State Bureaucratic Undermining

Introduction

Turbulence rocks the federal government, and it is now faddish to romanticize states as sites of resistance. California “strikes a bold pose as vanguard of the resistance.”1 New York “emerg[es] as the East . . . Coast headquarters of opposition to the incoming administration.”2 Federalism scholars tell us not to worry about decentralizing power from the federal government to states; states are more trustworthy now. We no longer live in “your father’s federalism.”3

Not so fast. States have long undermined federal civil rights,4 and they continue to do so in new ways. It is now politically expedient for politicians in New York and California to resist the conservative federal government with progressive policies and speeches. But even as they do, these same states fight federal civil rights laws in court and, through inattention and disorganization, thwart them on the ground.5 We need a more complete account of how federal rights fare—and fail—inside of state bureaucracy.

In fact, many of our federal rights administered by states are failing.6 These are important rights and high stakes, and the workings of state government provide some explanations. This Article describes how states undermine federal rights not from hostility, but because they are plural bodies that lack coordination. States diffuse their powers. They do so horizontally (across cabinet-level state actors, agencies, and commissions) and vertically (down to local governments like counties, towns, and special interest districts).7 State coordination problems are akin to a friction; intentional or not, they slow and stop federal law. Because this friction arises from large administrative structures, it appears in places you might not expect: large states without a history of intentionally resisting federal civil rights laws.8

Few federal rights acknowledge or accommodate state decentralization. They place obligations on “states” and do not specify how a state must administer the right. Here is an example. The National Voter Registration Act9 (NVRA) requires that “each State shall” provide voter registration opportunities in certain state offices: motor vehicle offices, public assistance offices, disability services offices, and military recruitment offices.10

“[E]ach State shall” is a deceptively simple command. Who, or what, is the “state”? The state secretary of state may be the state’s chief election official and technically responsible for the state’s compliance with the NVRA,11 but secretaries of state lack authority over motor vehicle offices, public assistance offices, disability services offices, and military recruitment offices—some of which are administered locally. Nor do secretaries of state control the state budget, necessary for vindicating a federal requirement that itself provides no funding for compliance. Statewide compliance with the NVRA, therefore, requires coordination across the state bureaucracy.

The NVRA is not alone; other federal rights—including pillars of our federal civil rights laws—require complex intrastate coordination to achieve statewide compliance. The Sixth Amendment right to counsel,12 for example, requires state legislatures to fund and design indigent defense programs, state executive officers to oversee those programs, and—because many states use local governments to administer their criminal justice systems—local offices to actually provide counsel. Cooperative-federalism programs that partner states with the federal government to administer programs like food stamps, cash assistance, disability assistance, and others also require intrastate coordination.13

State bureaucracy provides a natural resistance to this kind of coordination. Coordination requires cooperation and consent, or clear chains of legal command to resolve conflict.14 As Professors Jeffrey Pressman and Aaron Wildavsky put it in their foundational work on policy implementation, the “multiplicity of participants and perspectives,” even when everyone supports the policy outcome, creates an administrative “obstacle course.”15

In addition, states are now at their most complex and sheltered from federal intrusion. States have expanded and professionalized their bureaucracies.16 In doing so, they gained the faith of the federal government, which entrusted them to administer federal policies, growing them even further.17 As states took on a more prominent role in federal policy administration, the Supreme Court protected state sovereignty from federal intrusion in new ways.18

These trends now collide in our federal system: state bureaucracy is larger than ever before; states administer more federal laws and rights than ever before, and those rights sprawl across state bureaucracies in complex ways; and state control over its bureaucracy is more insulated from the federal government than ever before.

This Article takes that collision as its starting point. It describes the topography of state bureaucratic undermining, or the relationships between state actors that federal rights demand: interagency, interbranch, and state-local. It provides a typology that categorizes the kinds of bureaucratic roadblocks that arise: alienation between state actors, conflict between state actors, and role confusion.

Although these descriptions may not be flashy, they are both long overdue and crucial to understanding the reach of federal power and the nature of inequality. Despite Professor Owen Fiss’s declaration in 1979 that state bureaucracy was “a new unit of constitutional law,”19 state bureaucracy remains unforgivably understudied.20 What the literature lacks is a systematic, realistic discussion of states as plural bodies, including the kinds of state coordination that federal law demands and how those demands both create noncompliance with federal rights and shape administration of those rights.21

I also intend this Article to provide a necessary counterpart to recent federalism scholarship that has provided novel accounts of benefits of federal policy decentralization but has paid little attention to its contemporary costs. That scholarship has described complex pathways between the federal government and states.22 It has discovered that federal-state cooperation and conflict extend far beyond the formal markers of statutory and constitutional law into administrative realms23 and even more informal interactions.

This scholarship takes as a premise that, on the whole, state and local governments administer federal policies in ways that do not systematically reinforce or compound inequality and racial stratification. Or, if they do, the federal government can easily detect and correct that bias.24 Perhaps for this reason, this scholarship tends not to closely examine structural racism and inequality.25 When it examines state conflict with the federal government, it looks at ways states shape existing federal law through their own lawmaking, rather than examining patterns of actual state noncompliance with federal rights. This lowers the stakes of state-federal conflict: it’s easier to celebrate that conflict when the downside is state action that is still lawful.

To the extent contemporary federalism scholarship has considered in depth the drawbacks of decentralization caused by state misbehavior, it has worried about “local racists”26 and state “resistance and rebellion.”27 These are worthy fears!—indeed, local racism and state rebellion are both thriving and particularly visible in today’s climate28 —but incomplete ones. The bureaucratic undermining described in this Article does not depend on individual state actors to be racist, nor does it depend on those actors to engage in rebellion or resistance. It does not require state actors to intend conflict with the federal government at all.

This Article proceeds in four parts. Part I provides a brief history of state bureaucracy, beginning with the difficulties that states faced in responding to the Great Depression. It describes how states grew their bureaucracies, both as administrators of federal laws born during the New Deal (and beyond) and as an effort to professionalize and regain the trust of the American people.

Part II dives into state bureaucracies and describes the bureaucratic barriers that block full administration of federal law. It draws from contemporary litigation against state actors to examine how noncompliance with federal laws arises.

Part III provides a political economy account of state bureaucratic undermining and discusses the effects of bureaucratic undermining on racial minorities and low-income populations. It argues that although state coordination problems may not reflect the same kind of politics as intentional resistance to federal law, they are nonetheless political problems. Because any successful state coordination requires political will, federal rights that require extensive state bureaucratic coordination can fail to reach politically marginalized populations that lack political power.

Part IV surveys remedies reached in institutional suits against states to understand how parties and courts address and prevent bureaucratic undermining. These remedies—which this Article calls “coordination remedies”—take the form of choreography. They differ from traditional civil rights injunctions,29 which essentially legislate substantive compliance with federal laws in considerable detail. Coordination remedies, on the other hand, take functional government as their goal. They create channels through state government by fostering integration between state agencies (to combat agency alienation and conflict) and role clarity (to combat role confusion).30

I. The Fall and Rise of State Bureaucracy

The story of states from the 1930s to the present is one of growth and professionalization. Unable to deal with the economic consequences of the Great Depression without significant federal help and discredited by the desegregation and civil rights battles of the 1950s and 1960s, states had lost the trust of the American people and the federal government. In more recent years, they have sought to regain that confidence by growing their bureaucracies and professionalizing their legislatures. That effort has largely succeeded. States now partner with the federal government on a wide range of policy areas. Having shaken the image of racist parochialism, states now enjoy more protection from the judiciary and more open-mindedness from the academy. In short, states are at their largest and strongest.

This Part briefly tells that story. The way that states have developed informs the federalism issues that affect civil rights and constitutional litigation today, and yet it is rarely discussed in the federalism scholarship. The context of this history, however, sheds light on a problem—the coordination problem—that I argue explains and complicates the current place of states in the federal system.

The states emerged bruised from the Great Depression. Terry Sanford, a former governor and US senator from North Carolina, and former president of Duke University, described it in this way:

[I]n the early 1930s, the depression all but submerged the states. The economic collapse was so massive the states could not feed the hungry or find jobs for the unemployed. The states had no means for boosting the economy or saving the banks.

. . .

From the viewpoint of the efficacy of state government, the states lost their confidence, and the people their faith in the states; the news media became cynical, the political scientists became neglectful, and the critics became harsh.31

Compared to the states’ “reluctance and timidity,” the national government took “action.”32

Americans were willing to accept a strong federal executive at that time, even as they had been historically more suspicious of strong state executives.33 “The American governorship was conceived in mistrust and born in a strait jacket, the creature of revolutionary assemblies.”34 During this period of state weakness, governors had limited power to make appointments; they presided over “fragmented, poorly organized” department branches and small staffs with limited education and professional experience.35

At the same time, states—and especially southern states—became associated with segregated schools and civil rights violations. The Supreme Court in the 1950s and ’60s handed down a series of decisions, including Brown v Board of Education of Topeka,36 that required schools to end racial segregation.37 States became “lumped together” in what became known as the “‘Alabama syndrome,’ implying that the conservative, racist administration of Alabama Governor George Wallace in the 1960s [was] the norm.”38

State bureaucracy played a particularly important role in resisting civil rights laws during that era. When Brown I came down, states and state officials in the South placed desegregation into the hands of state and local bureaucracy to give a legitimizing sheen to otherwise fierce efforts to prevent desegregation.39 For example, bureaucratic tactics included pupil reassignment laws enacted at the state level that allowed local bureaucrats to control student school change requests by applying ostensibly race-neutral factors that were, in practice, crafted to preserve segregation.40 To these Southerners, “the path was clear: bureaucratization could accomplish most of what overt resistance had not.”41

In an effort to regain trust, states began to professionalize their governments.42 These reforms sought to strengthen the power of the governor and increase the capacity of the bureaucracy. They included lengthening gubernatorial terms, increasing the governor’s appointment and removal powers, increasing gubernatorial control over the state budget, and providing additional veto powers.43 Reform efforts to streamline authority in state bureaucracy, however, have had mixed results. The number of state actors elected separately from the governor remained essentially unchanged from 1955 to 1994.44 Scholars disagree as to whether these reforms have achieved their goals.45

The sheer size of state bureaucracies has increased as well. In 1954, state governments employed 1.1 million employees; by 2015, that number had increased to 5.4 million.46 Monthly payrolls increased from $300.7 million to $21.6 billion.47 State bureaucrats have also gotten younger, more educated, and more experienced in public service.48

As states reformed, they became more attractive as administrative partners for the federal government.49 In the 1930s and ’40s, the federal government began to rely on states to administer federal laws—in particular, New Deal programs like public assistance laws.50 Since then, the federal government has partnered with the states in a wide range of policy areas: the environment, telecommunications, health care, and so forth.51

State government grew as states accepted financial incentives from the federal government and organized their institutions to administer those programs. Some state agencies, created by states to administer federal programs, “are effectively creatures of federal policy.”52 Federal laws fund and shape these agencies, in part by funding state employee training and directing the policy mission of the bureaucracy.53 As Professor Abbe Gluck has noted, “Cooperative federalism [ ] is a phenomenon that feeds on itself: each federal program that gives money and implementation authority to the states makes those states more reliable, and relied-upon, partners with the federal government.”54 In 1955, states received approximately 17 percent of their revenue from the federal government; that number had increased to approximately 26 percent by 1992.55

By the late 1970s, public opinion had shifted away from the national government toward the states. Americans became disillusioned with the federal government because of several failures at the federal level: the Vietnam War, the costly and perceived-as-ineffective War on Poverty, and the Watergate scandal. The result of these failures was a “crisis of confidence in the national government.”56 In one 1979 survey, only 19 percent of citizens rated the US Congress as “excellent” or “pretty good,” whereas 31 percent gave those same ratings to state legislatures.57

Around that time, the idea of decentralizing federal policy to states became an important part of the political milieu. President Richard Nixon in the 1970s and President Ronald Reagan in the 1980s expanded the autonomy and policymaking authority of states and local governments by converting some categorical grants—federal funding that required states to perform detailed policy tasks—to more open-ended block grants.58 This “New Federalism” rested on “a respectful vision of local government as competent to make intelligent choices about local needs.”59

As states have grown and professionalized, the judiciary has shielded them from federal interference. By the 1990s, “the idea of states’ rights was no longer a shibboleth for segregation,” and conservative justices became concerned that a strong federal government overly undermined state autonomy.60

The Supreme Court has diminished the federal government’s authority to direct state action in at least five ways, resulting in a “rollback of federal legislative power.”61 First, it restricted Congress’s ability to use the Fourteenth Amendment to prevent state-based discrimination. In a series of cases beginning in the 1990s, the Supreme Court struck down provisions of federal civil rights laws—enacted pursuant to Congress’s Fourteenth Amendment enforcement authority62 —that sought to prevent state discrimination on the basis of religion,63 age,64 disability,65 and gender.66 Second, the Court limited Congress’s once nearly plenary authority to enact legislation pursuant to the Commerce Clause.67 Third, the anticommandeering doctrine limited the actions Congress could force states to take to administer federal law.68 Fourth, the Court limited Congress’s ability to coerce states to take action by threatening to remove federal funding.69 Finally, the Court limited the ways that Congress could use federal law to dictate how states organize their internal subdivisions.70

The Supreme Court has also limited the remedies available to plaintiffs to correct civil rights violations by states. During the civil rights era of the 1960s and early ’70s, remedies against states took the form of massive consent decrees and injunctions, hundreds of pages long, that essentially legislated substantive compliance with federal laws in considerable detail.71 These cases sought “total transformation[ ]” and “reconstruction of an ongoing social institution,” and were the first cases to grapple with “a new unit of constitutional law—the state bureaucracy.”72

From the late 1970s through the ’90s, the Supreme Court circumscribed its previous broad approval of these remedies.73 Public law remedies began to take a lighter touch; instead of manually reforming state institutions, they set forth procedures for reaching substantive compliance. These decrees “emphasize broad goals and leave the defendants substantial latitude to determine how to achieve them; mandate precise measurement and reporting with respect to achievement; and institutionalize ongoing mechanisms of reassessment, discipline, and participation.”74

No case better illustrates the broad idea that newly well-behaved states deserve greater protection from the federal government than Shelby County v Holder.75 There, the Supreme Court disabled § 5 of the Voting Rights Act of 1965,76 long one of the most effective and transformative civil rights statutes in America’s history.77 Section 5 required some state and local governments to get preapproval (“preclearance”) for any change to their elections system, from redistricting to changing a polling place.78 Section 5 targeted those jurisdictions with histories of voting discrimination—largely, but not all, southern jurisdictions.79 Although the Court had repeatedly upheld the constitutionality of § 5 and the coverage formula,80 in 2013 it held that “[t]hings have changed dramatically” in the South.81

As the image of states has softened and grown distant from ineptitude, segregation, and intransigent rights violations, academic scholarship on the place of states in the federal system has softened as well. In the 1970s, several influential scholars promoted lawsuits against states and state bureaucracy that forced significant structural reform.82 By the 1980s, however, liberal defenders of structural lawsuits against states had taken on an “anxious tone.”83 These scholars worried about the costs to state sovereignty and the judicial legitimacy of structural injunctions that cut deeply into the workings of state government.84 More recently, federalism scholars on the left have argued that state resistance to the federal government actually creates productive discourse around national policymaking—rather than creating civil rights crises—and ultimately serves nationalist ends.85

States have thus rebounded. State governments are larger than ever, in part because of the federal funding they receive to administer federal programs and in part because of an effort to professionalize state government and regain trust. As a consequence, states also enjoy more legal protections than they have in years.

II. Bureaucratic Barriers

We arrive, therefore, at a high point in state government. Amidst a growing and professionalizing state bureaucracy and renewed respect for state sovereignty, states are now responsible for administering more federal law than ever.86 And yet state bureaucracy, especially as it operates to administer federal law, remains something of a mystery.

This Part examines one aspect of that administration: how state actors must coordinate to comply with federal law.87 Any federal law that uses states as administrators requires states to take action. Federal election laws require states to offer voter registration opportunities at state offices.88 Federal environmental laws require states to achieve federal environmental goals.89 Federal public assistance laws require states to administer welfare and assistance programs.90 And so on.

The forms of these policies vary. Federal statutes include both Spending Clause laws, which offer federal dollars to states for administering federal policy,91 and unfunded federal mandates, which require states to take action on their own without any financial incentive.92 Constitutional mandates on states do not treat states as administrators, exactly, but still require states to take action and therefore implicate intrastate coordination. These mandates include the Sixth Amendment right to counsel and the Eighth Amendment constraints on prison conditions.93

States decentralize their powers, so complying with these laws requires intrastate coordination. I treat coordination broadly in this Article: a federal law implicates intrastate coordination whenever it requires more than one state actor to take action and whenever at least one state actor relies on another to achieve full compliance. The actual substance of coordination can take different forms. It may take the form of coercion: one state actor exercising power over another, especially when the two actors do not share a common purpose.94 But coordination may also be another term for “consent”: a process of bargaining and agreement between two actors that may share common purpose but disagree on means.95

Whether coordination takes the form of coercion or consent depends on the power relationship between the actors.96 The interaction may be hierarchical or nonhierarchical. Hierarchical relationships exist when one state actor has clear state-law authority to coerce another to take the action necessary for compliance, as between a state and its local government, for example. Hierarchical relationships are often not as clear as state law suggests, however.97

Nonhierarchical relationships exist between two actors with an indeterminate power relationship: between two cabinet members, for example, or between a governor and a state commission whose members are appointed by nongubernatorial actors. Two actors in nonhierarchical relationships often answer to different constituencies, and they may be of different political parties or possess conflicting policy and political ambitions.98 And because state law does not provide a clear line of authority between these actors, conflicts between them may be intractable without intervention.

With these baseline concepts in mind, this Part uses examples of conflict and coordination between state actors to set out the topography and typology of state bureaucratic barriers. The topography describes the relationships between state actors that federal rights demand: interagency, interbranch, and state-local. The typology characterizes the kinds of bureaucratic roadblocks that arise—alienation between state actors, conflict between state actors, and role confusion—and explains how these barriers differ from those that arise in the federal bureaucracy.

A. Topography

This Section maps the state bodies—the topographical features of the state terrain—that must coordinate to comply with federal laws. It provides examples of the kinds of laws that require coordination and describes how those laws give rise to conflict—that is, coordination problems—between state actors.

1. Interagency.

States decentralize their power in multiple ways, even at the horizontal, state government level. Unlike the federal government, states largely have not adopted the unitary executive model. In 2017, thirty-five states elected their secretaries of state separately from their governor.99 Forty-four states separately elected their attorney general.100 State government is also populated by commissions, created by executives and legislatures, with authority over state policymaking.101 Because those commissioners can be appointed by a variety of state actors, they may be politically insulated and cause coordination troubles.

Federal rights often require multiple independent executive officers—sometimes of different political parties—to coordinate with each other. Conflict between those officers can create barriers to full compliance with federal law.102 Conflict can arise for many reasons, further described below,103 including legitimate disagreements about policy and political incentives to take adverse positions.

As an example, return to the NVRA, which requires coordination among several state actors. Each state must designate a “chief State election official to be responsible for coordination of State [NVRA] responsibilities.”104 Many states have designated their secretaries of state in that role.105 Actual compliance with the NVRA, however, brings in other state actors by requiring public assistance agencies, disability services offices, and DMV offices to register applicants to vote.106 But secretaries of state do not administer those offices: public assistance offices and state disability services offices are administered by other cabinet-level directors. Those directors are often appointed by governors—forty-one state social services agency directors are107 —but can be appointed by commissions, agency heads, or cabinet directors as well.108 Still another state official administers DMV offices. None of these actors is subject to the authority of the secretary of state.

The NVRA implicates horizontal interagency conflicts at the local level as well. Because California’s elections system is highly decentralized—voter registration happens at the local level—and because local public assistance and DMV offices collect the voter registration forms, those local agencies must coordinate with local registrars’ offices to deliver the forms.109 This kind of coordination may seem mundane—it is—but it is a barrier that has prevented full compliance with the NVRA in California.110

2. Interbranch.

Compliance with some federal rights requires coordination between state executives and legislatures. That is, executives rely on legislative action to comply with these rights. Compliance with the right may require the state legislature to provide funding, or it may require the legislature to amend state law to empower the bureaucracy to administer the right.

Many federal obligations on states cost money. Some—those derived from cooperative-federalism programs like food stamps—come with federal money that states receive for agreeing to implement the program.111 Other rights, however, require states to foot the bill. The NVRA, for example, is a right (and obligation on states) that the federal government does not fully fund.112 The Sixth Amendment right to counsel—and obligation on states to provide counsel—similarly requires states to provide funding.113

State executives’ reliance on state legislative funding sometimes plays out in public. A recent example: the Montana Public Defender Commission wrote an article in the Great Falls Tribune discussing the challenges of maintaining a statewide agency responsible for providing counsel to indigent criminal defendants.114 The Commission, composed of individuals appointed by a mix of state actors—the Montana Supreme Court, the state bar, the senate president, etc.115 —noted the budget challenges it faced116 and stated that it was “committed to working with the governor’s office and the legislature during the upcoming session to assure that the public defender system continues to function efficiently, and admirably to fulfill this legal representation obligation.”117 Other state-level public assistance administrators have made similar pleas for additional funding from state legislatures in recent years.118

State executive officials can feel hamstrung by state law outside the context of budgetary constraints as well. They commonly argue in litigation that they cannot be responsible for noncompliance with a federal right because they are not empowered by state law to administer it.119 Whether officials make those arguments in good or bad faith is difficult to determine. I have encountered what I believe to be good faith iterations of the argument in litigation. Either way, federal rights that demand interbranch coordination permit states to avoid compliance for some time by playing this game of liability hot potato.120

3. State-local.

Federal law can also require coordination between states and their local governments. States often delegate (and, in some cases, abdicate)121 their federal responsibilities to local governments. When that happens, local governments play the role of state agencies.122

Local governments are quirky hybrids. They have no place in the constitutional structure,123 and yet they are the governments that lie closest to people’s everyday lives.124 Legally, they are creations of their states.125 But they are also, in many respects, independent actors that exercise their own autonomy.126

Further, local governments often operate with little oversight from their state creators. Professor Richard Briffault, who built the scaffolding that supports much local government scholarship, has observed that state-local delegations of authority are “often quite broad and [ ] rarely revoked. In most states, local governments operate in major policy areas without significant external legislative, administrative, or judicial supervision.”127 Despite being creations of the state in theory, local governments “function as representatives of local constituencies and not as field offices for state bureaucracies” in practice.128 Because of the variety of legal frameworks that states employ to organize their local governments, however, state-local interactions are wildly diverse across the country and difficult to characterize holistically.129

Given both the independence of local governments from their states and the significant ways that states rely on local governments as administrators, state-local coordination problems arise frequently in the context of federal rights. States administer many of their federal public assistance responsibilities, such as food stamps, Medicaid, and cash assistance programs, by delegating those duties to local governments.130 States also delegate many election administration responsibilities, such as voter registration and absentee balloting, to local governments.131

In other areas, federal standards govern existing state programs already administered by local governments. States don’t explicitly delegate their responsibilities under the Americans with Disabilities Act of 1990132 (ADA), for example, but because local government jails can house state prisoners who may require accommodations for disabilities, local governments end up administering the ADA.133 Similarly, states don’t always explicitly delegate their obligations under the Sixth Amendment to provide counsel to indigent criminal defendants, but because local prosecutors administer and enforce state criminal laws, many local governments end up funding compliance with the Sixth Amendment.134

These state-local delegations increase the coordination burden on the state executive. To achieve statewide compliance, states must monitor local compliance with these federal laws. But once delegated, states will deny that they remain responsible and disclaim responsibility for local noncompliance.135 Ensuring statewide compliance requires the state to engage in extensive trainings, closely monitor local governments, and possibly fund expensive government programs. And local administrators, who may be separately elected by their own local constituents, often resist strong state oversight. Moreover, only a small number of federal laws that straddle state-local relationships acknowledge and accommodate that additional coordination burden.136

On the flip side, the state-local relationship implicates coordination when local governments rely on state government to provide them with information crucial to administering a federal right. So in Armstrong v Schwarzenegger,137 for example, California had failed to transmit to local jails disability information about prisoners, making it impossible for those local jails to comply with the ADA.138 A similar information-sharing problem can arise in the context of the Help America Vote Act of 2002,139 which requires both that states maintain a current voter registration list140 and that a state or local official verify whether a provisionally cast ballot should be counted, based on current registration status.141 This postelection verification cannot happen if the state does not maintain an accurate voter list or if it fails to make that list available to local officials.

Coordination difficulties like these have resulted in noncompliance in the context of all the federal policy areas mentioned above. Because states resist responsibility for the actions of their local governments—even in the context of explicit federal obligations on states—litigation against states for local noncompliance can last years.142 And courts are inconsistent in how and whether they find state liability for local noncompliance, making these coordination problems difficult to resolve.143

B. Typology

In this Section, I canvass the barriers that arise within the terrain of state coordination and find that state bureaucracy can frustrate federal rights in three primary coordination-related ways. First, state actors and agencies can become alienated from one another and fail to communicate to achieve compliance. Second, state actors can be in conflict with one another and refuse to coordinate. Third, state actors may not be aware of the roles they must play as administrators and either crowd each other out or neglect the federal rights altogether.

These are not complicated concepts, but they also aren’t widely observed. Intrastate unrest is difficult to detect in litigation because agencies are often codefendants, not direct adversaries, and they are often represented by the same state attorney general. Outside of litigation, agencies rarely air their dirty laundry publicly, and even if they did, state-level press is particularly weak.144 The public is rarely privy to conflicts between state agencies.145

1. Alienation.

In 2014, the Pew Charitable Trusts set out to research and publicize the effectiveness of the Motor Voter law—one part of the NVRA—twenty-one years after it was passed.146 Instead of issuing a report containing statistics and compliance numbers, as it had hoped, Pew concluded that systematic study of Motor Voter compliance was impossible because of “deep and varied problems with the data (including how and if it [was] collected).”147 Pew found that state officials were largely unaware of whether their motor vehicle agencies complied with the NVRA and unable to provide accurate compliance numbers.148

In other words, agency alienation prevented Pew from assessing compliance with the NVRA. Before the NVRA, election officials and motor vehicle officials, housed in separate parts of the state, had no reason to work together. They have different (and rarely compatible) data systems, which makes integration and, ultimately, compliance difficult.149 Reaching a similar conclusion, the nonpartisan Presidential Commission on Election Administration stated that “Motor Voter is not working as intended” and noted that the absence of integrated registration information played a large role.150

Interestingly, Michigan is one of the only states that, prior to the NVRA, happened to house elections officials and motor vehicle officials inside a single state agency: the office of the secretary of state.151 It is no surprise, then, that both Pew and the Presidential Commission on Election Administration praised Michigan’s Motor Voter compliance as “seamless” and for “streamlin[ing] the voter registration process for the motor vehicle agency” by “reducing transaction times, standardizing names and addresses,” and improving user experience.152

The NVRA offers even more opportunities for alienation. As described above, for most states, NVRA compliance sprawls across state government both horizontally and vertically.153 It also sprawls diagonally. Depending on how the state has set up NVRA administration, state-level election officials may need to coordinate with local-level motor vehicle and public assistance agencies. So even if state-level election officials are in communication with other state-level directors, they may be alienated from local-level administrators outside of their control. This was recently a problem in California, where local public assistance offices ignored the secretary of state and instead looked for guidance from their own state-level directors.154

Organizational theory suggests that intra- and interagency alienation, and a general disconnect between agency heads and low-level administrators, is a feature of complex institutions.155 Professor Joanna Schwartz, drawing from this literature, has disputed the idea that complex organizations are always aware of noncompliance in the organization.156 Because “information [within complex organizations] is decentralized and held by a number of different people and entities,” and because “gaps in [organizational] design and implementation can frustrate information collection efforts,” “those at the highest levels of governance may not learn about incidents of wrongdoing or critical details of those incidents.”157

Agency alienation is a problem that is particularly acute at the state, rather than federal, level. It is a kind of structural mismatch between what federal laws require and what administrative capabilities state governments possess. As a body, federal legislators have a better idea about the capabilities of federal bureaucracy than state bureaucracy. Congress can tailor its delegation to the competencies of the federal administration or intentionally legislate interagency coordination and overlap.158 Alternatively, Congress can create new federal bureaucracy to suit its administrative needs. It can do none of this tailoring at the state bureaucratic level.

2. Conflict.

Sometimes intrastate conflicts are more contentious and difficult to resolve than those within the federal government. State executives and cabinet officers are often elected independently and may come from different political parties.159 Unlike federal administration officials, state officers and agency heads elected separately are not responsible to one another; they are not even necessarily beholden to the same sets of voters. Whereas federal officials are all working toward the same goal—and can be removed if they are not160 —state officers may have very different goals. An elected state secretary of state, for example, may rely on a coalition of voters that includes public assistance recipients and will have a strong interest in enforcing the provision of the NVRA that requires states to offer voter registration at public assistance agencies. A governor in the same state who relies on a coalition of voters more geared toward car owners will have a stronger interest in enforcing the provision of the NVRA that requires states to offer voter registration at motor vehicle offices. And since full compliance with the NVRA requires these two state officials to coordinate, compliance will suffer in a way that it would not if it were administered by multiple federal actors.

Consider the interagency conflict that gave rise to Virginia Office for Protection and Advocacy v Stewart.161 Two federal laws—the Developmental Disabilities Assistance and Bill of Rights Act of 2000162 and the Protection and Advocacy for Mentally Ill Individuals Act of 1986163 —offer federal funds to states to create an organization, either state- or privately run, designed to protect the rights of people with disabilities or mental illness.164 Virginia created a state agency—the Virginia Office for Protection and Advocacy (VOPA)—to fill that role.165 The dispute underlying the case arose when VOPA attempted to investigate mistreatment of disabled patients at state-run mental institutions.166 The state officers responsible for those hospitals refused to give information to VOPA, invoking peer privilege.167 VOPA sued.

VOPA and the state hospital directors were accountable to different actors; no hierarchical relationship existed between them. VOPA was governed by an eleven-member board: three appointed by the governor, five by the speaker of the state house, and three by the state senate committee on rules.168 Meanwhile, the state director of the mental hospitals was appointed by the governor.169 Further, while the state mental hospital received its funding from the state coffers, VOPA was funded federally.170 Given these differences, we might expect that these two state actors may not be willing to accommodate each other’s requests. And though the Supreme Court ultimately ruled that VOPA could sue the state hospital director, the conflict between those two state agencies frustrated federal disability rights law for nearly four years.171 Soon after the lawsuit, Virginia changed its law to consolidate administration of those federal laws into one commission fully appointed by the state bureaucracy.172

The concept of “picket fence federalism” partly explains this conflict. Picket fence federalism describes the strength of vertical relationships (the picket) among federal, state, and local administrators working on the same substantive issue.173 Because federal money often funds state bureaucrats that administer federal programs (as in VOPA), those state bureaucrats have a divided loyalty: not only to the state government, but also to the federal administrators and laws that fund them.174 State bureaucrats share vocational and mission-based priorities with federal administrators, and they can be “a surreptitious force for undermining the policymaking discretion of elected nonfederal generalists like state legislators, city councilors, county commissioners, mayors, and governors.”175 These state-level bureaucrats can refuse—or be unable—to be responsive to state elected officials176 because federal funding insulates them from the state-level political pressures that motivate elected politicians.177

An example of intrastate conflict involving both the state executive and legislature (and my favorite coordination-based example because of the sheer number of state bodies involved) comes from New York’s failure to comply with the Uniformed and Overseas Citizens Absentee Voting Act178 (UOCAVA). An amendment to UOCAVA requires states to transmit absentee ballots to military and overseas voters at least forty-five days prior to a federal election.179 The provision setting the forty-five-day requirement first became effective for the 2010 federal primary and general elections.180

New York’s election calendar posed a problem for compliance with the forty-five-day requirement. State law set New York’s federal primary election for September 14, 2010, just forty-nine days before the general election on November 2, 2010. The New York executive branch believed it was unable to meet UOCAVA’s forty-five-day requirement for the general election, so it obtained a hardship waiver from the Department of Defense on the condition that it comply with UOCAVA in other ways and change its election calendar going forward.181

But New York failed to change its election calendar to prevent similar problems in 2012, and 2014, and 2016. The New York State Board of Elections argued that it could not change New York’s primary election date unilaterally,182 and the state assembly and senate were unable to agree on a new date.183 The state assembly proposed a date in June to more reliably meet UOCAVA’s requirements;184 the state senate proposed a date in August, believing a June primary would inconvenience its members.185 Each of these proposals was adopted by two of the four members of the board of elections, who submitted these proposals to the court.186 The state senate even moved to intervene as a party, believing its interests were not well represented by the named defendants: the state (represented by the attorney general) and the board of elections (represented by its own counsel).187

Channeling Cool Hand Luke, the court quipped that “[w]hat we’ve got here is a failure to communicate.”188 The failure of state actors to coordinate stood in the way of statewide UOCAVA compliance. In its motion to intervene, the state senate expressed confusion at the different positions taken by different state actors.189 Because the state parties were never able to agree, the court continued setting the dates of New York’s elections through injunctions and consent decrees that made statewide compliance with UOCAVA possible.190

In one order, the court acknowledged New York’s coordination troubles outright and stated that compliance with UOCAVA depended on compliance by all state actors: “[T]he court observes that while the [state board of elections] filed the calendar, New York State is also a defendant. The court expects full compliance by all defendants, regardless of how they choose to effectuate such compliance.”191

Like agency alienation, agency conflict is also a particularly difficult barrier to overcome when it happens at the state, rather than federal, bureaucratic level. Like state agencies, federal agencies come into conflict with one another.192 But unlike state agencies, the federal bureaucracy has a clear line of authority to resolve these disputes, with cabinet secretaries and the president at the top.193 No such line can exist in states where state officers are separately elected, and in those circumstances, no ultimate authority (akin to a president) exists to force an intransigent state actor to comply.

3. Role confusion.

Another state bureaucratic barrier to compliance with federal laws is that, without sufficient direction from either state or federal law, state actors are not always clear on who is responsible for administering the law. That lack of clarity can prevent those actors from taking responsibility for compliance and makes state accountability more difficult to achieve.

Because every state is different, federal laws rarely specify a state actor to administer the program.194 Whereas federal lawmakers can craft legislation that relies on the specific structure of the federal bureaucracy—and often do195 —they are limited in their specificity when it comes to states as administrators.

Federal laws that rely on states take a variety of approaches to coordinating state administration. On one end of the spectrum, some federal rights simply create obligations on states and provide no detail as to how state officers should work together as administrators.196 On the other end of the spectrum, some federal laws or regulations—especially those that require state buy-in—identify specific roles that state officers must take as administrators. In those cases, it is up to the states themselves to determine which state officer plays which role.

Any federal law that does not instruct states on how to coordinate internally is susceptible to crowding and neglect. That is, it can lead to multiple state actors stepping on one another’s feet or the opposite problem: state actors assuming that someone else is responsible. Courts are left to sift through state law to determine hierarchies of authority between state actors, a task that carries its own challenges.197

A 2017 lawsuit alleging noncompliance with the Sixth Amendment right to counsel by South Carolina’s municipal courts is illustrative.198 Municipalities in South Carolina may create their own local courts, which have authority to prosecute low-level misdemeanor crimes.199 However, South Carolina state officials are unclear on which state officer bears ultimate responsibility for these courts. A recent New York Times article paraphrases the administrator for South Carolina’s Office of Court Administration as saying that “her office played no role in oversight of municipal courts and that the State Supreme Court was responsible.”200 The state supreme court clerk “said that the Office of Court Administration was responsible.”201

Even when federal laws do specify roles for state actors, and even when state actors assume those roles, those actors can feel constrained by the limits that state laws place on their authority to administer. Whether federal laws can provide state actors with powers state law does not already grant is an open question.202 So even highly specific federal laws can fall prey to state coordination roadblocks in ways that federal laws harnessing the federal bureaucracy do not.

Litigation in Virginia over compliance with the federal Food Stamp Act of 1977203 highlights the compliance challenges of role confusion. In Robertson v Jackson,204 plaintiffs eligible for food stamps sued a Virginia state official—the commissioner of the Virginia Department of Social Services—for local noncompliance with timeliness and other requirements of the Food Stamp Act.205 The commissioner argued that he could not be held responsible for local noncompliance around the state because the plaintiffs had failed to sue both the local agencies that administered the food stamp program and another state actor, the State Board of Social Services, which also played an oversight role.206 The commissioner alleged that it was the local social service agencies that were actually responsible for administering the federal program. He also argued that it was the state board, and not the commissioner, that was empowered by state law to both implement the food stamp program in the state and “to require the local authorities to perform the duties imposed by law.”207

Here, then, the defendant state commissioner was making role-confusion arguments in three different ways: interagency confusion, as between the commissioner and the state board; interbranch confusion, as among the legislature, the commissioner, and the state board; and state-local confusion, as between state actors and local agencies.

This barrier is another that is more salient at the state level than at the federal level. Whereas the federal bureaucracy has little legal claim to control how it organizes itself when otherwise directed by Congress, states do. As described above, the Supreme Court has shielded states from federal interference.208 In doing so, it has noted that how a state chooses to structure its government is at the core of its sovereignty.209 In ways that the federal bureaucracy cannot, states use the language of state sovereignty in litigation to resist compliance with federal law when it requires states to reorganize their government.210

III. Coordination, Politics, and Inequality

Bureaucratic undermining carries few of the traditional markers we associate with state-federal struggle. It does not appear to be correlated with geography in the way that, say, Congress determined that vote discrimination was an especially serious problem in the South.211 Coordination-based noncompliance cases arise in a geographically diverse set of states.212 Nor does bureaucratic undermining appear to be driven by political party or ideology:213 state bureaucratic undermining occurs in states across the political spectrum.214

If not politics or geography, what explains state bureaucratic undermining? This Part explores practical and theoretical explanations for the bureaucratic barriers described above. It argues that because intrastate coordination requires political will, federal rights requiring extensive coordination will not always effectively serve the people they seek to benefit. Those with little political power—including racial minorities and low-income populations—struggle to force state bureaucratic compliance. States, as compared with the federal government and local governments, are often uniquely unreceptive fora for prorights political efforts.

Breaking down any of the three barriers described above—alienation, conflict, and role confusion—requires coordination in the form of either consent or coercion.215 Either form requires expending political capital.216 Take a federal law that demands coordination between two state actors with a hierarchical relationship (state and local actors, for example). Given a baseline of inaction or noncompliance, statewide compliance requires the state-level actor to convince the local-level actor to vindicate the right. The state actor may be able to convince the local actor to comply without coercion—perhaps by offering funding or reminding the local actor of the legal obligation. If unsuccessful, the state actor will need to coerce, which implicates the legal architecture governing that state-local relationship. If the state actor has removal power over the relevant local actors, it may exercise that power. If it no longer wants to entrust local actors with administration responsibilities, it could seek authority from the legislature to centralize administration of the right. It might even bring a lawsuit against the local agency in cases of local intransigence.217

All of these options are costly. Funding compliance at the local level imposes financial costs. Suing a local administrator imposes litigation costs. Removing a local administrator may impose political costs if perceived as an intrusion on local autonomy, especially if the local administrator is popular or the right is unpopular at the local level. Requesting additional authority from the state legislature to recentralize administration of the right at the state level could involve negotiations with the legislature and long-term political costs.

The options available to state actors in a nonhierarchical relationship are also costly. Conflicts between cabinet-level actors that can’t be resolved will require intervention either by a higher state power, if one exists, or the courts. Even conflicts that can be resolved through consent require one actor to identify the issue and convince the second actor to coordinate.

Less costly coordination actions tend to be less effective. A recent study about compliance with the NVRA found that minimally obtrusive state actions—like well-timed trainings for local officials and emails reminding them to comply—were only minimally effective in fostering local compliance, and only in local offices that were already partly compliant.218 The interventions failed to increase compliance in noncompliant local offices.219

In cases where these kinds of thorny intrastate conflicts have undermined federal rights, courts have observed that political will is an important part of full compliance. In the UOCAVA case against New York described above, New York’s existing election calendar prevented the state from transmitting ballots to military and overseas voters with enough time before the election.220 Between 2010 and 2015, New York’s state board of elections, state assembly, and state senate were unable to come to an agreement on amending New York’s election laws to make compliance possible. “Having had ample opportunity to correct the problem,” the federal court noted, New York “failed to find the political will to do so.”221 That lack of political will required the court to “becom[e] embroiled in [New York’s] election schemes,” despite normal considerations of comity and federalism.222

Prison overpopulation in California is another example. Federal courts ordered California to reduce its overcrowded prison population. California’s plan to reduce its prison population (called “realignment”) required state-local coordination. Because it involved sending state prisoners to local jails, Governor Jerry Brown needed to secure the cooperation of local governments.223 One of the judges in the case described this process as follows: “It’s clear to me . . . that realignment is a political deal, in which the Governor went to the fifty-eight counties and got something that every county could live with.”224

On the other side of the ledger, costs to the state for noncompliance are relatively low. The Eleventh Amendment immunizes states from retroactive damages claims even when the state or a state actor violates a federal right.225 Plaintiffs may obtain prospective injunctive relief,226 but that amounts only to compliance with the statutory or constitutional right that the state was already responsible for.227

In addition, state actors are unlikely to suffer political costs for failing to comply with federal rights that seek to empower the disenfranchised. As a general matter, state actors do not face much political pressure from low-income populations—those most likely to be the beneficiaries of federal rights. Voter turnout falls as income falls.228 Public assistance recipients, who stand to benefit from compliance with the NVRA, lack political power and influence. As Professor Laurence Lynn put it, “Human services have no constituency. Their clients do not vote, at least not in large enough numbers to make a difference to the average legislator.”229 And it is well-established that policy outcomes are largely unresponsive to the preferences of low-income individuals.230

Not unlike low-income communities, racial minorities also on average lack political power as compared to whites.231 The data show that blacks and Hispanics struggle to influence governmental policy: “Spending on a given item is more apt to decrease when blacks and Hispanics favor a rise, and more apt to increase when they favor a fall.”232

Lacking political power is particularly meaningful in the state context. Our system of federalism already makes change difficult to achieve by creating layers of government with overlapping jurisdiction and competency. Professor Lisa Miller has argued that this structure disadvantages those without political power: “American federalism reinforces existing race and class stratification by creating a system of multiple political venues that makes it difficult for poorly resourced groups to navigate and by driving several layers of government between citizens who experience these problems and lawmakers who have the capacity to ameliorate them.”233 Miller is writing about criminal law and justice, but her argument is doubly powerful in the context of decentralized federal laws. By separating principal from agent and decentralizing those legal relationships down through the levels of government, these federal laws already obscure lines of authority and accountability and make reform complex and burdensome.234 State coordination adds an additional “venue” to the topography of state governmental actors.

But states are uniquely bad places to add additional venues. State government is particularly opaque to outside interests—in Professor Miriam Seifter’s words, “largely invisible. Whereas a ‘synopticon’ of watchdogs monitors the federal executive branch at every turn, state bureaucracy does not operate in a fishbowl.”235 Because state government is overall less transparent than the federal government,236 state residents are less aware of how states allocate power internally than they are of how power is allocated at the federal level. States privatize and delegate many of their own functions, making internal markers of responsibility difficult to find. And state media has contracted in recent years, leaving the inner workings of state government less scrutinized and understood than other levels of government.237

In addition, internal state politics make state governments particularly bad fora to find the political will to administer federal rights. Professor Sheryll Cashin has written about the “political economy of state fiscal decisionmaking” in the context of federal redistribution programs and has demonstrated that states are less interested than the federal government in income redistribution, in part because “state political actors, most critically governors, are rationally compelled toward the provision of ‘middle class’ services.”238 Professor Briffault has similarly argued that states are both more attuned to suburban interests than they are to urban ones239 and reluctant to “displace local authority when considerations of equity or efficiency make it appropriate to do so.”240

Cashin also notes that racial minorities are more likely to be subject to discrimination and unfavorable treatment at the state level than at the federal level. Again in the public assistance context, she found that state administrators were more likely than federal administrators to adopt racist attitudes toward African Americans receiving welfare.241 And more broadly, of course, “the history of racial subordination in the United States has been marked by a great deal of state sponsorship or acquiescence in racist acts and policies.”242

Finally, states are more susceptible to lobbying from business interests than they are from public interest organizations. That is, they are susceptible to “capture” by business interests. “Capture” refers to the idea that interest groups—often corporate interests—are able to further their interests through administrative action at the expense of the public interest.243 State agencies are more likely to subordinate the public interest to industry interests for two reasons. First, public interest groups have been less active at the state level than at the federal level and have not effectively balanced out the influence of corporate interests at the state administrative level.244 Second, although governors, like presidents, are better able to juggle interest group priorities than are individual state agencies, state agencies enjoy less executive oversight because state executive branches are less centralized than the federal executive branch.245

These three reasons—state opacity, a political slant away from the interests of racial minorities and low-income groups, and a skew toward business interests—all make state government especially difficult to navigate for disenfranchised communities. Federal rights that demand intrastate coordination, in concert with state governments not built for that coordination, increase state-level complexity of administration.

Extensive intrastate coordination requirements can therefore serve as an avenue for existing inequality to manifest as unequal administration, even without any obvious racial animus.246 This resonates with accounts of structural inequality. Professor Richard Ford, for example, has put forth a structural argument that

[r]ace-neutral policies, set against an historical backdrop of state action in the service of racial segregation and thus against a contemporary backdrop of racially identified space—physical space primarily associated with and occupied by a particular racial group—predictably reproduce and entrench racial segregation and the racial-caste system that accompanies it.247

Professor Rachel Moran has noted that

the model of racial animus that animated Brown seem[s] to be growing obsolete. . . . Yet, even with these reported gains in racial tolerance and understanding, residential segregation has remained a commonplace feature of American life, and significant gaps in educational attainment, earnings, and wealth persist between White and non-White Americans.248

Although the idea of structural racism and inequality is not a new one,249 we are continuing to learn about how racism and income inequality have become so deeply a part of our institutions that those institutions now reproduce inequality inadvertently. Statisticians are beginning to demonstrate how existing racial inequality biases computer-based criminal justice tools designed to take race out of the equation. Computer programs that determine bail amounts based on supposedly unbiased factors rely on arrest history, for example, but prior arrests are determined by racially discriminatory arrest patterns.250 Sentencing algorithms also contain “hidden biases” that disfavor racial minorities.251 These programs, designed to display a veneer of neutrality, become biased against racial minorities because they absorb the racial bias of the rest of the criminal justice system.252

IV. Constructing the Channels of Political Change

Professor John Hart Ely’s classic metaphor for the role of courts in policing the law of democracy resonates here. Ely wrote that courts should err on the side of “clear[ing] the channels of political change” to prevent power from becoming entrenched in the hands of the powerful.253

Before you can keep channels clear, you must create them. State agency alienation, conflict, and role confusion represent different kinds of nonexistent, blocked, or damaged channels through state bureaucracy. Effectively administering a federal obligation requires states to construct these channels, even through rocky terrain: between actors of different political parties, between actors that have no other reason to communicate, or between actors averse to the obligation in the first place.

In this Part, I catalog remedies—including injunctions, consent decrees, and out-of-court settlements—that seek to address state coordination problems by creating and repairing these channels. These remedies reflect the state coordination problem in their form and focus. Unlike the traditional injunctions of the civil rights era that were often hundreds of pages long254 and “took the form of highly detailed regulatory codes embracing vast provinces of administration,”255 these remedies take the form of choreography. They treat functional government as the goal and create pathways between state actors by specifying how they must work together to achieve compliance.

I call these features “coordination remedies.” I find that recent coordination remedies primarily address intrastate coordination issues in two ways: agency integration and role clarity. I describe these two remedial tools using examples of remedies from suits against states and suggest a third tool that I have not seen executed: a state agency employee dedicated to representing the interests of beneficiaries of federal rights who are not sufficiently represented in state government already.

Understanding these remedies is valuable for four reasons. First, they describe the tools practitioners believe are necessary to address state coordination problems and vindicate federal rights. Because many of these remedies are negotiated consent decrees and out-of-court settlements, they should also provide insight into the kinds of barriers state bureaucrats themselves see and the changes they believe are necessary to clear the channels. This on-the-ground perspective is impossible to glean from the statutory language itself.

Understanding these remedies is also helpful for Congress to understand both what kinds of structures undermine its laws and what it takes to solve these problems. Though this Article focuses on state noncompliance, it is at its heart a critique of the federal decision to decentralize. As described above, Congress currently takes a variety of approaches to coordinating state bureaucracy, from doing nothing at all to requiring states to provide something akin to a coordination plan.256 But Congress could do more. Concerns like state bureaucratic undermining should make Congress think twice about relying on states to administer complex statutory rights that join extensive intrastate coordination with politically marginalized beneficiaries unlikely to motivate state actors themselves.

Second, even if Congress chooses to decentralize, it can design its policies to better accommodate contemporary challenges of state administration. As Professor Heather Gerken points out, fully recentralizing decentralized federal rights likely requires a level of national political will that, almost by definition, does not exist—if it did, the policy would not be decentralized in the first place.257 But numerous reforms short of full recentralization exist as well. As described in this Part, those reforms include legislative tools and requirements for constructing channels of state authority and efficiency258 as well as mechanisms for motivating state actors like proxy-representation requirements or monetary penalties.259

Third, I hope that synthesizing these remedies will be helpful for advocates, potential plaintiffs, and potential defendants. When a state appears to be out of compliance with a federal right, these remedial trends can serve as diagnostic tools: top-line locations within the state topography to examine to find the root of the problem. Compiling these remedies may also be useful if a defendant challenges the intrusions into state sovereignty a coordination remedy makes.260 In this Part, I describe a set of courts that have already approved, or even issued, remedies of this kind.

Finally, understanding this set of contemporary remedies helps us understand remedies trends generally. Public law litigation is “[p]rotean,”261 and public law remedies have evolved over the past fifty years. Contextualizing coordination remedies within the broader legal and political milieu may help us identify how states and state institutions are changing as well.

A. Agency Integration

Remedies for coordination-based noncompliance with federal law commonly seek to integrate two state actors so that they are both able and forced to communicate effectively. Coordination remedies take two approaches to integration, both of which address agency alienation and conflict.

The first is specific coordination instructions that can be as concrete as requiring a state actor to work with local actors to transmit absentee ballots262 or as ambitious as requiring state actors to work together to change state law.263 These instructions are mundane and mechanical, and yet state actors are unable to execute them. In lieu of specific instructions to these actors, recent coordination remedies can require that the state actors direct a state employee to coordinate between agencies.264

Second, coordination remedies promote agency integration by requiring state actors to monitor compliance with the federal right and report that data to codefendants, the plaintiffs, and the court.265 Requiring a state to actively monitor compliance among the relevant administrators not only forces the state to own its compliance numbers,266 but also strengthens the relationships among those state actors. Unless federal laws or remedies require regular and functional communication between state bodies, that communication is unlikely to occur on its own. As one voting rights advocate put it, remedies that mandate state supervision over local governments help local “responsibilities become institutionalized as part of the agency infrastructure—just like any other agency responsibility.”267

An example comes from an out-of-court NVRA settlement reached between advocacy organizations and two California state actors: the secretary of state and the executive director of Covered California, the state’s health care exchange under the Affordable Care Act268 (ACA). Four years after President Barack Obama signed the ACA, the ACLU and other advocacy groups noticed that California was not offering voter registration forms with applications for federally subsidized health coverage through California’s exchange,269 in violation of the NVRA.

California was a sensible, sympathetic target state for the ACLU. The state secretary of state is California’s chief election official270 and is specifically designated as responsible for complying with the NVRA. Although the secretary—Debra Bowen, at the time—and the governor were not accountable to one another, they were both Democrats.271 The secretary, though responsible under California law for complying with the NVRA,272 is not responsible for administering California’s state ACA exchange, however. That responsibility rests with the executive director of Covered California.273 The secretary and the executive director would need to coordinate to comply with the NVRA; they were not coordinating.274

The potential plaintiffs were able to reach an out-of-court settlement with the two state officials.275 The document clearly intends to foster integration between the two state actors in both of the ways described above. It spells out the many ways that the secretary and Covered California must interact to ensure compliance. Covered California must provide the secretary with a draft letter it intends to send to its clients about voter registration;276 it must tell the secretary how many voter registration forms it needs to send with its public assistance applications;277 the secretary must supply those forms to Covered California in a timely manner;278 and so forth.

It also makes the two actors accountable to one another. It requires the two actors regularly to “communicate,”279 to send each other administrative updates on compliance,280 and to collect and send each other enrollment and registration statistics.281 These requirements force the two state actors to provide each other with up-to-date status reports on their compliance.

B. Role Clarity

Reflecting the role confusion that can arise when states are not naturally organized to administer a federal law, coordination remedies carefully define the roles, responsibilities, and powers of the defendants in the litigation. They ensure that a clear line of authority exists among state bureaucratic actors, and they provide role specificity when more than one state actor is available to play a role.

As an illustration, consider this 2007 institutional suit that indigent criminal defendants brought against the state and governor of New York alleging that the state violated the defendants’ Sixth Amendment right to counsel because five counties failed to provide adequate representation.282 After seven years of litigating whether the state and governor of New York were proper defendants to the action (they were),283 the parties settled in 2014.284 As New York is a large state and compliance with the Sixth Amendment requires, like many federal laws, a great deal of cooperation between state and local actors, the parties entered into a coordination remedy.

The remedy choreographed cooperation between state actors, beginning with role clarity. It acknowledges the state administrative bodies designed to improve the quality of indigent defense in New York—the Office of Indigent Legal Services (ILS) and the Indigent Legal Services Board (ILSB)—and states that these bodies “have the legal authority to monitor and study indigent legal services in the state, to recommend measures to improve those services, to award grant monies to counties to support their indigent representation capability, and to establish criteria for the distribution of such funds.”285 The remedy states that “the parties agree that ILS is best suited to implementing, on behalf of the State, certain obligations arising under this Agreement.”286 Other state actors include the governor and the five counties where noncompliance arose.287

Perhaps reflecting a concern about the role of the ILS, which was not a party to the lawsuit, the remedy states that the ILSB directed the ILS to comply with the terms of the remedy and includes, as an appendix, a document entitled “Authorization of the Indigent Legal Services Board and the New York State Office of Indigent Legal Services Concerning Settlement of the Hurrell-Harring v. State of New York Lawsuit.”288 This document acknowledges that some of the representations in the remedy require implementation by the ILS, direct the ILS to implement the requirements, and state that the ILS agrees to implement the requirements.289

This document institutionalizes the remedy’s requirements into the state’s administrative infrastructure and recognizes that state agencies may maintain independence from the governor, despite the governor’s best efforts. Here, the ILSB is an independent agency. Its chair is the chief judge of New York’s Court of Appeals,290 and its members are unpaid.291 The remedy brings the ILSB into the fray and aligns the mission of the ILSB with the mission of the decree.

Finally, the remedy creates a direct line of authority and supervision among state actors. It requires the governor to “coordinate and work in good faith with the Office of Court Administration [ ] to ensure, on an ongoing basis, that each judge and magistrate within the Five Counties, including newly appointed judges and magistrates, is aware of the responsibility to provide counsel to Indigent Defendants at Arraignments.”292

These remedies fill in the gaps of existing state law. In this sense, coordination remedies are judicial equivalents to state-enabling legislation. State legislatures commonly enact state-enabling laws that translate the general terms and requirements of federal laws onto the specific contours of state structures.293

C. Representation

These remedies—integration and role clarity—address the bureaucratic barriers of agency alienation, conflict, and role confusion that I describe above.294 But they do not address the effects that the coordination problem can have on racial minorities and low-income groups.295 In fact, while surveying these remedies, I have not seen any that specifically guard against coordination’s potential to reinforce racial and income stratification.

Scholarship on representation in the federal administrative state provides the beginnings of a solution. Administrative law scholars have recently proposed the idea of “proxy representation”: government-funded administrators stationed within the federal bureaucracy that give voice to underrepresented groups in the rulemaking process.296 These proxies are meant to counterbalance the existing influence of business interests on federal rulemaking.

Coordination remedies should make use of proxy representation as well. As described above, business interests have an outsized impact on decisionmaking in state government—even more so than at the federal level.297 To promote stakeholder representation from within state government, coordination remedies could require states to appoint an internal proxy representative to protect the interests of politically marginalized groups. These remedies already commonly require states to appoint one or more internal employees responsible for facilitating coordination298 —either appointing another employee to represent stakeholder interests during the administrative process or adding that responsibility to existing coordinators would not be particularly burdensome. Congress could—inside of federal legislation that uses states as administrators—require states to appoint proxy representatives at the outset.

I see proxy representation as a way to institutionalize responsiveness to marginalized voices, especially voices with no clear way to make political change. The representative should be in touch with the broader set of stakeholders around the state or local government. The power of the representative could vary, from a required consultation, to someone who must be in the room when administrative decisions are made, to someone with actual veto power over a limited set of administrative decisions that affect politically marginalized groups.

This kind of representation comes with its own administrative challenges. It does not escape me that adding an additional node to the state bureaucratic policymaking process itself has the potential to contribute to the coordination problems that cause undermining. In addition, deciding how to incorporate proxy representation inside of state government requires choosing some stakeholders over others and could lead to further exclusion of groups with little political power. Any given court, state government, or federal statute would need to balance the benefits of improved representativeness with the costs of additional bureaucracy.

Conclusion

I intend this Article to offer a new critique of decentralization. A word of warning at a time when states appear to be the best venue for progressive change. And an invitation for scholars to dive more deeply into state and local bureaucracy to better understand both how federal laws filter down to people themselves and why those laws have struggled to correct persistent inequality. More specifically, I understand the state coordination problem to be an understudied mechanism by which rights lose their effectiveness and racial and income stratification remain entrenched. Extensive coordination requirements allow intrastate politics and bureaucracy to undermine federal rights.

I expect this undermining to grow, not recede. State coordination problems may seem small when viewed in the context of federalism writ large, and I concede that not all federal rights implicate state bureaucracy and coordination. But as described in Part I, state bureaucracy is an increasingly important vehicle for federal policy priorities. It is increasingly large. And it is increasingly protected by courts from unwanted federal interference.299

Stepping back, I conclude by posing two questions. First, is state bureaucratic undermining a form of resistance to federal law? It hasn’t been treated as such, certainly by the academy—although there is a massive literature on intentional state resistance to federal law during the past seventy years, almost nothing exists that documents this subtler, but still devastating, bureaucratic undermining. Nor does undermining resemble the kind of resistance we are familiar with: “[o]rganized . . . opposition to an invading, occupying, or ruling power.”300

But there is another way to use the word “resistance”—a second definition: “[t]he impeding or stopping effect exerted on an object or substance by another, or by a force.”301 This is state bureaucratic undermining—resistance as friction.

Expanding our idea of state resistance to federal law to include this second definition is one way to acknowledge and address structural—rather than overt—inequality and racism within our system of decentralized law. It ups the stakes of this kind of noncompliance, not unlike how employment discrimination scholars have developed theories of disparate impact that expand our understanding of how workplace and societal structures cause discrimination.302 To the extent that state-sponsored discrimination is evolving, and state bureaucracy provides new avenues for compounding existing inequality, our federalism vocabulary should reflect that evolution.

Second, is state bureaucratic undermining something we should condemn, or is it a feature of federal decentralization that serves a useful function in our federalist government?

As a preliminary matter, the partisan valence of undermining may be less obvious than the case studies in this Article suggest. I used examples from federal rights that get more support from the left than the right: indigent defense, public assistance, prison reform, etc. The military voting statute, which joins military vote outreach with ballot access, may be more politically ambiguous. Even so, Democrats would likely support vigorous federal oversight of the collection of rights highlighted here, whereas Republicans might hope for, defend, or even build in opportunities for undermining.

But you could easily imagine a decentralized federal law that the left would want to undermine and the right would want to vigorously enforce. Consider a federal immigration or deportation law that requires states to act and in doing so, requires otherwise alienated state agencies to work together. The politics of enforcement would be flipped. That said, for the reasons I laid out in Part III, I would expect undermining to harm those with the least political power to make state government work.

With the political and policy valence of undermining potentially up for grabs and undertones of racial and income inequality, process considerations—rather than substantive or political ones—will be most helpful in evaluating undermining. That is, undermining is least likely to serve a useful function when it has no expressive or dialogue-creating properties because it is hidden deep inside state government or in some other media-poor environment, like small local governments; when the state policymaking process excludes stakeholder voices; or when actual noncompliance with federal law is at issue, rather than rights-narrowing state action.

These process considerations help contextualize state undermining on the spectrum of state resistance to federal law more broadly and differentiate it from the examples of state and local resistance I began this Article with. California’s “sanctuary state” legislation and its compliance with the NVRA, for example, are two different kinds of resistance (understood broadly) that arise from decentralized policymaking. Whereas California’s immigration policy is expressive, set to be litigated in the public square, and a valuable national conversation starter, its violations of the NVRA are hard to detect, exclude stakeholder input, and mute disagreement through political disempowerment.

State undermining might therefore serve a useful role in our federal government, but not if it entrenches existing political or income inequality. Before we celebrate state and local resistance, we should make sure to unearth what ordinarily lies hidden: interactions between state actors that provide new explanations for some of our most intractable inequalities. We should not take our eyes off a growing, changing state bureaucracy.

• 1Adam Nagourney, California Strikes a Bold Pose as Vanguard of the Resistance (NY Times, Jan 18, 2017), online at http://www.nytimes.com/2017/01/18/us/california-strikes-a-bold-pose-as-vanguard-of-the-resistance.html (visited Apr 18, 2018) (Perma archive unavailable). See also Editorial, California Looks to Lead the Trump Resistance (NY Times, Dec 9, 2016), online at http://www.nytimes.com/2016/12/09/opinion/california-looks-to-lead-the-trump-resistance.html (visited Apr 18, 2018) (Perma archive unavailable).
• 2Gabriel Debenedetti, Trump Opposition Sets Up Blue-State Headquarters (Politico, Jan 8, 2017), archived at http://perma.cc/MUM5-NDHD.
• 3Heather K. Gerken, Federalism 3.0, 105 Cal L Rev 1695, 1696 (2017). See also Heather Gerken, We’re about to See States’ Rights Used Defensively against Trump (Vox, Jan 20, 2017), archived at http://perma.cc/7CVV-L3KM.
• 4See William J. Stuntz, Unequal Justice, 121 Harv L Rev 1969, 1980 (2008) (“State and local power were key sources of resistance to [the civil rights] movement.”); Erwin Chemerinsky, The Values of Federalism, 47 Fla L Rev 499, 513 (1995).
• 5See Part II.
• 6For general discussion of seemingly intractable noncompliance, see Lisa L. Miller, The Invisible Black Victim: How American Federalism Perpetuates Racial Inequality in Criminal Justice, 44 L & Society Rev 805, 805–06 (2010) (reviewing statistics and asking “[h]ow are we to make sense of [racial] disparities half a century after Brown v. Board of Education (1954), the Civil Rights Act of 1964, and the Voting Rights Act of 1965?”). More specifically, noncompliance with many of the federal rights this Article focuses on—in particular the right to counsel and federal voting rights—is widespread. Nationally, states have failed to fulfill the promise set out by the Sixth Amendment. See David Carroll, Gideon’s Despair: Four Things the Next Attorney General Needs to Know about America’s Indigent Defense Crisis (The Marshall Project, Jan 2, 2015), archived at http://perma.cc/W6AH-EL26; Cara H. Drinan, The Third Generation of Indigent Defense Litigation, 33 NYU Rev L & Soc Change 427, 429 (2009) (“[D]espite voluminous documentation of the indigent defense crisis, the crisis persists.”). And the recent Presidential Commission on Election Administration report, as well as other recent surveys, have described widespread noncompliance with federal election laws. Presidential Commission on Election Administration, The American Voting Experience: Report and Recommendations of the Presidential Commission on Election Administration 15–18 (2014), archived at http://perma.cc/4R5D-KNBT. See also Justin Weinstein-Tull, Election Law Federalism, 114 Mich L Rev 747, 759–61 (2016) (describing widespread noncompliance with a set of federal election laws).
• 7To see how states have decentralized their powers horizontally across state actors, see Council of State Governments, 49 The Book of the States 189–94 (2017) (listing the appointment method of various state-level directors in each state). For discussion of broad state delegation to local governments, see Justin Weinstein-Tull, Abdication and Federalism, 117 Colum L Rev 839, 847–64 (2017) (describing state delegation of administrative responsibilities to local governments in the contexts of elections, public assistance, indigent defense, prison conditions, and education); Richard Briffault, Our Localism: Part I—The Structure of Local Government Law, 90 Colum L Rev 1, 1 (1990) (“State legislatures, often criticized for excessive interference in local matters, have frequently conferred significant political, economic and regulatory authority on many localities.”).
• 8In this paper, I highlight examples from California, New York, Illinois, and other states.
• 9Pub L No 103-31, 107 Stat 77 (1993).
• 1052 USC § 20504(a)–(c) (mandating that each state shall provide voter registration opportunities with driver’s license applications and renewals in motor vehicle offices); 52 USC § 20506(a)(2)(A) (public assistance agencies); 52 USC § 20506(a)(2)(B) (state disability offices); 52 USC § 20506(a)(3)(B) (other state and local agencies, such as public libraries, public schools, city and county clerks’ offices, fishing and hunting license bureaus, government revenue offices, and unemployment compensation offices); 52 USC § 20506(c) (military recruitment offices). Other federal election laws make similarly vague requirements on states. The Uniformed and Overseas Citizens Absentee Voting Act (UOCAVA), Pub L No 99-410, 100 Stat 924 (1986), for example, requires that “[e]ach state shall” send absentee ballots to military and overseas voters who request them no fewer than forty-five days before an election. 52 USC § 20302(a)(8).
• 11National Conference of State Legislatures, Election Administration at State and Local Levels (June 15, 2016), archived at http://perma.cc/AJ5B-QFYH.
• 12See Gideon v Wainwright, 372 US 335, 343–45 (1963) (requiring states to provide counsel to indigent criminal defendants who cannot afford their own).
• 13See Part II.A.
• 14See Jeffrey L. Pressman and Aaron Wildavsky, Implementation: How Great Expectations in Washington Are Dashed in Oakland; Or, Why It’s Amazing That Federal Programs Work at All, This Being a Saga of the Economic Development Administration As Told by Two Sympathetic Observers Who Seek to Build Morals on a Foundation of Ruined Hopes 133–35 (Berkeley 3d ed 1984).
• 15Id at 102.
• 16See Part I.
• 17See Part I.
• 18See William N. Eskridge Jr, Some Effects of Identity-Based Social Movements on Constitutional Law in the Twentieth Century, 100 Mich L Rev 2062, 2319 (2002) (noting that the Supreme Court sought to protect state sovereignty by conducting a “rollback of federal legislative power” in the 1990s as states regained their credibility after the civil rights era). See also notes 61–81 and accompanying text (describing doctrinal changes protecting states from federal intrusion).
• 19Owen M. Fiss, Foreword: The Forms of Justice, 93 Harv L Rev 1, 4 (1979).
• 20See Miriam Seifter, Gubernatorial Administration, 131 Harv L Rev 483, 485 (2017) (noting a dearth of scholarship on American governors); Jim Rossi, Overcoming Parochialism: State Administrative Procedure and Institutional Design, 53 Admin L Rev 551, 553 (2001) (noting the lack of scholarship on state administrative law). Penetrating work on internal conflicts between and within federal agencies is just starting to appear. See Daniel A. Farber and Anne Joseph O’Connell, Agencies as Adversaries, 105 Cal L Rev 1375, 1387–1408 (2017) (describing recent research that examines inter- and intra-agency cooperation and conflict in the federal bureaucracy and setting out a helpful typology of these conflicts); Jennifer Nou, Intra-Agency Coordination, 129 Harv L Rev 421, 451–73 (2015); Bijal Shah, Uncovering Coordinated Interagency Adjudication, 128 Harv L Rev 805, 820–30 (2015); Jason Marisam, Interagency Administration, 45 Ariz St L J 183, 189–200 (2013); Jody Freeman and Jim Rossi, Agency Coordination in Shared Regulatory Space, 125 Harv L Rev 1131, 1145–51 (2012). Professor Bertrall Ross has also done fascinating recent work on how federal agencies protect vulnerable groups by administering civil rights statutes. See Bertrall L. Ross II, Administering Suspect Classes, 66 Duke L J 1807, 1829–39 (2017); Bertrall L. Ross II, Embracing Administrative Constitutionalism, 95 BU L Rev 519, 527–30 (2015).
• 21With a few exceptions. See Bridget A. Fahey, Health Care Exchanges and the Disaggregation of States in the Implementation of the Affordable Care Act, 125 Yale L J F 56, 59–64 (2015) (describing intrastate conflict in the context of the Affordable Care Act); Bridget A. Fahey, Consent Procedures and American Federalism, 128 Harv L Rev 1561, 1573–75 (2015); Roderick M. Hills Jr, Dissecting the State: The Use of Federal Law to Free State and Local Officials from State Legislatures’ Control, 97 Mich L Rev 1201, 1232 (1999) (considering how courts should treat state law that prevents state and local officials from administering federal law).
• 22See Heather K. Gerken, Foreword: Federalism All the Way Down, 124 Harv L Rev 4, 21–44 (2010). Other scholars have examined the ways that states energize, resist, and work around federal laws—in particular, cooperative-federalism programs like health care, telecommunications law, and public assistance laws. See Abbe R. Gluck, Intrastatutory Federalism and Statutory Interpretation: State Implementation of Federal Law in Health Reform and Beyond, 121 Yale L J 534, 584–94 (2011) (noting that the Affordable Care Act contains at least five different kinds of federal-state cooperation); Jessica Bulman-Pozen and Heather K. Gerken, Uncooperative Federalism, 118 Yale L J 1256, 1285–87 (2009) (describing how disagreement among the federal, state, and local governments can result in productive dialogue and disagreement); Jim Rossi, Dual Constitutions and Constitutional Duels: Separation of Powers and State Implementation of Federally Inspired Regulatory Programs and Standards, 46 Wm & Mary L Rev 1343, 1370–83 (2005) (suggesting that state law and state courts, rather than federal courts employing the Supremacy Clause, are the better fora for reconciling state constitutional constraints with federal obligations); Philip J. Weiser, Towards a Constitutional Architecture for Cooperative Federalism, 79 NC L Rev 663, 677–81 (2001) (explaining why theories of state empowerment based on contract and preemption fail to sufficiently authorize state agencies to comply with the Telecommunications Act).
• 23See Jessica Bulman-Pozen, Executive Federalism Comes to America, 102 Va L Rev 953, 994–1001 (2016) (describing how state and federal bureaucracies negotiate and interact to create beneficial, state-differentiated national policy in some cooperative-federalism schemes).
• 24See Gerken, 124 Harv L Rev at 65 (cited in note 22) (noting that a strong national majority can reverse local majorities when it chooses, mitigating costs associated with decentralizing power to states or local governments that oppress racial minorities “in defiance of a national majority”). I have previously questioned the strength of this national “trump card.” See Weinstein-Tull, 117 Colum L Rev at 889 & n 311 (cited in note 7).
• 25It sometimes relegates questions about systemic stratification to an aside. See, for example, Roderick M. Hills Jr, The Political Economy of Cooperative Federalism: Why State Autonomy Makes Sense and “Dual Sovereignty” Doesn’t, 96 Mich L Rev 813, 890 (1998) (conceding that his argument against federal commandeering might fail if “state and local governments systematically underrepresent racial minorities or other disadvantaged or disorganized constituencies even more than the national government”). This exception would seem to swallow the rule for some areas of the law.
• 26Gerken, 124 Harv L Rev at 59 (cited in note 22).
• 27Id at 66.
• 28See, for example, recent voting rights cases finding that states intentionally discriminated against racial minorities. Perez v Abbott, 267 F Supp 3d 750, 794–96 (WD Tex 2017), affd in part, revd in part, and remanded, 138 S Ct 2305 (2018); North Carolina State Conference of the NAACP v McCrory, 831 F3d 204, 215 (4th Cir 2016).
• 29Professors Abram Chayes and Owen Fiss did pioneering work on the characteristics, promise, and pitfalls of these remedies. See generally Abram Chayes, The Role of the Judge in Public Law Litigation, 89 Harv L Rev 1281 (1976); Owen M. Fiss, The Civil Rights Injunction (Indiana 1978); Owen M. Fiss, 93 Harv L Rev 1 (cited in note 19).
• 30See Part IV.
• 31Terry Sanford, Storm over the States 20–21 (McGraw-Hill 1967).
• 32Id at 22, citing Harold J. Laski, The Obsolescence of Federalism, 39 The New Republic 367 (May 3, 1939). See also Alice M. Rivlin, Reviving the American Dream: The Economy, the States & the Federal Government 92 (Brookings 1992) (describing how the states were unable to cope with the Great Depression and how, especially in comparison to a federal government augmented from its experience in World War II, some saw states as “anachronisms that might eventually fade from the American governmental scene”).
• 33Ann O’M. Bowman and Richard C. Kearney, The Resurgence of the States 14 (Prentice-Hall 1986).
• 34William H. Young, The Development of the Governorship, 31 State Government 178, 178 (1958).
• 35Rivlin, Reviving the American Dream at 93 (cited in note 32). For a history of gubernatorial powers generally, see Seifter, 131 Harv L Rev at 492–99 (cited in note 20).
• 36347 US 483 (1954) (“Brown I”).
• 37Id at 495.
• 38Bowman and Kearney, Resurgence of the States at 1–2 (cited in note 33). See also Ross Sandler and David Schoenbrod, Democracy by Decree: What Happens When Courts Run Government 17 (Yale 2003) (“The image of state and local officials as villains in need of judicial correction began with the South but spread to officials across the country during the era of antiwar demonstrations and urban riots.”).
• 39This was especially true when, the year after Brown I, the Supreme Court issued its opinion in Brown v Board of Education of Topeka, 349 US 294 (1955) (“Brown II”), which took a more cautionary approach to the actual act of desegregation. The Court held that local authorities must undertake “good faith implementation” of desegregation, based on “local conditions,” “practical flexibility,” and “a facility for adjusting and reconciling public and private needs.” Id at 299–300. States must desegregate their schools not immediately, the Court held, but “with all deliberate speed.” Id at 301. Brown II emboldened state officers to continue resisting desegregation. “The overall message [of Brown II] to the South seemed to be that it could take as long as it wanted to desegregate schools. To many, this meant never.” John A. Kirk, “Massive Resistance and Minimum Compliance”: The Origins of the 1957 Little Rock School Crisis and the Failure of School Desegregation in the South, in Clive Webb, ed, Massive Resistance: Southern Opposition to the Second Reconstruction 76, 81 (Oxford 2005).
• 40See Mary L. Dudziak, The Little Rock Crisis and Foreign Affairs: Race, Resistance, and the Image of American Democracy, 70 S Cal L Rev 1641, 1712–14 (1997).
• 41Id at 1714.
• 42See Todd E. Pettys, Our Anticompetitive Patriotism, 39 UC Davis L Rev 1353, 1389 (2006) (noting that, after the 1950s and ’60s, “the states worked hard to regain the people’s trust, such as by improving their tax systems, professionalizing their staffs, and strengthening their governors’ powers”).
• 43See Bowman and Kearney, Resurgence of the States at 54–65 (cited in note 33).
• 44Thad Beyle and Margaret Ferguson, Governors and the Executive Branch, in Virginia Gray and Russell L. Hanson, eds, Politics in the American States: A Comparative Analysis 192, 207 (CQ 9th ed 2008).
• 45See, for example, James K. Conant, In the Shadow of Wilson and Brownlow: Executive Branch Reorganization in the States, 1965 to 1987, 48 Pub Admin Rev 892, 898 (1988) (noting that, although the recent spate of state bureaucratic reforms increased gubernatorial power over state bureaucracies, the economic efficiency and administrative effectiveness of those reforms remain unknown).
• 46Council of State Governments, 49 Book of the States at 409 (cited in note 7).
• 47Id.
• 48See Bowman and Kearney, Resurgence of the States at 67 (cited in note 33).
• 49See Gluck, 121 Yale L J at 552 (cited in note 22) (noting that state bureaucratic evolution—including the “dramatic increase in the professionalization of state agencies”—is both “widely noted” and “made state administrators more attractive and competent implementation partners”).
• 50See Weiser, 79 NC L Rev at 669 (cited in note 22).
• 51See id at 669–70.
• 52Roderick M. Hills Jr, The Eleventh Amendment as Curb on Bureaucratic Power, 53 Stan L Rev 1225, 1236 (2001).
• 53Id at 1236–37.
• 54Gluck, 121 Yale L J at 552 (cited in note 22). See also John P. Dwyer, The Practice of Federalism under the Clean Air Act, 54 Md L Rev 1183, 1221–24 (1995) (noting that, even by 1970, state bureaucracies were incapable of developing effective environmental policies, but that, since then, “federal funding and federal environmental legislation have promoted the development and growth of state environmental bureaucracies and expertise”).
• 55See Russell L. Hanson, Intergovernmental Relations, in Virginia Gray and Herbert Jacob, eds, Politics in the American States: A Comparative Analysis 35, 54 (CQ 6th ed 1996).
• 56Bowman and Kearney, Resurgence of the States at 11–12 (cited in note 33).
• 57Id at 12.
• 58Rivlin, Reviving the American Dream at 101–02 (cited in note 32).
• 59Sheryll D. Cashin, Federalism, Welfare Reform, and the Minority Poor: Accounting for the Tyranny of State Majorities, 99 Colum L Rev 552, 570 (1999).
• 60Eskridge, 100 Mich L Rev at 2319 (cited in note 18).
• 61Id.
• 62For a general discussion of this line of cases, see Kenji Yoshino, The New Equal Protection, 124 Harv L Rev 747, 768–73 (2011) (discussing the history of Congress’s Commerce Clause and Fourteenth Amendment enforcement authority and the cases limiting them); Robert C. Post and Reva B. Siegel, Legislative Constitutionalism and Section Five Power: Policentric Interpretation of the Family and Medical Leave Act, 112 Yale L J 1943, 1952–66 (2003) (discussing the Court’s constriction of Congress’s Fourteenth Amendment lawmaking authority).
• 63See City of Boerne v Flores, 521 US 507, 511 (1997) (striking down the Religious Freedom Restoration Act of 1993).
• 64See Kimel v Florida Board of Regents, 528 US 62, 67 (2000) (striking down the provision of the Age Discrimination in Employment Act of 1967 that created state liability for state-based employment discrimination on the basis of age).
• 65See Board of Trustees of the University of Alabama v Garrett, 531 US 356, 360 (2001) (striking down the provision of the Americans with Disabilities Act of 1990 that created state liability for state-based discrimination on the basis of disability).
• 66See United States v Morrison, 529 US 598, 619–27 (2000) (striking down the provision of the Violence against Women Act of 1994 that created civil liability for gender-motivated violence as beyond Congress’s Fourteenth Amendment enforcement power).
• 67The Court struck down the Gun-Free School Zones Act of 1990, United States v Lopez, 514 US 549, 551, 567–68 (1995), and a provision of the Violence against Women Act of 1994, Morrison, 529 US at 607–19, as exceeding Congress’s Commerce Clause authority. For a general discussion of this line of cases and the ways in which Lopez and Morrison diminished Congress’s lawmaking authority, see John O. McGinnis, Reviving Tocqueville’s America: The Rehnquist Court’s Jurisprudence of Social Discovery, 90 Cal L Rev 485, 511–16 (2002).
• 68See Printz v United States, 521 US 898, 925–33 (1997) (striking down a part of the Brady Handgun Violence Prevention Act that required state and local law enforcement officers to conduct a background check on prospective handgun buyers); New York v United States, 505 US 144, 151–54, 161 (1992) (holding that Congress cannot compel a state “to enact and enforce a federal regulatory program” and striking down a federal statute that forced states to take title to low-level radioactive waste unless the state was able to dispose of that waste—either itself or through an interstate compact—by a certain date), quoting Hodel v Virginia Surface Mining & Reclamation Association, Inc, 452 US 264, 288 (1981).
• 69See National Federation of Independent Business v Sebelius, 567 US 519, 575–85 (2012) (holding that Congress could not coerce the states to expand their Medicaid programs by threatening to take away their existing Medicaid funding). See also Fahey, 128 Harv L Rev at 1587–89 (cited in note 21) (describing the development of the coercion doctrine).
• 70See Nixon v Missouri Municipal League, 541 US 125, 129, 140 (2004) (holding that the federal Telecommunications Act of 1996 did not preempt a Missouri law that regulated its subdivisions because of the “working assumption that federal legislation threatening to trench on the States’ arrangements for conducting their own governments should be treated with great skepticism, and read in a way that preserves a State’s chosen disposition of its own power”); Gregory v Ashcroft, 501 US 452, 460–61 (1991) (holding that federal law disrupting the traditional federal-state balance—which includes permitting the states to structure their government as they see fit—must make a “plain statement” of its intention to do so because “[t]hrough the structure of its government . . . a State defines itself as a sovereign”). See also Nestor M. Davidson, Cooperative Localism: Federal-Local Collaboration in an Era of State Sovereignty, 93 Va L Rev 959, 984–90 (2007) (describing the doctrine of state control over its subdivisions).
• 71See Fiss, 93 Harv L Rev at 23–24 (cited in note 19). Scholars termed these documents “civil rights injunctions” or “structural injunctions,” and they emerged from the desegregation cases that followed the Supreme Court’s decision in Brown I. Structural reform cases spread from school desegregation to a wider variety of state institutions: police, prisons, psychiatric hospitals, and public assistance programs. See id at 2–4 (cited in note 19). See also Sandler and Schoenbrod, Democracy by Decree at 45–46, 63 (cited in note 38) (describing two consent decrees that totaled 515 pages in a lawsuit designed to reform New York City’s special education program).
• 72See Fiss, 93 Harv L Rev at 3–4 (cited in note 19).
• 73See id at 4–5 (describing how the Burger Court mounted a “counterassault” against structural injunctions); Theodore Eisenberg and Stephen C. Yeazell, The Ordinary and the Extraordinary in Institutional Litigation, 93 Harv L Rev 465, 472–73 (1980) (noting that federalism doctrines that protect state institutions from federal interference “flourishe[d] in federal courthouses”).
• 74Charles F. Sabel and William H. Simon, Destabilization Rights: How Public Law Litigation Succeeds, 117 Harv L Rev 1016, 1032 (2004).
• 75570 US 529 (2013).
• 76Pub L No 89-110, 79 Stat 437, codified as amended at 52 USC § 10101 et seq.
• 77See Samuel Issacharoff, Is Section 5 of the Voting Rights Act a Victim of Its Own Success?, 104 Colum L Rev 1710, 1712–14 (2004).
• 78Shelby County, 570 US at 537.
• 79Id.
• 80See Lopez v Monterey County, 525 US 266, 282–83 (1999); City of Rome v United States, 446 US 156, 173–78 (1980); Georgia v United States, 411 US 526, 536–39 (1973).
• 81Shelby County, 570 US at 547.
• 82See Fiss, The Civil Rights Injunction at 91–95 (cited in note 29); Chayes, 89 Harv L Rev at 1284, 1288–1304 (cited in note 29); Fiss, 93 Harv L Rev at 28–38 (cited in note 19).
• 83Sabel and Simon, 117 Harv L Rev at 1018 (cited in note 74).
• 84See Paul Gewirtz, Remedies and Resistance, 92 Yale L J 585, 676–80 (1983); William A. Fletcher, The Discretionary Constitution: Institutional Remedies and Judicial Legitimacy, 91 Yale L J 635, 649 (1982).
• 85See Bulman-Pozen and Gerken, 118 Yale L J at 1284–91 (cited in note 22).
• 86See, for example, Weinstein-Tull, 117 Colum L Rev at 847–64 (cited in note 7) (describing the state role in administering federal election, public assistance, indigent defense, prison condition, and education laws); Gluck, 121 Yale L J at 582–94 (cited in note 22) (describing the various state roles in administering federal health care law); Weiser, 79 NC L Rev at 668–81 (cited in note 22) (describing the rise of cooperative-federalism programs that use states to administer federal public assistance, health care, environmental, and telecommunications laws).
• 87Throughout, I use “federal right” and “federal law” largely interchangeably. This Article is primarily concerned with federal laws that seek to enfranchise, empower, or provide a benefit or civil right.
• 88See notes 9–10 and accompanying text.
• 89See Dwyer, 54 Md L Rev at 1184–85, 1209–10 (cited in note 54) (noting that the Clean Air Act and other federal environmental statutes require state action in exchange for federal funding); Margaret H. Lemos, State Enforcement of Federal Law, 86 NYU L Rev 698, 715–16 (2011).
• 90See Weiser, 79 NC L Rev at 668–69 (cited in note 22) (noting that most federal public assistance programs offer states funding in exchange for state administration of the program).
• 91Like federal public assistance laws, for example. See id.
• 92See, for example, the Americans with Disabilities Act of 1990, Pub L No 101-336, 104 Stat 327, codified at 42 USC § 12101 et seq.
• 93I share Professors Theodore Eisenberg and Stephen Yeazell’s skepticism that the source of the obligation should affect the extent of a court’s enforcement authority. See Eisenberg and Yeazell, 93 Harv L Rev at 488 (cited in note 73):

  To the extent that institutional litigation is viewed as questionable because judges lack the capacity and resources to make policy for large institutions and to supervise their administration, it is difficult to understand why the existence of a relevant statute would make a difference. A legislative enactment does not magically enhance judges’ administrative and policymaking abilities.

  That said, the source of the obligation will determine the enforcement structure. Whereas a constitutional guarantee is likely to be enforced by private actors, federal statutes may be enforced privately, cooperatively through interactions with federal agencies, or through federal enforcement litigation. The nature of the remedy will vary as well if the federal statutes specify what remedial measures are appropriate.

• 94See Pressman and Wildavsky, Implementation at 133–35 (cited in note 14).
• 95Id at 134.
• 96I draw here from Professors Daniel Farber and Anne Joseph O’Connell’s deep research on federal interagency conflicts. They provide a helpful typology of four kinds of federal administrative relationships: hard hierarchical, soft hierarchal, monitoring, and symmetrical. See Farber and O’Connell, 105 Cal L Rev at 1389 (cited in note 20). I adapt their framework to focus on hierarchical and nonhierarchical relationships.
• 97In a case against Alabama for failing to transmit ballots to military and overseas voters, the state claimed it was powerless to force local election authorities to comply with the federal law. It claimed that “[i]f a local official refuses to cooperate or provide information to the Secretary of State, the Secretary has no authority to compel the action of a local official.” The state claimed that “persuasion” was sometimes effective, “but the fact remains that the Secretary cannot be in 67 counties at once, and cannot compel a local official to mail a ballot by a particular date.” State Defendants’ Response to the United States’ Motion for Summary Judgment, Declaratory Judgment, and Permanent Injunctive Relief, United States v Alabama, No 2:12-cv-00179, *5 (MD Ala filed Dec 4, 2013).
• 98When both are elected, cabinet-level officials, for example.
• 99See Council of State Governments, 49 Book of the States at 189–94 (cited in note 7).
• 100Id.
• 101See Seifter, 131 Harv L Rev at 512–15 (cited in note 20) (describing the legal ambiguities of state independent agencies, their relationships with governors, and the lack of contemporary literature on these agencies). For a similar discussion at the federal level, see generally Jennifer Nou, Governing Agencies (unpublished manuscript, 2017) (on file with author) (describing the many types of governing structures of independent agencies at the federal level).
• 102Most scholarship on interagency disputes has looked at the relationship between governors and state attorneys general. See generally, for example, William P. Marshall, Break Up the Presidency? Governors, State Attorneys General, and Lessons from the Divided Executive, 115 Yale L J 2446 (2006); Note, Appointing State Attorneys General: Evaluating the Unbundled State Executive, 127 Harv L Rev 973 (2014). That relationship does tend to create high-profile disputes, in part because the two are prominent political offices with built-in structural incentives to compete. See Marshall, 115 Yale L J at 2453 (noting that the state attorney general position is often a stepping stone to the governorship and that political incentives may encourage disputes between those two officers whether they share a political party or not).

  More recent scholarship has examined how some cooperative-federalism programs require state actors to work together to consent to the federal program in the first place and how the federal government works around conflicts between state actors to implement the Affordable Care Act (ACA). See Fahey, 128 Harv at 1572–82 (cited in note 21) (describing the variety of consent procedures employed by cooperative-federalism programs that require the participation of various state actors); Fahey, 125 Yale L J F at 59–64 (cited in note 21) (describing how federal bureaucrats found ways to work with state insurance commissioners to expand state administration of the ACA even in the face of governors wanting no part in the program).

• 103See Part II.B.
• 10452 USC § 20509.
• 105National Conference of State Legislatures, Election Administration at State and Local Levels (cited in note 11) (noting that twenty-four states have designated their secretary of state as the state’s chief election official).
• 106See 52 USC §§ 20504 (motor vehicle offices), 20506 (public assistance offices and disability services offices).
• 107Council of State Governments, 49 Book of the States at 189–94 (cited in note 7).
• 108Id.
• 109Alternatively, California could require its public assistance and DMV offices to transmit those voter registration forms to their central state offices, which would then transmit them to local county registrars, most likely through the state secretary of state’s office. That pathway has its own coordination demand.
• 110Interview with Raúl Luévano Macías, Attorney, ACLU of Northern California (Sept 16, 2016) (on file with author) (“Macías Interview”).
• 111See Weiser, 79 NC L Rev at 668–69 (cited in note 22) (describing the funding structure of cooperative-federalism programs).
• 112The NVRA provides no funding. The Help America Vote Act, Pub L No 107-252, 116 Stat 1666, provides some election administration funding, but it is “quite limited”; states and local governments bear the financial burden of funding elections, including compliance with the NVRA. See Alec C. Ewald, The Way We Vote: The Local Dimension of American Suffrage 4 (Vanderbilt 2009).
• 113See Drinan, 33 NYU Rev L & Soc Change at 429–30, 464–66 (cited in note 6) (noting that problems with indigent defense stem from funding issues, which are problems of state legislatures). States take a variety of approaches: nineteen states require their local governments to either fully fund or provide most of the funding for indigent criminal defense; fewer than half of the states fully fund their indigent defense systems. Holly R. Stevens, et al, State, County and Local Expenditures for Indigent Defense Services Fiscal Year 2008, *5 (American Bar Association, Nov 2010), archived at http://perma.cc/48Q5-S64Y.
• 114Office of the State Public Defender Commission, Montana Public Defenders Commission Propose Funding Help (Great Falls Tribune, Dec 29, 2016), archived at http://perma.cc/ZW5U-LHXB.
• 115Public Defender Commission Membership (Office of the State Public Defender, Oct 2016), archived at http://perma.cc/9Z5Q-AKRL.
• 116See Office of the State Public Defender Commission, Montana Public Defenders Commission Propose Funding Help (cited in note 114) (“[I]ncreases in workload are not addressed through corresponding increases in funding, staff or attorneys.”).
• 117Id.
• 118See, for example, a letter written by the director of the Missouri State Public Defender (MSPD) system to the Missouri governor, stating that MSPD was unable to perform its job without additional funding and taking the extraordinary and elegant step of appointing the governor as a public defender. Matt Ford, A Governor Ordered to Serve as a Public Defender (The Atlantic, Aug 4, 2016), archived at http://perma.cc/NQU2-SCGK. See also, for example, Idaho Governor Butch Otter, who publicly asked his state legislature for funding to provide adequate indigent defense counsel after being sued for statewide violations of the Sixth Amendment. KBOI News Staff, Gov. Butch Otter Seeking Nearly 8 Percent Increase in Education Budget (KBOI News, Jan 11, 2016), archived at http://perma.cc/EL2K-3ANG (“Please join me in a commitment to ensuring that all Idaho citizens in every one of our 44 counties can avail themselves of this fundamental constitutional right. My budget recommends $5 million to implement the changes that you approve.”).
• 119See Weinstein-Tull, 117 Colum L Rev at 870–71 (cited in note 7).
• 120See id at 871–72.
• 121See generally id (describing how states delegate responsibilities to local governments, cease monitoring local compliance with those responsibilities, and then disclaim ultimate responsibility themselves).
• 122See generally id (describing local administration of state responsibilities in the election law, indigent defense, prison administration, public assistance, and education contexts). See also Briffault, 90 Colum L Rev at 7 (cited in note 7) (“The formal legal status of a local government in relation to its state is summarized by the three concepts of ‘creature,’ ‘delegate’ and ‘agent.’”); Aaron Saiger, Local Government as a Choice of Agency Form, 77 Ohio St L J 423, 431–33 (2016) (describing the conceptual similarities between local governments and state agencies).
• 123Local governments are not mentioned in the Constitution. See Richard Thompson Ford, The Boundaries of Race: Political Geography in Legal Analysis, 107 Harv L Rev 1841, 1864 (1994) (“Local government exists in a netherworld of shifting and indeterminate legal status.”); Joan C. Williams, The Constitutional Vulnerability of American Local Government: The Politics of City Status in American Law, 1986 Wis L Rev 83, 85 (“This Article’s basic thesis concerning the constitutional vulnerability of cities begins from the fact that cities—unlike the states or federal government—have no set place in the American constitutional structure.”).
• 124See David J. Barron, The Promise of Cooley’s City: Traces of Local Constitutionalism, 147 U Pa L Rev 487, 491 (1999).
• 125See Hunter v City of Pittsburgh, 207 US 161, 178 (1907) (noting that local governments are “subdivisions of the State, created as convenient agencies for exercising such of the governmental powers of the State as may be entrusted to them”).
• 126Professor Richard Schragger has described the existence of a “‘shadow doctrine’ that treats localities as sovereign political entities entitled to constitutional protection.” Richard C. Schragger, Reclaiming the Canvassing Board: Bush v. Gore and the Political Currency of Local Government, 50 Buff L Rev 393, 396 (2002). See also Briffault, 90 Colum L Rev at 85–86 (cited in note 7):

  Without according local governments formal constitutional rights against their states, the Court has affirmed the importance of localism in our political culture. The Court has treated localities as active, locally responsive governments, not just administrative arms of the states. The Court has endorsed the appropriateness of local decision-making and has frequently promoted localism and protected the considerable autonomy local governments enjoy under state law.

• 127Richard Briffault, “What About the ‘Ism’?” Normative and Formal Concerns in Contemporary Federalism, 47 Vand L Rev 1303, 1318 (1994).
• 128Id.
• 129See generally Dale Krane, Platon N. Rigos, and Melvin B. Hill Jr, Home Rule in America: A Fifty-State Handbook (CQ 2001) (describing the wide variety of state-local relationships in America, varying from highly centralized states that largely retain power over local governments to highly decentralized states that delegate broad powers to independent local governments).
• 130See, for example, Heather Hahn, et al, A Descriptive Study of County- versus State-Administered Temporary Assistance for Needy Families Programs *65 (Department of Health and Human Services, May 2015), archived at http://perma.cc/2LKP-FFRQ (noting that about half of the states surveyed delegate their administrative responsibilities under the federal cash assistance program to local governments).
• 131See Weinstein-Tull, 114 Mich L Rev at 778–80 (cited in note 6) (describing extensive state delegations to local government in the elections context).
• 132Pub L No 101-336, 104 Stat 327, codified at 42 USC § 12101 et seq.
• 133See Weinstein-Tull, 117 Colum L Rev at 852–55 (cited in note 7). In recent years, for example, California has given county jails jurisdiction over certain state prisoners in an effort to reduce the state prison population. See Joan Petersilia, California Prison Downsizing and Its Impact on Local Criminal Justice Systems, 8 Harv L & Pol Rev 327, 332–34 (2014) (describing “realignment” and detailing the number of state prisoners handled by county jails as a consequence of this initiative).
• 134As noted above, nineteen states require their local governments to provide all or most of the funding for indigent criminal defense, and fewer than half of the states fully fund their indigent defense system. See note 113. Even states that fully or partly fund their indigent defense systems delegate significant administrative responsibilities to their local governments. See generally Stevens, et al, State, County and Local Expenditures for Indigent Defense Services Fiscal Year 2008 (cited in note 113).
• 135See Weinstein-Tull, 117 Colum L Rev at 847–64 (cited in note 7) (describing these state arguments in a wide variety of federal policy areas).
• 136For example, the Food Stamp Act accounts for states that decentralize their system of public assistance administration by broadly defining “state agency” under the statute to include “the local offices thereof, which ha[ve] the responsibility for the administration of the federally aided public assistance programs within such State.” 7 USC § 2012(s). Medicaid imposes monitoring requirements on states that delegate administrative responsibilities to local governments. See 42 CFR § 435.903.
• 137622 F3d 1058 (9th Cir 2010).
• 138Id at 1063–64, 1074.
• 139Pub L No 107-252, 116 Stat 1666.
• 14052 USC § 21083(a).
• 14152 USC § 21082(a).
• 142See, for example, Duncan v Michigan, 832 NW2d 761, 765–66, 778 (Mich App 2013) (describing the procedural history of a Sixth Amendment case against the state of Michigan for local noncompliance in which resolution of state-local coordination difficulties caused a trial on the merits to be delayed by almost six years).
• 143See Weinstein-Tull, 117 Colum L Rev at 866–68 (cited in note 7).
• 144See Seifter, 131 Harv L Rev at 525 (cited in note 20).
• 145Id at 523–25.
• 146Pew Charitable Trusts, Measuring Motor Voter *1 (May 2014), archived at http://perma.cc/L4PK-UWKK.
• 147Id at *2.
• 148Id.
• 149See id.
• 150Presidential Commission on Election Administration, American Voting Experience at *17, 30 (cited in note 6) (calling the NVRA the “most often ignored” election law statute).
• 151Pew Charitable Trusts, Measuring Motor Voter at *7 (cited in note 146).
• 152Id; Presidential Commission on Election Administration, American Voting Experience at *30 (cited in note 6) (noting that Michigan “seamlessly integrate[s] the Motor Voter transaction into the DMV driver’s license application program in such a manner as to keep a large number of voter records current”).
• 153See notes 104–10 and accompanying text.
• 154Macías Interview (cited in note 110).
• 155Joanna C. Schwartz, Introspection through Litigation, 90 Notre Dame L Rev 1055, 1060.
• 156Id (noting that the “presumption” that agency leaders have access to the information necessary to diagnose noncompliance “ignores the facts of institutional life”).
• 157Id at 1060–61 (arguing that litigation against complex organizations can be revelatory information-gathering exercises even for the organization itself).
• 158See generally Freeman and Rossi, 125 Harv L Rev 1131 (cited in note 20).
• 159See Seifter, 131 Harv L Rev at 526–27 (cited in note 20).
• 160See Farber and O’Connell, 105 Cal L Rev at 1436–37 (cited in note 20); Rachel E. Barkow, Insulating Agencies: Avoiding Capture through Institutional Design, 89 Tex L Rev 15, 27–30 (2010) (discussing for-cause and at-will removal provisions).
• 161563 US 247 (2011) (“VOPA v Stewart”).
• 162Pub L No 106-402, 114 Stat 1677, codified at 42 USC § 15001 et seq.
• 163Pub L No 99-319, 100 Stat 478, codified as amended in various sections of Title 42.
• 164VOPA v Stewart, 563 US at 250–51.
• 165Id at 251–52.
• 166Id at 252.
• 167See Virginia v Reinhard, 2008 WL 2795940, *1 (ED Va), revd, Virginia v Reinhard, 568 F3d 110 (4th Cir 2009), revd and remd, VOPA v Stewart, 563 US 247 (2011).
• 168VOPA v Stewart, 563 US at 251–52.
• 169See Va Code § 37.2-301 (stating that the governor appoints the commissioner of the Department of Behavioral Health and Developmental Services).
• 170VOPA v Stewart, 562 US at 250.
• 171The parties settled the case on September 6, 2011. Settlement Agreement, Virginia v Stewart, No 3:07-CV-734 (ED Va filed Sept 6, 2011). The complaint was initially filed on December 3, 2007. Complaint and Motion for Preliminary Injunction, Virginia v Reinhard, No 3:07-CV-734 (ED Va filed Dec 3, 2007).
• 172Virginia Board for People with Disabilities, About the Board, archived at http://perma.cc/TNN3-ZWND.
• 173See Hills, 53 Stan L Rev at 1227 (cited in note 52).
• 174See Hanson, Intergovernmental Relations at 51–52 (cited in note 55).
• 175Hills, 53 Stan L Rev at 1227 (cited in note 52).
• 176See Hanson, Intergovernmental Relations at 52 (cited in note 55).
• 177See Hills, 53 Stan L Rev at 1236 (cited in note 52) (“The very independence of state agencies from the control of elected state politicians is frequently a product of federal law, which requires federal grants to be administered by a ‘single agency’ that is insulated from political control.”).
• 178Pub L No 99-410, 100 Stat 924 (1986), as amended by the Military and Overseas Voter Empowerment Act (MOVE Act), Pub L No 111-84, 123 Stat 2318 (2009).
• 17952 USC § 20302(a)(8).
• 180MOVE Act, § 579(c), 123 Stat at 2324.
• 181Specifically, New York agreed to maintain the forty-five-day window between transmission and ballot deadline by transmitting absentee ballots on October 1, 2010 but extending the deadline for receiving the ballots by thirteen days. However, New York failed to transmit the ballots as required. See United States v New York, 2012 WL 254263, *1–2 (NDNY).
• 182Memorandum Decision and Order, United States v New York, No 1:10-CV-1214, *3 (NDNY Feb 9, 2012) (“NY Memorandum Decision and Order”) (noting that the state board of elections “suggests that [it] lacks authority to recommend any necessary modifications to the resulting calendar”).
• 183United States v New York, 2012 WL 254263 at *2.
• 184Id.
• 185Id:

  [T]he Senate urges the court to consider the economic implications of the primary date, the hardship of candidates to obtain signatures in the winter months, and that June is at the end of the legislative session. More specifically, the Senate points out that a June primary would force its members to have to weigh their elected responsibilities against the need for political presence in their district.

• 186See NY Memorandum Decision and Order at *2–3 (cited in note 182).
• 187Motion by the New York State Senate to Intervene and Memorandum in Support Thereof, United States v New York, No 1:10-CV-1214, *13–15 (NDNY filed Oct 11, 2011) (available on Westlaw at 2011 WL 5134793) (“NY Senate Motion to Intervene”).
• 188NY Memorandum Decision and Order at *2 (cited in note 182).
• 189NY Senate Motion to Intervene at *13–15 (cited in note 187).
• 190See Supplemental Remedial Order, United States v New York, No 1:10-CV-1214, *1–2 (NDNY Oct 29, 2015); Supplemental Remedial Order, United States v New York, No 1:10-CV-1214, *1–2 (NDNY Dec 12, 2013); United States v New York, 2012 WL 254263 at *3.
• 191NY Memorandum Decision and Order at *5 (cited in note 182) (emphasis added).
• 192See Farber and O’Connell, 105 Cal L Rev at 1378–84 (cited in note 20) (providing numerous examples of federal interagency conflict).
• 193I am oversimplifying conflict resolution at the federal level. See id. But even so, federal agencies have, at the very least, a theoretical process and capacity for resolution, whereas states may not.
• 194But they do sometimes require the states themselves to designate one state actor as responsible for administration. See, for example, 52 USC § 20509 (“Each State shall designate a State officer or employee as the chief State election official to be responsible for coordination of State responsibilities under this chapter.”).
• 195See Jacob E. Gersen, Overlapping and Underlapping Jurisdiction in Administrative Law, 2006 S Ct Rev 201, 207–16.
• 196See, for example, the Sixth Amendment right to counsel or the ADA.
• 197See Krane, Rigos, and Hill, Home Rule in America at 4 (cited in note 129) (“[E]ven though state courts typically hold the state-local relationship to be unitary and hierarchical, the political reality is that the relationship is more complicated.”); David J. Barron, A Localist Critique of the New Federalism, 51 Duke L J 377, 393 (2001) (“Local autonomy—or, at least, something widely perceived to be local autonomy—is alive and well under state law despite an overwhelming state constitutional premise that localism is to be the exception rather than rule.”).
• 198Class Action Complaint, Bairefoot v City of Beaufort, No 9:17-cv-02759, *1 (D SC filed Oct 11, 2017).
• 199See SC Code Ann § 14-25-5.
• 200Timothy Williams, Courts Sidestep the Law, and South Carolina’s Poor Go to Jail (NY Times, Oct 12, 2017), online at http://www.nytimes.com/2017/10/12/us/south-carolina-jail-no-lawyer.html (visited Apr 20, 2018) (Perma archive unavailable).
• 201Id. We know so little about state and local courts. See generally Stephen C. Yeazell, Courting Ignorance: Why We Know So Little about Our Most Important Courts, 143 Daedalus 129 (2014). My future work will explore state judicial hierarchies as administrators of federal rights and local court reform. It will explore different kinds of state-level oversight of local courts, how effectively they play the role of overseer, what local court misbehavior gets lost in the state judicial organizational structure, and what kinds of judicial administrative structures we would want to see constructed.
• 202See Abbe R. Gluck, Our [National] Federalism, 123 Yale L J 1996, 2037–38 (2014).
• 203Pub L No 95-113, 91 Stat 913, codified as amended at 7 USC § 2011 et seq.
• 204972 F2d 529 (4th Cir 1992).
• 205Id at 529–30.
• 206Brief of Appellant, Jackson v Robertson, No 91-2580, *8–17 (4th Cir filed Sept 11, 1992).
• 207Id at *9–10.
• 208See notes 61–81 and accompanying text.
• 209See Gregory v Ashcroft, 501 US 452, 460 (1991) (“Through the structure of its government . . . a State defines itself as a sovereign.”).
• 210See Weinstein-Tull, 117 Colum L Rev at 883–86 (cited in note 7) (collecting and describing sovereignty-based state arguments in suits alleging violations of federal law).
• 211See Nicholas Bagley, Federalism and the End of Obamacare, 127 Yale L J F 1, 7 (2017) (“The Voting Rights Act and other laws adopted pursuant to the Reconstruction Amendments, for example, reflect the fear that elected officials and voters in many states, especially in the South, will be systematically inattentive to the interests of minority groups.”).
• 212This Article highlights examples from California, New York, Alabama, Illinois, and Rhode Island.
• 213Professor Jessica Bulman-Pozen argues that partisanship explains much state resistance to federal laws. Jessica Bulman-Pozen, Partisan Federalism, 127 Harv L Rev 1077, 1082 (2014). That division does not seem to explain bureaucratic undermining.
• 214Coordination-based state noncompliance cases have arisen in blue states like California, New York, Illinois, and Rhode Island, as well as red states like Alabama, Louisiana, and Oklahoma.
• 215See notes 94–95 and accompanying text.
• 216Viewing state actors as rational helps to understand how and whether coordination happens. See Daniel B. Rodriguez, The Positive Political Dimensions of Regulatory Reform, 72 Wash U L Q 1, 43 (1994).
• 217The power of a state to bring a federal suit against a local government for noncompliance with a federal right is as yet undetermined. See United States v Missouri, 535 F3d 844, 851 (8th Cir 2008) (noting that the NVRA provided no cause of action for Missouri to sue its noncompliant local election agencies and that the plaintiff could not force Missouri to do so).
• 218Douglas R. Hess, Michael J. Hanmer, and David W. Nickerson, Encouraging Local Compliance with Federal Civil Rights Laws: Field Experiments with the National Voter Registration Act, 76 Pub Admin Rev 165, 172 (2015).
• 219Id.
• 220See notes 181–91 and accompanying text.
• 221United States v New York, 2012 WL 254263, *1–2 (NDNY).
• 222Id.
• 223See Margo Schlanger, Plata v. Brown and Realignment: Jails, Prisons, Courts, and Politics, 48 Harv CR–CL L Rev 165, 185–88 (2013) (describing the process by which Governor Brown convinced local governments to support his prison “realignment” effort).
• 224Id at 181.
• 225See Edelman v Jordan, 415 US 651, 678 (1974).
• 226See Ex parte Young, 209 US 123, 158–59, 168 (1908).
• 227Professors Charles Sabel and William Simon argue that the consequence to states for noncompliance with federal rights is a “loss of independence and increased uncertainty” under a court-ordered remedy. Sabel and Simon, 117 Harv L Rev at 1055 (cited in note 74).
• 228See Sean McElwee, Why the Voting Gap Matters *2 (Demos, 2014), archived at http://perma.cc/TD5G-UY3L.
• 229Laurence E. Lynn Jr, The State and Human Services: Organizational Change in a Political Context 178 (MIT 1980).
• 230See Bertrall L. Ross II and Su Li, Measuring Political Power: Suspect Class Determinations and the Poor, 104 Cal L Rev 323, 345–46 (2016) (noting that “[r]ecent studies of the comparative political influence of the poor have found evidence that the poor have significantly less influence on public policy and legislators’ roll call votes than members of other income classes” and collecting sources). After conducting empirical analysis, Professor Bertrall Ross and Su Li found “no evidence that the poor influenced their Congressional representatives’ support for legislative actions favorable to the group. Instead, [their] findings suggest that something other than political power, perhaps ideology or partisanship, motivates legislative actions favorable to the poor.” Id at 373–74. See also Nicholas O. Stephanopoulos, Political Powerlessness, 90 NYU L Rev 1527, 1577–79 (2015) (citing sources).
• 231As Professor Ian Haney López has argued, analyzing stratification is best done by examining class and race as interconnected: “[R]ace and class in the United States inextricably interdigitate such that neither can be engaged without sustained attention to the other.” Ian F. Haney López, Post-Racial Racism: Racial Stratification and Mass Incarceration in the Age of Obama, 98 Cal L Rev 1023, 1051 (2010). See also Cheryl I. Harris, Fisher’s Foibles: From Race and Class to Class Not Race, 64 UCLA L Rev Discourse 648, 658 (2017) (noting that, in the context of critiquing a “class not race” approach to affirmative action, “race and class are co-constitutive yet distinct” and that “race has structured class relations and class relations have structured race, albeit in ways that are not symmetrical or constant”); id at 653–54 n 17 (collecting academic and popular sources that discuss the relationship between class and race).
• 232Stephanopoulos, 90 NYU L Rev at 1573 (cited in note 230), citing John D. Griffin and Brian Newman, Minority Report: Evaluating Political Equality in America 64–69 (Chicago 2008). See also Nicholas Stephanopoulos, The False Promise of Black Political Representation (The Atlantic, June 11, 2015), archived at http://perma.cc/5SKA-K8YW (“[O]n vital issues like crime, welfare, and housing[,] [blacks’ policy preferences] are too often ignored by elected officials when they conflict with whites’ preferences.”).
• 233Miller, 44 L & Society Rev at 835 (cited in note 6).
• 234See Hills, 96 Mich L Rev at 828 (cited in note 25) (“[E]rosion of political accountability is endemic to all forms of cooperative federalism; whenever the federal government induces states to act, whether with block grants or categorical grants, there is a considerable risk that voters will be confused about which level of government imposed the regulatory burdens of the program.”).
• 235Miriam Seifter, Further from the People? The Puzzle of State Administration, 93 NYU L Rev 107, 109–10 (2018) (citation omitted).
• 236Id at 131–34.
• 237Id at 141–42.
• 238Cashin, 99 Colum L Rev at 583 (cited in note 59). Professor Jon Michaels has also noted that states can be unreliable administrators of federal welfare reform efforts. See Jon Michaels, Deforming Welfare: How the Dominant Narratives of Devolution and Privatization Subverted Federal Welfare Reform, 34 Seton Hall L Rev 573, 604 (2004).
• 239See Richard Briffault, Our Localism: Part II—Localism and Legal Theory, 90 Colum L Rev 346, 354–56, 447–54 (1990).
• 240Id at 452.
• 241Cashin, 99 Colum L Rev at 591–94 (cited in note 59) (describing racist characterizations of black welfare recipients as lazy).
• 242Id at 592–93 (“African-Americans have suffered most when the institutions of American social policy have been parochial, and they have benefited the most when those institutions have been national.”), quoting Robert C. Lieberman, Shifting the Color Line: Race and the American Welfare State 230 (Harvard 1998).
• 243See Michael A. Livermore and Richard L. Revesz, Regulatory Review, Capture, and Agency Inaction, 101 Georgetown L J 1337, 1340 (2013).
• 244See Matthew D. Zinn, Policing Environmental Regulatory Enforcement: Cooperation, Capture, and Citizen Suits, 21 Stan Envir L J 81, 122–23 (2002) (describing state administrative capture in the context of environmental regulations); Cashin, 99 Colum L Rev at 598 (cited in note 59) (“[T]he rent-seeking lobbying efforts of wealthy, well-organized groups are more likely to hold sway in state legislative bodies and, therefore, to undermine possibilities for coalitions that support redistributive spending.”). See also id at 596 (citation omitted):

  The national arena also offers low-income and anti-poverty interest groups the strategic advantage of being able to focus energies on one political forum, with attendant economies of scale. At the national level, research, public education, and advocacy on poverty issues may be undertaken much more cheaply than if such activities are undertaken repeatedly in multiple fora. Moreover, to the extent that the self-interested tendencies of voters are more pronounced at state and local levels, anti-poverty interest groups must work harder at those levels to overcome these political barriers.

• 245See Rossi, 53 Admin L Rev at 560–62 (cited in note 20). But see Philip J. Weiser, Chevron, Cooperative Federalism, and Telecommunications Reform, 52 Vand L Rev 1, 36–37 (1999) (arguing that the structure of cooperative-federalism programs guards against state regulatory capture by introducing competition between state actors and other institutional checks on capture, such as requirements that state agencies conduct objective analyses before taking certain actions).
• 246See Miller, 44 L & Society Rev at 835 (cited in note 6):

  While multiple legislative venues may provide a more open political system in some respects, it also creates a political context that perpetuates racial hierarchy by creating opportunities for highly resourced groups to control the terms of the debate and forcing the less organized onto their terrain, no matter how they may initially frame the problem.

• 247Ford, 107 Harv L Rev at 1845 (cited in note 123). See also id at 1852.
• 248Rachel F. Moran, Whatever Happened to Racism?, 79 St John’s L Rev 899, 900 (2005).
• 249See, for example, Samuel R. Bagenstos, The Structural Turn and the Limits of Antidiscrimination Law, 94 Cal L Rev 1, 4–5 (2006) (describing the key facets of the structural approach to employment discrimination law).
• 250Laurel Eckhouse, Big Data May Be Reinforcing Racial Bias in the Criminal Justice System (Wash Post, Feb 10, 2017), online at http://www.washingtonpost.com/opinions/big
  -data-may-be-reinforcing-racial-bias-in-the-criminal-justice-system/2017/02/10/d63de518-ee3a-11e6-9973-c5efb7ccfb0d_story.html (visited Apr 20, 2018) (Perma archive unavailable).
• 251Andrea Roth, Trial by Machine, 104 Georgetown L J 1245, 1275 (2016) (“Poorly constructed actuarial instruments used in so-called ‘evidence-based sentencing’ can lead judges in jurisdictions without sentencing guidelines to base sentences on questionable and racially skewed proxies for future dangerousness and blameworthiness, such as family stability and length of prior record.”).
• 252See Eckhouse, Reinforcing Racial Bias (cited in note 250).
• 253John Hart Ely, Democracy and Distrust: A Theory of Judicial Review 105–34 (Harvard 1980).
• 254See, for example, Sandler and Schoenbrod, Democracy by Decree at 45–46, 63 (cited in note 38) (describing a lawsuit challenging New York City’s special education program in which the parties adopted two consent decrees that totaled 515 pages).
• 255Sabel and Simon, 117 Harv L Rev at 1024 (cited in note 74).
• 256See Part II.A.
• 257See Gerken, 105 Cal L Rev at 1710 (cited in note 3) (“Academics often unthinkingly blame decentralization for shortfalls in our equality norms. This simplistic formulation ignores the fact that the turn to decentralization is a sign of weakness in the norms themselves. We adopt a decentralized solution only when our national norm is to tolerate shortfalls.”).
• 258See Part IV. As another example from the public assistance context, Cashin has suggested that federal welfare reform should include greater tools for state accountability and standardizing state processes—an option short of recentralizing. Cashin, 99 Colum L Rev at 618–22 (cited in note 59).
• 259For two illuminating works that contemplate and complicate the relationship between monetary penalties and administrative decisionmaking, see generally Joanna C. Schwartz, Myths and Mechanics of Deterrence: The Role of Lawsuits in Law Enforcement Decisionmaking, 57 UCLA L Rev 1023 (2010); Daryl J. Levinson, Making Government Pay: Markets, Politics, and the Allocation of Constitutional Costs, 67 U Chi L Rev 345 (2000).
• 260As I note in Part I, courts increasingly protect states from federal remedial intrusion. See notes 61–81 and accompanying text.
• 261Sabel and Simon, 117 Harv L Rev at 1021 (cited in note 74).
• 262See, for example, Consent Decree, United States v Illinois, No 1:15-cv-02997, *6 (ND Ill Apr 14, 2015) (requiring the state of Illinois and the Illinois Board of Elections to work with local election authorities to transmit absentee ballots to military and overseas voters by a certain date in a case alleging that an Illinois special election would fail to comply with UOCAVA).
• 263See, for example, id at *8:

  [The state and the board of elections agreed to] take such actions as are necessary to assure that all future special elections for Federal office are conducted in accordance with UOCAVA, including proposing legislation and taking any administrative actions needed to alter Illinois’ statutorily imposed timetable for conducting special elections for filling vacancies in the office of United States Representative in Congress. Specifically, the State Board of Elections will recommend amendments to the Election Code as required to enlarge the time period for conducting such elections sufficiently to guarantee that special primary election and special election ballots can be transmitted to UOCAVA voters at least 45 days before the date of the election. . . . Defendants shall file with the Court a status report on this proposed legislation no later than June 2, 2015.

• 264A recent out-of-court settlement between voting rights advocates and Oklahoma state actors, for example, leans heavily on coordination to ensure compliance with the NVRA. The settlement requires the Oklahoma State Election Board to appoint a staff person to coordinate statewide compliance with the NVRA. That coordinator must work with the Oklahoma Department of Human Services, the Oklahoma Department of Health, and the Oklahoma Healthcare Authority to ensure NVRA compliance, including by providing trainings, monitoring compliance, and issuing reminders about voter registration deadlines. Settlement Agreement between the Metropolitan Tulsa Urban League and the Oklahoma State Election Board, *5 (July 28, 2015), archived at http://perma.cc/QL4Q-SW6E. The remedy also requires each piece of the relevant Oklahoma state-level bureaucracy to appoint staff to coordinate between the state offices and their local affiliates. Id at *6–10.
• 265See note 157 and accompanying text.
• 266Just the act of monitoring and creating compliance data can improve compliance. See, for example, Heather K. Gerken, The Democracy Index: Why Our Election System Is Failing and How to Fix It 82–89 (Princeton 2009).
• 267Email from Lisa J. Danetz to Justin Weinstein-Tull (Apr 11, 2016) (on file with author).
• 268Patient Protection and Affordable Care Act, Pub L No 111-148, 124 Stat 119 (2010).
• 269See Gluck, 121 Yale L J at 581–82 (cited in note 22) (describing expansions to Medicaid, “the federal health-insurance program for low-income individuals that is administered in cooperation with the states”).
• 270Cal Election Code § 10 (“The Secretary of State is the chief elections officer of the state.”).
• 271Democrats tend to support efforts to expand voter registration. See Zachary Roth, Voter Registration at the Center of Partisan Voting War (MSNBC, Sept 21, 2015), online at http://www.msnbc.com/msnbc/voter-registration-the-center-the-voting-wars (visited Jan 12, 2018) (Perma archive unavailable) (noting that Democratic states have sought to expand access to voter registration).
• 272Cal Election Code § 2402(a) (designating the secretary of state “the chief state elections official responsible for coordination of the state’s responsibilities under the federal National Voter Registration Act of 1993”).
• 273Covered California: The California Health Benefit Exchange, Peter V. Lee, Executive Director, archived at http://perma.cc/NN7F-M8TD.
• 274Macías Interview (cited in note 110).
• 275Settlement Agreement Regarding Interim Remedial Measures and System Integration of Covered California’s Voter Registration Obligations (Mar 21, 2014), archived at http://perma.cc/6M2W-7Q36 (“Cal Settlement Agreement”). A close reading of the document suggests that Covered California was more enthusiastic about the settlement. For example, the document is signed by all parties, but only Covered California “expressed a desire to ensure that all applicants since October 1, 2013, and moving forward, receive the opportunity to register to vote as required by law.” The document does not include any similar mention by the secretary. Id at *1, 12–13.
• 276Id at *4.
• 277Id.
• 278Id at *4, 7–8.
• 279Cal Settlement Agreement at *4 (cited in note 275).
• 280Id at *5 (requiring Covered California to provide monthly reports to the secretary detailing the number of health care recipients offered voter registration opportunities and requiring the secretary to provide Covered California with monthly reports on the number of completed or incomplete voter registration forms received by health care recipients).
• 281Id at *9–10.
• 282Hurrell-Harring v State, 930 NE2d 217, 219 (NY 2010).
• 283When sued, the State of New York moved to dismiss the charges as nonjusticiable because (1) New York’s decision to delegate indigent defense to local governments was a legislative decision, not remediable by courts; and (2) local governments were indispensable parties that must be joined for litigation to proceed. Memorandum of Law in Support of Defendant’s Motion to Dismiss, Hurrell-Harring v State, No 8866-07, *16–17, 27–28 (NY Sup filed Apr 4, 2008). See also Brief for Respondents, Hurrell-Harring v State, No 2010-0066, *31–36 (NY App filed Dec 4, 2009) (available on Westlaw at 2009 WL 6409872) (arguing, again, that plaintiffs’ claims were nonjusticiable before the court of appeals after prevailing in state trial court).

  New York’s highest court ruled against the state and held that the legislative decision to decentralize provision of indigent defense did not insulate the state from the Sixth Amendment. Hurrell-Harring, 930 NE2d at 227–28 (noting that “[i]t is, of course, possible that a remedy in this action would necessitate the appropriation of funds and perhaps, particularly in a time of scarcity, some reordering of legislative priorities” but also that “this does not amount to an argument upon which a court might be relieved of its essential obligation to provide a remedy for violation of a fundamental constitutional right”).

• 284Stipulation and Order of Settlement, Hurrell-Harring v State, No 8866-07, *1 (NY Sup Oct 21, 2014) (“NY Settlement”).
• 285Id at *2.
• 286Id.
• 287Id at *3.
• 288NY Settlement at Exhibit A (cited in note 284).
• 289Id at *2–3, Exhibit A.
• 290NY Exec Law § 833(1)(a).
• 291NY Exec Law § 833(5).
• 292NY Settlement at *6 (cited in note 284). For another example, see a coordination remedy in a case that the United States brought against the State of Rhode Island, the Rhode Island Board of Elections, and various state actors for violating the NVRA. The consent decree states that the state secretary of state is the chief state elections official and ultimately responsible for coordinating compliance with the NVRA, but it also describes the very specific responsibilities of the state board of elections in administering the state’s elections programs. Consent Decree, United States v Rhode Island, No 1:11-cv-00113-S, *3, at ¶ 5 (D RI Mar 18, 2011):

  Notwithstanding the Secretary of State’s designation as the chief state election official, the Defendant Rhode Island Board of Elections has the powers and duties to: Arrange and make provisions for the registration of voters pursuant to the [NVRA]. . . . The state board shall formulate programs to assist those persons or organizations desiring to register voters and shall provide, pursuant to procedures, rules, and regulations it shall adopt, voter registrations services which may include training sessions, registration materials, manuals and other services for the purpose of registering to vote eligible Rhode Island citizens.

  See also Consent Decree, United States v New York, No 1:10-cv-01214, *2–3, at ¶¶ 4–6 (NDNY Oct 19, 2010) (differentiating and clarifying the election administration roles of the state of New York, the New York State Board of Elections, and local elections officials in a case alleging that New York violated UOCAVA).

• 293Tellingly, many ACLU affiliates focus just as much on legislative change at the state level as they do on litigation against states. Macías Interview (cited in note 110). It is beyond the scope of this Article, but I suspect that for any given federal requirement, the presence of state-enabling legislation that clearly defines the roles of various state actors prevents coordination-based noncompliance.
• 294See Part II.B.
• 295See Part III.
• 296See, for example, Daryl J. Levinson, Foreword: Looking for Power in Public Law, 130 Harv L Rev 31, 117–18 (2016) (describing examples of proxy representation); Shirin Sinnar, Institutionalizing Rights in the National Security Executive, 50 Harv CR–CL L Rev 289, 294–300 (2015) (describing ways that the national security bureaucracy has institutionalized stakeholder representatives inside of the administrative apparatus).
• 297See notes 243–45 and accompanying text.
• 298See notes 262–64 and accompanying text.
• 299See notes 61–81 and accompanying text. Although outside the scope of this Article, criminal law is increasingly bureaucratized as well. See generally Daphna Renan, The Fourth Amendment as Administrative Governance, 68 Stan L Rev 1039 (2016) (describing compliance with the Fourth Amendment as an administrative challenge and sketching an administrative framework for addressing that challenge). See also Stuntz, 121 Harv L Rev at 1980 (cited in note 4) (noting that “America’s justice system already is bureaucratized” and that “the most important bureaucracies—police forces and district attorneys’ offices—are governed by local politics and politicians”).
• 300Oxford English Dictionary, online at http://www.oed.com/view/Entry/163661 (visited Apr 21, 2018) (Perma archive unavailable).
• 301Id.
• 302See, for example, Stephanie Bornstein, Reckless Discrimination, 105 Cal L Rev 1055, 1057–72 (2017) (theorizing a “recklessness” standard for employment discrimination, “arguing for liability where an employer acts with reckless disregard for the consequences of implicit bias and stereotyping in employment decisions,” and summarizing three existing, contemporary models of disparate impact that “[m]oderniz[e]” how disparate treatment is evaluated under Title VII).