The Supreme Court recently killed one of the largest private-employer civil rights class actions in American history, Wal-Mart Stores, Inc. v. Dukes. After a decade of litigation, the case ended before it even began. Former and current female employees brought a class action against Wal-Mart Stores, Inc. (Wal-Mart), on behalf of roughly 1.5 million women, alleging nationwide gender discrimination in violation of Title VII of the Civil Rights Act of 1964. The plaintiffs contended that Wal-Mart gives its local managers undue discretion when making pay and promotion decisions, resulting in women being underpaid and disproportionately denied promotions. They argued that there is “a strong and uniform ‘corporate culture’” that permits gender stereotyping, and bias to taint, perhaps subconsciously, personnel decisions made throughout the company.
The plaintiffs’ complaint relied on both disparate impact and disparate treatment theories of discrimination. “Disparate impact” occurs when there is a neutral policy that has an adverse impact on a protected class of workers, such as women. “Disparate treatment” occurs when an employer intentionally treats a class of employees adversely because of their membership in a protected class. The Dukes plaintiffs contended that the company knew about the disparate impact its practices were having on its female employees and failed to do anything about it—amounting to intentional discrimination. The class representatives sought an injunction, declaratory relief, punitive damages, and back pay. The Court did not decide the merits of the case—i.e., whether in fact Wal-Mart discriminated against its female employees. Rather, the Court concluded that the case failed to meet the requirements of a class action, contrary to the Ninth Circuit’s en banc decision to affirm the district court’s certification.
The Court’s opinion—while silent on the actual merits of whether there is systemic gender discrimination at Wal-Mart—is a major blow to the plaintiffs’ case because of the unique and powerful role of a class action. There is strength in numbers, especially when that number is 1.5 million. But there is also strength in due process, especially when there are billions of dollars at stake and the defendant that is being accused of massive wrongdoing is one of the largest companies in the world. The Dukes decision is important because it attempts to draw a boundary line. On the one hand, the class action is a procedural asset that promotes efficiency and court access—enabling plaintiffs with small claims and resources to jointly challenge widespread misconduct in a single suit. On the other hand, the class action is a procedural anomaly that is granted only under limited circumstances—enabling defendants to adequately defend themselves and class members to bring separate cases when their individual interests diverge. This Essay contends that the Court drew a boundary line that favors large, powerful employers over everyday workers alleging systemic discrimination.
A class action is “an exception to the usual rule that litigation is conducted by and on behalf of the individual named parties only.” Because a representative action runs counter to this fundamental principle, the legislature and judicial system have established rigorous criteria to ensure that departure from the norm is justified. The federal class action rule—Rule 23 of the Federal Rules of Civil Procedure—sets out the requirements for when a party can represent others so that efficiency and due process are served. The courts must conduct a rigorous analysis to make certain the Rule’s requirements are met.
For a case to be certified as a class action, all of the Rule 23(a) provisions and one of the Rule 23(b) provisions must be met. Rule 23(a) requires numerosity, commonality, typicality, and adequacy of representation. Cases certified under Rule 23(b)(1) and (b)(2) are mandatory class actions, in which class members are not required to receive notice and an opportunity to exclude themselves from the litigation. The class members are bound by a court judgment they may not have known about, much less consented to. This extraordinary situation is justified by the class’s homogeneity and cohesiveness. Rule 23(b)(1) allows a class action when there is a risk that, in the absence of a class action: (a) the party opposing the class will be subject to inconsistent obligations, or (b) as a practical matter, piecemeal litigation of individual class members will impair the interests of other class members who are not parties to the individual lawsuits, as is the case in a trust fund. Rule 23(b)(2) permits a class action when there is class-wide conduct that makes “final injunctive relief or corresponding declaratory relief” appropriate for the whole class, as is the case for many civil rights claims. Alternatively, a class action brought under Rule 23(b)(3)—the “catch-all” provision—requires that class members be provided notice and the right to opt-out because the connection between the class is not nearly as strong. A Rule 23(b)(3) class—the most common type—is permitted when common questions predominate over individual ones and a class action is superior to other methods of resolving a dispute.
One of the challenges to aggregate litigation is determining what due process is required when claims for monetary relief are involved. When seeking back pay or monetary damages, class members’ interests may diverge, thereby breaking down the homogeneity and cohesion that usually make notice and opt-out rights unnecessary. The Due Process Clause of the Fifth and Fourteenth Amendments forbids deprivation of “life, liberty, or property, without due process of law.” Therefore, a court must be careful not to deprive class members of their property—i.e., money—when requiring them to be in a mandatory class action. As the Supreme Court stated in dicta in Ortiz v. Fibreboard Corp., the “inherent tension between representative suits and the day-in-court ideal is only magnified if applied to damages claims gathered in a mandatory class” because the “legal rights of absent class members . . . are resolved regardless of either their consent, or . . . their express wish to the contrary.” When certifying a class action, a court must also be careful not to deprive a defendant of due process. A defendant must be able to adequately defend itself from individual claims whose aggregation may mask important distinctions and available defenses.
With such safeguards in place, the class action device plays a critical role in the American civil justice system. Not only does aggregate litigation save judges and parties substantial time and cost by resolving similar claims all at once, it also promotes law enforcement. For many employees and others, a class action is their only meaningful access to the courts. Those with small claims and limited resources are unlikely to challenge powerful corporations on their own, effectively immunizing companies from complying with the law. Individually, the litigation costs and attorney’s fees may exceed the value of the recovery—i.e., a negative-value suit—resulting in employees forgoing litigation. Collectively, employees can share the risks and burdens of litigation and pool their resources, making it economically feasible to challenge misconduct in court. Even if individuals are able to seek redress for individual harms, they cannot effectively challenge widespread misconduct in the absence of collective action. While individual cases may motivate employers to change their relevant policies and practices to avoid similar suits in the future, this pales in comparison to the remedies and scope of injunctive relief plaintiffs can craft in class actions. And government agencies—burdened by budgetary and political constraints—often cannot fill the gap left by the lack of private enforcement. Not insignificantly, employers also enjoy the efficiency and global peace that class settlements often provide. Thus, class certification should be demanding, but not so demanding that it compromises the numerous advantages aggregate litigation has to offer.
The Dukes class certification standard jeopardizes potentially meritorious challenges to systemic discrimination. By redefining the class certification requirements for employment discrimination cases in two major areas, discussed below, the Court compromises employees’ access to justice.
At the outset, the Dukes majority decertified the class by determining that there was not enough glue to hold the case together as a class action. Under Rule 23(a), a party seeking class certification must demonstrate that there is enough in common between the class representatives and members of the class to justify aggregate litigation. In the other words, there must be “questions of law or fact common to the class.” In a 5–4 decision written by Justice Scalia, the conservative majority raised the bar for commonality—arguably one of the easiest class action thresholds. Conceding that all it takes is a single common question to satisfy the requirement, the Court concluded that this criterion was not met.
Not surprisingly, the Court appropriately reiterated that the class certification standard is tough. A judge must rigorously analyze whether a case should be a class action, including deciding merits issues if they overlap with class certification. And the party moving for class certification must actually prove that the Rule 23 requirements are met. But what was surprising was the Court’s further step—requiring the Wal-Mart plaintiffs to prove, with significant evidence, that there exists a general policy of discrimination as a condition of class certification. Deconstructing the term “question” in Rule 23(a)(2), the majority concluded that it was not enough for plaintiffs to pose the question of whether there was a pattern or practice of discrimination to satisfy commonality. Now plaintiffs must know the answer.
As stated in General Telephone Co. of the Southwest v. Falcon, there is a wide gap between an individual case and a class action. Dukes aptly observed this fact. Just because an individual woman may have experienced gender discrimination at work does not mean that a class of women has experienced the same. But Dukes widened the gap. Relying on dicta in footnote fifteen of the Falcon decision, the Court required the plaintiffs to demonstrate commonality with “‘significant proof’ that Wal-Mart ‘operated under a general policy of discrimination.’” This interpretation of commonality goes beyond prior Title VII class action jurisprudence. The Ninth Circuit majority criticized its dissent for similar reasons:
The dissent . . . seeks to create a new class action requirement based on a hypothetical in one sentence of Supreme Court dicta; conflates, or at least fails to distinguish, the posture of Falcon and the present case; [and] ignores the weight of the many cases in other circuits arriving at the same [class certification] standard we have described . . . .
Dukes states that class members must each suffer the “same injury” rather than the same Title VII violation. The level of generality at which “injury” is defined is critical. Tellingly, Dukes further explains that “[t]heir claims must depend upon a common contention,” such as “discriminatory bias on the part of the same supervisor.” This exemplar suggests that the level of generality is narrow. The common contention could just as well have been discriminatory bias as a result of a policy of excessive subjectivity—the issue in Dukes. As if it was not enough that the Court unearthed and elevated the “significant proof” standard for commonality from a footnote in Falcon, the Court’s application of this standard confirms the formidable climb necessary to reach commonality in the future.
But the Court found that the evidence provided by the plaintiff class did not satisfactorily establish a common claim of discriminatory bias. To demonstrate commonality, plaintiffs proffered three types of evidence: statistics showing gender disparities in pay and promotions; 120 affidavits from female employees reporting discrimination; and testimony from a sociologist, who concluded that Wal-Mart was vulnerable to gender discrimination because of its corporate culture and personnel practices.
At the outset, by analyzing each type of evidence in isolation to determine whether it provided “significant proof” of a general discriminatory policy, the Court diminished the overall import of plaintiffs’ evidence. Any one type of evidence—statistics, affidavits, or expert testimony—by itself may fail to provide sufficient justification for aggregate litigation. But the evidence in toto creates a clearer picture of the common thread that holds the class together. The statistical disparity, anecdotal accounts, and “social framework” analysis, taken together, provide the glue necessary to bind 1.5 million separate stories. Disaggregation of the evidence, on the other hand, effectively disables a judge from connecting the dots and seeing the commonality necessary for class certification.
The Court gave each type of evidence short shrift. To the extent that the Court credited the statistical proof showing a gender disparity, the Court favored storewide statistics over regional and national ones. The 120 affidavits were viewed as a drop in the bucket in comparison to the size and geographic dispersion of the class. And the sociologist’s expert testimony—the “only evidence of a ‘general policy of discrimination’” according to the Court—was disregarded on the grounds that he could not discern to what extent stereotyped thinking and bias influenced pay and promotion decisions at Wal-Mart nationwide. Given the Court’s analysis, it is hard to imagine what type and quantity of evidence would satisfy the new “significant proof” standard for an employment discrimination case of this scope and magnitude.
First, the Court concluded that national and regional statistics demonstrating a gender disparity did not establish that plaintiffs’ theory of discrimination could be proved on a class-wide basis. Such statistics did not necessarily reflect storewide disparities, upon which commonality depended. Even if such statistical evidence demonstrated storewide gender disparities, the Court concluded that this still would not establish a common issue because each store manager would proffer a different explanation for its shortfall. The Court claimed that merely proving that a discretionary system resulted in a statistical disparity was insufficient to demonstrate commonality, absent identification of a specific employment practice. But precedent makes clear that an employer’s “undisciplined system of subjective decisionmaking” is an “employment practice” that may give rise to Title VII claims under disparate impact and disparate treatment theories. The Court was unsatisfied with this and concluded that this employment practice was not enough to tie together the claims of the class.
Second, plaintiffs’ anecdotal evidence did not help clear the commonality threshold. Despite sworn testimony of discrimination by over a hundred employees in a half-dozen states, the Court concluded that this evidence fell woefully short because of the sheer size and geographic dispersion of the class. In International Brotherhood of Teamsters v. United States, the Government, as plaintiff, produced forty affidavits from individuals testifying about race discrimination, in support of the plaintiff’s allegation of company-wide discrimination. Dukes favorably cited Teamsters for the fact that there was one anecdote for every eight class members. Had the plaintiffs collected affidavits in the same proportion—i.e., a one-to-eight ratio—they would have had to produce 187,500 affidavits for a class of 1.5 million members. This suggested proportion (or one even in the ballpark) effectively ensures that no plaintiff will be able to allege systemic discrimination by an employer the size of Wal-Mart using anecdotal evidence. The most affluent class counsel cannot realistically finance and staff a case requiring this kind of showing—even without additional types of evidence—merely to cross the commonality threshold.
Moreover, because the Court required plaintiffs to “demonstrate that the entire company ‘operate[s] under a general policy of discrimination,’” to satisfy commonality, even if all 120 accounts of gender discrimination were true, they were too weak to create an inference of class-wide discrimination. The conflation of class certification and the pattern-or-practice liability criteria catapulted commonality to an unattainable level. The amount of anecdotal evidence required to demonstrate commonality for a class this large is effectively out of reach.
Finally, the Court concluded that even the “only evidence of a general policy of discrimination”—expert testimony that Wal-Mart’s discretionary policy resulted in biased personnel decisionmaking—was “worlds away” from meeting the “significant proof” standard. First, the fact that Wal-Mart’s policy gave local supervisors unfettered discretion over employment decisions called into question whether there was a uniform employment practice that could be challenged on a class-wide basis. The notion that unchecked, local, subjective decisionmaking at 3,400 separate stores across the country could provide the common thread for a single action is admittedly counterintuitive. This is true, of course, if one focuses on the trees rather than the forest. Commonality depends on the locus of analysis. If the locus is the thousands of supervisors in the field, making myriad decisions affecting 1.5 million separate employees, it is easy to conclude that there is no common question whose resolution would decide the case. But if the locus is the company—giving its agents the authority to make biased employment decisions while looking the other way—it is easier to see how the case can be resolved on a class-wide basis. By focusing on the employer, the entity culpable under Title VII and the one responsible for systemic harm, the uniform employment practice becomes apparent. The various ways the discrimination plays out as a result of this practice is a red herring, unnecessary to the threshold commonality determination under Rule 23(a)(2).
Furthermore, even assuming Wal-Mart operated an “undisciplined system of subjective decisionmaking” potentially actionable under Title VII, the Court did not find that sufficient to meet the commonality test. This stems from the majority’s skepticism, if not disbelief, that a majority of Wal-Mart’s managers might act—even subconsciously—in a way that disfavors women’s employment prospects. The Court stated, without support, that “left to their own devices most managers in any corporation—and surely most managers in a corporation that forbids sex discrimination—would select sex-neutral, performance-based criteria for hiring and promotion that produce no actionable disparity at all.” The Court insisted that plaintiffs identify a “common mode” of how supervisors exercise their discretion throughout the company but then rejected the statistics, affidavits, and expert evidence indicating that gender bias might be the answer.
The inability of plaintiffs’ expert to discern to what extent gender bias permeated supervisors’ decisionmaking fueled the Court’s skepticism. The expert could not answer “whether 0.5 percent or 95 percent of the employment decisions at Wal-Mart might be determined by stereotyped thinking.” Because the Court insisted that the answer to this question—rather than the question itself—was the basis for plaintiffs’ commonality theory, the Court found the sociologist’s testimony useless. But the answer to this “essential question” is not only unknown, but unknowable. Given the subtle, complex and sometimes even unconscious nature of modern discrimination, it would be practically impossible to determine with any specificity how much gender bias infected the workplace.
The Court’s skepticism was also bolstered by the fact that Wal-Mart has an “announced” policy forbidding gender discrimination. Juxtaposing Wal-Mart’s official non-discrimination policy with its policy of giving local supervisors unfettered discretion to make employment decisions, the Court was not persuaded that plaintiffs had provided “significant proof” sufficient to bridge the gap between an individual and class case. But surely the mere presence of a written anti-discrimination policy should not be able to destroy commonality in putative employment discrimination class actions. Otherwise, all employers could bulletproof themselves from liability by inserting boilerplate language into their employee handbooks. Given contemporary societal attitudes about flagrant gender discrimination, one would expect most employers to have official anti-discrimination statements in their personnel materials. It would be naïve to presume the absence of gender bias in the workplace because of such a statement. Moreover, the statement would have no bearing in disparate impact cases, where plaintiffs need not prove intent.
In contrast, the dissent—comprised of all the female justices and Justice Breyer—had no problem concluding that Wal-Mart’s discretionary policy may have resulted in systemic bias:
The practice of delegating to supervisors large discretion to make personnel decisions, uncontrolled by formal standards, has long been known to have the potential to produce disparate effects. Managers, like all humankind, may be prey to biases of which they are unaware. The risk of discrimination is heightened when those managers are predominantly of one sex, and are steeped in a corporate culture that perpetuates gender stereotypes.
Finding no error of law or abuse of discretion, the dissent deferred to the district court’s findings indicating potential class-wide disparate impact and disparate treatment gender discrimination. In finding sufficient commonality, the district court relied on plaintiffs’ evidence suggesting a system of discretionary decisionmaking that operates uniformly across stores, a corporate culture that suffuses gender bias, a company failure to counter such bias, and pay and promotion disparities that “can be explained only by gender discrimination and not by . . . neutral variables.”
The implications of this close, highly controversial portion of the Dukes opinion are varied. On the one hand, the Court’s ruling may have little impact on employment discrimination class actions. While the Court’s insistence that plaintiffs put forth more evidence to satisfy commonality will likely discourage employees from bringing class actions of this magnitude and scope, cases the size of Dukes are rare. With 1.5 million potential class members nationwide, Dukes unquestionably tested the outer bounds of what it takes to hold a class together. Smaller classes are bound to be more successful.
On the other hand, the Court’s ruling has the potential to cut short a number of employment discrimination class actions premised on the theory of excessive subjectivity as a discriminatory policy. Although Dukes has received a lot of attention because of its extraordinary facts, it is not an outlier when it comes to the plaintiffs’ underlying theory of liability. The reliance on Wal-Mart’s discretionary policy as a vehicle for disparate impact and disparate treatment discrimination litigation makes the case prototypical, rather than exceptional. Consequently, to satisfy commonality generally, judges may now require a stronger causal connection between an employer’s discretionary decisionmaking policy and a disparity or adverse employment action. This shift will make it harder for employees relying on this theory to act collectively.
II. Back Pay and Other Monetary Relief
The conservative majority decertified the class action by concluding that there was not commonality under Rule 23(a)(2). Because a class action cannot survive without commonality, the Court could have stopped there. Indeed, the Court did refrain from ruling on any of the other Rule 23(a) criteria for this very reason. Despite the majority’s admission that going any further was unnecessary, it decided that it was inappropriate for plaintiffs to seek back pay under Rule 23(b)(2).
Like many employees challenging systemic discrimination, the Dukes plaintiffs sought not only an injunction to curtail the company’s allegedly discriminatory policies and practices, but also back pay. Back pay compensates employees for earnings they would have received in the absence of discrimination. Back pay not only makes victims of discrimination “whole.” More importantly, it encourages voluntary compliance with the law. In enacting Title VII of the Civil Rights Act, Congress made clear that back pay, like injunctive and declaratory relief, is essential to law enforcement. Because of the importance of back pay, there is even a presumption in favor of it when discrimination is established.
The Court’s unanimous conclusion that back pay was not appropriate for the type of class action certified in Dukes was surprising. This gratuitous decision effectively reversed almost a half-century of Title VII jurisprudence permitting back pay under such circumstances. In the first and only Title VII post-Dukes case to date, the federal district court concluded: “In so holding, a unanimous Supreme Court reduced to rubble more than forty years of precedent in the Courts of Appeals, which had long held that backpay is recoverable in employment discrimination class actions certified under Rule 23(b)(2).” Courts have regularly permitted back pay for civil rights cases under Rule 23(b)(2) on the grounds that this monetary relief is equitable and critical to Title VII’s remedial scheme. Even appellate courts with the toughest class certification standards have recognized that back pay is consistent with the Rule’s strictures.
Despite this history, the Court found the equitable nature of back pay irrelevant and conditioned the availability of back pay on whether or not it was incidental to the injunctive or declaratory relief sought. The Court concluded that Wal-Mart was entitled to have back pay determined individually rather than formulaically, thereby making such relief non-incidental to the class-wide injunction.
The Court determined that back pay could not be calculated on an aggregate basis (contrary to the approach taken by many courts ); rejected the formulaic method proposed by the Dukes plaintiffs; and reiterated Wal-Mart’s right to raise individual affirmative defenses to each class member’s claim following a finding of a pattern or practice of discrimination. Unlike Hilao v. Estate of Marcos—another Ninth Circuit decision, in which the court allowed the defendant to present individual defenses to a random sampling of cases used to extrapolate class-wide compensatory damages—Dukes disapproved of this “novel project.” The Court concluded that, should back pay be calculated on the basis of “Trial by Formula” rather than through individualized hearings, Wal-Mart would lose its statutory right to defend itself from individual claims in violation of Title VII. Thus, it was the individualized—rather than the monetary—nature of back pay that made it inappropriate for Rule 23(b)(2) certification.
Once Dukes concluded that back pay had to be calculated individually, it was not a stretch for it to find such monetary relief non-incidental, and thus inappropriate for Rule 23(b)(2) certification—at least under the standard established by the Fifth Circuit in Allison v. Citgo Petroleum Corp. This shift in Title VII jurisprudence is significant because of the greater difficulty employees have in aggregating their monetary claims under the alternative—Rule 23(b)(3). Civil rights plaintiffs have historically challenged systemic discrimination under Rule 23(b)(2), in part to avoid the more onerous burdens and costs associated with this alternative. Rule 23(b)(3) certification is available if common issues predominate over individual ones and a class action is a superior mechanism for resolving the dispute. Because this type of class action is less cohesive and homogeneous than one brought under Rule 23(b)(2), class members must be provided notice of the class action and an opportunity to opt out of the action. Not surprisingly, if a court decides that back pay must be calculated on an individualized basis, the court is more apt to conclude that individual issues predominate over common ones, thereby foreclosing Rule 23(b)(3) certification as well.
But even in those jurisdictions where plaintiffs may be able to clear the (b)(3)-certification hurdle under such circumstances, the cost of sending out class notices—which can reach hundreds of thousands of dollars—may be prohibitive. Consequently, between the heightened certification requirement for Rule 23(b)(2) cases involving monetary relief, and the more demanding burden and costs for Rule 23(b)(3) cases, some employees alleging systemic discrimination may not be able to successfully bring a class action at all. As class action expert, Professor John C. Coffee, Jr., notes, this may be the most significant problem Dukes poses for future employment discrimination cases:
The simple truth is that employment discrimination litigation cannot normally be certified under Rule 23(b)(3) because of the “predominance” requirement of that rule . . . . Even in a far simpler, more streamlined case than [Dukes], there will still typically be a host of individual issues that will make it difficult (and usually impossible) to satisfy that predominance standard.
. . . .
. . . [T]he liberal wing of the [C]ourt may not have recognized how procedurally cut off and trapped employment discrimination victims are if back pay cannot be obtained as a form of “incidental” relief under Rule 23(b)(2). Reinforcing this sense is the casual assertion by Justice Ruth Bader Ginsburg in her [Dukes] dissent that the case should be remanded to the district court for a determination as to whether it could be certified under Rule 23(b)(3). That idea is a non-starter. In all circuits, the predominance standard has long been the Grim Reaper of putative class actions, and the sprawling character of the [Dukes] class . . . doomed it from the start—if the predominance standard applied.
The Court’s unanimous decision to narrow the availability of back pay to very limited circumstances under Rule 23(b)(2) alters the prospect of Title VII claims being aggregated at all.
Moreover, what is interesting about Dukes is not only what the Court decided, but what it did not. First, the Court did not answer one of the questions for which it actually granted review: whether any monetary relief is appropriate for a Rule 23(b)(2) action, which mentions only injunctive or declaratory relief. Rule 23(b)(2) states that a class action is authorized where “the party opposing the class has acted or refused to act on grounds that apply generally to the class, so that final injunctive relief or corresponding declaratory relief is appropriate respecting the class as a whole.” The Court recognized that one possible reading of the Rule foreclosed monetary relief altogether—an interpretation the Court did not adopt. While the Court did not go so far as to prohibit monetary relief, it did not explicitly permit such relief either. Instead, it opted to prohibit individualized relief—whether monetary, injunctive, or declaratory.
Second, the Court surprisingly did not subscribe to any “predominance” test when determining whether back pay should be allowed under Rule 23(b)(2). All of the federal courts of appeals to have addressed the question of whether monetary relief—more specifically, damages—is allowed under the Rule have concluded that such relief is permitted so long as it does not predominate. While Rule 23(b)(2) itself does not mention anything about predominance, the Advisory Committee, which drafted the rule, did. The Advisory Committee Notes plainly state that, so long as the appropriate final relief does not relate “exclusively or predominantly to money damages,” Rule 23(b)(2) certification is proper. Because of the Rule’s silence on the matter, the circuit courts have uniformly relied on the Advisory Committee Notes for guidance. Their only disagreement has been how predominance should be determined. The circuit courts are split, creating three predominance tests: the Fifth Circuit’s popular “incidental” test, the Second Circuit’s ad hoc balancing test, and the Ninth Circuit’s “objective effects” test —each with its own advantages and disadvantages.
Oddly, the Court in Dukes applied the Fifth Circuit’s incidental test without formally adopting it or its underlying premise—predominance. Instead, the Court flatly rejected the predominance concept on the grounds that it was not in the Rule’s text; it encouraged class representatives to forego potentially valid monetary claims to maximize class certification and it required continual evaluation of class membership. Moreover, the Court did not provide a comparative analysis of the three predominance tests used by the circuit courts. This was a lost opportunity for the Court to explain why the Fifth Circuit’s incidental approach is superior to those of the Second and Ninth Circuits. The Court adopted the most onerous test for determining the propriety of monetary relief in a Rule 23(b)(2) class action without offering why this test trumped the others.
Third, the Court did not decide whether the absence of notice and a right to opt out violates due process for a mandatory class action involving monetary claims that do not predominate. In other words, the Court has left unanswered the question posed in 1985 in Phillips Petroleum Co. v. Shutts. Rather than shut the door on this matter once and for all, the Court left open a crack of doubt, concluding that there is a “serious possibility” that the absence of notice and an opt-out right “may” violate due process where monetary claims do not predominate. While on a practical level the matter may be resolved, the loophole remains unsatisfying.
The implications of this unanimous portion of the Dukes opinion are mixed. On the one hand, the Court’s ruling on back pay will make it more difficult for employees alleging systemic misconduct under Title VII to seek monetary relief. This is because it is now harder for plaintiffs to use the Rule 23 provision designed for such cases—(b)(2). First, the Court makes no distinction between equitable and non-equitable monetary relief as a basis for Rule 23(b)(2) certification. This means that back pay—normally favored because of its equitable nature—enjoys no preference over compensatory and punitive damages sought under this class action provision. Second, any monetary relief that is not incidental to the injunctive or declaratory relief cannot survive Rule 23(b)(2) certification. For those cases brought in the Second and Ninth Circuits, or in circuits that have not ruled on the matter, a court must use the harshest standard established for evaluating the availability of relief that is neither injunctive nor declaratory. Third, courts are more likely to conclude that back pay must be determined on an individual, rather than an aggregate, basis. The individualized nature of the relief, in turn, makes it non-incidental, and thus unavailable under the Rule 23(b)(2) provision.
Because of the more difficult certification standard under Rule 23(b)(2), employees may decide to seek only injunctive or declaratory relief, or to forego bringing a Title VII class action altogether. For those who seek certification under the alternative class action provision, Rule 23(b)(3), its more onerous burdens and costs may also foreclose aggregate litigation. These outcomes risk underenforcement.
On the other hand, while Dukes may have made class certification harder, it did not eliminate Title VII class actions. Employees bringing a pattern-or-practice employment discrimination case involving monetary relief can seek certification solely under Rule 23(b)(3) or a hybrid—where injunctive or declaratory relief is sought under Rule 23(b)(2) and monetary or individualized relief is sought under Rule 23(b)(3). This approach gives employees notice of the litigation and the option of not being part of a class whose cohesion is admittedly compromised by varied monetary interests, thereby addressing any due process concerns. Moreover, because Dukes eliminated predominance as a prerequisite for injunctive or declaratory relief under Rule 23(b)(2), plaintiffs need not cherry pick which monetary relief to pursue under the Rule to enhance class certification. Plaintiffs can instead bring all of their claims for monetary relief under Rule 23(b)(3).
Finally, given the very fact-specific nature of Dukes’s back pay ruling, it may have a limited impact on cases brought in other substantive areas and under statutes other than Title VII.
In sum, Dukes has redefined the class certification requirements for Title VII cases in ways that jeopardize potentially meritorious challenges to systemic employment discrimination. Although the ultimate scope and magnitude of Duke’s impact is unclear, it is clear that Dukes has tipped the balance in favor of powerful employers over everyday workers.
*. Associate Professor of Law, The Catholic University of America, Columbus School of Law. Much gratitude goes to Research Fellow Christina K. Setlow and Research Assistant Cara M. Swan for their excellent research and editorial assistance. I am also indebted to Dean and Professor Veryl V. Miles and the Columbus School of Law for their generous funding of this project. This article is dedicated to my sister, Suzanne M. Malveaux, and to all those women who have the courage and strength to dream big and challenge wrongdoing wherever it exists. Thank you for inspiring me.
1. 131 S. Ct. 2541 (2011).
2. Id. at 2547 (citing Civil Rights Act of 1964, 42 U.S.C. §§ 2000e–1 to 2000e–17 (2006)).
3. Id. at 2548; id. at 2563 (Ginsburg, J., dissenting) (noting that the district court considered methods used by Wal-Mart to “maintain a ‘carefully constructed . . . corporate culture,’ such as frequent meetings to reinforce the common way of thinking, regular transfers of managers between stores to ensure uniformity throughout the company,” and closely monitoring stores on a constant basis).
4. Id. at 2548.
8. See Fed. R. Civ. P. 23(a) (link). Rule 23(a) is satisfied when: (1) the class is so numerous that joinder would be impracticable; (2) the class shares common questions of law or fact; (3) the representative parties’ claims or defenses are typical of the class; and (4) the representative parties will fairly and adequately represent the class. Id.
9. See id. 23(c)(2)–(3).
10. See id. 23(b)(1).
11. See id. 23(b)(2).
12. See id. 23(c)(2)(B).
13. See id. 23(b)(3).
16. See Carnegie v. Household Int’l, Inc., 376 F.3d 656, 661 (7th Cir. 2004) (Posner, J.) (“The realistic alternative to a class action is not 17 million individual suits, but zero individual suits, as only a lunatic or a fanatic sues for $30.”).
17. Title VII employment discrimination cases may also be negative-value suits, despite the fact that their claims are not de minimis—as is often the case in consumer actions. See Suzette M. Malveaux, Fighting to Keep Employment Discrimination Class Actions Alive: How Allison v. Citgo’s Predomination Requirement Threatens to Undermine Title VII Enforcement, 26 Berkeley J. Emp. & Lab. L. 405, 429 (2005) (“Even if plaintiffs may have a greater incentive to pursue their individual claims because of Title VII’s $300,000 damage cap, plaintiffs are not able to spread the costs of litigation as class members would be able to in the class action context.”).
18. See 131 S. Ct. 2541, 2560–61 (2011).
19. Fed. R. Civ. P. 23(a)(2). A federal class action must meet the other criteria under Rule 23(a) as discussed in the Introduction of this Essay.
20. See Dukes, 131 S. Ct. at 2556–57.
21. Id. at 2551–52.
22. Id. at 2551.
23. Id. at 2553–56 & n.9.
24. See id. at 2552.
25. 457 U.S. 147, 157–58 (1982). Falcon involved a lead plaintiff who alleged discrimination against Mexican-Americans. He alleged that he and a class of Mexican-American employees were subjected to intentional discrimination in promotions and that a class of Mexican-American applicants was subjected to disparate impact discrimination in hiring. Id. at 162 (Burger, J. concurring in part and dissenting in part). The Court found that there were no common questions between the plaintiff and the applicant class. Id. at 157–58 (majority opinion).
26. Dukes, 131 S. Ct. at 2553 (quoting Falcon, 457 U.S. at 159 n.15).
28. 603 F.3d 571, 594–95 (9th Cir. 2010), rev’d, 131 S. Ct. 2541. Moreover, the Court noted that:
Falcon’s discussion of two distinct processes—hiring and promotion—for which “significant proof’’ could prove sufficient to certify a single class, is an unusually high standard that Plaintiffs here need not meet because they did not present the distinct legal theories of recovery that the Falcon plaintiffs, both employees and applicants, had pursued together in one class.
Id. at 595.
29. Dukes, 131 S. Ct. at 2551 (quoting Falcon, 457 U.S. at 157).
31. Id. at 2549.
32. Id. at 2555–56. There was a dispute over whether the plaintiffs’ expert did, in fact, provide storewide statistics. See id. at 2564 & n.5 (Ginsburg, J., dissenting).
33. Id. at 2556 & n.9 (“[W]hen the claim is that a company operates under a general policy of discrimination, a few anecdotes selected from literally millions of employment decisions prove nothing at all.”).
34. Id. at 2553 (emphasis added).
35. Id. at 2553–54 (holding that the sociologist’s testimony “does nothing to advance [plaintiffs’] case” and concluding that “we can safely disregard what he has to say”).
36. Id. at 2555.
38. Id. at 2555–56 (emphasizing that “[o]ther than the bare existence of delegated discretion” by Wal-Mart, plaintiffs did not identify a specific employment practice).
39. Id. at 2565 (Ginsburg, J., dissenting) (quoting Wards Cove Packing Co. v. Atonio, 490 U.S. 642, 657 (1989)) (internal quotation marks omitted).
40. See id. at 2554–55 (majority opinion).
41. Id. at 2555–57.
43. Dukes, 131 S. Ct. at 2556 (citing Teamsters, 431 U.S. 324, 338 (1977)).
44. With only 120 affidavits submitted, given the additional cost of experts, discovery, and attorney time, lead counsel for plaintiffs spent $7 million on the case. Leigh Jones, U.S. Law Firm Spent $7 Million to Sue Wal-Mart, Reuters.com, June 21, 2011, http://www.reuters.com/assets/print?aid=USTRE75K77C20110621 (link).
45. Dukes, 131 S. Ct. at 2556 (alteration in original) (emphasis added).
47. Id. at 2553–54. The expert’s testimony barely escaped analysis under the rigorous standard for admission established in Daubert v. Merrell Dow Pharmaceuticals, Inc. See id. (citing 509 U.S. 579 (1993)). The district court concluded that the Daubert standard for admitting testimony of an expert witness under Federal Rule of Evidence 702 did not apply at the class certification stage, and the Supreme Court—although not deciding the issue—“doubt[ed]” this was so. Id.
48. Id. at 2554.
49. The employer cannot exhibit reckless indifference to the discriminatory impact its policies have on its employees.
50. Id. at 2567 (Ginsburg, J., dissenting) (“Wal-Mart’s delegation of discretion over pay and promotions is a policy uniform throughout all stores. The very nature of discretion is that people will exercise it in various ways. A system of delegated discretion . . . is a practice actionable under Title VII when it produces discriminatory outcomes.”).
51. The individual stories of alleged discrimination are relevant to a Rule 23(b)(3), where the court must determine if common questions predominate over individual ones. See Fed. R. Civ. P. 23(b)(3). The dissent correctly noted that the majority conflated the requirements of Rule 23(a)(2) and Rule 23(b)(3). See Dukes, 131 S. Ct. at 2565–67 (Ginsburg, J., dissenting).
52. Dukes, 131 S. Ct. at 2554 (majority opinion) (quoting Watson v. Fort Worth Bank & Trust, 487 U.S. 977, 990–91 (1988)).
53. See id.
54. Id. (emphasis added).
55. See id. at 2554–55 (“[Plaintiffs] have not identified a common mode of exercising discretion that pervades the entire company—aside from their reliance on Dr. Bielby’s social frameworks analysis that we have rejected.”). Additionally, the Court noted that “[i]n a company of Wal-Mart’s size and geographical scope, it is quite unbelievable that all managers would exercise their discretion in a common way without some common direction. [Plaintiffs] attempt to make that showing by means of statistical and anecdotal evidence, but their evidence falls well short.” Id. at 2555.
56. Id. at 2553 (quoting Dukes v. Wal-Mart, Inc., 222 F.R.D. 189, 192 (N.D. Cal. 2004)) (internal quotation marks omitted).
57. Id. at 2554.
59. Id. at 2553. The district court also recognized that Wal-Mart imposes penalties for equal employment opportunity violations. Id.
61. Id. at 2564 (Ginsburg, J., dissenting) (concluding that “[i]t is hardly surprising that for many managers, the ideal candidate [is] someone with characteristics similar to their own”).
62. Id. at 2562–64.
63. See id. at 2564–65.
64. Moreover, to the extent that Dukes elevates the proof necessary to demonstrate commonality under Rule 23(a)(2), this elevation may arguably apply to all class actions. See id. at 2565–66 (“Because Rule 23(a) is also a prerequisite for Rule 23(b)(1) and Rule 23(b)(2) classes, the Court’s ‘dissimilarities’ position is far reaching.”).
65. Id. at 2551 n.5 (majority opinion) (“In light of our disposition of the commonality question . . . it is unnecessary to resolve whether [plaintiffs] have satisfied the typicality and adequate-representation requirements of Rule 23(a).”).
66. See 118 Cong. Rec. 7168 (1972) (remarks made by Sen. Williams in a section-by-section analysis of The Equal Employment Opportunity Act of 1972); see also United States v. N.L. Indus., Inc., 479 F.2d 354, 379 (8th Cir. 1973) (describing the compensatory and deterrent functions of back pay) (link).
67. Albemarle Paper Co. v. Moody, 422 U.S. 405, 419–21 & n.12 (1975) (link); see also 1 Janice Goodman, Mary Ann Oakley, Alice D. Bonner, Edith Barnett & Suzanne Sangree, Employee Rights Litigation § 2.10[a][i] (2010) (“Back pay is the most common form of monetary relief in Title VII cases . . . [and is] . . . routinely granted barring extraordinary circumstances.”); id. (“[T]he denial of back pay to prevailing plaintiffs is a minor exception rather than the rule.”).
68. United States v. City of New York, No. 07-CV-2067, 2011 WL 2680474, at 698 (E.D.N.Y. July 8, 2011); see 5 Lex K. Larson, Employment Discrimination § 92.11 (2d ed. 2011) (citing cases to support the assertion that “the majority of courts have had little difficulty fitting an action for back pay and injunctive relief into Rule 23(b)(2)”).
69. City of New York, 2011 WL 2680474 at *7–8 & n.3; see, e.g., Robinson v. Metro-North Commuter R.R. Co., 267 F.3d 147, 169 (2d Cir. 2001) (collecting cases) (link); Allison v. Citgo Petroleum Corp., 151 F.3d 402, 415–416 & n.10 (5th Cir. 1998) (“Of course, to the extent the district court applied an incidental damages standard to plaintiffs’ claims for back pay, its analysis was flawed.”) (link); id. at 425 (“[W]e hold that nonequitable monetary relief may be obtained in a class action certified under Rule 23(b)(2) only if the predominant relief sought is injunctive or declaratory.” (emphasis added)); see also Suzette M. Malveaux, Class Actions at the Crossroads: An Answer to Wal-Mart v. Dukes, 5 Harv. L. & Pol’y Rev., Mar. 21, 2011, http://hlpronline.com/2011/03/class-actions-at-the-crossroads/ (discussing same) (link).
70. Dukes v. Wal-Mart Stores, Inc., 603 F.3d 571, 618–19 & n.40 (9th Cir. 2010), rev’d, 131 S. Ct. 2541 (2011) (citing cases adopting the “consensus view” and noting that “it is . . . well accepted, even by circuits that are generally restrictive in certifying classes seeking monetary damages under Rule 23(b)(2), that a request for back pay in a Title VII case is fully compatible with the certification of a Rule 23(b)(2) class”); see also, e.g., Thorn v. Jefferson-Pilot Life Ins. Co., 445 F.3d 311, 331 (4th Cir. 2006) (“[W]e do not hold, nor have we ever held, that monetary relief is fundamentally incompatible with Rule 23(b)(2).”) (link); Allison, 151 F.3d 4 at 415 (concluding that Rule 23(b)(2) permits monetary relief that is equitable, and “[b]ack pay, of course, had long been recognized as an equitable remedy under Title VII”); Pettway v. Am. Cast Iron Pipe Co., 404 F.2d 211, 257–58 (1974) (concluding that, under the circumstances, ‘an award of back pay, as one element of the equitable remedy, conflicts in no way with the limitations of Rule 23(b)(2)’ (quoting Robinson v. Lorillard Corp., 444 F.2d 791, 802 (4th Cir. 1971))) (link); Johnson v. Ga. Highway Express, Inc., 417 F.2d 1122, 1125 (5th Cir. 1969) (“The demand for back pay is not in the nature of a claim for damages, but rather is an integral part of the statutory equitable remedy, to be determined through the exercise of the court’s discretion, and not by a jury.”) (link).
71. Wal-Mart Stores, Inc. v. Dukes, 131 S. Ct. 2541, 2560 (2011).
72. The Court held that monetary relief may not be certified under Rule 23(b)(2) where such relief is not “incidental to the injunctive or declaratory relief.” Id. at 2557. The Court, however, refrained from deciding if any forms of incidental monetary relief would be allowed. See id at 2560.
73. Although the Court never explicitly adopts the Fifth Circuit’s “incidental” test for determining whether monetary relief is permitted under Rule 23(b)(2), which is set forth in Allison v. Citgo Petroleum Corp., the Court subsequently applies its incidental test to plaintiffs’ back pay claims, and concludes that such relief is impermissible because it failed the test. See id. at 2560–61 (citing Allison, 151 F.3d 402, 415 (5th Cir. 1998)).
75. See, e.g., Dukes v. Wal-Mart Stores, Inc., 603 F.3d 571, 619 & n.41, rev’d, 131 S. Ct. 2541 (2011); Thorn v. Jefferson-Pilot Life Ins. Co., 445 F.3d 311, 331 (4th Cir. 2006); In re Monumental Life Ins. Co., 365 F.3d 408, 418 (5th Cir. 2004) (link); Coleman v. Gen. Motors Acceptance Corp., 296 F.3d 443, 449–50 (6th Cir. 2002) (link); see also Larson, supra note 68, § 92.11 (“It is sometimes impractical, when a large class is involved, to calculate back pay for each member on an individual basis. The approaches . . . adopted by . . . various circuits confirm that a court may . . . award . . . back pay on a class-wide, rather than individual, basis.”); id. (“[C]ourts generally have employed a class-wide remedy when individual determinations are impracticable.”).
76. The Supreme Court established a method for trying a pattern-or-practice case in International Brotherhood of Teamsters v. United States. 431 U.S. 324, 360–62 (1977). The trial is divided into two phases: liability and remedy. At the liability phase, plaintiffs have the burden of proving a pattern or practice of discrimination. If plaintiffs meet this burden, this creates an inference that each class member who suffered an adverse employment action was a victim of the discriminatory pattern or practice. Id. at 360. At the remedy phase, the burden shifts to the defendant who has the opportunity to prove otherwise. Id. at 360–61. Teamsters states that following a pattern or practice liability finding, “a district court must usually conduct additional proceedings . . . to determine the scope of individual relief.” Id. at 361 (emphasis added). Teamsters, however, does not require such proceedings. See id. at 361–62.
78. 131 S. Ct. at 2561.
79. Id. It would also violate the Rules Enabling Act for the Court to interpret Rule 23 in a way that would abridge the company’s substantive rights. See id. (quoting 28 U.S.C. § 2072(b) (2006)).
80. See 151 F.3d 402, 412–415 (5th Cir. 1998).
81. A class action may also be brought under Rule 23(b)(1), but this is inapplicable for most civil rights actions.
82. See, e.g., Allison, 151 F.3d at 412–13.
83. See Malveaux, supra note 17, at 427.
84. Id. at 425–26.
85. John C. Coffee, Jr., “You Just Can’t Get There From Here”: A Primer on Wal-Mart v. Dukes, 80 U.S.L.W. 93 (2011) (internal citation omitted).
86. The question presented on petition for a writ of certiorari was “[w]hether claims for monetary relief can be certified under Federal Rule of Civil Procedure 23(b)(2)—which by its terms is limited to injunctive or corresponding declaratory relief . . . .” Petition for Writ of Certiorari at i, Wal-Mart Stores, Inc. v. Dukes, 131 S. Ct. 795 (2010) (No. 10-277), 2010 WL 3355820, at *i.
87. Fed. R. Civ. P. 23(b)(2).
88. Dukes, 131 S. Ct. at 2557 (“[A]t a minimum, claims for individualized relief (like the backpay at issue here) do not satisfy the Rule[23(b)(2)].”). The Court relied on the text, history, and structure of Rule 23 in coming to this conclusion. Id at 2557–59.
89. Malveaux, supra note 69; see, e.g., Dukes v. Wal-Mart Stores, Inc., 603 F.3d 571, 615–17 (2010); Robinson v. Metro-North Commuter R.R. Co., 267 F.3d 147, 162–64 (2d Cir. 2001) (relying on the Advisory Committee’s note); Allison v. Citgo Petroleum Corp., 151 F.3d 402, 411–415 (5th Cir. 1998) (same).
91. The predominance test commonly applied in cases certified under Rule 23(b)(2) should not be confused with the predominance test used in cases certified under Rule 23(b)(3). The question for Rule 23(b)(2) certification is whether monetary relief predominates over injunctive or declaratory relief. The question for Rule 23(b)(3) certification is whether common issues predominate over individual ones. See Fed. R. Civ. P. 23(b)(3).
92. See Allison, 151 F.3d at 411, 415.
93. See Robinson, 267 F.3d at 162–64.
94. See Dukes, 603 F.3d at 615–17.
95. See Malveaux, supra note 69 (describing advantages and disadvantages of each predominance test).
96. Wal-Mart Stores, Inc. v. Dukes, 131 S. Ct. 2541, 2559 (2011). Moreover, the Court concluded that the predominance analysis violated Rule 23’s structure. Id. The Court reasoned that Rule 23(b)(3)—with its additional procedural protections—better accommodates monetary claims, and the fact that such claims are non-predominant does not change this. See id.
97. Id. When structuring a Rule 23(b)(2) class action, class counsel may not seek all potentially valid claims of monetary relief for fear that it will predominate over injunctive or declaratory relief. By doing so, class counsel risks sacrificing the monetary relief not sought because individual class members may be precluded from pursuing it under the collateral estoppel doctrine. In Dukes, this risk existed as to compensatory damages, which were not pursued class-wide.
98. Id. at 2559–60. The Court’s conclusion that “[t]he predominance test would also require the District Court to reevaluate the roster of class members continually,” id. at 2559, is overstated. Class membership is fluid, often changing as members leave employment, die, or no longer fall within the class definition. While the district court has the ongoing responsibility to ensure that class certification remains appropriate, it is not responsible for micromanaging the litigation. Courts manage class actions in broad strokes. Thus, the Court correctly noted that continual re-evaluation of class membership would waste the district court’s time. Id. at 2560. However, the Court’s solution—“that the backpay claims should not be certified under Rule 23(b)(2) at all”—overreaches. Id.
99. A comparative analysis of the three predominance approaches taken by the circuit courts is beyond the scope of this paper. For an analysis of these approaches, see Malveaux, supra note 69, at 27–37.
100. Dukes, 131 S. Ct. at 2559.
102. Dukes, 131 S. Ct. at 2559.
103. For an explanation of why the Second and Ninth Circuit approaches are preferable, see Malveaux, supra note 69, at 27–37.
104. The dissent suggested as much. See Dukes, 131 S. Ct. at 2561 (Ginsberg, J., dissenting) (noting that “[a] putative class of this type may be certifiable under Rule 23(b)(3)” and reserving the matter for consideration on remand).
Copyright 2011 Northwestern University
Cite as: 106 Nw. U. L. Rev. Colloquy 34 (2011), http://www.law.northwestern.edu/lawreview/colloquy/2011/18/LRColl2011n18Malveaux.pdf.
Persistent URL: http://www.law.northwestern.edu/lawreview/colloquy/2011/18