The Yale Law Journal

October 2015

Saving 60(b)(5): The Future of Institutional Reform Litigation

Civil ProcedureCivil-Rights Law

abstract. Institutional reform decrees are one of the chief means by which federal courts cure illegal state and federal institutional practices, such as school segregation, constitutionally inadequate conditions in prisons and mental hospitals, and even insufficient dental services under Medicaid. The legal standards governing federal courts’ power to modify or dissolve institutional reform decrees, a crucial tool that can be used to safeguard or sabotage these decrees’ continued vitality, are rooted in Federal Rule of Civil Procedure 60(b)(5). In Horne v. Flores, the Supreme Court tweaked Rule 60(b)(5) to make it easier for state and local institutions to modify or dissolve the institutional reform decrees to which they are bound. This Note argues that Horne has introduced considerable confusion and divergence among lower court approaches to the modification and dissolution of reform decrees, and has made it too easy for institutional defendants to escape federal oversight. At the same time, however, Horne rested on legitimate policy critiques of institutional reform litigation. This Note attempts to chart a middle ground between the doctrine’s detractors and defenders by making concrete proposals about how courts should resolve the confusion introduced by Horne. These recommendations would align the institutional reform doctrine with the policy critiques highlighted by the Court in Horne while still allowing for the effective vindication of constitutional rights.

author. Yale Law School, J.D. 2015. Many thanks to Kristin Collins for her thoughtful commentary and sustained support, Nicholas Parrillo for introducing me to the topic and for his suggestions during the writing process, and Judith Resnik for her consistent instruction, guidance, and stimulation. Thanks also to Noah B. Lindell, Mike Clemente, Charlie Bridge, Elizabeth Ingriselli, Rebecca Lee, the participants in the 2014 Advanced Civil Procedure Seminar at Yale Law School, and the editors of the Yale Law Journal.


On June 11, 1963, George Wallace, then Governor of Alabama, stepped into the doorframe of the Foster Auditorium at the University of Alabama and refused to move.1 He was attempting physically to block the entry of two black students, Vivian Malone and James Hood, who were admitted to the University by court order. Governor Wallace sought to keep the pledge he made in his inauguration speech—“segregation now, segregation tomorrow and segregation forever”—but he failed to honor his promise after President Kennedy federalized the Alabama National Guard and Deputy Attorney General Nicholas Katzenbach commanded Wallace to step aside. The encounter, which is now remembered as the “Stand in the Schoolhouse Door,” came to symbolize what was then a fledgling movement to vindicate the constitutional rights of African Americans through the use of court-ordered desegregation decrees.

Today, the school district where Wallace made his ill-fated stand bears a closer resemblance to his inauguration pledge than one would expect. Recently released reports documenting the resegregation of the Tuscaloosa, Alabama public schools2 confirm what academics have long known: America’s fight for desegregation is not over, and its supporters are losing ground.3 The data show that when school districts are released from court-ordered desegregation plans, they gradually resegregate.4 This phenomenon suggests that the court orders provided structural support necessary for desegregation efforts, and that the orders were prematurely vacated.

Behind the decline of the desegregation project lays a subtler trend: the steady unraveling of protections against governments’ attempts to modify or dissolve institutional reform decrees in federal court. Institutional reform decrees are one of the chief means by which federal courts cure illegal local, state, and federal institutional practices, including school segregation, English language learning programs that are deficient under the Equal Educational Opportunities Act,5 constitutionally inadequate conditions in prisons and mental hospitals,6 and even insufficient dental services covered by Medicaid.7 Although some scholars suggest that the era of institutional reform litigation (IRL) has passed,8 many public institutions still operate under orders issued and supervised by federal courts. For example, in 2000, twenty-five percent of state prisoners were incarcerated in institutions subject to court orders.9 Institutional reform decrees still play an important role in the life of America’s public institutions.10

Institutional reform decrees bring federal courts into the administration of state affairs, implicating a unique feature of American federalism. As decrees become easier to modify or dissolve, state institutions become more likely to escape federal court oversight and operate autonomously. Some scholars support the continued erosion of the standards governing modification of these decrees, arguing that state institutions should have flexibility in administering their public policy and that federal courts lack both the expertise and the authority to craft that policy.11 Scholars who support a tougher standard for modification argue that the decrees provide an essential tool for the effective vindication of the rights of persons ill-equipped to effectuate public policy change on any level.12

While an abundance of literature has debated the propriety of institutional reform from a policy perspective,13 evaluating the relative merits and demerits of federal court oversight of state institutions, there is little documentation of the Supreme Court’s jurisprudence on when an institutional reform decree should be modified or dissolved.14 This is surprising, given that attempts to modify or dissolve institutional reform decrees often revolve around the very policy issues that scholars debate. Additionally, the Court’s jurisprudence is particularly important given the sparse guidance provided by the procedural vehicle parties use to modify decrees: Federal Rule of Civil Procedure 60(b)(5). This Rule merely instructs that a decree may be modified when “applying it prospectively is no longer equitable.”15 The gap in the literature on reform decrees has become even more apparent after the Supreme Court decided Horne v. Flores in 2009.16 Horne substantially modified the standards that federal courts use under Rule 60(b)(5) to modify decrees, and it created considerable confusion regarding the circumstances under which dissolution of such decrees is appropriate.

The literature’s focus on policy issues—and the resulting dearth of attention to the legal and doctrinal elements of the institutional reform debate—has allowed positions that were developed in theoretical terms, often along party lines, to shape the doctrine.17 However, moving forward, courts ought to exhibit moderation, compromise, and attention to the fact-specific nature of institutional reform decrees. Courts must strike the proper balance between the legitimate policy critiques of institutional reform decrees and the need for courts to maintain oversight over institutions operating in violation of individuals’ federal rights. Horne was one attempt to strike such a balance, but the Court’s standard swung the pendulum too far in the direction of state and local power. Critics of institutional reform are correct that ongoing decrees pose problems for democratic accountability in local government. The current doctrine, however, does not narrowly tailor the standards for modification to target the situations that pose the greatest threat to democratic accountability. Supporters of institutional reform, for their part, are correct that courts have a role to play in regulating aberrant institutions, but they should acknowledge that this role must accommodate new policy insights and the considered judgments of elected officials years after decrees have been entered. The Court’s failure to strike the proper policy balance in Horne—and the muddied, confusing balance it did strike—has produced a nebulous and uncertain terrain for courts to navigate when evaluating motions to modify decrees. Since the stakes of the policy debate have been clearly established, the next step is to begin clearly and carefully hewing the standards courts apply to the areas of debate over which there is consensus.

This Note addresses the gap in the literature between policy and doctrine by discussing the legal standards for, and jurisprudence of, modifying and dissolving court-ordered institutional reform decrees. Part I introduces the history, policy issues, and legal jurisprudence surrounding IRL. Part II introduces Horne v. Flores, the most recent and important Supreme Court case on IRL, and details both the reform the Court sought to introduce and its interpretation of Rule 60(b)(5). Part III examines the confusion that Horne’s innovations have wrought by examining problems, divergences, and mistakes in lower court decisions applying Horne. Part IV proposes several changes to the Rule 60(b)(5) doctrine that will clear up the confusion from Horne and, more importantly, will help align the doctrine with the proper balance of policy values while protecting the rights and remedies secured by IRL.

I. adjudicating institutional reform

IRL, and the policy and jurisprudential debates surrounding it, evolved out of the dynamic history of the federal courts’ equitable power to issue injunctions and regulate behavior. This power, and the justifications undergirding it, varies with the nature of the behavior regulated, the identity of the parties, the level of consent, and the federal court’s conception of its authority to involve itself in policymaking. This Part traces the rise of IRL and lays out a roadmap of competing frameworks for evaluating its propriety. Part I.A discusses the historical context of IRL and the Supreme Court’s interpretation of Rule 60(b)(5), which governs motions to modify or dissolve injunctions. Part I.B canvasses the policy debate on the appropriateness of continued federal court involvement in state public policy, a conversation that is very much alive today. Part I.C then examines the legal doctrines governing court-ordered injunctions, consent decrees, and modifications to those orders and decrees in the institutional reform context.

A. The Rise of Institutional Reform Litigation

The first Supreme Court cases adjudicating the modification of judicial decrees involved injunctions and consent decrees that were designed not to reform institutions, but to end anticompetitive behavior.18 These cases were the first to interpret the then-new Federal Rule of Civil Procedure 60(b)(5), which provides scant guidance for determining when a decree may be modified or dissolved:

On motion and just terms, the court may relieve a party or its legal representative from a final judgment, order, or proceeding for the following reasons: . . . (5) the judgment has been satisfied, released, or discharged; it is based on an earlier judgment that has been reversed or vacated; or applying it prospectively is no longer equitable . . . .19

At first, the Court interpreted the phrase “no longer equitable” in an exacting way: “Nothing less than a clear showing of grievous wrong evoked by new and unforeseen conditions should lead us to change what was decreed after years of litigation with the consent of all concerned.”20 The Court’s “grievous wrong” standard treated these early antitrust consent decrees like contracts, hesitating to modify agreements reached through mutual consent21 or to divine a “purpose” from the decree.22 In other words, a “contract paradigm” undergirded the early doctrine on the modification of consent decrees: courts construed and modified consent decrees as part of an attempt to enforce the bargain struck by the parties.23 The contract paradigm emphasizes the rights of private parties to enter into efficient, bargained-for settlements of their disputes.24 The decree’s authority derives from the parties’ consent,25 and the court’s role in interpreting and modifying the decree is delimited by that consent.26

While the Court tended to treat antitrust consent decrees like contracts, it adopted a different paradigm for the modification of litigated injunctions. These injunctions look less like contracts: first, they are litigated with a finding of liability, whereas a consent decree might not entail an admittance of fault. Second, courts enter the decree as an order, not an agreement, making courts less worried about breaching a bargained-for contract.27 Thus, when litigated injunctions were involved, the Court dropped the “grievous wrong” standard and replaced it with a more lenient approach allowing courts to modify injunctions (for instance, by establishing the conditions required for competition between manufacturers).28 This “judicial act paradigm” allowed courts to retain authority and control over these injunctions, as long as enforcement of the decree was appropriate to accomplish the decree’s purpose and to vindicate the rights and duties of the parties.29 This paradigm emphasizes the role of the court in determining rights and duties, making findings of liability, and advancing the public interest.30

The judicial act-contract spectrum worked well for remedies arising out of antitrust disputes, in part because the remedy was aimed at curtailing a private party’s behavior. Soon after the Supreme Court established the judicial act and contract paradigms, however, a new category of cases challenged them. This new category, IRL, differs from private law disputes in several important ways.31 The crucial distinction is that IRL is aimed at modifying government policy and practice, not the conduct of private parties. Litigants seek to use IRL to influence issues such as education,32 prisons,33 mental institutions,34 foster care systems,35 access to Medicaid,36 and more. For this reason, IRL is often referred to as “public law litigation.”37 Efforts to affect policy by suing local and state actors create a host of important issues, including the effectiveness of the democratic process, blame-shifting within and between administrations, the limits of judicial competence, vertical federalism, and the separation of powers.

B. Policy Issues in Institutional Reform Litigation

IRL implicates a number of policy concerns. In particular, the three main debates on the appropriateness of IRL concern the proper role of federal courts in formulating public policy, the potential harm that reform decrees may inflict on democratic accountability, and the capacity of courts to supervise institutional reform.

1. Federal Judicial Involvement in State and Local Policymaking

First, critics and supporters disagree on the extent to which federal courts should become involved in substantive local policy. Courts are sensitive to the potentially intrusive nature of federal judicial control over state and local governments, particularly when social service institutions are the subject of IRL. Critics of IRL often cite Ruiz v. Estelle,38 in which federal courts were accused of dismantling a “Texas style” prison system and installing a “California style” regime, only to see violence and gang membership increase in Texas prisons.39 The federal court’s choice to implement another state’s prison control program demonstrates how IRL can exacerbate horizontal and vertical federalism issues.

However, defenders of IRL respond that federal courts, as a “coordinate” branch of government, have an obligation to make policy that effectuates and gives meaning to public values.40 These defenders further observe that, as the modern state becomes increasingly bureaucratic, federal court remedies to cure constitutional violations by these government entities will need to restructure the bureaucracies that give rise to the infringing conduct. Complex IRL decrees are thus a proportionate response to an increasingly complex administrative apparatus.41 Such authors have noted that in most prison litigation reform cases like Ruiz, there was no cry of usurpation of state rights—and prison conditions improved.42

2. Political Accountability

Second, and relatedly, critics and defenders of IRL diverge on the extent to which federal court involvement in effectuating policy change prevents local constituents from holding elected officials accountable for their policy choices. Critics argue that federal court monitoring of state policy can short-circuit democratic checks on local elected officials in several ways. Federal court control removes policy issues from the purview of locally elected government officials, which in turn removes those issues from democratic control by constituents. A court order can thus “lock in” a particular policy and make it immune to democratic control.43 Additionally, federal court orders provide a litany of options for elected officials to escape accountability: officials can implement unpopular reforms they like while publicly criticizing them;44 lock programs in place against future governments by entering into a consent decree;45 pass responsibility on to the next administration to enforce the decree;46 and protect reform budgets against competing claims, since court-ordered reforms are usually at the top of any priority list.47 These fears are exacerbated by the extended duration of IRL: decrees may remain on a court’s docket for decades, and violations may give rise to additional decrees, subsequent violations, and extensions of the duration of federal court control.

On the other hand, defenders of IRL argue that consent by all parties involved improves the administration of the decree, that accountability problems with courts are overstated, and that detractors exaggerate the “lock-in” effect of court orders. First, they argue, consent from the defendant-institution generates buy-in within the regulated institution, creating possibilities for better enforcement of the decree. Insider knowledge from the institution also helps prevent surprises in the administration of the decree. Second, the institutional differences between courts and executive agencies are often overstated: judges tend to have similar backgrounds and training to administrative professionals,48 their decisions are often aligned with popular opinion, and their actions are constrained by the legislative and executive branches.49 Finally, the lock-in effect of court orders is exaggerated, both because a decree may be modified and because other branches of local government may use alternative means to alleviate the underlying problems.50

3. Courts’ Capacity To Administer Reform

Finally, critics and defenders dispute the competence of courts to formulate and administer institutional reform. Critics argue that the limited capacity, knowledge, and budget of the judiciary make it a bad candidate for institutional reform.51 In particular, critics claim that (1) courts usually deal with a strictly delimited set of parties and circumstances, not polycentric disputes with multiple parties and interests; (2) courts generally focus on whether existing laws were violated in the past, not on what policies or laws will work in the future; and (3) courts are attuned to the violation of rights and appropriate remedies, not the public policy considerations involved in IRL decrees.52 The result, critics claim, is a “privatization” of policymaking that transfers decision-making power over policy issues from traditional political mechanisms to national advocacy organizations and litigating parties. This creates a pattern of similar lawsuits in several jurisdictions, all controlled and guided by the policy of a national advocacy organization.53

However, supporters of IRL might respond that no other branch of government will solve the underlying problem, that the institutional disadvantages of courts are exaggerated, and that national norm formation and enforcement is not always undesirable. First, as a general matter, the fact that a court has been asked to cure a systematically dysfunctional institution is usually an indication that the political branches have failed to solve the problem, giving rise to an inference that the court is the only branch in a position to resolve the problem. Second, the critique of judicial involvement in policymaking underestimates the variety of roles courts play, including the management of their own operations, administration of bankrupt estates, and supervision of compliance with court orders. Political scientists have recognized for decades that courts regularly make policy,54 an observation that should temper any categorical rejection of courts’ role in public policy formation. Third, a court’s recognition of its disadvantages is its principal weapon against them. As Edward Rubin and Malcolm Feeley note, courts’ awareness of their own limits gave rise to “circumspection and self-abnegation,” with courts reaching out to adopt standards developed by, for example, correctional professionals.55 Finally, while the privatization argument rightly laments the transfer of some public policy responsibility away from elected officials, it neglects that courts are still deeply involved in the reform either way, and that the elected branches from whom responsibility has been shifted have already failed to resolve the underlying issue on their own. Even if a legislature resolved the issue instead of a court, it would likely rely on many of the same private organizations for information collection and policy planning, organizations which promulgate standards for their member groups whether or not those standards are incorporated into judicial decrees.

C. Legal Doctrine in Institutional Reform

The doctrine on IRL falls into three categories that correspond to different steps in the IRL process: standards governing the entry of decrees, standards governing the modification and dissolution of decrees, and standards governing the propriety of prophylactic relief.

It is well established that courts have the power to enter injunctions or approve consent decrees that provide the kind of structural relief sought through IRL. The entry of injunctive relief is governed by a strict four-factor test, which requires that the plaintiff prove that (1) she suffered an irreparable injury, (2) the injury cannot be relieved by traditional legal remedies, (3) the balance of hardships between the plaintiff and the defendant warrants an injunction, and (4) the injunction does not disserve the public interest.56 These standards derive from the “drastic and extraordinary” nature of injunctive relief,57 and reflect federal courts’ preference for awarding traditional monetary awards instead of injunctions to restore plaintiffs to their rightful position.

In contrast to injunctions, the standards governing entry of a consent decree are relatively lax. A consent decree must “spring from, and serve to resolve, a dispute within the court’s subject-matter jurisdiction; must come within the general scope of the case made by the pleadings; and must further the objectives of the law upon which the complaint was based.”58 Courts apply a less exacting level of scrutiny to consent decrees because the power of a decree is not animated only by “the law which forms the basis of the claim,” but also by “the parties’ consent.”59 The level of judicial involvement may also hinge on other factors, such as whether or not the case is a class action—in which case Federal Rule of Civil Procedure 23 imposes additional requirements on a court’s approval of a consent decree.60

Until recently,61 the standards governing the modification of decrees—whether litigated injunctions or consent decrees—were clear.62 The Court, over a twenty-year period,63 developed a test for modifications of decrees that sought to balance the preference for state and local control with the need for continuing federal court oversight of entities that systematically violate federal law. This standard was definitively formulated in Rufo v. Inmates of Suffolk County Jail, which involved a motion to modify a consent decree requiring Suffolk County to comply with constitutional standards by building additional jails.64 In Rufo, the Court announced a two-step test for evaluating requests to modify or vacate injunctions. First, a change in factual or legal circumstances is required as a preliminary condition. Second, in order to justify modification, the change in law or fact must make compliance with the decree substantially more onerous, render the decree unworkable because of unforeseen conditions, or make the decree detrimental to the public interest.65

When entering an order, a court has the power to award prophylactic relief.66 Prophylaxis refers to precautionary measures that are used both to provide more effective relief and to prevent the harm that could be caused by potential future violations before it develops.67 Prophylactic relief goes beyond what is required by law in order to ensure that the law is followed in the future. Examples include monitoring measures that require a party to report conditions, provide access to an investigator, or provide for ongoing oversight of an institution’s performance.68 Courts are most likely to award such relief in three circumstances: when monitoring of the relief will be difficult,69 when the court does not trust the defendant-institution to administer the relief adequately,70 or when restoration of the plaintiff to its rightful position is impossible and the court chooses to award more (rather than less) relief.71 Prophylactic measures are closely tied to the development and success of IRL in the past several decades, particularly given the problem of recalcitrant local governments, institutions, and norms.72

The doctrine gives courts a great deal of discretion to decide on the extent of prophylactic relief to award, because the decision is highly contextual and dependent on the trustworthiness of the parties involved. However, when ordering prophylactic relief, courts employ the same standard governing the entry of consent decrees. First, the relief must “spring from, and serve to resolve” a dispute that the court has jurisdiction to hear.73 Specifically, the court may not enter relief to solve a problem that is not before that court. Second, the relief must “come within the general scope of the case made by the pleadings.”74 In other words, the court must justify the relief based on arguments made by the parties. Finally, the relief must “further the objectives of the law upon which the complaint was based.”75 The purpose of the relief must be closely related to the objectives of the law governing the parties’ dispute.

The constitutional status of prophylactic relief has fluctuated over the years, in part due to the advances and retreats of each side in the broader policy debate over IRL itself. As early as 1977, in Milliken v. Bradley, the Supreme Court noted that federal court decrees “must directly address and relate to the constitutional violation itself.”76 This means that a court’s entry of a decree will “exceed appropriate limits” if the decree is “aimed at eliminating a condition that does not violate the Constitution or does not flow from such a violation.”77 While the Court in Milliken addressed the initial order of prophylactic relief, the Court in Rufo created a different standard for modification. The Rufo Court nonetheless affirmed the capacity for consent decrees to mandate relief beyond what the Constitution requires or what a court would have ordered absent the settlement.78 As the Rufo Court observed: “[A] proposed modification should not strive to rewrite a consent decree so that it conforms [only] to the constitutional floor.”79

It is important to note that the standards for entering prophylactic relief, entering an injunction, and modifying an injunction are all separate and independent legal tests. A court may only award prophylactic relief if the standard for entering an injunction is satisfied (because the vehicle for prophylactic relief is an injunction). A court may only modify an injunction if one has been previously entered, but may modify that injunction whether or not it contains prophylactic measures. When modifying an injunction, a court may award prophylactic relief if the standards for prophylactic relief are satisfied.

II. horne v. flores

The doctrinal and policy issues swirling around modifications to IRL decrees came to a head in Horne v. Flores.80 In Horne, members of the Arizona state legislature intervened in an ongoing case seeking to modify an injunction governing English Language Learner (ELL) programs in the state.81 The convoluted facts of the case brought the policy critiques of IRL to the fore, and gave five Justices an opportunity to make IRL decrees easier to modify—and, ultimately, to dissolve. Part II.A describes the facts of the case and the Supreme Court’s decision, while Part II.B analyzes the decision’s effects on Rule 60(b)(5) doctrine and the ambiguities it left unresolved.

A. The Decision

In 2000, ELL students in Nogales, Arizona, won a declaratory judgment holding the State, the Arizona State Board of Education, and Arizona’s Superintendent of Public Instruction liable for violating the Equal Educational Opportunities Act (EEOA).82 The defendants did not appeal the order. The court then entered a series of injunctions over several years. One order in 2001 applied the declaratory judgment statewide,83 another in January 2005 gave the State ninety days to fund its ELL program adequately,84 and yet another held the State in contempt in December 2005.85 These orders were also not appealed.86 After accruing $20 million in fines, the State Legislature passed HB 2064, which increased funding and provided for substantive changes to ELL programs, and the Attorney General moved for accelerated court approval of the law as adequate to satisfy the decree.87 When the principal defendants to the original declaratory judgment—the Governor, the State, and the State Board of Education—joined the plaintiffs in opposing HB 2064,88 Arizona’s Speaker of the House and President of the Senate intervened89 and moved for relief under Rule 60(b)(5) on the basis of the progress made.90 The district court denied that motion, the court of appeals affirmed, and the Supreme Court granted certiorari.91

Justice Alito, writing for a Court divided five to four, stressed the federalism issues at play, explaining that such concerns are “elevated” when “different state actors have taken contrary positions.”92 The opinion emphasized the need for a “flexible approach” to Rule 60(b)(5) motions in the institutional reform context, indicating that this standard required courts to “remain attentive to the fact that ‘federal-court decrees exceed appropriate limits if they are aimed at eliminating a condition that does not violate [federal law] or does not flow from such a violation.’”93 Justice Alito referenced academic literature criticizing public officials who “consent to, or refrain from vigorously opposing, decrees that go well beyond what is required by federal law.”94 He then indicated that the injunction at issue—of which the state executive officers approved—ought to be treated as a collusive consent decree, rather than the result of a truly adversarial process.95 The concern for undue federal control over state actors, the Court said, means that courts should ensure that the effect of equitable decrees is limited to remedying violations of federal law, and no more.96

After discussing the problems created by consent decrees (and insufficiently opposed injunctions), Justice Alito clarified the doctrine that would guide the Court’s Rule 60(b)(5) analysis. He explained that the correct inquiry is the one prescribed in Rufo—namely, whether “a significant change either in factual condition or in law” renders continued enforcement “detrimental to the public interest,”97 such that “applying it prospectively is no longer equitable” under Rule 60(b)(5). But he interpreted the “changed circumstances” language broadly to include not only changes in law or fact, but also “new policy insights” that might warrant a modification to the underlying judgment.98 Justice Alito, purporting to follow Rule 60(b)(5) but in substance modifying the standard, wrote: “To determine the merits of this claim, the Court of Appeals needed to ascertain whether ongoing enforcement of the original order was supported by an ongoing violation of federal law . . . .”99 Having established this standard for the modification of equitable decrees, Justice Alito found “significant change in facts or law [that] warrant[ed] revision of the decree” under Rufo,100 leaving little doubt about what the lower courts ought to do on remand.101

B. Changes to the Rule 60(b)(5) Doctrine

Horne made it easier for state and local institutions to escape federal court supervision through three important doctrinal innovations to the Rufo test. First, Horne held that “new policy insights” may constitute the kind of changed circumstances that satisfy the first step of the Rufo test.102 Before Horne, only changes in law or fact could trigger review. By adding “new policy insights,” the Court made it possible for federal, state, or local institutions to manufacture the conditions needed for courts to dissolve injunctions against such institutions. For example, in Horne, the Court held that federal, state, and local policies enacted since the original lawsuit constituted policy insights sufficient to satisfy its new standard.103 The Court’s addition of “new policy insights” could also mean that, contrary to the Court’s earlier holdings in Rufo and United States v. Swift & Co., changed circumstances need not be unforeseen in order to trigger review of an injunction.104 The Court made clear that institutions may avoid the strictures of an injunction by enacting legislation that avoids the decree’s mandate: “[P]etitioners argue that Arizona is now fulfilling its statutory obligation by new means that reflect new policy insights and other changed circumstances. Rule 60(b)(5) provides the vehicle for petitioners to bring such an argument.”105

Second, Horne held that if the changed circumstances prong has been met, a decree is no longer equitable and is thus subject to modification “as soon as a violation of federal law has been remedied.”106 In other words, if new policies have been implemented in a manner that brings the offending institution into line with constitutional or statutory minimums, the court order may be dissolved even if it has not yet been satisfied. This doctrinal innovation would functionally eliminate the power of prophylactic decrees: a defendant could comply with federal law and then request that the court vacate the remaining prophylactic aspects of its decree, including monitoring requirements, reporting systems, and mechanisms of internal review. On the other hand, the Court also noted that vacatur is proper “[i]f a durable remedy has been implemented,”107 language that may preserve room for prophylactic remedies to operate until such a long-term solution is in place.108 However, lower courts have ignored the “durable remedy” aspect of Horne’s test, leaving the future of prophylactic injunctions uncertain.

Relatedly, Horne alsoheld that continued enforcement of an order is improper—and thus, that modification is required—if the order is not supported by an ongoing violation of federal law.109 Although the Court relied on Milliken v. Bradley for this proposition, Milliken held that the entry of an order, not its modification, is improper in such circumstances.110 By holding that an existing order must be supported by an ongoing violation of federal law, Horne has created confusion over whether the Rule 60(b)(5) equity test need even be applied once the defendant-institution has come into compliance with federal minimums.111 Either way, the Court has created a new burden of proof for the party responding to a motion to terminate a decree: it must in effect show an ongoing violation of federal law.

Third and finally, Horne creates a presumption that if the Rufo test is satisfied, a decree should be vacated. Previously, modification or vacatur of a decree was subject to the court’s discretion, even if the Rufo test were satisfied.112 However, the Horne Court noted that if “the objective of the District Court’s . . . order . . . has been achieved,” and a durable remedy is in place, “continued enforcement of the order is not only unnecessary, but improper.”113 Reversing the presumption to operate against continued enforcement not only restricts courts’ traditional discretion in formulating equitable relief, but also further incentivizes parties subject to a decree to manufacture circumstances that would require vacatur of that decree.

III. confusion in the lower courts after horne

Horne adjusted the Rufo standards in several ways and introduced academic criticism of IRL into the law. The translation from academic critique to legal standard, however, has not been without friction. Horne privileged the desire for state and local control over the desire for federal oversight, but it did not explicitly disavow prior decisions that employed stricter standards for modifying decrees. This ambiguity has left a number of slippages and vagaries in the law, and lower courts have struggled to ascertain the standards applicable in various cases. This Part addresses some of those ambiguities, and explores lower court opinions that reveal and confirm the existence of disagreements and confusions over how to interpret Horne’s new standards. The next Part makes concrete suggestions on how to resolve these disagreements, informed by the need to balance competing policy values.

A. Horne’s Scope: Injunctions, Consent Decrees, Neither, or Both?

Horne involved a peculiar set of facts: the underlying decree was a fully litigated injunction, but it exhibited many of the characteristics of consent decrees because there was evidence of collusion and insufficient contestation of the suit by the Arizona executive branch.114 This fact pattern has contributed to confusion on the scope of Horne’s holding: does it apply to fully litigated decrees, or just to potentially collusive ones, as in Horne? Does it apply to all injunctions and consent decrees? To declaratory judgments? To all court decrees, or just those that are more protective than what federal law requires?

These questions are not merely academic; they reflect actual confusion in the courts. The District Court for the District of Connecticut recently held that Horne did not alter the Rule 60(b)(5) landscape because the underlying order in Horne involved a declaratory judgment as opposed to a consent decree.115 However, other district courts have come to the opposite conclusion.116 Meanwhile, a panel of district and circuit judges recently held that Horne only applies to consent decrees that are more protective than what federal law would otherwise require, or when defendants have remedied underlying constitutional violations without satisfying a decree.117

B. The Status of Prophylactic Institutional Reform Decrees

Horne also left unclear the status of decrees that go beyond the requirements of federal law. For example, at least one court has held that Horne applies only to decrees that exceed what is required by federal law—that is, to decrees with prophylactic elements.118 Before Horne, courts could enter prophylactic consent decrees as long as the decrees “spring from, and serve to resolve, a dispute within the court’s subject-matter jurisdiction”; “come within the general scope of the case made by the pleadings”; and “further the objectives of the law upon which the complaint was based.”119 After Horne, some courts have applied a much stricter standard to adjudicate the propriety of prophylactic relief.

Much of the confusion revolves around a line in Horne indicating that, when a party petitions to dissolve a decree, “a critical question . . . is whether the objective of the District Court’s . . . order . . . has been achieved. If a durable remedy has been implemented, continued enforcement of the order is not only unnecessary, but improper.”120 Similar threats to prophylactic relief abound in Horne, including languagestating that continued enforcement of an order is improper if the order is not supported by an ongoing violation of federal law,121 and that a decree is no longer equitable “as soon as a violation of federal law has been remedied.”122 Each of these passages describes circumstances in which enforcement of prophylactic relief is improper, but provides little guidance to courts on when to modify or vacate prophylactic remedies. They raise two distinct questions: (1) what constitutes “the objective” of a decree and the achievement of that objective; and (2) what is the constitutional or statutory minimum with which a defendant must come into compliance?

1. What Constitutes the Objective of the Decree?

Horne held that “whether the objective of the District Court’s . . . order . . . has been achieved” is an essential factor in determining whether continued enforcement of the order is necessary.123 Lower courts have taken this instruction seriously, but there is divergence in how to interpret the “objective” of the decree and whether it has been “achieved.” Whether a court reads the objectives of the decree narrowly or broadly determines whether or not the decree and its prophylactic means of implementation will survive a Rule 60(b)(5) motion. However, it is not clear what set of tools a court should bring to bear on the question. Should the decree be read like a contract, such that anything outside the four corners of the decree is irrelevant? Or should the litigation that gave rise to the order help the court infer what the objective is? And how should “achievement” of the objective be measured?

Some courts have evaluated the objective of the decree in terms of the actual language of the decree, incorporating all of the requirements into a broad determination of an objective. At least one court used this approach when a decree’s prophylactic elements had not yet been satisfied, but the underlying constitutional violation had been remedied.124 The court noted that a consent decree “embodies an agreement of the parties,” one that they “desire and expect will be reflected in, and be enforceable as, a judicial decree.”125 Other courts have applied a substantial compliance standard, equating a decree’s objective with performance of the essential requirements of the decree and a lack of intentional deviation. For example, the Court of Appeals for the Second Circuit indicated that substantial compliance with a decree’s objectives could justify termination, but also suggested that the district court was not required to terminate the decree on that basis.126The Court of Appeals for the Sixth Circuit upheld a decision by the District Court for the Middle District of Tennessee to dissolve a consent decree after finding the state of Tennessee to be in “current,substantial compliance” as required by the decree, even though it had violated the decree several times in the past.127 The language of substantial compliance can also be written into a decree: one court took a substantial compliance approach in part because the terms of the decree stipulated that any party may seek to dissolve the decree when “further supervision by the Court is not necessary.”128 Other courts have rejected the substantial compliance approach129: one district court denied a defendant’s motion to dissolve a consent decree when the decree required eighty-five percent compliance and the defendant had demonstrated only eighty-four percent compliance.130

At the same time, ambiguities within Horne have created confusion around whether meeting the objectives of the decree is even relevant to the Rule 60(b)(5) inquiry. Most of these ambiguities stem from sentences in Horne suggesting that once the underlying violation of federal law has been remedied, the decree is no longer proper.131 If this were true, satisfaction of the law would obviate an inquiry into whether the objectives of the decree have been achieved. At least two courts have implied that the defendant need only come into compliance with federal or constitutional law, not with the decree, in order for a court to dissolve the decree.132

2. What Is the Minimum Under Federal Law?

If a decree is no longer proper once the violation of federal law has been remedied, courts need a means to determine what actions violated federal law and what would constitute compliance with the law. These issues are often resolved in the litigation process. But when parties enter a consent decree, or if the court enters a preliminary injunction without reaching the merits of the case,133 there typically is no finding of liability. In these scenarios, questions of whether the defendant violated the law and what specific behavior constituted the violation are often never reached. And when the parties settle or enter into a consent decree, the question of what specific actions are required to remedy the violation and comply with federal law goes unanswered. One prominent example is Floyd v. City of New York, a series of decisions finding that New York City’s stop-and-frisk policy violated the Fourth Amendment.134 In a well-publicized series of orders, the Second Circuit remanded the remedial portion of those decisions to another judge,135 but then reversed course, limiting the remand and encouraging the parties to settle.136 New York City had a change in administration while the case was on appeal, and the new mayor indicated that he would end the stop-and-frisk policy and drop the appeal.137 Thus, while the new mayor is in power, there is presumably no risk of violating the Fourth Amendment, since the stop-and-frisk policy will cease entirely. However, the settlement (which, functionally, resembles a consent decree138) leaves the status of the decision providing remedial measures in an awkward limbo, and the settlement agreement sunsets five years after it was entered.139 No other legal mechanism prevents a future administration from crafting a stop-and-frisk policy similar to the one declared unconstitutional. By settling voluntarily, the city avoided a court decision that would set a strong constitutional standard and impose remedies lasting beyond the current mayoralty’s agreement with the defendants in that case.140

The inverse scenario is also a possibility: a consent decree could go farther than federal law might require, but with no party sure of exactly what federal law would require because the court never reaches that question. Some lower courts have confronted this issue, but they have not resolved it adequately. One court pithily canvassed the issues at stake when it wrote: “Horne cannot mean that when a defendant agrees to a series of measures designed to remedy constitutional violations, these agreements are necessarily unenforceable because the measures exceed some ill-defined constitutional floor.”141 That court held that, because Horne did not directly address the question of whether a defendant must satisfy the decree or merely satisfy the constitutional minimum, the more stringent Rufo—and not Horne—controlled the Rule 60(b)(5) process.142 It also rejected the defendant’s argument that the court should adopt a definition of the constitutional minimum from a similar case.143 Another court recognized the predicament,144 but held that the defendant had complied with the consent decree’s termination requirements, and therefore vacated the decree without determining whether the underlying violation had been remedied.145 The lower courts have yet to settle on a standard for what, exactly, a defendant must do to dissolve a decree.

C. The Role of “New Policy Insights”

The two areas of confusion discussed above addressed the standard for modifying or dissolving a decree and the scope of allowable remedies. But there is yet another wrinkle that Horne added to the doctrine on Rule 60(b)(5), this one concerning the evidence used to meet the modification or withdrawal standard. The law governing Rule 60(b)(5) motions prior to Horne contemplated two kinds of changed circumstances—changes in law and changes in fact—that could justify vacating or modifying injunctions. Once changed circumstances were demonstrated, the defendant still needed to demonstrate that continued enforcement of the decree would be inequitable absent modification. One innovation of Horne is the contemplation of another category of changed circumstances: the availability of new policy insights. Just like “changes in the nature of the underlying problem” and “changes in governing law or its interpretation by the courts,” the Horne Court observed that “new policy insights” might “warrant reexamination of the original judgment.”146 As of September 19, 2015, lower courts had mentioned Horne’s line about “new policy insights” only ten times, and not always in a way that makes use of that new interpretation of Rule 60(b)(5).147 Yet courts have relied on policy insights in various ways when deciding whether to grant a motion to modify or dissolve a court order.

The least surprising interpretation of the new policy insight language in Horne supposes that demonstrating or implementing new insight into the problem that a decree was entered to resolve will satisfy the first prong of the Rufo test. Under this interpretation, once new policy insights are demonstrated, a defendant may successfully challenge the decree by demonstrating that the new insights make “compliance with the decree substantially more onerous . . . [and] unworkable because of unforeseen obstacles . . . or . . . detrimental to the public interest.”148 One court found that new scientific evidence regarding the best use of an area of the Everglades constituted changed circumstances that justified modification of an ongoing environmental decree.149 Another court made a similar finding on the U.S. Department of Agriculture’s new forestry plan,150 and a third held that new state legislation made continued enforcement of a decree “substantially more onerous.”151 Other courts have implied, but have not stated outright, that new policy insights would suffice to show changed circumstances. For example, one court rejected a defendant’s remedy as inadequate in part because the defendant failed to specifically allege that new policy insights justified vacatur of the decree.152

There is disagreement over whether a policy implemented to resolve the same issue that the original court decree was intended to address can be relied upon as changed circumstances. The issue lurking in the background is the possibility that an institutional defendant may simply ignore an IRL decree and implement its favored policy response to the underlying problem, and then petition the court to modify or dissolve the decree on the basis of its “new policy insight.” At least one court has said that a state’s efforts to comply with a decree did not constitute changed circumstances, because the changes were foreseeable and involved the objective of the decree itself.153 On the other hand, another court held that reshuffling agency procedures to address “precisely those concerns that the [original] order was crafted to remedy constitute[d] a factual change.”154 The court vacated the decree, not because it was inequitable, but because the new agency procedures made continued enforcement of the decree “substantially more onerous.”155 The Horne Court’s cryptic addition of the “new policy insights” concept has confounded the lower courts, and has created a potential loophole for IRL defendants to exploit.

D. What Is a “Durable Remedy,” and When Is It Required?

As Horne explained, “[i]f a durable remedy has been implemented, continued enforcement of the order is not only unnecessary, but improper.”156 Justice Breyer’s dissent also referred to the “well-established principle” that, in order to dissolve a decree, the moving party must show that the prohibited acts are unlikely to occur again.157 This language seems to condition dissolution of an unsatisfied decree on the existence of a durable remedy—but the Court did not explain how that condition interacts with the new standards that allow defendants to avoid prophylactic measures when there is no ongoing violation of law. Indeed, it is difficult to distinguish between aspects of a “durable remedy” and what would be considered prophylactic relief. In short, the Court muddied the waters by implicitly referring to the necessity of prophylactic measures to create a “durable remedy,” while at the same time making it easier to dissolve decrees with prophylactic elements.

Courts have treated the “durable remedy” language in a number of ways. Some have held that a “durable remedy” is one that gives the court confidence that a defendant will not resume violation of a right once judicial oversight ends.158 This requires courts to find some sort of prophylactic mechanism that ensures future compliance with the terms of the decree. Even this standard, however, entails considerable flexibility in what constitutes a durable remedy. For example, one court overturned a district court’s finding that the defendant needed to establish testing mechanisms to ensure future compliance (a typical form of prophylactic assurance), noting that the defendant’s compliance for the last several years was likely a sufficient guarantee of durability.159 On the other hand, another court found that there could not be a durable remedy without adequate reporting systems in place.160

Moreover, some courts have reached the question of whether to vacate an unsatisfied decree without addressing the durable remedy issue from Horne.Several courts have skipped the durability inquiry. One court, when adjudicating a motion to modify a desegregation decree, expanded a lower court’s modification of the decree to make it even less restrictive and exempt additional school districts.161 The court did not inquire into the durability of existing remedies, even after acknowledging that vestiges of segregation still remained.162 Other courts skipped the durability question altogether, asking only whether or not the objective of the challenged decree had been achieved.163 As with the constitutional minimum versus decree fulfillment question, lower courts appear to be hopelessly divided over the meaning of Horne’s “durable remedy” requirement. Read together with other portions of the opinion, it might mandate, allow, or forbid prophylactic remedies.

IV. saving 60(b)(5): policy and doctrine

Horne v. Flores left lower courts with insufficient direction to resolve a new set of issues in the Rule 60(b)(5) and IRL landscape. However, Horne also introduced serious policy-based critiques of IRL’s tendency to hamstring state and local governments and to short-circuit democratic control over those institutions. The path forward must combine Horne’s concern for policy valueswith fine-grained doctrinal adjustments. These adjustments must preserve legitimate authority over continuing injunctions while narrowly tailoring the scope of that authority to those measures that further the proper balance of policy values. This Part begins to develop the path forward, offering concrete, moderate suggestions about how courts should approach motions to modify decrees in light of the substantive criticisms on both sides of the policy debate.

A. Apply Horne Flexibility Only to Consent Decrees and Collusive Injunctions, Not Properly Litigated Injunctions

Consent decrees involving government institutions pose a threat to democratic accountability: parties may negotiate public policy behind closed doors, and politicians may lock in future administrations, pander to private interests, and seek political cover.164 The underlying injunction in Horne was fully litigated, but the Supreme Court rightly suspected that it presented many of the same issues posed by consent decrees, as the defendants may not have opposed the proceedings with the zealousness required to protect the public interest or to inform the court of opposing viewpoints. Consent decrees and potentially collusive injunctions thus present circumstances in which the Horne Court’s heightened concern for democratic accountability rightly applies, and courts should tailor the Rule 60(b)(5) standards to the democratic and institutional issues at play. Horne’s doctrinal innovations, such as its presumption that decrees may be modified in certain circumstances,165 are well tailored to the unique issues posed by consent decrees and potentially collusive injunctions.

In contrast, litigated injunctions pose far weaker threats to political accountability, and deference to the proponent of modification is not appropriate for that reason alone. Instead, when a party moves to modify a decree, the party should show that a change in circumstances has (1) made compliance too onerous, (2) made the decree unworkable, or (3) rendered continued enforcement of the decree detrimental to the public interest.166 This is the test that preceded the Horne decision, and it was crafted with attention to the need for continued court oversight of changes in fact or law that militate in favor of modification.167 When the underlying decree has already been subjected to extensive fact finding, arguments by adverse parties, and continued oversight by courts, deference to the defendant’s claim that the decree is no longer equitable is improper.168

This approach strikes a sensible middle ground between interpretations by lower courts. It is more restrained than the position taken by some courts, which would apply Horne’s new standard to every equitable decree entered by a court. But this approach is broader than, for example, the panel’s position in Coleman v. Brown,which held that Horne flexibility applied only to consent decrees that are more protective than what federal law requires.169 The Coleman approach would sacrifice most of what Horne sought to protect, given that there will not always be a clear federal precedent for the underlying conduct, nor always a finding that the defendant is liable for violating that standard.

However, since the existence of more efficient alternatives does not implicate the political accountability concerns that are specific to consent decrees and collusive injunctions, the existence of new policy insights should count as changed circumstances regardless of whether the underlying decree is a negotiated consent order or a fully litigated injunction. Giving effect to new policy insights also helps protect the policymaking function of legislatures and prevents court orders from freezing outdated knowledge into place.170 The incorporation of new policy insights could be packaged into the part of the Rufo inquiry that asks whether “enforcement of the decree without modification would be detrimental to the public interest.”171 Defendants’ increased leverage to modify decrees would be balanced by the fact that, even when there are new policy insights, the proponent of modification would still need to show that those circumstances justify modification, and that the proposed modification is tailored to the change in circumstances.

B. Clarify That the Standard for Vacatur Is the Objective of the Underlying Decree, Not the Constitutional or Statutory Minimum

As described above, the doctrine is unclear on how to weigh a defendant’s efforts to comply with a decree. Specifically, should a court look for satisfaction of the decree or satisfaction of federal law?172 Horne stated both rules in different places and relied on different authorities for each proposition. When the Court discussed the objective of the decree, it referenced Frew ex rel. Frew v. Hawkins, a case involving a motion under Rule 60(b)(5) to modify a decree.173 When the Court discussed the federal minimum, it referenced Milliken v. Bradley, a case that involved the entry of a decree.174 The case law prior to Horne thus supports both the standards discussed in Horne, but applies them to two different cases: the object of the decree is relevant to Rule 60(b)(5) modifications, and the satisfaction of federal minimums is relevant to the entry of the decree in the first place. When evaluating motions under Rule 60(b)(5) for modification, courts should continue to assess whether the objectives of the decree have been satisfied rather than whether an ongoing violation of federal law exists. This distinction is critical to striking the proper balance of policy values, for several reasons.

First, it is imprudent to apply a more permissive standard to the modification of a decree than to its entry. Such an approach would diminish the incentives for parties to bargain for settlement or consent decrees by creating the potential for courts to roll back decrees by later modifications.175 The availability of prophylactic relief thus promotes efficiency, as the Court in Rufo recognized: plaintiffs may bargain for a consent decree that goes beyond what a court could require, rather than risk a litigated injunction that awards only what is constitutionally required.176 Making prophylactic relief vulnerable to vacatur as soon as a defendant is constitutionally compliant would remove this incentive to settle.177 Additionally, a more permissive standard for modification would risk nullifying prophylactic relief entered by a court or agreed to by the parties, even if the standard for entry of a decree allowed it. If a decree may be modified or vacated as soon as there is no longer an ongoing violation of federal law, then any prophylactic aspects of the decree are at immediate risk. Courts should prefer the availability of prophylactic relief, not only because it incentivizes settlement as described above, but also because it can be crucial to protecting rights in the complex administrative schemes that are often the subject of IRL.178

Second, Horne’s use of the federal law minimum, as opposed to the objective of the decree, forces courts to navigate an interpretive labyrinth in order to determine what federal law requires without having adjudicated the merits of the underlying case in the first instance. As numerous courts have found,179 it is nearly impossible to certify that a defendant has complied with the law with no finding of liability, no determination of what the law required or whether it was violated, and no articulation of what conduct is required to come into compliance with the law.180 In contrast, courts can establish the objective of the decree relatively easily using the substantial compliance standard, which has been used by courts for years.181 This approach allows courts to act efficiently while preserving the intent of the court that entered the decree originally.

C. Require a Durable Remedy When Dissolving a Decree

In addition to its Janus-faced standard for modification, the Horne Court was unclear about the role and relevance of the durable remedy and, in particular, when a durable remedy is required for vacatur. Some courts have subsequently skipped the inquiry altogether.182 Going forward, as Justice Breyer argued in dissent, courts should continue to use the “well-established principle” that, to dissolve a decree, the moving party must show both that the decree’s objects have been attained and “that it is unlikely, in the absence of the decree, that the unlawful acts it prohibited will again occur.”183 While the specifics of what constitutes a durable remedy will likely vary based on the facts of each case, it is important that courts consider whether dissolution will roll back the progress the decree created.

The durability issue will likely have important repercussions for the status of thousands of decrees, particularly if combined with the more permissive rule that only the constitutional minimum must be satisfied, rather than the objective of the decree. All the new standards introduced in Horne make it much easier for state and local governments, even those with a history of neglect, to escape court orders. Jettisoning the durability requirement will have a particularly significant impact on desegregation decrees, as many of these decrees go beyond curing de jure segregation and instead seek to eliminate vestiges of segregation (and thus to ensure a durable remedy for the de juresegregation).184 For example, whether or not Justice Alito’s language about the durability remedy is taken to constitute an independent requirement will determine whether or not the Court’s decision in Board of Education of Oklahoma City Public Schools v. Dowell will be abandoned. In that case, the Court held that the purpose of a desegregation order would be achieved only if “the Oklahoma City School District was being operated in compliance with the commands of the Equal Protection Clause of the Fourteenth Amendment” and if “it was unlikely that the Board would return to its former ways.”185 If this latter requirement resembling a durable remedy were to drop out of the analysis, Horne will have significantly loosened the standards for a state or local government actor to dissolve or modify one of the hundreds of desegregation decrees currently in place.

D. Allow for Experimental Modification of Decrees When New Policy Insights Are Available

A potentially more useful innovation from the Horne decision is the concept of “new policy insights.” By introducing new policy insights as a form of changed circumstances, the Horne Court sought to make room for defendants’ good-faith efforts to construct alternative means of satisfying the objective of a court’s decree. This would allow for the flexibility necessary to incorporate the best available information on how to resolve the legal problems identified in the decree. It would also allow for maximal democratic input into court-monitored compliance.

It would be a mistake to dismiss these intentions as merely ideological. Scholars have recognized that flexibility between schemes of compliance with court-ordered relief is important to effective reform of historically and systematically rights-violative institutions.186 Making rules open to revision, negotiation, and reassessment allows decrees to better respond to changing conditions and gives stakeholders buy-in into the compliance regime.187

However, excessive deference to administrators’ requests to modify decrees will allow defendant institutions to chip away at court-ordered remedies and water down public law protections. Horne attempted to balance the importance of court monitoring against the values of new policy insights with a heavy hand, instead of crafting a means of testing both the adequacy of the new policy insights and the good-faith intentions of the actors implementing them. Instead of allowing new policy insights to constitute changed circumstances under Rufo, courts should allow experimental modification of decrees that permits defendants to implement new insights, but automatically reverts to the prior decree if the new insights fail to adequately safeguard the objectives of the decree.

If a court modifies a decree and the modification makes the decree less effective, the plaintiff is at a severe disadvantage: in order to restore the original, effective decree, the plaintiff would need to file a Rule 60(b)(5) motion, show changed circumstances, and convince the judge to again modify the decree. Instead, when asked to modify an ongoing IRL decree, courts should create “hammer” provisions that cause a decree to automatically revert to its original form if the modifications prove to be less effective within a certain period of time.188 This provision would allow the court to introduce a modification to the decree and monitor its results closely: if those results show that the modification created unlawful results, the order would reinstate the previous state of affairs automatically, without the need for another motion to modify the decree again. The same logic could be applied to modifications in the other direction. For example, a court could approve a modification making a school desegregation plan even more aggressive, as long as the court specifies circumstances under which the original decree should “snap” back into place. However, in this scenario, the criteria for determining whether to revert to the original decree are less clear, whereas experimental modifications that loosened a decree would presumably snap back when an underlying federal law violation reappeared.

When combined with monitoring and measuring requirements, this experimentalist model has several advantages. First, it encourages defendants to petition for modification only when there are good-faith policy insights that would improve the efficacy or efficiency of compliance mechanisms, since modifications that create lower levels of compliance will be rolled back. This saves courts time and energy as well, since it shifts the burden to petitioning defendants to ensure that proposed modifications are effective. Second, it keeps courts out of the business of policymaking, returning the task of administrative oversight to elected officials and professional bureaucrats. This helps ameliorate one of the chief criticisms of IRL: that it uses the courts, instead of the legislative or executive branches, to effectuate public policy changes. Third, and most importantly, experimental modification of decrees allows agencies and institutions to respond to changes in information without reducing the level of compliance with public law that the decree requires. Courts can promote flexibility by erring on the side of modification with less concern about the consequences of ineffective remedies, and plaintiffs can have firmer and speedier guarantees that their rights will be vindicated.


A commitment to civil rights has been an important aspect of American life for generations. Yet the oft-told success story of America’s school desegregation movement is not complete: the story continues in towns all over the country, not just Tuscaloosa. In fact, desegregation decrees are in place in over 300 school districts in the United States.189 But school desegregation is only a small, if highly visible, aspect of the rise and fall of the IRL project in the United States: the standards brought to bear on motions to modify court-ordered civil rights protections affect millions of individuals every day. Decades of results, spurred by successful litigation, are now threatened by new standards that make it easier for past offenders to roll back those victories. The results of these victories do not need to be dramatic to be important. IRL decrees help to keep prisoners safe, provide dental services to those on Medicaid, and help underprivileged children learn English.

This Note has sought to identify emerging areas of doctrinal confusion in IRL and to suggest some solutions to the real-world problems courts face when adjudicating proposed modifications or dissolutions of IRL decrees. The critics of IRL do have a point: these decrees can hamstring local governments, sever the ties of political accountability, and block the implementation of new policy insights. However, Horne went too far toward making decrees easier to change or vacate, and left considerable confusion in several important areas of law. The path forward lies in an even-handed balance between the benefits of IRL and the legitimate doctrinal innovations that Horne began to usher in.

In large part, courts are in the best position to balance these interests, since the myriad factual patterns they face require the use of fact finding, discretion, and judgment to determine when, for example, a durable remedy is in place. However, the policy issues surrounding IRL decrees also raise questions about the capacity and authority of federal courts to do the balancing. Another option, of course, would be to modify Rule 60(b)(5) itself. This avenue would involve input from a number of sources, as the Rules Enabling Act requires that a proposed change be (1) subject to public comment; (2) approved by committees that include procedural scholars, judges, members of the bar, and representatives of the Executive; (3) adopted by the Supreme Court; and (4) subjected to a six-month period of congressional consideration.190 This process can respond to some policy concerns regarding accountability, transparency, expertise, and the separation of powers.191 The formal rule-making process would also provide an opportunity for further empirical study of the effectiveness of institutional reform decrees.192

However, a congressional response might suffer from some of the same pitfalls as the policy debate—namely, breaking along party lines. The Prison Litigation Reform Act,193 which reduced the standard for terminating institutional reform decrees in prison cases, was passed in large part because of a Republican push to, in Senator Dole’s words, “restrain liberal Federal judges who see violations o[f] constitutional rights in every prisoner complaint and who have used these complaints to micromanage State and local prison systems.”194 The Democrats, on the other hand, largely opposed the bill. They warned that it “would radically and unwisely curtail the power of the Federal courts to remedy constitutional and statutory violations in prisons, jails, and juvenile detention facilities,”195 and would “set a dangerous precedent of stripping the Federal courts of the ability to safeguard the civil rights of powerless and disadvantaged groups.”196 Two other attempts to reduce the standards for termination of decrees have also been proposed by Republicans in Congress, but both failed.197

Ultimately, however, this is not the place to resolve the debate regarding the merits and demerits of legislative versus judicial approaches to doctrinal change. What gets lost in the debates over institutional reform, whether they take place on the floors of Congress or in academic journals such as this one, is the very real impact on human lives that these decrees have. Both sides of the debate have the best interests of those lives in mind, whether they think they are best served through continued court oversight of state functions or through an increase in states’ freedom to innovate in their policies. Courts should be free to implement the best observations from each side of the IRL debate. That can be done by applying a more flexible standard for modification of consent decrees and collusive injunctions than for litigated injunctions, ensuring that the objectives of a decree have been satisfied before dissolving it, and requiring that a durable remedy be in place before dissolving a decree. Only by striking this balance can courts both effectively and efficiently safeguard civil rights going forward, and protect the future of institutional reform in the United States.