Academic journal article Notre Dame Law Review

The Malleability of Collective Litigation

Academic journal article Notre Dame Law Review

The Malleability of Collective Litigation

Article excerpt

In Wal-Mart v. Dukes, (1) Wal-Mart avoided class action because employment decisions were made by local supervisors. However, it was Wal-Mart who chose to delegate discretion; by doing so, it made class litigation less likely. Wal-Mart's choice of business administration, then, substantially reduces its expected liability. This is but one example of a broader, overlooked phenomenon. Mass defendants can control, before the occurrence of damages, the scope of future collective litigation. Collective litigation procedures are malleable, sensitive to the defendant's pre-damages choice of actions. This Article develops and substantiates this insight.

This Article elaborates on two manifestations of this phenomenon. First, defendants can avoid class actions by "individualizing" the prospective class, injecting individual differences that preclude class treatment. Second, defendants can selectively contract with future victims, buying out the stronger, leaving only weak victims with a claimable right, and reducing the prospective class's capacity to litigate. Against this backdrop, this Article proposes an array of mechanisms to strengthen collective litigation procedures, including shifting the burden to defendants to justify the business action that prevented collective litigation, and taxing defendants for making the plaintiffs' case weaker.


     A. Mass Defendants' Inherent Litigation Advantages
        1. Unequal Economies of Scale
        2. Selective Settlements--"Cherry-Picking" Plaintiffs.
     B. Defendants' Litigation Advantages as a Social Problem
     C. The Solution: Collective Litigation
     A. The Crucial Class Certification Standards
        1. Overview
        2. Individual versus Common Questions
     B. Inducing Factual Differences--"Individualizing" the Class
        1. Individualizing Written Contracts
        2. Creating Choice of Law Differences
        3. Varying Oral Communication
        4. Heterogeneous Plaintiffs and Products
        5. Decentralizing Action
     A. Ex ante Divide-and-Conquer
     B. Examples
        1. Nuisances
        2. Products Liability: Disclaimers and Standard-Form
     A. Litigatory Damages
     B. Judicial Discretion
        1. Certifying Individualized Classes
        2. Suspicion toward Waivers and the Contemporary
     C. Class-Wide Solutions
        1. Individualizing Course of Action
        2. Selective Pre-Damages Contracts


During the 2011 term the Supreme Court ruled on the largest civil rights class action suit in U.S. history. (2) In "one of the most expansive class actions ever," (3) hundreds of thousands of women claimed that Wal-Mart--the world's largest private employer (4)--discriminated against them on pay and promotions in "literally millions of employment decisions." (5) However, the main issue before the Justices was not substantive, but procedural. Courts do not automatically authorize class litigation; they first have to certify a lawsuit as a class action. The Wal-Mart certification debate centered on the following question: whether these scores of women "have enough in common to join together in a single lawsuit." (6) As Wal-Mart conferred pay and promotion discretion on its local managers, the plaintiffs' claims might be too individualized to be pursued collectively. While Wal-Mart argued that plaintiffs "do not have enough in common to warrant class-action treatment," the plaintiffs, naturally, stressed the centralized, company-wide policy behind pay and promotion decisions. (7) These are the rules of the game.

This Article moves beyond the Wal-Mart case, which was decided against the plaintiffs, (8) to examine the rules of the game more closely. As Wal-Mart illustrates, the level of individual vis-a-vis common questions is a crucial factor in the decision to authorize class litigation. However, courts and scholars have overlooked that this essential factor--the commonality of the class--is often under the defendant's control. Defendants can, so to speak, "individualize" the prospective class. Wal-Mart chose to delegate discretion to local supervisors; by doing so, it made class litigation less likely. When class treatment is denied, plaintiffs have to pursue the far less effective individual litigation. Wal-Mart's choice of business administration, then, substantially reduces its expected liability regardless of its actual fault. This is but one example of a broader, and largely undiscussed, phenomenon. In virtue of their position, mass defendants can control, before any damages occur, the scope of prospective collective litigation.

The legal definition and structure of collective litigation procedures, therefore, are not exogenous, given facts of life; rather, they are manipulable and sensitive to the defendant's pre-damages choice of actions. I refer to this phenomenon as the ex ante malleability of collective litigation. This Article develops and substantiates this insight, and discusses its legal implications. In doing so, it elaborates on two ways in which mass defendants can frustrate, in advance, collective litigation. The first, a direct extension of the Wal-Mart example, stems from doctrinal gaps that enable defendants to avoid class actions; the second results from the capacity of prospective defendants to selectively contract with future victims.

The perspective taken in this Article is in contrast with existing views, which tend to take certification standards and, more broadly, the very existence of collective litigation, as independent of the defendant's behavior. More generally, this Article attempts to shift attention to what defendants can do to avoid litigation ex ante, before the occurrence of damages. Marc Galanter famously pointed to mass defendants' capacity to "come out ahead," using their post-damages litigation advantages. (9) This Article discusses how mass defendants can come out ahead ex ante as well, even before the occurrence of damages. In fact, although the literature has focused on the post-damages setting, taking pre-damages prophylactic measures to reduce the odds of successful litigation may well be more effective, from defendants' perspective, than resisting collective litigation after damages occur. (10)

The notion that firms conduct their businesses with an eye toward collective litigation does not mean that the desire to block litigation is the only motivation for their behavior. Wal-Mart--although in general notorious for its centralized decision making (11)--may well have had independent commercial reasons, other than preventing class actions, to decentralize pay and promotion decisions. Nevertheless, collective litigation considerations presumably factor into mass defendants' calculus. There is no reason to think that, anticipating liability, defendants do not attempt to avoid costly collective litigation, at least to some extent. Defense lawyers, for example, advise their clients to change their course of action to reduce the risk of collective litigation. (12) Contractual waivers of class litigation, which are inserted to standard-form contracts to reduce liability, proliferate after the Court authorized such waivers. (13) And after Wal-Mart the incentives for firms to decentralize are likewise larger. (14)

This ex ante perspective to collective litigation procedures, the focus of this Article, entails various legal implications. The capacity of defendants to manipulate, in advance, collective litigation procedures is socially undesirable. Collective litigation procedures (or the threat thereof) are aimed at flattening the litigation inequalities between mass defendants and individual plaintiffs, promoting deterrence and fairness; in their absence defendants can often pay less than the harm they inflicted on the group of plaintiffs. Defendants who frustrate collective litigation procedures create, in essence, litigatory damages--the plaintiffs' case is worth less, and sometimes much less, than it should be. In order to rectify this externality, this Article offers a menu of responses, ranging from attention to concrete attempts to avoid collective litigation, to a greater inclination to overcome individual differences among the would-be plaintiffs.

This Article proceeds as follows. Part I presents the conventional view regarding the role of collective litigation. Where one defendant faces individual plaintiffs, the defendant has inherent litigation advantages--it enjoys better economies of scale and can settle selectively with plaintiffs. These litigation advantages mean that the defendant pays less than the harm it inflicted on the plaintiffs. Collective litigation is a procedural tool that largely overcomes these difficulties. This is the common perception, viewing collective procedures as a vehicle to effectively vindicate rights that are otherwise not worth pursuing.

Against this backdrop, this Article further shows that in many situations the defendant is in the position to control, before the occurrence of damages, the creation of successful collective litigation--i.e., the very procedures that are aimed at preventing the defendant from exploiting its litigation advantages. Part II discusses doctrine-based malleability--how defendants can manipulate the commonality requirements to avoid class certification. Courts--Wal-Mart is but one example--are hesitant to authorize class litigation where it raises too many individual questions. Defendants, however, can individualize the class, injecting factual differences among the prospective plaintiffs to avoid future certification. Decentralizing discretion is one way to do so. Other examples include inducing modifications in written contracts, creating artificial choice-of-law differences among future plaintiffs, and orally communicating with the would-be plaintiffs. The literature has given scant attention to the capacity of defendants to shield themselves from class litigation by manipulating certification standards.

The discussion on the malleability of the commonality requirements leads to a deeper form of malleability. Part III shows how, regardless of the law of class actions, mass tort defendants can affect the structure and formation of collective litigation. Briefly, mass defendants can identify and cherry-pick strong victims, who are going to be the moving force behind future litigation. Once the defendant locates the strong would-be plaintiffs, it is better off settling with them, as these settlements make the remaining collective of plaintiffs weaker--a "divide-and-conquer" strategy. While the literature has given attention to this problem in the context of selective settlements, it has overlooked that the defendant can take a similar approach before the occurrence of damages. Conceptually, there is no difference between settling with trailblazing plaintiffs and buying out strong would-be plaintiffs. Both adversely affect the remaining victims and reduce the defendant's overall liability. One example of pre-damages selective contracts is buying out the residents in the vicinity of a nuisance, as these neighbors are those who would have the greatest incentives to vigorously litigate against the defendant. Excluding them leaves the remaining victims worse off and reduces the defendant's overall liability. In a similar way, the defendant can detect strong would-be plaintiffs through standard-form contracts, encouraging them to sign liability waivers. This Article, then, makes a novel link between liability waivers and collective litigation.

The normative implications follow. If defendants can avoid liability through pre-damages manipulations, one should look for ways to rectify this externality and improve existing collectivization procedures. A general, comprehensive solution is taxing defendants for the harm they created, i.e., for making the plaintiffs' case weaker. As this solution seems unlikely in practice, Part IV further delineates other directions. One set of responses is judicial case-by-case attention to ex ante malleability. In the context of class certification, courts can be more liberal toward certification where defendants attempted to circumvent certification standards. Courts can do so, for example, by shifting the burden to the defendant to justify its "individualizing" course of action. In the context of selective buy-outs, judges can be more suspicious toward liability waivers. When this perspective is taken, the results are contradictory to the existing law of disclaimers. While currently judges are more inclined to honor liability waivers between the defendant and sophisticated and informed consumers, the ex ante malleability approach is more suspicious exactly toward such agreements, as they are the most likely to harm remaining victims. A second set of responses involves broader, class-wide solutions. In the context of doctrine-based malleability, one direction may be a shift to class-wide determinations of liability, which obviate defendants' desire to bypass certification by individualizing the class. In the context of ex ante buy-outs, possible class-wide reactions might be prohibiting selective waivers, or assigning collective rights to a third party.


Where one defendant harms multiple plaintiffs, the defendant enjoys inherent litigation advantages over individual plaintiffs. This means that the value of the plaintiffs' claims is lower than it should be--victims are not fully compensated for their harms, and the defendant is not obliged to pay the full costs for its wrongdoing. Litigation inequalities between plaintiffs and defendants are, therefore, a social problem. Procedures of aggregate litigation are considered to be a tool to overcome these disparities. This Part presents this conventional story. It surveys the inherent adjudicative advantages mass defendants have, and how procedures of collective litigation can overcome these inequalities.

A. Mass Defendants' Inherent Litigation Advantages

This section discusses the litigation advantages defendants have when they face multiple plaintiffs, litigating them individually. The first is unequal economies of scale; the second is the ability to selectively settle with plaintiffs.

1. Unequal Economies of Scale

One fundamental advantage defendants have stems from the fact that compared to each individual plaintiff, mass injurers have more to gain (and lose) from litigating each case. The reason is that many questions that have to be litigated in different cases are correlated; hence, much of the work the defendant invests in litigating one case is valuable for other cases. Examples include investment in legal research, scientific survey, and expert witnesses. The defendant enjoys "economies of scale by investing once-and-for-all in the common questions and spreading the cost of that investment across all claims." (15) As the defendant has more at stake, it systematically invests more in litigating each case. (16) Hence, there is "a much greater chance that the defendant ... will prevail on the common questions." (17) It is true that under certain circumstances defendants may prefer collectivization to litigating separately, in order to achieve "global peace." (18) This may happen where individual plaintiffs can extract more than the value of their claims. (19) However, normally the problem of unequal economies of scale creates a fundamental bias in favor of mass injurers who litigate individually. (20) Indeed, mass defendants usually struggle to avoid class litigation.

The problem of uneven incentives to litigate is more severe under certain circumstances. These situations include mass injuries in which the correlation across cases is higher (more of the legal work is common), and where the class is more fragmented (individuals have smaller cases compared to the defendant). In fact, in some cases the result of this bias might be zero liability: individual harm that falls short of a certain threshold, possibly in the range of thousands of dollars, might well be too low to pursue individually. (21) In other cases, where individual claims are sufficiently large to pursue individually, unequal economies of scale reduce the ultimate value of the plaintiffs' claims, making the defendant's liability lower.

2. Selective Settlements--"Cherry-Picking" Plaintiffs

Another way in which mass defendants can avail themselves of their position is by striking selective, strategic settlements. Plaintiffs vary: some have better evidence, others suffer higher damages. Those with the better evidence and higher damages are more likely to initiate a lawsuit. Upon litigating, these "strong types"--whose cases have a higher monetary value--are likely to invest more effort in pursuing their claims. Plaintiffs with weak evidence and low damages, "weak types," are less likely to sue; and if they do, they are likely to put less effort into litigation. Defendants can use this variance to settle selectively--to use "divide-and-conquer" strategies. (22) In particular, defendants should find it worthwhile to settle first with the strongest plaintiffs, i.e., those with the greatest incentives to litigate.

Because individual cases are correlated, each plaintiff that litigates individually affects the remaining ones. (23) Strong plaintiffs who litigate individually confer a benefit on subsequent plaintiffs; the stronger the plaintiff, the greater the value to the group. This benefit can be, for example, a favorable holding for the first plaintiff which has a precedential value for fellow plaintiffs. (24) In other examples the value the first plaintiffs create is their attorneys' work-product, which can be used by later plaintiffs. (25) In yet other cases, it is the mere presence of early claimants that enables others to "name and blame" the defendant. (26)

Be it through precedential value, legal work-product, or mere information regarding the tortfeasor's identity, plaintiffs affect each other. Settlements, compared to trials, reduce the value that litigants add to the remaining plaintiffs. Hence, the defendant is better off settling first with the strongest; these plaintiffs, in case trial occurs, are likely to confer on other plaintiffs the largest benefits. On the other hand, when weak, losing plaintiffs produce negative common value, the defendant might be better off taking them to trial. (27) In short, as Karl Llewellyn observed, defendants' lawyers use the "strategy of presenting cases in favorable series, settling the unfavorable cases." (28)

Settling first with strong plaintiffs puts the remainder in a worse position. In some cases, the elimination of one strong plaintiff can make litigation infeasible for the remaining, weak plaintiffs. (29) In other cases, litigation is still worth pursuing. However, the absence of the strongest makes the claims of the remaining plaintiffs less worthy, as now it is harder for them to achieve successful legal results. Hence, excluding the stronger from the pool of plaintiffs reduces the expected compensation that the remaining plaintiffs can get, and, as before, makes the defendant's liability smaller. The more important the strong plaintiffs are to the remaining ones, the defendant will find this divide-and-conquer strategy more useful. (30)

In addition to lower liability, another outcome of strategic settlements is transfer of wealth from weak to strong litigants. (31) Typically, strong plaintiffs can extract some of the gain the defendant makes from settling with them first. (32) In a sense, the defendant and the strong plaintiffs are better off at the expense of weaker claimants. (33)

Strategic settlements are an actual problem. Defendants do seek to settle as a means to resist potential mass torts and products liability suits. (34) Confidential settlements with first plaintiffs are a particularly effective tool to avoid the spillover of valuable information to subsequent litigants and reduce the odds of future successful lawsuits. (35) Likewise, when litigation is underway, defendants can attempt to distinguish between strong and weak cases, preferring to settle the former and litigate the latter. (36) Similarly, strategic settlements can be used to push toward a favorable precedent, hence making repeat players "come out ahead." (37)

B. Defendants' Litigation Advantages as a Social Problem

Mass defendants, then, can use their better position vis-a-vis individual plaintiffs. They do so through unequal economies of scale and selective settlements. These litigation advantages reduce defendants' expected liability. The normative implications are clear. First, victims are not fully compensated for their damages. This, however, might not be a substantial problem to worry about. Many perceive compensating victims as a secondary goal of litigation. (38) Individual harms may be low in value, and, even where they are not, plaintiffs can often insure themselves against the vicissitudes of life. However, litigation does more than fulfill individual goals--it serves a social benefit by making the wrongdoers fully pay for the harm they inflicted. (39) Hence, regardless of inadequate compensation, flawed litigation is a social problem: defendants are not optimally deterred from taking a harmful course of action.

While there might be an over-deterrence argument--i.e., litigation makes defendants pay more than they should and hence ineffective litigation is a blessing--this claim does not seem to be true across the board. Whether collective litigation creates over-deterrence is eventually an empirical question, which is presumably context-specific. To the extent the over-deterrence problem exists, its scope has contracted since the recent decisions heightening pleading standards. (40) Furthermore, where abusive litigation is a problem, courts should directly address it, (41) for example, by a more aggressive use of sanctions on frivolous suits. (42)

In principle, then, litigation inequalities are a social problem. One should look for ways to restore equality between defendants and plaintiffs. Misalignment between the legal standard and the defendant's actual liability "should ring a warning bell that the law is probably inefficient and should be modified." (43) This is all the more correct when litigation is the only effective means of deterring wrongdoing. (44)

C. The Solution: Collective Litigation

Collective litigation--in the form of class action, multidistrict litigation (consolidating pretrial proceedings and discovery), or informal aggregation by plaintiffs' lawyers--overcomes most of these pitfalls. (45)

In the context of unequal economies of scale, collectivization raises the stakes for the plaintiffs' side, making it as eager to collectively invest in the case as the defendant. Put differently, collectivization levels the litigation playing field. Class actions are a particularly effective collectivization tool. (46) "By aggregating hundreds, thousands, or even millions of claims, the class action can make small claims viable and empower claimants...." (47)

In the context of selective settlements, collective litigation makes it harder, and less valuable, for defendants to cherry-pick strong plaintiffs to settle. Class actions can empower plaintiffs and hence remedy the problem of biased precedents, mitigating the "haves come out ahead" concerns. (48) Formal collectivization mechanisms can solve the difficulties that are created when strong plaintiffs individually contract with the defendant. (49) Likewise, plaintiffs' attorneys can informally aggregate and share information, (50) thereby mitigating attempts to pick strong plaintiffs and/or conceal settlements.

Collective litigation procedures in general, and class actions in particular, then, serve a social goal as they enable plaintiffs to vindicate their rights; in their absence, plaintiffs would recover much less, if anything. Consistent with this logic, a denial of class certification is often described as "the 'death knell' of the litigation." (51) Likewise, it is not surprising that defendants, by and large, fiercely resist the use of collective litigation, (52) even at the legislative level. (53)

Indeed, collective litigation mechanisms are widely considered to be a procedural vehicle to overcome the problems created by the litigation advantages that mass defendants enjoy. "Class actions are a conventional device for flattening [the adjudicative] inequities [between defendants and plaintiffs]." (54) Likewise, the American Law Institute asserts that "[a]ggregation is a means by which courts may promote justice under law more fully." (55) This conventional view implies that collective litigation mechanisms--enabling victims to overcome the advantages defendants have and vindicate their rights--are independent of the parties' behavior. Indeed, these procedures are often depicted as nothing more than tools that are plugged onto "class members' preexisting bundle of rights." (56) They are a vehicle, a form of delegation of legislative power, and thus are bound by substantive rights. (57)

This view, however, overlooks the fact that these collective litigation procedures--which are supposed to suppress litigation disparities--are plagued by inequalities that are similar to the ones they are aimed at eliminating. This Article demonstrates that defendants are often in the position to regulate, before the occurrence of damages, the scope of the procedural vehicle itself. The defendant, in essence, sets the scene: the odds and form of successful aggregation are typically under its domain. The procedural tool, then, is not independent of the parties' choice of action. This simple observation--that the current design of collective litigation is malleable, vulnerable to myriad forms of ex ante manipulations--has not been made in the literature. It is the subject of the following Parts. Part II discusses how defendants can affect, ex ante, the odds of certifying class litigation. I refer to this phenomenon as doctrine-based malleability. Part III shows how defendants can use the variance among future plaintiffs to frustrate, before the occurrence of damages, the formation of successful collective litigation.


A. The Crucial Class Certification Standards

1. Overview

As mentioned above, class actions are considered an effective tool to vindicate victims' rights. While class litigation overcomes the problems that are associated with the inherent litigation advantages mass defendants have, it has its own flaws. Most notably, class action is a formal collectivization mechanism that binds absent plaintiffs to an outcome of a legal process in which they did not participate. This is a salient exception to a "principle of general application in Anglo-American jurisprudence." (58) Class actions, in short, harm the plaintiffs' due process rights. In order to respect victims' individual rights, legal systems define a minimum threshold to initiate class action. These are the certification requirements--the gatekeeper to the world of class litigation. Certification, then, embodies the balance between the need to achieve the advantages of collective litigation and the desire to respect individual rights.

Rule 23 of the Federal Rules of Civil Procedure delineates this balance. Among the basic prerequisites to certification, courts have to ensure that the class of victims is too large to voluntarily join into a collective; (59) that "there are questions of law or fact common to the class;" (60) and that the class is adequately represented. (61) These and additional requirements that Rule 23 sets are crucial. The stakes are high. In the absence of class certification, individual litigation is often impractical; and, as a matter of fact, once a class is certified the parties are likely to settle for sizeable amounts. No wonder, then, that Rule 23's standards are debated: defendants typically argue that plaintiffs fail to meet these requirements, and plaintiffs argue that certification is appropriate.

While Rule 23's requirements are a sensible apparatus to determine the borderline between class and individual litigation, these requirements are vulnerable to manipulation by the defendant. The task of this section is to substantiate this argument, focusing on the requirement that the questions raised by the plaintiffs are not too individual.

2. Individual versus Common Questions

Many of the inquiries that judges ought to take before certification center on whether there are "too many" individual questions. The aggregation of the plaintiffs' cases can raise common questions, which are similarly solved for all the members of the class (e.g., whether the harmful product was negligently designed); it can also raise individual questions that are resolved differently for each member of the class (e.g., whether, and to what extent, each plaintiff negligently contributed to the damages). (62)

Courts, in principle, disfavor certifying a class in which there are too few common questions relative to individual ones--i.e., the class is not sufficiently "cohesive." The tension between individual and common questions appears in several contexts. First, Rule 23 (a) demands, as a basic prerequisite to certification, that "there are questions of law or fact common to the class" and that the representative's claims are typical of the common ones. (63) Second, the most prevalent clash between individual and common inquiries is the predominance requirement. In order to certify a Rule 23(b)(3) class action, the court should find "that the questions of law or fact common to class members predominate over any questions affecting only individual members." (64) Finally, even where common questions predominate over individual ones, courts deny certification of class actions that are not "manageable" due to a high proportion of individual questions. (65)

The crucial role that individual differences play in class certification makes some sense. The more individual questions there are, the greater the effort the court will have to invest in adjudicating the class;

and vice versa: more common questions mean a more easily resolved class action. (66) Similarly, more individual questions indicate that the court should be more cautious toward plaintiffs' procedural rights. "The presence of important noncommon issues ... raises doubts about the administrative efficiency of combining claims into a class action and about the infringement of plaintiffs' individual autonomy and due process rights." (67) In addition, for the reasons discussed above, more common questions aggravate the litigation advantages that mass defendants have. More common questions indicate a higher correlation across cases; collectivization is presumably more necessary the more correlated the cases. (68) On the other hand, one can question all these arguments. The existence of relatively more individual questions does not make collectivization futile. More commonality among claims makes class litigation more efficient, but collectivization of cases with at least some common questions is presumably more efficient than the baseline of litigating each case individually. (69) Likewise, as long as there are some common questions, mass defendants can have a substantial litigation advantage. The relevant question for policy makers to ask is the extent to which defendants are likely to be in a better position when litigating individually. While the degree of common questions is one factor that affects the importance of class litigation in equalizing the litigation playing field, it is by no means the only one. Collectivization is particularly important, for example, where the alternative is likely to be no litigation and zero deterrence (and compensation), as is common in low-value claims, regardless of the common/individual questions ratio. A greater portion of individual questions in the mix of common issues, then, does not eliminate the necessity of collectivization.

Certification standards reflect a balance between conflicting interests. The predominance requirement and the way courts have interpreted Rule 23 embody an essential aspect of this balance--class actions should not consist of "too many" individual questions vis-a-vis common questions. While there are pro and con arguments, this is a defensible position. However, under the conventional view this balance assumes for each class litigation a given, exogenous mixture of individual and common questions. (70) It is here that the predominance requirement becomes more tenuous. As I will describe below, the individual/common questions ratio is not predetermined; rather, in many cases it is the defendant's ex ante choice of action that dictates the level of individual questions--defendants can easily "individualize" the prospective class. From this perspective, it is far less obvious to decide whether to certify a class based on a criterion controlled by one of the parties.

B. Inducing Factual Differences--"Individualizing" the Class

Current doctrine, I argue, enables defendants to influence in advance the level of individual questions, and, as a result, whether courts would certify class litigation. Doctrine allows defendants to "individualize" the class. The way to do so for defendants is to choose, before the occurrence of damages, a course of action that induces factual differences. In many cases, as case law demonstrates, these factual differences frustrate class certification. The following paragraphs provide paradigmatic examples in which class certification is sensitive to the course of action the defendant had chosen.

1. Individualizing Written Contracts

A supplier can offer its customers a uniform contract. Alternatively, it can tweak the standard contract, injecting individual differences. Some courts find these differences destructive of class certification. Against this backdrop, one court denied certification of contract claims in a case where physicians alleged that the defendants--health maintenance organizations (HMOs)--breached contracts, systematically underpaying the doctors for medical services:

   The algorithms ... [used by the defendants' computer programs]
   appear to be ... varied and complicated.... Instead of applying one
   specific universal rule to cheat all doctors (e.g., automatically
   deducting $100 from everyone's claim), the ... programs are instead
   alleged to apply a variety of more individually tailored rules....
   Even if the plaintiffs were to prove that computer systems
   "sometimes" [underpays the plaintiffs] this fact would do nothing
   to further any of the plaintiffs' individual breach of contract
   claims. (71)

Importantly, it seems that the individualized class in this example is in a similar need of collectivization as the hypothetical, non-individualized one, in which defendants would apply "one specific universal rule to cheat all doctors." The cases in the individualized class are a bit less correlated, but litigating them individually is likely to result in substantially lower, if not zero, overall liability. Hence, the policy justifications for class litigation in the individualized scenario are (almost) equally important.

Many other examples of individualized contracts exist. (72) While individualizing the class may be a demanding task in certain circumstances, in many cases--as the HMOs example suggests--it simply requires programming slightly different contracts. (73) As one defense attorney advises:

   Remember, class actions thrive on similarity.... If you can make
   changes that introduce significant variability, you can limit your
   risk of class actions. So look for ways to modify the material
   terms of the contract.... Change the wording. Move paragraphs
   around. There is almost never just one way to say something....
   [M]ix it up a little. If you do this correctly, you can limit class
   action liability.... (74)

Defendants, then, can often introduce individual differences in a relatively easy manner. The certifiability of class actions that are justified on policy grounds is, at least in certain cases, completely under the defendant's control.

2. Creating Choice of Law Differences

A related opportunity to avoid, ex ante, class certification, is inserting choice of law differences. Divergence in applicable law means, in essence, differences among the plaintiffs, as there are more questions that are not completely common to all the members of the class. (75) Courts are indeed reluctant to certify a class where different state laws apply. "The existence of significantly differing state laws currently poses a virtually insuperable obstacle to certification of multistate, diversity class actions.... [C]ourts regard the potential management difficulties and diseconomies of this 'daunting enterprise' sufficient to tip the balance against class certification." (76) However, uniformity in applicable law is not an objective, pre-existing standard. Rather, it can be set, to some extent, by the defendant. Consider the following:

   Firms can create, pre-damages, differences in applicable law. One
   way to do so is using different types of standard-form contracts,
   each of which dictates a different law governing the transaction.
   Another way is stipulating that the applicable law is the law of
   the state where the consumer resides. The result is similar: after
   damages occur, courts will be reluctant to authorize a unified
   action due to the multiplicity of applicable law. (77)

As before, these induced variations allow defendants to reduce the odds of prospective certification, at least in certain circumstances. (78)

3. Varying Oral Communication

Mass service providers often have a choice as to the method to convey the message to their customers. Oral representation is an individualized method of communication, which, compared with written representation, is more likely to escape prospective class certification. In one paradigmatic case, for instance, where patients who received an implant that was not approved by the Food and Drug Administration (FDA) brought a class action against the physician, the trial court denied certification of informed consent claims:

   Given the distinctly factual nature of the informed consent
   question, however, the court hesitates to find commonality on this
   issue.... The emotional state and injury of each proposed
   plaintiff, not to mention possible variations in the conversations
   with [the Doctor], renders this area unsuitable for certification.

As this example shows, oral, individual communication has the same "individualizing" capacity as the contractual modifications in the previous examples. Both can prevent inference of sufficient commonality and hence block class certification. (80) In this example the individualizing force of the oral consultations with the doctor was strong enough to overcome the uniform consent form, which did not disclose that the implant was not FDA approved, (81)

Similar examples, illustrating the individualizing role of oral communications, regularly appear where plaintiffs claim breach of orally-made (or orally-negotiated) contractual obligations (82) and in fraud cases. (83) As before, both the "individualized" and the "non-individualized" versions of the cases seem to justify class treatment--in the absence of class action, individual litigation is likely to extract far less compensation, if any, from the defendant.

4. Heterogeneous Plaintiffs and Products

The more diverse and heterogeneous the class, the less likely is certification. Similar to the previous examples, the range of damages and the plaintiffs' identity--both predict certification--are often under the defendant's control. Hence, as before, the class is ex ante malleable. Consider the following decision to deny certification of fraud allegations against gaming machine manufacturers, due to the inherent differences within the pool of customers:

   Gamblers do not share a common universe of knowledge and
   expectations.... Some players may be unconcerned with the odds of
   winning, instead engaging in casual gambling as entertainment or a
   social activity. Others may have played with absolutely no
   knowledge or information regarding the odds of winning.... Still
   others, in the spirit of taking a calculated risk, may have played
   fully aware of how the machines operate. (84)

Targeting a more diverse audience is therefore valuable for the prospective defendant, as it reduces the odds of class action if and when damages materialize, (85) Likewise, creating variability in damages can block certification--as one court reasoned its decision to deny certification of antitrust claims, "the prices for some customers are going up while the prices of other customers are not." (86) Similar logic makes the defendant better off when its activity potentially harms plaintiffs in different times and locations--the more dispersed the activity is, courts are generally more reluctant to certify. (87) A related reasoning applies to products heterogeneity. "[W]here the allegedly defective 'product' is actually a line of products that are similar, yet have distinguishable formulations, compositions, or configurations, courts refuse to certify on the grounds of factual variation." (88)

As in the previous examples, targeting diverse plaintiffs and disseminating heterogeneous products can individualize the class, making it uncertifiable; denying certification, in turn, allows the defendant to pay less than the harm it allegedly inflicted on the victims.

5. Decentralizing Action

Class action defendants are often hierarchical organizations. Their structures matter. Courts are less likely to certify class litigation where more discretion resides at the defendant's lower ranks, as individual differences are then more salient. Hence, by decentralizing decision making, an organization can avoid class litigation. Take the following example:

   Facing potential future liability, a retail giant chooses to
   delegate authority to local managers. To avoid class action based
   on discriminatory policies, for instance, local supervisors hold
   discretion regarding hiring and promotion.

The recently decided Wal-Mart case presents an essentially similar question. (89) As Wal-Mart's counsel concluded at the oral argument, "because the plaintiffs' claims in this case hinge on the delegation of discretion to individual managers throughout the country, they cannot meet the cohesion requirements." (90) Employment discrimination is but one example of molding class certification ex ante through decentralization. (91) In other areas, decentralized discretion leads to similar individualization of the class. (92) "[A] company could significantly limit its risk of class actions by introducing more autonomy in how sales and customer service personnel deal with consumers...." (93) Again, policy justifications point in similar directions in both the decentralized and the centralized cases--in both, without class litigation the defendant's overall liability would fall short of the damages it inflicted on the plaintiffs.

This list of mechanisms, from injecting contractual differences to decentralizing discretion, is not exhaustive. (94) Some of the mechanisms may overlap with others. Nonetheless, these mechanisms do illustrate the very same point--ex ante choices of business organization can make prospective certification harder and hence reduce liability. As the aforementioned quotations from defense lawyers suggest, there is no reason to think that firms that face the risk of multi-million dollar class actions are oblivious to these considerations and to the opportunity to reduce prospective liability. While this Part focuses on the manipulability of the commonality requirements, the very same considerations motivate suppliers to draft the more commonly discussed mandatory arbitration clauses, which prohibit the use of class actions. (95) Indeed, in light of the straightforward effect of these mandatory arbitration class action waivers, the Court's decision that authorizes their use has been attacked by academics, legislators, and judges; (96) and it may lose some of its sweeping application as a result of this critical reaction. The lesson of this Article, however, is that a similar pre-litigation, collectivization-avoidance logic can proliferate in numerous domains without any visible response.

It is true that aggregation is sometimes in the interest of mass defendants. (97) It is also true that there are often independent business reasons to introduce individual differences. (98) Likewise, firms that do inject individual differences to reduce liability may simultaneously harm the efficient administration of their own businesses. (99) Nevertheless, these are qualifications to the general argument. The main message of this Part is that the level of individual differences, from defendants' perspective, is a choice variable--they are able to set it in order to achieve optimal outcomes. The strategic value of individual differences, along with other business considerations, factors into mass defendants' decision making. (100) Firms can choose, in each case, whether to "individualize" the prospective class or not. While sometimes firms are better off not changing their ordinary course of action, presumably there are many circumstances in which it is valuable for mass defendants to individualize the prospective class. (101) The unilateral capacity of mass defendants to manipulate collective litigation procedures, if they find it worthwhile, resurrects the one-sidedness that these procedures are aimed at rectifying.

The success of the foregoing mechanisms in individualizing the class stems from the doctrinal reluctance to certify classes that raise too many individual questions. Doctrine, as always, is far from being neat and well-settled. The rules are context-specific. While some courts are more skeptical of class litigation, others are willing to certify a class notwithstanding minor individual differences. However, even when courts do authorize a class, the exact boundary between "sufficiently cohesive" and "too individual" is hotly contested. (102) Plaintiffs argue that there is uniformity, defendants stress individual differences, and the court ultimately draws the "commonality line." (103) While this might be a reasonable judicial approach to apply law to facts, it is a less convincing one when the legal standard for adjudicating a class--sufficiently few individual questions--is controlled by one party. As the foregoing examples demonstrate, in many areas defendants have the capacity to manipulate, ex ante, the level of individual questions. From this perspective, the struggle to find the line between overly individualized and sufficiently uniform seems much less useful, given that the policy reasons behind class litigation apply to individualized and non-individualized classes in an almost similar force. Why should one commercial activity be exempt from effective enforcement while a similar activity is liable? Why should defendants avoid liability based on their choice of organizational structure?

The following Part demonstrates another manifestation of the same phenomenon--the malleability of collective litigation procedures.

Author Advanced search


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.