Emory Law Journal

Rethinking the Theory of the Class Action: The Risks and Rewards of Capitalistic Socialism in the Litigation Process
Martin H. Redish Louis and Harriet Ancel Professor of Law and Public Policy, Northwestern University School of Law. This article formed the basis for a talk at a seminar sponsored by the American Law Institute in the summer of 2013. However, the views expressed are solely those of the author.

Abstract

Despite all of the controversial scholarship that has been published in recent years concerning the modern class action, it is both puzzling and disappointing how little of it has sought to grasp the deep structural precepts underlying the device. All too often, the scholarly debate, not to mention the political debate, has broken down along ideological lines: the political left has reflexively favored the device and the political right has reflexively opposed it. Virtually all of even the serious scholarly work done on the subject has, for the most part, been superficial, failing to pursue, much less to grasp, the practice’s underlying foundational purposes. The goal of this Article is to seek to understand those foundational purposes. The Article argues that the DNA of the modern class action fundamentally differs from that of the traditional one on-one litigation process. The relationship between class attorney and class member, for example, is significantly different from the normal relationship between attorney and client. Recognition of these foundational differences should force us to recognize that the attorney–client relationship in the class action context is more like a guardian–ward relationship than a traditional relationship between client and attorney.

This insight does not necessarily mean that the class action is inherently improper. Indeed, in some ways recognition of the guardian model of the modern class action underscores the procedure’s value. But it also underscores the need to recognize the inherently capitalistic nature of that guardian relationship. Where profit incentives for the attorney are in accord with the interests of the class members, the practice should work well. However, all too often, modern class action procedure is plagued by externalities and perverse economic incentives, allowing class attorneys to profit even when class members will benefit virtually not at all. The goal of class action doctrine and rulemaking, then, should be to remove those economic perversions to insure that the capitalistic nature of the process functions effectively.

Introduction

With all the controversy that has pervaded legal scholarship on the subject of class actions over the years, it is perhaps surprising that relatively little of it has attempted to place the modern class action (i.e., post-1966 amendment to Rule 23 of the Federal Rules of Civil Procedure) within the broader theoretical framework of our nation’s system of litigation. To be sure, those who are critical of the way in which the class action rule has been interpreted and applied generally assume—albeit implicitly—a particular foundational rationale and conceptual framework, while those who support expanded use of the class action implicitly assume a very different theoretical rationale for the device. But only rarely (and never effectively) has any scholar sought to undertake a detailed and coherent critique of the underlying theory of the modern class action, or to attempt to place it within the broader theoretical framework of the litigation process.

This issue is not one solely of intellectual or academic concern, as interesting as it is on those levels. A proper understanding of the foundational theoretical rationale of the modern class action should logically have a significant impact on the shaping of the structure of the class action device as well as many of its ancillary doctrinal appendages. The goal of this Article is to provide an explanation and critique of the alternative approaches to this inquiry and to advocate an entirely new theoretical rationale as the proper conceptual foundation of the modern class action. It is my belief that when the modern class action is properly understood, its only even arguably legitimate rationale is as an entirely new procedural animal, one conceptually distinct from all the procedural devices to have proceeded it. I label this underlying theory the “Guardianship Model” because it recognizes that, as a practical matter, plaintiff class attorneys are functioning more as capitalistically driven guardians of the absent class members than as one side of a traditional attorney–client relationship.

Before I can reach this conclusion, however, it is incumbent upon me to explain why the existing available rationales fail to grasp the underlying DNA of the modern class action device, either conceptually or practically. Therefore, in this Article, I initially provide a description of the theories of the underlying rationale of today’s class action and provide a critique of those rationales. 1See infra Part II. None of the preexisting theoretical models represents a valid explanation of the modern class action—either because they are incomplete 2See discussion infra Part II. or because they represent an alchemy-like transformation of the underlying substantive law, which the Federal Rules of Civil Procedure are denied power to do by the Rules Enabling Act. 328 U.S.C. § 2072 (2012); see discussion infra Part II.

Up to now, the only available alternative to these seriously flawed theoretical models was what could best be described as the “Aggregation Model”—a view of the modern class action as nothing more than one of a number of procedural aggregation devices available to the federal courts. From this perspective, the class action would differ little, if at all, from the Federal Rules creating the practices of impleader, 4Fed. R. Civ. P. 14. interpleader, 5Fed. R. Civ. P. 22. intervention, 6Fed. R. Civ. P. 24. or compulsory joinder. 7Fed. R. Civ. P. 19. It is true that, in the past, I myself have expressly advocated adoption of such a perspective on the modern class action, mostly as a default to the fatally flawed alternatives, which have been suggested. 8See Martin H. Redish, Wholesale Justice: Constitutional Democracy and the Problem of the Class Action Lawsuit 11 (2009). I still maintain that the Aggregation Model is far preferable to those alternative theoretical models. However, I now recognize that the Aggregation Model is itself flawed because it fails to reflect the unique nature of the relationship between attorney and client in the modern class action. The nature of the relationship between clients and attorneys in the modern class action is like no form of litigation our system has ever seen. For that reason, it is incorrect to equate the class action device with more traditional forms of procedural aggregation. Instead, it is necessary to recognize the modern class action as a fundamentally new and different procedural animal—a wholly original model of litigation with its own unique procedural DNA. Properly understood, the modern class action must be seen as a wholly distinct form of attorney–client relationship from anything ever seen before, where the attorneys function as much as guardians as they do attorneys. In sum, in this Article, I conclude that (1) existing theories of the class action and its proper role are poorly thought out and inaccurate; (2) the modern class action is appropriately viewed as an entirely new procedural animal, distinct from any procedural device that has preceded it; (3) that fact in no way categorically invalidates the class action device; indeed, recognition of the procedure’s true role actually underscores its unique importance in our procedural system; (4) however, recognition of the procedure’s underlying DNA reveals or emphasizes certain litigation pathologies to which the class action is vulnerable and should lead to important reforms designed to curb the potential for those abuses.

The first Part of the Article describes the origins of the modern class action, including its group litigation predecessors and the creation of the modern form of the concept in the 1966 revision of Rule 23. 9Infra Part I. By contrasting the post-1966 version with its predecessors, Part I will underscore the current version’s conceptually unique underpinnings. Part II explores the existing theoretical models of the modern class action and explains their conceptual and practical inadequacy as explications of the underlying theoretical rationale of the modern class action. 10Infra Part II. In Part III, the Article explains why the Guardianship Model provides the only plausible rationale for the modern class action and explores the comparative risks and benefits of creating a multiparty litigation device grounded in such a theoretical model. 11Infra Part III. The final Part explores ways that the procedural pathologies to which a group litigation device grounded in the Guardianship Model may be circumvented or reduced in order to assure the fair and efficient functioning of the modern class action as an implementation of the goals of the Guardianship Model. 12Infra Part IV.

I. The Development of the Modern Class Action: Evolution or Creation?

It has long been widely understood that the modern class action finds its origins in historical practice, beginning with English group litigation during the Middle Ages, followed by the long-established equity practice originated by Justice Story, and culminating in adoption of the original Rule 23 of the Federal Rules of Civil Procedure, promulgated by the Supreme Court in 1938. 13See Redish, supra note 8, at 5–9. Given these venerable historical origins, one might assume that the modern class action, as embodied in the 1966 amendment to Rule 23, constitutes nothing more than the natural evolution of group litigation practice. In reality, however, nothing could be further from the truth. All of the preexisting forms of group litigation differed dramatically from the current form of litigation practice in one enormously important respect: None of the prior forms of group litigation bound absent litigants who lacked a prelitigation substantive legal connection to the named parties. 14See id. The same cannot, of course, be said of the post-1966 version of the federal class action.

Group litigation in Medieval England fit the pre-litigation substantive intersection requirement because individuals functioned as parts of established groups. It was only those in charge of these groups who were authorized to represent the members in court. 15Id. at 6. Justice Story’s famed Equity Rule provided no res judicata impact to any form of representative group litigation, 16Id. at 7. and the original version of Rule 23 gave res judicata effect solely to representative group litigation in which the parties’ legal interests were united substantively in the prelitigation state. 17 Fed. R. Civ. P. 23 (1938). The Rule divided class actions into three subcategories: True, Hybrid, and Spurious. See Redish, supra note 8, at 8; see also Harry Kalven, Jr. & Maurice Rosenfield, The Contemporary Function of the Class Suit, 8 U. Chi. L. Rev. 684, 702–03 (1941). True class actions—which did bind absent class members—involved claims in which prelitigation held rights were substantively joint or common. Redish, supra note 8, at 8. Hybrid class actions, which similarly bound absent class members, involved the sole adjudication of claims that were substantively shared by class members in the prelitigation setting: individually held in rem claims in the same property. Id. While Spurious class actions did not require prelitigation intertwining of rights, resolution of the suit did not bind the absent class members. Id. In short, every form of representative litigation to precede the 1966 amendment to Rule 23 solely involved aggregation of claims that were already linked by substantive law prior to suit. 18See Redish, supra note 8, at 5–9.

It was, then, not until the promulgation of the modern form of Rule 23 in 1966 that our procedural system included a device that allowed representative parties to bind absent class members when their rights possessed no prelitigation substantive link. Though perhaps the drafters of the amendments to Rule 23 did not themselves realize it, 19While there exists no affirmative basis to support the conclusion that the drafters of the 1966 version of Rule 23 were unaware of the dramatic change they were making in representative group litigation practice, one could reasonably assume that if they were in fact aware of so dramatic a conceptual and practical alteration in class group litigation practice they would have noted it in the Advisory Committee notes that accompanied the revised rule. See Fed. R. Civ. P. 23 advisory committee’s note (1966). they were quite clearly creating a procedural mechanism previously unheard of in Anglo-American law. The result of this dramatic alteration in group litigation practice has been the growth of a device that has exhibited conflicting qualities reminiscent of something out of Jekyll and Hyde. If utilized properly in a manner consistent with the underlying substantive law and with appropriate regard for the due process rights of the absent class members, the class action device is capable of fostering important procedural values by overcoming extremely high transaction costs to the enforcement and implementation of substantive rights. 20See Redish, supra note 8, at 54. In a similar manner, guardians perform an important function by looking out for the interests of those who, for whatever reason, are unable to effectively protect their own interests. 21See id. at 39. However, this new procedural animal is particularly dangerous because it should hardly be controversial to assert that plaintiff class action attorneys are for the most part not driven by purely altruistic concerns. Thus, the only means by which those attorneys can function as effective guardians is if their economic incentives overlap with those of their “wards,” the absent class members. Under present circumstances, that is often far from the reality.

II. The Underlying Theory of the Modern Class Action: A Taxonomy

To this point, it would be accurate to assert that practitioners, scholars, and jurists have either proposed or implicitly assumed one of three underlying theories of the modern class action: (1) the Aggregation model; (2) the Entity model; or (3) the Private Attorney General model.

The latter two models—at least in the ways in which scholars have employed them—must be categorically rejected. While the Private Attorney General Model implements legitimate purposes served by the modern class action, it is grossly incomplete and therefore leads to a skewed perspective on the class action, which effectively alters the underlying substantive law being enforced in democratically and constitutionally impermissible ways. 22See id. at 29–35. The Entity Model, on the other hand, has absolutely nothing to recommend it. On the contrary, the model represents an alchemy-like transformation of the nature of underlying substantive rights in a manner that violates core notions of American democratic theory. 23See infra text accompanying notes 30– 35.

The Private Attorney General Model, on one hand, does in fact grasp an important element of the foundational theory of the modern class action because a class action may well perform a law enforcement function by both punishing and deterring unlawful private or governmental behavior. Professor John Coffee has correctly noted that “[p]robably to a unique degree, American law relies upon private litigants to enforce substantive provisions of law that in other legal systems are left largely to the discretion of public enforcement agencies.” 24John C. Coffee, Jr., Understanding the Plaintiff’s Attorney: The Implications of Economic Theory for Private Enforcement of Law Through Class and Derivative Actions, 86 Colum. L. Rev. 669, 669 (1986). But individual litigation may often not be in a position to perform this function effectively because the individual damages are simply too small to justify bringing suit. 25See Redish, supra note 8, at 31–32. In such situations, the class action is uniquely suited to perform the private attorney general function because of its unique aggregative efficiencies. 26It should be noted that the idea of the class action as a private attorney general was introduced long before adoption of the 1966 version of Rule 23. See Kalven & Rosenfield, supra note 17, at 717–21; see also Stephen C. Yeazell, From Medieval Group Litigation to the Modern Class Action 232 (1987). However, the Private Attorney General Model is often relied upon to justify use of the class action device even when absent class members are unlikely to receive the compensation to which they are lawfully entitled. 27See Redish, supra note 8, at 34. The rationale for support of such a structure is that the class action may perform its private attorney general function just as effectively, whether class members are actually compensated or not. 28See id. at 29–35. Such an approach, however, improperly transforms the underlying substantive law by transforming that law from a compensation framework into either a bounty hunter or civil fine framework (as where the uninjured attorneys become the primary financial beneficiaries or cy pres is invoked). 29See id.

On the other hand, the Entity Model, which has been suggested at different points by Professors David Shapiro 30See David L. Shapiro, Class Actions: The Class as Party and Client, 73 Notre Dame L. Rev. 913, 919 (1998). and Sam Issacharoff, 31See Samuel Issacharoff, Preclusion, Due Process, and the Right To Opt Out of Class Actions, 77 Notre Dame L. Rev. 1057, 1060 (2002). presumes that once the claims of individual litigants are grouped together as part of a class, they are mysteriously transformed from individually held claims into a kind of organic entity. As a result, the model’s proponents argue, the due process rights normally accorded an individual litigant—for example, notice of suit or the right to control the litigation of one’s own legal claim—magically disappear. 32For a detailed analysis and critique of the Entity Model, see Redish, supra note 8, at 148–56. But no basis exists, either legally or practically, for such a transformation. 33See id. If the class members’ claims were individually held prior to certification, they must remain so, lest Rule 23 violate the Rules Enabling Act’s prohibition on a rule’s modification of substantive rights, 3428 U.S.C. § 2072(b) (2012). not to mention the Due Process Clause of the Fifth Amendment. 35U.S. Const. amend. V.

This leaves the Aggregation Model. It is, I suppose, this model that I have long implicitly assumed represents the accurate description of the class action’s underlying theoretical framework. From this perspective, the theoretical underpinnings of the class action are fundamentally no different from any of those underlying other aggregation devices provided for in the Federal Rules—whether permissive 36Fed. R. Civ. P. 20. or compulsory joinder, 37Fed. R. Civ. P. 19. interpleader, 38Fed. R. Civ. P. 22; see also 28 U.S.C. § 1335. impleader, 39Fed. R. Civ. P. 14. or intervention. 40Fed. R. Civ. P. 24. In a purely relative sense, of course, I continue to stand by my instinct. Whatever problems the Aggregation Model suffers from (and, as will be seen, there are many), at least when viewed relative to the two alternative discredited models, it should still prevail. Recently, however, I have come to recognize that the Aggregation Model itself suffers from serious flaws. More careful consideration has demonstrated to me that the modern class action differs in important ways from a normal aggregation device. Indeed, as previously explained, the class action is significantly different from any legal device our litigation system has ever known, primarily because of the fundamental ways in which the attorney–client relationship differs from the way in which that relationship has traditionally operated as part of our procedural system. 41See supra Part I.

III. The Unique Nature of the Attorney–Client Relationship in the Modern Class Action

To understand the fundamental ways in which the modern class action differs from both traditional individual litigation and other forms of procedural aggregation (such as permissive 42Fed. R. Civ. P. 20. or compulsory joinder, 43Fed. R. Civ. P. 19. interpleader, 44Fed. R. Civ. P. 22; see also 28 U.S.C. § 1335 (2012). intervention, 45Fed. R. Civ. P. 24. or impleader 46Fed. R. Civ. P. 14. ), consider the numerous and important ways in which the relationship between attorney and client in the modern class action is unique:

  • Unlike the overwhelming majority of traditional individualized or aggregative litigation, the attorneys representing a plaintiff class are invariably the starting and driving forces in the creation of the class and the bringing of suit. In the class action context, the overwhelming majority of clients are inactive through the entire litigation process. 47See Redish, supra note 8, at 147. In contrast, in the traditional litigation context the client generally approaches the attorney with a legal problem or responds to an attorney’s generic advertisement offering services.

  • Class attorneys will likely never have even met, much less consulted, absent class members, even though in every meaningful sense of the term those absent class members are their clients because resolution of the class proceeding will automatically resolve their legal rights, for good or ill. 48It is true that absent class members in Rule 23(b)(3) class actions are permitted to opt out of the litigation, and in such a case resolution of the class proceeding will not necessarily resolve their legal rights. See Redish, supra note 8, at 24; see also Fed. R. Civ. P. 23(c)(2)(B)(v) (requiring notice to (b)(3) class members which “clearly state[s]” that “the court will exclude from the class any member who requests exclusion”). However, class members who have opted out are no longer members of the class and are therefore irrelevant to this analysis. In the traditional context, attorney and client have at the very least met and spoken and may well continue to do so throughout the term of the attorney’s representation.

  • There will often exist potential conflicts of interest between class attorneys and their clients inherent in the class action procedure, which never arise in the traditional attorney–client relationship (e.g., inventorying, which allows class attorneys to bargain away the interests of some claimants in exchange for settlement of a distinct class action, or the incentive to argue for (b)(1) or (b)(2) classification, which would reduce the rights of claimants). 49See Redish, supra note 8, at 24–25.

  • As both practical and legal matters, class attorneys stand to gain more financially than any one of their clients. 50See id. In the overwhelming majority of traditional situations (with the exception of cases involving a few fee-shifting statutes), this will never even be a possibility.

  • A class is capable of existing, and class attorneys may receive substantial fees, even though relatively few of the absent class members will ever be compensated when the class proceeding is successful. In the traditional attorney–client context, in contrast, it is all but inconceivable that this will occur.

In light of these stark differences in the attorney–client relationship between the modern class action and more traditional forms of litigation, there is no way to view the modern class action as directly comparable to a simple aggregation device. More importantly, in the class action context, there is no way to rely on the traditional ethical and personal bonds between attorney and client to avoid or deter the harms that an attorney can cause to her highly vulnerable clients. 51See id. Indeed, in important ways the modern class action may be viewed more as a type of third-party financing of litigation than as a traditional attorney–client relationship. In the case of third-party financing, interested actors are introduced into the litigation who have no connection with or allegiance to the clients; instead, they solely have a financial interest in the outcome, which may or may not always overlap with the interests of the clients. 52See John H. Beisner & Gary A. Rubin, Stopping the Sale on Lawsuits: A Proposal to Regulate Third-Party Investments in Litigation (2012), http://www.instituteforlegalreform.com/uploads/sites/1/TPLF_Solutions.pdf; John Beisner, Jessica Miller & Gary Rubin, Selling Lawsuits, Buying Trouble: Third-Party Litigation Funding in the United States (2009), http://legaltimes.typepad.com/files/thirdpartylitigationfinancing.pdf; see also Joanna M. Shepherd, Ideal Versus Reality in Third-Party Litigation Financing, 8 J.L. Econ. & Pol’y 593, 595 (2012); Ass’n of Trial Lawyers of Am., The Ethics of Financing Cases, in ATLA Winter Convention Reference Materials (2004), available at Westlaw AAJCLE. The main difference between the class action and third-party financing is that in the latter situation the attorney serves as a bridge or buffer between the client, on the one hand, and the third-party financier, on the other. 53See supra note 52. In contrast, there is no buffer in the class action context; the attorneys are the third-party financiers. 54See supra note 52. Thus, if the modern class action is to assure that those who represent the victims of legal harm do so fully, fairly, and adequately, we must conceptualize that procedure in some manner other than as a form of traditional litigation.

In this Article, I attempt to do just that. I label this new model the “Guardianship Model” because in order to prevent the harms caused by the departures from the traditional attorney–client relationship, the class attorneys must be deemed to function as guardians of the interests of the absent class members. By imposing on class attorneys the fiduciary obligations that guardians have to their wards, our procedural system should be able to obtain the benefits of the modern class action while simultaneously avoiding at least most of the pathologies to which the current practice gives rise. 55See infra Part V. I should emphasize that the analogy to a legal guardian is by no means perfect. In some ways, class attorneys function as traditional attorneys, qualitatively different from the traditional guardian model. Thus the Guardianship Model of the modern class action represents a synthesis of traditional procedural aggregation and a form of guardianship, reminiscent of the fiduciary manner in which guardians legally function under long-established principles of equity. 56See 39 Am. Jur. 2d Guardian and Ward § 1 (2008) (defining a guardianship as “a trust relation of the most sacred character, in which one person, called a ‘guardian,’ acts for another, called the ‘ward,’ whom the law regards as incapable of managing his or her own affairs”); id. § 2 (“The relationship between a guardian and a ward is a fiduciary one . . . .”); id. § 99 (2008 & Supp. 2014) (“A fiduciary duty, in a guardianship case, means that one party occupies such a relation to the other party as to justify the latter in expecting that his interests will be cared for.”); see also 39 C.J.S. Guardian & Ward § 80 (2014) (“Generally, in the custody and management of the ward’s estate, a guardian is subject to the control and supervision of the courts, and his or her acts are always open to the rigid scrutiny of an equity court whose duty it is to see that the ward’s interests are not rendered valueless by unauthorized dealings with guardianship trusts.” (footnote omitted)); 4 John Norton Pomeroy, A Treatise on Equity Jurisprudence § 1313, at 886 (Spencer W. Symons ed., 5th ed. 1941) (“When the special statutory jurisdiction has been exercised, a person has been adjudged or ‘found’ a lunatic or otherwise non compos mentis, and a committee or guardian has been appointed, the general jurisdiction of equity extends over such committee or guardian, for the purpose of calling him to an account of his trust, in the same manner as over all other strictly fiduciary persons.”).

IV. The Benefits of the Guardianship Model: The Unique Advantages of Capitalistic Socialism

To suggest that the modern class action is appropriately viewed as a form of guardianship is in no way intended to categorically condemn the device. To the contrary, it underscores the unique value served by the modern class action—what I refer to, paradoxically (or, arguably, oxymoronically), as “Capitalistic Socialism.” I employ this paradoxical phrase because the class action—at least the modern (b)(3) class action 57Fed. R. Civ. P. 23(b)(3) (“[T]he court finds that the questions of law or fact common to class members predominate over any questions affecting only individual members, and that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy. The matters pertinent to these findings include: (A) the class members’ interests in individually controlling the prosecution or defense of separate actions; (B) the extent and nature of any litigation concerning the controversy already begun by or against class members; (C) the desirability or undesirability of concentrating the litigation of the claims in the particular forum; and (D) the likely difficulties in managing a class action.”). —necessarily combines the two normally warring economic theories. The class action functions as a form of litigation socialism because it is, for the most part, designed to redistribute wealth from large concentrations of economic power (either government or corporations) that have allegedly violated the legal rights of smaller entities and individuals. 58See Redish, supra note 8, at 166. I proceed on the assumption that the defendants in (b)(3) class actions are generally wealthy for the simple reason that there is generally no point to bring a damages class action unless the defendant is capable of paying. The practice is capitalistic because the attorneys bringing the proceeding have financial gain as at least one of their motivations. The practice, in theory, fits the old capitalistic mantra that people can do well by doing good.

Concededly, this is a somewhat oversimplified description of the underlying motivations for filing class actions. In certain instances, plaintiff class attorneys may be as motivated by concerns of social justice as much as they are motivated by financial gain. 59See William B. Rubenstein, On What a “Private Attorney General” Is—And Why It Matters, 57 Vand. L. Rev. 2129, 2136–37 (2004), for a discussion of the history of the phrase “private attorney general.” Rubenstein notes a distinction between “the ideological plaintiffs’ attorney and the fees-driven plaintiffs’ attorney.” Id. For example, “the NAACP attorney, paid a small salary for her intense efforts, is generally driven by different incentives than those that motivate the plaintiffs’ attorney subsisting on the fees she can draw from her portfolio of class action lawsuits.” Id.; see also Michael L. Rustad, Smoke Signals from Private Attorneys General in Mega Social Policy Cases, 51 DePaul L. Rev. 511, 515 (2001) (noting a difference between “class actions to bring about social change” and class actions that are financially self-serving to the plaintiffs’ attorneys). But apart, perhaps, from civil rights attorneys who pursue equitable class actions under Rule 23(b)(2) (who may be compensated under the Civil Rights Attorneys Fees Awards Act), 6042 U.S.C. § 1988(b) (2012). it is probably accurate to assert categorically that financial gain plays at least a significant, if not dominant, role in motivating attorney behavior. But again, this is not intended to suggest that the process is inherently illegitimate or corrupt. Those motivated, even exclusively, by considerations of financial gain are often capable of both advancing the interests of social justice and obtaining compensation for injured victims who are not in a position to vindicate their individual rights by themselves. 61See Redish, supra note 8, at 31.

Most important is the fact that the class action device is capable of circumventing the normal inertia and lack of knowledge that would otherwise plague these potential suits. Because the class attorneys function as a type of incentivized guardian, they are able to reduce transaction costs and vindicate individual rights that would otherwise almost certainly go without remedy. But just as the Guardianship Model possesses numerous benefits, so too is it fraught with potential pathologies—in other words, ways in which the class attorneys are perversely incentivized to undermine or ignore the rights and interests of their clients (or, if you will, “wards”), the absent class members. The problem—and the factor that renders the model a hybrid mixture of guardianship and aggregation models—is that capitalistic guardianship is not identical to altruistic or governmental guardianship. 62Cf. id. at 39 (discussing governmental guardianship). Capitalistic guardianship, in contrast to the other two forms, is, in the end, driven predominantly, if not exclusively, by concerns of profit. As already explained, this profit motive works well when the capitalistic interests of the guardian are inherently intertwined with the interests of the “wards,” but it is easily recognizable that this is not always so in the case of the modern class action. The device is vulnerable to a variety of externalities—in a number of situations, the guardian’s profit incentive and the best interests of the absent class members are likely to diverge. 63See id. at 30–32. In other words, situations all too often arise in which the class attorneys may achieve their goal of financial gain without vindicating the rights of or achieving compensation for those whose interests they purportedly represent. 64See id. at 24.

This relationship is in stark contrast to the nature of the attorney–client relationship in traditional one-on-one litigation for several reasons. First, it is simply difficult to conceive of situations in the context of individualized litigation in which the attorney could profit without benefiting the client. Second, even if it were possible to hypothesize such a situation, the attorney–client bond established in such a situation will usually deter such behavior, if only due to the fact of a facetoface meeting. Neither of these “speed bumps,” however, exists in the class action context. As will be seen in the following Part, 65See infra Part V. there exist numerous factors unique to the modern class action that enable or invite attorney profit without significant benefit to class members. Moreover, such a breach of the guardian’s fiduciary obligation will not be deterred by the bond established in the one-on-one attorney–client relationship. 66See discussion supra Part III. For that reason, it is vitally important to recognize and isolate the relevant externalities or perverse incentives and insert legally prophylactic devices designed to avoid them. For the class action device to be employed most effectively, then, our goal should be to allow the beneficial purposes of capitalistic socialism to be successfully achieved while, at the same time, removing the perverse incentives or economic externalities that have traditionally plagued the behavior of class attorneys in the modern class action. It is to this task that the analysis now turns.

V. Use of the Guardianship Model to Avoid the Pathologies of the Modern Class Action

The stark reality of the modern class action is that in a number of contexts, class attorneys’ interests are not necessarily intertwined with those of their clients. 67See discussion supra Part III. The class attorneys may make a great deal of money out of what superficially appears to be the successful resolution of a class action, while at the same time an overwhelming number of the absent class members benefit not at all. 68See also Redish, supra note 8, at 34. A substantial award may be obtained in litigation or settlement, with the money for all practical purposes being split between the class attorneys and some charitable institution, which was not a client and was in no way a victim of the defendant’s unlawful behavior. 69See Martin H. Redish, Peter Julian & Samantha Zyontz, Cy Pres Relief and the Pathologies of the Modern Class Action: A Normative and Empirical Analysis, 62 Fla. L. Rev. 617 (2010).

Of equal importance is the fact that the financial interests of class attorneys and those of the absent class members will often directly clash. For example, class attorneys may fight with all their resources so that their clients will be denied the protections of notice and opt-out. Indeed, this is exactly what took place in the famed Wal-Mart litigation. 70Wal-Mart Stores, Inc. v. Dukes, 131 S. Ct. 2541 (2011). In what appears to be a stroke of tremendous irony, the very same factors that render the class attorneys so important as a guardian of the legal interests of absent class members simultaneously plant the seed for the possibility of grave abuse. 71See supra notes 62–64 and accompanying text. In both cases, it is the transaction costs involved in the litigation of small individual claims, caused by inertia and lack of information, that are plaguing the absent class members. Moreover, because absent clients are often faceless and nameless to the class attorneys, the disincentive to abuse normally created by the bond growing out of the attorney–client relationship is dangerously absent. 72See supra notes 62–64 and accompanying text.

Some might respond that the system has already taken account of these dangers and protected against them. For example, the certification requirements of Rule 23(a) already include prerequisites that the named parties (and their attorneys) adequately represent the class. 73Fed. R. Civ. P. 23(a)(4). Every settlement of a certified class proceeding must be approved by the court. 74Fed. R. Civ. P. 23(e). But surely it is easy to recognize how feeble these mechanisms are in preventing the abuses that plague the Guardianship Model. This is especially true when the defendant and the class attorneys present a united front to the court as the result of a settlement class action. 75For a constitutionally based critique of the settlement class action, see generally Martin H. Redish & Andrianna D. Kastanek, Settlement Class Actions, the Case-or-Controversy Requirement, and the Nature of the Adjudicatory Process, 73 U. Chi. L. Rev. 545 (2006). As a totally passive institution, the court counts on the litigants to provide all relevant information, and where both sides are already in agreement, neither party has the incentive to provide the counter to the pro-certification argument. 76See id. at 550. Additional remedies are required to protect both the absent class members and, on occasion, the party opposing the class.

VI. Proposals for Remedying the Pathologies of the Modern Class Action by Means of the Guardianship Model

A. Overview

In this Article, I propose a variety of remedies to protect absent class members against the perverse economic incentives or economic externalities that plague the modern class action. In this way, the system will be able to prevent undermining attainment of the goals of the Guardianship Model. These proposals include the following:

B. Modifying the Certification Requirements of Rule 23(a)

The greatest concern about the modern class action from a systemic perspective is what I have referred to in previous writing as the “faux” class action. 81See Redish, supra note 8, at 24–35. On its face, the faux class action appears strikingly similar to a wholly proper class action: There are named plaintiffs and a theoretically categorizable group of injured absent plaintiffs who can be compensated as the result of successful class proceeding. 82See id. But either because the individual damage claims of the absent plaintiffs are so small as not to justify the effort required to file a complaint, or the costs of identifying specific absent class members and their location are prohibitive, or both, these class actions often amount to nothing more than the equivalent of a cardboard cutout of a real class action. 83Id. at 25.

The proceeding is resolved in favor of the plaintiff class, but virtually no absent class members receive compensation. To be sure, a significant award may be justified on the merits, or the defendants may be more than happy to settle in order to impose a res judicata ban on any injured claimant, and the attorneys will likely be compensated substantially on the basis of the sum total of that award or settlement. 84See Redish & Kastanek, supra note 75, at 549–50. However, since attorney compensation is determined by the amount of the award or settlement rather than exclusively by the amount actually awarded to injured claimants, the economic incentive of the attorneys and the interests of those for whom they act as guardians need not overlap. 85See supra notes 51–56 and accompanying text; see also Redish & Kastanek, supra note 75, at 551. As a result, the attorneys’ incentive to ascertain the existence and location of specific absent claimants is all but non-existent. As a result, there is, at the very least, a real danger that the attorneys will operate as the equivalent of bounty hunters, rather than exclusively on behalf of the interests of those who are effectively unable to help themselves. 86Redish, supra note 8, at 26. In short, the system allows the corruption of capitalistic socialism by introducing what is, at the very least, an economic externality and arguably even a perverse economic incentive that undermines their obligation, even as wholly profit-driven actors, to operate in the best interests of those for whom they act as guardians.

It would not be all that difficult to avoid this difficulty. The Supreme Court could simply add to the certification requirements of Rule 23(a) that those seeking certification establish that “meaningful relief” will be provided to the absent class members, assuming a settlement is reached or a judicial award is made. Indeed, it is somewhat surprising that no such requirement exists now, since it is, after all, the absent class members on whose behalf the suit is presumably brought.

The response of many will no doubt be that even if absent class members cannot effectively be compensated, the class action still may perform an extremely valuable function as a type of private attorney general action. 87See discussion supra Part II; see also Deborah R. Hensler, Revisiting the Monster: New Myths and Realities of Class Action and Other Large Scale Litigation, 11 Duke J. Comp. & Int’l L. 179, 182–83 (2001); Thomas D. Rowe, Jr., The Legal Theory of Attorney Fee Shifting: A Critical Overview, 1982 Duke L.J. 651, 662. In other words, the class action may achieve social justice without compensating class members. But as already noted, 88See discussion supra Part II. there are structurally fatal problems with this argument. Initially, it is unresponsive to the contention that allowing certification of such faux class actions will threaten the fiduciary relationship between class attorneys and class members by reducing the incentive to find absent class members who are difficult to locate. Moreover, it ignores the systemic difficulty that the underlying substantive law—the law being enforced in the class proceeding—authorizes attainment of this social justice goal exclusively through the means of victim compensation. To authorize bounty-hunter/faux class actions pursuant to Rule 23 clearly modifies the underlying substantive law in direct violation of the clear directive of the Rules Enabling Act. 8928 U.S.C. § 2072 (2012) (“(a) The Supreme Court shall have the power to prescribe general rules of practice and procedure and rules of evidence for cases in the United States district courts (including proceedings before magistrate judges thereof) and courts of appeals. (b) Such rules shall not abridge, enlarge or modify any substantive right. All laws in conflict with such rules shall be of no further force or effect after such rules have taken effect. (c) Such rules may define when a ruling of a district court is final for the purposes of appeal under section 1291 of this title.”). This practice also violates core notions of American democratic theory. 90For a detailed discussion of the implications of such a model for democratic theory, see Redish, supra note 8, at 21–61. If a rule of procedure is allowed to transform the remedial element of underlying substantive law from victim compensation to a form of bounty-hunter incentivization, the substantive law will have been altered without the electorates’ awareness. Instead, the law will have been transformed through the smokescreen of procedure.

One might respond to my argument in support of the insertion into Rule 23(a) of a new certification requirement that if my goal were truly to protect the absent claimants, it makes little sense to prohibit certification. My reply to such a response is twofold. Initially, in addition to the interests of the absent claimants, I also recognize important structural and systemic interests in not permitting an aggregation device to modify underlying substantive law through the back door of procedure. Secondly, insertion of this additional certification requirement would incentivize class attorneys to bring only those class proceedings in which they could feasibly compensate injured parties.

It is true that even if one were to accept my replies to these criticisms, the problem of the inherent ambiguity of the concept of “meaningful relief” would remain. Even where the overwhelming portion of absent claimants will not receive compensation, no doubt at least a few of them will. I believe, however, that it is appropriate to leave the task of defining the phrase in an individualized, inductive manner to the certifying court. For the most part, defining the concept should not be all that difficult. Common sense should dictate, for example, that where at most three or four percent of the absent class is likely to receive compensation, the relief given to the class is not “meaningful.” Using that postulate as a baseline, it should not be all that difficult to determine the rough contours of the concept. The fact that disputes will arise at the margins in no way distinguishes this guideline from many long established legal standards.

C. Modifying Attorney Compensation Methodology

An alternative method for deterring the bringing of faux class actions by economically driven class attorneys would be to modify the mode of attorney compensation in order to internalize the need for actual client compensation. At the present time, class attorneys’ compensation is effectively, if not technically, determined as a percentage of the classwide award or settlement as a whole. 91See 7B Charles Alan Wright, Arthur R. Miller & Mary Kay Kane, Federal Practice and Procedure § 1803.1, at 345 (3d ed. 2005), whose authors noted that in class actions the “approaches taken to setting fees [have] varied between cases in which the courts have used what is referred to as the ‘lodestar’ method to set fees, to cases in which a ‘percentage-of-recovery’ method has been utilized, and to yet other cases in which courts have experimented with competitive bidding.” Id. Difficulties associated with the lodestar method “[have] persuaded some courts to abandon the lodestar method and adopt a percentage of recovery approach or some combination of the two.” Id. at 349 n.22; see also Blum v. Stenson, 465 U.S. 886, 900 n.16 (1984) (“[U]nder the ‘common fund doctrine,’ . . . a reasonable fee is based on a percentage of the fund bestowed on the class. . . .”); In re Twinlab Corp. Sec. Litig., 187 F. Supp. 2d 80, 85 (E.D.N.Y. 2002) (“The trend in the Second Circuit is to use the percentage method.”); Ferdinand S. Tinio, Annotation, Attorneys’ Fees in Class Actions, 38 A.L.R.3d 1384, § 3 (1971 & Supp. 2014) (explaining that “[u]nder the percentage of fund method . . . a court shapes the counsel fee based on what it determines is a reasonable percentage of the fund recovered for those benefited by the litigation” and several factors may be considered to ensure the fee is reasonable). Thus, compensation turns not at all on either the number of claimants who were actually compensated, or the total amount of the award or settlement paid out to claimants. 92See Tinio, supra note 91, § 3. The economic incentive of class attorneys, then, is fully satisfied the minute the award or settlement as a whole is approved. 93See Andrew Rosenfield, An Empirical Test of Class-Action Settlement, 5 J. Legal Stud. 113, 115 (1976). Achieving actual compensation of absent class members at that point is transformed into an economic externality. 94See id.

Putting aside the question whether this externality threatens the constitutionally protected due process rights of absent class members, 95For a discussion of the due process rights of absent class members, see generally Redish, supra note 8, at 135–75. one can assert without serious doubt that it undermines the performance of the attorneys’ capitalistically driven guardian function. This is so for the simple reason that under the present framework attorneys’ compensation is not at all affected by the fact of claimants’ compensation. 96See Tinio, supra note 91, § 3.

In a certain sense, the suggestion that attorney compensation be measured solely by actual class member compensation is analogous to the important modification of attorney compensation achieved by a provision of the Class Action Fairness Act of 2005. 9728 U.S.C. § 1712 (2012) (referring to the substantive coupon settlement provisions). Prior to the enactment of that legislation, class settlements were often approved when class compensation came exclusively in the form of coupons providing discounts of some sort for purchase of defendants’ products or services. 98See Martin H. Redish, Class Actions and the Democratic Difficulty: Rethinking the Intersection of Private Litigation and Public Goals, 2003 U. Chi. Legal F. 71, 78 (“[W]hen the litigation dust settles, even in cases in which the plaintiff class prevails (either by means of judgment or settlement), often the overwhelming majority of class members never receive anything approaching meaningful compensation for the defendants’ violation of their substantive rights. Instead, they are frequently ‘awarded’ the opportunity to receive some form of discount coupon for purchase of a product or service already provided by the defendant in the normal course of business.”). Under the new legislation, in any federal class action in which settlement comes in the form of coupons, attorney compensation likewise must come in the form of coupons. 99See 28 U.S.C. § 1712. Similarly, I am suggesting that a compensatory method be developed in which class attorneys are compensated only to the extent that claimants themselves have been compensated.

If such a compensatory methodology were developed, class attorneys’ economic incentives would trigger one of two alternative courses of behavior: (1) vigorous efforts to ascertain and compensate individual class members, or (2) a refusal to bring the class action in the first place because of a determination that the likelihood of significant compensation is so small as not to justify the effort involved. Either way, the pathology brought about by the externality caused by existing compensation methods would be substantially reduced.

Exactly when and how class attorneys should be compensated under this revised philosophy is beyond the scope of this Article. Suffice it to say that once it is understood that the goal should be to reduce or remove the economic externality brought about by the existing compensatory methodology, it should not be too difficult to develop a functional alternative.

D. Abolition of Cy Pres Relief

Cy pres relief refers to what in prior scholarship my coauthors and I described as “an effort to provide unclaimed compensatory funds to a charitable interest that is in some way related to either the subject of the case or the interests of the victims, broadly defined.” 100Redish, Julian & Zyontz, supra note 69, at 620. While the doctrine has venerable origins in the law of trusts, 101Id. at 624. The phrase derives from the French expression, cy pres comme possible, which means “as near as possible.” Id. For a description of the doctrine’s historical development in the law of trusts, see id. at 624–30. its use in the modern law of class actions goes back only to the 1970s. 102See id. at 630–38. By the early 2000s, the concept’s use expanded dramatically. 103See id. at 634–38. In recent years, federal courts of appeals appear to have developed a certain level of buyers’ remorse over the rapid expansion of cy pres in class action litigation. 104See, e.g., In re Baby Prods. Antitrust Litig., 708 F.3d 163 (3d Cir. 2013); Dennis v. Kellogg Co., 697 F.3d 858 (9th Cir. 2012); Klier v. Elf Atochem N. Am., Inc., 658 F.3d 468 (5th Cir. 2011). Moreover, even when courts of appeals have approved cy pres settlements, they have often imposed restrictions on their future use. 105See, e.g., In re Lupron Mktg. & Sales Practices Litig., 677 F.3d 21 (1st Cir. 2012).

As my coauthors and I demonstrated in our prior work on the subject, cy pres in the class action context is fraught with serious practical and constitutional problems. 106See Redish, Julian & Zyontz, supra note 69, at 641–51. For present purposes, however, it is important to focus on the manner in which cy pres gives rise to a pathological externality that undermines the capitalistic guardian model of the modern class action.

Recall that under the Guardian Model, as I have described it here, at least as a general matter, class attorneys function as a capitalistically driven guardian on behalf of the interests of the absent class members. 107See Tinio, supra note 91, § 3. For the most part, they lack whatever bonds develop between client and attorney based on the traditional one-on-one relationship for the simple reason that the overwhelming portion of the absent class remains faceless to class attorneys. 108See discussion supra note 48. While in a few instances, class attorneys in Rule 23(b)(3) damages class actions will be motivated by predominantly ideological considerations, it is safe to say that these instances will be sufficiently small in number that it is impossible, as an ex ante matter, to assume the existence of such a motivation. 109See discussion supra note 50. As a practical matter, then, the motivation for the effective performance of the guardian function will be largely capitalistic: if the absent class members benefit economically, the attorneys will also benefit economically. 110See supra text accompanying notes 65–66. Thus, any situation in which the attorneys may benefit economically when their “wards” do not undermines the capitalistic incentives so essential to the Guardian Model of the modern class action. 111See supra notes 55–56, 62–64 and accompanying text.

It is difficult to imagine a more obvious economic externality than cy pres. The attorneys receive compensation whether or not absent class members are compensated. While of course the same could be said under current standards of all class actions, 112See supra notes 59–61 and accompanying text. the addition of cy pres gives rise to an entirely distinct problem because it creates the illusion of compensation. 113Redish, Julian & Zyontz, supra note 69, at 623. Where no one receives compensation except the attorneys, it is surely the case that eyebrows would be raised. However, when cy pres is invoked, the class proceeding takes on the patina of compensation because a charity is benefitted as a result. 114See id. at 620. But the charity was never injured; the absent class members were the defendant’s victims. Yet they are, for the most part, not the ones receiving compensation. 115See id. at 621–22. Even in a more idealistically based guardian–ward relationship in which financial gain plays a far less prominent role, a system in which the guardian is fully compensated even when he fails to perform his assigned function of protecting the interests of his ward would have to be deemed the creation of a dangerous conflict of interests. 116See supra notes 62–64 and accompanying text. In a situation in which the guardians’ incentives are confined to the existence of significant economic benefit, making those benefits available to the guardians without requiring them to perform the function of protecting the interests of their wards either makes no sense, or cynically renders the guardian relationship little more than a sham. When cy pres is invoked, it functions to provide camouflage to that sham.

In my previous scholarship on the subject, I suggested that under limited circumstances there might be an appropriate role for cy pres—for example, when the large majority of the class has been compensated but there is still money left over from the award or settlement fund. 117Redish, Julian & Zyontz, supra note 69, at 621–22. I now realize, however, that cy pres is so dangerous, both as a perverse economic incentive for the class attorneys and as a disguise to the court as well as to outside observers, that, if only as a prophylactic measure, the cy pres concept must be categorically rejected. Any unclaimed money must either be returned to the defendant or escheat to the government, so that all involved can be reminded of what the true purpose of the class proceeding is—to benefit the victims whose legal rights have been violated.

E. Recognizing Class Attorneys as the Real Parties in Interest for Purposes of Res Judicata

If I am correct in my assertion that class attorneys are appropriately viewed more as guardians than as traditional legal representatives of clients, it is appropriate to modify the traditional allocation of responsibility between attorney and client for purposes of the law of judgments.

Note that I do not argue that class attorneys should be deemed the real parties in interest for all purposes. In most senses, for purposes of the real party in interest inquiry, class attorneys will be treated as attorneys, rather than as guardians. This represents an application of the hybrid nature of the model. However, in this one instance, because it is the class attorneys making the strategic decisions and named plaintiffs in a large class are totally fungible, the disincentive to seek multiple certifications needs to be imposed on the attorneys, not the class members. This represents a practical, rather than a formalistic, application of legal definitions. Just as a corporation may be a “citizen” for one provision of the Constitution but not for another, so too should the law be able to define legal terms with regard to the practical consequences of that definition in a particular context. Surely, this type of sociological jurisprudence should not be deemed shocking in the twenty-first century.

Because I have explored this issue in more detail in other recent scholarship, 118See generally Martin H. Redish & Megan B. Kiernan, Avoiding Death by a Thousand Cuts: The Relitigation of Class Certification and the Realities of the Modern Class Action, 99 Iowa L. Rev. 1659, 1662 (2014) (“[C]lass attorneys should be deemed real parties in interest on the certification issue, and therefore the direct estoppel impact of a certification denial should bar subsequent certification attempts not only by the prior named plaintiffs, but also by the class attorneys.”). I have chosen not to delve into the issue in detail here. However, it is important to contrast this pathology with those previously discussed. 119See Redish, supra note 98. While those other pathologies, if uncorrected, threaten to undermine the fiduciary relationship between guardian and ward inherent in adoption of the Guardianship Model, this pathology is more of a perverse byproduct of the Guardianship Model itself.

The Guardianship Model itself finds nothing inherently wrong with the extraordinary degree of power vested in class attorneys, as long as the economic incentives function free from externalities. 120See discussion supra at Part VI.C. But even when the Guardianship Model is operating at a level of full effectiveness, potentially perverse impacts exist on others outside of the guardian–ward relationship, which need to be taken into account. 121To a certain extent, the proposals put forth in this Article focus not exclusively on the protections of the absent class members but also on the interests of the system, by assuring that faux class actions not be permitted. See discussion supra at Part VI.B. In the case of res judicata and the real party in interest, that outsider is the defendant who loses the protections for which res judicata has been designed. Traditional res judicata doctrine was established on an assumption of the existence of a traditional attorney–client relationship, in which the client is the actor in the litigation process who both makes all final decisions and is impacted most directly and substantially by negative litigation decisions. Traditional res judicata thus aims its negative incentives at the litigant, rather than the attorney. But in the modern class action, as we have already seen, both the primary decision making and the primary financial impact are borne not by the litigants—especially the absent class members—but rather by the attorneys. 122See supra text accompanying notes 47–50. While this is permissible from the perspective of the Guardianship Model, it results in the effective emasculation of res judicata as a protection of defendants against wasteful and harassing redundant litigation. Thus, in addition to the modifications of the modern class action necessary to make the Guardianship Model an effective protection of absent class members, it is also necessary to modify the target of the negative incentives imposed by traditional res judicata doctrine. This goal is achieved by viewing the class attorneys, as well as the named class plaintiffs, as the real parties in interest and therefore as the target of the doctrine’s negative incentivization.

Conclusion: The Need to Restructure Our Understanding of the Modern Class Action

As is so often the case, focusing on first principles may well have a significant impact on the law in the trenches. This Article is designed to illustrate this linkage in the specific context of class actions.

By recognizing how dramatically different the modern class action is from all preexisting procedural formats and zeroing in on the true nature of that new form, we are able to see both the benefits and dangers that accompany this procedural revolution. Our goal should naturally be to accentuate the benefits and avoid the pathologies of this new procedural animal. The way to accomplish that goal, I have argued, is to view the modern class action not simply as an aggregative form of traditional litigation but rather as a guardianship arrangement in which class attorneys are viewed as the fiduciary bound guardians of the absent class members. 123See discussion supra at Part V. Once this relationship is recognized as the foundational legitimate description of the modern class action’s underlying theory, we should be in a position to recognize how current law must be modified in order to preserve the guardians’ capitalistic incentives to further the interests of their wards. 124See discussion supra at Part VI.B–E.

Some may disagree with exactly how I propose to accomplish that goal. Even were my recommendations to be accepted, important details would remain to be worked out. However, even if this Article is able to accomplish no more than to begin a dialogue about the unique theoretical foundation of the modern class action, it will have achieved an important goal.

Footnotes

Louis and Harriet Ancel Professor of Law and Public Policy, Northwestern University School of Law. This article formed the basis for a talk at a seminar sponsored by the American Law Institute in the summer of 2013. However, the views expressed are solely those of the author.

1See infra Part II.

2See discussion infra Part II.

328 U.S.C. § 2072 (2012); see discussion infra Part II.

4Fed. R. Civ. P. 14.

5Fed. R. Civ. P. 22.

6Fed. R. Civ. P. 24.

7Fed. R. Civ. P. 19.

8See Martin H. Redish, Wholesale Justice: Constitutional Democracy and the Problem of the Class Action Lawsuit 11 (2009).

9Infra Part I.

10Infra Part II.

11Infra Part III.

12Infra Part IV.

13See Redish, supra note 8, at 5–9.

14See id.

15Id. at 6.

16Id. at 7.

17 Fed. R. Civ. P. 23 (1938). The Rule divided class actions into three subcategories: True, Hybrid, and Spurious. See Redish, supra note 8, at 8; see also Harry Kalven, Jr. & Maurice Rosenfield, The Contemporary Function of the Class Suit, 8 U. Chi. L. Rev. 684, 702–03 (1941). True class actions—which did bind absent class members—involved claims in which prelitigation held rights were substantively joint or common. Redish, supra note 8, at 8. Hybrid class actions, which similarly bound absent class members, involved the sole adjudication of claims that were substantively shared by class members in the prelitigation setting: individually held in rem claims in the same property. Id. While Spurious class actions did not require prelitigation intertwining of rights, resolution of the suit did not bind the absent class members. Id.

18See Redish, supra note 8, at 5–9.

19While there exists no affirmative basis to support the conclusion that the drafters of the 1966 version of Rule 23 were unaware of the dramatic change they were making in representative group litigation practice, one could reasonably assume that if they were in fact aware of so dramatic a conceptual and practical alteration in class group litigation practice they would have noted it in the Advisory Committee notes that accompanied the revised rule. See Fed. R. Civ. P. 23 advisory committee’s note (1966).

20See Redish, supra note 8, at 54.

21See id. at 39.

22See id. at 29–35.

23See infra text accompanying notes 30– 35.

24John C. Coffee, Jr., Understanding the Plaintiff’s Attorney: The Implications of Economic Theory for Private Enforcement of Law Through Class and Derivative Actions, 86 Colum. L. Rev. 669, 669 (1986).

25See Redish, supra note 8, at 31–32.

26It should be noted that the idea of the class action as a private attorney general was introduced long before adoption of the 1966 version of Rule 23. See Kalven & Rosenfield, supra note 17, at 717–21; see also Stephen C. Yeazell, From Medieval Group Litigation to the Modern Class Action 232 (1987).

27See Redish, supra note 8, at 34.

28See id. at 29–35.

29See id.

30See David L. Shapiro, Class Actions: The Class as Party and Client, 73 Notre Dame L. Rev. 913, 919 (1998).

31See Samuel Issacharoff, Preclusion, Due Process, and the Right To Opt Out of Class Actions, 77 Notre Dame L. Rev. 1057, 1060 (2002).

32For a detailed analysis and critique of the Entity Model, see Redish, supra note 8, at 148–56.

33See id.

3428 U.S.C. § 2072(b) (2012).

35U.S. Const. amend. V.

36Fed. R. Civ. P. 20.

37Fed. R. Civ. P. 19.

38Fed. R. Civ. P. 22; see also 28 U.S.C. § 1335.

39Fed. R. Civ. P. 14.

40Fed. R. Civ. P. 24.

41See supra Part I.

42Fed. R. Civ. P. 20.

43Fed. R. Civ. P. 19.

44Fed. R. Civ. P. 22; see also 28 U.S.C. § 1335 (2012).

45Fed. R. Civ. P. 24.

46Fed. R. Civ. P. 14.

47See Redish, supra note 8, at 147.

48It is true that absent class members in Rule 23(b)(3) class actions are permitted to opt out of the litigation, and in such a case resolution of the class proceeding will not necessarily resolve their legal rights. See Redish, supra note 8, at 24; see also Fed. R. Civ. P. 23(c)(2)(B)(v) (requiring notice to (b)(3) class members which “clearly state[s]” that “the court will exclude from the class any member who requests exclusion”). However, class members who have opted out are no longer members of the class and are therefore irrelevant to this analysis.

49See Redish, supra note 8, at 24–25.

50See id.

51See id.

52See John H. Beisner & Gary A. Rubin, Stopping the Sale on Lawsuits: A Proposal to Regulate Third-Party Investments in Litigation (2012), http://www.instituteforlegalreform.com/uploads/sites/1/TPLF_Solutions.pdf; John Beisner, Jessica Miller & Gary Rubin, Selling Lawsuits, Buying Trouble: Third-Party Litigation Funding in the United States (2009), http://legaltimes.typepad.com/files/thirdpartylitigationfinancing.pdf; see also Joanna M. Shepherd, Ideal Versus Reality in Third-Party Litigation Financing, 8 J.L. Econ. & Pol’y 593, 595 (2012); Ass’n of Trial Lawyers of Am., The Ethics of Financing Cases, in ATLA Winter Convention Reference Materials (2004), available at Westlaw AAJCLE.

53See supra note 52.

54See supra note 52.

55See infra Part V.

56See 39 Am. Jur. 2d Guardian and Ward § 1 (2008) (defining a guardianship as “a trust relation of the most sacred character, in which one person, called a ‘guardian,’ acts for another, called the ‘ward,’ whom the law regards as incapable of managing his or her own affairs”); id. § 2 (“The relationship between a guardian and a ward is a fiduciary one . . . .”); id. § 99 (2008 & Supp. 2014) (“A fiduciary duty, in a guardianship case, means that one party occupies such a relation to the other party as to justify the latter in expecting that his interests will be cared for.”); see also 39 C.J.S. Guardian & Ward § 80 (2014) (“Generally, in the custody and management of the ward’s estate, a guardian is subject to the control and supervision of the courts, and his or her acts are always open to the rigid scrutiny of an equity court whose duty it is to see that the ward’s interests are not rendered valueless by unauthorized dealings with guardianship trusts.” (footnote omitted)); 4 John Norton Pomeroy, A Treatise on Equity Jurisprudence § 1313, at 886 (Spencer W. Symons ed., 5th ed. 1941) (“When the special statutory jurisdiction has been exercised, a person has been adjudged or ‘found’ a lunatic or otherwise non compos mentis, and a committee or guardian has been appointed, the general jurisdiction of equity extends over such committee or guardian, for the purpose of calling him to an account of his trust, in the same manner as over all other strictly fiduciary persons.”).

57Fed. R. Civ. P. 23(b)(3) (“[T]he court finds that the questions of law or fact common to class members predominate over any questions affecting only individual members, and that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy. The matters pertinent to these findings include: (A) the class members’ interests in individually controlling the prosecution or defense of separate actions; (B) the extent and nature of any litigation concerning the controversy already begun by or against class members; (C) the desirability or undesirability of concentrating the litigation of the claims in the particular forum; and (D) the likely difficulties in managing a class action.”).

58See Redish, supra note 8, at 166.

59See William B. Rubenstein, On What a “Private Attorney General” Is—And Why It Matters, 57 Vand. L. Rev. 2129, 2136–37 (2004), for a discussion of the history of the phrase “private attorney general.” Rubenstein notes a distinction between “the ideological plaintiffs’ attorney and the fees-driven plaintiffs’ attorney.” Id. For example, “the NAACP attorney, paid a small salary for her intense efforts, is generally driven by different incentives than those that motivate the plaintiffs’ attorney subsisting on the fees she can draw from her portfolio of class action lawsuits.” Id.; see also Michael L. Rustad, Smoke Signals from Private Attorneys General in Mega Social Policy Cases, 51 DePaul L. Rev. 511, 515 (2001) (noting a difference between “class actions to bring about social change” and class actions that are financially self-serving to the plaintiffs’ attorneys).

6042 U.S.C. § 1988(b) (2012).

61See Redish, supra note 8, at 31.

62Cf. id. at 39 (discussing governmental guardianship).

63See id. at 30–32.

64See id. at 24.

65See infra Part V.

66See discussion supra Part III.

67See discussion supra Part III.

68See also Redish, supra note 8, at 34.

69See Martin H. Redish, Peter Julian & Samantha Zyontz, Cy Pres Relief and the Pathologies of the Modern Class Action: A Normative and Empirical Analysis, 62 Fla. L. Rev. 617 (2010).

70Wal-Mart Stores, Inc. v. Dukes, 131 S. Ct. 2541 (2011).

71See supra notes 62–64 and accompanying text.

72See supra notes 62–64 and accompanying text.

73Fed. R. Civ. P. 23(a)(4).

74Fed. R. Civ. P. 23(e).

75For a constitutionally based critique of the settlement class action, see generally Martin H. Redish & Andrianna D. Kastanek, Settlement Class Actions, the Case-or-Controversy Requirement, and the Nature of the Adjudicatory Process, 73 U. Chi. L. Rev. 545 (2006).

76See id. at 550.

77See infra Part VI.B.

78See infra Part VI.C.

79See infra Part VI.D.

80See infra Part VI.E.

81See Redish, supra note 8, at 24–35.

82See id.

83Id. at 25.

84See Redish & Kastanek, supra note 75, at 549–50.

85See supra notes 51–56 and accompanying text; see also Redish & Kastanek, supra note 75, at 551.

86Redish, supra note 8, at 26.

87See discussion supra Part II; see also Deborah R. Hensler, Revisiting the Monster: New Myths and Realities of Class Action and Other Large Scale Litigation, 11 Duke J. Comp. & Int’l L. 179, 182–83 (2001); Thomas D. Rowe, Jr., The Legal Theory of Attorney Fee Shifting: A Critical Overview, 1982 Duke L.J. 651, 662.

88See discussion supra Part II.

8928 U.S.C. § 2072 (2012) (“(a) The Supreme Court shall have the power to prescribe general rules of practice and procedure and rules of evidence for cases in the United States district courts (including proceedings before magistrate judges thereof) and courts of appeals. (b) Such rules shall not abridge, enlarge or modify any substantive right. All laws in conflict with such rules shall be of no further force or effect after such rules have taken effect. (c) Such rules may define when a ruling of a district court is final for the purposes of appeal under section 1291 of this title.”).

90For a detailed discussion of the implications of such a model for democratic theory, see Redish, supra note 8, at 21–61.

91See 7B Charles Alan Wright, Arthur R. Miller & Mary Kay Kane, Federal Practice and Procedure § 1803.1, at 345 (3d ed. 2005), whose authors noted that in class actions the “approaches taken to setting fees [have] varied between cases in which the courts have used what is referred to as the ‘lodestar’ method to set fees, to cases in which a ‘percentage-of-recovery’ method has been utilized, and to yet other cases in which courts have experimented with competitive bidding.” Id. Difficulties associated with the lodestar method “[have] persuaded some courts to abandon the lodestar method and adopt a percentage of recovery approach or some combination of the two.” Id. at 349 n.22; see also Blum v. Stenson, 465 U.S. 886, 900 n.16 (1984) (“[U]nder the ‘common fund doctrine,’ . . . a reasonable fee is based on a percentage of the fund bestowed on the class. . . .”); In re Twinlab Corp. Sec. Litig., 187 F. Supp. 2d 80, 85 (E.D.N.Y. 2002) (“The trend in the Second Circuit is to use the percentage method.”); Ferdinand S. Tinio, Annotation, Attorneys’ Fees in Class Actions, 38 A.L.R.3d 1384, § 3 (1971 & Supp. 2014) (explaining that “[u]nder the percentage of fund method . . . a court shapes the counsel fee based on what it determines is a reasonable percentage of the fund recovered for those benefited by the litigation” and several factors may be considered to ensure the fee is reasonable).

92See Tinio, supra note 91, § 3.

93See Andrew Rosenfield, An Empirical Test of Class-Action Settlement, 5 J. Legal Stud. 113, 115 (1976).

94See id.

95For a discussion of the due process rights of absent class members, see generally Redish, supra note 8, at 135–75.

96See Tinio, supra note 91, § 3.

9728 U.S.C. § 1712 (2012) (referring to the substantive coupon settlement provisions).

98See Martin H. Redish, Class Actions and the Democratic Difficulty: Rethinking the Intersection of Private Litigation and Public Goals, 2003 U. Chi. Legal F. 71, 78 (“[W]hen the litigation dust settles, even in cases in which the plaintiff class prevails (either by means of judgment or settlement), often the overwhelming majority of class members never receive anything approaching meaningful compensation for the defendants’ violation of their substantive rights. Instead, they are frequently ‘awarded’ the opportunity to receive some form of discount coupon for purchase of a product or service already provided by the defendant in the normal course of business.”).

99See 28 U.S.C. § 1712.

100Redish, Julian & Zyontz, supra note 69, at 620.

101Id. at 624. The phrase derives from the French expression, cy pres comme possible, which means “as near as possible.” Id. For a description of the doctrine’s historical development in the law of trusts, see id. at 624–30.

102See id. at 630–38.

103See id. at 634–38.

104See, e.g., In re Baby Prods. Antitrust Litig., 708 F.3d 163 (3d Cir. 2013); Dennis v. Kellogg Co., 697 F.3d 858 (9th Cir. 2012); Klier v. Elf Atochem N. Am., Inc., 658 F.3d 468 (5th Cir. 2011).

105See, e.g., In re Lupron Mktg. & Sales Practices Litig., 677 F.3d 21 (1st Cir. 2012).

106See Redish, Julian & Zyontz, supra note 69, at 641–51.

107See Tinio, supra note 91, § 3.

108See discussion supra note 48.

109See discussion supra note 50.

110See supra text accompanying notes 65–66.

111See supra notes 55–56, 62–64 and accompanying text.

112See supra notes 59–61 and accompanying text.

113Redish, Julian & Zyontz, supra note 69, at 623.

114See id. at 620.

115See id. at 621–22.

116See supra notes 62–64 and accompanying text.

117Redish, Julian & Zyontz, supra note 69, at 621–22.

118See generally Martin H. Redish & Megan B. Kiernan, Avoiding Death by a Thousand Cuts: The Relitigation of Class Certification and the Realities of the Modern Class Action, 99 Iowa L. Rev. 1659, 1662 (2014) (“[C]lass attorneys should be deemed real parties in interest on the certification issue, and therefore the direct estoppel impact of a certification denial should bar subsequent certification attempts not only by the prior named plaintiffs, but also by the class attorneys.”).

119See Redish, supra note 98.

120See discussion supra at Part VI.C.

121To a certain extent, the proposals put forth in this Article focus not exclusively on the protections of the absent class members but also on the interests of the system, by assuring that faux class actions not be permitted. See discussion supra at Part VI.B.

122See supra text accompanying notes 47–50.

123See discussion supra at Part V.

124See discussion supra at Part VI.B–E.