Printer Friendly

Front-end fiduciaries: precertification duties and class conflict.


     A. The Back-End Focus
     B. Means and Ends
     A. CAFA-Created Conflict
     B. Binding Stipulations to Avoid CAFA Removal: A Case Study
     C. The Need for a Framework
     A. The Contours of a Front-End Fiduciary Duty
     B. The Framework Applied to Binding Stipulations
        1. Binding stipulations prejudice substantive legal rights
        2. Probability of success multiplied by potential judgment
        3. Ex post judicial oversight
        4. The inadequacy of opt-out

"[A] lawyer must never forget that he is the master. He is not there to do the client's bidding.... [T]he lawyer must serve the client's legal needs as the lawyer sees them, not as the client sees them." (1)


The traditional wisdom among attorneys and jurists has been that while clients decide the ends of lawsuits, their attorneys control the means of achieving those ends. Clients decide whether to settle suits or plead guilty to crimes, but their attorneys decide, sometimes contrary to clients' wishes, which legal arguments are made. Accordingly, commentators have given many means decisions less attention: which claims are pleaded in a complaint, which witnesses are called at trial, and whether an issue is raised on appeal are all decisions that are frequently, and without objection, made by an attorney. Instead, the focus has conventionally been on ends: whether the outcome of litigation was fair and in line with what was sought by the client.

The potential for conflicts of interest in class actions between attorneys and class members has prompted renewed attention to decisionmaking about litigation ends. In class actions, where most "clients" are absent, decisions about ends--in particular, whether to settle--must be made by an attorney. Much has been said of the potential for conflicts to emerge between attorneys and absent class members in these circumstances. (2) Scholars have outlined the misincentives that cause conflicts to emerge, the extent to which these conflicts undermine the class action as a device for vindicating claims, the problems that arise when settling claims, and whether, as a result, settlements accurately reflect the merit of the suit. (3) Myriad solutions have also been suggested for dealing with these conflicts--ranging from applying more scrutiny under the adequacy requirement of Rule 23(a) of the Federal Rules of Civil Procedure, to providing a more robust notice and opt-out regime, to requiring more active informed consent from unnamed class members. (4)

Taken together, this scholarship can largely be defined by its focus on settlements. Because many analyses of conflicts of interest are retrospective, making ex post judgments about the fairness of a lawsuit's outcome, almost all of the relevant commentary deals with ethics at the end of litigation, following class certification. As a result, most of the discussion about solutions for dealing with these conflicts takes place in a Rule 23 certification framework. Few commentators have recognized the possibility of conflicts of interest at the beginning of class litigation--or, as Linda Mullenix has described it, "at the front end" (5)--and those who have done so have only urged that Rule 23 certification take place sooner. (6) Less attention has been paid to precertification conflicts or to the fiduciary duties, if any, a plaintiff's attorney owes to a class prior to certification. At the same time, while some courts recognize a fiduciary duty owed by class counsel to the unnamed class members before certification, courts are not uniform on this point. (7) Even assuming that courts do recognize a precertification duty, the contours of that duty are unclear.

This Note fills two gaps in the literature about conflicts of interest in class actions. First, there has been no scholarly treatment of the scope and contours of an attorney's fiduciary duty to class members prior to class certification that is, outside the strictures of Rule 23. Precertification conflicts are far more difficult to address because no Federal Rules-based framework exists for considering such conduct. Second, this is the first academic treatment of means-based decisionmaking in class actions. Unlike postcertification inquiries into conflicts of interest concerning settlements, this inquiry is particularly complicated because there is often inadequate information about likely outcomes when certain means are employed. Accordingly, as discussed in this Note, there is considerably more gray area surrounding means-related decisionmaking. In the precertification stage, without information about how the litigation will run its course, attorneys make decisions that could either credibly be defended as in the best interest of the class, or be criticized as in breach of the attorney's fiduciary obligations to those class members.

These questions concerning the contours of an attorney's precertification fiduciary duty to class members when making strategic decisions are not merely academic. They are recurring and yet often unaddressed in a variety of circumstances in class action litigation.

The Supreme Court recently decided a case that could have significantly implicated the fiduciary duties of class counsel at the precertification stage. In Standard Fire Insurance Co. v. Knowles, the Court held that a plaintiff filing a class action complaint in state court may not avoid the federal jurisdictional reach of the Class Action Fairness Act of 2005 by stipulating, prior to certification of the class, that the class will not seek damages that exceed $5 million (the jurisdictional threshold for removal of the action to federal court). (8) The Court reached this holding by concluding that a "precertification stipulation does not bind anyone" but the named class plaintiff and therefore does not reduce the value of the putative class members' claims to an amount below the federal jurisdictional threshold. (9) In so holding, the Court did not address another issue arguably raised by the case: the extent to which this practice implicates fiduciary duties at the precertification stage. (10) An investigation into the contours of such fiduciary duties is important, though, because the same issues recur in a variety of contexts in class actions. For instance, it is becoming increasingly common for class counsel to jettison legal claims prior to class certification to make the certification process easier. Or, in some cases, attorneys have divided claims into separate time periods so as to craft suits with small enough amounts in controversy that they avoid federal jurisdiction. These devices to avoid federal jurisdiction raise sticky questions of fiduciary duties and the role of class counsel.

This Note focuses on just one of these practices--the use of binding stipulations, likely now defunct--as an example for analyzing precertification fiduciary duties. More generally, this Note aims to offer a formulation of the scope of attorneys' precertification fiduciary duties: an attorney breaches his fiduciary duty to absent class members when he makes a decision that prejudices the substantive legal rights of those class members without notice and opportunity for objection. When an action potentially prejudices or does prejudice a substantive legal right of absent class members, an attorney should have an opportunity to offer a good faith defense--that the course of conduct was undertaken in a good faith belief that it would maximize the class's recovery. That defense, in turn, can be evaluated in terms of whether it is legitimate and genuine, or pretextual.

This Note proceeds in three Parts. Part I offers a critical assessment of the existing theoretical treatment of conflicts of interest in class actions. I begin by reviewing the existing back-end focus in class conflict (11) analysis and describing the ways in which this analytic framework is ill suited for front-end conflicts. Part I also unpacks the theoretical debate about whether courts and litigants should be concerned with the means by which class actions are prosecuted or the ends that they are designed to achieve.

Part II contextualizes these theoretical treatments within the context of the Class Action Fairness Act of 2005 (CAFA). I discuss how, since CAFA's passage, plaintiffs' attorneys have developed new strategies for keeping class actions in state courts, and how some of these strategies have the potential to prejudice class members' interests. As a result, questions of conflicts of interest and fiduciary duties have become increasingly important at the front end of litigation. Part II illustrates this phenomenon by focusing on one such stratagem to avoid CAFA removal, which the Supreme Court recently disallowed: the use of binding stipulations to limit the amount in controversy to a total below CAFA's jurisdictional threshold.

Part III offers a framework for treating potential breaches of a class counsel's precertification fiduciary duties. I apply this framework to the example of binding stipulations and argue that in many cases the use of such binding stipulations (even were they permissible to avoid removal to federal court) would constitute a breach of class counsel's fiduciary duties. Lastly, utilizing recent empirical research on class member opt-out rates and applying the teachings of behavioral economics, I respond to the argument that class members' later opportunity to opt out of the class is a sufficient check.


A. The Back-End Focus

In his seminal article, Class Action Accountability." Reconciling Exit, Voice, and Loyalty in Representative Litigation, John Coffee asked the important question of how best to hold class counsel accountable to the class members whom they represent. (12) Arguing that class actions are an organizational form involving classic principal-agent conflicts, (13) Coffee suggested that the vocabulary of "exit," "voice," and "loyalty" are appropriate typologies for analyzing the alternative mechanisms by which to minimize agency costs. (14) Applied to class actions, "exit" includes an enhanced right to opt out of a class and pursue an individual action, "voice" encompasses an expanded opportunity to select counsel or participate in decisionmaking, and "loyalty" comprises duties owed by class counsel to class members. (15)

Myriad approaches have been suggested for limiting the conflicts or agency costs between class counsel and class members. The vast majority of these approaches fit within the exit/voice/loyalty framework. For instance, some suggestions include expanding "exit" opportunities for class members through increased opt-out rights. (16) Proponents of "voice" mechanisms for reducing agency costs have, on the other hand, discussed implementing voting and intervention rules to minimize informational asymmetries, (17) requiring certain mandatory disclosures to expose potential conflicts between class counsel and class members, (18) and increasing informal cooperation between courts and lawyers to resolve conflicts. (19) A slew of "loyalty"-based alternatives have also been set forth. These include enforcing clearer mandates on class counsel than those provided by existing procedural and ethical rules, (20) addressing the adequacy of representation earlier in litigation, (21) appointing additional counsel to safeguard class interests against the possibility of collusive settlement, (22) using guardians ad litem to monitor performance, (23) and applying presumptions against certain settlement configurations. (24)

While these proposals vary widely, and have been met with mixed approval and skepticism, they are similar insofar as they largely depend on "back end determinations." (25) That is, existing scholarship focuses on conflicts of interest at the end of litigation, often in the context of settlement. (26) As a result, many class action conflict solutions are ill suited for potential conflicts of interest occurring at the beginning of class litigation. Moreover, almost all of the appellate-level treatment of class conflict has been on the back end, concerned with the possibility of conflicts of interest at settlement. (27) This back-end focus by courts has in turn reinforced the back-end focus in scholarship, with countless articles offering a take on class conflict and the leading cases on the subject. (28) Yet despite the depth of that discussion, it says little about instances in which these conflicts emerge on the front end of litigation.

Front-end conflicts between class counsel and class members are distinct from these back-end conflicts in two respects. First, front-end and back-end inquiries differ from an informational standpoint. Back-end inquiries focus on the outcome of litigation--often a proposed settlement--and therefore any analysis is retrospective or ex post, with more complete information about the litigation and claims available for assessing the fairness of the outcome. While the sheer quantity of articles addressing principal-agent conflicts in class action settlements is a testament to the fact that these back-end inquiries are not easy, front-end inquiries are even more complex because analysis is often about the means by which the litigation is conducted. The inquiry is therefore prospective or ex ante and, consequently, dependent on imperfect information as to the result of certain means-based decisions.

Second, front-end and back-end conflicts also differ because of the structural circumstances in which they take place. All conflicts occurring between class counsel and class members at the back end of litigation are governed by the requirements of Rule 23. Proposed settlements are assessed pursuant to the requirements of Rule 23(e), which mandates notice and opportunity for class members to opt out of a settlement or object to its terms and also calls for a fairness inquiry into the terms of the settlement. (29) Rule 23 not only provides standards for assessing fairness--that is, "adequacy" is required of representation and "fairness" is required of a settlement--but also provides a clear opportunity for a court to scrutinize potential conflicts of interest. Such is not the case with front-end conflicts occurring before class certification. In such cases, there is no available Rule 23 inquiry into fairness. In fact, it is not clear how potential conflicts of interest or principal-agent conflicts are brought to the attention of a court before Rule 23 certification. It is equally unclear which type of actor--class members, defendants, or courts--has a duty to look for potential breaches of an attorney's fiduciary duties before class certification.

B. Means and Ends

Traditionally, under the Model Rules of Professional Conduct, a plaintiff's attorney has a duty to consult with his client about decisions concerning the ends of litigation--whether to settle a case, what sort of relief will be acceptable in such a settlement, and so forth. (30) Decisions about the means an attorney selects for reaching such outcomes are trickier. While an attorney must "reasonably consult with the client about the means by which the client's objectives are to be accomplished," (31) considerable discretion is vested in the attorney on technical, legal, and tactical matters. (32)

Class actions complicate these ethical rules. (33) In the class action context, an attorney lacks a client with whom to consult regarding the means or ends of litigation. A named plaintiff has only a nominal effect on litigation decisions; often that party was selected by class counsel and is seldom prepared to exercise oversight in the lawsuit. Moreover, hornbook law has taken a conservative approach to defining attorney-client relationships and the scope of fiduciary duties in aggregate litigation. While a named plaintiff is considered the "client" of class counsel and owed a fiduciary duty in many respects, (34) the extension of that relationship to absent class members is less clear. The Third Restatement of the Law Governing Lawyers presumes an attorney-client relationship for purposes of treating conflict-of-interest (35) and confidentiality (36) concerns, but little more.

The imperfect fit of the applicable rules of professional ethics in the context of aggregate litigation has prompted academic treatment. (37) Scholars have largely concerned themselves with two questions: Which members of a class are clients to whom a fiduciary duty is owed and to whom the ethical rules apply? And once the client is defined, on which issues must an attorney seek consultation in aggregate litigation?

As for the first question, the debate has largely turned on how to define the "client." Rule 23 is silent on the subject and does not characterize the relationship. A number of unitary and multiple client theories have been suggested within the literature. (38) One view is that the relationship between class counsel and absent class members is a constructive attorney-client relationship, where each member is a constructive client for purposes of professional ethics rules. (39) Critics object that conceiving of representation in this way results inevitably in violations of Model Rule 1.7, which, when construed strictly, prevents lawyers from undertaking class actions where the lawyer's ability to take action on behalf of a class or subclass may be adversely affected by the lawyer's responsibilities to another "client" (in this context, an absent class member). (40) An alternate prominent view is that the class should be conceived of as an entity, (41) much like a corporation or voluntary association is considered an entity client. (42) Conceiving of the class collectively as an entity, with fiduciary duties and ethical responsibilities owed to the class, and not to individual members, removes the difficulties associated with Model Rule 1.7. What might be considered concurrent conflicts of interest under the constructive-client view are mere agency problems that can be managed through Rule 23 (but do not preclude representation) under the entity approach. The counter is that class actions lack many of the characteristics of other associations recognized as legal entities for purposes of suing or being sued. Litigation classes, unlike other associations, lack consent to the organization by its members, majoritarian mechanisms for decisionmaking, prior association or commonality of interests, and homogenous preferences. (43) These differences in organizational form are especially acute in the precertification context, where the notice and implied consent mechanisms of Rule 23 are not present.

Setting aside the question of whether Model Rule 1.7 precludes representation, conceiving of the attorney-client relationship from the constructive or entity perspective has, in many cases, minimal practical importance in the later stages of litigation. Often the requirements of Rule 23 require as much as (if not more than) what would be required by professional ethics rules: all absent members have an opportunity to weigh in on the ends of litigation, such as whether a settlement is appropriate. And if a class member is unhappy with the ultimate objective of the litigation, she can opt out. These issues are far more vexing earlier in litigation, when there is no Rule 23 overlay. Who in the precertification stage are a lawyer's "clients," and to what extent must a lawyer consult with those individuals on the means selected for reaching litigation outcomes? Under the most limited view, an attorney need only consult with the named representative about the means of litigation pursuant to Model Rule 1.4. Nancy Moore advocates viewing the putative class as a prospective client in the precertification stage and later assessing whether the precertification actions were taken in the best interest of the class (as arguably contemplated by Rule 23(g)). (44) Viewing the attorney-client relationship through the lens of the constructive-client approach would require an assessment of an attorney's actions as affecting each absent class member. While the constructive view would not presumably require actual consultation with each absent class member as to the means of litigation, it would require that an attorney act in a manner consistent with his fiduciary duty to each class member.

Which view of the attorney-client relationship is adopted has real implications for how the means of class action litigation are regulated. During the precertification stage, outside of the confines of Rule 23, a narrow view of the attorney-client relationship strips professional ethics rules of much force and limits the fiduciary obligations of class counsel to absent class members at the beginning of the litigation. On the other hand, a robust view that creates a fiduciary relationship and implies ethical requirements between an attorney and all class members reduces the discretion class counsel has to select the means by which claims are litigated.

Once the "client" in aggregate litigation is defined, the question then becomes how to apply the various professional ethics rules that provide for client consultation and control over litigation. Conventional wisdom holds that diverse preferences over the ends of litigation, or "fundamental preferences," are far more troubling than differences over means, or "instrumental preferences," because fundamental preferences require separate representation or subclassing. (45) Yet while preferences about ends may require more drastic remedies, there are also more existing judicial mechanisms for managing these diverging preferences, (46) and more guidance as to how much consent and client control is required. On the other hand, the extent to which a client must be consulted with and have control over instrumental decisions is far less clear. Rule 23 does not provide guidance, and professional ethics rules, even outside the class action context, are not a model of clarity. The ends, after all, are the objectives of litigation over which class members have "ultimate authority"; means are merely the "technical, legal, and tactical matters" to be controlled by counsel. (47) As a result, little guidance is available regarding the extent to which class action clients have control over the means by which their claims are aggregated and prosecuted. (48)

Complicating all of this further, in some instances, certain instrumental preferences will necessarily undermine certain fundamental preferences. That raises a final question: when preferences as to the means of class action litigation conflict with, or are at least an obstacle to, preferences concerning the ends of that litigation, which preferences should be preferred? Framed another way, when maintaining robust due process protections for absent class members undermines efforts at achieving "global peace," (49) which course is preferable? Despite the Supreme Court's rejection of settlements in Amchem and Ortiz, arguably prioritizing means-based considerations (e.g., the protection of absent class members' due process rights) over ends-based considerations (e.g., the global peace arising from the resolution of multiple claims at once), (50) scholarship has moved in a different direction. Criticizing the Supreme Court's jurisprudence, scholars have argued in favor of relaxing means-based inquiries in order to achieve a preferable end--finality. (51) Setting aside the merits of such a proposal, it remains unclear how such principles are to be applied when ends-versus-means considerations arise at the front end of litigation. Moreover, questions abound regarding the particular instances in which due process considerations must be relaxed in order to achieve finality.

Collectively, these lurking questions regarding the scope of fiduciary duties and professional ethics rules and their application to instrumental preferences, or means-related decisions, have confounded the resolution of class conflict at the front end of litigation. This is due in part to undertheorization of precertification conflicts of interest and fiduciary obligations; the result has been that courts have applied varying standards for resolving (or in some cases not resolving) disputes. It is the aim of the remainder of this Note to identify those conflicts, investigate the levels of complexity inherent in such problems, and offer a proposed framework for resolution.


A. CAFA-Created Conflict

For years, state courts were class counsels' and clients' preferred forum, and they were easily accessible. State courts presented considerable advantages for class plaintiffs. In state courts, the standards for class certification were (and, in many cases, still are) less rigorous. (52) The Supreme Court's decisions in Amchem and Ortiz further ratcheted up the level of scrutiny applied to class certification and settlement in the federal courts. By contrast, there are considerably fewer procedural hurdles to settlement in state court, and state judges are more willing to give deference to settlements agreed to by the parties.

Concerned with these dynamics, Congress passed CAFA in 2005, significantly expanding federal jurisdiction over both class actions and mass actions. (53) The law expanded the diversity jurisdiction of the federal courts by amending the amount-in-controversy requirement, permitting the aggregation of individuals' claims in calculating that amount, and imposing a minimal rather than complete diversity requirement. (54) In doing so, CAFA eased the process by which class actions may be removed to federal court. These amendments, in turn, had a significant effect on class action litigation. The pre-CAFA complete diversity requirement had allowed plaintiffs' counsel to evade federal jurisdiction by naming additional local plaintiffs or defendants in order to defeat diversity. (55) One particularly sensational account of this practice was offered at the Senate Committee's 2002 hearing on class actions, where a witness testified that her drug store was named as a defendant in "hundreds of lawsuits" so that "the lawyers could keep the case in a place known for its lawsuit-friendly environment." (56) Requiring only minimal diversity was intended to remedy these problems.

Additionally, before CAFA, relying on Supreme Court precedent requiring that each named class plaintiff seek damages in excess of the statutory minimum for diversity jurisdiction to apply, (57) plaintiffs' attorneys would name a plaintiff with a claim below the jurisdictional threshold in order to avoid federal jurisdiction. (58) CAFA effectively removed these devices for keeping a class action in state court, thereby easing a defendant's path toward removal to federal court. No longer could a plaintiff name a nondiverse party or a nominal plaintiff to destroy diversity.

But, not surprisingly, plaintiffs' attorneys adapted. Since CAFA, they have come up with increasingly novel ways to evade removal to the federal courts. To start, despite the broad language of CAFA, the Act does contain a series of exceptions. Specifically, CAFA contains a "home state" exception for cases in which the principal defendant and two-thirds of the class members are citizens of the state in which the action is filed, and a "local controversy" exception that extends the "home state" exception to confer jurisdiction over any action in which a defendant is a citizen of the state and the injuries in question are local to the state. (59) These exceptions, in effect, provide roadmaps to plaintiffs' attorneys for keeping class actions in state court. Further, plaintiffs' attorneys have devised other procedural innovations to evade CAFA jurisdiction. For instance, a practice has emerged whereby attorneys have divided claims into separate time periods so as to limit the amount in controversy to below the jurisdictional threshold of CAFA. In Freeman v. Blue Ridge Paper Products, Inc., the plaintiffs divided their suit into five separate suits covering distinct six-month time periods, limiting the total damages for each suit to less than CAFA's jurisdictional threshold. (60) Plaintiffs' attorneys have also styled class actions as counterclaims or third-party complaints in order to avoid removal under CAFA. (61)

One stratagem to defeat removal, predating CAFA but revived since the Act's passage, is the use of binding stipulations to limit a class action's amount in controversy to a sum below the jurisdictional threshold. (After the Court's decision in Knowles, however, this practice is unlikely to be successful.) Since long before CAFA's passage, it was well established that a plaintiff is the master of his complaint. (62) Control of the complaint has traditionally meant that a plaintiff may select the forum in which a diversity suit is brought. And where the amount in controversy would otherwise allow for the removal of an action to federal court, the longstanding rule of St. Paul Mercury Indemnity Co. v. Red Cab Co. allows a plaintiff to limit the damages sought to below the jurisdictional threshold in order to avoid removal. (63) This principle makes sense to the extent a plaintiff's recovery is bound by the amount of damages requested in his complaint, as he is effectively waiving any claim to a greater amount of damages. Relying on the Red Cab language, some pre-CAFA courts allowed an individual plaintiff to limit her recovery to stay out of federal court. (64) Nevertheless, because other strategies--like fraudulent joinder (65)--existed to prevent removal to federal court, the use of binding stipulations was infrequent in the class action context prior to CAFA. (66)

After the passage of CAFA, plaintiffs' attorneys were forced to look for new innovations to avoid federal jurisdiction, and the use of Red Cab-style stipulations to prevent removal caught on. The use of binding stipulations was just one such innovation. It is the focus of this Note, not because it is the sole strategy adopted by plaintiffs' attorneys to win the removal battle, but because it is arguably the most high profile, having recently been addressed by the Supreme Court, and because it raises questions as to the means by which class counsel litigate class actions. (67)

B. Binding Stipulations to Avoid CAFA Removal: A Case Study

The stipulations themselves are simple. Typically they contain a short declaration from the named class plaintiff and her counsel, swearing not to seek damages in excess of $5 million, inclusive of interest and costs, at any time during the case whether it is removed, remanded, or otherwise. (68) The specific wordings of these stipulations are seldom litigated, and when they are, it is to little effect. (69) Rather, at the level of the federal circuit courts, litigation over the validity of these stipulations has focused on whether they are permissible under CAFA, consistent with attorneys' fiduciary obligations to classes, and binding in state court following remand. The majority of the discussion of stipulations has occurred at the district court level, focusing on whether plaintiffs will be judicially estopped from seeking damage awards in excess of $5 million in state court. Little discussion is given to the implications these stipulations have on class plaintiffs' or counsels' fiduciary duties or duties of good faith.

Five circuits--the Third, (70) Sixth, (71) Eighth, (72) Ninth, (73) and Eleventh (74)--held prior to Knowles that binding stipulations could effectively limit a class's recovery to an amount below the jurisdictional threshold so as to avoid removal. Many of these cases entirely avoid the issue of fiduciary duties. Of those cases in which the defendant has raised the issue as a ground for opposing remand, the courts have dismissed the argument as premature at the removal stage. (75) Moreover, courts have rejected fiduciary duty and "bad faith" arguments by noting a class member's right to later opt out of the class. (76)

Meanwhile, the Fifth (77) and Seventh (78) Circuits have noted in passing that a binding stipulation that waives a portion of a class's recovery violates the fiduciary duties of a named plaintiff and a class counsel to the class. As Chief Judge Frank Easterbrook has stated, a class representative "can't throw away what could be a major component of the class's recovery," and, as a result, what the named plaintiff or class counsel "is willing to accept thus does not bind the class and therefore does not ensure that the stakes fall under $5 million." (79) In addition, at least one district court in the Eight Circuit--the circuit in which the use of binding stipulations was most prevalent--questioned the validity of this practice, noting that "[p]laintiff is not merely asserting her claims: she is also asserting the claims of a class" and she has "no right to limit or compromise the recovery of the class without Court approval, particularly before she has even been approved as a representative for the class." (80)

The Supreme Court resolved this division between the circuits in Standard Fire Insurance Co. v. Knowles. (81) There, the Court held that because these stipulations are not binding upon the entire class, they cannot destroy federal jurisdiction. In a short opinion for a unanimous Court, Justice Breyer reasoned that "a proposed class action cannot legally bind members of the proposed class before the class is certified." (82) Accordingly, the Court held that Knowles, the named plaintiff in the case, "lacked the authority to concede the amount-in-controversy issue for the absent class members." (83)

The Supreme Court paid little attention to the role of fiduciary duties in the dispute. The Court briefly referenced, in considering the argument that Knowles would be an inadequate representative by capping the class's recovery, the Seventh Circuit's language in Back Doctors that a class representative has a "fiduciary duty not to 'throw away what could be a major component of the class's recovery.'" (84) Yet the Court did not elaborate any further on the role of fiduciary duties, and a majority of circuit decisions predating Knowles similarly declined to address objections by defendants relating to fiduciary duties. Only the Seventh Circuit in Back Doctors truly considered the issue. There, the court reasoned that allowing a named plaintiff or class counsel to bind the entire class to relief of less than $5 million is in tension with the fiduciary duty that almost all courts have recognized to exist between class action attorneys and absent class members. (85) That fiduciary duty requires class counsel to seek the maximum possible recovery on behalf of the class and not to prejudice the class members' ability to vindicate their rights. (86)

The responses to this argument have been numerous: First, some circuit courts pre-Knowles accepted the argument that CAFA does not change the proposition that a plaintiff is the master of her own case, and she may therefore limit claims substantively or financially to keep the case out of federal court. (87) Second, adequacy-of-representation issues are not properly assessed when determining federal jurisdiction. Rather, if a binding stipulation constitutes a breach of an attorney's fiduciary duty, a state court should deny class certification on adequacy grounds at the certification stage. (88) Third, some courts concluded that given the opt-out rights of all class members, if a binding stipulation were to prejudice an absent class member's rights, the class member could easily opt out of the class and pursue his own recovery. (89) The most novel argument in support of the use of binding stipulations advanced in the lower courts was that they can be wholly consistent with an attorney's fiduciary duties and in the best interest of the class. While binding stipulations limit the total potential payout of a class action, they can also counterintuitively increase the claims' value by causing them to be litigated in state courts. (90) In other words, a higher probability of a lower recovery could be more valuable to the class than a lower probability of a higher recovery--or as one court put it, "accepting a damage cap in return for less rigorous certification law may be a wise tactic." (91) An attorney would act in accordance with his fiduciary duties to the class when he makes a good faith judgment that the expected value of the class action (the probability of success multiplied by the amount of the potential judgment) is greater in state court.

Courts have largely ignored the arguments on each side of this issue, and as a result, doctrinal inconsistency persisted until the Supreme Court's recent decision.

C. The Need for a Framework

While binding stipulations are unlikely to be used after the decision in Knowles, the questions involved in Knowles and in lower court cases in this area illustrate more general issues relating to fiduciary duties at the front end of litigation. As already noted, when class action suits do end up in federal court, plaintiffs have adopted various maneuvers for easing the burdens of Rule 23 certification. For instance, in addition to counterclaim class actions and claim splitting, (92) it has now become a somewhat common practice to jettison causes of action so that only claims that can be easily certified remain in the case. (93) And in some cases, ends-related decisions arise at the precertification stage, (94) making the defining of precertification duties all the more important.

Binding stipulations should be viewed as just one of many such innovations by plaintiffs' attorneys to avoid removal under CAFA. At best, a whack-a-mole method for dealing with these strategies will result in the same issues recurring in different forms. At worst, an ad hoc approach without a broader conceptual framework risks inconsistencies within the law. The result could be courts approving of certain mechanisms for evading federal jurisdiction while admonishing parties for employing others.

The challenge, then, is identifying a framework to resolve loyalty problems at the front end of class action litigation.


A. The Contours of a Front-End Fiduciary Duty

In a traditional lawsuit, where an attorney files a complaint on a single client's behalf, the existence of a fiduciary duty between the attorney and the client is unquestioned. Class actions lack this sort of clarity. While there is agreement that counsel owes a fiduciary duty to a certified class, the existence of such a relationship in the precertification stage is far from clear. The majority view is that before class certification, the putative class members are not "represented" by class counsel and thus are not owed a fiduciary duty. (95) Yet a number of courts have held that, even in the absence of class certification, class counsel owes a fiduciary duty to unnamed class members in the precertification period. (96) Regardless of the position taken, these decisions are largely bereft of reasoning. More than anything, courts and commentators have relied on tradition and strained readings of professional ethics rules to reach these conclusions.

This Note argues that, contrary to conventional wisdom, a fiduciary duty between class counsel or the named class plaintiff and each individual class member ought to attach at the time a class action complaint is filed. In other words, a precertification fiduciary duty should attach at the time of filing, and it is owed not just to named plaintiffs or the class as an entity, but to each potential unnamed class member.

The notion of a "fiduciary" has roots in principles of equity and the law of trusts. (97) The word "fiduciary" is derived from "fiduciarius," and bound up in its definition is the idea of "one who is trusted." (98) In its traditional common law context, a fiduciary relationship was one in which a trustee held title to property on behalf of a beneficiary. As trustees took responsibility for the beneficiary's title and interest, the law imposed a fiduciary standard of acceptable conduct, restricting self-dealing and other self-interested conduct by the trustee. (99) Central to such a relationship are reliance on the trustee, and the trustee's de facto control and dominance over the beneficiary's asset. (100) In other words, a heightened obligation is imposed on the trustee or agent because the beneficiary or principal is uniquely vulnerable to the trustee's abuse of power. For this reason, fiduciary relationships arise where a person takes control over an aspect of another person's life or property with the understanding that the trustee will exercise control for the benefit of that person.

What do these historical roots tell us about the existence of a precertification fiduciary duty? Regular attorneys are fiduciaries. Attorneys who bring class action lawsuits are fiduciaries, too. But while oftentimes an attorney acts as an agent for his client--he is hired and the client exercises some degree of control over him--in the class action context an attorney more closely resembles a trustee. (101) More than in the traditional context, the putative class relies on its attorney to make the decisions about how the lawsuit will be prosecuted: where it will be filed, what claims will be pled, what motions will be made, when settlement will occur, and so on. Class counsel has de facto control and dominance over these litigation decisions, and the class members are uniquely vulnerable to such control. (102)

The general existence of a fiduciary duty between class counsel and individual class members is not in dispute. (103) Rather, the question is whether that duty should exist prior to Rule 23 class certification--beginning at the time a complaint asserting claims on behalf of a class is filed. Decisions denying the existence of a precertification duty draw a distinction between the status of class members before and after certification. The theory is that before certification, class members are not as vulnerable because negative rulings will not necessarily have a binding effect, (104) and they have the opportunity to opt out of the class later on if necessary. Moreover, courts could conceivably deny class counsel status to an attorney who undertakes particularly inappropriate behavior prior to certification.

But imposing different pre- and postcertification fiduciary duties is an artificial, counterintuitive distinction. Nothing changes in terms of an absent class member's reliance on an attorney before and after certification. Likewise, the level of control an attorney has over an absent class member's relevant asset--that is, his claim--remains the same pre- and postcertification. Making this type of distinction between pre- and postcertification duties is particularly anomalous because, in effect, it would mean that a fiduciary duty arises out of Rule 23, as opposed to a relationship of trust that would ordinarily be defined by state law. The bases for distinguishing between pre- and postcertification duties are not factors that change the relationship between an attorney and a class member. Rather, they are mitigating factors that either escalate or minimize the inherent vulnerabilities in the fiduciary relationship. For instance, the fact that an absent class member may not be bound by a negative opinion from a court prior to certification does not undermine the existence of a legal relationship to class counsel. Rather, it is a reason why the potential class member may be less vulnerable. Conversely, the absence of class definition, judicial supervision, and opt-out rights before certification makes the legal relationship more risky, but neither creates nor denies that relationship's existence.

Further, as this Note has already suggested and will cover in more detail shortly, at all times in a lawsuit, an attorney can make strategic decisions that will ultimately affect the outcome of class litigation and class members' recovery. This fact is equally true before and after certification. Imposing fiduciary duties only after certification shifts the burden of exercising caution from attorney to client. With postcertification relationships, the onus is on the attorney not only to comply with Rule 23 but to act with the utmost faithfulness to the class members. But without fiduciary duties prior to certification, that oversight burden is borne by absent class members who will be forced to acquire information about an attorney's conduct before certification and to police potential malfeasance. In all likelihood, given what we know about the oversight exercised by class members, that means attorneys are left unchecked before certification if they do not have some duty of loyalty. This difference makes little sense. If anything, a greater fiduciary duty should be imposed prior to certification, when class members are vulnerable to decisions that are not immediately reviewable under Rule 23 or its state counterparts. In short, the correct rule, and the rule that courts are increasingly endorsing, is that a fiduciary duty exists at the time a complaint is filed. (105)

So what, then, do these fiduciary duties entail? Fiduciary law is one of the most indefinite and ambiguous categories of legal obligation. (106) For years, courts and scholars have relied on colorful descriptions of the fiduciary relationship-most notably Justice Cardozo's description of fiduciary duties in Meinhard v. Salmon. (107) Yet while "the punctilio of an honor the most sensitive" (108) is provocative, almost no one knows what this didacticism actually means. Scholars have dissected all elements of the doctrine, and there has been considerable theoretical discussion of the obligation, (109) but case law remains contradictory and muddled. (110)

Defining the contours of a precertification duty is difficult for two reasons. First, in addition to the inherent ambiguity in defining the scope of a fiduciary duty between attorney and client, aggregate litigation adds a wrinkle of complexity. Second, while discussions of the fiduciary duties of class action attorneys are often conflated with Rule 23's adequacy requirement, there is no Federal Rules-based roadmap for defining the contours of the duty at the precertification stage. In defining precertification duties, it would, of course, be entirely unrealistic for a class member to make many litigation decisions or be consulted about every part of the case. Judge Leonard Sand of the Southern District of New York, therefore, has offered this helpful insight about the scope of the precertification duty (a formulation that this Note adopts):

   [W]e may venture a few statements about the scope of the fiduciary
   duty owed by class counsel to putative class members prior to class
   certification. In short, the scope of those duties is limited to
   protecting the substantive legal rights of putative class members
   that form the basis of the class action suit from prejudice in an
   action against the class defendant resulting from the actions of
   class counsel. Where the actions of class counsel put those rights
   at risk, class counsel must at a minimum put absent class members
   on notice and provide them with an opportunity to object. Where
   they fail to do so, class counsel exposes itself to potential
   liability for breach of its fiduciary duties. (111)

Judge Sand separately referred to this protection of substantive legal rights as guarding the "due process rights of absent class members." (112) And thus, in his framework, the fiduciary duties of class counsel prior to certification are judged in accordance with due process principles: class counsel may not prejudice the substantive legal rights of absent class members, and when they do, class counsel must, at a minimum, afford class members notice and opportunity to object, or else it is a breach of their fiduciary duty.

Thus, the preliminary inquiry into whether there has been a breach of an attorney's fiduciary duty prior to certification entails two steps. First, does the decision or action of the class action attorney "prejudice" the "substantive legal rights" of a class member? If so, then, second, has that absent class member been given appropriate notice and an adequate opportunity to object to the attorney's course of conduct? If not, then the makings of a breach of an attorney's precertification fiduciary duty to the client have been established.

But should the inquiry end there? That is, when an attorney prejudices substantive legal rights of would-be absent class members, does that constitute a per se violation of his fiduciary duty, or should the attorney be given an opportunity to mount defenses or rebut a presumption? There is a temptation to favor a per se rule for means-related decisions since it is not certain how those decisions will ultimately influence the outcomes of class litigation. But a blanket rule that any prejudicing of substantive legal rights constitutes a breach of fiduciary duties may unfairly prohibit conduct that is actually beneficial for a class. That is, there are some instances--discussed further below--in which an attorney may prejudice substantive legal rights in order to maximize class recovery. Thus, a more flexible rule is necessary--one that allows attorneys to argue that they did not breach their fiduciary duties to the class because they had a good faith basis for prejudicing some substantive rights.

The analysis should therefore play out as follows: First, a court must assess whether an action (or means-based decision) prejudices a substantive legal right of the proposed class. If that is the case, and the class is not provided some form of notice and an opportunity to object--both of which are unlikely in the precertification context--there is a presumption that the attorney has breached his fiduciary duty to the class. The attorney then ought to have the opportunity to defend the decision or course of conduct as consistent with his fiduciary duty. In doing so, the attorney must argue that the decision was based on a good faith belief that he was maximizing class recovery. Such inquiry should not delve into what the attorney actually believed, but rather what a reasonable good faith belief would have been given the circumstances. In assessing such a defense, courts must be wary of disingenuous justifications or subjective pretexts for the attorney's decisions. A defense of a good faith judgment should not overcome objective evidence that it was clear at the time the decision was made that the decision would prejudice substantive legal rights and undermine class recovery.

B. The Framework Applied to Binding Stipulations

While perhaps more straightforward than Judge Cardozo's prescription in Meinhard, this adaptation of Judge Sand's description of the contours of a precertification fiduciary duty is still an abstraction. What, at the front end of litigation, is a "substantive legal right" and in what instances is it "prejudiced"? What constitutes a good faith defense, how does it play out, and what evidence may be offered? Notably, before certification, many traditional risks to substantive legal rights do not exist. Therefore, the relevant question--when are substantive rights prejudiced?--arises when means-based decisions regarding the way the suit is prosecuted prejudice substantive legal rights. And when those rights are prejudiced, what does appropriate notice and an adequate opportunity to object look like? Does an attorney need to take extraordinary steps to secure the consent of absent class members when rights are prejudiced before certification, or is the later opportunity to opt out sufficient for the purposes of a fiduciary duty analysis? Rather than continue with abstraction, this Note answers these questions by contextualizing them within the context of the example presented in Part II--namely, the pre-Knowles use of binding stipulations and an attorney's fiduciary duties to absent class members in the precertification stage of litigation.

1. Binding stipulations prejudice substantive legal rights

To review, binding stipulations are typically documents resembling affidavits signed by a class action attorney and a named plaintiff, and attached to a class action complaint when filed, warranting that the class will at no time seek more than $4.9 million in class recovery. The only purpose of these stipulations is to limit the amount in controversy to below CAFA's $5 million jurisdictional threshold so as to prevent removal from state to federal court. Despite some defendants' contentions that these stipulations cannot actually bind absent class members, courts prior to Knowles overwhelmingly found them effective in limiting recovery. These limitations are significant. In most cases, claims are brought on behalf of nationwide classes with potential recoveries far in excess of $5 million. (113) In other words, in most cases, binding stipulations have the effect of slashing the class's aggregate potential recovery, and significantly limiting each individual class member's recovery. The Supreme Court rejected the use of these stipulations in Knowles, so their validity is now a settled issue. But as an example of a technique used by plaintiffs' attorneys, they are a useful prism though which to evaluate precertification fiduciary duties.

Based on the first level of analysis, these stipulations were a breach of an attorney's fiduciary duty to absent class members. To put it simply, a binding stipulation constitutes an attorney signing away a large percentage of a class's potential recovery in order to keep a case in state court. These stipulations were intended to be binding, meaning they cannot be undone, and constituted a full waiver of a portion of a class member's claim. If such conduct occurred outside of the class action context, the breach would be obvious. That is, if a lawyer, without the client's consent, agreed with opposing counsel to limit her client's claim by fifty percent, a clear breach of the lawyer's fiduciary duty to her client would be established. Here it is no different. Determining the value of a class member's legal claim unquestionably affects a substantive legal right, and a binding stipulation prejudices it.

But the analysis should not end there. The next question is whether there is a good faith basis for the use of binding stipulations--that is, in some cases, could attorneys maximize class recovery by limiting the total amount in controversy and keeping the case in state court?

2. Probability of success multiplied by potential judgment amount

One court, in a streak of realism, suggested that a binding stipulation could be consistent with an attorney's fiduciary duties because while the total potential recovery in state court would be capped at a lower amount, plaintiffs might have a higher probability of success in state court. (114) In other words, a binding stipulation is consistent with an attorney's fiduciary obligation when the probability of success in state court multiplied by the potential capped recovery in state court is greater than the probability of success in federal court multiplied by the uncapped potential recovery. (115)

There are at least two problems with such an argument. First, no matter how the issue is cut, an attorney is waiving a portion of a class member's claim up front based on a (hopefully educated) guess that the class will have better luck in state court. Second, making this judgment is difficult. To start, plaintiffs' attorneys lack perfect information about the value and merits of claims. The decision to file a binding stipulation occurs before filing, which means attorneys are making calculations as to their probability of success and potential recovery before having the benefit of any discovery. That discovery could increase the value of a class's claim (further undermining the argument for a binding stipulation) or decrease it. Further, even if plaintiffs' attorneys had near-perfect information, would they value a claim correctly? Personal biases, overconfidence, or unfamiliarity with the law all are reasons why an attorney might not value a claim correctly. Even if they could, class members and judges almost certainly could not. That means that in assessing an attorney's fiduciary obligation to the class, courts and class members would have to rely exclusively on the value judgments and probability assignments of interested class action attorneys.

Despite those hurdles, it is conceivable that the use of a binding stipulation--if it were still a valid tactic for avoiding federal jurisdiction--could be consistent with an attorney's fiduciary duty. Consider the following hypothetical: An attorney assesses a potential class action and values it at $6 million. Various procedural hurdles in federal court make the case difficult to certify, and thus the attorney decides that filing in state court will better facilitate certification of the class and settlement. Recognizing that the probability of certification and settlement is considerably higher in state court, the attorney stipulates to limit the amount in controversy to just below $5 million, effectively waiving just over a million dollars of the potential recovery in favor of a much higher probability of success. Such a case is an example of an instance in which the attorney likely acted consistently with his fiduciary duty to the class. While he would prejudice substantive legal rights, he would do so in order to maximize the class's ultimate recovery.

In sharp contrast with such a case are lawsuits in which the amount in controversy greatly exceeds $5 million, and attorneys use stipulations to make it easier to obtain quick certification and quick settlements of "sweetheart deals" in state court that give class counsel considerable returns (often to the detriment of the class's recovery) for very little work. Thus, courts must separate those instances where there is a reasonable good faith basis for the attorneys' conduct from those where the stipulation is merely a tool to enrich attorneys.

Given the uncertainty and ambiguity tied up in these decisions, it is tempting to adopt a wait-and-see approach. Many courts have advocated just that--that is, that state courts assess whether the stipulation is consistent with an attorney's fiduciary duties after certification, or that potential class members simply opt out after certification if they feel their interests have been prejudiced. As we will see next, such ex post solutions are ultimately unhelpful.

3. Ex post judicial oversight

A common response to the position that binding stipulations are a breach of an attorney's fiduciary duty, and accordingly should not be allowed, is that a state court could later decertify the class if it found that the stipulation constituted a breach of the attorney's fiduciary duty. As a preliminary matter, there is no guarantee that a state court will exercise the same amount of circumspection that a federal district court would. In some states, certification standards are notoriously lax. (116) Even if state certification guidelines track the federal rule in form and practice, courts and commentators have recognized that ex post judicial regulation of class counsel is largely ineffective at reducing agency costs, as judges lack both the necessary resources (117) and incentives in many cases. (118)

In addition, in many cases, courts will be certifying classes for settlement purposes and will have insufficient information to fully exercise oversight ex post. (119) A lack of information is compounded by hindsight bias: because judges struggle (like any person would) with assessing the ex ante likelihood of outcomes, they cannot determine the complete reasonableness of certification and settlement ex post. (120)

4. The inadequacy of opt-out

Many courts have relied on an absent class member's later opt-out right as a basis for approving the use of binding stipulations. Many of the problems with ex post judicial oversight apply to the reliance on an opt-out right. An opt-out, after all, is really just an ex post form of oversight and rejection by class members. But this ex post oversight is unlikely to be effective. For instance, the lax state certification standards are likely to affect opt-outs as well. Recent empirical work on class action opt-outs reveal that the vast majority of class members do not opt out of class actions. (121) This "silence" is not a function of approval by class members, but a result of ignorance about the terms of a proposed settlement and an insufficient amount of time to object. (122) Some scholarship suggests that notices to class members are inadequate, (123) and in this case it is questionable whether they would apprise class members of the existence of a binding stipulation. Deborah Rhode made this point about the unintelligibility of formal notices with an anecdote about an antitrust case seeking damages from major drug companies on behalf of antibiotics purchasers:

   Class members received notices stating that unless they indicated a
   desire to opt out of the litigation, they would be bound by its
   result. Of the responses received, many if not most evinced some
   degree of misunderstanding. Some reflected a level of confusion
   that graphically illustrates the limitations of formal notice:

      Dear Sir:

        I received your pamphlet on drugs, which I think will be of
      great value to me in the future.

        Due to circumstances beyond my control I will not be able to
      attend this class at the time prescribed on your letter due to
      the fact that my working hours are from 7:00 until 4:30.

      Dear Sir:

        Our son is in the Navy, stationed in the Caribbean some place.
      Please let us know exactly what kind of drugs he is accused of

        From a mother who will help if properly informed.

           A worried mother,
           Jane Doe (124)

Quite simply, would class members understand the effect the binding stipulation had on their claim? Even when class members are informed, objecting or opting out may not appear to be cost beneficial: the marginal benefit of a greater recovery would be outweighed by the costs of litigating an individual action. (125) This is especially true in a case involving binding stipulations; class members will not opt out because even though they could recover more through trial or settlement, (126) that marginal increase in recovery cannot be justified by the cost of nonaggregate litigation. Lastly, absent class members are likely to suffer from hindsight bias; they will be unable to later assess whether, absent the binding stipulation, they really would have been better off.

In sum, in many cases the use of binding stipulations constitutes a breach of an attorney's fiduciary duties to a class in the precertification stage. By stipulating, attorneys waive or prejudice substantive legal rights of class members (that is, the portion of the value of their claims), and no adequate voice or exit option exists to escape such a fiduciary breach. But the analysis must not end there. Courts must be careful to distinguish between those cases in which an attorney's tactic (the stipulation in our example) is designed truly for the purposes of maximizing class recovery and those cases where it is a mere tool for enriching plaintiffs' attorneys. While there should be a presumption that tactics like binding stipulations are a breach of an attorney's fiduciary duties, since they prejudice substantive legal rights to recovery, a more thorough analysis must take into account the complete range of circumstances that existed when the decision was made.


This Note seeks to fill a void in the legal literature on attorney ethics, fiduciary obligations, and attorney-client conflicts in class actions. While a tremendous amount has been written about these issues, almost all of the commentary is backward looking and back-end focused. Scholars have been concerned with the adequacy of representation when cases come to a close. As a result, they have lost sight of the role and importance of fiduciary duties on the front end of litigation. This Note is the first investigation into such duties and argues that plaintiffs' attorneys owe fiduciary duties to all class members, named or absent, at the time a class action complaint is filed, irrespective of when or if the class is certified. The simple test for reasoning through whether means-based decisions violate an attorney's fiduciary duty, borrowed from Judge Sand, should be whether the decision prejudices a class member's substantive legal rights, and if it does, whether the attorney had a reasonable good faith basis for believing the decision maximized class recovery. Such an inquiry is particularly well suited for front-end conflicts because it is ex ante. Courts need only decide whether an attorney's act has the potential to prejudice substantive legal rights and whether the attorney made a reasonable, good faith judgment, not whether the action was actually prejudicial or what the attorney's actual state of mind was.

The elegance of this approach is evident from this Note's analysis of the recently resolved use of binding stipulations to keep class actions out of federal court. Close examination of this practice reveals it in many cases to be a violation of attorneys' fundamental fiduciary duties to absent class members. This Note also traces the rare instances in which this practice may be consistent with an attorney's fiduciary duties. More generally, this examination of the use of binding stipulations illustrates the need for greater awareness of class conflict at the front end of litigation, before class certification.

(1.) Clement F. Haynsworth, Jr., Professionalism in Lawyering, 27 S.C.L. REV. 627, 628 (1976).

(2.) See, e.g., Samuel Issacharoff, Class Action Conflicts, 30 U.C. DAVIS L. REV. 805, 812-13, 820-22, 832 (1997) (discussing mandatory class actions, settlement class actions, and class actions made up of preexisting client inventories); Geoffrey P. Miller, Conflicts of Interest in Class Action Litigation: An Inquiry into the Appropriate Standard, 2003 U. CHI. LEGAL F. 581,598-628 (identifying conflicts between absent class members and other absent class members, between absent class members and class counsel, between absent class members and the representative plaintiff, among multiple class counsels, and among multiple representative plaintiffs); Deborah L. Rhode, Class Conflicts in Class Actions, 34 STAY. L. REV. 1183, 1186-88, 1204-15 (1982) (discussing conflicts between attorneys and class members in class actions seeking structural reforms in public and private institutions); Richard G. Stuhan & Sean P. Costello, Robbing Peter to Pay Paul." The Conflict of Interest Problem in Sibling Class Actions, 21 GEO. J. LEGAL ETHICS 1195, 1199-1202 (2008) (discussing the conflict of interest problems associated with sibling class actions); Gregg H. Curry, Comment, Conflicts of Interest Problems for Lawyers Representing a Class in a Class Action Lawsuit, 24 J. LEGAL PROF. 397, 399-407 (2000) (discussing various potential conflicts with attorneys and the classes they represent); Sylvia R. Lazos, Note, Abuse in Plaintiff Class Action Settlements: The Need for a Guardian During Pretrial Settlement Negotiations, 84 MICH. L. REV. 308, 311-19 (1985) (discussing abuses in the presettlement stage); Genine C. Swanzey, Note, Using Class Actions to Litigate Mass Torts." Is There Justice for the Individual?, 11 GEO. J. LEGAL ETHICS 421, 423-32 (1998) (reviewing conflicts of interest between present and future claimants, among individual clients, and between lawyers and class members).

(3.) See, e.g., Janet Cooper Alexander, Do the Merits Matter? A Study of Settlements in Securities Class Actions, 43 STAN. L. REV. 497, 536 (1991) (arguing that inherent in class actions is "a significant possibility that litigation decisions will be made in accordance with the lawyer's economic interests rather than those of the class"); Bruce L. Hay, Asymmetric Rewards. Why Class Actions (May) Settle for Too Little, 48 HASTINGS L.J. 479, 497-501 (1997) (discussing the effect of conflicts on settlement values); Susan P. Koniak & George M. Cohen, Under Cloak of Settlement, 82 VA. L. REV. 1051, 1057-1102 (1996) (discussing examples of class action settlements in which the conduct engaged in by class counsel was particularly egregious, yet was nevertheless blessed through a court's approval of the settlement); Miller, supra note 2, at 598-628 (outlining the various conflicts that may emerge).

(4.) See infra notes 16-24 and accompanying text.

(5.) Linda S. Mullenix, Taking Adequacy Seriously: The Inadequate Assessment of Adequacy in Litigation and Settlement Classes, 57 VAND. L. REV. 1687, 1733 (2004).

(6.) See id. at 1693 (arguing that "the best way to make settlement classes attack-proof is to ensure adequate representation at the outset of class litigation").

(7.) See infra notes 95-96 and accompanying text.

(8.) Standard Fire Ins. Co. v. Knowles, No. 11-1450, 2013 WL 1104735, at * 4-6 (U.S. Mar. 19, 2013).

(9.) Id. at *4.

(10.) The Court did make reference to the possibility that a "court might find that [a named plaintiff] is an inadequate representative due to the artificial cap he purports to impose on the class' recovery." Id. In doing so, the Court cited an opinion from the Seventh Circuit, Back Doctors Ltd. v. Metropolitan Property & Casualty Insurance Co., in which the circuit court noted that a class representative's fiduciary duty encompasses not "throw[ing] away what could be a major component of the class's recovery." Knowles, 2013 WL 1104735, at *5 (quoting Back Doctors Ltd. v. Metro. Prop. & Cas. Ins. Co., 637 F.3d 827, 830-31 (7th Cir. 2011) (internal quotation marks omitted)). But the Court never discussed the issue further, nor did it make mention of the role the class's attorney plays in all of this.

(11.) While for some "class conflict" may be a term of art, throughout this Note I use the phrase to describe conflicts between attorneys and potential or actual class members. Such conflicts might manifest as a classic conflict-of-interest problem between attorney and client or as one of the potential fiduciary duty issues described here.

(12.) John C. Coffee, Jr., Class Action Accountability: Reconciling Exit, Voice, and Loyalty in Representative Litigation, 100 COLUM. L. REV. 370, 370 (2000).

(13.) Id. at 375.

(14.) Id. at 376-78.

(15.) Id. at 377-78.

(16.) See Mark C. Weber, A Consent-Based Approach to Class Action Settlement. Improving Amchem Products, Inc. v. Windsor, 59 OHIO ST. L.J. 1155, 1197-1200 (1998).

(17.) See, e.g., Lewis A. Komhauser, Control of Conflicts of Interest in Class-Action Suits, 41 PUB. CHOICE 145, 172 (1983); Patrick Woolley, Rethinking the Adequacy of Adequate Representation, 75 TEX. L. REV. 571,604 (1997).

(18.) See Rhode, supra note 2, at 1197-1202.

(19.) See Mary Kay Kane, Of Carrots and Sticks: Evaluating the Role of the Class Action Lawyer, 66 TEX. L. REV. 385,405 (1987).

(20.) See Rhode, supra note 2, at 1258.

(21.) See Mullenix, supra note 5, at 1733-38.

(22.) See David L. Shapiro, Class Actions." The Class as Party and Client, 73 NOTRE DAME L. REV. 913,959 n.132 (1998).

(23.) See Susan P. Koniak, Feasting While the Widow Weeps: Georgine v. Amchem Products, Inc., 80 CORNELL L. REV. 1045, 1093 & n.219, 1122 (1995).

(24.) See Issacharoff, supra note 2, at 832-33.

(25.) Linda Mullenix uses the term "back end determinations" to refer to inquiries into adequacy of representation that take place at the end of class action litigation, often during the certification of a settlement. See Mullenix, supra note 5, at 1713, 1715-16. She makes the normative argument that back-end determinations are unlikely to provide an adequate check on conflicts of interest and reduce agency costs because "earlier deficiencies in proving adequacy are likely to be carried over into the settlement approval process without further probing of the adequacy requirement" and "without a front end adequacy determination, none of the parties at the back end have any compelling interest in challenging adequacy." Id. at 1716. This Note borrows her terms for the descriptive purpose of differentiating between temporal periods in class action litigation.

(26.) See, e.g., Hay, supra note 3, at 479; Bruce Hay & David Rosenberg, "Sweetheart" and "Blackmail" Settlements in Class Actions: Reality and Remedy, 75 NOTRE DAME L. REV. 1377, 1377-79 (2000); Koniak & Cohen, supra note 3, at 1053-56.

(27.) The Supreme Court has twice dealt with conflicts in the class action context. See Ortiz v. Fibreboard Corp., 527 U.S. 815, 821,848-49, 852-53 (1999); Amchem Prods., Inc. v. Windsor, 521 U.S. 591,597, 619-20, 626-27 (1997). In Amchem and Ortiz, the Court rejected settlements of asbestos-related class actions because of conflicts of interest. Both decisions provide barebones guidance as to the contours of Rule 23(a)'s adequacy of representation requirement for settlement classes.

(28.) See sources cited supra note 2. Of course, all of this naturally tends to assume that the purpose of class litigation is to provide adequate compensation to the class. This is by no means a given, and a prominent view in the literature is that deterrence or punishment of wrongdoers is the more important policy goal. See Patrick A. Luff, Bad Bargains: The Mistake of Allowing Cost-Benefit Analyses in Class Action Certification Decisions, 41 U. MEM. L. REV. 65, 80-84 (2010) (discussing generally the deterrence-related arguments). If deterrence is the primary goal, then completely maximizing class recovery becomes less important. Then again, to the extent a breach of an attorney's fiduciary duty reduces the value of a judgment or settlement, it may also undermine the class litigation's deterrent effect.

(29.) FED. R. CIV. P. 23(e)(2). Moreover, the Supreme Court in Amchem clarified that the certification requirements of Rule 23(a)-(b), such as adequacy of representation, still apply to settlements. See 521 U.S. at 620-21.

(30.) MODEL RULES OF PROF'L CONDUCT R. 1.2(a) (2012) (requiring that a lawyer "abide by a client's decisions concerning the objectives of representation").

(31.) Id. R. 1.4(a)(2).

(32.) See id. R. 1.2 cmt. 2.

(33.) See, e.g., Nancy J. Moore, "Who Should Regulate Class Action Lawyers?," 2003 U. ILL. L. REV. 1477, 1478 ("There has been considerable difficulty applying existing rules of conduct to these situations, partly because of confusion regarding the relationship among class counsel, the named class representatives and absent members of the class."); Brian J. Waid, Ethical Problems of the Class Action Practitioner: Continued Neglect by the Drafters of the Proposed Model Rules of Professional Conduct, 27 LOY. L. REV. 1047, 1048 (1981).

(34.) See 1 RESTATEMENT (THIRD) OF THE LAW GOVERNING LAWYERS [section] 14 cmt. f (2000).

(35.) See id. [section] 125 cmt. f.

(36.) See id. [section] 70 cmt. c.

(37.) See, e.g., Koniak, supra note 23, at 1121 ("[T]he ethics rules cannot be mechanically applied to class actions.").

(38.) See 5 ALBA CONTE & HERBERT B. NEWBERG, NEWBERG ON CLASS ACTIONS [section] 15.3 (4th ed. 2002); Developments in the Law--Conflicts of Interest in the Legal Profession, 94 HARV. L. REV. 1244, 1449-54 (1981).

(39.) 5 CONTE & NEWBERG, supra note 38, [section] 15.3.

(40.) See Waid, supra note 33, at 1071-72 (quoting MODEL RULES OF PROF'L CONDUCT (proposed final draft 1981)).

(41.) See Moore, supra note 33, at 1482-89; Shapiro, supra note 22, at 923-34. But see Mandujano v. Basic Vegetable Prods., Inc., 541 F.2d 832, 834-35 (9th Cir. 1976) (stating that "a proposed class ... is not a legal entity," and that the "class attorney continues to have responsibilities to each individual member of the class").

(42.) See MODEL RULES OF PROF'L CONDUCT R. 1.13(a) (2012) ("A lawyer employed or retained by an organization represents the organization acting through its duly authorized constituents.").

(43.) See Coffee, supra note 12, at 381-84.

(44.) See Moore, supra note 33, at 1486.

(45.) See Elizabeth Chamblee Burch, Optimal Lead Plaintiffs, 64 VAND. L. REV. 1109, 1173-74 (2011) (citing SCOTT E. PAGE, THE DIFFERENCE: HOW THE POWER OF DIVERSITY CREATES BETTER GROUPS, FIRMS, SCHOOLS, AND SOCIETIES 1 1-12 (2007)).

(46.) Rule 23, for example, provides mechanisms for notice, objections to settlements, opt outs, and subclasses. See FED. R. CIV. P. 23.

(47.) MODEL RULES OF PROF'L CONDUCT R. 1.2 cmts. 1-2.

(48.) As Lynn Baker and Charles Silver have noted:

   Chief among the decisions to be made regarding "technical and legal
   tactics" and the "means" by which the objectives of the
   representation are to be achieved are the decisions where and when
   to file the client's lawsuit. Given the goal of maximizing the
   client's gross recovery, the attorney can be expected to consider a
   number of factors in deciding where among the available options to
   file the lawsuit....

Lynn A. Baker & Charles Silver, Fiduciaries and Fees." Preliminary Thoughts, 79 FORDHAM L. REV. 1833, 1855 (2011).

(49.) Richard A. Nagareda, Autonomy, Peace, and Put Options in the Mass Tort Class Action, 115 HARV. L. REV. 747, 751 & n.8 (2002) (exploring the idea of "global peace," that is, the peace that comes from resolving multiple claims).

(50.) See Ortiz v. Fibreboard Corp., 527 U.S. 815, 821, 848-49, 856-59, 864-65 (1999); Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 597, 628-29 (1997).

(51.) Most notably, Richard Nagareda has argued that ideals of due process in litigation that emphasize individual autonomy and conflict-free representation as prerequisites for effective representation conflict with real-world dynamics and effective administration of settlements in mass torts. See RICHARD A. NAGAREDA, MASS TORTS IN A WORLD OF SETTLEMENT 159-60 (2007).

(52.) Compare Wal-Mart Stores, Inc. v. Dukes, 131 S. Ct. 2541, 2551 (2011) (noting that Rule 23 requires actual evidence and "rigorous analysis" by the district court), with, e.g., Gen. Motors Corp. v. Bryant, 285 S.W.3d 634, 641 (Ark. 2008) (affirming that the Supreme Court of Arkansas had "previously rejected any requirement of a rigorous-analysis inquiry" in class certification proceedings).

(53.) Class Action Fairness Act of 2005, Pub. L. No. 109-2, 119 Stat. 4 (codified as amended in scattered sections of 28 U.S.C.). The Senate began considering CAFA in the 105th Congress when the Senate Judiciary Subcommittee on Administrative Oversight and the Courts held hearings in October 1997. S. REP. NO. 109-14, at 1-2 (2005), reprinted in 2005 U.S.C.C.A.N. 3, 3. The Subcommittee heard testimony from a number of experts on the unfairness of class action settlements, attorneys' fees, and abuses in state courts. See id. at 2, reprinted in 2005 U.S.C.C.A.N. at 3-4. The law was introduced on September 28, 1998, and was reintroduced five more times, culminating with the introduction of the "Class Action Fairness Act of 2005" on January 25, 2005. Id. at 2-3, reprinted in 2005 U.S.C.C.A.N. at 4-5 (discussing CAFA's legislative history). The law was ultimately passed by both houses and signed into law by President George W. Bush on February 18, 2005. See Remarks on Signing the Class Action Fairness Act of 2005, 41 WEEKLY COMP. PRES. DOC. 265 (Feb. 18, 2005), available at

(54.) See Class Action Fairness Act [section] 4(a), 119 Stat. at 9-10 (codified as amended at 28 U.S.C. [section] 1332 (2011)). Compare 28 U.S.C. [section] 1332(d) (requiring an aggregate amount in controversy of $5 million for class actions), with id. [section] 1332(a) (requiring an amount in controversy of $75,000 for civil actions generally). CAFA dispensed with the rule that all plaintiffs must be diverse from all defendants. See Class Action Fairness Act [section] 4(a), 119 Stat. at 9. Compare 28 U.S.C. [section] 1332(d)(2)(A) (prescribing that diversity jurisdiction in class actions exists when, inter alia, "any member of a class of plaintiffs is a citizen of a State different from any defendant"), with id. [section] 1332(a)(1) (requiring that the parties in civil actions in general be "citizens of different States" for federal diversity jurisdiction to exist), and Strawbridge v. Curtiss, 7 U.S. (3 Cranch) 267, 267 (1806) (interpreting the Judiciary Act of 1789 to require complete diversity for federal diversity jurisdiction).

(55.) This was a particularly adept move because, under the pre-CAFA regime, the class's counsel could voluntarily dismiss the claim against the local party after one year had elapsed, but removal was still prohibited (despite the presence of diversity) under the nonwaivable one-year time limitation of 28 U.S.C. [section] 1446(b)-(c). See Russaw v. Voyager Life Ins. Co., 921 F. Supp. 723, 725 (M.D. Ala. 1996) (declining to exempt fraudulent joinder from the one-year limit on removal).

(56.) S. REP. No. 109-14, at 10, reprinted in 2005 U.S.C.C.A.N. at 11 (internal quotation marks omitted).

(57.) See Zahn v. Int'l Paper Co., 414 U.S. 291,300-01 (1973) (requiring each member of a class to meet the federal jurisdictional amount in controversy for the court to have jurisdiction). The rule set forth in Zahn was ultimately done away with in Exxon Mobil Corp. v. Allapattah Services, Inc., which construed 28 U.S.C. [section] 1367 (enacted after Zahn) to allow federal courts to exercise supplemental jurisdiction over joined claims that do not individually meet the amount-in-controversy requirement of 28 U.S.C. [section] 1332, provided that at least one claim meets the amount-in-controversy requirement. 545 U.S. 546, 549 (2005).

(58.) See Kline v. Avis Rent A Car Sys., Inc., 66 F. Supp. 2d 1237, 1240 (S.D. Ala. 1999) ("[P]laintiff is still master of her own claim.... [T]he court will not consider ...

waived claims in determining whether the amount in controversy exceeds $75,000." (internal quotation marks omitted)). Some courts, however, rejected this approach. See Torreblanca de Aguilar v. Boeing Co., 47 F.3d 1404, 1411 (5th Cir. 1995) ("[W]e hold that if a defendant can show that the amount in controversy actually exceeds the jurisdictional amount, the plaintiff must be able to show that, as a matter of law, it is certain that he will not be able to recover more than the damages for which he has prayed in the state court complaint. Such a rule is necessary to avoid the sort of manipulation that has occurred in the instant case."); In re Norplant Contraceptive Prods. Liab. Litig., 918 F. Supp. 178, 180 (E.D. Tex. 1996) (reaching the same conclusion).

(59.) See Kevin M. Clermont & Theodore Eisenberg, CAFA Judicata: A Tale of Waste and Politics, 156U. PA. L. REV. 1553, 1557-58 (2008).

(60.) 551 F.3d 405, 406 (6th Cir. 2008); see also Proffitt v. Abbott Labs., No. 2:08-CV151, 2008 WL 4401367, at *2 (E.D. Tenn. Sept. 23, 2008) (noting that plaintiffs had divided their suit into eleven class actions, each for a one-year period, for the specific purpose of evading diversity jurisdiction).

(61.) In Palisades Collections LLC v. Shorts, for example, a collection agency brought an action on behalf of a telephone company to collect a nominal sum in unpaid charges. The defendant counterclaimed with a class action claim against the collection agency. Because the collection agency was a counterdefendant, not a defendant, the court held that there was no right of removal under CAFA. 552 F.3d 327, 328-29 (4th Cir. 2008); see also Ford Motor Credit Co. v. Jones, No. 1:07 CV 728, 2007 WL 2236618, at *3 (N.D. Ohio July 31, 2007) (confirming that a third-party defendant's class action cannot be removed under CAFA).

(62.) See The Fair v. Kohler Die & Specialty Co., 228 U.S. 22, 25 (1913) ("Of course the party who brings a suit is master to decide what law he will rely upon...."); 14B CHARLES ALAN WRIGHT ET AL., FEDERAL PRACTICE AND PROCEDURE [section] 3702 (3d ed. 1998) ("[T]he plaintiff is the master of his or her claim....").

(63.) 303 U.S. 283, 294 (1938) (stating in dicta that if a plaintiff "does not desire to try his case in the federal court he may resort to the expedient of suing for less than the jurisdictional amount, and though he would be justly entitled to more, the defendant cannot remove"); see also 1A JAMES WM. MOORE ET AL., MOORE'S FEDERAL PRACTICE [paragraph] 0.158 (2d ed. 1996) (citing Red Cab as an example of preventing removal "by resorting to the expedient of suing for less than the jurisdictional minimum"); WRIGHT ET AL., supra note 62, [section] 3702 ("[P]laintiff is the master of his or her own claim; if the plaintiff chooses to ask for less than the jurisdictional amount, only the sum actually demanded is in controversy.").

(64.) See In re Shell Oil Co., 970 F.2d 355, 356 (7th Cir. 1992) (per curiam) ("Litigants who want to prevent removal must file a binding stipulation or affidavit with their complaints."); McCoy v. Erie Ins. Co., 147 F. Supp. 2d 481, 484 (S.D.W. Va. 2001) ("Many courts have seized on the ... Red Cab dicta as a bright-line rule compelling remand where a specific sum less than the jurisdictional amount is stated."); Russell D. Jessee, Pleading to Stay in State Court: Forum Control, Federal Removal Jurisdiction, and the Amount in Controversy Requirement, 56 WASH. & LEE L. REV. 651, 651 (1999) (stating that generally, if plaintiffs prefer a state court forum and are willing to seek damages below the federal jurisdictional amount, they may prevent a geographically diverse defendant from removing to federal court).

(65.) "Fraudulent joinder" generally refers to a practice by plaintiffs of attempting to defeat diversity jurisdiction, and thereby avoid removal, by joining a local or nondiverse defendant who has no real connection to the case. See E. Farish Percy, Making a Federal Case of It: Removing Civil Cases to Federal Court Based on Fraudulent Joinder, 91 IOWA L. REV. 189, 191 (2005).

(66.) In fact, there was a pre-CAFA split of authority as to the legitimacy of the use of binding stipulations in class actions. Compare Tovar v. Target Corp., No. Civ. A. SA04CA0557XR, 2004 WL 2283536, at *3 (W.D. Tex. Oct. 7, 2004), Clark v. Pfizer Inc., No. Civ. A. 04-3354, 2004 WL 1970138, at *3 (E.D. Pa. Sept. 7, 2004), and Spann v. Style Crest Prods., Inc., 171 F. Supp. 2d 605, 611 (D.S.C. 2001) (approving of the use of stipulations), with Feldman v. N.Y. Life Ins. Co., No. Civ. A. 97-4684, 1998 WL 94800, at *5 (E.D. Pa. Mar. 4, 1998) (citing Torreblanca de Aguilar v. Boeing Co., 47 F.3d 1404, 1409-10 (5th Cir. 1995)), Adkins v. Gibson, 906 F. Supp. 345, 348 (S.D.W. Va. 1995), abrogated on other grounds by McCoy, 147 F. Supp. 2d 481, and Dunn v. Pepsi-Cola Metro. Bottling Co., 850 F. Supp. 853,855 (N.D. Cal. 1994) (rejecting the use of stipulations).

(67.) Stephen Shapiro, studying removal under CAFA, noted several years before Knowles that "[o]ne question that courts have not yet clearly answered is whether or not plaintiffs could limit the amount in controversy by stipulating that they would not accept any more than five million dollars in total recovery for the class." Stephen J. Shapiro, Applying the Jurisdictional Provisions of the Class Action Fairness Act of 2005." In Search of a Sensible Judicial Approach, 59 BAYLOR L. REV. 77, 116 (2007).

(68.) For example, the stipulation in Tomlinson v. Skechers U.S.A., Inc. reads as follows:

   I do hereby swear and affirm that I do not now, and will not at any
   time during this case, whether it be removed, remanded, or
   otherwise, seek damages for myself or any other individual class
   member in excess of $75,000 (inclusive of costs and attorneys'
   fees) or seek damages for the class as alleged in the complaint to
   which this stipulation is attached in excess of $5,000,000 in the
   aggregate (inclusive of costs and attorneys' fees).

Civil No. 11-5042, 2011 U.S. Dist. LEXIS 142862, at *4 (W.D. Ark. May 25,2011).

(69.) See, e.g., Smith v. Am. Bankers Ins. Co. of Fla., No. 2:11-cv-02113, 2011 WL 6090275, at *5 (W.D. Ark. Dec. 7, 2011) (rejecting defendant's argument that the specific wording of the stipulation was not adequately binding on class counsel); Knowles v. Standard Fire Ins. Co., No. 4:11-CV-04044, 2011 WL 6013024, at *4 (W.D. Ark. Dec. 2, 2011) (rejecting defendant's argument that the language "will not ... seek" did not effectively disclaim recovery in excess of the jurisdictional threshold since the stipulation did not indicate that the plaintiff would "refuse[] to accept" an award in excess of that amount (first alteration in original) (internal quotation marks omitted)), leave to appeal denied, No. 11-8030, 2012 WL 3828891 (8th Cir. Jan. 4, 2012), reversed, No. 11-1450, 2013 WL 1104735 (U.S. Mar. 19, 2013); McClendon v. Chubb Corp., No. 2:11-CV-02034, 2011 WL 3555649, at *7 (W.D. Ark. Aug. 11, 2011) (stating that a damage disclaimer need only "serve the same function" as a binding stipulation to be effective).

(70.) See Morgan v. Gay, 471 F.3d 469, 471, 477 (3d Cir. 2006) (permitting the use of a disclaimer in the named plaintiffs complaint to limit the class's recovery to no more than $5 million in order to avoid diversity jurisdiction).

(71.) See Freeman v. Blue Ridge Paper Prods., Inc., 551 F.3d 405, 409 (6th Cir. 2008) ("[P]laintiffs can avoid removal under CAFA by limiting the damages they seek to amounts less than the CAFA thresholds."); Smith v. Nationwide Prop. & Cas. Ins. Co., 505 F.3d 401, 407 (6th Cir. 2007) ("[A] plaintiff may sue for less than the amount [he] may be entitled to if [he] wishes to avoid federal jurisdiction and remain in state court." (second and third alterations in original)); see also McClendon v. Challenge Fin. Investors Corp., No. 1:11 CV 1597, 2011 WL 5361069, at *5 (N.D. Ohio Nov. 3, 2011) ("[A] class representative can avoid CAFA jurisdiction by expressly limiting the class members' damages.").

(72.) See Rolwing v. Nestle Holdings, Inc., 666 F.3d 1069, 1073-74 (8th Cir. 2012) ("[W]e conclude that Missouri's well-established judicial estoppel doctrine makes these stipulations binding, ... and remand based on CAFA's amount-in-controversy requirement was appropriate."); Bell v. Hershey Co., 557 F.3d 953, 958 (8th Cir. 2009) ("In order to ensure that any attempt to remove would have been unsuccessful, [plaintiff] could have included a binding stipulation with his petition stating that he would not seek damages greater than the jurisdictional minimum upon remand.... "). A number of district courts within the Eighth Circuit also approved of this practice. See Smith v. Am. Bankers Ins. Co. of Fla., 2011 WL 6090275, at *5; Knowles, 2011 WL 6013024, at *4; Thompson v. Apple, Inc., No. 3:11-CV-03009-PKH, 2011 WL 2671312, at *2 (W.D. Ark. July 8, 201 l); Tomlinson, 2011 U.S. Dist. LEXIS 142862, at *6; Murphy v. Reebok Int'l, Ltd., No. 4:11-cv-214-DPM, 2011 WL 1559234, at *2 (E.D. Ark. Apr. 22, 2011); Tuberville v. New Balance Athletic Shoe, Inc., No. 1:11-cv-01016, 2011 WL 1527716, at *2-3 (W.D. Ark. Apr. 21, 2011); E-Shops, Corp. v. U.S. Bank Nat'l Ass'n, Civil No. 10-4822 (DSD/JJK), 2011 WL 1324574, at *2 n.3 (D. Minn. Apr. 7, 2011).

(73.) See Lowdermilk v. U.S. Bank Nat'l Ass'n, 479 F.3d 994, 999 n.5 (9th Cir. 2007) ("A plaintiff may ... stipulate to damages in order to avoid federal jurisdiction...."); see also Guglielmino v. McKee Foods Corp., 506 F.3d 696, 705 n.6 (9th Cir. 2007) (O'Scannlain, J., concurring) (noting that a "party might file a binding stipulation, prior to removal, that it will not seek more in recovery than the jurisdictional threshold" in order to prevent removal).

(74.) See Thomas v. Countrywide Home Loans, No. 3:11-CV-399-WKW, 2012 WL 1190895, at *5 (M.D. Ala. Feb. 17, 2012) ("[O]utside the context of CAFA, it is settled in [the Eleventh C]ircuit that a plaintiff's binding stipulation limiting damages is effective for determining the jurisdictional amount in controversy.").

(75.) See, e.g., Murphy, 2011 WL 1559234, at *3 (reasoning that Reebok's attack on the stipulations went more to Murphy's adequacy as a class representative and counsel's adequacy as class counsel than to good faith, and therefore concluding that the issue could be addressed after remand).

(76.) See, e.g., Morgan v. Gay, 471 F.3d 469, 476 n.7 (3d Cir. 2006) ("The availability of opting out by unnamed class members assuages any concerns that [the named plaintiffs] damage limitation harms these other class members."); Murphy, 2011 WL 1559234, at *3.

(77.) See Ditcharo v. United Parcel Serv., Inc., 376 F. App'x 432, 437 (5th Cir. 2010) ("[Binding stipulations] do not provide [a named plaintiff] with the authority to deny other members of the[] putative class action the right to seek an award greater than [the jurisdictional threshold].").

(78.) See Back Doctors Ltd. v. Metro. Prop. & Cas. Ins. Co., 637 F.3d 827, 830-31 (7th Cir. 2011); see also Fiore v. First Am. Title Ins. Co., No. 05-CV-474-DRH, 2005 WL 3434074, at *3 (S.D. Ill. Dec. 13, 2005) ("Plaintiff cannot in good faith place a $5,000,000 limitation on the recovery of the putative class....").

(79.) Back Doctors Ltd., 637 F.3d at 830-31.

(80.) Bass v. Carmax Auto Superstores, Inc., No. 07-0883-CV-W-ODS, 2008 WL 441962, at *2 (W.D. Mo. Feb. 14, 2008).

(81.) No. 11-1450, 2013 WL 1104735 (U.S. Mar. 19, 2013).

(82.) Id. at *3.

(83.) Id. at *4.

(84.) Id. (quoting Back Doctors Ltd., 637 F.3d at 830-31).

(85.) See Back Doctors Ltd., 637 F.3d at 830-31.

(86.) Id.

(87.) See Morgan v. Gay, 471 F.3d 469, 474 (3d Cir. 2006); Brill v. Countrywide Home Loans, Inc., 427 F.3d 446, 449 (7th Cir. 2005); cf. Rolwing v. Nestle Holdings, Inc., 666 F.3d 1069, 1072-73 (8th Cir. 2012) (noting generally that courts have upheld the use of binding stipulations). In Knowles the Court clarified that a plaintiff's right to be the master of her own complaint does not extend so far as to allow her to legally bind all plaintiffs.

(88.) McClendon v. Chubb Corp., No. 2:11-CV-02034, 2011 WL 3555649, at *5 (W.D. Ark. Aug. 11,2011) ("Any arguments Defendants may have as to the named Plaintiffs' adequacy as class representatives may be addressed after remand."); Murphy v. Reebok Int'l, Ltd., No. 4:11-cv-214-DPM, 2011 WL 1559234, at *3 (E.D. Ark. Apr. 22, 2011) ("[Defendant's] attack on the stipulations goes more to [plaintiff's] adequacy as a class representative and counsel's adequacy as class counsel than to good faith. These issues can be addressed after remand."). The Court in Knowles cited the fact that "a court might find that Knowles is an inadequate representative due to the artificial cap he purports to impose on the class' recovery" in concluding that stipulations are in effect contingent and cannot bind for purposes of assessing an amount in controversy. Knowles, No. 11-1450, 2013 WL 1104735, at *4.

(89.) See Smith v. Am. Bankers Ins. Co. of Fla., No. 2:11-cv-02113, 2011 WL 6090275, at *8 (W.D. Ark. Dec. 7, 2011) ("[P]utative class members may simply opt out of the class and pursue their own remedies if they feel that the limitations placed on the class by Plaintiff are too restrictive."); Knowles v. Standard Fire Ins. Co., No. 4:11-CV-04044, 2011 WL 6013024, at *6 (W.D. Ark. Dec. 2, 2011) (same), leave to appeal denied, No. 11-8030, 2012 WL 3828891 (8th Cir. Jan. 4, 2012), rev'd, No. 11-1450, 2013 WL 1104735; McClendon, 2011 WL 3555649, at *5 ("[C]lass members who do not agree with the way Plaintiffs have structured their claims are free to opt out of this action and bring their own suit structured in the manner they see fit."); Tuberville v. New Balance Athletic Shoe, Inc., No. 1:11-cv-01016, 2011 WL 1527716, at *4 (W.D. Ark. Apr. 21, 2011) ("[P]utative class members could simply opt out of the class and pursue their own remedies or join a different ongoing class action if they feel that the limitations placed on the class by the Plaintiff are too restrictive.").

(90.) See Murphy, 2011 WL 1559234, at *2 (citing Kenneth S. Gould, A Dynamic Development Under the Arkansas Rules of Civil Procedure." Arkansas's Favorable Approach to Class Actions, ARK. LAW., Fall 2010, at 20).

(91.) Id.

(92.) See supra notes 60-61 and accompanying text.

(93.) See Edward F. Sherman, "Abandoned Claims" in Class Actions: Implications for Preclusion and Adequacy of Counsel, 79 GEO. WASH. L. REV. 483, 483 (2011).

(94.) The possibility for precertification fiduciary duty breaches relating to decisions about ends arise when settlements are entered into prior to class certification. Cf. Mullenix, supra note 5, at 1718 (discussing adequacy issues in settlement classes).

(95.) See 2 RESTATEMENT (THIRD) OF THE LAW GOVERNING LAWYERS [section] 99 cmt. 1 (2000) ("[P]rior to certification, only those class members with whom the lawyer maintains a personal client-lawyer relationship are clients."); see also Garrett v. Metro. Life Ins. Co., No. 95 CIV. 2406 (PKL), 1996 WL 325725, at *6 (S.D.N.Y. June 12, 1996) ("[B]efore class certification, the putative class members are not 'represented' by the class counsel...."); Babbitt v. Albertson's, Inc., No. C-92-1883 SBA (PJH), 1993 WL 128089, at *4 (N.D. Cal. Jan. 28, 1993) ("[T]he putative class members in the instant case are not represented by class counsel for the purpose of application of the disciplinary rules.").

(96.) See, e.g., In re Bluetooth Headset Prods. Liab. Litig., 654 F.3d 935, 946 (9th Cir. 2011) ("Prior to formal class certification, there is an even greater potential for a breach of fiduciary duty owed the class during settlement."); In re Gen. Motors Corp. Pick-Up Truck Fuel Tank Prods. Liab. Litig., 55 F.3d 768, 801 (3d Cir. 1995) ("Beyond their ethical obligations to their clients, class attorneys, purporting to represent a class, also owe the entire class a fiduciary duty once the class complaint is filed."); Schick v. Berg, No. 03 Civ. 5513 (LBS), 2004 WL 856298, at *6 (S.D.N.Y. Apr. 20, 2004) (finding that "pre-certification class counsel owe a fiduciary duty not to prejudice the interests that putative class members have in their class action litigation" because "class counsel acquires certain limited abilities to prejudice the substantive legal interests of putative class members even prior to class certification"), aff'd, 430 F.3d 112 (2d Cir. 2005).


(98.) See id.

(99.) See id. at 2.

(100.) See United States v. Chestman, 947 F.2d 551, 568 (2d Cir. 1991).

(101.) See Parker v. Anderson, 667 F.2d 1204, 1211 (5th Cir. Unit A 1982) ("The duty owed to the client sharply distinguishes litigation on behalf of one or more individuals and litigation on behalf of a class.").

(102.) See Culver v. City of Milwaukee, 277 F.3d 908, 910 (7th Cir. 2002) ("The class action is an awkward device, requiring careful judicial supervision, because the fate of the class members is to a considerable extent in the hands of a single plaintiff ... whom the other members of the class may not know and who may not be able or willing to be an adequate fiduciary of their interests. Often the class representative has a merely nominal stake ..., and the real plaintiff in interest is then the lawyer for the class....").

(103.) See, e.g., id. at 913 (collecting cases that support the proposition that class counsel's breach of fiduciary duty to the class renders him an inadequate representative).

(104.) See Smith v. Bayer Corp., 131 S. Ct. 2368, 2379 (2011) ("[A]n unnamed member of a certified class may be considered a 'party' for the [particular] purpos[e] of appealing an adverse judgment. But ... no one ... was willing to advance the novel and surely erroneous argument that a nonnamed class member is a party to the class-action litigation before the class is certified." (second and third alterations in original) (citation and some internal quotation marks omitted)).

(105.) See supra note 96 and accompanying text.

(106.) See D. Gordon Smith, The Critical Resource Theory of Fiduciary Duty, 55 VAND. L. REV. 1399, 1400 (2002).

(107.) 164 N.E. 545, 546 (N.Y. 1928) ("A trustee is held to something stricter than the morals of the market place. Not honesty alone, but the punctilio of an honor the most sensitive, is then the standard of behavior.").

(108.) Id.

(109.) See, e.g., MATTHEW CONAGLEN, FIDUCIARY LOYALTY: PROTECTING THE DUE PERFORMANCE OF NON-FIDUCIARY DUTIES 106-39 (2011); TAMAR FRANKEL, FIDUCIARY LAW 101-83 (2011); LEONARD I. ROTMAN, FIDUCIARY LAW 53-150 (2005); Gregory S. Alexander, A Cognitive Theory of Fiduciary Relationships, 85 CORNELL L. REV. 767, 774-78 (2000); Rob Atkinson, Obedience as the Foundation of Fiduciary Duty, 34 J. CORP. L. 43, 47-82 (2008); Henry N. Butler & Larry E. Ribstein, Opting Out of Fiduciary Duties: A Response to the Anti-Contractarians, 65 WASH. L. REV. 1, 28-32 (1990); Robert H. Sitkoff, The Economic Structures of Fiduciary Law, 91 B.U.L. REV. 1039, 1040-45 (2011).

(110.) See Andrew Trask, The Ten Most Interesting Class Action Articles of 2011, CLASS ACTION COUNTERMEASURES (Dec. 29, 2011), ("Various courts have recognized that named plaintiffs, counsel, and even judges have fiduciary duties to absent class members. But what are the precise contours of those duties?").

(111.) Schick v. Berg, No. 03 Civ. 5513(LBS), 2004 WL 856298, at *6 (S.D.N.Y. Apr. 20, 2004) (emphasis added), aff'd, 430 F.3d 112 (2d Cir. 2005).

(112.) Id. at *5. Due process considerations are nothing new in class actions. See, e.g., Richard A. Nagareda, Administering Adequacy in Class Representation, 82 TEX. L. REV. 287, 288 (2003). Since the Supreme Court's decision in Phillips Petroleum Co. v. Shutts, 472 U.S. 797 (1985) (holding that due process requires that absent class members be notified of class litigation and their right to opt out before they can be bound by a judgment), courts have been concerned with plaintiffs' due process rights in class actions. And even before then, Deborah Rhode commented that "Rule 23's mandate of adequate representation is of constitutional dimension. In essence, this requirement embodies a fundamental tenet of due process: that judicial procedure fairly protect 'the interest of absent parties who are to be bound by it.'" Rhode, supra note 2, at 1192. This Note diverges from this position because it does not argue that a precertification duty creates a due process right, or that an attorney's duty prior to class certification perfectly tracks principles of due process.

(113.) See, e.g., Rolwing v. Nestle Holdings, Inc., 666 F.3d 1069, 1071 (8th Cir. 2012) (concerning a shareholder suit, where, under plaintiffs' theory, Nestle "would be exposed to damages well in excess of the $5 million threshold"); Thompson v. Apple, Inc., No. 3:11CV-03009-PKH, 2011 WL 2671312, at *1-2 (W.D. Ark. July 8, 2011) (concerning a complaint alleging that all iPhones, which range in price from $99 to $599, are "worthless"); Murphy v. Reebok Int'l, Ltd., No. 4:11-cv-214-DPM, 2011 WL 1559234, at *2 (E.D. Ark. Apr. 22, 2011) (concerning a complaint over Reebok's muscle-toning shoes, for which the company had grossed up until then more than $100 million in sales nationally); Tuberville v. New Balance Athletic Shoe, Inc., No. 1 : 11-cv-01016, 2011 WL 1527716, at *5 (W.D. Ark. Apr. 21, 2011) (noting that defendants had argued that the amount in controversy was $75 million).

(114.) See Murphy, 2011 WL 1559234, at *2 (citing Gould, supra note 90, at 20).

(115.) You might express this calculus as follows: it would be proper for class counsel to file a binding stipulation when (probability of success in state court) x (stipulated limit) > (probability of success in federal court) x (total potential recovery).

(116.) See, e.g., Gould, supra note 90, at 20 ("The ... prerequisites [of the Arkansas rule governing class action certification] are generally easily met."). Proponents of tort reform have long described certain jurisdictions--such as Madison County, Illinois--as "judicial hellholes" or "magnet jurisdictions" that are particularly plaintiff friendly. See AM. TORT REFORM FOUND., JUDICIAL HELLHOLES 2004, at 14-18, 28 (2004), available at

(117.) Frank H. Easterbrook, What's So Special About Judges?, 61 U. COLO. L. REV. 773, 778-79 (1990) (reciting the resource limitations affecting judicial decisionmaking).

(118.) See In re Activision Sec. Litig., 723 F. Supp. 1373, 1374 (N.D. Cal. 1989) ("[Y]he court is abandoned by the adversary system [when certifying a class for settlement purposes].... Rarely do the settling defendants ... offer any counterpoint; rarely do members of the class come forward with any response or opposition to the fees sought. There are no amici curiae who volunteer their advice.").

(119.) See In re Oracle Sec. Litig., 131 F.R.D. 688, 692 n.6 (N.D. Cal.) ("The critical factors in evaluating a settlement are the timing of settlement opportunities and amounts left 'on the table.' A court will almost never have reliable information on these factors."), modified, 132 F.R.D. 538 (N.D. Cal. 1990).

(120.) See Jeffrey J. Rachlinski, A Positive Psychological Theory of Judging in Hindsight, 65 U. CHI. L. REV. 571,597, 625 (1998) (explaining the operation of hindsight bias in assessing negligence).

(121.) See Theodore Eisenberg & Geoffrey Miller, The Role of Opt-Outs and Objectors in Class Action Litigation: Theoretical and Empirical Issues, 57 VAND. L. REV. 1529, 1532 (2004) ("Opt-outs from class participation and objections to class action resolutions are rare: on average, less than 1 percent of class members opt-out and about 1 percent of class members object to class-wide settlements.").

(122.) See Christopher R. Leslie, The Significance of Silence: Collective Action Problems and Class Action Settlements, 59 FLA. L. REV. 71, 90-91 (2007).

(123.) See Shannon R. Wheatman & Terri R. LeClercq, Majority of Class Action Publication Notices Fail to Satisfy Rule 23 Requirements, 30 REV. LITIG. 53, 54 (2010).

(124.) Rhode, supra note 2, at 1235.


(126.) While technically a binding stipulation would not control a settlement, which is the way most class actions are disposed of, it changes the settlement dynamics because any defendant knows the maximum the class could recover is $5 million.

Nick Landsman-Roos, J.D. Candidate, Stanford Law School, 2013. My thanks to Nora Engstrom, Lily Katz, Anuja Yhatte, Vaughn Walker, Matthew Woleske, and Charles Yablon for helpful comments and edits on previous drafts. All errors are mine.
COPYRIGHT 2013 Stanford Law School
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2013 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Landsman-Roos, Nick
Publication:Stanford Law Review
Date:Apr 1, 2013
Previous Article:Marks, morals, and markets.
Next Article:Federalism and the state civil jury rights.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters