November 18, 2002
Class Actions, Competition, and Reflections on Federalism
There are two auto mechanics, graduates of Yale, who are contemplating an engine repair, and one says to the other “we know it will work in practice, but will it work in theory?” Today I am going to be speaking about something that does not work well either in practice or in theory. The practice that I will be speaking about is a particular kind of class action practice that involves large segments of the population as potential plaintiffs, such as in the area of product liability. My focus will be on overlapping, competing and duplicative class actions involving the same parties and the same claims that are now so often filed at the same time in federal and state courts, indeed, sometimes in multiple state courts and multiple federal courts. The theory that I will be speaking about is federalism because, oddly, it is federalism that is invoked by lawyers and policy makers to defend the role of state courts in regulating national commerce and activity.
I will contend that the competition between federal and state courts in these class action cases is not healthy competition leading to efficiency but quite the reverse. And as I shall try to explain, under any conceivable theory of federalism, it is the national government that should regulate national commerce. I will argue that federalism is very much a two-way street, not just a rhetoric or theory that preserves state power from encroachment by Washington, and that it violates “our” federalism when state courts reach out to apply local law and procedures on a nation-wide basis. In our federal system, national policy and nationally applicable regulation of interstate conduct and business should occur at the national level, not in select state courts as the by-product of class action litigation. Put another way, states ought not to intrude on one another by attempting to regulate activities that have deeper connections to other states and by disposing of the rights of people or entities who have little or no connection to the forum state.
Perhaps this seems obvious to you. I should hope so given the work that the Manhattan Institute has done in this area. As an admirer said to Adlai Stevenson following one of his speeches: “Your remarks were simply superfluous.” But it may surprise you that there is no consensus on these issues within the judiciary or the academy and that some of the trends that I identify as problems are the result of decisions by the United States Supreme Court.
Before I go much further certain disclaimers are in order. First, I am speaking here today on my own behalf. For the past several years, I have been the Chair of the Advisory Committee on the Federal Rules of Civil Procedure. But I do not speak for the Committee today. Nor do I speak as a scholar or as a political scientist but simply as one who has been involved in the rulemaking process and who has heard much over the past 8 years from scholars and lawyers about the workings of the class action rule, Rule 23. Second, although I will be talking about forum shopping and the filing of competing class actions, I do not in any way challenge the good faith or motivation of the plaintiffs’ class action lawyers who bring the actions or, for that matter, the defense lawyers who seek to remove these cases to federal court. This is an area of law that is very sensitive to incentives and to litigation tools. If we provide those incentives and tools, we should not complain of those who respond to them or take them up. Third, although I will be speaking of certain state courts and making the point that some few of the state courts are selected because they are perceived as friendlier to plaintiffs and less rigorous on such critical matters as the qualification of expert witnesses, I mean no disrespect to the judges of those courts. It can be assumed that they are acting in ways that are intended by the citizens of the state who have selected them. It is a policy choice to have elected judges, or a low threshold for the admissibility of expert witnesses and evidence, or lenient rules on class certification. The question I would like to consider is whether these policy choices should apply to conduct and injury occurring outside of the state’s boundaries and this is a question that does not depend for answer on one’s appraisal of the relative competency of state and federal judges. In fact, we have remarkably capable judges throughout the state and federal judiciaries at all levels.
Indeed, in thinking about these matters I suggest that it would be healthy entirely to depersonalize the rhetoric of federalism. The Supreme Court sometimes uses this kind of rhetoric in referring to the States and will speak, for example, of “affronts” to the states’ “dignity.” But this is just a turn of phrase, a kind of short hand particularly where sovereign immunity is at issue. This kind of expression draws on a tradition that has its roots in a time when the sovereign was in fact a person. The Court does not truly imagine that the State is actually offended or embarrassed or sad or happy or that the sovereign in our system of government is anything other than “we the people.” However, when the proponents of state regulatory authority over national activity use such rhetoric, it is not just a turn of phrase but an effort to engage in interest group politics within the judiciary. Rather than address whether a certain kind of matter is most appropriately handled in state or federal court, in light of objective principles of federalism and objective features of the different judicial systems, they will argue that the dignity and competency of state court judges will be slighted if federal jurisdiction is provided. But federalism is not about making the particular state or federal actors happy or sad or ensuring their self esteem. The founders had something much more profound in mind in fragmenting authority among the different units of government. Satisfying the career goals of state and federal judges should be low on our list of priorities in effectuating the federal plan of government that the founders bequeathed to us.
I have one final disclaimer. Although I will be speaking about state court and federal court, much of what I address is equally applicable horizontally among the state courts. If it infringes on the national interest when a state court applies state law and procedures to nationwide class actions, so too does it infringe on the interests of other states when a state court applies its law and procedures to multi-state class actions, covering conduct and injury that occurred elsewhere involving residents of other states. Perhaps there is a version of Gresham’s Law at work here where bad procedure may drive out good and overreaching choice of law decisions in one state may drive out more cosmopolitan choices in another.
I will first describe the problems in practice. I will then address the problems in theory.
Our story begins in 1966. In that year the advisory committee on civil rules re wrote the existing class action rule by, among other things, adding a new provision known as 23(b)(3) which permitted damage class actions in federal court. The new damage class action would be an opt out class, meaning that class members were in unless they responded to whatever notice they received by affirmatively opting out. This change made all the difference. Large consumer damage class actions now became possible, aggregating claims that were too small to justify the expense of litigation on their own. Other possibilities for aggregated litigation opened up as well. This was a heady time in the social history of the country, and the rulemakers imbibed deeply of the same spirit. They were eminent lawyers and judges -- the Chair of the Committee was Dean Acheson -- and they were prepared to think big and to exert their influence. As to the (b)(3) opt out damage class, one member of the Committee later characterized it as “the most radical bit of rulemaking since the original rules created one cause of action and abolished key distinctions between law and equity.”
In the Committee debate over whether to include a mandatory damage class action provision, Professor Moore, the eminent proceduralist from Yale law school, expressed the concern that any compulsory damage class would infringe on an individual litigant’s right to control the litigation and that defendants would use the class action device to rig a class to take a dive. We now call this the problem of the “reverse auction.” According to the transcript of the meeting, Judge Wyzanski had an epiphany. He responded to Professor Moore, ‘Would you be satisfied Professor Moore, if the class could never include anybody who specifically protests within a given period?’ Professor Moore responded, “Well, I think that would be helpful.” The late John Frank chimed in:. “May I say that if that were done my problem would fully evaporate.”
Unfortunately the problem did not quite evaporate but the Rule took its shape and began its amazing career. Eventually mass tort cases and other kinds of damage class actions would be brought under the Rule in addition to small claim consumer class actions. As Professor Arthur Miller explains, whatever the members of the Committee may have expected, the Rule, perhaps by design perhaps by serendipity, caught the wave of “the most incredible upheaval in federal substantive law in the history of the nation between 1963 and 1983.” The Rule was also copied in the state courts, most of whom have a similar procedural device. Thus, a mere rule of joinder, instituted by a committee of the judicial conference without fanfare or direct congressional participation thereby became the means by which private attorneys may, depending on one’s point of view, tame or break even the most well established business interests.
Since the 1960s the Rule has been widely used. For a time in the 1980s, it appeared that class action litigation might be waning. But in the last decade there has been a surge of class action filings. We can document this in federal court. According to a recent study by the Federal Judicial Center, the number of class actions filed in federal court has doubled in the period from 1994 to 2001. While there are no statistics kept for the state courts as a whole, few would doubt that the number of filings in state court has easily outpaced even the surge in federal filings.
Of course many class actions are routine, present few problems, and achieve modest recoveries in the $300 to $500 range for class members who have sustained a modest injury. This seems an understandable, if somewhat expensive, supplement to whatever public regulatory resources are committed to such matters. The difficulties arise from the overuse of the device and, in particular, from the filing of multiple, overlapping, duplicative class actions in multiple courts, state and federal.
By their very scope these multistate or national class actions can be filed almost anywhere. But multiple filings in numerous venues is just the outcome that the class action device was designed to prevent. The purpose of the class action is to eliminate repetitive individual litigation, promote judicial efficiency, permit small claims to find a forum, and achieve uniform results in similar cases. But as Professor Edward Cooper explains “duplicative class litigation is destructive of just these goals . . . . Multiple filings can threaten appropriate judicial supervision, damage the interests of class members, hurt conscientious class counsel, impose undue burdens of multiple litigation on defendants, and needlessly increase judicial workloads.”
The American Law Institute’s 1994 Complex Litigation Project highlighted the problems caused by multiple class action filings as did RAND’s 2000 study entitled Class Action Dilemmas. In its close study of 10 cases, the RAND found that in four cases, class counsel filed parallel actions in other courts. In five of the ten, other groups of plaintiffs’ attorneys filed competing actions in other jurisdictions.
The Manhattan Institute has been helpful to our understanding by commissioning a study on interstate class actions. In a study of three county courts, their research finds a disproportionate number of class action filings, given the size of the population and general docket, and finds that the number of nationwide class actions substantially increased in each court during the period from 1998 to 2000. Madison County, Illinois, with a population of 250,000, had more class actions filed than any other state court except for Cook County (Chicago) and Los Angeles County. Most significantly, the research found that a “large percentage of the cases involved plaintiffs and/or defendants that were not residents of the counties where the class actions were filed.” The study also found many instances of copy cat class actions in which a class action complaint filed in one court is simply refiled elsewhere with minor cutting and pasting such that multiple suits will pend against the same defendants on behalf of the same plaintiffs in multiple jurisdictions. After looking at some of the kinds of cases brought as nationwide class actions in Madison County, the study concludes that “Madison County judges have been asked over the last few years to set national policy on issues that could affect the daily lives of millions of Americans throughout the country – from what water they drink to how much they pay for their next insurance policy or telephone bill – all from a small courthouse in southwest Illinois.”
Some of the problems that arise from multiple, overlapping filings cause unfairness within the context of the particular litigation. There is the problem of the reverse auction in which class counsel in one court compete against other class counsel in other courts for the defendant’s favor of a settlement. In this race to the bottom, class counsel have an incentive to accept terms less favorable to the class in return for reaping the rewards that flow to class counsel able to bring in a settlement.
Multiple actions also may provide an escape from more exacting judicial supervision. There is the potential of such filings to frustrate judicial scrutiny of certification motions, settlements, and fee requests. Thus, a settlement rejected in one court may be shopped around until another court is found that will approve it. The best known example of this is the GM litigation in which the Third Circuit rejected a settlement of a class action filed in Philadelphia as unfair to class members only to see what many would characterize as substantially the same settlement approved in the 18th Judicial District for the Parish of Iberville, Louisiana. A similar phenomenon may occur at the certification phase of the litigation. The careful decision of one court to deny certification because, for example, the class is too diffuse or disparate or the representation inadequate can be circumvented by filing the same action in another court some of which permit ex parte certification motions. As we attempt to improve judicial supervision of the class action process in federal court, from the decision to certify, through review of the proposed settlements, to scrutiny of attorneys fees, and as leading state courts attempt to do the very same things, it is predictable that the cases will shift into those state courts where such supervision is perceived to be lacking. This is already happening as the Manhattan Institute study demonstrates. And while it is true that the practices of some particularly notorious local courts have been addressed and corrected by the state supreme courts, this has only led to filings in other more hospitable county courts.
In sum, there are manifold inefficiencies and inequities that result from the filing of duplicative, overlapping class actions in multiple jurisdictions. Where the actions are filed in multiple federal courts, the Judicial Panel on Multi-District Litigation acts to co-ordinate the litigation and to control the excesses. When the actions are pending in multiple state and federal jurisdictions, however, there is no such panel or control. And this lack of control creates the opportunity for destructive forum shopping in which unfair advantage may be taken of class members, of other class counsel, of defendants, and of the judicial system taken as a whole, a system that can ill afford to expend resources on duplicative, complex litigation.
We have a problem in practice, and the problem is much bigger than just a few adventurous jurisdictions.
In an article on Class Actions and Social Issue Torts in the Gulf South, Professor McGovern describes some of the ways in which those who bring state court class actions attempt to regulate interstate and national activity. He notes that there are now “law firms with sufficient clout to take on entire industries,” that the lawyers in these firms “have decided to stay in the litigation arena to advance their social agenda,” that they have joined forces with state attorneys general to push the agenda and that “[t]hey have learned . . . that aggregating many cases in a plaintiff-friendly jurisdiction not only precludes other plaintiffs’ counsel from pursuing the identical litigation but also creates a “piling on” strategy that almost inevitably leads to settlement.” According to Professor McGovern, forum selection is all important to this agenda because “[i]n many mass torts the choice of a court bec[omes] the single, most critical, determinative factor.”
What Professor McGovern describes would not seem generally troubling were these actions confined to particular states, such that the residents of the state would bear the burdens and receive the benefits of the litigation. If they do not like the state’s regulatory environment they can try to change it or they can vote with their feet. This is the competitive face of federalism. But the program described by Professor McGovern is co-ordinated to have national effect. Whether it is tobacco or antitrust litigation or a state court’s nationwide automobile recall order, we are seeing a new trend, powered by the class action device, in which state law and procedures will be used to achieve settlements and outcomes affecting citizens in all 50 states. Because these actions reach beyond the state’s boundaries and traditional area of control, the likelihood of conflict with other courts, state and federal, is maximized.
How this came to pass would take more than the time remaining, if it could be explained at all. The way was paved by a 1985 Supreme Court case that endorsed the power of state courts to handle nationwide litigation. In Phillips Petroleum v. Shutts, 472 U.S. 797 (1985), an oil and gas class action, the court permitted a nationwide class action in Kansas state court even though “over 99% of the gas leases and some 97% of the plaintiffs in the case had no apparent connection to the State of Kansas except for this lawsuit.” Id. at 814-15. For reasons that are not clear, Phillips based its argument entirely and somewhat disingenuously on the due process rights of the class members rather than on its own right to litigate in other forums. The Court’s answer to Phillips’ solicitude for class members was predictable: no plaintiff was forced to litigate in Kansas because each class member was given the opportunity to opt out of the class thus satisfying due process. But the court did not address, because it was not asked to, why it was acceptable to expose a defendant to suit in Kansas for activities that happened largely elsewhere. Nor does the opinion confront the decisive role of procedure and choice of forum, matters that are equally or more important to outcome than variations in substantive law or choice of law rules. Ask a plaintiff’s lawyer why he or she chose a certain forum and the answer is likely to involve matters other than the state’s substantive law, for example, the receptiveness of the particular judicial officer and the standards for the admission of expert opinion.
Curiously, Shutts is now relied upon to justify multi state mandatory, non-opt out, injunctive relief class actions brought in state court under state equivalents to 23(b)(2).
While preserving state court power over class action litigation in Shutts and later in Matsushita v. Epstein, the Court has taken a tough line on federal diversity jurisdiction in class action litigation, closely guarding the doors to federal court. As is often pointed out, it is anomalous to permit the federal courts to hear a minor tort action involving a single plaintiff and single defendant, where the claim barely exceeds $75,000, but at the same time refuse jurisdiction over a class action involving many millions of dollars and hundreds or thousands of plaintiffs from various states simply because one minor defendant is a resident of the forum state or because while the sum of class members’ claims is huge, their individual loss is less than $75,000.
I hope that I have fairly described what is now occurring in a limited number of very important class action cases affecting millions of people and large chunks of the economy. Does this practice make sense in theory? Does it make sense for state courts to handle multi-state or nationwide class actions, at least where the state’s interests are not clearly paramount over all other jurisdictions? After all, the state could decline to certify classes of nationwide scope. It could limit the class to its own residents or to activity occurring within the state thereby reducing the opportunity for conflict with other jurisdictions. No doubt many state courts do exactly that.
The nationwide state class action does not find support in any theory of federalism or public policy that we would generally subscribe to. Sometimes there is an appeal to tradition. It is said that the states have traditionally taken the lead in the field of torts. But surely that is a weak argument in this context. The nationwide class action does not have much in the way of tradition behind it. Shutts is from 1986. Before the Erie decision in 1938 the federal courts also participated in the development of tort law.
It may be said that the states should be given the opportunity to experiment and to develop alternative ways of handling these cases. This is the experimental, competitive theory of federalism we associate with Justice Brandeis. Under this theory, the states compete with each other for business and for residents and may try different approaches to governance and regulation. Residents and businesses will exit if they deem the experiment a failure. But none of this thinking supports imposition of one state’s experiment on other states or the nation as a whole. Indeed, it rests on the contrary premise. Nor does this theory of federalism support a state’s reaching out to non resident defendants and forcing them to litigate in the state against plaintiffs who were not injured in the state.
There is the access theory of federalism. Placing certain decisions within local control brings government closer to the voters, may increase participation and may guarantee respect for local choices and culture. But this theory assumes a community of participants who are deciding for themselves. It does not support the claims of that community to govern those who are outside of it. A nationwide class action brought in a state forum to take advantage of its evidence rules or its judicial selection procedures does not enhance democratic participation by those who have no voice in the state.
Finally, we might consider the Madisonian theory of federalism -- the man, not the county, mind you. Under Madison’s theory we use federalism to preserve liberty by fragmenting authority and the power of interest groups. But the main thread of Madison’s argument is that interests groups can achieve power at the local level far more readily than at the national level. Federalist 10 provides no support for placing control of the national economy in the hands of interest groups that may dominate certain local legal systems or cultures.
I hope that it is not too simple to agree with James Wilson’s argument to the Pennsylvania ratifying convention that “[w]hatever the object of government extends, in its operation, beyond the bounds of a particular state, should be considered as belonging to the government of the United States’ and “Whatever object of government is confined in its operation and effect, within the bounds of a particular State, should be considered as belonging to the government of that State.” Applying this theory, could we agree that an in-state class action, applying the law of one state to residents of that state or persons injured in the state, usually belongs in state court whereas a multi-state or nationwide class action involving many non-resident plaintiffs, where no one state’s interests are paramount belongs in federal court? This is the position that the Advisory Committee has taken by adopting a resolution in support of federal jurisdiction “for large, multi-state class actions, in which the interests of no one state are paramount, with appropriate limitations or threshold requirements so that the federal courts are not unduly burdened and the states’ jurisdiction over in-state class actions is left undisturbed.” This resolution is consistent with the position taken by the Federal Courts Study Committee in 1990.
All societies have to decide whether to repose certain kinds of decisionmaking in local structures or in a central government. There are always tensions between the benefits of fragmented decisionmaking and the efficiency and uniformity of consolidation. We have a system that is so potentially versatile that we should be able to reap many of the benefits of both diffusion and consolidation. There will always be some parallel litigation between state and federal court. This may advance understanding even if it requires new forms of co-operation among state and federal judges. But there should not be conflict between cases that purport to represent the very same litigants. Nor should local units make decisions binding on the nation. I suggest that the trend toward regulation of national activity through state court litigation is not a good one. It is for the Congress to set national standards for interstate commerce and interstate conduct. We very much need the Congress to act in the mass tort area lest the asbestos debacle become the model.
The federal courts have something important to contribute and the multistate, multidistrict class action should have a high claim on our attention.