Paul Karlsgodt is a partner in the Denver office of Baker Hostetler and editor and primary contributor to the legal blog www.classactionblawg.com [4]. Mr. Karlsgodt’s law practice focuses on class action defense and other complex commercial litigation.
There can be little doubt that the past year was a good year for class action defendants in the U.S. Supreme Court. Although plaintiffs and defendants split on the total number of decisions two to two, arguably the two more significant decisions went defendants’ way. Smith v. Bayer [5] paves the way for multiple bites at the apple for class action plaintiffs seeking certification, but there are practical barriers keeping this from happening too often in practice, including the expense and effort involved in pursuing multiple class actions, the probability of removal under the Class Action Fairness Act (CAFA) in most cases, and the persuasive precedential effect of any court’s prior decision denying class certification. Erica P. John Fund v. Halliburton Corporation [6] clarifies that a plaintiff cannot be required to prove an element of a claim on the merits in order to justify class certification, but saying a plaintiff is not required to prove an element of a claim is a far cry from saying that plaintiffs do not have to meet all of the requirements of Federal Rule of Civil Procedure 23. By contrast, AT&T Mobility LLC v. Concepcion [7] opens the door to a way for at least some defendants to avoid having to defend class actions at all, while Wal-Mart Stores, Inc. v. Dukes [8] sets a high burden that a plaintiff must meet in order to justify certification.
A popular reaction from the media, practitioners, and scholars alike after Concepcion and Dukes is that the decisions marked a paradigm shift in class action jurisprudence in favor of defendants. Commentators mused that Concepcion could be the death knell of the consumer class action. Dukes was hailed as a landmark decision that would help to end costly and frivolous class-action litigation.
Yet, a review of the scores of lower court decisions issued after Concepcion and Dukes reveals that very little has changed. Dukes has not become a magic talisman for warding off class certification orders. While courts that may have been predisposed to refuse certification before Dukes may view the decision as an additional reason justifying that decision, courts that were predisposed to certify class actions before Dukes are still finding ways to do it. Very few decisions reflect that the decision was pivotal to the outcome of a class-certification motion. At best, the impact so far seems to be that Dukes could tip the scales toward the defendant in a case in which the court is on the fence. Meanwhile, a clear majority of the lower courts to consider the validity of a class arbitration waiver after Concepcion have found other ways to declare the provisions unenforceable.
However, just because Concepcion and Dukes do not appear to be causing the sea change in class actions that many predicted, that does not mean that the decisions are not valuable tools in defending class actions. The trick will be in understanding when and how best to use them. Although that is a question that will no doubt evolve over time, here are some observations based on what the courts have been doing so far.
AT&T v. Concepcion
One thing that has become clear in the few months since Concepcion was announced: the facts of that case do matter. Justice Scalia’s broadly worded opinion does not expressly turn on the consumer-friendly nature of AT&T Mobility’s arbitration clause. However, the subsequent lower-court decisions have been anything but unanimous in upholding class arbitration waivers. In fact, perhaps the only lower-court decision to actually enforce a class arbitration waiver following Concepcion is the Eleventh Circuit’s decision in Cruz v. Cingular Wireless LLC, which involved the very same language that was at issue in Concepcion.
Other arbitration waivers appear to remain open to a variety of different attacks. For example,  in Chen-Oster v. Goldman Sachs, Inc. [9], the U.S. District Court of the Southern District of New York relied on the federal common law of arbitrability rather than state-law public policy in invalidating an arbitration provision, an issue not addressed in Concepcion.  In NAACP of Camden County East v. Foulke Management Corp., a New Jersey state court found an agreement unenforceable on state-law grounds other than public policy.  According to the reasoning of U.S. District Judge Edward M. Chen of the Northern District of California in Kanbar v. O’Melveny & Myers [10], even the state law of unconscionability can be a basis to invalidate an arbitration provision, provided that the provision is found to be unconscionable for reasons other than a state policy against arbitration itself. Â
Many of the decisions so far may also be distinguishable from Concepcion because they involved employment-related claims, not consumer claims, and it is possible that defendants will find more success in cases involving consumer contracts. Still, the safest way to withstand scrutiny in any context is to place enough safeguards in the provision to ensure that a consumer with a real dispute has a meaningful and cost-effective way to bring his or her claims in an arbitration forum. Depending on the likely sizes of the disputes that a particular company might face, this may require many of the elements included in the AT&T Mobility provision, such as one-way cost shifting, a $10,000 minimum recovery if the claimant recovers more than the company’s final offer, the consumer’s right to elect certain cost-saving procedures, and a choice of forum convenient to the consumer. The prospect of having to provide a consumer-friendly arbitration clause in exchange for peace from class action litigation is enough to give many companies pause. Instead of fighting one or a handful of expensive class actions, the company may instead be faced with the prospect of thousands of expensive individual arbitration proceedings.
Even assuming that more companies begin to adopt arbitration clauses similar to the one used by AT&T Mobility, the long-term impact of Concepcion may simply be to shift the focus of consumer class action litigation in the United States to other industries, such as insurance, to which the Federal Arbitration Act does not apply, or to situations in which companies have not interfaced directly with consumers and therefore have no arbitration agreement to enforce.
Wal-Mart Stores, Inc. v. Dukes
From a defense practitioner’s perspective, it is helpful to think about Dukes as a Swiss army knife as opposed to a sledge hammer in terms of its usefulness in opposing a motion for class certification at the trial court level. A citation to Dukes will probably not win a class certification motion by itself, but the decision can be a very helpful tool in explaining the proper analysis on a variety of class certification issues.
The standard of proof. If there were any doubt before the decision, Dukes makes clear once and for all that the rigorous analysis standard requires the federal courts to look behind the pleadings and make findings of fact where the facts bearing on an element of Rule 23 are in dispute. It also illustrates how the courts should give practical consideration to what exactly the plaintiff will have to prove at trial and whether the claims can actually be proved through common evidence. In light several key court of appeals decisions prior to Dukes (including the Second Circuit in In re IPO Securities Litigation, the Third Circuit in In re Hydrogen Peroxide Antitrust Litigation, and even the Ninth Circuit’s opinion in Dukes [11] itself), this analytical model is not groundbreaking, but Dukes puts an end to any possible reluctance by the federal courts to resolve an issue of fact simply because it might overlap with the merits.
Commonality. One significant contribution that Dukes has made to the defense lawyer’s repertoire is that it gives teeth to the commonality element of Rule 23(a). Prior to Dukes, attacking commonality was a waste of time, since any one common question of law and fact was sufficient to satisfy the element. After Dukes, one common question of law or fact is still sufficient to satisfy commonality, but the decision clarifies what commonality means in an important way. It is not whether a question common to all class members exists that is important; it is whether the question can be answered in a way that is common to all class members. In other words, commonality asks whether a question can be answered in a “single stroke†for all class members.
However, the fact that the commonality standard has been clarified and given some teeth does not guarantee victory for defendants. Recently, in Gray v. Hearst Communications, Inc., the Fourth Circuit applied the newly clarified standard and found that the issue of whether a phone book publisher had breached a distribution obligation in various contracts with advertisers was an issue that could be answered in “one stroke†with respect to all plaintiffs.
Expert testimony. The majority opinion in Dukes makes clear that a court should conduct a Daubert analysis to evaluate the reliability of expert testimony proffered by plaintiffs before relying on that testimony in finding that there is a way to decide a question of law or fact through common evidence. This is consistent with several court of appeals decisions issued shortly before Dukes, in particular the Seventh Circuit’s decision in American Honda Motor Co. v. Allen. After Dukes, defendants should have a much easier time convincing federal courts to scrutinize expert testimony offered by plaintiffs to show that a particular issue is susceptible to common proof.
Unfortunately, Dukes apparently hasn’t resolved this issue once and for all because the discussion in Dukes about requiring a Daubert analysis at the class certification phase is dicta. The Eighth Circuit has already exploited this shortcoming in In re Zurn Pex Plumbing Products Litigation. There, the court decided to apply a “tailored Daubert analysis†that does not require an exhaustive analysis of the Daubert factors or a ruling on the admissibility of the proferred expert’s testimony. So, while it appears that the federal courts will all give some scrutiny to expert testimony in conducting the rigorous analysis required under Rule 23, it is not clear that all federal courts will rule conclusively on the admissibility of expert testimony at the class certification stage.
Statistical evidence. In one sense, the decision is helpful to defendants merely by drawing attention to an issue that many lower courts had previously refused to give much analysis. Dukes is one of the first class action decisions to go into any significant detail in examining a plaintiff’s attempt to meet the class certification standards through statistical proof.
In many cases, especially in those outside the employment context, it may be possible for plaintiffs to distinguish the specific analysis in Dukes regarding the statistical proof relied on by the plaintiffs. However, even in cases where the facts may be distinguishable, Dukes illustrates several key ways in which statistical evidence offered in support of class certification can be attacked: 1) the expert cannot demonstrate to any reasonable degree of certainty that all or substantially all plaintiffs have been subjected to a common policy or practice by the defendant; 2) the statistics fail to show any causal connection between the alleged common practice and harm suffered by the class as a whole; 3) the statistics do not demonstrate harm to all class members as opposed to some subset or percentage of them; 4) the expert has failed to rule out other explanations for the harm allegedly suffered by class members other than the defendant’s conduct, policy or practice; and 5) the expert has not proposed a way of proving injury by common proof but has instead proposed a “trial by formula†where individual claims are aggregated and then sampled to determine a typical or average claim.
Part of the challenge for both sides in the fight over statistical evidence will be in helping the court to understand what the statistics purport to show. This problem is reflected in Justice Ginsberg’s dissenting opinion in Dukes, which accuses the majority of misunderstanding the statistical evidence in the first place. One would think that the majority and dissent could at least agree on the factual underpinnings of the statistical analysis being scrutinized, but they obviously did not. There is no simple solution to this problem, since most judges have not been trained as scientists or mathematicians. However, the problem underscores the importance of a clear factual presentation by both sides about the statistics at stake, what they show, what they don’t show, and how they relate to the facts and issues in the case.
Claims for monetary relief Under Rule 23(b)(2). The most uncontroversial portion of Dukes was the nine-to-zero holding that Rule 23(b)(2) could be used to certify individualized claims for monetary relief.  The majority held that claims for monetary relief under Rule 23(b)(2) cannot be certified unless they are “incidental†to injunctive or declaratory relief sought on behalf of class as a whole. Somewhat predictably, however, the hitch in the lower courts has quickly become deciding that claims for monetary relief are “incidental.â€Â One reading offered by Justice Scalia in the majority decision is that Rule 23(b)(2) “does not authorize class certification of monetary claims at all.â€Â Another second test for whether claims are “incidental†is that damages not allowed if they are individualized but may be allowed if they are uniform and either owed to all or none of the class members. However, even before Dukes, this approach had met with some skepticism. A third possibility is that the monetary relief is “incidental†if damages susceptible to objective determination without any exercise of judicial discretion. Expect disputes about what constitutes “incidental†monetary relief to continue into the future.
Another potential loophole in the Dukes analysis on claims for monetary relief under Rule 23(b)(2) is the possibility of issue certification of the liability phase of a class action. This is the analysis employed by the Eastern District of New York in United States v. The City of New York [12]. However, the decision in The City of New York may be limited to the specific context of pattern-and-practice disparate treatment cases, which require an initial phase in which the plaintiff must show a prima facie case of discrimination.
Parting thoughts
One key issue that I will be keeping my eye on over the coming year is how the state courts respond to this year’s Supreme Court developments, particularly Dukes. Some state courts have already cited Dukes approvingly in denying motions for class certification under state class action rules. However, many state courts applied relaxed class certification standards before Dukes. As the Smith decision makes clear, state courts are free to apply their own class certification rules and standards. Notably, shortly after issuing Dukes, the Court denied certiorari [13] in the case of Phillip Morris USA, Inc. v. Jackson (No. 10-735), in which the defendants had argued that the Louisiana Supreme Court’s decision to permit class certification in the face of individual questions of whether consumers relied on statements made by the tobacco industry violated due process. In my own state of Colorado, there are four cases involving class certification issues now fully briefed, argued, and awaiting a ruling from the state supreme court: Garcia v. Medved (09SC1080), Jackson v. Unocal (09SC668), BP America v. Patterson (10SC214), State Farm v Reyher (10SC77). Decisions in those four cases are expected this fall.
If the scrutiny with which state courts and federal courts review class certification begins to diverge as a result of post-Dukes developments, expect forum selection battles to come to the forefront in class action practice. Less restrictive class action standards in the state courts could also further encourage smaller, more limited classes, as plaintiffs’ lawyers attempt to keep the amount in controversy under the $5,000,000 limit under the Class Action Fairness Act.