recent manifestations, seems at first glance to be a symptom of the broader polarization in national politics.[7] It is no accident that AG Abbott—a Republican—made his comment about suing the federal government during the Obama Administration. Now that the political tables have turned, the homepage for the Democratic Attorneys General Association reads, in large orange font, “Democratic Attorneys General are the first line of defense against the new administration.”[8] These partisan divides play out across the policy spectrum. For much of the last decade, for example, coalitions of blue states committed to stricter environmental safeguards have litigated to prod the federal Environmental Protection Agency to more stringently regulate emissions of greenhouse gases.[9] Over the same period, red-state coalitions have likewise litigated to prevent such regulation.[10] Similarly, red- and blue-state coalitions confronted one another over the constitutionality of the ACA’s individual mandate and Medicaid expansion.[11] And state amicus curiae filings in the same-sex marriage litigation likewise reflect a passionate red/‌blue divide over the pace of social change with respect to sexual orientation and family relationships.[12] In many instances, state public-law litigation is a vehicle for expressing the same divisions that convulse American politics generally.

As state litigation has grown in volume and prominence, it has drawn more attention in both the academic literature and the popular press. Much of that attention has been negative.[13] At least since the multistate tobacco litigation of the 1990s, critics have argued that state suits may effectively result in national lawmaking by settlement, coercing defendants and circumventing federal lawmaking processes.[14] But new lines of critique have emerged in more recent years, suggesting links between AG litigation and trends in partisanship and polarization. Critics contend that AGs are abandoning their traditional role “as representatives of their states,” in which the goal of litigation was to vindicate the long-term, institutional interests of states qua states.[15] Rather than focusing on threats to state autonomy, AGs today can be found pushing for more federal regulation[16] or supporting claims “of individuals as opposed to the states themselves.”[17] And, as noted, they often are doing so in partisan clusters rather than banding together as states to promote state interests in a politically neutral manner. James Tierney, a former Maine AG and leading observer on these matters, worries “that the AGs become seen as one more lawyer . . . on the make, and that undercuts the credibility of the office itself.”[18]

We have some sympathy for those critiques, but we think the picture is far more complicated than critics acknowledge—in part because the concept of “state interests” is itself complicated. To understand state litigation, it helps to situate it within broader theories of federalism. When most people think of federalism, they imagine “vertical” conflicts between the states and the federal government, conflicts in which states typically are resisting assertions of federal power so as to maximize their own regulatory autonomy. But our federal system also addresses “horizontal” conflicts in which powerful states (or groups of states) attempt to impose their will on others. Vertical conflicts are, for the most part, about who decides—the states or the federal government. Horizontal conflicts are about what policies will prevail.

From this perspective, the critiques of state litigation are easy to understand. When states challenge federal policy in vertical cases, they are performing their traditional role in a federal system—throwing off the federal yoke so that they can govern themselves. To the extent that state AGs argue in favor of federal law in such cases, they look like traitors to the cause. Horizontal cases, similarly, appear to be at odds with the states’ shared interest in autonomy. When state AGs argue in favor of individual claims of constitutional right, for example, or use federal law to reform widespread business practices, they seem to be vindicating their home states’ regulatory interests—interests that tend to track partisan divisions—at a cost to the broader institutional interests of the states as such.

This contrast between vertical and horizontal conflicts is a helpful frame for considering state public-law litigation. But the line between such conflicts—and between states’ institutional and regulatory interests—is often fuzzy and contested. As we explain, many seemingly vertical conflicts have horizontal aspects and vice versa. For regulatory challenges that cannot be solved without collective action—certain environmental issues, for example—pro-regulatory states have little choice but to push for nationwide solutions. In such circumstances, states’ institutional and regulatory interests merge, and states exercise their sovereignty by appealing to federal power. Things look different to anti-regulatory states, and those disagreements will often play out along partisan lines. It does not follow, however, that the state AGs on either side of the case are putting politics before state interests. Likewise, under our contemporary model of federalism, states have an interest not only in doing their own thing but also in participating in national politics—an interest that may aggravate horizontal conflict.

State litigation must also be viewed in the evolving context of public-law litigation generally in American law. Our legal system is exceptional in its reliance on litigation and courts to resolve conflicts and articulate policies that, in other systems, would fall into political or bureaucratic channels.[19] And rather than rely exclusively on enforcement by the national Executive, federal law frequently authorizes entrepreneurial litigation by private attorneys general. When state AGs enforce federal law, they play a similar entrepreneurial role to class action attorneys or public interest organizations. The same thing is true when state AGs rely on their own state laws but cooperate to secure nationwide judgments or settlements that impose a de facto national regulatory solution on a particular industry.

An important response to criticisms of state litigation, then, is to ask “compared to what?” When states sue to enforce the Clean Air Act or the securities laws, or to challenge the ACA or the Trump travel bans, they are playing a similar role to the Sierra Club, the ACLU, or class action plaintiffs’ lawyers. If states were precluded from bringing such suits, their private analogs would remain. Yet, as we explain, there are good reasons—grounded in democratic accountability and in state governments’ unique institutional perspectives—to prefer state litigation to purely private mechanisms for aggregating diffuse interests.

Part I of this Article offers a sketch of polarization in the federal and state governments, tracing the relationship between polarization, national policymaking, and policy autonomy at the state level. We suggest that state autonomy can sometimes be a “safety valve” for polarized conflict at the national level. Part II then turns to state litigation. It charts the institutional development of state AGs’ offices and the expansion of doctrinal and statutory rights to sue, which have helped state AGs emerge as a particularly powerful group of lawyers. We then map the different sorts of claims that states use to shape national policy.

Part III turns to the relationship between state litigation and polarization. Few scholars have sought to study polarization in the work of state AGs, but the available evidence suggests that state litigation is indeed becoming more “political” in the sense that Democratic and Republican AGs increasingly are pursuing different causes or are lining up on opposite sides of the same cases. The impact on our broader politics, however, will often turn on the nature of a given lawsuit. We use the distinction between vertical and horizontal conflicts as a framework for normative assessment of state public-law litigation in an era of intense political polarization. Finally, we take up the comparative question in Part IV, situating state litigation within the broader phenomenon of public-law litigation as a mode of American governance. Although we think suits by state AGs compare favorably to other mechanisms of aggregate litigation, we warn that overly aggressive state public-law litigation may result in a judicial or political backlash that might undermine the benefits of this valuable institutional mechanism.

I. Polarization and Federalism

Contemporary American politics displays a level of political polarization that, while hardly unprecedented, is significantly greater than anything in recent memory.[20] For decades, American political scientists lamented the lack of clear programmatic differences between the major political parties; that state of affairs, they complained, deprived American voters of a meaningful choice at election time.[21] The present era thus plays out the old adage, “Be careful what you wish for.” (Alternatively, it embodies the Chinese curse: “May you live in interesting times.”) American politics—and the underlying society—finds itself divided between quite different conceptions of the good life, with strong and contrary implications for government regulation, fiscal policy, and individual rights.[22] The temperature of political debate has called into question whether our national political institutions can mediate and resolve these conflicts. And these changes in political climate have affected the weather at the U.S. Supreme Court, influencing both the sorts of cases brought before the Justices and the types of parties that bring them.

For students of American federalism, there is a certain irony to all this. A decade ago, prominent voices in the federalism literature took the position that American federalism is meaningless and unnecessary because American society lacks the kind of basic divisions that make federalism necessary in, say, Canada or Iraq.[23] This line of thought surely represented the conventional wisdom in terms of its basic assumptions, even if not everyone accepted the conclusion that America’s federal structure could safely be junked.[24] Scholars looking to defend federal structures were left searching for glimmers and vestiges of state identity that might sustain autonomous subnational institutions.[25] The question now, by contrast, sometimes seems to be whether Americans can find sufficient common ground to move forward together on common problems.[26] Federalism, we suggest, can help.

A. Polarization in National Politics

The Democratic and Republican Parties are more polarized today than they have been in decades—maybe more than a century, according to some measures.[27] What that means, in part, is that the contemporary Congress is marked by high levels of partisan sorting: Members are more easily sorted by party today than they were in the past.[28] There are fewer conservative Democrats and fewer liberal Republicans.[29] As a result, there is little or no overlap between members of the different parties.[30] Second, and closely related, is the notion of ideological divergence, which refers to the distance between the party medians.[31] That distance today is greater than at any time since the end of Reconstruction.[32]

A vigorous debate exists as to whether this polarization of politicians reflects a broader polarization of the public at large. One group views the public as basically moderate in its views but sees a fundamental disconnect between those views and a highly polarized political class.[33] Another group holds that polarization reaches much further down into the electorate.[34] But even if the public’s policy views remain moderate, surveys reveal high degrees of “affective” polarization.[35] Simply put, Democrats and Republicans don’t like each other very much—much less, it seems, than in the 1980s and 1990s.[36] And this partisan dislike has translated into a polarization of political trust, so that partisans report radically low levels of trust in government when the other party dominates the national government.[37]

Such polarization translates into sharp and sustained disagreement and a refusal to compromise across party lines. Evidence suggests that this effect is most severe when institutions are closely divided, as either party can realistically hope to gain control and neither can afford to give the other a victory.[38] When different parties control the House and Senate, the probable effect is gridlock—an inability to get things done because there’s no common ground for consensus.[39] The same is often true when one party controls both houses of Congress and the other party controls the White House. Unless the dominant party in Congress has a veto-proof majority, the President can block major legislation.

These obstacles can sometimes be overcome by appeals to the public at large. But low levels of political trust make it difficult for a president to go over the heads of partisan opponents in the Congress and appeal to moderates in the other party, as President Reagan was able to do in the 1980s.[40] The consequences are well known: gridlock means that Congress is likely to produce less federal legislation, and the bills that do emerge are likely to be less consequential.[41] Rather than addressing big, contentious questions, a gridlocked Congress will tend to enact symbolic legislation or to leave the critical choices to agencies.[42]

Things look different under unified government, of course. When the same party controls both houses of Congress and the Presidency, it can—in theory at least—accomplish quite a lot.[43] In times of unified government, the consequence of polarization should be more extreme legislation.[44] In that sense, polarization raises the stakes of control of the national government: if one party can win control of both Congress and the Presidency, it can dictate policy on virtually every issue people might care about and need not compromise with the minority party.

It is not obvious that the polarization we have described is a bad thing. After all, as we have already noted, American political scientists at the middle of the twentieth century longed for ideologically pure parties that would offer voters a clear choice; this, they thought, was the key to truly responsible government.[45] To assess whether political polarization is good or bad for the Republic would require its own article (or book), and we cannot offer a rigorous analysis here. Briefly, we would emphasize several specific concerns developed elsewhere in the literature. Some political scientists argue that unified government can produce legislation that is more extreme than many of the majority party’s own constituents would want—and thus inconsistent with the preferences of the majority of voters.[46] Moreover, our separation of powers effectively imposes supermajority requirements on most legislative action; as a result, polarization combined with a close division of the electorate results either in gridlock or diversion of government action into constitutionally dubious channels.[47] Finally, recent literature suggests that contemporary polarization is more “affective” than policy-driven; in other words, Americans have developed a strong dislike for persons on the other political “team” even though the actual policy differences between the parties are often minor.[48] It is hard to see any upside to polarization once it reaches that point. In any event, we sketch these reasons simply to give a sense of our priors. Our argument here must largely presuppose, rather than defend, the proposition that polarization is worrisome and in need of mitigation. Our question is whether federalism—and, in particular, state litigation—is likely to mitigate or exacerbate the polarization that worries us.

B. Polarization and the States

Federalism can operate as an important safety valve in polarized times, lowering the temperature on contentious national policy debates and creating opportunities for policymaking that may be impossible at the national level.[49] In evaluating this claim, it will help to distinguish between polarization within states and polarization among states. Some states, at least, seem to have less polarized politics than we see at the national level. In these states, bipartisan resolutions to divisive issues may well be possible. But even if states have similar political cultures to that at the national level, the distribution of political preferences is geographically uneven. This polarization among states—the now familiar divide between red and blue states—makes it possible to act on divisive issues in ways that avoid the all-or-nothing nature of national solutions.

1. Polarization Within States.—The patterns of polarization that define national politics today are not replicated in all of the states. In Massachusetts, for example, Democrats and Republicans can agree on a generous level of social provision and broadly libertarian social policies,[50] while Texas Republicans and Democrats tend to share a general commitment to a low-tax, small-government model.[51]

Precisely why this is so is difficult to pin down, but the available evidence suggests two (complementary) answers. First, state politicians may themselves be less polarized—in the sense of ideological distance—than their federal counterparts. Second, even if Democrats and Republicans are miles apart ideologically, the unique features of state government may dampen the effects of that distance.

To begin with, it appears that party identity varies across states: it means something different to be a Republican in Massachusetts than it does to be a Republican in Texas. In other words, partisan sorting is not as clear-cut at the state level as it is in Washington, D.C. Whereas there is vanishingly little overlap between the national representatives of the two parties, the picture looks different if one focuses on state legislators. Democrats elected to state office in Mississippi, Louisiana, and Arkansas, for example, are in some cases more conservative than Republicans in Delaware, Illinois, New Jersey, Rhode Island, Hawaii, Connecticut, New York, and Massachusetts.[52]

Ideological divergence is also muted—or at least more mixed—at the state level. A leading 2011 study of polarization in state legislatures found that the distance between party medians varied significantly from one state to the next.[53] California boasted the most polarized state legislature, leading a group of fifteen states in which ideological divergence was more pronounced than in Congress. The majority of state legislatures, however, were less polarized than Congress.[54] Similarly, five of the last six governors of Massachusetts—one of the bluest states there is—have been Republicans.[55] Last year, the current governor of the Bay State enjoyed the highest approval ratings in the nation, and several other Republican governors in blue states are similarly popular.[56] One recent analyst concluded that “Republican gubernatorial candidates are . . . able to be more moderate than Republican presidential candidates, and therefore tend to be more ideologically compatible with the Democrat-dominated electorates of blue states.”[57] Republican governors in Democratic states thus seem to do best when they take a moderate line on social issues and maintain significant separation from the national party.[58]

As this last point suggests, an inquiry into polarization in state government should heed, not only the ideological preferences of state officials, but also how those preferences translate into political behavior. Several characteristics of state government suggest that we might expect state politics to reflect less partisan conflict even if state officials are themselves fairly polarized.[59] For example, surveys indicate that state and local governments enjoy considerably higher levels of trust than the federal government. Researchers have been asking survey questions about trust in government for many decades, and trust has recently become central to some scholars of polarization.[60] Those scholars have generally focused on national-level measures of trust. But the survey questions have often included a comparative component that inquires whether citizens repose more trust in state or national institutions. This research concludes that “[c]itizens on average evaluate the performance of the federal government as significantly lower than that of the state and local governments, report less faith in the federal government to ‘do the right thing,’ have significantly lower confidence in the ability of the federal government to solve problems effectively, see the federal government as significantly less responsive than lower levels of government, and nearly 60% see the federal government as the most corrupt level of government.”[61] If polarization scholars are right that higher levels of trust make it more likely that partisans will make “ideological sacrifices” to create bipartisan solutions,[62] then that ought to be more likely at the state level.

Culture may also play a role in mitigating ideological polarization’s effects on state officials. State political cultures may be sufficiently distinctive that the range of partisan disagreement is narrower within them.[63] Daniel Elazar, for example, argued that certain states share a broader commitment to regulation and social provision based on having been originally settled by New England Puritans committed to those values.[64] Consistent with this view, Republican governors in New England have tended to support the more generous social welfare arrangements in those states while pushing fiscal conservatism around the edges.[65] We might further speculate that state political cultures include shared norms of political practice that inhibit the nastier forms of partisanship that entrench polarization.[66]

Or perhaps state and local governments deal with a large number of bread-and-butter issues—e.g., road maintenance, education, and crime control—on which the public may have limited tolerance for partisan posturing that prevents basic needs from being met.[67] We come from North Carolina—deep purple and closely divided. We just had a particularly nasty gubernatorial election and a fractious legislative session. But even in states like ours, pragmatic concerns like fixing potholes and reducing crime may moderate polarization’s effects. Unlike their federal counterparts, state politicians can’t spend all their time grandstanding; state governments have to get certain things done, and a lot of those things aren’t particularly ideological. Balanced budget requirements may further constrain them from the worst kinds of obstruction and kicking the can down the road. Successful Republican blue-state governors, after all, are frequently characterized as “pragmatic,” “non-ideological” managers who tend to decouple their own political fortunes from the national party.[68] Democratic governors in red states are, for now at least, fewer and further between. But the ones we have seem to have pursued a similar approach.[69]

Another important factor may be that (unlike North Carolina) many states are not as closely divided as the national government—they are not purple but consistently red or blue. As we’ve already noted, some research suggests that close divisions increase the incentives for political opportunism, as the minority party may hope to regain the majority if it can prevent the opposition from being successful.[70] In states where the minority is likely to remain in that position, by contrast, minority party-members may seek to have at least some voice through bipartisan cooperation. We might even be seeing some vindication of the Antifederalist notion that republican government—predicated on statesmanlike transcendence of narrow factional interest—is more likely to succeed in smaller communities. It’s hard to know for sure, and the question is ultimately an empirical one on which we presently lack much good evidence. We do have evidence, however, that polarization and its effects are less extreme at the state level.

One significant caveat is in order. A variety of research suggests that levels of polarization and mistrust are in part a function of the issue set that is salient to voters.[71] The comparatively sunny cast of some states’ politics may thus arise from the issues laid before state governments. If we were to devolve certain contentious issues from the national scene to state governments, that might well change the character of state politics. Perhaps habits of cooperation forged in filling potholes might bleed over into debates about transgender rights. But they also might not.

In sum, there are reasons to think that federalism can mitigate the effects of political polarization by offering alternative policymaking venues in which the hope of consensus politics is more plausible. As the next section details, taking divisive questions off the national agenda may moderate the overall polarization problem even if that is not true.

2. Polarization Among States.—The notion of an equally divided nation goes back all the way to the 2000 election.[72] But very few places in America are fifty-fifty in this way: “Geography matters.”[73] Within the states, relatively small differences in the correlation of partisan forces[74] significantly affect political outcomes, painting some state governments completely red and others blue.[75] In those states, even if state-level bipartisanship fails to generate effective policy on divisive issues, unified government might step in to fill the breach.

As of 2015, only nineteen states had divided government.[76] That number declined to eighteen after the 2016 elections, then slipped to seventeen after West Virginia Governor Jim Justice switched to the Republican party.[77] Thus, even those state legislatures with relatively high levels of polarization may be capable of avoiding gridlock and getting things done. In New Jersey, for example, Democrat Phil Murphy’s election as governor has made the Garden State one of eight states under unified Democratic control. “If Murphy has his way,” the Washington Post predicted, “New Jersey will become a proving ground for every liberal policy idea coming into fashion, from legalized marijuana to a $15 minimum wage, from a ‘millionaire’s tax’ to a virtual bill of rights for undocumented immigrants.”[78] Meanwhile, Republican-controlled states have “pursued economic and fiscal strategies built around lower taxes, deeper spending cuts and less regulation” and have adopted policies with respect to labor, education, and social issues that diverge sharply from blue-state strategies.[79]

To be sure, the combination of polarization and unified government can produce less compromise and more extreme policy in state governments, too.[80] But the stakes are lower for statewide, as compared to nationwide, solutions. At the very least, devolving decision-making authority to the states opens up opportunities for policy variation—not only among states, but also between the states and Congress. A flourishing federal system means that Democrats currently out of power in Washington, D.C. don’t just have to give up or focus on rearguard actions at the federal level; they can govern at the state level.[81] Especially when state government is unified, those Democrats can pursue a very different set of policies than those originating on Capitol Hill. The consequence may not be compromise, exactly, but it does offer a way to serve the preferences of people who identify with the minority party in Congress.[82]

A federalism-based modus vivendi is unlikely to satisfy devoted partisans on one side or another of any divisive issue. But as Michael McConnell has explained, “So long as preferences for government policies are unevenly distributed among the various localities, more people can be satisfied by decentralized decision making than by a single national authority.”[83] Moreover, when different jurisdictions can implement (and not simply advocate) their preferences, they can have significant and unexpected effects on the national debate. When the same-sex marriage issue became salient in the mid-1990s, for example, most states used the autonomy that the federal system afforded them to explicitly outlaw the practice. But some states permitted same-sex marriages to go forward, and over time the example of those new families helped bring about one of the most remarkable shifts in public opinion in American history.[84] State-by-state diversity may thus break up rigidly polarized political patterns over time, even if state political cultures are not significantly more warm and fuzzy than the national one.

C. How Federal Polarization Affects Federalism—And How State Litigation Can Help

We’ve argued that states can serve as safety valves for polarized national politics. In order for states to play those roles, however, the federal government must leave them room to maneuver. And there’s the rub: while polarization highlights the benefits of federalism, it also poses a distinct threat to state autonomy.

This point is most obvious under conditions of unified national government. As we explained above, polarization plus unified government is likely to produce more extreme policy. That means more federal overreaching—statutes that trench on state interests or that are more broadly preemptive in scope. Where that is true, states may find they have less space to act, and the benefits outlined above will be lost.

Divided government at the federal level can also hold threats to state autonomy, though the reason is less intuitive. At first blush, polarization plus divided government may seem like a boon for federalism: the less Congress is able to do, the more that’s left for the states.[85] But congressional gridlock may also produce more unilateral action by the federal Executive, in the form of executive orders and guidance, gentle and not-so-gentle nudges directed at agencies, and so on. This dynamic was reflected in President Obama’s “We Can’t Wait” campaign, for example. The campaign started with a speech in which the President said, “[W]e can’t wait for an increasingly dysfunctional Congress to do its job. Where they won’t act, I will.”[86] And it became a yearlong theme in the lead-up to the 2012 election cycle. In one year alone, the President announced more than forty executive actions packaged under the “We Can’t Wait” brand.[87] President Trump has shown few signs of retreating from executive-branch unilateralism, notwithstanding unified Republican control of government; he has used executive orders for (among other things) his controversial travel bans and efforts to strip “sanctuary cities” of federal funding.[88]

From a federalism perspective, there’s a lot not to like about unilateral executive action. Most obviously, it’s easier to do than running formal legislation through two chambers of Congress and the President. Many people believe that state interests are protected in the national political process through the close ties between national and state parties and politicians and the representation of states through their congressional delegations.[89] Others emphasize the many “veto-gates” in Congress that stand in the way of legislation.[90] These are the so-called political and procedural safeguards of federalism. And to the extent that states get protection in the legislative process, one might worry about federal policy being made in a far more streamlined fashion and centered in the executive branch, where states have no special voice.[91] Granted, states may find considerable freedom to shape federal policy in its bureaucratic interstices by proposing innovative ways to implement federal mandates or by dragging their feet on locally unpopular requirements.[92] But despite the practical importance of implementation authority, the leeway afforded is unlikely to be broad enough to accommodate the basic ideological conflicts that often characterize our polarized national debates.

This brings us to an additional way in which states can mitigate the effects of polarization—not through legislation and regulation, but through litigation: states can challenge federal action that arguably goes too far.[93] Anthony Johnstone has observed that “[i]f the primary virtue of federalism in these politically polarized times is the accommodation of diverse policy preferences . . . then attorneys general are uniquely qualified to give voice to those preferences in federalism litigation.”[94] This role is not unique to states, of course—private litigants can bring federalism-based legal challenges as well.[95] As we explain below, however, considerations of expertise, institutional capacity, and democratic accountability suggest that states may be particularly well-situated to spearhead such litigation. Indeed, states have been at the forefront of some of the most consequential challenges to federal policy in recent years, including not only the constitutional challenge to the ACA but also more recent challenges to the Trump Administration’s travel bans.

Those examples are merely the tip of the state-litigation iceberg, but they capture a feature that has drawn significant attention in popular commentary: the states’ challenges to the ACA and the travel bans have been decidedly partisan affairs. The ACA litigation was led by red states; the ongoing travel-ban litigation is dominated by blue states.[96] One might well wonder, therefore, whether in practice state litigation mitigates polarization or instead exacerbates it. The remainder of this Article is devoted to that question. We begin by surveying the landscape of state litigation, mapping the many different ways that state AGs can shape national policy, and describing some of the institutional and doctrinal changes that have caused such litigation to flourish. We then examine how the various categories of state litigation relate to polarization at both the federal and state levels.

II. The Flowering of State Public-Law Litigation

In recent decades, state AGs have emerged as a uniquely powerful cadre of lawyers. As the chief legal officers for their respective states, AGs are responsible for enforcing state law and defending the state against legal challenges; in many areas, they also share responsibility with federal agencies for enforcing federal law.[97] Independently elected in forty-three states, AGs stand at the top of organizational hierarchies that operate alongside—and sometimes in opposition to—other institutions for state policymaking.[98]

Although state public litigation goes back considerably further, state AGs’ work first grabbed the national spotlight in the 1990s, when AGs from different states banded together to take on Big Tobacco. Although AGs were by no means the first lawyers to sue the tobacco companies, they succeeded where others had failed, securing a settlement that required substantial changes in tobacco marketing and payments to the states totaling more than $206 billion.[99] In more recent years, AGs have targeted, and ultimately disrupted, settled industry practices by paint producers, toy manufacturers, pharmaceutical companies, and auto companies—among others. As one corporate lobbyist put it, “In some ways, [AGs are] more powerful than governors . . . . They don’t need a legislature to approve what they do. Their legislature is a jury. That’s what makes them frightening[.]”[100]

State litigation is not just practically significant; it is also politically salient. And as AGs have become increasingly active and entrepreneurial, they have also attracted criticism from various quarters—including from other AGs.[101] Critics claim that state litigation is driven by partisan ambitions rather than a desire to vindicate the interests of the states qua states. We take up those critiques in Part III. Our goal here is to provide a positive account of what state public-law litigation is, and what makes it possible.

Before proceeding, a few words on terminology and scope: we use the term “state public-law litigation” because we want to address a particular subset of litigation by state AGs. We do not focus on government-contracts litigation involving the state, ordinary civil enforcement of state regulatory laws, or most individual criminal prosecutions. Rather, our subject is more like the category of impact litigation undertaken by public-interest lawyers. Just as public-rights cases brought by nongovernmental organizations seeking broad reforms became a critical category of litigation in the late twentieth century, requiring courts and scholars to rethink a litigation model predicated on the enforcement of private rights,[102] so too litigation by state governments has increasingly taken on a public-law cast.

That said, the category remains fuzzy. Although one can easily identify examples of state public-law litigation, such as the state lawsuits challenging the ACA or the Trump travel bans, delimiting principles are harder to come by.[103] Because our interest is in the practical impact of state litigation on American politics and the federal system, we want to define the relevant category fairly loosely. What we have in mind is (1) litigation activity (not only filing lawsuits but also defending them and participating as amici) (2) by states[104] that is (3) intended to have a legal and/or political impact that transcends the individual case and the jurisdiction where the action takes place.

A. The Engines of Expanding State Litigation

Prior to the 1980s, most state AG offices could be described as “[p]lacid and reactive.”[105] Things changed dramatically over the next few decades. The “New Federalism” of the Reagan Administration devolved countless regulatory and administrative responsibilities from the federal government to the states.[106] As the workload of state agencies increased, so too did their litigation exposure—with the burden of defense falling on state AGs.

Recognizing their AGs’ significant new responsibilities, states allocated more resources to them.[107] Higher budgets and greater responsibilities, in turn, drew a new breed of attorney to the AG’s office. Increasingly, the “state’s law firm” was staffed with “a younger, better educated, and more ambitious caliber of attorney.”[108]

As institutional capacity expanded, so too did the opportunities to use it. When federal agencies decreased their enforcement activities in the 1980s, state-level enforcers rushed in to fill the void.[109] Areas like antitrust and consumer protection, once dominated by the federal government, became enclaves of aggressive state enforcement.[110] Many AGs established specialized units and task forces to handle their new responsibilities, thereby “enhanc[ing] the role of the attorney general as a ‘public interest lawyer’ and offer[ing] many opportunities to improve the quality of life for citizens of the states and jurisdictions.”[111]

Meanwhile, new provisions of federal law facilitated state litigation by authorizing state AGs to enforce federal statutes, often by suing as parens patriae to protect the rights of state citizens.[112] The common law doctrine of parens patriae dates back to early English practice, in which the King exercised certain royal prerogatives as “parent of the country.”[113] In its more modern form, the doctrine allows states to vindicate sovereign or quasi-sovereign interests, including an “interest in the health and well-being . . . of [their] residents in general.”[114] Today, many state and federal statutes explicitly authorize states to sue as parens patriae.[115] Others can be read to authorize state suits implicitly by creating broad rights of action for citizens whom the states represent.[116] And even absent specific statutory authorization, state AGs may (depending on state law) have common law or constitutional authority to litigate as parens patriae on behalf of citizens.[117]

The 1990s tobacco litigation built on, and spurred, expansions in AG authority. Prior to the states’ assault on Big Tobacco, countless private plaintiffs had sued under a variety of tort and warranty theories—all seeking to hold the industry accountable for peddling an unreasonably dangerous product. None succeeded.[118] Many plaintiffs were simply outspent by the defendants; others were turned away on the ground that they had assumed the risk of smoking; and still others were thwarted by courts’ refusal to permit large numbers of smokers to sue together as class actions.[119]

Then came the states, which were able to avoid the pitfalls of earlier litigation and bring the tobacco companies to the bargaining table. Most states pursued restitution actions, seeking reimbursement for Medicaid expenses incurred in the treatment of smoking-related illnesses.[120] By shifting the focus from individual smokers to the states’ own losses, the state suits were able to cut off the tobacco companies’ prime defense strategy: blaming individual smokers. As Mississippi AG Mike Moore put it, “This time, the industry cannot claim that a smoker knew full well what risks he took each time he lit up. The state of Mississippi never smoked a cigarette. Yet it has paid the medical expenses of thousands of indigent smokers who did.”[121] Similarly, the states’ strategy allowed them to avoid the challenges of class certification: “[I]nstead of millions of plaintiffs, there would only be one. Concerns over common issues of fact, which doomed earlier class actions to fail the predominance and superiority tests of federal and state class action statutes, would be finessed.”[122] Ultimately, forty-six states joined the Master Settlement Agreement, which required the tobacco companies to pay the states more than $200 billion over twenty-five years and to agree to an array of regulatory constraints.[123]

Although the tobacco litigation is in some ways sui generis, it highlights several features that have helped fuel state litigation more broadly. First, the tobacco suits entailed an “unprecedented” degree of interstate cooperation among AGs, and their success made clear—to AGs as well as to potential defendants—the power of concerted multistate action.[124] Second, the litigation demonstrated the value of cooperation between AGs and private attorneys. The states’ suits benefited from substantial assistance and financing from private lawyers—a pattern that has been repeated in many subsequent actions. By teaming up with private counsel (particularly those willing to work for a contingent fee), state AGs can expand their reach into litigation that would otherwise be prohibitively expensive or resource-intensive, or would require specialized expertise.[125] Third, the staggering size of the settlement—“the largest transfer of wealth as a result of litigation in the history of the human race”[126]—revealed just how lucrative state litigation could be. In the years since the tobacco litigation, state AGs have become adept at using large monetary recoveries to publicize the financial contributions they make to the state and its citizens.[127] In many states, moreover, AG offices can retain certain types of financial recoveries, making litigation a self-sustaining endeavor.[128]

Finally, the states’ legal theories in the tobacco cases created a template for future actions against industries that cause widespread harm to state citizens.[129] The recoupment strategy alone is a powerful tool for recovering the states’ own expenses[130] and becomes more powerful still when combined with the states’ authority to sue as parens patriae to address harms to their citizens.[131] In the ongoing state efforts against opioid manufacturers, for example, the states have asserted various common law tort claims and are seeking recovery for harms to citizens and to their own proprietary interests, including “billions of dollars in damages to the State related to the excessive costs of healthcare, criminal justice, education, social services, lost productivity; and other economic losses as a direct result of the illicit use of these dangerous drugs caused by opioid diversion.”[132]

Courts—state and federal—have also played a role in the growth of state AG litigation. Perhaps most importantly, they have taken an expansive view of state standing. In Massachusetts v. EPA, the Supreme Court cited Massachusetts’s “stake in protecting its quasi-sovereign interests” as a reason for “special solicitude” in the standing analysis.[133] Long before those words were penned, lower federal courts had held that states can sue as parens patriae to vindicate their citizens’ rights under the federal constitution, even in circumstances in which the citizens themselves would lack standing. For instance, whereas the rule of City of Los Angeles v. Lyons makes it difficult for private parties to seek injunctive relief from sporadic instances of official misconduct,[134] courts have permitted states to sue in equivalent cases.[135] Similarly, as noted above, courts recognized states’ standing to sue the tobacco companies to recoup the expenses they had incurred as a result of smoking-related illnesses suffered by their citizens. When unions and other private organizations asserted similar claims, however, courts ruled that their injuries were too remote to establish standing.[136]

Representative suits by states also enjoy a host of other procedural advantages over their closest private analogues, class actions. Whereas class actions are governed by a complex set of procedural requirements designed to promote judicial economy and protect the interests of absent class members, courts have declined to apply those rules to similar suits by states—even as they have tightened up the requirements for private suits.[137] Courts have likewise refused to subject parens patriae suits to the jurisdictional requirements of the Class Action Fairness Act[138] or to mandatory arbitration clauses.[139] And when faced with simultaneous suits by states and by private class counsel, courts have often denied certification to the private class action on the ground that the state suit is the “superior” method of adjudication.[140] As one court put it, “[T]he State should be the preferred representative” of its citizens.[141]

It is not surprising, then, that state litigation activity has increased markedly in both volume and visibility in recent decades. For example, the number of Supreme Court cases in which states are parties has shot up since the 1980s—spurred in part by the creation in 1982 of the National Association of Attorneys General (NAAG) Supreme Court Project.[142] Even more notable is the increase in states’ filings as amici. Such filings are not command performances but represent AGs’ discretionary decisions to devote limited resources to Supreme Court advocacy.[143] The most comprehensive study of state litigation in the Supreme Court reports that since 1989 states have “become exceptionally active amicus curiae participants. They account for 20% of all certiorari petitions accompanied by an amicus brief and 18% of the amicus briefs on the merits.”[144] Today, states’ participation in the Supreme Court—both as direct parties and as amici—is second only to that of the federal government.[145]

The Supreme Court may be the most prominent venue for state litigation, but it is hardly the only one. States also have become more frequent litigants in the state and lower federal courts. Texas’s Greg Abbott sued the Obama Administration “at least 44 times”;[146] AG Maura Healy of Massachusetts reportedly “led or joined dozens of lawsuits and legal briefs” challenging the Trump Administration in 2017 alone.[147]

And states are now far more likely to band together in litigation in order to maximize their impact. For example, Paul Nolette found a marked increase in “coordinated AG litigation”—defined as filed lawsuits as well as preliminary investigations involving coordinated activity by at least two AGs—from 1980 to 2013. Professor Nolette reports: “From a consistently low number of one to four cases a year throughout the 1980s, the quantity of multistate cases . . . gradually increased, reaching twenty for the first time in 1996, thirty in 2002, and forty in 2008.”[148] The number of AGs participating in such cases also has grown, with a greater proportion of multistate cases involving sixteen or more states in recent years.[149] As Nolette explains, “Litigation involving over half of the nation’s AGs, once an unusual event, represents over 40% of all the multistate cases conducted since 2000.”[150] For many observers, AG activism amounts to “a major shift in how political fights are waged.”[151]

B. Mapping State Litigation

We know states are doing more litigation, but the aggregate numbers can only tell us so much. Although discussion of high-profile state litigation sometimes treats it as a unitary category, that perspective obscures important variation within the genre. This section maps state litigation into several discrete types, based on the nature of the claims asserted. We begin with the kinds of cases observers typically associate with state public-law litigation—cases in which states are pitted against the federal government. These include (1) claims that federal government action exceeds the limits of national regulatory authority, as in the state challenges to the ACA; (2) claims that federal government action violates aspects of the national separation of powers, as in state challenges to President Obama’s immigration policies; and (3) claims that federal government action violates individual federal rights, as in the state lawsuits against President Trump’s travel bans. It bears emphasis, however, that states can also shape policy outside their borders by targeting primary behavior directly, in suits against private actors alleging violations of either (4) state or (5) federal law.

To be sure, many prominent lawsuits will fall within more than one of these categories. For example, challenges to President Trump’s travel bans have sometimes included both claims that the bans violate individual rights and claims that the President has exceeded the scope of his lawful executive authority.[152] And state amicus briefs concerning the validity of the federal Defense of Marriage Act (DOMA) raised both federalism and individual rights arguments.[153]

Each category also includes legal claims and arguments asserted by states in a variety of settings—including, for example, not only lawsuits but also amicus filings by state AGs. We define our categories by the legal claim asserted, not the form in which that claim is advanced. And, while we have framed our categories as challenges to the legality of either federal governmental or private action, we also include states’ assertion of arguments—often in opposition to other states—affirming the legality of those actions.[154]

1. Federal Power Claims.—This category contains claims that federal action exceeds the legal limits of national authority. The paradigmatic claims are those about the reach of Congress’s enumerated powers.[155] For example, minutes after President Obama signed the ACA, thirteen states filed suit arguing that Congress lacked power under the Commerce Clause to require individuals to buy health insurance.[156] Sometimes states raise these sorts of claims as a preemptive strike on federal legislation, as in the ACA case. Perhaps more often, these issues are raised by private parties as defenses to the imposition of federal requirements or penalties,[157] or in suits for a declaratory judgment or an injunction seeking to bar enforcement of federal law.[158] States then come in as amici—sometimes on both sides of the case.[159]

These cases are high visibility but, we want to suggest, of limited practical importance. They’re just not very promising, given the Court’s capacious understanding of national enumerated powers.[160] The Commerce Clause is very, very broad—and even where it’s not broad enough, there is the Necessary and Proper Clause to fill most gaps.[161] (In the healthcare case, the Taxing Clause saved the day for the ACA.)[162] We may see occasional wins for states here, but they’re likely—as in Lopez—to be mostly symbolic in their importance.[163]

The more significant cases are those in which Congress seeks to enlist state officials to implement federal law but arguably lacks power to do so. Most federal programs rely on state and local officials for enforcement and implementation. Polarization makes states governed by the party that is out of power in Washington particularly likely to want to opt out of such programs. Under the Anti-Commandeering Doctrine, Congress can’t require state officials to implement federal policy.[164] Instead, Congress typically conditions federal benefits (usually money) on state cooperation.[165] Challengers therefore argue that federal spending conditions are insufficiently clear or amount to federal coercion, as in the Medicaid Expansion portion of the healthcare case[166] or in the current challenges to the Trump order on sanctuary cities.[167] Alternatively, states’ claims may focus on whether certain federal requirements really amount to commandeering.[168]

This latter class of cases operates within a cooperative federalism context rather than a model of federalism where states have their own exclusive sphere of regulatory jurisdiction outside of federal authority.[169] But rather than seeking to control the content of federal policy, these cases generally try to preserve states’ ability to opt out. The Printz litigation that established the anti-commandeering principle for state executive officers did not try to strike down the federal Brady Act; it simply protected the right of state and local officials not to participate in its enforcement.[170] Likewise, the Medicaid expansion decision established an opt-out right for states.[171]

Finally, an important class of federal-power claims involves state immunities from federal regulation. These claims arise defensively, typically in response to claims by private litigants.[172] For a brief period during the late 1970s and early 1980s, state and local governments asserted immunities from federal regulation itself under the now-defunct National League of Cities doctrine.[173] More enduring principles shield state governments from certain judicial remedies when they violate federal requirements. A line of cases stretching back over a century—but intensifying under the Rehnquist Court—recognized a broad principle of state sovereign immunity shielding states from damages claims brought by individuals for violations of federal law.[174] More recent cases have constricted federal civil rights claims against state and local officers for violations of federal statutory requirements.[175] States have participated in these cases as both party defendants and extensively as amici (again, often on both sides).[176]

These immunity cases differ from most of our examples of state public-law litigation in that they arise defensively—they are not, as it were, examples of AGs like Texas’s Greg Abbott going into work and suing the federal government. Nonetheless, they do seem part of a systematic effort to expand protections for state and local governments under federal law. It seems fair to view Jeffrey Sutton’s successful advocacy of an expansive view of state sovereign immunity in cases like University of Alabama v. Garrett[177] and Kimel v. Florida Board of Regents,[178] for example, as an extension of his entrepreneurial tenure as State Solicitor of Ohio.[179]

2. Federal Separation of Powers Claims.—It’s less intuitive to think of States making separation of powers arguments, but one can find examples reaching way back: in 1970, for example, Massachusetts filed an unsuccessful original action in the Supreme Court challenging the constitutionality of the Vietnam War.[180] Separation of powers claims have become far more prevalent over the past decade or so. As we’ve noted, polarization tends to cause gridlock, even with a nominally unified government in Washington. And gridlock encourages the President to reach for his pen and phone to get things done.[181] Resulting challenges sound in separation of powers, not federalism. But the litigation is motivated by states that are either seeking to protect their own autonomy or to find a way to participate in a national lawmaking process that has shifted from Congress to the Executive Branch.

United States v. Texas—the immigration case—is a good example.[182] When President Obama extended lawful presence to millions of additional undocumented aliens, it was hard to argue that the deferred-action programs (Deferred Action for Parents of Americans (DAPA) and Deferred Action for Childhood Arrivals (DACA)) fell outside the authority of the national government as a whole. Instead, state challengers contended that the President lacked the authority to—as Obama himself put it—“change the law” without going to Congress.[183] As was clear to all involved, Congress’s general intransigence on the immigration issue meant that a decision against executive authority would be—for all intents and purposes—a decision against federal authority more generally.

A separate set of process arguments are statutory but serve a constitutional purpose. Again, the immigration case is a good example. Texas’s successful argument in the district court was simply that Obama’s policy change had failed to comply with the Administrative Procedure Act (APA) because it had not gone through notice and comment. Notice and comment isn’t an insurmountable hurdle for agency lawmaking, but it does delay implementation of national policy. More importantly, it allows states—like anybody else—to insist on direct input into the federal lawmaking process. It allows states to be heard at the agency just as they are supposedly heard in Congress, although without any special status vis-à-vis other participants. Provisions in the APA for notice and comment, as well as for judicial review of process failures at the agency, effectively operate as separation of powers-type constraints on the administrative state.[184]

The separation of powers principle that Congress—not the President—makes the law also generates a second kind of challenge to federal action. That challenge argues that executive action—like the immigration order or the travel ban or the EPA’s clean power plan—is substantively inconsistent with the underlying statute.[185] Polarization can cause such claims to multiply. The longer gridlock persists, the more likely that new executive initiatives will stray from the obvious purview of the original legislation. So, for example, states challenged the Obama Administration’s transgender bathroom guidance on the ground that its definition of gender discrimination differs from that of the Congress that enacted Title IX of the Civil Rights Act.[186] Likewise, when federal agencies promulgated broad “preemption preambles” during the George W. Bush Administration, a coalition of states, as well as a state governmental association, filed amicus briefs arguing that these preambles exceeded the agencies’ statutory mandate.[187]

A more difficult class of cases involves litigation challenging federal government inaction. Federal administrative law generally presumes that agency inaction—at least in the form of agency refusals to initiate enforcement proceedings—are not subject to judicial review.[188] But this presumption can sometimes be overcome, as it was by Massachusetts v. EPA’s holding that states could challenge the agency’s denial of rulemaking petitions authorized by statute.[189] Given Congress’s continued failure to act on climate change, “EPA regulation pursuant to [Massachusetts v. EPA] . . . has served as the core of the US federal efforts on climate change.”[190] And where an incoming administration seeks to overturn previous executive action—thus arguably returning to the status quo ante of inaction—states may find greater leverage to challenge this departure from the prior baseline. Recent litigation over the Trump Administration’s “repeal” of President Obama’s DACA policy, for example, has gotten significant traction by arguing that the repeal rested on improper reasons.[191] State litigation to enforce the Executive’s statutory obligations can thus force adoption and continuation of executive policies even where national-level gridlock would otherwise foreclose them.

3. Federal Rights Cases.—Some state challenges to federal action rely not just on structural principles but also on individual rights arguments. In the travel ban cases, for instance, state governments assert parens patriae standing to raise the rights of their citizens. Sometimes states assert proprietary interests as well; some of the state plaintiffs in the travel ban cases argued that their state universities had been deprived of faculty and students from abroad.[192] And sometimes the states participate as amici to express a view on the scope of federal individual rights, as in the same-sex marriage cases.[193]

This category also includes state litigation activity contesting federal rights. For example, numerous states have participated as amici opposing Equal Protection challenges to affirmative action in state universities.[194] It is even more common to see states opposing rights claims by criminal defendants.[195] Similarly, states often play defense against federal civil rights claims brought by private litigants. (These two categories are often related, as many federal civil rights claims involve allegations of improper actions by state or local law enforcement.) In this latter set of cases, state governments are often the defendants; even where they are not (in the many cases against municipalities and their officers, for instance), they may well play a prominent role as amici.[196] And in all such cases, other states may support the party asserting federal rights as amici. When he was AG of Minnesota in the early 1960s, for example, Vice President Walter Mondale filed a brief on behalf of twenty-two states urging the Supreme Court to expand the right to counsel in Gideon v. Wainwright.[197]

As we discuss in more detail in the following Part, these rights cases create the potential for conflicts among states. Whenever state AGs support claims of constitutional rights, they are—in a very real sense—arguing against their own state’s power. More than that, they are seeking to impose a particular rule on all states. Like the statutory challenges described above, then, individual rights cases often involve interstate conflicts over control of federal policy. Those conflicts, moreover, can often be coded as red versus blue. And because they frequently involve “hot button” issues, these cases raise particular risks of politicizing the AG’s office.

4. State Enforcement of State Law that Creates National Regulation.—As we have already noted, the tobacco litigation of the 1990s was a critical watershed for state public-law litigation. To be sure, states have sought to enforce their own laws in ways that affect conditions outside their jurisdictions for a very long time.[198] And local governments have also been active in this sort of litigation—for example, in suits against the firearms industry during the 1990s.[199] But the most successful efforts have been undertaken by states. Most observers seem to agree that the tobacco litigation ushered in a new era of state activism that then spread to other regulatory areas and types of litigation.[200]

The tobacco litigation and its contemporary analogs share two related features that differentiate them from ordinary state enforcement of state law against private parties. The first is that rather than a single state suing a defendant within its jurisdiction for torts that harmed its citizens, the tobacco litigation featured a broad coalition of states—ultimately including all of them.[201] And the Master Settlement Agreement that ended the litigation eventually came to include nearly all manufacturers of tobacco in the American market. The litigation thus aimed at global peace—that is, a comprehensive settlement among all the relevant players.

The second point is that the tobacco settlement essentially created a nationwide regulatory regime governing cigarettes.[202] It includes, for example, not only payments by the defendants for past harms but also agreements to strengthen warning labels and restrictions on advertising. Because it applies throughout the United States and governs the activities of virtually all tobacco companies doing business here, one could fairly say that it might as well be a federal law.

Similar multistate litigation efforts have imposed quasi-regulatory regimes via comprehensive settlements with major industry players in the pharmaceutical and other industries.[203] We expect this phenomenon will continue. In the fall of 2017, for example, the Commonwealth of Massachusetts sued the credit-reporting company Equifax following announcement of a data breach that allegedly affected over 140 million consumers.[204] Massachusetts brought the suit under its own data privacy statute, as well as a more general consumer protection statute. If other states and credit reporting firms are drawn into this litigation, one might well see another comprehensive settlement with terms that would effectively act as, and possibly obviate, national regulation.

5. State Enforcement of Federal Law.—State AGs also can, and do, enforce many aspects of federal law. State enforcement of federal law is pervasive, from antitrust to consumer protection to environmental law.[205] As we explained above, this can happen either through explicit statutory authorization or through states relying on more general private rights of action, often asserting parens patriae standing to sue on behalf of their citizens.[206]

On its face, this category of cases may not seem particularly empowering for states, given that AGs are merely enforcing policies that already have been written into federal statutes and regulations. Yet the level of enforcement can have profound consequences for what the law means in practice, and for how regulated entities view their options. That is true even when the law’s substantive requirements are perfectly clear: higher levels of enforcement are likely to increase deterrence by raising the expected sanction for violations.[207] And when the relevant statutory or regulatory commands are somewhat less than pellucid—as is often the case—state AGs can shape policy on a national scale by pushing particular interpretations of vague or ambiguous federal laws.[208]

Thus, the most interesting instances for our purposes are those where state enforcement reflects a disagreement with national enforcement policy. The most salient recent example was Arizona’s effort to ramp up enforcement of federal immigration laws in response to what it saw as an abdication by federal authorities.[209] Another example, with a different political valence, would be Eliot Spitzer’s effort in New York to enforce federal environmental laws more aggressively than the federal EPA had previously been willing to do.[210]

Like the multistate cases described above, state enforcement of federal law can create the equivalent of regulatory policy nationwide. Given the interconnectedness of the national market, it’s hard to confine the effects of state enforcement within a particular state’s borders. If New York aggressively pursues Microsoft, Washington may feel aggrieved. And if pro-environment states undermine the fortunes of big oil companies, the oil-producing states may share in the consequences.

III. State Litigation, Politics, and Polarization

As state AGs have gained prominence, they have also attracted critics. A prominent theme in the critiques is that state litigation has moved away from its traditional core of defending “state interests” and into an uncertain new realm dominated by politics, partisanship, and policy debates.[211] Indeed, such critiques sparked the creation of a dissident AG organization, the Republican Attorneys General Association (RAGA), in 1999 as a way to “stop what they called ‘government lawsuit abuse’ and redirect state legal efforts away from national tort cases and back to traditional crime fighting.”[212] The creation of RAGA didn’t do much to stem state litigation, but it did help balance the political membership of AGs’ offices. AGs used to be overwhelmingly Democratic; there is now a much closer mix of Democrats and Republicans—due in part to aggressive campaign contributions and ads by the Chamber of Commerce and similar groups.[213] Many of those newly elected Republican AGs have themselves become active litigants, particularly during the Obama Administration.

The consequence is that it’s easy to paint state litigation as a partisan affair, with blue-state AGs challenging national policies or business practices that are defended by their red-state counterparts—or vice versa. Viewed from that perspective, the work of AGs seems destined to exacerbate, rather than ameliorate, the trends toward polarization that define our national politics.

We think the picture is considerably more complicated, as this Part explains. We begin by surveying what we know about partisanship and polarization among state AGs themselves, and then address the question that animates this Article: to the extent that state litigation is “political,” what should we make of that fact?

A. Polarization, State AGs, and State Litigation

When RAGA was founded in 1999, there were only twelve Republican AGs.[214] Today there are twenty-seven.[215] In the intervening years, AG elections have not only gotten more competitive,[216] they have also become more high-profile and more expensive. Drawing on data from the Database on Ideology, Money in Politics, and Elections (DIME), Figures 1 and 2 show the median and mean total campaign contributions reported by AG candidates in races from 1990 to 2012.[217] As the difference between the medians and means suggests, there are outliers in both directions—but particularly at the high end. Not all AG elections are expensive today, but some are very expensive. In 2012, for example, seven AG candidates reported fundraising totaling more than $1 million; Greg Abbott topped that list at $13.9 million.

Source: Adam Bonica, Database on Ideology, Money in Politics, and Elections: Public Version 2.0 (2016). Stanford, CA: Stanford University Libraries.

Note: In Figure 1, the top line is referencing “Median Dem” and the middle line is referencing “Median.”

Source: Adam Bonica, Database on Ideology, Money in Politics, and Elections: Public Version 2.0 (2016). Stanford, CA: Stanford University Libraries.

Note: In Figure 2, the top line is referencing “Mean Dem” and the middle line is referencing “Mean.”

These numbers must be taken with a grain of salt, particularly prior to 2000, when the data were spotty. But they are consistent with reports that more money is flowing into AG races, much of it from out of state.[218] And there is good reason to believe that the numbers have gone up (perhaps sharply) since 2012. RAGA, for example, raised $16 million in 2014—up from $470,000 in 2002.[219] Both RAGA and DAGA reported raising record sums during the first half of 2017 (up 45% and 73%, respectively, from the same point in the prior election cycle).[220] Both groups are also deploying their money more aggressively, after announcing in 2017 that they would end their longstanding “handshake agreement that they wouldn’t target seats held by incumbents of the other party.”[221] The effects were immediate: in one 2017 race alone, RAGA and DAGA collectively spent about $10 million.[222]

If AG races are more contentious than they once were, and if partisan associations like RAGA and DAGA are playing a more significant role in those elections, what are the consequences for AGs themselves? Do AGs reflect the same kind of partisan sorting and ideological divergence that characterize polarization at the federal level? Measuring polarization in AGs is no easy task, given the absence of conventional measurement tools—such as roll-call votes, which are the dominant tool for measuring ideology (and, thus, polarization) in Congress. But the available evidence suggests that the more general trends toward political polarization have not passed AGs by. To the best of our knowledge, the only current measure of AG ideology is from the DIME project, from which we drew the data on campaign contributions above. DIME is the brainchild of Stanford political scientist Adam Bonica, and it is more than a repository of information on campaign finance. Professor Bonica uses the contribution data to estimate the ideology of candidates based on the contributions they receive—“[t]he pattern of who gives to whom.”[223] Because many donors give to candidates at all levels of government, the ideology measures—known as CFscores—can compare the ideology of politicians in different types of offices (e.g., legislators vs. governors) as well as comparing different inhabitants of the same office (e.g., AGs from different states or AGs from the same state in different years).[224]

The limited information on AG races prior to 2000 makes it difficult to draw any meaningful conclusions about trends over time, but the data do suggest that partisan sorting is no less pronounced among AGs than among other elected officials. Figure 3 shows the CFscores for AGs elected in 2000–2012: positive values are more conservative, and negative values are more liberal. As is true in Congress today, there is no overlap between the most conservative Democrats and the most liberal Republicans.[225] Professor Bonica’s own analysis of the data paints a similar picture for ideological divergence. Focusing on state-level ideology during the 2009–2010 election cycle, Bonica found that AGs in thirty-five states were more ideologically extreme than “the mean state legislator from their respective party” and that the distance between the mean Democrat and Republican AGs was similar to the ideological divergence in Congress at the time.[226]

Source: Adam Bonica, Database on Ideology, Money in Politics, and Elections: Public Version 2.0 (2016). Stanford, CA: Stanford University Libraries.

Note: In Figure 3, the second line from the top is referencing “Median Rep” and the fifth line from the top is referencing “Median Dem.”

The fact that AGs from different parties are divided is not terribly surprising, though the suggestion that they are more ideologically extreme than most state legislators may be. The operative question for our purposes, however, is whether trends in political polarization are being reflected in state litigation. It’s easy to see why the answer might be yes. Some observers predict, for example, that the changes in AG elections will sharpen partisan divides and reduce bipartisan cooperation: “It’s hard to work cooperatively with your fellow AGs if you’re always wondering what they’re going to use to try to target you in the next election.”[227] Or, to put it more bluntly: “As each cycle goes by, the presumption is going to be that the AG across the table is going to destroy you if he or she can.”[228]

Similarly, the trend toward unified government in the states is likely to produce more polarization, and less bipartisanship, in state litigation. Until relatively recently, it was not uncommon to find Democratic AGs in otherwise red states.[229] And, because most states had divided government, most AGs had to contend with an opposite-party legislature or governor. It stands to reason—and there is some evidence to support this notion, discussed below—that AGs who hail from a different party than other state leaders will tend to take a more moderate approach to litigation than those who work in states with more one-sided politics. But those “purple” seats are becoming less common, as more states turn to unified government and more AG races follow suit. Of the thirty-one states that had unified government in 2017, at least twenty-seven had same-party AGs.[230]

Here too, it is easier to hypothesize about polarization than to measure it, but what we know about state litigation suggests that partisanship is playing a more dominant role. For example, research on state amicus briefing indicates that AGs from different states increasingly articulate opposing interests. Writing in 1987, Thomas Morris reported that states appeared on opposite sides of only 2% of the cases argued before the Supreme Court.[231] Such findings reinforced the view that the political developments of the 1980s and early 1990s “helped forge a new sense of shared interest between the states . . . . [N]ot only have state attorneys general become more active, they have increasingly sought to influence policy qua states in the collective sense rather than as individual state actors.”[232]

That sense of shared interest may have eroded in recent years. A 2014 study by Professor Nolette found significantly more interstate conflict, particularly during the Obama Administration. Focusing on cases decided by the Supreme Court between 1993 and 2013, Nolette examined instances in which multiple AGs filed briefs, either as amici or parties, at the cert or merits stage. He found a “large spike” in interstate conflicts during the last four years of the sample.[233] In 35% of the cases during that period, states either squared off against each other or collaborated on briefs with a strong partisan slant.[234]

In other work, Professor Nolette also documented partisan patterns in multistate litigation in the lower federal courts. Whereas state suits against corporations have been largely bipartisan affairs, Nolette found “wide partisan splits among AGs” in what he calls “policy-forcing” suits—cases in which states have “attempted to force [federal agencies] to take a more active regulatory approach.”[235] He found partisanship to be playing a dominant role in “policy-blocking” litigation as well—a category of litigation that he defines as “state legal challenges to regulatory actions by federal policymakers”[236]—though the roles were reversed. Whereas Democratic AGs had taken the lead in “policy-forcing” litigation since the George W. Bush Administration, Republican AGs were at the forefront of “policy-blocking” litigation under President Obama.

Studies like Nolette’s are illuminating, but they raise important questions about how to measure partisanship and polarization in the litigation context. One might try to code the positions advanced by AG briefs as liberal or conservative and then determine the partisan affiliation of the AGs who sign each brief. The difficulty, of course, is devising a system for coding substantive positions that is both valid and reliable.[237] Instead, most researchers have focused on the identity of the AGs who participate in the relevant case or brief. Nolette identifies partisanship by a head count of participating AGs.[238] That approach avoids the difficulties of categorization that bedevil attempts to code positions by ideology, but it has its own problems: it is insensitive to the ratio of Democratic and Republican AGs in office, and (relatedly) focuses on the AGs who participate in a given case rather than the AGs who opt to sit it out. The upshot is that a brief signed by twenty Democrats and five Republicans registers the same way regardless of whether there are twenty Democratic AGs in office or forty-five.

A different approach is to code polarization based on the number of (say) Republican AGs participating in a case compared to the number of Republicans then in office, as a means of calculating whether the coalition of AGs was more Republican than would be expected by chance. A recent study by Margaret Lemos and Kevin Quinn took that approach, focusing on the coalitions of AGs who joined or opposed each other in amicus briefs filed in the Supreme Court between 1980 and 2013.[239] If state amicus activity were partisan, one would expect cosigners to be from the same party and opposing briefs to be filed by AGs from different parties. Professors Lemos and Quinn found some partisan clustering (meaning that the group of AGs joining or opposing a brief was significantly more or less Republican than would be expected from a random draw of AGs then in office), but only in recent years, and—for the most part—only in cases in which groups of AGs weighed in on both sides.[240] When AGs appeared as amici on only one side of a case, they tended to do so in bipartisan coalitions.[241] (There were a number of years, however, in which there were significantly polarized Republican coalitions—mostly in criminal procedure cases in which Republican AGs joined an amicus brief and Democratic AGs did not participate at all.)[242]

Partisan patterns do not, of course, prove that partisanship is causing AGs to act.[243] Virtually no researchers have sought to tease out different drivers for state litigation. The leading exception is Colin Provost, whose studies of state consumer-protection litigation have controlled for factors such as the magnitude of harm caused to state citizens by the defendant’s conduct, the presence of consumer groups in the state, citizen ideology, median income, and more.[244] His findings are too complicated to summarize briefly here, but they underscore the need for caution before drawing conclusions about the motivations for state litigation. Provost found, for example, that AGs’ own party affiliation did not have a significant effect on the probability of their joining a consumer-protection lawsuit, but that the number of consumer groups in the state did—as did the ideology of state citizens, but only in cases involving Fortune 500 companies.[245]

Taken together, the existing studies suggest two important points for our purposes. First, context matters: the extent to which state litigation reflects polarization among AGs depends on the kind of litigation at issue. For example, state litigation against business interests tends to be more bipartisan than state litigation against the federal government.

Second, AGs’ own partisanship may interact with other considerations in ways that are difficult—if not impossible—to tease out from the data alone. For example, Professor Nolette’s finding that state litigation against corporations tends to be bipartisan might reflect the fact that some suits are more “political” than others. But (as Nolette acknowledges) the pattern also may be explained by more prosaic concerns: when a major company is already settling with a large group of states, and when the main consequence of non-participation is exclusion from the settlement proceeds, other state AGs may see little advantage to sitting it out.[246]

As Professor Provost’s study indicates, moreover, AGs’ own partisan affiliations may be less significant in some cases than the ideological commitments of the state’s citizens—or, perhaps, of other state officials.[247] It follows that we might expect to see different behavior from a Democratic AG in an otherwise heavily Republican state than from a Democratic AG in a resoundingly blue state. And, as more states become more solidly red or blue, we might expect AGs to act in an increasingly partisan manner—as some of the data suggest.

In sum, the mere fact of partisan versus bipartisan coalitions can only tell us so much about the causes and effects of state litigation, or whether AGs are “playing politics” rather than seeking to vindicate the interests of their states. In order to make those kinds of assessments, we need a better understanding of how state litigation interacts with state interests—both institutional and regulatory. We also need a better understanding of when, and why, “politics” should matter. We take up those questions next.

B. Horizontal and Vertical Litigation

In assessing the impact of state public litigation on polarized political debates, it will help to distinguish between two types of conflict in federal systems.[248] The classic conflict is a vertical struggle between the national government and the states. When the national government tries to extend the reach of its Commerce or Spending powers, or when states band together to oppose the practice of “unfunded mandates,” these disputes qualify as predominantly vertical in character.

Our federal system was originally concocted, however, to keep a lid on a different sort of conflict—that is, horizontal conflict among states (or groups of states). Powerful groups of states frequently try to impose their preferences on other states. Creating a national government limited this conflict somewhat, but it also created a potent new weapon for states to use against one another. That weapon was the national government itself, which one group of states may use as an instrument to impose its preferences on a dissenting minority group of states. Classic examples here are the fugitive slave laws, which the slaveholding states that dominated the national government before the Civil War enacted to force the abolitionist North to go along with slavery.[249]

Vertical conflict is primarily about each state’s right to go its own way on particular questions. When Alfonso Lopez successfully challenged Congress’s authority under the Commerce Clause to restrict guns in schools,[250] that didn’t affect Texas’s own right to decide whether to permit them (it doesn’t). But it did leave the decision up to Texas. And it certainly didn’t prejudice the right of other states to restrict guns in schools. Generally speaking, the same will be true of other “federal power” claims, as we defined them in the previous Part.

Horizontal conflict, on the other hand, now mostly takes the form of fights for the right to control national policy. Both Texas’s challenge to DAPA and the blue states’ efforts to protect DACA from repeal by the Trump Administration are arguable examples, given the federal government’s plenary power over immigration matters.[251] Similarly, in Massachusetts v. EPA, one group of states[252] thought that the EPA should regulate greenhouse gases as air pollutants under the Clean Air Act; another group[253] thought it should not. Both were trying to make policy for the whole country—and still are, in extensive litigation concerning President Trump’s environmental policies.[254] (And lest deregulation seem to leave the issue open to state experimentation, industry and sometimes the federal government have argued that lax federal standards often preempt more rigorous ones at the state level.)[255] Thus, these sorts of claims often involve conflict among states over the content of national policy rather than carving out space for state policy diversity.[256] The same is true of cases in which states seek to enforce federal rights—constitutional or statutory—or use state law to create what is effectively a nationwide regulatory regime.

Although one can always find exceptions and odd cases, we think we can safely say that, generally speaking, vertical conflicts are about who decides, while horizontal conflicts are about what is to be decided. If that’s right, then the state interests at stake in vertical cases are likely to be institutional ones. Those interests may cash out in either a liberal or conservative direction in any given situation, but the interests themselves—the preference for state-level autonomy rather than top-down direction from the federal government—are politically neutral.[257] Although many observers have traditionally ascribed a conservative political valence to state autonomy in general, thoughtful scholars on the Left have recognized that to be a mistake at least since the George W. Bush administration.[258] By contrast, the interests in horizontal cases—where the parties dispute what the uniform federal rule should be—are more likely to be shorter-term regulatory interests with an identifiable political valence.[259]

To the extent that these observations are true, they suggest several normative propositions—propositions that, we believe, many critiques of state litigation today imply but rarely make explicit. The first is that in vertical conflicts we ought to see more cooperation among states across partisan lines to defend the institutional interests of state governments. One terrific example is then-Alabama Solicitor General Kevin Newsom’s amicus brief on behalf of Alabama, Louisiana, and Mississippi, supporting the pot-smokers in Gonzales v. Raich.[260] Here’s how Newsom led off that brief:

The Court should make no mistake: The States . . . do not appear here to champion . . . the public policies underlying California’s so-called “compassionate [marijuana] use” law. As a matter of drug-control policy, the amici States are basically with the Federal Government on this one. . . .

From the amici States’ perspective, however, this is not a case about drug-control policy. . . . This is a case about “our federalism” . . . . Whether California and the other compassionate-use States are “courageous” – or instead profoundly misguided – is not the point. The point is that, as a sovereign member of the federal union, California is entitled to make for itself the tough policy choices that affect its citizens.[261]

If we view states as safety valves for polarized national politics—as Part I suggested—then we should celebrate briefs like this, where states put policy disagreements aside to assert their shared institutional interests in limiting national power.

A second normative proposition is that AGs should focus less of their time and resources on horizontal conflicts. When states argue in vertical cases that particular disputes should be left up to them, they are clearing space for different jurisdictions to reach different conclusions on our most divisive questions. That lowers the stakes of national politics and mitigates the effects of polarization. But when states argue in horizontal cases that national law must adopt their own political or moral vision and impose it nationwide, they are participating in polarized conflict. There may be times when the moral imperative to do that is too strong to resist. But there is a cost, because this sort of state litigation undermines our federal system’s ability to manage polarization.

We think there is a lot of truth to these propositions, but they are not the full story. For a variety of reasons, many state public-law lawsuits will not fall cleanly into one category or the other. And even in clearly vertical cases, states may have legitimate structural interests that favor national action. Finally, horizontal litigation may serve either individual rights or other structural values—principally separation of powers—that are independently worth promoting.

First, many cases have both vertical and horizontal dimensions. For instance, the ACA litigation seemed like a vertical conflict: Congress tried to impose the ACA’s requirements on the states, and the challenger states wanted out. Striking down the ACA would not, on its face, prevent individual states from adopting a similar regime or even a single payer system. But many argued that the interstate healthcare market is so interconnected that no state could feasibly impose these requirements on its own.[262] From this perspective, if we were to have an ACA-type regime expanding healthcare coverage for all, it could only be done at the national level. This effectively made the conflict a horizontal one: blue states favoring such a regime had to use the federal government to achieve it by requiring dissenting states to conform. And by arguing the national government lacked power to enact the ACA, the red states effectively sought to force the blue states to stick with the prior, less universal regime.

Likewise, cases that look horizontal may have an important vertical dimension. As we explained above, many state challenges to national policy nowadays rely on separation of powers theories. These state challenges concede that the national government has power to act but argue that it has violated constitutional or statutory principles dividing labor among the branches of the federal government.[263] These cases may seem horizontal because they don’t purport to limit national authority overall. But, particularly in a world of polarization and gridlock, they often render federal action impossible (or at least vanishingly unlikely) as a practical matter.[264]

Second and closely related, there can be legitimate disagreement even in clearly vertical cases about where the institutional interests of the states lie. International relations scholars have argued that contemporary nations exercise their sovereignty by entering into cooperative arrangements with other nations to address problems, like climate change or the international drug trade, that they cannot effectively address alone.[265] The American states are similarly interdependent, and they have interests that can only be vindicated by national cooperation. If pollution generated in Ohio is causing acid rain in New Hampshire, New Hampshire’s autonomy may actually be enhanced by cooperative arrangements that restrict pollution that New Hampshire, acting alone, would be powerless to control. That cooperative arrangement is generally called “the federal government.”[266] For this reason, states have an institutional interest in ensuring that the national government is strong enough—and has broad enough powers—to help them out with regulatory problems they can’t effectively address on their own.

The upshot is that we may see conflict among different groups of states over issues—like the scope of the Commerce Clause in Raich—that are vertical in their structure, and both groups of states may be defending their institutional interests. That will not always be true, of course; it depends on whether the relevant policy challenge could be addressed effectively by state-level regulation, or whether it demands collective action. And that, in turn, is a question on which reasonable minds will often differ.

Legitimate disagreement also exists about whether vertical claims asserting immunities against federal remedies actually foster the sort of autonomy that can mitigate national polarization. One of us has argued that the Supreme Court’s expansive state sovereign immunity jurisprudence does little for state autonomy because it simply shields states from certain federal remedies (principally money damages) rather than restricting the scope of federal regulation altogether.[267] Cases like Garrett and Kimel, for example, did not take the potentially contentious issue of disability rights off the national agenda; they simply allowed state institutions to get away with violating federal law without paying damages.[268] Yet this argument may underappreciate the extent to which removing the threat of damages awards enables the state officials administering federal regulatory regimes to reshape those regimes to conform more closely to states’ preferences. To the extent that state sovereign immunity or limits on the scope of § 1983 shield instances of “uncooperative federalism,”[269] immunity claims may play a role similar to other assertions of vertical autonomy rights. We suspect this possibility is minor[270] but cannot discount it entirely.

All of this helps refine the normative propositions we outlined above. For those who believe that AGs have a critical role to play in vindicating the states’ long-term institutional interests—and we count ourselves as members of that camp—it is not enough to ask whether AGs are litigating in bipartisan coalitions or trying to “block” federal policy rather than to “force” it. Likewise, when we observe AGs lining up on both sides of a case, or seeking to impose their own views of good policy on the rest of the nation, we cannot (without more) conclude that any of the participating AGs is putting short-term political or policy gains above state interests. There is no substitute for parsing the particular merits issues in each individual case.

As we have explained, what matters from a federalism perspective is whether the state’s litigating position could, if successful, enable that state to “go its own way.” In an interconnected economy such as ours, one state’s autonomy will sometimes be another state’s shackles, and we should not be surprised that many such policy disagreements play out along partisan lines. It does not follow, however, that state AGs are doing anything other than playing their traditional role as “representatives of their states”[271] when they push for more rather than less federal law.

That said, there have been—and will continue to be—important cases in which state AGs use litigation to lock in particular policies in ways that run counter to state autonomy. For example, although United States v. Windsor seemed to emphasize the importance of the states’ right to define marriage for themselves and condemned the federal Defense of Marriage Act’s interference with state family law,[272] the Court’s decision in Obergefell imposed a single national answer to the question of same-sex marriage.[273] That resolution left people concentrated in blue states very happy, but it imposed a piece of their social vision on an unwilling group of red states. And in arguing against the rights claims in Windsor and in favor of the claimants in Obergefell, some of the states made arguments that may be invoked to undermine state interests in future litigation concerning state autonomy.[274]

Perhaps not surprisingly, the briefing patterns in Windsor and Obergefell were decidedly partisan. Thirty-six states filed amicus briefs in Obergefell: nineteen in support of same-sex marriage rights and seventeen opposed. The AGs supporting the rights claim were all Democrats, while those opposing the claim were all Republicans.[275] The states arguing against the rights were, moreover, pretty solidly red: fourteen had a unified Republican government; two had Republican-controlled legislatures and Democratic governors; and one had a Republican-controlled legislature and an independent governor.[276] The states on the pro-rights brief were more mixed: only seven had a unified Democratic government; six had Republican governors; and in nine the Republican party controlled one or both houses of the legislature.[277] Thus, in many of those states, the Democratic AG was taking a position on marriage that other members of state government (perhaps a majority) may have opposed. It is surely no coincidence that same-sex marriage was already legal in every one of the pro-rights states, either as a result of a court ruling or (more commonly) a state statute.[278] Similarly in Windsor, the seventeen Republican AGs who signed the state brief defending DOMA—a restriction on state power—hailed from states with constitutional provisions or legislation consistent with the Act.[279]

This is horizontal conflict in action: states using federal law (statutory or constitutional) to extend their own vision of the good nationwide. Even if one thinks—as we do—that justice required the result in Obergefell, one might nevertheless acknowledge its significant federalism costs, as well as the polarizing effects of state participation in social conflict. The federalism side of the equation is complicated by the fact that one of the most important things states do is to define themselves as moral and political communities by taking positions on issues that matter to their citizens. Same-sex marriage is one of those defining questions that affirms a community’s sense of itself as progressive and inclusive on the one hand or traditionalist and religious on the other. And given that Obergefell came down to a disagreement about the definition of “marriage,” it was arguably appropriate for the institutions chiefly charged with defining that concept in our system—state governments—to weigh in.[280] The fact that the pro-rights states were making a statement against interest, to some extent, may have made their arguments that much more weighty.[281] More broadly, it may well promote state interests in the long term for states to be recognized as sources of important insights on questions like this.[282] At the very least, we acknowledge that the cost to federalism may be worth paying in important cases.

The same may be true of other horizontal claims, such as separation of powers challenges to executive unilateralism that do not foreclose federal legislation. If national policymaking is likely one way or another, there may be no immediate payoff from the perspective of federalism. Yet the longer-term effect of such litigation may be to reinforce structural limitations on federal executive authority that, on the whole, work to the benefit of the states.[283] As in the individual rights setting, moreover, there may be an independent value to state participation in fundamental questions about the structure of American government.

The question remains whether distinctly partisan litigating patterns by state AGs might deepen the social and political cleavages that mark this era of intense polarization. We have no doubt that much state litigation is motivated by partisan considerations—either the need to generate partisan support for a particular AG’s future ambitions or the desire to vindicate sincere views about the law that happen to correspond to the positions taken by one’s political party.[284] Even clearly vertical claims asserting institutional interests can be brought for partisan reasons, because a particular state (or its AG) is simply opposed to national policy on political grounds and wants to be free of it.

We think that’s fine, actually. The objection to partisan motivations for state litigation seems to be that they render that litigation opportunistic. But it is hard to say why this sort of opportunism is necessarily a bad thing. In Federalist 51, Madison says that we’re counting on the selfish interests of particular officials to create incentives to protect the institutional interests of the various parts of the government. Opportunism, in other words, is the foundation of both separation of powers and federalism.[285]

Nevertheless, there is a somewhat different reason to worry about partisan motivations—a reason that goes to the heart of our exploration of state public-law litigation and polarization. One might grant the point about Madisonian contestation and still worry about the consequences of branding contentious legal issues as “red” or “blue.” It bears emphasis that the relevant cases would, for the most part, be brought with or without the states: most litigation in which AGs participate either already involves other plaintiffs—typically private individuals or organizations[286]—or could be brought by private parties instead of states. One might therefore take the view that state participation serves only to exacerbate the ill effects of polarization, by bringing explicitly partisan warfare to the courts. That may well be a cost, but it depends on a comparative assessment, not only of different categories of state litigation—our focus thus far—but also of state litigation and its alternatives, a question we take up in the next Part.

IV. State Governments as Public-Law Litigators

Any normative assessment of state public-law litigation must contend with a comparative question: state litigation as compared to what? One obvious alternative to litigation (regardless of the parties) is regulation, and a common strain in critiques of state litigation is that it crowds out other more democratic means of resolving contested policy questions.[287] That critique might gain force to the extent that state litigation itself appears to be a partisan affair, as partisan political issues seem best resolved through political processes. But the force of the objection depends on whether more democratic—and straightforwardly political—modes of policymaking are in fact meaningful alternatives, and on how litigation by states compares to the alternative of litigation by private individuals and groups. This Part explores those comparative questions, situating state litigation within the broader phenomenon of public-law litigation generally.

A. State Lawsuits as a Subset of Public-Law Litigation

As we have seen, state litigation sometimes has the practical effect of setting policy for the nation as a whole, and it generally does so outside the normal lawmaking processes for establishing federal regulatory norms. The states’ tobacco settlement, for example, established nationwide rules for tobacco companies that bound basically the entire industry—all without the enactment of any federal statute or regulation. Critics have seized on this feature of state litigation, arguing that AGs are taking contested, and often deeply partisan, issues off the democratic table and throwing them instead to the courts.

Such criticisms find longstanding analogues in critiques of public-law litigation, even when it is undertaken by private parties and nongovernmental organizations. At least since the 1960s, public-law litigation has been a central part of the American legal landscape. Some of this litigation has been constitutional, such as the NAACP’s campaign against Jim Crow, and some has been statutory, such as litigation by the Sierra Club, the National Resources Defense Council, and other groups to enforce environmental standards. But there is no doubt that American public law counts on nongovernmental actors to develop and enforce critical constitutional and statutory norms. Elaboration and implementation of legal norms through adversarial litigation is, in Robert Kagan’s memorable phrase, “the American way of law.”[288]

In his seminal article four decades ago, Abram Chayes observed that “the dominating characteristic of modern federal litigation is that lawsuits do not arise out of disputes between private parties about private rights. Instead, the object of litigation is the vindication of constitutional or statutory policies.”[289] Similarly, the influential Hart and Wechsler casebook documents a shift from a “dispute resolution” model of judicial power (in which judicial articulation and implementation of norms arise out of deciding concrete disputes between private parties) to a “public rights” or “law declaration” model (in which litigation is a vehicle for articulating public norms of broad applicability beyond the parties to the case).[290] Resistance to the public rights model of litigation has often centered on concerns about courts’ role in governance.[291] As Professor Chayes acknowledged, public-law litigation tends to produce relief that is not “confined in its impact to the immediate parties; instead, it is forward looking, fashioned ad hoc on flexible and broadly remedial lines, often having important consequences for many persons including absentees.”[292] Just as state AGs’ settlements over tobacco and prescription drug pricing effectively produced national regulatory regimes,[293] public-law litigation brought by private parties and NGOs has often involved not only judicial lawmaking but also the establishment of ongoing remedial regimes affecting large swaths of society. And it has been criticized accordingly: like state litigation, public-law litigation is often charged with blurring the line between litigation and legislation and with establishing ongoing regulatory regimes outside the normal lawmaking process.[294]

A related set of criticisms focuses on the practical impact of public-law litigation on governance. Using litigation to articulate and implement legal norms is an important aspect of what Robert Kagan called “adversarial legalism,” which he decried as “a markedly inefficient, complex, costly, punitive, and unpredictable method of governance and dispute resolution.”[295] The “complexity, fearsomeness, and unpredictability” of American-style litigation “often deter the assertion of meritorious legal claims and compel the compromise of meritorious defenses”; worse, Kagan suggests, “Adversarial legalism inspires legal defensiveness and contentiousness, which often impede socially constructive cooperation, governmental action, and economic development, alienating many citizens from the law itself.”[296]

Finally, the notion—often implicit in critiques of programmatic state litigation—that AGs should stick to more prosaic and uncontroversial functions, like enforcing the auto lemon laws, likewise echoes the broader literature on public-law litigation. Critics of the federal Legal Services Corporation (LSC) during the 104th Congress, for example, prohibited LSC grantees from filing class actions on the ground that “impact litigation” was a distraction from providing bread-and-butter services to individual indigent clients.[297] The class action literature warns that cause-oriented lawyers may be poorly situated to represent the interests of some of their clients, who may be more concerned about more immediate interests than advancing the broader cause.[298] More generally, some commentators have defended the private dispute resolution model as better suited to the institutional competences and legitimacy of courts.[299] State AGs, in other words, are hardly the only people involved in public-law litigation who have been urged to stick to a less grandiose conception of their institutional role.

The fact that common criticisms of state public litigation apply, for the most part, to public-law litigation generally does not mean those criticisms are unimportant. But it does raise the “compared to what?” question we flagged at the beginning of this Part. Robert Kagan grounds the adversarial legalistic structure of our governance in the combination of Americans’ demand for justice and distrust of government.[300] Similarly, Sean Farhang attributes the pervasiveness of private enforcement of public regulation to political polarization and frequent bouts of divided government.[301]

None of these features of American public life are going away anytime soon. On the contrary, as Part I suggested, current rates of polarization may make conventionally “political” solutions to contested policy questions especially unlikely, especially at the federal level. Meanwhile, given the central role that public-law litigation plays in our legal system, reining in state AGs would not (in most cases) result in less adversarial legalism, but simply a shift from state litigation to litigation by class action lawyers, NGOs, and the like.

In short, the alternative to much public litigation by states is probably not—or at least often not—resolution of the underlying controversy by political or bureaucratic means. In our view, then, the more salient question is one that has been largely ignored in the literature to date: how do states compare with other institutional options for pursuing public-law litigation?

B. States as Aggregate Litigants

Public-law litigation typically asserts claims on behalf of diffuse interests, such as consumers, racial minorities, or persons exposed to environmental harms. One of the central questions in American procedural law is how to facilitate litigation by numerous and diffuse persons—such as citizens who benefit from a clean environment—who would likely not have either the incentives or the wherewithal to bring individual lawsuits. The class action is the classic solution, though there is also multi-district litigation, the mass action permitted under some states’ laws, and the rule that organizations can have standing to sue on behalf of their members. We think it makes sense to view state governments as another such mechanism, and so it will be useful to compare state governmental plaintiffs to other means for aggregating diffuse interests in litigation.

We begin with points of similarity. States have many of the same interests that private parties do, and in many cases state litigation will have private analogs (or may be brought contemporaneously with private parties). States own property, for example, and they enter into contracts. And not surprisingly, when they suffer injuries to these sorts of proprietary interests, states have no trouble establishing their standing to sue.[302]

What may be less obvious is that these sorts of interests may support important forms of public-law litigation against the national government. For instance, in Massachusetts v. EPA, the Bay State and several other state governments sought to force the EPA to regulate greenhouse gases under the Clean Air Act.[303] Although the Supreme Court’s ruling on standing relied importantly on the Commonwealth’s sovereign interests, the Court noted that “Massachusetts does in fact own a great deal of the ‘territory alleged to be affected’” by climate change.[304]

Likewise, an important category of legal conflict between the national and state governments involves cooperative federalism programs in which states participate in exchange for federal funds. The Court frequently likens these statutory regimes to contracts between the national government and the states, and it seems clear that state governments could challenge federal administration of the regime based on their contractual interest in enforcing the terms of the deal as the states understand them.[305]

States also have a range of non-proprietary interests that arise out of being governments. Such interests are divvied up into confusing categories of “sovereign” and “quasi-sovereign” interests,[306] though most are relatively straightforward without the terminology. Governments often have responsibilities and prerogatives—regulatory and otherwise—with respect to property they do not own; hence Massachusetts had an interest in “preserv[ing] its sovereign territory” in the climate change case.[307] Governments also have responsibilities to provide benefits to their citizens that can be increased by harmful activity; recall that, in the tobacco litigation, states sued to redress their increased Medicaid expenses arising from their citizens’ tobacco use.[308] And because governments have regulatory responsibilities, they suffer cognizable injuries when they are prevented from enforcing their own laws. That is why, for example, a state government that intervenes in litigation contesting the validity of a state statute has standing to appeal a judgment striking the statute down, even if neither of the original parties files an appeal.[309]

Similarly, state governments can be injured by actions that change or make it more difficult to perform their regulatory responsibilities.[310] In the Texas immigration case, the state argued that it had certain legal responsibilities to all persons lawfully present within its jurisdiction; it was required, for example, to issue such persons drivers’ licenses at a net cost to the State of about $130 per license.[311] This example simply illustrated concretely the basic truth that expanding the population for which a state is responsible inevitably increases the burdens of educating, policing, and otherwise supporting that population.[312] Similarly, Massachusetts’s recent challenge to the Trump Administration’s expansion of religious exemptions to the ACA’s contraceptive mandate stressed that, under state law, reductions in employers’ federal insurance coverage obligations would trigger corresponding costs as the Commonwealth became obligated to fill any resulting gaps.[313] More generally, because state governments are pervasively involved in cooperative federalism arrangements with federal agencies—sharing regulatory responsibilities over benefits programs, education, environmental protection, homeland security, and any number of other areas—changes in federal regulation will often impact the rights and obligations of state governments under these schemes.

Finally, in addition to pursuing their own interests, state governments frequently sue as parens patriae on behalf of their citizens. Parens patriae standing typically requires that the state assert a “quasi-sovereign” interest—that is, “a set of interests that the State has in the well-being of its populace.”[314] The Supreme Court has acknowledged that parens patriae is a “judicial construct that does not lend itself to a simple or exact definition.”[315] But the concept becomes somewhat more tractable when considered alongside more conventional (private) forms of claim aggregation. When a state like Massachusetts or Texas files a lawsuit on behalf of its citizens and relies on injuries to their interests to support its claim to standing, it is typically doing something akin to what the NAACP and the Sierra Club do when they file lawsuits on behalf of their members.[316]

The hornbook doctrine of organizational standing allows an association or other membership organization to sue on behalf of its members so long as “(a) its members would otherwise have standing to sue in their own right; (b) the interests it seeks to protect are germane to the organization’s purpose; and (c) neither the claim asserted nor the relief requested requires the participation of individual members in the lawsuit.”[317] Many national organizations that frequently file claims in federal court are comparable in size to the states. The Sierra Club, for example, claims three and a half million members—enough to be the thirtieth most populous state in the Union, just behind Connecticut and ahead of Iowa.[318] The American Association of Retired Persons (AARP) is roughly the size of California.[319] Such organizations may sue when they can show that at least one member has suffered (or will suffer) an injury in fact.[320] In order to establish standing as parens patriae, by contrast, a state must show that the claimed injury affects a “sufficiently substantial segment of [the state’s] population.”[321] That requirement is not terribly demanding,[322] but it does erect a hurdle that private organizations need not overcome.

Parens patriae cases also markedly resemble private class actions, as state AGs represent the interests of citizens who are not themselves formally parties to the suit. The resemblance holds regardless of whether AGs are pursuing monetary remedies for citizens[323] or seeking injunctive or declaratory relief.[324] Indeed, “parens patriae and private class actions often proceed in tandem, with public and private attorneys working together” to pursue common goals.[325]

These many points of similarity between state public-law litigation and private alternatives underscore the need to situate the work of state AGs within the broader litigation landscape. In many cases—though not all, a significant qualification to which we return below—state litigation will operate as a supplement to, or a substitute for, similar litigation by private individuals or groups. Understanding state litigation that way helps highlight its comparative strengths, while also focusing attention on potential weaknesses.

C. Democratic Litigation?

The most obvious, and important, difference between state and private litigation is that states are democratic governments. The overwhelming majority of state AGs are independently elected, and those who are not are usually accountable to an elected governor.[326] State law generally provides other checks and balances, such as legislative oversight, budgetary controls, or sunshine laws requiring some degree of public transparency. These mechanisms are by no means perfect,[327] but they do suggest that a state AG should be more accountable to a state’s citizens than the leaders of an organization like the Sierra Club are to its members.

In an ideal world, moreover, one might imagine that AGs’ obligation to represent diverse constituencies of voters might cause them to adopt more moderate litigating positions than private groups—thereby ameliorating some of the concerns about partisanship and polarization that we explored above. A state AG represents the whole state—not just the party that elected her.[328] And although partisan assumptions surely shape every AG’s conception of the public interest, we have little doubt that AGs do frame that interest more broadly. State AGs’ responsibilities cut across a wide range of issues, from criminal enforcement to consumer welfare to environmental protection to preventing terrorism.[329] They are, therefore, accountable and responsive to broader interests than the subset of their citizens directly affected by a particular lawsuit. And because state AGs increasingly act on a national stage—collaborating with other states, taking part in cooperative federalism schemes with national officials, and soliciting campaign contributions from national interest groups—they cannot afford to take too parochial a perspective on their activities.

Thus, to the extent that one is concerned that public-law litigation is less democratic than alternate modes of policymaking, one might find good reason to prefer state litigation to analogous litigation by private parties. This point comes with several essential caveats, however. The political and national pressures bearing on state AGs may be a double-edged sword. As we’ve shown, AG campaigns are a lot more expensive than they were in the 1990s, and the imperatives of campaign fundraising may push AGs to espouse more extreme—or simply more consistently red or blue—positions. The more general literature on polarization suggests, after all, that politicians taking highly partisan positions may be responding more to funders than to voters.[330]

Likewise, although one might hope that AGs consider the interests of all citizens, AGs’ incentives to do so are, at the very least, questionable. Every state contains large numbers of both Republicans and Democrats, and to the extent that state public-law litigation has a partisan slant, state citizens not from the AG’s party may strongly prefer that the litigation not be brought. State AGs (or the governors who appoint them) are elected on the same at-large, first-past-the-post system as other statewide officials, which necessarily leaves the minority party unrepresented even where the margin between majority and minority is small.[331] Even if high-profile public lawsuits become campaign issues in AG elections—and they sometimes do—AGs in many states may have little or no incentive to worry about the preferences of citizens from the other party.[332] Representation of all the states’ citizens often depends on the partisan alignments in the state, which will determine whether the AG must compete for the median voter or play to her party base.

This representation problem is, of course, endemic to all unitary decisionmakers elected on a winner-take-all basis. Many Republicans felt shut out of government under the Obama Administration, just as many Democrats do now. Federalism is a partial answer to that problem, as it gives the national out-party the opportunity to control at least some states where it remains a majority,[333] and further decentralization may address it at the state level.[334] But we think the problem feels different when the relevant elected official is a lawyer, and the people of the state are not just his constituents but his clients. The interests (perhaps “preferences” is a better word in this context) of Republicans and Democrats may in many instances be irreconcilable, and it is probably impossible to ask an AG to “represent” all the citizens in many scenarios. At the same time, we find it deeply problematic for a lawyer purporting to act on behalf of all the state’s citizens to ignore the preferences of a large portion of them.[335]

At first blush, private class actions seem preferable on this score—though here, too, matters prove to be more complicated than they first appear. Just as AGs have an obligation to represent the “state,” or “the people,” or “the public interest,” so too class counsel are obligated to represent all the members of the class. The latter obligation is, at least in theory, easier to enforce. Rule 23 of the Federal Rules of Civil Procedure requires judges in class actions to ensure that class counsel can “fairly and adequately represent the interests of the class.”[336] The Supreme Court has observed that “[t]he adequacy inquiry under Rule 23(a)(4) serves to uncover conflicts of interest between named parties and the class they seek to represent.”[337] And if there are conflicts of interest within the class, there are mechanisms to deal with them. Rule 23(c)(5) permits a court to “divide[] [a class] into subclasses”[338] when the class contains members “whose interests are divergent or antagonistic.”[339]

The protections of Rule 23 may not help with the ideological conflicts we have in mind, however. Class counsel is duty-bound to protect the “interests” of absent class members, but—as we hinted above—there is a difference between legal “interests” and “preferences” about law and policy. Class action doctrine tends to conceive of interests in objective terms, analogous to the goals embodied in substantive law. Thus, one might have an “interest” in obtaining a certain form of relief if there is a colorable argument that the law so provides; whether or not one actually wants that relief is largely irrelevant to the adequacy-of-representation inquiry.[340] Derrick Bell’s work on school-desegregation litigation provides an illustration. Bell’s account makes clear that many African-American families opposed such litigation because they thought race-discrimination lawsuits should pursue school quality over integration. But that kind of conflict—over how best to understand the law and what to do about it—is not the kind of conflict of interest that Rule 23 has in mind. On the contrary, as David Marcus has explained, “[j]udges [in school-desegregation litigation] dealt with the problem of conflicts in litigant preferences among class members by denying their relevance. Really at stake, they reasoned, were group rights, and individuals did not matter all that much.”[341]

This feature of class-action litigation has led some commentators to search for means to make class-action litigation more “democratic,” to take better account of individual preferences. Bill Rubenstein, for example, has suggested “[r]ules that require[] individuals or experts filing group-based cases to demonstrate that some level of community dialogue preceded the decision to file, or to show some level of community participation in the filing, or to establish approval for their filings from democratically elected representatives.”[342] AGs are, of course, one category of democratically elected representative. And while existing mechanisms of democratic accountability for state AGs—including independent elections, interdependent relationships with other arms of state government, and various checking and transparency mechanisms grounded in state constitutions and statutes—leave ample room for improvement, they nevertheless remain an important advantage for state litigation as compared to its private alternatives.

D. The Litigation Safeguards of Federalism

In assessing the role that state governments can play in public-law litigation, it is also worthwhile to consider the impact of such litigation on the states’ role in our federal system. Writing in this vein, Daniel Francis has argued that state litigation is one of the “political safeguards of federalism.”[343] Just as Herbert Wechsler argued that states participate in contemporary federalism through their representation in the national legislative branch,[344] and Heather Gerken, Jessica Bulman-Pozen, and Gillian Metzger have contended that states protect their interests through their bureaucratic interactions with the national executive,[345] so Professor Francis argues that states realize their role in modern federalism in part through activity before the judicial branch.[346]

It is easy to appreciate these “litigation safeguards of federalism” when states argue that the national government lacks power to intrude on state policy choices.[347] But the point extends to cases involving horizontal conflicts among states, or cases in which states use litigation to protect their citizens from business practices they deem harmful (or to protect businesses from regulatory demands they deem harmful). Prior to the New Deal, states presided over a purportedly exclusive sphere of state autonomy, and their primary federalism interest was in guarding the boundaries of that sphere. But we now live in an age of concurrent jurisdiction and cooperative federalism, wherein states act in the same policy space as the national government and, much of the time, serve as partners in the same regulatory regimes.[348] States therefore have an interest not just in safeguarding their autonomy to act independently of the national government but also in participating within the broader system of national policymaking and implementation. As Professor Francis puts it, the institutional arrangements of federalism must protect “the ability of the states to participate saliently in governance, regulation, and political life, and to do so independently—that is, neither with the prior permission nor at the direction of the federal government.”[349]

Litigation is one way that states can find a public forum to oppose, support, or seek to shape national policy. As Professor Francis points out, litigation has several advantages in this regard. Filing a lawsuit affords state AGs the opportunity to force their concerns onto the national agenda, in a public setting in which factual claims are submitted to adversarial testing and where decision of the particular issue will not be “bundled” (as in elections) with any number of other issues.[350] Litigation also can clarify the lines of accountability that the Supreme Court often says are critical to a well-functioning federalism, by making clear which governments (or government officials) are responsible for particular policies.[351]

In all these ways, litigation compares favorably to Professor Wechsler’s legislative representation in Congress (which may or may not actually care about state institutional interests)[352] and to forms of bureaucratic “uncooperative federalism” (which are usually not very transparent or public, and which may tend toward prolonged recalcitrance rather than legal resolution). Most of these benefits, Francis emphasizes, are independent of how the cases actually come out; the important point is the availability of the courts as a public, responsive, and relatively level playing field for states to articulate their views.[353]

State litigation may thus be a valuable mechanism for state participation in our federal system generally, without regard to the particular type of lawsuit involved. Given their concurrent jurisdiction over regulatory matters and involvement in cooperative federalism schemes, state governments are important stakeholders in the national political process. In Albert Hirschman’s terms, vertical litigation protects states’ right to “exit” that process and pursue their own vision, while horizontal litigation is more like “voice” within the national process.[354] We do not say that litigation is always or even mostly superior to other forms of involvement, such as political representation in Congress, connections between state and national political parties, the intergovernmental lobby, or bureaucratic consultation and infighting.[355] But federalism has always been about finding more than one basket for one’s eggs.

E. Judicialization and Backlash

State public-law litigation is not only more democratic than many forms of private litigation; it may also be more powerful. As Part II explained, state AGs enjoy various advantages in the litigation realm that may make state public-law litigation more formidable, or simply more feasible, than its private analogues—a consequence that will strike some observers as entirely desirable and others as cause for regret. The key point for present purposes is that there will not always be a private analog to state suits: state litigation has a broader reach given the more expansive scope of state interests and the favorable procedural rules for states.[356]

Consider questions of standing, for example. Even when AGs are asserting the same sorts of interests as private parties, the scope of the state’s interests may be broader than those for the average individual or firm, due to the breadth of states’ activities and holdings. As we described above, states can also establish standing based on interests that flow from their status as governments—interests that lack any private equivalent. In the Texas immigration case, for instance, it is difficult to imagine a private plaintiff who could claim a concrete injury from the Obama Administration’s deferred-action programs. Similarly, some commentators have suggested that state AGs may be “the only plaintiffs who have a shot at standing” to pursue Emoluments Clause challenges against President Trump.[357] At the very least, it seems clear that the AGs’ theory—that the Emoluments Clauses were “material inducements to the states entering the union,” giving states an interest in enforcing “the terms under which [they] participate[] in the federal system”—would not be available to a private individual or group.[358]

AGs also can claim significant advantages when they sue on behalf of individuals, as in parens patriae cases. Because State AGs need not file a class action in order to represent their citizens, they are not bound by Rule 23 and can bring suit much more easily than can a class action attorney. The tobacco cases are a prime example. Hundreds of private suits had foundered on the shoals of class certification before the states stepped in. Among other things, the states were able to avoid difficult questions of predominance that doomed damages class actions requiring individualized evidence of injury or causation.

In addition to these procedural benefits, AGs derive practical advantages from their governmental status. AGs have investigatory powers, such as the ability to issue subpoenas, that enable them to gather information from potential adversaries in the absence of formal discovery.[359] AGs also have tools of publicity that may not be available to private parties and attorneys. A press conference by a state AG, or group of AGs, is likely to carry more weight and capture more attention than a statement by a private legal-advocacy organization or class-action attorney.[360] The publicity associated with AG investigations and litigation may, in turn, enhance the leverage AGs can bring to the bargaining table. And AGs have significant resources at their disposal. Even if their budgets are limited (and in many states they are), AGs can and often do team up with private attorneys with sizeable war chests. In some cases, moreover, AG litigation has been subsidized by private donations: for example, the red-state challenges to the ACA were financed largely by a private lobbying organization.[361]

All of this suggests that state public-law litigation may sweep more broadly than litigation by private individuals and groups. It follows that as state litigation increases, so too does the number of contentious policy issues that will be resolved by litigation (and settlement) rather than via more conventional political processes of legislation and regulation. Whether that is a good or a bad thing depends, of course, on one’s view of the appropriate bounds of “adversarial legalism”—a question we do not purport to answer here.

Instead, we want to make a somewhat different point. The advantages that states currently enjoy in the litigation field are not set in stone, and they could be trimmed back—by courts, by state legislators, or even by federal law. Opponents of state standing already suggest that states should face unique obstacles to standing that ordinary litigants need not confront. In the Texas immigration case, for example, the United States asserted that Texas’s injury was “self-inflicted” and that it was somehow “offset” by benefits that it would experience under the federal policy the state sought to challenge.[362] But there is no general doctrine of self-inflicted injury or offsetting benefits in standing law. Likewise, in Massachusetts v. Mellon, the Supreme Court suggested that state governments cannot assert parens patriae standing to assert their citizens’ federal constitutional rights in a suit against the national government.[363] That is not a disability that any private membership organization would face, even though Mellon’s assertion that the United States itself is the primary representative of its citizens in federal matters would seem to apply there as well.

State AGs also face potential backlash from others within state government. State legislators have the power to slash AGs’ budgets—as has happened in our home state of North Carolina.[364] Legislators might also impose limitations (such as requirements of legislative or gubernatorial approval) on AGs’ ability to initiate suit. Or, to take another example from North Carolina, state legislators might attempt to assert control over the conduct of certain categories of state litigation[365] or to vest litigation authority in government attorneys outside the AGs’ office.[366]

Finally, we can imagine a variety of federal-law responses. As we have noted, a significant number of federal statutes explicitly authorize state governments to sue to enforce federal law, and where this is true, Congress would be free to restrict or condition such suits as it sees fit. Likewise, states sometimes avail themselves of broad general rights to sue under the APA and similar laws, and these general rights could be modified to specify the circumstances under which state AGs may sue. Because most state public-law litigation is brought in federal court, the federal rules of procedure could also be amended to limit the circumstances in which state governments may file suits. And just as the “special solicitude” for states’ standing under Massachusetts v. EPA was a judicial innovation, so too the federal courts may decide to craft special limitations on state lawsuits. We would not rule out the possibility that principles of constitutional federalism might limit federal law’s ability to systematically make it more difficult for states to file lawsuits than other parties, but we suspect the range of action open to Congress and the federal courts on this point is relatively broad.

That state legislatures, Congress, or the federal courts could limit state lawsuits hardly means that they should. A central thrust of our argument has been that state litigation is—on the whole—a uniquely valuable contribution to national debate about matters of shared public concern. We think it would be counterproductive to hamstring state AGs in the ways suggested above, and we think that most concerns about contemporary litigation should be directed at reforming public-law litigation generally, rather than focusing on states.[367] But we do worry that as states take a more prominent and aggressive role in public-law litigation, AGs may invite a backlash that could limit their authority. Indeed, the threat of such a backlash strikes us as directly related to the themes of partisanship and polarization that we have explored in this Article.[368] To the extent that state litigation is viewed as “political” in a pejorative sense, it may be especially vulnerable to retrenchment by political opponents. And courts may refuse to extend favorable treatment to state litigation if they come to see it as a form of political grandstanding, or if it forces them to confront a host of divisive issues they would otherwise avoid.

We close, then, with two points about the future of state public-law litigation. First, we want to sound a note of caution for AGs and others involved in state public-law litigation. State AGs—like any other litigants—have to balance the costs and benefits of potential litigation when deciding whether to proceed. In the previous Part, we argued that the states’ long-term institutional interests deserve significant weight in that calculus, though they will sometimes be trumped by competing imperatives. It bears emphasis that the states’ institutional interests include an interest in maintaining litigation as a distinctive mode of state power. That interest will sometimes counsel restraint, even when the short-term gains of successful litigation would be sizeable.

Second, to the extent that new restrictions are proposed for state litigation, those restrictions should be informed by a careful assessment of the role that state public litigation plays in our federal system and our national politics. We hope the analysis in this Article can contribute to that assessment.


American federalism can be—and in fact nearly always has been—a safety valve for political and social divisions that might otherwise threaten national unity. The states have contributed to the health of our body politic in a wide variety of ways over the course of our history (and at other times they have undermined it). While it seems unlikely that the Founders envisioned the entrepreneurial state litigation of the past twenty years, such litigation has become an important mechanism for state participation in American politics. Like any other institutional feature of our government, that litigation has upsides and downsides. Done right, however, we think state public-law litigation can be a force for easing the political polarization that afflicts our national politics.

  1. *Robert G. Seaks LL.B. ’34 Professor, Duke Law School.
  2. **Alston & Bird Professor, Duke Law School.This paper grows out of a presentation that we gave at the 2017 summer meeting of the National Association of Attorneys General in Big Sky, Montana. We are particularly grateful to Dan Schweitzer and the NAAG for inviting us to participate and for spurring our thinking on these matters. We are also grateful for the opportunity to present this work to the Rockefeller Center for Public Policy at Dartmouth College, the Florida Conference of District Court of Appeal Judges, the Federalist Society’s 2017 Texas Chapters Conference, the Rappaport Center for Law and Public Policy at Boston College Law School, the Federalist Society’s Sixth Annual Executive Branch Review Conference, the Judicial LLM Program at Duke Law School, and at faculty workshops at Duke and Berkeley. And we are indebted to Tom Barnico, Erin Blondel, Caitlin Halligan, Anthony Johnstone, Katherine Mims-Crocker, Elbert Lin, Jonathan Mitchell, Dan Schweitzer, Neil Siegel, Michael Tigar, and Mark Tushnet for helpful comments; to Jane Bahnson and Guangya Liu of the Duke Law Library for help with the data; and to Paul Brzyski, Abby Frisch, Grace Hitzeman, Colleen O’Leary, and Emani Walks for research assistance. None of these people should be assumed to endorse our conclusions here. All the views expressed, as well as any errors, remain ours alone.
  3. .Sue Owen, Greg Abbott Says He Has Sued Obama Administration 25 Times, Politifact (May 10, 2013), [].
  4. .See infra section II(B)(4).
  5. .See infra section II(B)(5).
  6. .See infra subpart I(A).
  7. .See, e.g., Ben Christopher, For California Attorney General, Suing Trump Again and Again Is a Team Sport, L.A. Daily News (Nov. 30, 2017), [] (describing blue-state suits against the Trump Administration); Alan Neuhauser, State Attorneys General Lead the Charge Against President Donald Trump, U.S. News (Oct. 27, 2017), [] (describing blue-state suits against the Trump Administration and red-state suits against the Obama Administration); Owen, supra note 1 (describing red-state suits against the Obama Administration).
  8. .Democratic Att’ys Gen. Ass’n (Feb. 9, 2018), [].
  9. .E.g., Massachusetts v. EPA, 549 U.S. 497, 509 (2007).
  10. .See, e.g., West Virginia v. EPA, 136 S. Ct. 1000 (2016) (granting application for stay of EPA’s rule in petition filed by twenty-nine states); Lawrence Hurley & Valerie Volcovici, U.S. Supreme Court Blocks Obama’s Clean Power Plan, Sci. Am. (Feb. 9, 2016), [] (noting that “coal producer West Virginia and oil producer Texas,” along with “several major business groups,” led the effort).
  11. .Compare Brief of the States of Maryland, California, Connecticut, Delaware, Hawaii, Illinois, Iowa, New Mexico, New York, Oregon, and Vermont, the District of Columbia, and the Virginia Islands as Amici Curiae in Support of Petitioners, NFIB v. Sebelius, 567 U.S. 519 (2012) (No. 11-398), with Brief for State Respondents, NFIB v. Sebelius, 567 U.S. 519 (2012) (No. 11-398) (filed on behalf of Alabama, Alaska, Arizona, Colorado, Florida, Georgia, Idaho, Indiana, Iowa, Kansas, Louisiana, Maine, Michigan, Mississippi, Nebraska, Nevada, North Dakota, Ohio, Pennsylvania, South Carolina, South Dakota, Texas, Utah, Washington, Wisconsin, and Wyoming). That the great State of Iowa appears twice in these citations is not a typo: the case divided the Attorney General and the Governor, who appeared on different sides.
  12. .Compare, e.g., Brief of the Commonwealth of Virginia as Amicus Curiae in Support of Petitioners at 7, Obergefell v. Hodges, 135 S. Ct. 2584 (2015) (No. 14-556, 14-562, 14-571, 14-574), Brief of the State of Hawaii as Amicus Curiae in Support of Petitioners at 1, Obergefell, 135 S. Ct. 2584 (No. 14-556, 14-562, 14-571, 14-574), Brief of Massachusetts, California, Connecticut, Delaware, the District of Columbia, Illinois, Iowa, Maine, Maryland, New Hampshire, New Mexico, New York, Oregon, Pennsylvania, Rhode Island, Vermont, and Washington as Amici Curiae in Support of Petitioners at 1, Obergefell, 135 S. Ct. 2584 (No. 14-556, 14-562, 14-571, 14-574), and Brief of the State of Minnesota as Amicus Curiae in Support of Petitioners at 1, Obergefell, 135 S. Ct. 2584 (No. 14-556, 14-562, 14-571, 14-574) (all supporting the right of same-sex couples to marry), with Brief of Amicus Curiae State of Alabama in Support of Respondents at 1, Obergefell, 135 S. Ct. 2584 (No. 14-556, 14-562, 14-571, 14-574), Brief of Louisiana, Utah, Texas, Alaska, Arizona, Arkansas, Georgia, Idaho, Kansas, Montana, Nebraska, North Dakota, Oklahoma, South Dakota, and West Virginia as Amici Curiae Supporting Respondents at 1, Obergefell, 135 S. Ct. 2584 (No. 14-556, 14-562, 14-571, 14-574), and Brief of South Carolina as Amicus Curiae in Support of Respondents at 2, Obergefell, 135 S. Ct. 2584 (No. 14-556, 14-562, 14-571, 14-574) (all rejecting a right to same-sex marriage).
  13. .E.g., William H. Pryor, Jr., Comment, 31 Seton Hall L. Rev. 604, 607 (2001); Professor John Langbein, Panel Two: The Politics and Economics of Government-Sponsored Litigation (June  22, 1999), in Manhattan Inst., Regulation by Litigation: The New Wave of Government-Sponsored Litigation, [].
  14. .See, e.g., Pryor, supra note 11, at 608 (“The purpose of the tobacco litigation . . . was to establish through the action of several states a national policy that is properly reserved to state legislatures and to Congress in the exercise of its enumerated powers.”).
  15. .Paul Nolette, Federalism on Trial: State Attorneys General and National Policymaking in Contemporary America 200 (2015).
  16. .Id. at 30–31.
  17. .Id. at 200–01.
  18. .Neuhauser, supra note 5 (quoting former Maine AG, James Tierney).
  19. .See generally Robert A. Kagan, Adversarial Legalism: The American Way of Law (2001).
  20. .For an accessible overview of the major characteristics of today’s polarization, see Nolan McCarty, What We Know and Don’t Know About Our Polarized Politics, Wash. Post (Jan. 8, 2014), []. For more extended treatments, see Michael Barber & Nolan McCarty, Causes and Consequences of Polarization, in Am. Pol. Sci. Ass’n, Negotiating Agreement In Politics 38 (Jane Mansbridge & Cathie Jo Martin eds., 2013); Ronald Brownstein, The Second Civil War: How Extreme Partisanship has Paralyzed Washington and Polarized America (2007); Thomas E. Mann & Norman J. Ornstien, It’s Even Worse Than It Looks: How the American Constitutional System Collided with the New Politics of Extremism 43–58 (2012).
  21. .See, e.g., Toward a More Responsible Two-Party System: A Report of the Committee on Political Parties, Am. Pol. Sci. Rev., Sept. 1950, at 18–19; see also David A. Hopkins, Red Fighting Blue: How Geography and Electoral Rules Polarize American Politics 78–79 (2017) (discussing the APSA report). Later on, political scientists worried that the parties were dying out. See id. at 84–95. They weren’t. The cycles of social scientists’ fears suggest that current predictions about the necessarily enduring nature of polarization should also be taken with a grain of salt. We are thus partial to the prediction that Marc Hetherington and Thomas Rudolph offer: “Things Will Probably Get Better, but We Are Not Sure How.” Marc J. Hetherington & Thomas J. Rudolph, Why Washington Won’t Work 212 (2015).
  22. .It may also be the case that participants in American political debate are less willing to bracket disagreements about the nature of the good life than they once were. The rights-based liberalism of John Rawls and Ronald Dworkin—which was committed, in principle at least, to bracketing such disagreements—seems far less ascendant on the American Left than it was. See, e.g., Michael Sandel, Introduction to Liberalism and its Critics 3–4, 8–9 (Michael Sandel ed., 1984) (discussing the priority of the “right” over the “good” in late-twentieth-century liberalism). The traditional Right, of course, was never committed to this sort of bracketing, although the rise of libertarianism on the contemporary Right may amount to a move in that direction.
  23. .E.g., Malcolm M. Feeley & Edward Rubin, Federalism: Political Identity and Tragic Compromise 115 (2008).
  24. .See, e.g., Robert Schapiro, Polyphonic Federalism: Toward the Protection of Fundamental Rights 54–55, 92 (2009); Jessica Bulman-Pozen, Partisan Federalism, 127 Harv. L. Rev. 1077, 1080–81 (2014).
  25. .See, e.g., Ernest A. Young, The Volk of New Jersey? State Identity, Distinctiveness, and Political Culture in the American Federal System 6 (Feb. 24, 2015) (unpublished manuscript) (available on SSRN at []).
  26. .See, e.g., Joshua Holland, Under Trump, Red States Are Slashing the Safety Net and Blue States Are Fighting Back, The Nation (Jan. 30, 2018), [] (“Is America turning into two different republics sharing one set of borders?”).
  27. .See, e.g., Christopher Hare & Keith T. Poole, The Polarization of Contemporary American Politics, 46 Polity 411, 411–13 (2014) (concluding, based on roll-call votes, that “[p]olarization of the Democratic and Republican Parties is higher than at any time since the end of the Civil War”).
  28. .Cynthia R. Farina, Congressional Polarization: Terminal Constitutional Dysfunction?, 115 Colum. L. Rev. 1689, 1694 (2015).
  29. .Hare & Poole, supra note 25, at 416 fig.1 (showing ideological dispersion of the parties in Congress 1879–2013).
  30. .Farina, supra note 26, at 1694. According to the National Journal’s ideological rankings of members of Congress, for example, the number of Representatives located between the most liberal Republican and the most conservative Democrat in the House dropped from 344 in 1982 to four in 2013. Chris Cillizza, The Ideological Middle Is Dead in Congress. Really Dead., Wash. Post (Apr. 10, 2014), []. In the Senate, there were fifty-eight senators in this overlap-space in 1982; by 2013, none. Id.
  31. .Farina, supra note 26, at 1694.
  32. .The Polarization of the Congressional Parties, (Mar. 21, 2015), []; see David W. Brady & Hahrie Han, An Extended Historical View of Congressional Party Polarization (Dec. 2, 2004) (unpublished manuscript) (on file with author).
  33. .See, e.g., Morris P. Fiorina, Samuel J. Abrams & Jeremy C. Pope, Culture War? The Myth of a Polarized America 7–8 (3d ed. 2011).
  34. .See, e.g., Alan I. Abramowitz, The Disappearing Center: Engaged Citizens, Polarization, and American Democracy 83 (2010); Alan I. Abramowitz & Kyle L. Saunders, Is Polarization a Myth?, 70 J. Pol. 542, 553–54 (2008) (“The high level of ideological polarization evident among political elites in the United States reflects real divisions within the American electorate.”).
  35. .See Hetherington & Rudolph, supra note 19, at 28–33; Political Polarization in the American Public, Pew Res. Ctr. (June 12, 2014), [] (reporting that, in 2014, 27% of Democrats and 36% of Republicans saw the other party as “a threat to the well-being of the country”).
  36. .See Hetherington & Rudolph, supra note 19, at 30–31 (discussing increases in Democratic and Republican negativity towards the other party over time). Your humble authors remain a happy exception.
  37. .See id. at 73–91, 94.
  38. .See id. at 25; Frances E. Lee, Beyond Ideology: Politics, Principles, and Partisanship in the U.S. Senate 18–21 (2009) (explaining that political parties’ institutional interest in “winning elections and wielding power” can bring them into conflict, even when they are not ideologically opposed on an issue).
  39. .See McCarty, supra note 18 (“The combination of high ideological stakes and intense competition for party control of the national government has all but eliminated the incentives for significant bipartisan cooperation on important national problems. Consequently, polarization has reduced congressional capacity to govern.”). 
  40. .See Hetherington & Rudolph, supra note 19, at 40–42.
  41. .On the subject of gridlock, see generally, Symposium, The American Congress: The Legal Implications of Gridlock, 88 Notre Dame L. Rev. 2065 (2013).
  42. .See, e.g., Diana Epstein & John D. Graham, Polarized Politics and Policy Consequences (RAND Corp., Occasional Paper 2007), at 17, [].
  43. .There are important caveats here. The legislative process builds in enough veto-gates and effective supermajority requirements that the minority party can often still gum things up even under conditions of unified government. Moreover, unified government can sometimes expose fissures in the majority party, producing something reminiscent of gridlock under divided government. See, e.g., Louis Jacobson, The Year of Single Party Control and Supermajorities, Governing (Jan. 7, 2013), [] (discussing examples from state government). The first year of the Trump administration seemed to bear out that hypothesis, as the Republicans failed to move major legislation (including the much-promised repeal of the Affordable Care Act) through Congress. By the end of the year, however, the gigantic tax overhaul had changed the picture substantially. See, e.g., Naomi Jagoda, Trump Signs Tax Bill Into Law, The Hill (Dec. 22, 2017), [].
  44. .See, e.g., Morris P. Fiorina, America’s Missing Moderates: Hiding in Plain Sight, Am. Int., [] (discussing excesses by both parties during recent periods of unified government).
  45. .See supra note 19.
  46. .See, e.g., Jeffrey R. Lax & Justin H. Phillips, The Democratic Deficit in the States, 56 Am. J. Pol. Sci. 148, 164 (2011) (finding that states “tend to ‘overshoot’ relative to the median voter’s specific policy preferences”).
  47. .See, e.g., Thomas E. Mann, Foreword to American Gridlock: The Sources, Character, and Impact of Political Polarization, at xxii–xxiii (James A. Thurber & Antoine Yoshinaka eds., 2015).
  48. .See, e.g., Hetherington & Rudolph, supra note 19, at 28–33; Lilliana Mason, Uncivil Agreement: How Politics Became Our Identity 50–54 (2018) (“[C]urrent levels of partisan antipathy have moved beyond pure disagreements of principle. Partisans dislike each other to a degree that cannot be explained by policy disagreement alone.”).
  49. .See, e.g., Anthony Johnstone, Hearing the States, 45 Pepp. L. Rev. 575, 589 (2018) (“Federalism is the classic constitutional solution to reduce the costs of political contestation through policy decentralization.”).
  50. .See Susan Milligan, The Popular Republicans, U.S. News (Feb. 2, 2018), [] (discussing Massachusetts Governor Charlie Baker).
  51. .See, e.g., Erica Greider, Big, Hot, Cheap, and Right: What America Can Learn from the Strange Genius of Texas 32 (2013).
  52. .Boris Shor & Nolan McCarty, The Ideological Mapping of State Legislatures, 105 Am. Pol. Sci. Rev. 530, 540 fig.7 (2011).
  53. .Id.
  54. .Id. at 546 fig.15.
  55. .See Chris Cillizza, The Most Popular Governor in the Country Is a Republican from Massachusetts. Yes, Really., CNN (July 22, 2017), []; Former Governors’ Bios, Nat’l Governors Ass’n, [] (listing the party affiliations of Massachusetts’s current and prior governors).
  56. .See Cillizza, supra note 53 (reporting a 71% approval rating for Massachusetts Governor Charlie Baker); David Mark, Republican Governors Thrive in Blue States, Polling Shows, Morning Consult (July 18, 2017), [] (stating that Governors Charlie Baker of Massachusetts, Larry Hogan of Maryland, and Phil Scott of Vermont all had “enviable approval ratings” before their re-elections); Milligan, supra note 48 (reporting that “Nevada’s Brian Sandoval, Maryland’s Larry Hogan, Massachusetts’s Charlie Baker, and Vermont’s Phil Scott”—all Republicans in blue states—“remain among the most popular governors in the country”).
  57. .Kevin Deutsch, Why Blue States Elect Red Governors, 21 Wash. U. Pol. Rev., Nov. 11, 2014, [].
  58. .See Cillizza, supra note 53; Joshua Miller, Why Is Charlie Baker So Popular?, Bos. Globe (Sept. 27, 2017), []; Milligan, supra note 48.
  59. .The discussion here is exploratory; we make no strong claims about causation. There is widespread debate about what causes polarization generally. See Farina, supra note 26 (summarizing political science literature). We express no view on those broader questions.
  60. .See Hetherington & Rudolph, supra note 19, at 33–39.
  61. .See Cindy D. Kam & Robert A. Mikos, Do Citizens Care About Federalism? An Experimental Test, 4 J. Emp. Leg. Stud. 589, 598 (2007) (reporting results from the 2000 Attitudes Toward Government Study, but concluding that “[t]hese findings are consistent with those reported by other scholars, using other nationally representative surveys”); see also State Governments Viewed Favorably as Federal Rating Hits New Low, Pew Res. Ctr. (Apr. 15, 2013), []. The Pew Research Center explains that:Overall, 63% say they have a favorable opinion of their local government, virtually unchanged over recent years. And 57% express a favorable view of their state government – a five-point uptick from last year. By contrast, just 28% rate the federal government in Washington favorably. That is down five points from a year ago and the lowest percentage ever in a Pew Research Center survey.Id. Interestingly, levels of trust in the federal government themselves vary significantly from state to state. See Paul Brace & Martin Johnson, Does Familiarity Breed Contempt? Examining the Correlates of State-Level Confidence in the Federal Government, in Public Opinion in State Politics 19 (Jeffrey E. Cohen ed., 2006).
  62. . Hetherington & Rudolph, supra note 19, at 157–61.
  63. .See, e.g., Samuel C. Patterson, The Political Cultures of the American States, 30 J. Pol. 187, 195–96 (1968). Patterson argues that:No one would expect the American political culture to be uniformly distributed spatially; our evidence is adequate enough to show that the political culture of Mississippi is not the same as that of Iowa. Some states may stand out more distinctively than others, and some group themselves in sections or regions that are distinctive.Id. See also John J. Harrigan & David C. Nice, Politics and Policy in States and Communities 10 (10th ed. 2008) (observing that “numerous studies have found that political culture influences the kind of policies adopted by states”).
  64. .See Daniel J. Elazar, The American Mosaic: The Impact of Space, Time, and Culture on American Politics 58 (1994); see also David Hackett Fischer, Albion’s Seed: Four British Folkways in America 189–90, 200–01 (1989) (examining in depth the influence of Puritan folkways in New England).
  65. .See Cillizza, supra note 53 (observing that Massachusetts governor Charlie Baker ran “as basically a non-partisan manager” who would “watch the state’s pocketbook,” but that he favored “abortion rights and featured his brother’s coming-out story in a legendary campaign ad”); Milligan, supra note 48 (citing social welfare policies of New England Republican governors, such as expanding Medicaid).
  66. .On political norms and conventions generally, see, for example, Neil S. Siegel, Sustaining Collective Self-Governance and Collective Action: A Constitutional Role Morality for the Trump Era and Beyond, 107 Geo. L.J. (forthcoming 2018).
  67. .See Deutsch, supra note 55 (noting that “state and national politics are two different animals, with different issues at play”).
  68. .See, e.g., Milligan, supra note 48 (discussing four Republican governors working with Democratic-controlled legislatures); see also Cillizza, supra note 53 (describing the popularity of Massachusetts governor Charlie Baker).
  69. .See, e.g., Tyler Bridges, Can This Governor Teach Democrats How to Win in the South?, Politico (Sept. 2, 2017), [] (profiling Louisiana governor John Bel Edwards).
  70. .See supra note 36 and accompanying text.
  71. .See, e.g., Hetherington & Rudolph, supra note 19, at 44, 97–98 (hypothesizing that salience affects the influence that trust exerts on political opinions and giving examples); Hopkins, supra note 19, at 99–100 (“[T]he newfound salience of social and cultural concerns during the 1990s was the driving force behind the divergence of the blue Northeast and Pacific Coast from the red South and interior West.”).
  72. .See, e.g., Michael Barone, The 49 Percent Nation, 33 Nat’l J. 1710, 1710–12 (June 9, 2001), [] (emphasizing the narrow popular vote margins in recent elections).
  73. .Id. (formatting omitted).
  74. .Matthew S. Levendusky & Jeremy C. Pope, Red States vs. Blue States: Going Beyond the Mean, 75 Pub. Opinion Q. 227, 242–44 (2011).
  75. .See Hopkins, supra note 19, at 41–45 (giving examples).
  76. .State Partisan Composition, Nat’l Conf. St. Legislatures (Apr. 11, 2018), [].
  77. .Id.
  78. .David Weigel, Incoming N.J. Governor Plans a Swing to the Left—And a Model for the Country, Wash. Post (Jan. 15, 2018),–and-a-model-for-the-country/2018/01/13/25f06238-f7d7-11e7-a9e3-ab18ce41436a_story.html?noredirect=on&utm_term=.fa458987b2c4 [].
  79. .Dan Balz, Red, Blue States Move in Opposite Directions in a New Era of Single-Party Control, Wash. Post (Dec. 28, 2013), [].
  80. .See id. (“The risk is that with unified control, governors and their like-minded legislators push beyond the views of their citizenry, particularly in states where public opinion is more evenly divided.”); Lax & Phillips, supra note 44, at 149 (studying congruence between state policy and public opinion and finding that “state policy is far more polarized than public preferences”) (formatting omitted). As we noted above, some research suggests—somewhat counterintuitively—that extreme policy may be more likely in states like North Carolina, where the two parties are in pitched battle for control of state government, than in states in which the majority party can count on continuing supremacy. See supra note 36 and accompanying text.
  81. .See, e.g., Heather K. Gerken, Dissenting by Deciding, 57 Stan. L. Rev. 1745, 1755, 1783–86 (2005); Ernest A. Young, Welcome to the Dark Side: Liberals Rediscover Federalism in the Wake of the War on Terror, 69 Brooklyn L. Rev. 1277, 1301–05, 1311 (2004).
  82. .See, e.g., Bulman-Pozen, supra note 22 (suggesting that state challenges to federal law stem largely from partisanship). 
  83. .Michael W. McConnell, Federalism: Evaluating the Founders’ Design, 54 U. Chi. L. Rev. 1484, 1493 (1987).
  84. .See generally Ernest A. Young, Exit, Voice, and Loyalty as Federalism Strategies: Lessons from the Same-Sex Marriage Debate, 85 U. Colo. L. Rev. 1133, 1135–36, 1140–42 (2014).
  85. .See, e.g., Ernest A. Young, A General Defense of Erie Railroad Co. v. Tompkins, 10 J.L. Econ. & Pol’y 17, 68–69 (2013) (describing the “Legal Process” model of federalism, under which “[w]hat is ‘reserved’ to the States . . . is regulatory authority over matters upon which Congress has been unwilling or unable to legislate”); see also Henry M. Hart, Jr., The Relations Between State and Federal Law, 54 Colum. L. Rev. 489, 525–35 (1954) (developing this view).
  86. .See Kenneth S. Lowande & Sidney M. Milkis, “We Can’t Wait”: Barack Obama, Partisan Polarization and the Administrative Presidency, 12 Forum 3, 3 (2014).
  87. .Id. at 9. One of the tools Obama used was the conditional waiver—allowing states to avoid requirements of federal law, such as No Child Left Behind, only if they adopted new standards prescribed by the Obama Administration. Id. at 11–12.
  88. .Rebecca Harrington, Trump Signed 90 Executive Actions in His First 100 Days—Here’s What Each One Does, Bus. Insider (May 3, 2017), []. The one major exception is President Trump’s effort to replace the Obama Administration’s Deferred Action for Childhood Arrivals (DACA) immigration program, implemented through unilateral executive action, with a legislated immigration package. See generally Noah Rothman, Congress Doesn’t Want the Responsibility Anymore, Commentary (Feb. 22, 2018), [] (surveying executive action under the Trump administration and suggesting the cause is as much congressional abdication as executive overreach). Of course, if no legislation is forthcoming, Trump may likewise reach for his pen and phone.
  89. .On the political safeguards theory, see, for example, Jesse H. Choper, Judicial Review and the National Political Process: A Functional Reconsideration of the Role of the Supreme Court 175–84 (1980) (arguing that the states’ political representation obviates the need for judicial review in federalism cases); Larry D. Kramer, Putting the Politics Back into the Political Safeguards of Federalism, 100 Colum. L. Rev. 215, 233–34 (2000) (arguing that political parties protect states by linking the fortunes of national- and state-level politicians); Herbert Wechsler, The Political Safeguards of Federalism: The Role of the States in the Composition and Selection of the National Government, 54 Colum. L. Rev. 543, 543–46 (1954) (arguing that the states’ representation in Congress provides a powerful check on national action); Ernest A. Young, Two Cheers for Process Federalism, 46 Vill. L. Rev. 1349, 1350–52 (2001) (arguing that political safeguards are not sufficient to replace judicial review but nonetheless provide an important check on national action).
  90. . See, e.g., Bradford R. Clark, Separation of Powers as a Safeguard of Federalism, 79 Texas L. Rev. 1321, 1339 (2001) (“[T]he lawmaking procedures prescribed by the Constitution safeguard federalism in an important respect simply by requiring the participation and assent of multiple actors. These procedures make federal law more difficult to adopt by creating a series of ‘veto gates.’”); Young, Two Cheers, supra note 87, at 1361–63 (stressing the role of legislative inertia in protecting federalism).
  91. .See Clark, supra note 88, at 1393–94; Ernest A. Young, Executive Preemption, 102 Nw. U. L. Rev. 869, 869–71, 900 (2008).
  92. .See, e.g., Jessica Bulman-Pozen & Heather K. Gerken, Uncooperative Federalism, 118 Yale L.J. 1256, 1271–80 (2009) (describing this phenomenon as “uncooperative federalism”).
  93. .See, e.g., Daniel Francis, Litigation as a Political Safeguard of Federalism, 49 Ariz. St. L.J. 1023, 1025–26 (2017) (contending that the state litigation is an undervalued safeguard for federalism).
  94. .Johnstone, supra note 47, at 599.
  95. .See Bond v. United States, 564 U.S. 211, 223–24 (2011) (rejecting the United States’ contention that individuals lack standing to raise claims that a federal statute exceeds Congress’s enumerated powers).
  96. .See supra note 5 and accompanying text.
  97. .See generally Nat’l Ass’n of Attorneys General, State Attorneys General: Powers and Responsibilities 45, 84, 121–22, 234, 270–73 (Emily Myers ed., 3d ed. 2013) [hereinafter NAAG] (discussing the role of state AGs and areas of joint federal-state enforcement, such as antitrust and environmental law).
  98. .William P. Marshall, Break Up the Presidency? Governors, State Attorneys General, and Lessons from the Divided Executive, 115 Yale L.J. 2446, 2452 (2006). Interesting questions arise when a given state itself has divided government. In North Carolina, for example, the Republican-dominated legislature has jousted with the Democratic governor for control over litigation on behalf of the state. See Current Operations Appropriations Act of 2017, § 147-17, 2017 N.C. Sess. Laws 248, 261–66 (amending various provisions so as to strengthen the General Assembly’s control over litigation involving the constitutionality of state statutes). Those sorts of problems are not without analogs at the federal level, as when the Republican-controlled House of Representatives sought to defend the federal Defense of Marriage Act after the Obama Administration announced that it was unwilling to do so. See United States v. Windsor, 570 U.S. 744, 753–54 (2013) (discussing the role of the House’s Bipartisan Legal Advisory Group). But because most states lack a unitary executive, it is also not uncommon for the state governor and attorney general to be from different parties. See, e.g., Wikipedia, Government of Massachusetts, [] (listing Massachusetts’s governor as a Republican and its AG as a Democrat); Wikipedia, Government of Illinois, [] (stating that Illinois currently has a Republican governor and a Democratic AG). This creates thorny state separation of powers problems on which federal practice can provide little guidance. Cf. The Federalist No. 70, at 472–80 (Jacob E. Cooke ed., 1961) (Alexander Hamilton) (arguing against a plural executive). We do not explore those problems further here, other than to suggest that a non-unitary executive may make it easier for voters to weigh in on the litigation decisions of a state government, simply because those decisions are not folded into a simple up-or-down vote on the performance of the entire executive branch.
  99. .See infra notes 110–16 and accompanying text.
  100. .Alan Greenblatt, The Avengers General, Governing (2003), [].
  101. .See id. (describing Republican AGs’ critiques of entrepreneurial state litigation and the ensuing formation of the Republican Association of Attorneys General).
  102. .See generally Richard H. Fallon, Jr., John F. Manning, Daniel J. Meltzer & David L. Shapiro, Hart and Wechsler’s The Federal Courts and the Federal System 73–76 (7th ed. 2015) (describing the shift from a private “dispute resolution” model to a public rights or “law declaration” model of the judicial function).
  103. .Cf. Randy E. Barnett, Four Senses of the Public Law-Private Law Distinction, 9 Harv. J.L. & Pub. Pol’y 267, 268–72 (1986) (highlighting the difficulty of separating “public” and “private” law).
  104. .We focus here on actions by state AGs. But it bears emphasis that important litigation efforts have sometimes been led by governors or other state officials, by membership organizations representing state institutions (such as the National Governors’ Association), or by local governments. The leading challenges to President Trump’s effort to punish “sanctuary cities” acting contrary to federal immigration policy, for example, have been brought by the Cities of San Francisco and Chicago. Laura Jarrett & Tal Kopan, Federal Judge Again Blocks Trump from Punishing Sanctuary Cities, CNN (Sept. 15, 2017), [].
  105. .Cornell W. Clayton, Law, Politics and the New Federalism: State Attorneys General as National Policymakers, 56 Rev. Pol. 525, 538 (1994); see Thomas R. Morris, States Before the U.S. Supreme Court: State Attorneys General as Amicus Curiae, 70 Judicature 298, 299 (1987) (observing that “state attorneys general tended to look upon their role as being merely ministerial functionaries of the state administration”).
  106. .Eric N. Waltenburg & Bill Swinford, Litigating Federalism: The States Before the U.S. Supreme Court 45 (1999).
  107. .During the 1970s and early 1980s, AGs’ budgets expanded at rates that “outpaced the growth of general government spending in every state.” Cornell W. Clayton & Jack McGuire, State Litigation Strategies and Policymaking in the U.S. Supreme Court, 11 Kan. J.L. & Pub. Pol’y 17, 18 (2001). Between 1970 and 1989 the mean number of attorneys increased from 51 to 148, and the median budget from $612,089 to $9.9 million. Waltenburg & Swinford, supra note 104.
  108. .Clayton, supra note 103; see also Kevin C. Newsom, The State Solicitor General Boom, 32 App. Prac. 6, Winter 2013, at 7–8 (describing the rise of appellate attorneys with private experience in state solicitor general offices).
  109. .See William L. Webster, The Emerging Role of State Attorneys General and the New Federalism, 30 Washburn L.J. 1, 5 (1990) (“In short order the states asserted themselves in dramatic fashion. . . . Attorneys general were called ‘fifty regulatory Rambos’ by one individual.”).
  110. .Id.; see also Clayton, supra note 103, at 535–36 (describing states’ efforts to secure regulatory and enforcement authority in areas including antitrust and consumer protection).
  111. .NAAG, supra note 95, at 46.
  112. .See, e.g., Hart-Scott-Rodino Antitrust Improvements Act of 1976, Pub. L. No. 94-435, sec. 301, § 4(c), 90 Stat. 1383, 1394 (1976) (codified at 15 U.S.C. § 15(c) (2012)) (authorizing states to sue as parens patriae in federal court on behalf of their citizens to secure treble damages for a variety of federal antitrust violations); see also Margaret H. Lemos, State Enforcement of Federal Law, 86 N.Y.U. L. Rev. 698, 712 (2011) (“As state attorneys general assumed new prominence, provisions for state enforcement began to proliferate in Congress. New provisions have been enacted by virtually every Congress in the last two decades.”). 
  113. .Richard P. Ieyoub & Theodore Eisenberg, State Attorney General Actions, the Tobacco Litigation, and the Doctrine of Parens Patriae, 74 Tul. L. Rev. 1859,1863 (2000); Jack Ratliff, Parens Patriae: An Overview, 74 Tul. L. Rev. 1847, 1850 (2000).
  114. .Alfred L. Snapp & Son, Inc. v. Puerto Rico ex rel. Barez, 458 U.S. 592, 607 (1982).
  115. .Margaret H. Lemos, Aggregate Litigation Goes Public: Representative Suits by State Attorneys General, 126 Harv. L. Rev. 486, 495–96, 496–97 nn.39–40 (2012). Whether Congress could confer authority on state AGs to sue in circumstances where state law denies it is an interesting question, but beyond the scope of this article.
  116. .See, e.g., EEOC v. Fed. Express Corp., 268 F. Supp. 2d 192, 197 (E.D.N.Y. 2003) (citing Connecticut v. Physicians Health Servs. of Conn., Inc., 287 F.3d 110, 121 (2d Cir. 2002)) (“[S]tanding provisions in many . . . statutes implicitly authorize[] parens patriae standing by using language that permits any ‘person’ who is ‘aggrieved’ or ‘injured’ to bring suit.”); see also Massachusetts v. Bull HN Info. Sys., Inc., 16 F. Supp. 2d 90, 103 (D. Mass. 1998) (quoting 29 U.S.C. § 630(a)) (reasoning that AG has statutory standing to sue under Age Discrimination in Employment Act as “‘legal representative’ of the people of the [state] for the purposes of this action”); Minn. v. Standard Oil Co. (Ind.), 568 F. Supp. 556, 563–66 (D. Minn. 1983) (permitting state to sue as parens patriae under § 210 of Economic Stabilization Act of 1970, which permitted suit by any “person” because “when a state acts in its quasi-sovereign capacity in a parens patriae action, . . . [a] harm to the individual citizens becomes an injury to the state, and the state in turn becomes the plaintiff”).
  117. .See generally Ieyoub & Eisenberg, supra note 111, at 1864–75 (describing the contours of parens patriae doctrine and its grounding in common law).
  118. .Id. at 1860 (“Before the states’ litigation, the tobacco industry had not lost a smoking case . . . .”).
  119. .Anthony J. Sebok, Pretext, Transparency and Motive in Mass Restitution Litigation, 57 Vand. L. Rev. 2177, 2184–88 (2004) (describing the history of tobacco litigation).
  120. .Id. at 2189; see also id. (describing Minnesota’s consumer-fraud approach as a notable exception).
  121. .Mike Moore, The States Are Just Trying to Take Care of Sick Citizens and Protect Children, 83 A.B.A. J. 53, 53 (1997).
  122. .Sebok, supra note 117, at 2190.
  123. .Hanoch Dagan & James J. White, Governments, Citizens, and Injurious Industries, 75 N.Y.U. L. Rev. 354, 371–73 (2000). Four states settled separately for approximately $36.8 billion, bringing the total to roughly $243 billion. W. Kip Vicusi, The Governmental Composition of the Insurance Costs of Smoking, 42 J.L. & Econ. 575, 577 (1999).
  124. .Ieyoub & Eisenberg, supra note 111, at 1860 (“The scope of interstate attorney general cooperation was unprecedented.”).
  125. .See generally Margaret H. Lemos, Privatizing Public Litigation, 104 Geo. L.J. 515, 532–33, 538–46 (2016) (analyzing the costs and benefits of partnerships between public and private attorneys).
  126. .Michael DeBow, The State Tobacco Litigation and the Separation of Powers in State Governments: Repairing the Damage, 31 Seton Hall L. Rev. 563, 564 (2001). Critics are quick to note that the settlement is being financed largely by smokers, who now pay more for cigarettes. Id.; see also Sebok, supra note 117, at 2181 (“As an executive at R.J. Reynolds ironically put it, ‘[T]here’s no doubt that the largest financial stakeholder in the [tobacco] industry is the state governments.’”).
  127. .See Margaret H. Lemos & Max Minzner, For-Profit Public Enforcement, 127 Harv. L. Rev. 853, 855 & n.6 (2014) (offering examples); Lemos, supra note 110, at 732–33 & n.153 (same).
  128. .Lemos & Minzner, supra note 125, at 866–67 (describing “revolving fund[]” arrangements at the state level).
  129. .See Ieyoub & Eisenberg, supra note 111, at 1862 (arguing that “it is [the states’] legal theories, together with the precedent of concerted attorney general action, that have the greatest implications for joint action on other fronts”).
  130. .See Dagan & White, supra note 121, at 355–57 (focusing on the states’ restitutionary claims and describing similar claims against gun manufacturers and lead-paint makers).
  131. .See generally Ieyoub & Eisenberg, supra note 111, at 1862, 1875–83 (describing parens patriae standing as applied in the tobacco litigation and its potential for future suits). For a more critical take, see DeBow, supra note 124, at 565 (arguing that “the tobacco template could conceivably be applied to a wide range of industries in future government litigation—including, perhaps, makers of alcoholic beverages, fatty foods, and automobiles” and warning of a “substantial danger that state attorneys general and local government officials will regularly succumb to the temptation of the tobacco example, and will seek to achieve regulatory and tax outcomes through litigation . . . .”).
  132. .Complaint at 3, Ohio v. McKesson Corp., No. 1:12-cv-00185-RBW (Ohio Ct. Com. Pl. Feb. 26, 2018).
  133. .549 U.S. 497, 520 (2007).
  134. .46 U.S. 95, 105–07, 110 (1983) (holding that person subjected to illegal chokehold by police lacked standing to seek an injunction, as there was no guarantee that the plaintiff would be subjected to similar acts by police in the future); see also O’Shea v. Littleton, 414 U.S. 488, 490, 503–04 (1974) (denying that a case or controversy existed regarding discriminatory law enforcement practices on similar grounds).
  135. .See, e.g., Pennsylvania v. Porter, 659 F.2d 306, 314–15 (3d Cir. 1981) (holding that state had standing as parens patriae to enjoin police misconduct while noting that “many individual victims may be unable to show the likelihood of future violations of their rights”). Courts have reasoned that, because the state represents all of its citizens, it will typically have little trouble establishing that a harm that has occurred in the past will likely befall some citizens in the future. Id. This sort of probabilistic reasoning generally does not work for private litigants. See generally Summers v. Earth Island Inst., 555 U.S. 488, 491, 494–501 (2009) (denying standing to a private environmental organization that had asserted a statistical certainty that some of its members would be injured by some of the challenged Forest Service actions). We suspect the difference is that cases like O’Shea and Lyons are grounded importantly in concerns about judicial intervention in state and local governance—a concern that is radically less compelling when the state itself is the plaintiff.
  136. .John C. Coffee, Jr., “When Smoke Gets in Your Eyes”: Myth and Reality About the Synthesis of Private Counsel and Public Client, 51 DePaul L. Rev. 241, 241–42 (2001).
  137. .See Lemos, State Enforcement, supra note 110, at 500–10 (detailing the procedural requirements for private class actions versus the requirements for similar suits brought by the State).
  138. .Mississippi v. AU Optronics Corp., 571 U.S. 161, 164 (2014); cf. People v. Greenberg, 946 N.Y.S.2d 1, 7 (App. Div. 2012) (holding that suit by state AG was exempt from similar jurisdictional rules governing private securities actions).
  139. .See, e.g., EEOC v. Waffle House, Inc., 534 U.S. 279, 294 (2002) (holding that arbitration agreement between employee and employer did not bar EEOC from bringing enforcement action).
  140. .See Lemos, State Enforcement, supra note 110, at 505–06 (collecting cases).
  141. .Sage v. Appalachian Oil Co., Inc., No. 3:92-CV-176, 2:93-CV-229, 1994 WL 637443, at *2 (E.D. Tenn. Sept. 7, 1994).
  142. .See Douglas Ross, Safeguarding Our Federalism: Lessons for the States from the Supreme Court, 45 Pub. Admin. Rev. 723, 727–28 (1985) (describing NAAG’s genesis and functions). Another significant institutional response was the creation of the State and Local Legal Center (SLLC), which files amicus briefs on behalf of member associations. Id. at 728.
  143. .See Clayton, supra note 103, at 544 (“[T]he decision to participate as amicus curiae is determined largely by the personal interests and felt political pressures on individual attorneys general.”).
  144. .Waltenburg & Swinford, supra note 104, at 48. If anything, the number of state briefs filed understates the level of state activity. Thanks in large part to NAAG’s coordination efforts, states frequently band together on amicus briefs. A study of merits-stage state amicus briefs found that the average number of joining states jumped from 2.4 in the 1970s to 13.9 in the 1990s. Clayton & McGuire, supra note 105, at 24–25; see also Waltenburg & Swinford, supra note 104, at 48 (“NAAG’s focus on the coordination of state amicus activity has resulted in substantial levels of joining behavior. Accordingly, where it is rare to find more than two amici joining together on a pre-certiorari amicus brief, on average six states coalesce . . . .”). A more recent study of state amicus filings reveals similar joining behavior at the certiorari stage: using data on state certiorari filings compiled by Dan Schweitzer at NAAG, Greg Goelzhauser and Nicole Vouvalis report that “[d]uring the 2001–2009 terms, state-sponsored amicus briefs urging review in state-filed cases were joined by an average of about 18 states, and only 5 of the 88 briefs filed were signed by a single state.” Greg Goelzhauser & Nicole Vouvalis, State Coordinating Institutions and Agenda Setting on the U.S. Supreme Court, 41 Am. Pol. Res. 819, 825 (2013). One veteran state litigator attributes these changes in part to technological advances, noting that email has made it far easier for dispersed AGs’ offices to share drafts. See Letter from Tom Barnico, Dir. AG Program, Boston College Law School, to authors (July 20, 2018) (on file with authors).
  145. .Margaret H. Lemos & Kevin M. Quinn, Litigating State Interests: Attorneys General as Amici, 90 N.Y.U. L. Rev. 1229, 1235 (2015).
  146. .Dan Frosch & Jacob Gershman, Abbott’s Strategy in Texas: 44 Lawsuits, One Opponent: Obama Administration, Wall St. J. (June 24, 2016), [].
  147. .Steve LeBlanc & Bob Salsberg, Massachusetts’ Maura Healey Helping Lead Effort to Litigate Trump, (Dec. 18, 2017), [].
  148. .Nolette, supra note 13, at 21 app. at 221; see also id. fig.2.1.
  149. .Id. at 21–22 & fig.2.2.
  150. .Id. at 22.
  151. .Frosch & Gershman, supra note 144.
  152. .See Complaint at 11–12, Washington v. Trump, No. 2:17-cv-00141-JLR, 2017 WL 462040 (W.D. Wash. Jan. 3, 2017) (alleging individual rights violations as well as violations of the Administrative Procedure Act (APA)).
  153. .See Brief Addressing the Merits of the State of Indiana and 16 Other States as Amicus Curiae in Support of the Bipartisan Legal Advocacy Group of the U.S. House of Representatives at 4–8, United States v. Windsor, 570 U.S. 744 (2013) (No. 12-307), 2013 WL 390993 [hereinafter Windsor Pro-DOMA States’ Brief] (arguing that neither federalism nor equal protection analysis supported heightened scrutiny of DOMA); Brief on the Merits of the States of New York, Massachusetts, California, Connecticut, Delaware, Illinois, Iowa, Maine, Maryland, New Hampshire, New Mexico, Oregon, Rhode Island, Vermont, and Washington, and the District of Columbia as Amici Curiae in Support of Respondent at 3, United States v. Windsor, 570 U.S. 744 (2013) (No. 12-307), 2013 WL 840031 [hereinafter Windsor Anti-DOMA States’ Brief] (arguing that DOMA denied equal protection and infringed states’ authority to regulate marriage). There is, moreover, important diversity within categories. As we discuss further below, the relevant legal constraints in each of the first three categories—federalism and separation-of-powers principles and individual rights—may be either constitutional or statutory in character. We do not distinguish between constitutional and statutory claims because we think that both constitutional and statutory norms serve constitutive functions in many instances. See generally Ernest A. Young, The Constitution Outside the Constitution, 117 Yale L.J. 408, 464 (2007) (discussing the constitutive role of statutory and other non-entrenched norms in structuring the government and identifying individual rights).
  154. .See, e.g., Windsor Pro-DOMA States’ Brief, supra note 151, at 2–3.
  155. .These claims almost always concern the Commerce Clause—the catch-all, default power that sustains most federal legislation. But occasionally they involve other powers, such as Congress’s power to enforce the Reconstruction Amendments. City of Boerne v. Flores, 521 U.S. 507, 512 (1997). Boerne was a private claim brought against a local government by church officials under the federal Religious Freedom Restoration Act (RFRA). But the case drew state amici filings on both sides. See Brief of the States of Maryland, Connecticut, Massachusetts, and New York as Amici Curiae in Support of Respondent, City of Boerne v. Flores, 521 U.S. 507 (1997) (No. 95-2074), 1996 WL 10282 (defending RFRA); Brief for Amici States of Ohio, Arizona, Colorado, Delaware, Florida, Hawaii, Idaho, Mississippi, Nevada, New Hampshire, North Carolina, Oklahoma, the Commonwealth of Pennsylvania, and the Territories of American Samoa, Guam, and the Virgin Islands in Support of Petitioner, City of Boerne, Texas, City of Boerne v. Flores, 521 U.S. 507 (1997) (No. 95-2074), 1996 WL 695519 (attacking RFRA). And Ohio Solicitor Jeffrey Sutton was given oral argument time to argue against RFRA’s constitutionality.
  156. .14 States Sue to Block Health Care Law, CNN (Mar. 23, 2010), []; see generally NFIB v. Sebelius, 567 U.S. 519, 547–58 (2012) (opinion of Roberts, C.J.) (accepting those arguments).
  157. .In United States v. Lopez, 514 U.S. 549 (1995), for example, a criminal defendant prosecuted for possessing a firearm within 1,000 feet of a school argued (successfully) that the federal prohibition did not regulate interstate commerce. Id. at 551–52. In United States v. Morrison, 529 U.S. 598 (2000), an individual defendant in a civil case argued (again successfully) that the federal private right of action for victims of “gender-motivated violence” exceeded Congress’s power under both the Commerce Clause and Section Five of the Fourteenth Amendment. Id. at 601–02, 604.
  158. .See, e.g., Gonzales v. Raich, 545 U.S. 1, 15 (2005) (addressing claim by users of medicinal marijuana seeking declaratory and injunctive relief that the federal Controlled Substances Act, as applied to them, exceeded Congress’s Commerce power).
  159. .In Lopez, several states filed in support of the Gun Free School Zones Act. See Brief for the States of Ohio, New York, and the District of Columbia as Amici Curiae in Support of Petitioner, United States v. Lopez, 514 U.S. 549 (1995) (No. 93-1260), 1994 WL 16007793. No state filed in support of Mr. Lopez, but he did get a brief filed by several national organizations representing state and local governments. See Brief of the National Conference of State Legislatures, National Governors’ Association, National League of Cities, National Association of Counties, International City/County Management Association, and National Institute of Municipal Law Officers, Joined by the National School Boards Association, as Amici Curiae in Support of Respondent at 13, United States v. Lopez, 514 U.S. 549 (1995) (No. 93-1260), 1994 WL 16007619 (arguing that “the Commerce Clause does not authorize enactment of the Gun Free School Zones Act”).
  160. .See generally Raich, 514 U.S. at 15–19; Wickard v. Filburn, 317 U.S. 111, 118–29 (1942).
  161. .See, e.g., United States v. Comstock, 560 U.S. 126, 130 (2010) (upholding broad federal power to imprison sexual predators under the Necessary and Proper Clause); Raich, 545 U.S. at 34–36 (Scalia, J., concurring in the judgment) (arguing that the Necessary and Proper Clause allows Congress to regulate noncommercial activity that affects commerce).
  162. .See NFIB, 567 U.S. at 574 (upholding the ACA under the Taxing Clause).
  163. .See, e.g., Richard H. Fallon, Jr., The “Conservative” Paths of the Rehnquist Court’s Federalism Decisions, 69 U. Chi. L. Rev. 429, 476–77 (2002); Ernest A. Young, Just Blowing Smoke? Politics, Doctrine, and the Federalist Revival After Gonzales v. Raich, 2005 Sup. Ct. Rev. at 1, 39–40 (“A roll-back of the national regulatory state was never in the cards; there are simply too many precedential, institutional, and political constraints pressing the Court to uphold relatively broad federal power.”).
  164. .See Printz v. United States, 521 U.S. 898, 933 (1997); New York v. United States, 505 U.S. 144, 188 (1992).
  165. .See generally Lynn A. Baker, Conditional Federal Spending After Lopez, 95 Colum. L. Rev. 1911, 1918–19, 1923–31 (1995) (noting the broad potential of conditional spending to circumvent limits on Congress’s enumerated powers). The leading case remains South Dakota v. Dole, 483 U.S. 203 (1987).
  166. .See NFIB, 567 U.S. at 575.
  167. .See, e.g., City of Santa Clara v. Trump, 250 F. Supp. 3d 497, 507 (N.D. Cal. Apr. 25, 2017).
  168. .For example, a thorny question in the sanctuary cities litigation is the extent to which local officials are simply being asked to cooperate with federal law enforcement in the same way any private citizen would have to or are instead being “commandeered” into enforcing federal immigration policy. See, e.g., City of Philadelphia v. Sessions, 280 F. Supp. 3d 579, 597–99 (E.D. Pa. 2017); Alison Frankel, DOJ Wants to Change the Constitutional Conversation in Sanctuary Cities Cases, Reuters (Mar. 7, 2018), [].
  169. .See, e.g., Martin H. Redish, The Constitution as Political Structure 26 (1995) (contrasting “dual” and “cooperative” federalism); Philip J. Weiser, Towards a Constitutional Architecture for Cooperative Federalism, 79 N.C. L. Rev. 663, 665 (2001) (categorizing congressional acts that “invite state agencies to implement federal law” as “cooperative federalism” programs).
  170. .See Printz, 521 U.S. at 933–34.
  171. .See Nat’l Fed’n of Indep. Bus. v. Sebelius, 567 U.S. 519, 585–88 (2012) (opinion of Roberts, C.J.) (stating that states are free to opt out of the Medicaid expansion while remaining within the original Medicaid program). In some circumstances a robust opt-out right could kill a federal scheme that required cooperation, and at that extreme the difference between trying to limit the scope of federal policy and preserving a right of opt-out dissolves. This may have been Justice Story’s hope, for example, in Prigg v. Pennsylvania, 41 U.S. (16 Pet.) 539 (1842). Although Prigg upheld Congress’s power to enact the Fugitive Slave Law and broadly construed its preemptive force, Story may have hoped that the Court’s holding that Congress could not require state and local officials to participate in the law’s enforcement would gut its effectiveness. See id. at 532, 598, 672–73; David C. Currie, The Constitution in the Supreme Court: The First Hundred Years 1789–1888, 245 n.54 (1985). Unfortunately, he turned out to be wrong about that. See Paul Finkelman, Sorting Out Prigg v. Pennsylvania, 24 Rutgers L.J. 605, 664 (1993).
  172. .The convoluted saga of attempts to avoid state sovereign immunity also includes cases in which individuals with financial claims against states enlist various other sovereign entities, including state governments, to prosecute those claims on the individuals’ behalf. These efforts have not generally had much success. See, e.g., New Hampshire v. Louisiana, 108 U.S. 76, 88–89 (1883) (holding that New Hampshire could not pursue financial claims against another state where New Hampshire had no interest of its own).
  173. .See Nat’l League of Cities v. Usery, 426 U.S. 833, 852 (1976) (holding that, at least in some circumstances, Congress may not regulate state governmental entities performing traditional governmental functions), overruled by Garcia v. San Antonio Metro. Transit Auth., 469 U.S. 528, 531 (1985); see also Ernest A. Young, State Sovereign Immunity and the Future of Federalism, 1999 Sup. Ct. Rev. at 1, 31–32 (discussing claims under National League of Cities as a species of “immunity federalism”).
  174. .See, e.g., Seminole Tribe v. Florida, 517 U.S. 44, 72 (1996); Hans v. Louisiana, 134 U.S. 1, 18 (1890).
  175. .See, e.g., Gonzaga Univ. v. Doe, 536 U.S. 273, 276 (2002) (Federal Educational Rights and Privacy Act (FERPA) does not create enforceable private rights under 42 U.S.C. § 1983); Alexander v. Sandoval, 532 U.S. 275, 293 (2001) (concluding no implied right of action for disparate impact discrimination under Title VI).
  176. .See, e.g., Brief of Amicus Curiae States of California et al., Supporting the State of Florida, et al., at 4, Seminole Tribe v. Florida, 517 U.S. 44 (1996) (No. 94-12), 1995 WL 17008502 (May 3, 1995) (contending that a statute mandating state participation in federal programs was inconsistent with principles of federalism).
  177. .531 U.S. 356 (2001).
  178. .528 U.S. 62 (2000). Judge Sutton, then in private practice at Jones Day, argued both Garrett and Kimel on behalf of the state defendants. Id.; Garrett, 531 U.S. at 356.
  179. .See also Alexander, 532 U.S. at 276, in which Judge Sutton, in private practice, appeared as counsel of record on behalf of the State of Alabama successfully opposing recognition of a private right of action for disparate impact discrimination under Title VI of the Civil Rights Act of 1964.
  180. .Commonwealth of Massachusetts v. Laird, 400 U.S. 886 (1970); see also id. at 886 (Douglas, J., dissenting) (noting that Massachusetts had authorized the suit by a specific legislative enactment).
  181. .See CNN, Obama-I’ve Got a Pen and a Phone, YouTube, [] (recording a speech by President Obama, wherein he expressed frustration with congressional gridlock and his intent to take unilateral action).
  182. .See 136 S. Ct. 2271, 2272 (2016) (affirming the injunction of the DAPA program and DACA program expansions in Texas v. United States, 86 F. Supp. 3d 591, 606, 678 (S.D. Tex. 2015)).
  183. .Brief for the State Respondents, United States v. Texas, 136 S. Ct. 2271 (2016) (No. 15-674), 2016 WL 1213267, at *1.
  184. .For assessments of the so-called administrative safeguards of federalism, compare Gillian E. Metzger, Administrative Law as the New Federalism, 57 Duke L.J. 2023, 2028, 2101–09 (2008) (asserting that administrative law is well-suited to preserving federalism), with Stuart M. Benjamin & Ernest A. Young, Tennis with the Net Down: Administrative Federalism Without Congress, 57 Duke L.J. 2111, 2114, 2145–54 (2008) (arguing that federalism requires insistence that Congress play the primary role).
  185. .See, e.g., Michigan v. EPA, 135 S. Ct. 2699, 2706 (2015) (considering challenge by twenty-three states to EPA rule regulating air pollutants on the ground that the agency did not consider costs of regulation as required by statute).
  186. .See Texas v. United States, 201 F. Supp. 3d 810, 815–16 (N.D. Tex. 2016) (granting preliminary injunction on behalf of thirteen states and other plaintiffs).
  187. .See Brief of Amici Curiae Vermont, Alabama, Alaska, Arizona, Arkansas, California, Colorado, Connecticut, Delaware, Florida, Georgia, Hawaii, Idaho, Illinois, Indiana, Iowa, Kansas, Kentucky, Louisiana, Maine, Maryland, Massachusetts, Minnesota, Mississippi, Missouri, Montana, Nevada, New Hampshire, New Jersey, New Mexico, New York, North Carolina, North Dakota, Ohio, Oklahoma, Oregon, Pennsylvania, Rhode Island, South Carolina, South Dakota, Tennessee, Utah, West Virginia, Virginia, Washington, Wisconsin, and Wyoming in Support of Respondent at 4, Wyeth v. Levine, 555 U.S. 555 (2009) (No. 06-1249); Brief of the National Conference of State Legislatures as Amicus Curiae Supporting Respondents at 5, Wyeth v. Levine, 555 U.S. 555 (2009) (No. 06-1249), []; see also Brief of the Center for State Enforcement of Antitrust and Consumer Protection Laws, Inc. as Amicus Curiae Supporting Respondent at 6, Wyeth v. Levine, 555 U.S. 555 (2009) (No. 06-1249), []. On the preemption preambles, see Catherine M. Sharkey, Preemption by Preamble: Federal Agencies and the Federalization of Tort Law, 56 DEPAUL L.REV. 227 (2007).
  188. .Heckler v. Chaney, 470 U.S. 821, 832 (1985).
  189. .549 U.S. 497, 528 (2007).
  190. .Hari M. Osofsky & Jacqueline Peel, The Role of Litigation in Multilevel Climate Change Governance: Possibilities for a Lower Carbon Future? 30 Env’t & Plan. L.J. 303, 310 (2013).
  191. .See, e.g., Batalla Vidal v. Nielsen, 279 F. Supp. 3d 401, 407, 438 (E.D.N.Y. 2018) (granting preliminary injunction against repeal of DACA program in suit by New York and fifteen other states). Similar litigation challenges the Trump Administration’s effort to overturn President Obama’s “clean power plan.” See, e.g., Richard Valdmanis, States Challenge Trump Over Clean Power Plan, Sci. Am. (Apr. 6, 2017), [].
  192. .See Hawaii v. Trump, 859 F.3d 741, 765 (9th Cir. 2017), rev’d on other grounds, 138 S. Ct. 2392 (2018) (“EO2 harms the State’s interests because (1) students and faculty suspended from entry are deterred from studying or teaching at the University; and (2) students who are unable to attend the University will not pay tuition or contribute to a diverse student body.”).
  193. .See supra notes 10 (Obergefell briefs) and 151 (Windsor briefs).
  194. .See Lemos & Quinn, supra note 138, at 1257.
  195. .See id. at 1255–56 (observing that many Republican AG briefs filed in criminal procedure cases are not opposed by state briefs favoring the criminal defendant).
  196. .See, e.g., City of West Covina v. Perkins, 525 U.S. 234, 235 (1999) (Ohio SG Jeffrey Sutton, who had filed an amicus brief on behalf of twenty-nine states, arguing on the city’s behalf by leave of court).
  197. .372 U.S. 335 (1963). See Yale Kamisar, Gideon v. Wainwright and Related Matters: An Armchair Discussion Between Professor Yale Kamisar and Vice President Walter Mondale, 32 L. & Ineq. 207, 207 (2014) (discussing Mondale’s role in Gideon).
  198. .See, e.g., Georgia v. Tenn. Copper Co., 206 U.S. 230, 231, 236 (1907) (hearing the State of Georgia’s public nuisance claim against Tennessee copper companies for discharging noxious gases that crossed the border into Georgia).
  199. .See, e.g., Timothy D. Lytton, Using Tort Litigation to Enhance Regulatory Policy Making: Evaluating Climate-Change Litigation in Light of Lessons from Gun-Industry and Clergy-Sexual-Abuse Lawsuits, 86 Texas L. Rev. 1837, 1843 (2008) (“By the late 1990s, municipalities began suing the gun industry to recover the costs of law enforcement and emergency medical services related to gun violence.”).
  200. .See, e.g., Nolette, supra note 13, at 23–24.
  201. .Forty-six states, the District of Columbia, Puerto Rico, American Samoa, Guam, Northern Mariana Islands, and the Virgin Islands, joined the Master Settlement Agreement with the tobacco companies. Four other states—Florida, Minnesota, Mississippi, and Texas—settled their cases separately. Supra note 121 and accompanying text; see also NAAG, supra note 90, at 388.
  202. .Nolette, supra note 13, at 24. The tobacco companies, along with NAAG, petitioned Congress for a national legislative settlement, but no such legislation was ever enacted. Dagan & White, supra note 121, at 369–70.
  203. .See Nolette, supra note 13, at 49–59 (offering a detailed account of the pharmaceutical litigation); id. at 25 tbl.2.1 (listing the top fifteen industries targeted in multistate litigation).
  204. .See Sarah T. Reise, State and Local Governments Move Swiftly to Sue Equifax, Ballard Spahr Consumer Fin. Monitor (Oct. 3, 2017), [].
  205. .See generally Lemos, State Enforcement, supra note 105, at 707–17 (describing the contours of state enforcement of federal laws in a variety of areas).
  206. .See supra notes 105–10 and accompanying text.
  207. .See Lemos, State Enforcement, supra note 110, at 737–40 (describing the power of enforcement).
  208. .See, e.g., id. at 739–40 (describing how state enforcement has molded federal antitrust doctrine).
  209. .See Arizona v. United States, 567 U.S. 387 (2012) (holding much of Arizona’s effort preempted).
  210. .See Lemos, State Enforcement, supra note 110, at 743–44 (explaining that the EPA was embroiled with lawsuits at the time but that it adopted Spitzer’s legal strategy within a few weeks, bringing a suit against power plants that New York intervened in). We leave to one side here the converse scenario, which occurs when states refuse to enforce federal law or repeal state laws that parallel federal laws. These state decisions may also significantly undermine or affect federal policy. For example, Colorado’s decision to end state prohibition of most marijuana use made it significantly more difficult for federal authorities to further national drug policies in that state. See generally Ernest A. Young, Modern-Day Nullification: Marijuana and the Persistence of Federalism in an Age of Overlapping Regulatory Jurisdiction, 65 Case W. Res. L. Rev. 769, 774–76 (2015).
  211. .See, e.g., supra notes 13–15 and accompanying text; Nolette, supra note 13, at 200–01 (“The long-term effect of the federal government’s invitation for AGs to influence national policy has been to encourage AGs to define state interests much differently than in the past. A crucial element of this shift is that while AGs have traditionally acted as representatives of their states, they have increasingly claimed the ability to represent a broader range of interests. This includes representing the interests of individuals as opposed to the states themselves.”); Jim Copland & Rafael A. Mangual, Left-Wing AGs Are Playing Politics with the Law, Nat’l Rev. (Sept. 29, 2016), [] (“Left-wing state attorneys general are acting less like legal representatives of their constituents and more like partisan political activists.”); Anthony Johnstone, The Appeal of State Attorneys General in a Federal System, H-FedHist, H-Net Reviews (July 2017), [] (reviewing Nolette, supra note 13) (“As AGs become more responsive to national interests, they may become less responsive to their own states’ interests.”); Brooke A. Masters, States Flex Prosecutorial Muscle, Wash. Post (Jan. 12, 2005), [] (addressing how some business groups view AGs as “ambitious politicians more interested in making headlines than consistent, viable policy”); Walter Olson, Opinion, Partisan Prosecutions: How State Attorneys General Dove Into Politics, N.Y. Post (Mar. 30, 2017), [] (“These days, packs of red- and blue-team AGs roam the political landscape looking for fights to get into . . . .”).
  212. .Greenblatt, supra note 98.
  213. .Id.
  214. .Id.
  215. .Attorney General (state executive office), Ballotpedia, [] (showing party control of state AG seats).
  216. .Greenblatt, supra note 98 (noting that the formation of RAGA “brought the office of state attorney general back into political play around the country”).
  217. .Because the number of AG races in any given election cycle is not uniform, an overall tally of total receipts would be misleading.
  218. .See, e.g., Dan Levine & Lawrence Hurley, Trump Bump: Court Fights Draw Big Money into Attorney General Races, Reuters (July 31, 2017), [] (describing spending in AG races generally); see also Christopher R. Nolen, Election Law, 41 U. Rich. L. Rev. 121, 139 (2006) (reporting that out-of-state organization contributed $2.1 million to candidate for attorney general, prompting reforms); Andrew Brown, Big Money Funding Race for WV Attorney General, Charleston Gazette-Mail (Oct. 1, 2016), [] (describing RAGA’s significant contributions to the West Virginia AG race); Kathleen Gray, Campaign Cash Flowing into Races for Attorney General, Secretary of State in Michigan, Detroit Free Press (Jan. 31, 2018), [] (describing state office campaign spending in Michigan); Jon Lender, Jepsen Solicits Special-Interest Funds to Help Out-of-State Political Ally, Hartford Courant (Sept. 29, 2017), [] (describing spending by Connecticut lobbyists on Virginia AG race); Ben Wieder, Big Money Comes to State Attorney-General Races, Atlantic (May 8, 2014), [] (describing spending in various states’ AG races).
  219. .Steven Mufson, Conservatives Pour Money into Races for State Attorneys General, Wash. Post (Sept. 23, 2016), [].
  220. .Levine & Hurley, supra note 216.
  221. .Alan Greenblatt, State AGs Used to Play Nice in Elections. Not Anymore, Governing Mag. (Nov. 15, 2017), [].
  222. .Id.
  223. .Adam Bonica, Mapping the Ideological Marketplace, 58 Am. J. Pol. Sci. 367, 367 (2014). Bonica argues that:The idea underlying the ideological measures is straightforward. Contributors are assumed—at least in part—to distribute funds in accordance with their evaluations of candidate ideology. That is, contributors will on average prefer ideologically proximate candidates to those who are more distant. The pattern of who gives to whom allows me to simultaneously locate both contributors and recipients.Id. For a detailed description of methodology, see id. at 368–73.
  224. .See id. at 369 (“In any given state, between 70% and 90% of contributors who fund state campaigns also give to federal campaigns, providing an abundance of bridge observations . . . . Candidates who run for both state and federal office provide additional bridge observations.”).
  225. .The discerning reader will notice that the lines for the most conservative Democrats and most liberal Republicans appear to hit the same point (just above zero), though not at the same time. That is in fact one person: Louisiana AG Buddy Caldwell, who was elected as a Democrat in 2008 and as a Republican in 2012. (Because CFscores are based on lifetime contributions, they do not capture a candidate’s shift to the left or right.) It’s worth noting that Louisiana elected the most conservative Democratic AG in 2000 and 2004 as well. Georgia held that title in 2002 and 2006, before that AG seat likewise flipped red in 2010. See Adam Bonica, Database on Ideology, Money in Politics, and Elections: Public Version 2.0, Stan. U. Libr. 2016, []. It bears emphasis that the static nature of the CFscores we are using here—that is, the fact that they do not capture changes in a candidate’s contributor base from one year to the next—dampens our ability to glean trends in polarization from the DIME data. Many AGs serve multiple terms, and several states had the same AG through all or most of the period for which data are widely available. The trend lines for those incumbent AGs will be flat, even if the AGs’ contributors—or their litigation strategies—moved to the left or the right. That said, most states did experience turnover in the AG’s office between 2000 and 2012, meaning that new candidates—with new scores—were elected during that time. The lack of any discernible movement toward greater ideological divergence among Democratic and Republican AGs is therefore somewhat surprising, given trends in polarization in Congress and in other offices, and worthy of further study.
  226. .Bonica, supra note 221, at 376. The distance between the CFscores of the mean Democrat and Republican AGs was similar to (but slightly higher than) that for governors. Id. at 376–77. One interesting difference between AGs and other state officials is that the former seem to be divided more symmetrically than the latter. Professor Bonica’s data show higher (that is, more extreme) CFscores for Republicans than for Democrats in the U.S. House and Senate, and in state legislatures and governorships. State AGs, by contrast, are more evenly balanced—at least in terms of their contributors. Bonica, supra note 221, at 377 fig.2; see Johnstone, supra note 47, at 608 (observing that “Professor Adam Bonica’s study of campaign finance contributions finds attorneys general to be slightly more polarized than other state officials, but also demonstrates they are more balanced as a group across the ideological spectrum than other state or federal elected officials, state courts, or even federal circuit court judges.”).
  227. .Greenblatt, supra note 219 (quoting Paul Nolette).
  228. .Id. (quoting Jim Tierney).
  229. .See Greenblatt, supra note 98 (comparing the total number of Republican AGs in the United States in 1999 and 2003).
  230. .Compare Nat’l Conf. of State Leg., 2017 State & Legislative Party Composition (Mar. 1, 2017), [] (showing party composition of state legislatures and governors), with Attorney General (State Executive Office), Ballotpedia, [] (showing party control of state AG seats). We say “at least” because Hawaii’s AG is technically a non-partisan official, appointed by the state’s elected governor. See id.
  231. .Morris, supra note 103, at 302 (“Most of the divisions did not consist of a significant number of states on either side, but rather one or two states on either side or one or two dissenters from an otherwise large number of states.”). Perhaps not surprisingly, Morris found that Commerce Clause cases were the most common sites of interstate conflict. Id.
  232. .Clayton, supra note 103, at 539.
  233. .Paul Nolette, State Litigation During the Obama Administration: Diverging Agendas in an Era of Polarized Politics, 44 Publius 451, 455–56 (2014).
  234. .See id. at 455–57, 457 tbl.1 (discussing the increase in horizontal conflicts involving partisan participation among AGs from 1999 through 2013).
  235. .Nolette, supra note 13, at 30–31. Specifically, Nolette argues that “[s]ince the George W. Bush administration, policy-forcing litigation has chiefly been an avenue for Democratic AGs to expand national regulation beyond the level preferred by Congress or federal agencies.” Id. at 31.
  236. .Id. at 31–32.
  237. .For literature discussing the problems with efforts to code judicial decisions as “liberal” or “conservative,” see Anna Harvey & Michael J. Woodruff, Confirmation Bias in the United States Supreme Court Judicial Database, 29 J.L. Econ. & Org. 414, 415 (2013) (finding that the labeling of cases depended more on the preferences of the Court than on the disposition of the case); William M. Landes & Richard A. Posner, Rational Judicial Behavior: A Statistical Study, 1 J. Legal Analysis 775, 776–78, 780–81 (2009) (explaining the numerous variables involved in classifying a decision); Young, Blowing Smoke, supra note 161, at 11–12 (noting that the inconsistent nature of these classifications poses a significant problem in accurate coding).
  238. .In his study of amicus briefs, for example, Professor Nolette defines cases as partisan in which Republican or Democratic AGs constituted at least 80% of participating AGs. Nolette, supra note 231, at 455. Nolette does not specify how he identifies polarization in multistate litigation.
  239. .Lemos & Quinn, supra note 143, at 1233, 1243.
  240. .Id. at 1251–52.
  241. .Id. at 1268.
  242. .Id. at 1255–56.
  243. .A focus on brief-joining may also tend to overstate the importance of partisanship, in the sense that it may capture relatively low-stakes position-taking rather than truly impactful legal action. The AG who supplies the twentieth signature to an amicus brief is probably not devoting a great deal of her office’s resources to the case, and her decision to join is unlikely to have much impact on the law. Such brief-joining may offer opportunities for AGs to signal to and satisfy co-partisans—and such behavior may in turn have ripple effects for other aspects of AGs’ work—but nonetheless is meaningfully different from, say, spearheading litigation on behalf of the state as party.
  244. .See Colin Provost, An Integrated Model of U.S. State Attorney General Behavior in Multi-State Litigation, 10 St. Pol. & Pol’y Q. 1, 10, 14–15 (2010) [hereinafter Provost, Integrated Model]; Colin Provost, State Attorneys General, Entrepreneurship, and Consumer Protection in the New Federalism, 33 Publius, Spring 2013, at 37, 47–49 [hereinafter Provost, State Attorneys General]; Colin Provost, The Politics of Consumer Protection: Explaining State Attorney General Participation in Multi-State Lawsuits, 59 Pol. Res. Q. 609, 612–15 (2006) [hereinafter Provost, Politics of Consumer Protection].
  245. .Provost, Integrated Model, supra note 242, at 15–17.
  246. .Nolette, supra note 13, at 28 (“When a regulatory settlement will occur regardless of whether or not a particular AG participates, most AGs are likely to participate in order to get a share of the settlement proceeds even if they disagree with the underlying legal theories in the threatened lawsuit.”).
  247. .Provost, Integrated Model, supra note 242, at 17; see also Lemos & Quinn, supra note 143, at 1263–66 (making this point and using the states’ briefing in District of Columbia v. Heller as an example).
  248. .See Lynn A. Baker & Ernest A. Young, Federalism and the Double Standard of Judicial Review, 51 Duke L.J. 75, 109–10 (2001) (defining and contrasting “vertical” and “horizontal aggrandizements”).
  249. .See id. at 121–24.
  250. .United States v. Lopez, 514 U.S. 549, 567–68 (1995).
  251. .We say “arguable” because, to the extent that DACA and DAPA sought to centralize the discretionary judgment about whom to deport in the White House or Main Justice, the defeat of those policies might simply return us to a regime of more decentralized discretionary judgments. Those judgments would not belong to the states—they would be made by federal agency officials—but they might not result in any sort of centralized policy.
  252. .549 U.S. 497, 505 n.2 (2007) (listing California, Connecticut, Illinois, Maine, Massachusetts, New Jersey, New Mexico, New York, Oregon, Rhode Island, Vermont, and Washington).
  253. .Id. at 505 n.5 (listing Alaska, Idaho, Kansas, Michigan, Nebraska, North Dakota, Ohio, South Dakota, Texas, and Utah).
  254. .See Juliet Eilperin, NYU Law Launches New Center to Help State AGs Fight Environmental Rollbacks, Wash. Post (Aug. 16, 2017), []. Blue state AGs announced that they would sue to block President Trump’s rollback of President Obama’s “clean power plan” long before the new plan was unveiled in the summer of 2018. See Press Release, David J. Hayes, Exec. Dir., NYU State Energy & Envtl. Impact Ctr., State Attorneys General Ready to Sue EPA Over Clean Power Plan Repeal (Nov. 28, 2017), []; see also Press Release, Office of Attorney Gen. Maura Healey, AG Healey Leads Statement From 20 State Attorneys General Announcing Intent to Sue Over EPA Rollback of Clean Car Rule (Aug. 2, 2018), [].
  255. .See, e.g., Engine Mfrs. Ass’n v. Coast Air Quality Mgmt. Dist., 541 U.S. 246, 247, 257–58 (2004) (accepting industry argument, supported by the United States as amicus, that California’s rules requiring fleet operators of vehicles to purchase low-emissions vehicles were subject to preemption by more permissive federal standards).
  256. .Horizontal conflicts may also involve wealth transfers from one part of the country to another. Southerners objected to the national tariff in the nineteenth century on the ground that it protected infant industry in the North while resulting in higher prices for imported goods in the South. See, e.g., Daniel Walker Howe, What Hath God Wrought: The Transformation of America, 1815–1848, at 271–73 (2007). Likewise, many have argued that the 2017 national tax overhaul’s limit on the deduction for state and local taxes transfers wealth from blue to red states. E.g., Michael Hiltzik, The Republican Tax Plan is an Arrow Aimed at Blue States like California, L.A. Times (Nov. 3, 2017), []. For that reason, blue states have filed suit to challenge the tax overhaul. See Joseph De Avila, Democratic States Sue Trump Administration Over Tax Overhaul, Wall St. J. (July 17, 2018), But even these fights—ostensibly over money—were actually about far more substantive policy preferences. The tariff promoted one way of life (industrialization) over another (agrarianism), while the tax reform favors a low tax–low services model of state regulation over a high tax–high regulation model.
  257. .See Baker & Young, supra note 246, at 140–42, 152–55.
  258. .See, e.g., David J. Barron, Reclaiming Federalism, Dissent Mag. (Spring 2005), [] (“With all three branches of the national government in conservative hands, progressives have begun to wonder whether federalism might be useful after all.”); Heather K. Gerken, A New Progressive Federalism, Democracy J. (Spring 2012), [] (arguing that federalism “allows racial minorities and dissenters to act as efficacious political actors, just as members of the majority do”). Today’s California has taken up the mantle of resistance to national authority passed down from John C. Calhoun’s South Carolina. See, e.g., Don Thompson & Elliot Spagat, Jeff Sessions, California Governor Clash as Feud Escalates, U.S. News (Mar. 7, 2018), [] (reporting new U.S. suit to preempt California’s immigrant sanctuary laws). And the next great vertical federalism conflict may well take place over blue states’ efforts to legalize marijuana. See Sadie Gurman, Sessions Terminates US Policy that Let Legal Pot Flourish, AP News (Jan. 4, 2018), [] (reporting how Jeff Sessions’s lifting of Obama-era policy “now leave[s] it up to federal prosecutors to decide what to do when state rules collide with federal drug law”).
  259. .To be clear, we do not mean to say that interests in particular regulatory policies are inherently short-term. The blue states’ suit to force national limits on greenhouse gases in Massachusetts v. EPA, for instance, asserted a very long-term interest. And certainly, constitutional arguments are long term in their consequences if adopted. The more short-term factor is the litigating states’ expectations concerning the relative propensity of either the national or state governments to promote their favored policies at any given political moment.
  260. .545 U.S. 1 (2005).
  261. .Brief of the States of Alabama, Louisiana, and Mississippi as Amici Curiae in Support of Respondents, Gonzales v. Raich, 545 U.S. 1 (2005) (No. 03-1454), 2004 WL 2336486, at *1–3.
  262. .See, e.g., Neil S. Siegel, Free Riding on Benevolence: Collective Action Federalism and the Minimum Coverage Provision, 75 Law. & Contemp. Probs. 29, 33, 44, 46–47 (2012) (explaining the free-rider problem in the interstate healthcare market due to multistate insurance operations and cross-state hospital use); David L. Shapiro, Federalism: A Dialogue 35–50 (1995) (summarizing the traditional economic justifications for national authority).
  263. .See, e.g., Texas v. United States, 809 F.3d 134, 150 (5th Cir. 2015), aff’d per curiam by an equally divided court, 136 S. Ct. 2271 (2016) (summarizing the States’ claims that the President’s immigration policy violated both separation of powers and the APA); Marian Johns, 14-State Coalition Challenges Consumer Financial Protection Bureau’s Constitutionality, Legal NewsLine (Aug. 6, 2018), [] (describing claim by Texas and thirteen other states that the CFPB violates separation of powers principles requiring that executive officers not be unduly insulated from accountability to the President).
  264. .See generally Clark, supra note 88, at 1339–41 (explaining how enforcing the rules of the federal lawmaking process safeguards federalism).
  265. .See, e.g., Abram Chayes & Antonia Handler Chayes, The New Sovereignty: Compliance with International Regulatory Agreements 26–29 (1995).
  266. .See, e.g., Robert D. Cooter & Neil S. Siegel, Collective Action Federalism: A General Theory of Article I, Section 8, 63 Stan. L. Rev. 115 (2010) (arguing that the purpose of federal power generally is to solve collective action problems).
  267. .See, e.g., Young, State Sovereign Immunity, supra note 171, at 51–58; Ernest A. Young, The Rehnquist Court’s Two Federalisms, 83 Texas L. Rev. 1, 63–65, 112–15, 121 (2004).
  268. . Bd. of Trustees of the Univ. of Ala. v. Garrett, 531 U.S. 356, 360 (2001); Kimel v. Fla. Bd. of Regents, 528 U.S. 62, 91 (2000). States remain subject to federal requirements, however, and plaintiffs can often secure injunctive relief against them and (sometimes) damages from the responsible state officers. Mitchell N. Berman, R. Anthony Reese & Ernest A. Young, State Accountability for Violations of Intellectual Property Rights: How to “Fix” Florida Prepaid (And How Not To), 79 Texas L. Rev. 1037, 1095 (2001).
  269. .See Bulman-Pozen & Gerken, supra note 90, at 1310.
  270. .The § 1983 cases are probably more important than the state sovereign immunity decisions in this regard. Where the state asserts sovereign immunity, individual officers—the “uncooperative federalists” celebrated by Professors Bulman-Pozen and Gerken—may still be liable for money damages. But cases like Gonzaga University v. Doe leave state officials who are “uncooperatively” administering federal spending power regimes subject only to the cutoff of federal funds by the responsible federal agency. See 536 U.S. 273, 279, 283, 286–89 (2003) (rejecting private enforcement under § 1983 of a federal conditional spending statute). That remedy involves considerable political costs and, as a result, is rarely attempted in practice. See Rosado v. Wyman, 399 U.S. 397, 426 (1970) (characterizing the cutoff of federal funds as a “drastic sanction”).
  271. .See Nolette, supra note 13, at 200–01 (arguing that state AGs opposing state policymaking autonomy have departed from their traditional role).
  272. .570 U.S. 744, 766–68 (2013).
  273. .Obergefell v. Hodges, 135 S. Ct. 2584, 2608 (2015).
  274. .See, e.g., Brief of the Commonwealth of Virginia, supra note 10, at 33 (“[I]t is indisputable that whenever such conflicts arise [between federalism and individual rights], the Fourteenth Amendment trumps federalism” and suggesting that federalism only matters if it pushes in the same direction as individual rights claims); Windsor Pro-DOMA States’ Brief, supra note 151, at 7–8 (questioning whether “federalism ha[s] any residual connection to the equal protection standard applicable to the federal government” and objecting “to the idea of leveraging individual rights claims using the Constitution’s structural safeguards”).
  275. .See supra note 10 for briefs filed by state amici. Party affiliations of state AGs can be found at [].
  276. .See Nat’l Conf. St. Legislatures, supra note 74 (showing party control of state governments).
  277. .Id. In some of those states, the Republican party controlled both the governor’s seat and part of the legislature.
  278. .Same-Sex Marriage, State by State, Pew Res. Ctr. (June 26, 2015) [] (showing same-sex marriage laws over time).
  279. .Lemos & Quinn, supra note 143, at 1258. The briefing patterns in District of Columbia v. Heller—concerning the individual right to bear arms under the Second Amendment—are perhaps even more interesting in this regard. The states filed warring amicus briefs in Heller, and the anti-gun brief was signed by Democratic AGs only. Yet Democrats also accounted for fifteen of the thirty-one AGs who signed the pro-gun brief—“arguing not only against the typical Democratic position on guns, but also against state power.” The likely explanation is that “virtually all of the AGs on the pro-gun brief hailed from states in the West, Midwest, and South—where support for gun rights typically is strongest.” See id. at 1263–64. Similarly, the AGs who signed pro-gun amicus briefs in McDonald, arguing that the Second Amendment right to bear arms should be incorporated against the states, represented states that already guaranteed “an ‘individual’ right to keep and bear arms in their own constitutions, often in terms more expansive than those in the Second Amendment.” Joseph Blocher, Popular Constitutionalism and State Attorneys General, 122 Harv. L. Rev. F. 108, 111 (2011).
  280. .Cf. Ernest A. Young & Erin C. Blondel, Federalism, Liberty, and Equality in United States v. Windsor, 2012–2013 Cato Sup. Ct. Rev., at 117, 133–37 (2013) (discussing this aspect of the same-sex marriage debate).
  281. .See Michael E. Solimine, State Amici, Collective Action, and the Development of Federalism Doctrine, 46 Ga. L. Rev. 355, 366 (2012) (highlighting “those instances where a large number of SAGs file amicus briefs, often jointly, that take a position against the presumed state interest in a federalism dispute and when the Justices appear to take special note of that incongruence when rendering that decision”).
  282. .See generally Francis, supra note 91, at 1048–51 (arguing that states benefit from voicing their views on federal law through litigation); cf. Solimine, supra note 279, at 375–76 (offering a normative defense, grounded in a variation on political-safeguards theory, of justices’ apparent reliance on state amicus briefs).
  283. .See supra subpart I(C); see also Johns, supra note 261 (noting separation of powers challenge to the Consumer Financial Protection Bureau).
  284. .Empirical research on the law and politics divide has yet to come up with any good way to separate “legal” views about the content of the law from “political” or “partisan” views about how cases should come out. See generally Young, Blowing Smoke, supra note 161, at 8–17. Our friends Scot Powe and H.W. Perry have demonstrated, moreover, that each party has a relatively coherent set of views about the content of the law—the reach of the Commerce Clause, say, or the extent of the President’s unilateral authority—that correlate strongly to party but nonetheless represent coherent legal positions. See H.W. Perry, Jr. & L.A. Powe, Jr., The Political Battle for the Constitution, 21 Const. Commentaries 641, 645, 695 (2004). We are willing to describe a legal view that correlates strongly to party as “partisan” in an important sense, with the caveat that the term should not be pejorative in that context and is not necessarily an antonym to “legal.”
  285. .See Young, Dark Side, supra note 79, at 1308–10.
  286. .We say “typically” because state AGs sometimes litigate alongside federal agencies as well. See, e.g., Nolette, supra note 13, at 51–55 (describing coordinated litigation by state and federal agencies targeting pharmaceutical pricing practices); Philip A. Lehman, N.C. Dep’t of Justice, National Mortgage Settlement: Executive Summary of Multistate/Federal Settlement of Foreclosure Misconduct Claims 1, [] (describing joint state-federal settlement involving “robo signing” practices by mortgage banks).
  287. .E.g., Nolette, supra note 13, at 203–04 (“Rather than relying on typical policymaking processes such as legislation or rule making, the AGs use the tools of adversarial legalism to influence policy.”); Margaret A. Little, Pirates at the Parchment Gates: How State Attorneys General Violate the Constitution and Shower Billions on Trial Lawyers, Competitive Enter. Inst. Issue Analysis, 2017, at 3 (“These lawsuits violate the Constitution’s separation of powers, particularly the assignment of lawmaking, taxing, and expenditure powers to the legislature.”); supra note 11 and accompanying text.
  288. .Kagan, supra note 17.
  289. .Abram Chayes, The Role of the Judge in Public Law Litigation, 89 Harv. L. Rev. 1281, 1284 (1976).
  290. .Fallon et al., supra note 100, at 3–76; see also Owen M. Fiss, The Supreme Court, 1978 Term—Foreword: The Forms of Justice, 93 Harv. L. Rev. 1, 14 (1979) (“The task of a judge, then, should be seen as giving meaning to our public values and adjudication as the process through which that meaning is revealed or elaborated.”).
  291. .See, e.g., Sean Farhang, The Litigation State: Public Regulation and Private Lawsuits in the U.S. 233 (2010) (noting that “[a]mong the most frequent criticisms” of the “large role of courts and lawsuits in American policy implementation” is “that it is deeply undemocratic, unsuited to a political community committed to representative democracy and legislative supremacy”); Note, Implementation Problems in Institutional Reform Litigation, 91 Harv. L. Rev. 428, 428 (1977) (observing that institutional reform litigation places judges “in a new role: they become responsible for implementing broad reforms in complex administrative systems, without ordinarily having expertise in either public administration or the particular institutional field in question”).
  292. .Chayes, supra note 287, at 1302.
  293. .See Nolette, supra note 13, at 22–24, 45–53.
  294. .See, e.g., Robert F. Nagel, Separation of Powers and the Scope of Federal Equitable Remedies, 30 Stan. L. Rev. 661, 686, 707–12 (1978). Moreover, to the extent that state AGs can use the proceeds of prior litigation to fund their ongoing activities, they may be able to set their own agendas without the same level of supervision provided by the typical appropriations process. See supra note 126 and accompanying text (discussing “revolving-fund” statutes, which permit some state AGs to retain certain litigation proceeds); Lemos & Minzner, supra note 125, at 873–74 (suggesting how self-funding mechanisms for public enforcers might interact with the appropriation process).
  295. .Kagan, supra note 17, at 4; see also id. at 198–206 (emphasizing the expense, inefficiencies, and uncertainty resulting from a regulatory system that has litigation at its center). We take “adversarial legalism” to be a broader category than “public law litigation.” As Sean Farhang has pointed out, “[t]he vast bulk of private litigation enforcing federal statutes (well over 90 percent) is neither a story of impact litigation by interest groups seeking to make policy, nor of suits challenging the policymaking prerogatives of national authorities.” Farhang, supra note 289, at 11.
  296. .Kagan, supra note 17, at 4.
  297. .See Legal Services Corporation, Supplementary Information Regarding the Final Rule on Class Action Participation, 61 Fed. Reg. 63,754 (Dec. 2, 1996) (to be codified at 45 C.F.R. pt. 1617) (“The legislative history of this provision indicates an intent that legal services programs should focus their resources on representation of individual poor clients and not be involved in any class actions.”).
  298. .See, e.g., Derrick A. Bell, Jr., Serving Two Masters: Integration Ideals and Client Interests in School Desegregation Litigation, 85 Yale L.J. 470, 471, 490–91, 493 (1976) (discussing the potential conflict in civil rights cases between the ideological goals of class action lawyers and the best interests of their clients).
  299. .See, e.g., Lon L. Fuller, The Forms and Limits of Adjudication, 92 Harv. L. Rev. 353, 394–95, 400–05 (1978) (arguing that “polycentric” problems are best solved by legislators).
  300. .Kagan, supra note 17, at 229.
  301. .Farhang, supra note 289, at 216–17.
  302. .See, e.g., Ann Woolhandler & Michael G. Collins, State Standing, 81 Va. L. Rev. 387, 406–07 (1995).
  303. .549 U.S. 497 (2007).
  304. .Id. at 519. It is hard to tell whether Massachusetts could have established standing based on this ownership interest alone. We think, however, that the real difficulty with Massachusetts’s claim for standing involved the causation elements of standing. As we discuss further infra, the causes of climate change are so multifarious, and the likelihood that any given regulatory change would redress it are so murky, that “special solicitude” for the Commonwealth’s state-ness may have been necessary to get it over the hump. But that goes to causation, not to whether the ownership interest was sufficient to support the requisite “injury in fact.”
  305. .See, e.g., Arlington Cent. Sch. Dist. Bd. of Educ. v. Murphy, 548 U.S. 291, 296 (2006) (likening spending-power legislation to contracts between the federal and state governments); Bowen v. Massachusetts, 487 U.S. 879 (1988) (entertaining a state’s challenge to administration of a federal grant-in-aid program).
  306. .See, e.g., Alfred L. Snapp & Son, Inc. v. Puerto Rico ex rel. Barez, 458 U.S. 592, 601–02 (1982) (outlining the proprietary, sovereign, and quasi-sovereign interests of states).
  307. .Massachusetts, 549 U.S. at 519.
  308. .See, e.g., Complaint at 42, Florida v. Am. Tobacco Co., No. CL 95-1466 AH, 1996 WL 788371 (Fla. 15th Cir. Ct.), [].
  309. .See Maine v. Taylor, 477 U.S. 131, 137 (1986) (holding that “a State clearly has a legitimate interest in the continued enforceability of its own statutes”); see also Kathryn A. Watts & Amy J. Wildermuth, Massachusetts v. EPA: Breaking New Ground on Issues Other than Global Warming, 102 Nw. U. L. Rev. 1029, 1035 (2008) (noting that “the state . . . has a sovereign interest in preserving its own law” that “should be sufficient for Article III purposes”).
  310. .The Supreme Court has held, for example, that a private organization has Article III injury in fact when a defendant’s practices impair the organization’s ability to provide services to the population it serves. See Havens Realty Corp. v. Coleman, 455 U.S. 363, 379 (1982).
  311. .Brief for the State Respondents at 19, United States v. Texas, 136 S. Ct. 2271 (2016) (No. 15-674).
  312. .See Brief of Amici Curiae Federal Courts Scholars and Southeastern Legal Foundation in Support of Respondents at 7–9, United States v. Texas, 136 S. Ct. 2271 (2016) (No. 15-674). Mr. Young was counsel of record and primary author on this brief.
  313. .See Massachusetts v. U.S. Dep’t of Health & Human Servs., 301 F. Supp. 3d 248, 255–56 (D. Mass. 2018). The district court rejected this interest as insufficiently certain to support Article III standing, see id. at 258–65, and appeal is pending in the First Circuit as this Article goes to press. One of us has filed an amicus brief in support of the Commonwealth’s standing to sue, while remaining agnostic on any issues on the merits. See Brief of Professor Ernest A. Young as Amicus Curiae in Support of Plaintiff–Appellant Urging Reversal, No. 18-1514, Massachusetts v. U.S. Dep’t of Health & Human Servs. (filed Sept. 24, 2018).
  314. .Alfred L. Snapp & Son, Inc. v. Puerto Rico ex rel. Barez, 458 U.S. 592, 602 (1982).
  315. .Id. at 601.
  316. .Dissenting in Massachusetts v. EPA, for example, Chief Justice Roberts wrote that “[j]ust as an association suing on behalf of its members must show not only that it represents the members but that at least one satisfies Article III requirements, so too a State asserting quasi-sovereign interests as parens patriae must still show that its citizens satisfy Article III.” 549 U.S. 497, 538 (2007) (Roberts, C.J., dissenting). Roberts was objecting to the notion that a state’s unique character “dilutes the bedrock requirement of showing injury, causation, and redressability to satisfy Article III.” Id. But he offered no reason why a state that could meet those requirements should have less right to represent its citizens than an association has to represent its members.
  317. .Hunt v. Wash. State Apple Advert. Comm’n, 432 U.S. 333, 343 (1977).
  318. .Compare Who We Are, Sierra Club, [] (estimating that Sierra Club includes 3.5 million members and supporters), with Wikipedia, List States and Territories of the United States by Population, [] (reporting Connecticut’s population as 3,588,184 and Iowa’s as 3,145,711).
  319. .Compare Social Impact, Am. Ass’n of Retired Persons, [] (claiming “nearly 38 million members”), with Wikipedia, List States and Territories of the United States by Population, supra note 316 (reporting California’s estimated 2017 population as 39,536,653 persons).
  320. .See, e.g., Warth v. Seldin, 422 U.S. 490, 511 (1975) (establishing that an “association must allege that its members, or any one of them, are suffering immediate or threatened injury as a result of the challenged action of the sort that would make out a justifiable case had the members themselves brought suit”).
  321. .Alfred L. Snapp & Son, Inc. v. Puerto Rico ex rel. Barez, 458 U.S. 592, 607 (1982). That requirement serves to differentiate the state’s interest from “the interests of particular private parties,” and to ensure that the state is “more than a nominal party.” Id.
  322. .See Lemos, State Enforcement, supra note 110, at 495 & n.37 (describing courts’ treatment of the requirement).
  323. .Id. at 499 (emphasizing similarities between state litigation and damages class actions and citing sources).
  324. .See generally Margaret H. Lemos, Three Models of Adjudicative Representation, 165 U. Pa. L. Rev. 1743, 1757–63 (2017) (comparing public suits and injunctive class actions of the sort that public-interest groups often spearhead).
  325. .Lemos, State Enforcement, supra note 110, at 499.
  326. .In Tennessee, the state AG is appointed by the Supreme Court, considered an officer of the judicial branch, and serves an eight-year term. See Tenn. Const. Art. VI, § 5. This arrangement appears to be unique. Attorney General of Tennessee, Ballotpedia, [].
  327. .See generally Margaret H. Lemos, Democratic Enforcement? Accountability and Independence for the Litigation State, 102 Cornell L. Rev. 929 (2017) (discussing some weaknesses of public accountability mechanisms).
  328. .NAAG, supra note 95, at 45 (explaining that the AG is the “principal legal representative of the public interest for all citizens”).
  329. .In many instances, those responsibilities will constrict the opportunities for partisanship, or dampen its effects. For example, aside from occasional high-profile exceptions, AGs typically defend state legislation against constitutional attack, even if the legislation in question was the handiwork of an opposite-party legislature and runs counter to the AGs’ own policy preferences. Similarly, Democratic AGs defend criminal convictions; Republican AGs defend civil rights or environmental judgments—and so on. There may be cases to the contrary, and we do not know (and do not purport to suggest) that Democrats and Republicans handle the day-to-day demands of the job in precisely the same way. Nevertheless, there are likely to be large swaths of the job that lack any particularly sharp partisan valence, and where the tensions between “states’ interests” and partisan interests is relatively easy to resolve.
  330. .See, e.g., Brownstein, supra note 18, at 327–38 (2007) (recounting the rise of “netroots” organizations that raised large sums of money for Democrats and used their influence to push party politicians to the Left).
  331. .See, e.g., Hopkins, supra note 19, at 38–45 (2017) (discussing the effects of first-past-the-post rules).
  332. .We see some indications that AG elections involve different political dynamics from other statewide offices. The fact that five of the last six governors of Massachusetts have been Republicans suggests that state government races are competitive despite the State’s all-Democrat congressional delegation. But we are told that in fact, races for AG are not competitive, and the record seems to bear this out: The last Republican AG of Massachusetts was Elliott Richardson, who left the post in 1969. See Wikipedia, Massachusetts Attorney General, [] (showing political affiliation of Massachusetts AGs dating back to 1702). Hence, current AG Maura Healy can feel comfortable filing nearly a dozen lawsuits against the Trump Administration in 2017 alone notwithstanding her Republican governor’s 71% approval rating. David S. Bernstein, Maura Healey’s Trump Card, Bost. Mag. (Jan. 30, 2018), [] (estimating that Massachusetts AG Healy filed roughly fifteen lawsuits against the Trump Administration in 2017). Why AG politics is so different from gubernatorial politics is a mystery to us, but that mystery is outside the scope of this paper.
  333. .See, e.g., Gerken, Dissenting by Deciding, supra note 79, at 1783 (noting “federalism can be understood at least in part as a strategy for allowing would-be dissenters to govern in some subpart of a system”); Young, Dark Side, supra note 79, at 1286 (noting “the party that is ‘out’ in Washington will almost certainly be ‘in’ in at least a couple of dozen states and literally thousands of localities”).
  334. .See, e.g., Heather K. Gerken, Foreword: Federalism All the Way Down, 124 Harv. L. Rev. 4, 21–25 (2010) (arguing that federalism should encompass cities and local institutions).
  335. .See generally Lemos, State Enforcement, supra note 110, at 489, 512–13, 546 (developing these points and arguing that citizens should therefore not be bound by the judgments in representative state actions).
  336. .Fed. R. Civ. P. 23(g)(1)(B), (g)(4).
  337. .Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 625 (1997).
  338. .Fed. R. Civ. P. 23(c)(5).
  339. .Charles Alan Wright, Arthur R. Miller, Mary Kay Kane, Richard L. Marcus, A. Benjamin Spencer & Adam N. Steinman, 7AA Federal Practice and Procedure § 1790 (3d ed. Sept. 2018 update).
  340. .In damages class actions, the solution (in theory, at least) is to opt out. Thus, the problem is most stark in “mandatory” injunctive class actions.
  341. .David Marcus, Flawed But Noble: Desegregation Litigation and its Implications for the Modern Class Action, 63 Fla. L. Rev. 657, 690 (2011).
  342. .William B. Rubenstein, Divided We Litigate: Addressing Disputes Among Group Members and Lawyers in Civil Rights Campaigns, 106 Yale L.J. 1623, 1652–53, 1659 (1997); see also Deborah L. Rhode, Class Conflicts in Class Actions, 34 Stan. L. Rev. 1183, 1185 (1982) (arguing for “full disclosure of, although not necessarily deference to, class sentiment”).
  343. .Francis, supra note 91, at 1026, 1040–41.
  344. .See Wechsler, supra note 87, at 543–44, 546.
  345. .See Bulman-Pozen & Gerken, supra note 90, at 1286.
  346. .See Francis, supra note 91, at 1048.
  347. .See supra section II(B)(1).
  348. .See, e.g., Heather K. Gerken, Our Federalism(s), 53 Wm. & Mary L. Rev. 1549, 1557 (2012) (“States do not rule separate and apart from the system. . . . [T]hey serve as part of a complex amalgam of national, state, and local actors implementing federal policy.”); Ernest A. Young, “The Ordinary Diet of the Law”: The Presumption Against Preemption in the Roberts Court, Sup. Ct. Rev., 2011 at 253, 257–63 (tracing the change from dual federalism to an integrated system of concurrent jurisdiction).
  349. .Francis, supra note 91, at 1033.
  350. .See id. at 1044–45.
  351. .See id. at 1051–54; see also New York v. United States, 505 U.S. 144, 169 (1992) (emphasizing the importance of clear lines of political accountability in federal systems).
  352. .See, e.g., Lynn A. Baker, Putting the Safeguards Back into the Political Safeguards of Federalism, 46 Vill. L. Rev. 951, 958–60 (2001) (contending that representation in Congress does not protect states as institutions from federal aggrandizement); Saikrishna B. Prakash & John C. Yoo, The Puzzling Persistence of Process-Based Federalism Theories, 79 Texas L. Rev. 1459, 1477–78 (2001) (observing that temporary interests may prompt state representatives in Congress to “sacrifice [states’] rights as institutions”).
  353. .See Francis, supra note 91, at 1040–41.
  354. .See Albert O. Hirschman, Exit, Voice, and Loyalty: Responses to Decline in Firms, Organizations, and States 4–5, 30 (1970) (defining “exit” and “voice” as alternative courses of action when the quality of a regime declines).
  355. .See John D. Nugent, Safeguarding Federalism: How States Protect Their Interests in National Policymaking 116–17 (2009) (highlighting the intergovernmental lobby); Bulman-Pozen & Gerken, supra note 90, at 1255–56 (focusing on bureaucrats); Kramer, supra note 87, at 219 (prioritizing political parties); Wechsler, supra note 87, at 543–44 (emphasizing representation in Congress).
  356. .See Lemos, Privatizing Public Litigation, supra note 123, at 572–78 (discussing government advantages in litigation).
  357. .Cogan Schneier, After Defeat in New York, State AGs Are Next to Test Emoluments Challenge, Nat’l L.J. (Jan. 24, 2018), [] (quoting James Tierney, former AG of Maine).
  358. .Complaint at 6–32, District of Columbia v. Trump, No. 8:17-cv-01596-PJM (D. Md. June  12, 2017).
  359. .See, e.g., Stephanie Ebbert, Healey Wins Showdown with Exxon Mobil, Boston Globe (Jan. 11, 2017), [] (describing Massachusetts AG’s investigation of Exxon Mobil, which will compel Exxon to turn over “40 years of documents” on the company’s research on global warming).
  360. .See, e.g., Nolette, supra note 13, at 58–64 (describing how AG litigation shaped public opinion and changed the political climate on pharmaceutical pricing); Sebok, supra note 117, at 2177–79 (describing shifts in public opinion on smoking after the multistate suit); see also Rachel M. Cohen, The Hour of the Attorneys General, Am. Prospect (Mar. 22, 2017), [] (“When a state files a lawsuit, it invokes a special sort of gravitas that private entities don’t have. And when ten, or fifteen, or twenty states join together to sue a corporation or the federal government, it sends a powerful message—something AGs rarely overlook.”).
  361. .Charles Elmore, Lobbying Group Picks Up Costs of Florida’s Health-Care Legal Challenge, Palm Beach Post (Feb. 19, 2011), [].
  362. .Texas v. United States, 86 F. Supp. 3d 591, 617, 635 (S.D. Tex. 2015).
  363. .262 U.S. 447, 485–86 (1923). Mellon seems flatly inconsistent with Massachusetts v. EPA, 549 U.S. 497, 520–21 (2007), which allowed states to sue parens patriae to assert their citizens’ rights under the Clean Air Act. But that question is beyond the scope of our discussion here.
  364. .Anne Blythe, GOP Lawmakers Target Democrat Josh Stein with Surprise Budget Cuts, News & Observer (June 21, 2017), [].
  365. .See Act of June 28, 2017, ch. 57, sec. 6.7(l), § 120.32.6(b), 2017-3 N.C. Adv. Legis. Serv. 1, 19 (LexisNexis) (vesting legislative leaders with “final decision-making authority” over the litigation of cases in which the constitutionality of state law is challenged).
  366. .See Lemos, Democratic Enforcement, supra note 325, at 983–84 (describing arrangements in some states in which specialized agencies control certain categories of litigation).
  367. . Justice Thomas, for example, has recently called for limiting nationwide injunctions. See Trump v. Hawaii, 138 S. Ct. 2392, 2424–29 (2018) (Thomas, J., concurring). Such injunctions feature in many state lawsuits, see, e.g., id., but are not unique to them.
  368. .See Johnstone, supra note 47, at 609 (worrying that state AGs “cannot be part of the solution to national partisan polarization . . . if those forces of polarization extend to the state level”); Elbert Lin, States Suing the Federal Government: Protecting Liberty or Playing Politics?, 52 U. Rich. L. Rev. 633, 651 (2018) (worrying that the increasing frequency and polarized nature of state litigation risks “cheapening the brand” of the states as litigants).