Printer Friendly

Rethinking consent decrees: how federal-court decrees in child welfare can harm those they are supposed to help and upset the federal-state balance.

Executive Summary

Aconsent decree is a judge's order, entered after a voluntary agreement between parties to a law-suit. Unlike an ordinary judgment, which awards relief or dismisses a case and then ends the court's jurisdiction, a consent decree gives a judge ongoing supervisory power to enforce the decree, by contempt if necessary. Consent decrees became popular in the 1970s as a way of resolving class-action lawsuits filed to "reform" state and local governments programs or agencies, such as prisons, special-education programs, and child-welfare agencies. Such decrees were viewed as a win-win-win for plaintiffs, defendants, and courts: plaintiffs' attorneys could demand concrete changes to a government program's operations, defendant governments could use the court order as leverage to increase funding, and courts could avoid deciding the complicated issue of whether a government program was actually violating an individual's federal constitutional or statutory rights.

This report examines the efficacy of consent decrees as a tool for reforming government, with a special focus on decrees governing child-welfare agencies, a subset of consent decrees that has been a growth industry and serves as a straightforward example of a broader problem. The report concludes that consent decrees often fall short of expectations for numerous reasons. First, because of the court's ongoing supervision and the parties' lack of incentive to change the status quo, a consent decree can take on a life of its own, often outlasting its original signatories for years or even decades. Second, consent decrees require substantial government expenditures--monies that no matter how well spent, might be directed to other, higher public priorities if allocated as part of the ordinary legislative process. Third, consent decrees leave governmental officials with little recourse but to follow the plaintiffs' attorneys' road map for reform, no matter the efficacy or efficiency. Finally, consent decrees can harm the very beneficiaries they are supposed to help by creating barriers to adopting new service approaches and by instituting priorities that may be at odds with the best interests of a specific child. A consent decree that locks in a certain type of child-welfare practice can become a barrier to adopting a cutting-edge approach that would actually deliver more effective services.

This paper concludes that Congress could improve the consent-decree environment. Specifically, Congress would do well to consider reforms that require federal courts to reexamine, on a regular basis, whether a state or local government program is violating federal law. In the absence of such a violation, the court must vacate the decree and relinquish jurisdiction. Such reforms would strike the appropriate balance between protecting vulnerable citizens and allowing elected government officials to do their job of prioritizing need, implementing effective solutions, and delivering outstanding service to the citizenry.

An Introduction to Institutional Reform Litigation by Consent Decree

In their groundbreaking 2003 book, Democracy by Decree, (1) Ross Sandler and David Schoenbrod explained why prisons, schools, child-welfare agencies, and other important governmental institutions had come to be controlled by courts and attorneys rather than elected officials and executive-branch agencies. Using many real-life examples, they illustrated how this development often resulted in massive government expenditures and a staggeringly small return on investment--all at the expense of democracy itself, as governmental institutions found they had little recourse when a high-ranking governmental official capitulated and ceded institutional power to the federal courts.

In 2009, the United States Supreme Court issued an opinion, Horne v. Flores, 557 U.S. 433 (2009), that appeared to restore the federal-state balance in the context of federal-court consent decrees. Noting that "federalism and simple common sense require [a district] court to give significant weight" to the position of government officials moving to dissolve a consent decree, the Court made clear that when a state or local government is in compliance with federal law, the court must vacate the consent decree, even if the state is continuing to violate the decree's terms. In other words, once a state or local government complies with federal law by any means--even means other than those the injunction imposes--the federal court is obligated to return control to state and local officials.

Regrettably, the Horne decision has done little to stem the federal courts' enthusiasm for consent decrees. Focusing solely on child-welfare systems, numerous state and local agencies operate under some form of federal-court oversight. A number of these consent decrees have been in place in excess of 20 years, and the Illinois child-welfare system is burdened by 10 different consent decrees, including one that has lasted nearly 40 years. (See the Appendix.) (In all, Illinois operates under 80 consent decrees.) A comparison of these decrees reveals considerable variability about what is considered a "federal right," and a high level of micromanagement over the agencies' day-to-day operations.

The purpose of this paper is to re-examine the consent decree as a tool for institutional reform now that 13 years have passed since the publication of Democracy by Decree and lower federal courts have had time to digest and apply Horne. The paper is structured as follows:

I. What Is a Consent Decree? This part explores what consent decrees are and how they are supposed to function.

II. Consent Decrees Currently Governing Child-Welfare Systems. This part looks at current information on the content of mandates contained in child-welfare consent decrees and their overall use (which states are subject to such decrees, for how long they have been subject to those decrees, etc.), and the drivers of this expansion.

III. How Child-Welfare Consent Decrees Harm Children--and Federalism. This part examines how consent decrees in child-welfare cases help or harm the interests of the government, plaintiffs, and citizens.

IV. The Supreme Court and Consent Decrees. This part considers how the Supreme Court has ruled on questions related to consent decrees, and the extent to which those rulings have been implemented.

V. What Congress Could Do. This part sets forth a set of policy recommendations for solving the consent-decree problem (as well as the problem of institutional decrees entered by federal courts without consent) and explains how Congress could fix the problem.

I. What Is a Consent Decree?

Consent decrees result from litigation brought to reform state and local government agencies and institutions. Reform-minded attorneys identify a government-administered program that they believe needs to change, build a legal case around a federal statutory or constitutional provision that has been arguably violated, and file suit. The alleged violation of federal law then becomes a "legal hook" for imposing broad reform. (2) This template has been used to advocate for reform of nearly every type of government program, from special education to environmental protection to health care to prisons. (3) This special litigation cottage industry has resulted in court supervision of jails and prisons in all 50 states and of educational practices in more than 600 school districts. (4) The industry even has its own name: "institutional reform litigation." (5)

So why the continued popularity of consent decrees? State and local officials often view a decree as a solution to a thorny problem: an underperforming governmental agency. A consent decree may give an executive-branch member a hammer to demand more funding from the legislature. It allows the official to prove to the public that a serious problem is being addressed. And because the decree typically includes a detailed remedial plan--written to conform to a "road map" by which plaintiffs' attorneys seek to guide multiple systems in different states--a consent decree has all the appearances of a functional plan

Plaintiffs like consent decrees, too. Settlement avoids what could be a long and expensive trial and guarantees that plaintiffs' attorneys will be paid millions of dollars as a "prevailing party." Even if plaintiffs decided to roll the dice on a trial and won, the court would only be able to order specific remedies based on the facts; in contrast, a consent decree can go far beyond the facts and dictate all manner of government activity. And, because the focus of consent decrees is on what the parties agreed, rather than what the law requires, consent decrees are easy for plaintiffs to enforce and difficult for defendants to terminate

Consent decrees are also appetizing to judges. Reforming government agencies is difficult, requiring extensive information and expertise as well as balancing a myriad of policy, political, and budget concerns. Lacking the time and capacity to make such decisions, courts are only too happy to let the parties "work it out." And once a consent decree is in place, a judge feels that he or she is doing something to solve the problem and will often stick around to ensure progress continues to be made.

To be sure, institutional-reform litigation and consent decrees are sometimes needed to alleviate truly heinous situations, such as deliberate governmental indifference to prison conditions, or child abuse. They are needed tools. But there are also limits to an institutional consent decree's effectiveness.

Scholars from across the political spectrum have recognized the limited success of consent decrees in reforming governmental institutions. (6) And, in Democracy by Decree, Sandler and Schoenbrod debunked the assumption that consent decrees always achieve the vision their creators intended. The problems are numerous and are present from the get-go: the plaintiffs' attorneys are free to advance their own view of good public policy; it is possible for attorneys to advance an agenda at the expense of their clients; confidentiality in negotiations and re-negotiations ensures that the public will have no input on the public policy adopted; plaintiffs' attorneys have the de facto power to veto modifications in the plan because, in the view of the attorneys and most judges, consent decrees are like private contracts, and a deal is a deal; and because consenting parties cannot appeal, it is extremely difficult for public officials to obtain appellate review, even for agreements that were struck by predecessor officials in administrations that have long since left office.

As noted above, the limited success of consent decrees has been well studied and documented. (7) But equally troubling are the harms consent decrees can visit on the plaintiffs they are intended to help:
Lawyers for homeless citizens might, for example, obtain a decree to
improve homeless shelters, but their homeless clients may prefer other
services and detest shelters because they are dehumanizing. Lawyers for
prisoners obtain a decree requiring the closing of an old jail and
construction of a new one on an island, but many prisoners and their
families prefer the old jail because its in-city location permits
easier visits. Lawyers for African Americans obtain a decree calling
for special help for African-American children but oppose letting
lawyers for Chinese-American children who will be disadvantaged by the
decree intervene in the litigation. Lawyers for an environmental group
obtain a court order forcing a local government to spend its scarce
capital funds to meet a clean water act requirement that has limited
environmental benefit but that cause the local governments to delay
other capital improvements that have greater environmental benefits. (8)

Or, in the case of child-welfare systems, the subject of this report, lawyers for children in a state foster-care system may forbid the use of residential care without regard to the best interests of a particular child, or may impose overly rigorous requirements before a state is allowed to dispense psychotropic medicines. Under such circumstances, the state may be forced to deny such medication to children who have a demonstrable need for it

And when federal judges impose consent decrees, voters also lose. They are no longer able to hold their elected representatives accountable when governmental agencies fail. (9) Instead, those officials are supplanted, not by federal judges directing state and local governments to comply with the law, but by plaintiff groups controlled by well-intentioned lawyers who shape the rules concerning how governmental officials are to comply with the law and do so with the imprimatur--and threat of contempt proceedings--of the federal courts.

The relationship between federal courts and state and local governments was not always this way. In our country's earliest days, litigants sued governments and officials, but these lawsuits were generally limited to whether the government had improperly taken something from a citizen, such as property, a contract right, or some other property-based right. (10) Indeed, until very recently, the Supreme Court had always prohibited plaintiffs from using the Due Process Clause to claim a fundamental right to "get" something from the government in the form of status or benefits: "[T]he Due Process Clause generally confers no affirmative right to governmental aid, even where such aid may be necessary to secure life, liberty, or property interests." (11) "Although the liberty protected by the Due Process Clause affords protection against unwarranted government interference," "it does not confer an entitlement to such [governmental aid] as may be necessary to realize all the advantages of that freedom." (12)

As federal, state, and local governments grew and began dispensing ever-increasing benefits to provide welfare, education, housing, and health care, citizens took note and began suing over how these benefits were distributed and administered. The new entitlement sentiment crystallized in a 1964 article that appeared in the Yale Law Journal. (13) In it, the author made the case that these new government benefits, "the new property," were just as important to citizens and the legal system as the government's taking of private property. In other words, courts should give equal weight to a citizen who claims a government benefit as to a citizen who has forcibly relinquished a house or land.

But there is an enormous difference between the two types of litigation. Because violations of constitutional and common-law rights represent "sins of commission" rather than "sins of omission," those rights are easy to enforce. (14) If the government trespasses on, or takes, private property, the courts order the government to stop or pay money to the landowner. If a private party breaches a contract, courts rarely order specific performance but simply award damages.

In contrast, soft rights, such as a "right" to clean air, an adequate education, or a safe foster-care system typically involve positive rights; they tell government what to do, rather than what not to do. (15) These positive rights are often aspirational and thus impossible to fully vindicate, at least in a world of limited resources. Public officials have difficulty knowing how to obey a court order that enforces positive rights. (16) For example, public officials in New York City have spent decades trying to figure out how to comply with a court-enforced consent decree that required the city to provide a free appropriate education for all children with disabilities. (17) Despite increasing the city's special-education budget from $434 million to $2.685 billion over a 20-year period, the city remained under court supervision, notwithstanding the fact that the city's per-capita expenditures on special-education students are now nearly four times higher than on general-education students ($50,698 per special-education student versus $13,802 per general-education student in fiscal year 2015). (18)

In 1976, as plaintiffs' attorneys and judges continued to push for ever-more-sweeping court supervision of state and local governments, the Harvard Law Review published what remains one of the most sweeping defenses of the practice. (19) In it, Professor Abram Chayes contrasted the old way of doing things, so-called "private law litigation," with the new trend, which he named "public law litigation." (20) In Chayes' view, courts were often better than elected officials at resolving policy problems and would frequently produce better outcomes. (21) And at the time, Chayes' conclusion mirrored public perception. After all, the public was widely supportive of the Supreme Court's decision in Brown v. Board of Education and its progeny that successfully desegregated the public schools.

But Chayes appears to have misunderstood the role of the judge in public law litigation. Policy choices require the consideration of reams of information and the balancing of competing interests--quintessential legislative functions. Because courts lack the time and capacity to engage in such activity, they generally shift the policy-making function to the parties by encouraging them to consent to a decree. These result in what the authors of Democracy by Decree have called "the controlling group"--consisting of plaintiff lawyers, their experts, defendant lawyers, officials involved in the negotiations on the contents of the decree, and various court-appointed masters or monitors. (22) This controlling group undermines Chayes' assumption that judges are calling the shots:
[Chayes] wrote in 1976 that judges, moderated by the inherent
conservatism of the judicial community, would base public policy on
reasoned and principled decision making. This ideal is rarely achieved.
The bulk of the court orders in institutional reform cases result from
bargains that, like the legislation they most resemble, are not
necessarily logical or principled. When a proposed order is submitted
for judicial signature on consent of the parties, judges are freed from
having to choose among policies and can remain true to a still powerful
judicial culture based on the separation of powers, which expects
judges to let elected officials manage government.... The judge anoints
the controlling group and keeps it going by pressuring the parties and
holding the defendants and their successors in office to the bargain.

And once the controlling group has achieved entry of an initial consent decree, what the law requires no longer matters. The statutory or constitutional "legal hook" that was the impetus for the decree falls away, and the law becomes "what[ever] the controlling group says it is." (24) Agreed-upon plans that fail to work are toxic. The public officials charged with compliance can be held in criminal contempt if they attempt to deviate from the plan. And woe to the public official who seeks to modify the decree. Judges and plaintiffs' attorneys analogize consent decrees to contracts, and "[j]udges resist allowing modifications unless plaintiffs' attorneys consent, even if the term sought to be modified is unnecessary to correct a violation of law." (25)

II. Consent Decrees Currently Governing Child-Welfare Systems

With this background, it is appropriate to consider how plaintiffs' attorneys wielding consent decrees have taken over state and local child-welfare systems. In a 2000 study, the National Center for Youth Law identified some 57 child-welfare institutional-reform lawsuits involving 36 states, with consent decrees governing at least 35 of the lawsuits. (A current list of these decrees, as well as those that have been entered since 2000, appears in the Appendix.) Indeed, in only a handful of cases did a judge impose any remedy other than a settlement agreement to which the parties consented. (26)

Many of these lawsuits have now reached legendary status. In Connecticut, a plaintiff class sued the governor and the commissioner of the Department of Children and Families in 1989, reaching a consent-decree settlement in January 1991. (27) The original complaint alleged dangerous and unlawful practices, including the failure to provide adequate child protective services, the failure to make reasonable efforts to keep families together, and the failure to provide minimally adequate staffing. (28)

After the first 12 years of court oversight, the plaintiffs' attorneys filed a motion for contempt that led to the voluntary handover of express management of the system to a task force headed by a court monitor. That process resulted in a Modified Revised Exit Plan, which the court approved in 2006, focusing on required outcome improvements in 22 areas. In the most recent quarterly report on Connecticut's performance, covering the first quarter of March 2015, the court-appointed monitors note that Connecticut maintained compliance with only 15 of the 22 measures. (29) Nearly a quarter of a century after first stipulating to a consent decree, there is no end of court supervision in sight for the State of Connecticut. And one would be hard pressed to identify the federal statutory or constitutional provisions that Connecticut allegedly violates today. As the monitors explain, the most "critical" areas of noncompliance in the first quarter of 2015 had to do with Connecticut's "case planning process, meeting children's service needs, appropriate visitation..., excessive caseloads for Social Work staff and appropriate discharge outcomes (education, work, and military service) prior to discharge from DCF custody for older adolescent youth (ages 18+)." (30)

In the District of Columbia, a similar lawsuit was also filed in 1989. (31) In one of the rare cases that actually went to trial, the court found for the plaintiffs in 1991 and ordered reforms to be undertaken. After several plans and a period of receivership, the District entered into a so-called Implementation and Exit Plan in 2010. (32) But there is no exit in sight. According to the most recent monitoring report, covering the last six months of 2014, the District has met only 74 of its 88 Exit Standards, and the District must have perfect compliance to exit. (33)

Plaintiffs' attorneys sued the Governor of Wisconsin on behalf of all Milwaukee children in child-welfare custody in June 1993, with a final settlement reached in 2002. (34) In the past 13 years, the City of Milwaukee has made extraordinary improvements in its foster-care system, with outstanding results in safety, permanency, and timeliness of adoptions. Yet the city remains under court supervision because the program has yet to satisfy the final unfilled settlement requirement: The most recent settlement agreement report shows that 88 percent of children in foster care experienced three or fewer placements in the previous 36 months as of June 30, 2015. "[D]efendants need to reach 90 percent compliance before they are eligible to exit monitoring and enforcement." (35) There is no federal statutory or constitutional provision that guarantees a child's right to three or fewer foster-care placements over a 36-month period; yet based solely on that slim reed, the City of Milwaukee remains beholden to the plaintiffs' attorneys and court-appointed monitors, some 13 years after its original settlement agreement.

In Maryland, plaintiffs' attorneys filed a class action in 1984 on behalf of Baltimore children placed in Maryland's foster-care program. (36) The parties submitted a settlement agreement to the court in 1988 that called for reduced worker caseloads, an improved system for providing medical care to foster children, providing assistance to natural parents to avoid foster care when possible, and increased recruitment of new foster homes. It was anticipated that this process would take two years. (37) Twenty-one years later, in 2009, the parties entered into a modified consent decree that runs a full 41 pages. (38) The decree imposes a broad range of practice standards that are divorced from any federal statutory or constitutional requirement, including the requirement that all case reassignments occur within five working days and dictating the content of a transfer document that must accompany such reassignments. (39) Exit from court supervision is not available until Maryland has complied with every commitment in the document for a period of three consecutive six-month reporting periods. (40)

A similar class-action lawsuit was filed against the City and State of New York in 1995. (41) After nearly a year of settlement negotiations, the parties approved a consent decree entered in 1999. (42) By 2001, the parties had already returned to court on plaintiffs' motion seeking an order directing defendants to act more diligently in implementing a statewide information-management system. The district court extended the agreement's term and directed the state to file semiannual reports with the plaintiffs until the court determined that the state had fully complied with the parties' agreement. There is little activity that appears in the court's online docket after that ruling, save a report in July 2008. But, in mid-2015, the plaintiffs' lead lawyer announced that she was filing a new class-action lawsuit against New York City and the State of New York for keeping children in foster care too long. (43) All this despite the fact that the city has increased the foster-care system's budget by $100 million over the past two years, resulting in the hire of hundreds of additional workers, the introduction of new parent services based on the latest in childhood-development science, and the improved assessment of the needs of children in the foster-care system. (44)

Then there is the State of Michigan, where plaintiffs' attorneys filed a class-action lawsuit involving the state's foster-care system in 2006. (45) After several years of discovery, then-Governor Granholm capitulated to a detailed consent decree in 2008. After Governor Snyder was elected in 2010, the state averted a motion for contempt and renegotiated limited portions of the consent decree. In all, the decree mandated nearly 240 different substantive and procedural measures that the state needed to meet and hold for 18 months before it was eligible for exit.

In late 2014, the state filed a motion to vacate or modify the consent decree in substantial part. At the court's request, the state and plaintiffs' attorneys instead entered into another renegotiation of the decree, which did not conclude until nearly the end of 2015. Although Michigan has made remarkable progress since 2011, the provisions that remain reflect incredible micromanagement of the state's operation of its foster-care system. These measures will be discussed in more detail in Part III. (46)

III. How Child-Welfare Consent Decrees Harm Children--and Federalism

In public law litigation, consent decrees have become a regular occurrence. Such decrees have many unintended consequences, including diverting public resources and attention from where they might best be spent and focusing them on issues directed by a plaintiffs' attorney control group. But two of the most significant harms are to the very plaintiffs whom the decrees are supposed to protect, and to the sovereign authority of state and local governments.

Consider the potential harm to the children that child-welfare consent decrees are supposed to protect. Michigan's current consent decree provides numerous examples:

* It is now a standard section in nearly every child-welfare consent decree to dictate certain caseload ratios, such as 12 children or families for every one caseworker. Michigan's original, modified, and recently renegotiated consent decrees all include such provisions. Such provisions lock in a certain type of practice and preclude experimenting with different arrangements that might provide better services to children. For example, some state foster-care systems have started using a team-based approach to service. With this approach, a child is assigned a small team of workers who can collaborate to assure that the child's needs are being met. But Michigan would be unable to initiate, or test, this approach, because team-based assignments would cause the state to be out of compliance on caseload ratios.

* Most child-welfare consent decrees do not address so-called psychotropic medications--medicines that have the capability of affecting the mind, emotions, and behavior. (It is common for children in the child-welfare system to be taking such medications, which can be prescribed to treat everything from Attention Deficit Disorder to a Bipolar Disorder.) But Michigan's recently renegotiated consent decree includes numerous provisions involving psychotropic medications. For example, Michigan's child-welfare agency must ensure that informed consent is obtained and documented in writing in connection with each psychotropic medication prescribed to a child in the agency's custody. But if a child is in the agency's custody, it is likely because the child has been abused or neglected by his or her parent, and the only person who can provide informed consent is the parent. While there are workarounds, children with severe emotional and behavioral disorders could be denied needed psychotropic medications while the agency tries to comply with the decree's requirements.

* In the same section of Michigan's recently renegotiated agreement, the agency is reasonably required to ensure that a qualified physician completes and documents an oversight review of a child whenever one or more criterion is present. But the agreement requires that the review be completed using a particular form selected by the plaintiffs' attorneys. So if a doctor chooses to use her own form, the agency is out of compliance with the consent decree, even if the information on the two forms is exactly the same. The result is to divert agency resources away from servicing children and toward policing the forms that doctors use.

* In a different section of the recently renegotiated agreement, the decree requires that Michigan's medical, dental, and mental-health-services plans contain certain information listed in an agency policy. The provision includes no option for the agency to revise its policy if the cited information is either inadequate or superfluous. And if the agency falls short in complying with this provision, it is ineligible for exit--even if that lack of compliance resulted from a worker focusing on actually providing care and services to the child. This encourages workers to prioritize filling out paperwork rather than providing care.

* The recently renegotiated agreement also contains provisions that limit the use of emergency or temporary facilities (such as shelters) and residential-care placements. In some instances, however, these placements are actually in the best interests of the child. For example, depending on the circumstances, a residential-care placement may be an appropriate placement for a child at a given time and thus would be in the child's best interest. Yet the decree places agency workers in the position of deciding whether to risk an order of contempt or to do what is best for the child.

* Since 2001, the federal government's Administration for Children and Families has promulgated Child and Family Services Reviews, known as CFSRs, to measure state performance on various factors related to child safety, permanency, and well-being. (These are discussed in more detail in Part V.) The most recent revisions to the regulations, the so-called "round three" CFSRs, involve two factors related to safety and five related to permanency. Only four, very small, states are currently in compliance with all seven of these factors. But the plaintiffs' attorneys required Michigan to have reached compliance with all seven to be eligible for exit. This is problematic, because some of the five permanency factors pull in different directions. For example, Michigan has excellent performance in finding a permanent placement for a child-in-care within 12 to 23 months, but the state struggles to do so in fewer than 12 months due to court delays and a desire to get a placement right the first time, so the child's placement is truly permanent. As a result of Michigan's deliberateness, it has the second-best performance in the nation in the separate permanency metric, "re-entry to foster care in 12 months." (In other words, Michigan performs exceptionally well at addressing a child's needs the first time, preventing the child from returning to the child-welfare system after a first discharge.) Under the decree, however, Michigan will have to increase the number of children given a permanent placement within 12 months after they enter the system, in other words, place children more quickly. This is likely to result in less stable outcomes and a commensurate increase in the number of children who end up returning to the system, lowering Michigan's exemplary re-entry rate. It is not obvious that this is in a child's best interest, but is the natural consequence of what the decree requires.

In these and many other examples, consent decrees reflect the underlying principle that nearly every aspect of a government agency's performance should be subject to court-enforced rules. Such an approach almost invariably leads to more rules and may actually decrease the agency's substantive performance. A superior alternative is accountable officials exercising judgment: there could be no better protection for a child in difficulty than an accountable caseworker who cares deeply about the child and her welfare. (This point is made in a more comprehensive way by Philip Howard's recent book, which makes the case that government cannot work if its every choice is dictated by rule. (47))

What's more, the practical reality of consent decrees is that they divert millions of dollars to plaintiffs' attorneys and court-appointed monitors. Michigan has already paid more than $10 million to plaintiffs' attorneys and monitors, and continues to pay the monitors more than $1.5 million annually. Some of these dollars were undeniably well spent, as the monitors and attorneys have assisted Michigan in improving its service delivery to children in the child-welfare system. But state budgeting is a zero-sum game. And despite the substantial progress that Michigan has already made, every going-forward dollar spent on monitors, attorneys, and compliance with needless burdens could be redirected toward hiring additional staff to further advance the state's improvement goals. Indeed, there would be more general-fund sources for all manner of additional services to foster children. Those dollars do not include the hidden cost of completing paperwork necessary to show compliance with the consent decree and the cost to caseworker morale, effectiveness, and turnover, of spending time filling out forms rather than caring for children. Nor do they include the tens of millions of dollars that have been spent in other states.

Wholly apart from harms to children, child-welfare consent decrees--indeed, consent decrees more generally--have a separate and distinct harm on state and local government sovereignty. The Supreme Court has recognized this reality and instructed lower federal courts to be cognizant of allowing state and local governments to function on their own and to defer to local policy judgments. For example, in Milliken v. Bradley, the Court directed the federal courts to "take into account the interests of state and local authorities in managing their own affairs." (48) And in Lewis v. Casey, the Court observed that "it is not the role of courts, but that of the political branches, to shape the institutions of government in such fashion as to comply with the laws and the Constitution." (49) When courts allow plaintiffs' attorneys and public officials to enter broad consent decrees that are not narrowly tailored to specific violations of federal statutory or constitutional law, they invade the province of the state and local executive and legislative branches, removing authority from accountable, elected officials and transferring it to a control group that is not accountable to voters.

Such a system is antithetical to our federalist system of government. Under the Tenth Amendment to the United States Constitution, the "powers not delegated to the United States by the Constitution, nor prohibited by it to the States, are reserved to the States respectively, or to the people." (50) In other words, our country is founded on the principle that our federal and state systems of government are unique, distinct, and complementary. And although the federal government has amassed substantial power over the past century, the Constitution acknowledges that state and local governments are generally the best suited to address complex social policy. This federalism principle applies as equally to the federal courts as it does to Congress, the President, and federal agencies. Further, when federal courts give their imprimatur to consent decrees that dictate the most detailed functioning of state executive agencies and programs, they impair state and local sovereignty not only at great cost to efficiency and effectiveness, but also to the Constitution itself.

Consider the wide range of requirements imposed on state and local child-welfare systems in different jurisdictions' consent decrees. If child-welfare research and methodology dictated certain best practices, one would expect those practices to be reflected in every state's consent decree. But consent-decree requirements vary substantially and often appear to represent nothing more than the parties' relative bargaining power. A summary of some key child-welfare consent-decree provisions demonstrate this fact (see table 1).

In sum, consent decrees are not just a matter of semantics and taking control away from state and local officials. Rather, they have real-life consequences that sometimes present the individuals forced to implement the decree with a catch 22: do what is best for a child and place the government in a position where it is not complying with a court order, or follow the court's directive, even if at the expense of a child's best interests. No caseworker should be put in that position. But there is no safety valve, as the courts generally take a "hands off" approach and leave it to the parties to work out conflicts. This leaves state and local officials covered by an existing consent decree with little negotiating leverage and forces them to strike the best deal they can, even if it has deficiencies.

IV. The Supreme Court and Consent Decrees

The United States Supreme Court has recognized that state sovereignty and the importance of the federal-state balance of power limit the scope of federal-court remedial power. (51) Indeed, in "a system of federal courts representing the Nation, subsisting side by side with 50 state judicial, legislative, and executive branches, appropriate consideration must be given to principles of" state authority in determining the availability and scope of court-ordered relief. (52) As a result, "federal courts must be constantly mindful of the special delicacy of the adjustment to be preserved between" the federal power to dictate how a state must operate and the state's authority to administer its own programs. (53)

Federal courts must also keep state sovereignty in mind as part of their discretion when ordering states to operate in a certain way because of the extraordinary nature of the remedy. (54) In Pennsylvania v. Williams, for example, a federal court appointed a receiver for a failed building and loan association over the objection of the Pennsylvania banking secretary. (55) The Supreme Court reversed, holding the district court should have deferred to Pennsylvania's plan to supervise the association's liquidation: "It is in the public interest that federal courts...should exercise their discretionary power with proper regard for the rightful independence of state governments in carrying out their domestic policy." (56)

There are two fundamental reasons why state sovereignty affects courts' equitable discretion. To begin, the Supreme Court has said so. (57) In Rufo v. Inmates of Suffolk County Jail, the Court pronounced that "principles of federalism and simple common sense require the court to give significant weight" to the state government officials implementing the court's order. (58) And in Frew ex rel. Frew v. Hawkins, the Court noted as "legitimate" the concern that "enforcement of consent decrees can undermine the sovereign interests and accountability of state governments." (59)

In addition, failure to account for state sovereignty in dispensing equitable remedies creates a significant "dead hand" problem. Consent decrees undermine democratic accountability after state leadership changes. Although consent decrees result when government officials agree to the terms, such decrees bind those officials' successors, who frequently had nothing to do with the original decision to enter into an agreement. (60) That is certainly the case in institutional-reform litigation, where the parties' consent decree, though perhaps even renegotiated, is often a remnant of an agreement made by an administration that has been out of office for many years. "The Framers fully recognized that nothing would so jeopardize the legitimacy of a system of government that relies upon the ebbs and flows of politics to 'clean out the rascals' than the possibility that those same rascals might perpetuate their policies simply by locking them into binding contracts." (61) That is why, if "not limited to reasonable and necessary implementations of federal law, remedies outlined in consent decrees involving state officeholders may improperly deprive future officials of their designated legislative and executive powers.... A State, in the ordinary course, depends upon successor officials, both appointed and elected, to bring new insights and solutions to problems of allocating revenues and resources. The basic obligations of federal law may remain the same, but the precise manner of their discharge may not." (62)

Accordingly, the Supreme Court has directed federal courts to take a "flexible approach" to motions seeking to vacate or modify institutional consent decrees. (63) In doing so, "courts must 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.'" (64)

In Horne v. Flores, a federal district court entered an order against the State of Arizona. The court concluded Arizona had violated the federal Equal Educational Opportunities Act of 1974 by failing to provide sufficient funding to enable English Language Learner students to overcome language barriers. After the state had continued to violate the act for many years and had incurred a series of court-imposed contempt fines, Arizona enacted legislation that finally brought the state into compliance with the act but failed to allocate enough monies to comply with the court order's term. Plaintiffs filed yet again for contempt, and Arizona moved to vacate the injunction. (65)

The Supreme Court held that if a state is complying with federal law and has corrected the violations that resulted in the court order, then a district court should vacate that order, even if the state is continuing to violate it: "If a durable remedy has been implemented, continued enforcement of the order is not only unnecessary, but improper." (66) If the state complies with federal law by any means--even means other than those the court order imposes--the state satisfies the test for vacating court-ordered relief and the federal court is obligated to return control to state officials. In other words, "[W]hen the objects of the decree have been attained, responsibility for discharging the State's obligations [must be] returned promptly to the State and its officials." (67)

In so holding, the Supreme Court discussed the federalism principles at stake in institutional reform litigation, particularly the dead-hand effect. By "confining the scope of its analysis to that of the original order," the court of appeals "insulated the policies embedded in the order...from challenge and amendment." (68) Those policies "were supported by the very officials who" failed to appeal the order, "and, as a result, were never subject to true challenge." (69) "To determine the merits" of Arizona's Rule 60(b)(5) claim for relief, the court of appeals "needed to ascertain whether ongoing enforcement of the original order was supported by an ongoing violation of federal law." (70)

The Court criticized the court of appeals for discounting the state's "[s]tructural and management reforms." (71) The court of appeals "missed the legal import" of these changes, namely, that the reforms might have brought the state into compliance with federal law even if the state had not satisfied the injunctive order. (72) "A proper Rule 60(b)(5) inquiry should...ask whether, as a result of structural and managerial improvements," the State is now in compliance with federal law. (73) If it is, continued enforcement of the original order is inequitable and "relief is warranted." (74)

In Horne, the Court was dealing with a court-imposed decree, as opposed to a decree to which the State of Arizona had "consented," as is the case with a child-welfare consent decree. But the Court made clear in its opinion that its holding applied equally to consent decrees. For example, in discussing the serious federalism concerns that arise when a federal court undertakes supervisory control of a state institution, the Supreme Court referenced state "consent" or the term "consent decree" no fewer than 12 times. (75) And the Court specifically held that Rufo's "flexible approach" to considering Rule 60(b)(5) motions to vacate "allows a court to recognize that the longer an injunction or consent decree stays in place, the greater the risk that it will improperly interfere with a State's democratic processes." (76) So Horne's standard applies equally to a court order negotiated by a plaintiff class and a state defendant as it does to an order that a court imposes. (The Ninth Circuit did finally terminate the decree in Horne, but that action took an additional six years.) (77)

Perhaps the Supreme Court will someday hear a case that allows it to make that implicit conclusion explicit; consent decrees are almost never litigated all the way to the Supreme Court. But the United States Court of Appeals for the Sixth Circuit in John B. v. Emkes has already done so. (78) There, Tennessee sought to vacate a consent decree that resulted from a class-action lawsuit involving Tennessee's failure to provide certain services to children in violation of the Medicaid Act. In affirming the district court's decision to vacate the decree, the Sixth Circuit reiterated the two applicable questions when resolving a Rule 60(b)(5) motion to vacate: "first, whether the state has achieved compliance with the federal-law provisions whose violation the decree sought to remedy; and second, whether the State would continue that compliance in the absence of continued judicial supervision," (79) demonstrated by the implementation of structures and policies. (Keep in mind that states will often agree to a consent decree that goes well beyond the scope of the alleged violation of federal law simply to avoid the time, expense, and embarrassment of litigation.) The point that Emkes makes clear is that the Horne standard applies when a state seeks to vacate a consent decree: once a state is in compliance with federal law, court supervision must end. (80)

More important, as explained in the next section, there are many reasons why defendants choose not to seek to terminate a court's injunctive order, whether imposed by the court or entered with the consent of the defendants. And even when such motions are filed, courts are inclined to send the parties back to the negotiating table. A congressionally mandated review of such orders every four years would put appropriate pressure on all parties to fix what can be fixed and to exclude the superfluous in judicial decrees.

V. What Congress Could Do

Under Horne and the Supreme Court's other federalism-based rulings on institutional-reform-litigation decrees, one would expect that state and local governments would be able to quickly exit a consent decree once it becomes clear that there are no ongoing violations of federal law or constitutional rights. But a major impediment to exit can be courts themselves. In the authors' experience with consent decrees around the country--and as observed by Messrs. Sandler and Schoenbrod in Democracy by Decree--courts can become attached to the institutional reform they are overseeing and come to see themselves as the chief protector for victims of a system purportedly run by incompetent and uncaring government officials. This attitude creates two problems.

First, because judges can hold public officials in contempt for violating a court order, officials tend to be wary of seeking relief from a consent decree in the absence of full compliance, no matter how counterproductive or unnecessary the commands of the decree. Contempt can mean jail time, and even when contempt means only public censure and a fine, it can ruin a public official's career. Small wonder, then, that so few state and local governments have tried to take advantage of the Supreme Court's holding in Horne by moving to vacate a consent decree when federal violations have been remedied but additional stipulated-requirements remain. As the Supreme Court has observed, "[g]overnment officials, who always operate under fiscal and political constraints, 'frequently win by losing' " in institutional reform litigation. (81)

Second, when a state or local government threatens to bring a Horne motion, federal judges tend to shut them down. For example, a judge can refuse to consider a motion until the parties have attempted to sit down and craft a new "agreement" through renegotiation--a difficult task that takes time. Or a judge may view the court as the only thing standing between a state agency and a state legislature eager to save money by cutting the agency's budget. (This may be a real risk, but it is not one that can be remedied by a consent decree.)

An additional difficulty is that Horne itself was not a clear roadmap for federal district courts to rein in consent-decree abuse. The Court was divided 5-4; the Justices were divided over multiple issues; and the Justices' opinions were long and complicated, totaling nearly 26,000 words. Such complexity is a natural by-product of the legal process; it need not be so with a legislative prescription.

There is, however, a solution to this seemingly intractable problem: Congress could force the federal courts to periodically conduct a review of the consent decrees they oversee to ensure that there is a continuing foundation for court supervision, i.e., an ongoing violation of federal law or constitutional rights. In 2012, Congress considered such a measure, the Federal Consent Decree Fairness Act sponsored by Senator Lamar Alexander. (82)

As introduced by Senator Alexander, the proposed act amended the federal judicial code and used a three-pronged approach to address the problems caused by long-lasting consent decrees. First, the act authorized any state- or local-government official to file a motion to modify or terminate a federal consent decree upon the earlier of (a) four years after the consent decree was originally entered, or (b) the expiration of the term of office of the highest state-or local-government official who was a party to the consent decree. Second, the proposed act shifted the burden of proof to the party that originally filed the action to demonstrate that the court's denial of the motion to modify or terminate is necessary to prevent the violation of a federal requirement that (a) was actionable by such party, and (b) was addressed in the original consent decree. Third, the proposed act required the court, within 30 days after the filing of the motion, to enter a scheduling order that (a) limited the time of the parties to file motions and complete discovery, and (b) set the date or dates of any hearings necessary to resolve the motion. The proposal also authorized a court to stay the injunctive or prospective relief set forth in the consent decree if the party opposing the motion to modify or terminate sought any continuance or delay that would prevent the court from entering a final ruling on the motion within 180 days of its filing.

There is much to commend in the proposed Federal Consent Decree Fairness Act, which received broad support but ultimately was never voted out of committee. Consider each of the proposed act's three principal mechanisms for change and how they could be enhanced.

Timing of Federal Court Review

Outside the consent-decree context, elections ordinarily provide citizens an opportunity to assess official performance and to implement change. But because consent decrees almost always last much longer than the terms of those public officials who consented, and have the anti-democratic effect of binding successor officials, consent decrees are exempted from the ordinary democratic process. The Federal Consent Decree Act acknowledged this reality by tying judicial review to the election process.

The proposed act's timing provision was a laudable start. But it does not take into account that even successor officials--for political or other reasons--may be reluctant to file a motion that seeks to vacate or modify a consent decree. Consent decrees go to the core of state sovereignty and essentially involve plaintiffs' attorneys controlling many aspects of a state executive-branch agency's practice and policy, backed by the power of the federal judiciary. In such extraordinary circumstances, Congress should require federal courts to review standing consent-decree orders and determine on a regular basis whether there continues to be a violation of federal law or constitutional rights. Such review not only guarantees state sovereignty, it allows voters to know whether their governmental officials are continuing to violate citizens' rights. In the absence of any violation, the federal court must terminate the consent decree and close the case. This mandatory review should take place every four years following the date the consent decree was first entered.

Burden of Proof

There is only one reason the federal courts should be micromanaging state executive-branch agencies: to prevent the violation of federal constitutional or statutory rights. Under existing law, the burden is on the state or local government to prove that a decree is no longer necessary. And as a practical matter, federal courts usually place a much greater burden on the state or local government to prove that all (or nearly all) of the consent decree's provisions have been satisfied. Under such burdens, perpetual consent decrees are essentially guaranteed, as history and practice have demonstrated. Indeed, as a result of government officials caving to plaintiff demands, in many institutional-reform cases, a federal court has never made the requisite finding that a governmental agency is violating a plaintiff class's federal rights.

The proposed Federal Consent Decree Act addressed this problem by placing the burden on the plaintiffs to prove that the consent decree is still necessary to protect the plaintiffs' rights. Of course, once the plaintiff class made an initial showing that rights were being, or were likely to be, violated, the burden would shift to the state or local government to disprove that contention. So federal law will continue to vindicate the rights of plaintiff classes, but state and local governments will no longer be burdened by federal-court consent decrees absent proof of a federal statutory or constitutional violation.

Standard for Entering/Maintaining a Consent Decree

Again, because plaintiffs' attorneys and government officials frequently stipulate or consent to a decree, federal courts are rarely put in the position of having to decide whether governmental conduct actually violates, or threatens to violate, federal statutory or constitutional rights. The proposed Federal Consent Decree Act took a step in the right direction by requiring a court to make specific factual findings that federal law was, in fact, being violated as a prerequisite to entering or maintaining a consent decree. But this standard should be enhanced in three respects, the first with respect to alleged statutory violations, the second with respect to alleged constitutional violations, and the third with respect to remedies.

Regarding alleged statutory violations, Congress should make clear that a consent decree may not be entered or maintained in the absence of a governmental violation of a substantive right under a federal statute, a right that is subject to private enforcement. For example, the federal government's Administration for Children and Families, since 2001, has promulgated Child and Family Services Reviews, known as CFSRs, to measure state performance on various factors related to child safety, permanency, and well-being. The federal government uses the CFSRs to assess key state-plan requirements of Titles IV-B and IV-E that provide a foundation for child outcomes. If a state has not achieved substantial compliance with one or more of the assessed areas, the state must develop and implement a program-improvement plan to address those areas. The penalty if a state is unable to successfully complete its program-improvement plan is the federal government's withholding of a portion of the state's Title IV-B and IV-E funds. In other words, the CFSRs are a funding carrot to encourage states to achieve a certain level of performance. But, because states are free to forgo this funding, CFSRs are not a federally mandated substantive requirement and should never appear in a consent decree. Similarly, alleged state-law violations or standards created by private entities (such as the Child Welfare League of America's caseload standards) are not federally mandated substantive requirements and do not justify a consent decree.

As for alleged constitutional violations, Congress should require federal courts to specify in the decree what the legal standard is. In the ordinary case, it is not possible to say that a state's child-welfare system is violating a child's specific constitutional right. Instead, plaintiffs claim a generic violation of the class's substantive due-process rights. In other contexts, federal courts attempting to discern whether a governmental entity has violated substantive due process in the abstract has applied a "shock the conscience" analysis. (83) Applying this standard in the context of a challenge to the Massachusetts child-welfare system, the federal district court in Connor B. ex rel. Vigurs v. Patrick (84) used the "shock the conscience" standard and required plaintiffs to prove that (1) the state's "'presumptively valid' decisions" constituted "'such a substantial departure from accepted professional judgment, practice, or standards as to demonstrate that the person responsible actually did not base the decision on such a judgment,'" (85) and (2) "'the behavior of the governmental officer [wa]s so egregious, so outrageous, that it may fairly be said to shock the contemporary conscience.' " (86)

The Massachusetts district court's test in Connor B. is a faithful application of the Supreme Court's precedents in Youngberg v. Romeo (87) and County of Sacramento v. Lewis. (88) In Youngberg, the Supreme Court held that when considering the substantive due-process rights of those who have been involuntarily committed to a state's care, the legal standard must "reflect[ ] the proper balance between the legitimate interests of the State and the rights of the involuntarily committed to reasonable conditions of safety and freedom from unreasonable restraints." (89) Because governmental professionals' decisions are "presumptively valid," a plaintiff claiming a substantive due-process violation must show that the professional's decision "is such a substantial departure from accepted professional judgment, practice, or standards as to demonstrate that the person responsible actually did not base the decision on such a judgment." (90) In other words, the courts must ensure only that professional judgment was exercised, not which of "several professionally acceptable choices should have been made." (91) This Youngberg rationale "applies with equal force" in the child-welfare context. (92)

In Lewis, the Supreme Court reexamined its substantive due-process jurisprudence and held that "only the most egregious official conduct can be said to be 'arbitrary in the constitutional sense.'" (93) In addition, only conscience-shocking behavior can be sufficiently arbitrary and egregious to be of constitutional significance. (94)

But, rather than try to codify these standards, Congress should require federal district courts to articulate a standard. The federal courts of appeal can then determine whether that standard is appropriate. Requiring courts to set forth the precise standard in a decree would clear up substantial confusion. (95)

Finally, Congress should require that any court-imposed relief be narrowly tailored to remedy the specific violation of federal statutory or constitutional law that the district court identifies. Such a requirement would limit consent decrees, allowing them to go no further than necessary to ensure that the class plaintiffs will not suffer an illegal injury. And if a court is entering a decree based on evidence of past injury, plaintiffs must prove that the decree is essential to prevent future injury.


Federal-court injunctive decrees remain a valuable tool for ensuring that state and local governments provide a level of service to their citizens that federal law demands. But as the past several decades have demonstrated, there is a potential for abuse that deprives government officials of their flexibility to set priorities and implement program changes, denies voters the opportunity to influence how their government acts, and sometimes harms that class of citizens that the decree was intended to benefit. While it would be inappropriate to eliminate federal-court authority to enter decrees against state and local governments, it is well past time for Congress to enact legislation that recalibrates the system and ensures a proper balance between state- and local-government sovereignty on the one hand, and federal rights on the other. That legislation should:

* Require federal courts with jurisdiction over an institutional judicial decree to review the decree every four years, at minimum, to ensure the decree continues to be justified;

* Make clear that in any proceeding involving the entry or maintenance of a federal-court decree, the burden is always on the plaintiffs to prove that the decree is required to remedy or prevent a violation of federal constitutional or statutory rights; and

* Demand that federal courts make specific findings before entering or maintaining a consent decree that a state or local government is violating or will imminently violate (a) a substantive federal statutory standard that gives private plaintiffs a private right of enforcement, or (b) a plaintiff's federal constitutional rights. With respect to claims that a government is violating substantive due-process rights, the legislation should require that decrees define a clear standard. Current federal-court jurisprudence suggests the following standard: when a government agency's conduct is such a substantial departure from accepted professional judgment, practice, or standards as to demonstrate that the person responsible actually did not base the decision on such a judgment, and the behavior of the responsible person is so egregious and outrageous that it may fairly be said to shock the contemporary conscience.

* Require the federal courts to find that the decree to be entered or maintained is narrowly tailored to prevent future violations of federal statutory or constitutional rights.

Congress has a special obligation to fix federalcourt injunction practice involving state and local governments because these decrees largely are purportedly intended to remedy federal statutes that Congress itself put in place. While the vast majority of these statutes were supposed to vest discretion in state and local officials to do their job (e.g., the CFSR standards), judicial decrees remove that discretion. And, by confining state and local governments to the standards of practice that plaintiffs' attorneys prefer, judicial decrees deny the type of flexibility and experimentation that state sovereignty is supposed to guarantee. It is time for Congress to restore the proper balance.


Summary of Judicial Decrees Currently Affecting Child-Welfare Systems

Connecticut: Juan F. v. Malloy (25 years)

Litigation was filed in December 1989 and settlement reached in January 1991. The current exit plan (approved in July 2006) contains 22 outcome measures that all must be met and sustained for six months before exit.

District of Columbia: LaShawn A. v. Gray (23 years)

Plaintiffs filed a complaint in 1989 which resulted in a court-ordered decree in 1993. The current implementation plan has been in place since December 2010, and includes approximately 92 separately measured exit standards divided into outcomes to be achieved and outcomes to be maintained.

Georgia: Kenny A. v. Deal (11 years)

Litigation was filed in 2002 and a consent decree entered in 2005. The modified agreement currently in place contains 29 outcome measures that must be achieved simultaneously for three consecutive periods before exit.

Illinois: Multiple (39 years)

Illinois currently operates under more than 10 consent decrees/settlement agreements related to the child-welfare system. There are several coordinators and monitors embedded within the Department of Children and Family Services to oversee and monitor compliance.

Maryland: L.J. v. Massinga (28 years)

Litigation was filed 1984 on behalf of Baltimore children placed in Maryland's foster-care program. The parties entered into an initial consent decree in 1988, with the intent that it would be complete in two years. Twenty-one years later, in 2009, the parties entered into a modified consent decree. Exit from court supervision is not available until Maryland has complied with all commitments for 18 consecutive months.

Michigan: Dwayne B. v. Snyder (8 years)

Plaintiffs filed their lawsuit in 2006, and the parties entered into a consent decree in 2008 that was modified in 2011, and again at the end of 2015. The current agreement includes 11 outcome measures to be maintained and 56 measures to be achieved, with various measures rolling to exit when achieved for specified timeframes.

Mississippi: Olivia Y. v. Barbour (8 years)

Litigation was filed in 2004 and a consent decree entered in 2008. Modified settlement agreements were finalized in 2012, and in 2016 in response to the plaintiffs' motions for contempt and the appointment of a receiver for the entire foster-care system.

New Jersey: Charlie and Nadine H. v. Christie (12 years)

Litigation was brought in 1999, a first consent decree was entered in 2004, a modified settlement agreement was entered in 2006, and a sustainability and exit plan was entered in late 2015. The 2015 plan contains 48 outcome measures to be maintained or to be achieved before exit.

New York: Marisol A. v. Giuliani; Elisa W. v. New York City (18 years)

The Marisol A. class-action lawsuit was filed against the City and State of New York in 1995, and the parties approved a consent decree in 1998. Although the parties terminated monitoring, the court retained authority to issue injunctions and award damages. While the city has increased its foster-care system's budget by $100 million over the past two years, the Elisa W. class-action lawsuit was filed in early 2015 against the City and State of New York, and the state recently settled and agreed to the entry of another consent decree requiring the appointment of another monitor.

Ohio: Roe v. Staples (30 years)

Plaintiffs filed their lawsuit in 1983, a consent decree was entered in 1986, and a modified decree was entered in 2006. Substantive provisions include requirements involving needs assessments, case plans, placement, visitation, and service. Ohio finally resolved the monitoring component of the decree in 2015, 30 years after execution of the initial decree.

Oklahoma: D.G. v. Yarbrough (4 years)

The lawsuit was filed in 2008, and the court approved the parties' consent decree in 2012. The consent-decree monitors are, twice per year, to assess the state's "good faith efforts" to comply with the goals of the decree.

South Carolina: Michelle H. v. Haley (new)

Litigation was initiated in January 2015 and a settlement reached in early 2016. The consent decree requires the state to satisfy dozens of provisions relating to caseloads, investigations, placements, visitation, and health care.

Tennessee: Brian A. v. Haslam (15 years)

Litigation was initiated in 2000 and a settlement reached in 2001. Under the 2003 modified decree, the state must comply with approximately 136 different outcome measures.

Wisconsin: Jeanine B. v. Walker (14 years)

Plaintiffs' attorneys sued the Governor of Wisconsin on behalf of all Milwaukee children in child-welfare custody in June 1993, with a final settlement reached in 2002. The city remains under court supervision because the program has yet to satisfy the final unfilled settlement requirement regarding number of placements.


(1.) Ross Sandler & David Schoenbrod, Democracy By Decree (2003).

(2.) Id. at 3-4.

(3.) Id. at 4.

(4.) Id.

(5.) Id

(6.) Abram Chayes, The Role of the Judge in Public Law Litigation, 89 Harv. L. Rev. 1281, 1309 (1976); Richard A. Posner, The Federal Courts: Challenge & Reform 340-42 (Harv. U. Press, 1996); Susan P. Sturm, The Legacy & Future of Corrections Litigation, 142 U. of Pa. L. Rev. 639, 659 (1993); Donald L. Horowitz, The Courts & Social Policy 106-298 (Brookings Inst., 1977); Peter H. Schuck, Suing Government: Citizen Remedies of Official Wrongs 147-81 (Yale U. Press, 1983).

(7.) See generally Democracy By Decree and supra n.7.

(8.) See Democracy By Decree at 8. These examples are not some hypothetical "parade of horribles," but represent actual case studies discussed at length in Democracy By Decree.

(9.) Id. at 9.

(10.) Id. at 98

(11.) Deshaney v. Winnebago Cnty. Dep't of Soc. Servs., 489 U.S. 189, 196 (1989).

(12.) Id. (emphasis added). Accord, e.g., Lindsey v. Normet, 405 U.S. 56, 74 (1972) (rejecting claim for a positive right to a homeless shelter); San Antonio Indep. Sch. Dist. v. Rodriguez, 411 U.S. 1, 35 (1973) (rejecting claim for a positive right to a public education). But see Obergefell v. Hodges, 135 S. Ct. 2584 (2015) (requiring state governments to recognize as married and provide benefits to same-sex couples).

(13.) Charles A. Reich, The New Property, 73 Yale L.J. 733 (1964).

(14.) Democracy By Decree at 104-05.

(15.) Id.

(16.) Id.

(17.) Id. at 105.

(18.) NYC Dept. of Educ. System Wide Report #2 for Fiscal Year 2015: Function By Student Type--Public Schools Only, available at

(19.) Abram Chayes, The Role of the Judge in Public Law Litigation, 89 Harvard L. Rev. 1281 (1976).

(20.) Democracy By Decree at 114.

(21.) Id. at 115.

(22.) Id. at 118.

(23.) Id. at 119.

(24.) Id. at 123.

(25.) Id. at 127.

(26.) Id. at 122.

(27.) Juan F. v. Malloy, No. H-89-859 (D.Conn. 1991).

(28.) Complaint, Juan F. v. Malloy, No. H-89-859, (last visited June 1, 2016).

(29.) Exit Plan Quarterly Report January 1, 2015-March 31, 2015, Juan F. v. Malloy, No. H-89-859, (last visited June 1, 2016).

(30.) Id

(31.) LaShawn A. v. Fenty, No. 89-cv-1754 (D.C. Cir. 2010).

(32.) Implementation and Exit Plan, LaShawn A. v. Fenty, No. 89-cv-1754 (last visited June 1, 2016).

(33.) A Progress Report, July 1-December 31, 2014, LaShawn A. v. Fenty, No. 89-cv-1754, (last visited June 1, 2016).

(34.) Jeanine B. v. Walker, No. 93-C-0547 (E.D. Wisc. 2012).

(35.) Children's Rights, "Update Regarding Jeanine B v. Walker," (last visited June 6, 2016).

(36.) L.J. v. Massinga, No. 1:84-cv-04409, (D. Md).

(37.) L.J. v. Massinga, 699 F. Supp. 508 (1988), (last visited June 1, 2016)

(38.) Modified Consent Decree L.J. v. Massinga, No. 1:84-cv-04409, (last visited June 1, 2016).

(39.) Id.

(40.) Id.

(41.) Marisol A. v. Giuliani, No. 1:95-cv-10533 (S.D.N.Y. 1999).

(42.) Marisol A v. Giuliani, 185 F.R.D. 152 (1999), (last visited June 1, 2016).

(43.) Vivian Yee, Suit to Accuse New York City and State of Keeping Children in Foster Care Too Long, N.Y. Times (July 7, 2015),

(44.) Id.

(45.) Dwayne B. v. Snyder, No. 2:06-CV-13548-NGE-DAS (E.D. Mich.).

(46.) A chart summarizing judicial decrees currently governing child-welfare systems appears in Appendix A.

(47.) Philip Howard, The Rule of Nobody (2014).

(48.) U.S. 267, 280-88 (1977).

(49.) U.S. 343, 349 (1996).

(50.) U.S. Const., amend. X.

(51.) Rizzo v. Goode, 423 U.S. 362, 378 (1976); Rufo v. Inmates of Suffolk Cnty. Jail, 502 U.S. 367, 392 (1992); Horne v. Flores, 557 U.S. 433, 448 (2009).

(52.) Rizzo, 423 U.S. at 379 (emphasis added, quotation omitted).

(53.) Id. at 378 (quoting O'Shea v. Littleton, 414 U.S. 488, 500 (1974)).

(54.) E.g., Mass. State Grange v. Benton, 272 U.S. 525, 528 (1926); Pennsylvania v. Williams, 294 U.S. 176, 185 (1935); Rizzo, 423 U.S. at 379-80.

(55.) U.S. at 178-79.

(56.) Id. at 185.

(57.) Alan Effron, Federalism & Consent Decrees Against State Governmental Entities, 88 Colum. L. Rev. 1796, 1800 n.27 (1988); Pennhurst State Sch. & Hosp. v. Halderman, 465 U.S. 89, 104 n.13 (1984); City of Los Angeles v. Lyons, 461 U.S. 95, 112 (1983).

(58.) U.S. at 392 n.14.

(59.) U.S. 431, 441 (2004).

(60.) Compare with West River Bridge Co. v. Dix, 47 U.S. 507, 531-32 (1848) (holding that state contracts could not override the state's successive leaders from using eminent domain power).

(61.) United States Trust Co. v. New Jersey, 431 U.S. 1, 45 (1977) (Brennan, J., dissenting).

(62.) Frew ex rel. Frew v. Hawkins, 540 U.S. 431, 441-42 (2004).

(63.) Horne, 557 U.S. at 450.

(64.) Id. (quoting Milliken v. Bradley, 433 U.S. 267, 282 (1977)) (emphasis added).

(65.) Horne, 557 U.S. at 439-45.

(66.) Id. at 450.

(67.) Id. at 452 (quoting Frew, 540 U.S. at 442) (emphasis added).

(68.) Id.

(69.) Id.

(70.) Id. (emphasis added).

(71.) Id. at 465.

(72.) Id. at 466.

(73.) Id. at 468.

(74.) Id. at 470.

(75.) U.S. 447-53.

(76.) Id. at 453 (emphasis added).

(77.) 789 F.3d 994 (9th Cir. 2015).

(78.) F.3d 394 (6th Cir. 2013).

(79.) Id. at 412 (citations omitted, emphasis added).

(80.) F.3d at 413 (emphasis added) (quoting Horne, 557 U.S. at 450).

(81.) Horne, 557 U.S. at 449.

(82.) See H.R. 3041, 112th Cong. (2012).

(83.) Bell v. Ohio State Univ., 351 F.3d 240, 250 (6th Cir. 2003).

(84.) F. Supp. 2d 142 (D. Mass. 2011).

(85.) Id. at 162 (quoting Youngberg v. Romeo, 457 U.S. 307, 323 (1982)).

(86.) Id. (quoting County of Sacramento v. Lewis, 523 U.S. 833, 847 n.8 (1998)).

(87.) U.S. 307 (1982).

(88.) U.S. 833 (1998).

(89.) U.S. at 321.

(90.) Id. at 323.

(91.) Id. at 321 (emphasis added, quotation omitted).

(92.) Connor B., 771 F. Supp. 2d at 160 (numerous citations omitted); accord, e.g., Yvonne L. v. N.M. Dep't of Human Servs., 959 F.2d 883, 893-94 (10th Cir. 1992) (applying the professional judgment standard to foster children).

(93.) U.S. at 846.

(94.) Id. at 846-47 & n.8.

(95.) See, e.g., Connor B. v. Patrick, 774 F.3d 45, 52-53 (1st Cir. 2014) (affirming the Massachusetts district court but declining to adopt the district court's constitutional standard in toto because plaintiffs' evidence did not establish that even the Youngberg standard was met).

JUNE 2016
Table of Contents
EXECUTIVE SUMMARY                                               v
CONSENT DECREE                                                  1
I. WHAT IS A CONSENT DECREE?                                    3
FEDERALISM                                                      8
V. WHAT CONGRESS COULD DO                                      15
    Timing of Federal Court Review                             16
    Burden of Proof                                            16
    Standard for Entering/Maintaining a Consent Decree         16
SUMMARY                                                        19
APPENDIX                                                       20
NOTES                                                          22
ABOUT THE AUTHORS                                              25

Table 1. Key Child-Welfare Consent-Decree Provisions

Item                    Michigan                Connecticut

Parent-child visits     100% must have          No requirement
                        one face-to-face visit
                        per month
Child Protective        100% completed          85% completed
Services: completion    within 30 days          within 45 days
of investigation
Use of psychotropic     Detailed system         No requirement
medications             of approvals and

Search for relatives    No requirement          Six-month search
when a child is                                 period that must
removed from                                    be conducted and
home                                            documented in at
                                                least 85% of cases
Investigator case-load  95% shall have no       No more than a
ratios                  more than a 12:1        17:1 ratio
Caseworker case-load    95% of caseworkers      In-house treatment
ratios                  shall have no more      workers shall have
                        than a 15:1 ratio       no more than a 15:1
                                                ratio; out-of-home
                                                workers a 20:1
                                                ratio, and adoption
                                                workers a 20:1 ratio

Item                    Washington, DC        Tennessee

Parent-child visits     85% must have         50% must have visits
                        weekly visitation     twice per month

Child Protective        90% completed         No requirement
Services: completion    based on policy
of investigation        time frames
Use of psychotropic     No requirement        Psychotropic medications
medications                                   cannot be
                                              used as a method of
Search for relatives    Reasonable effort     Diligent search
when a child is         to locate and invite  within 30 days and
removed from            known relatives to a  at 3 and 6 months
home                    family team meeting   post-custody
                        in 90% of cases
Investigator case-load  90% shall have no     No requirement
ratios                  more than a 12:1
Caseworker case-load    90% of caseworkers    15:1 or 20:1 ratio
ratios                  shall have no more    depending on the
                        than a 15:1 ratio     type of caseworker

Source: Authors' review of consent decrees imposed on state and local
COPYRIGHT 2016 The American Enterprise Institute
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2016 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Bursch, John; Corrigan, Maura
Publication:AEI Paper & Studies
Article Type:Report
Geographic Code:1USA
Date:Jun 1, 2016
Previous Article:AEI public opinion studies: polls on patriotism.
Next Article:Designing performance-based incentives when service providers compete for users to help.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters