Recent Trends In Rule 23 Class Certification Expert Analyses

Tuesday, September 1, 2009 - 01:00

Almost a decade into the new millennium, the right of parties to a rigorous class-certification contest is increasingly entrenched in most federal circuits and many states, and indeed is appearing in other countries, such as Canada. It is also spreading to new causes of action. For some such causes of action, the new rulings might well be game changers, and in other cases the rulings at least move the certification goal posts. Legal teams will be looking at the class certification strategies and analytical tools deployed in a wide variety of causes of action, in case an approach taken in employment class actions (for example) might be germane to antitrust. In doing so, they will evaluate which questions can be addressed by experts that provide rigor at the certification stage but remain separate from the merits.

The Elements Of A Class Certification Analysis

In the past, courts have sometimes granted class certification based on little more than the pleadings.As a result, plaintiffs could realistically succeed so long as the opinions of their experts were "not fatally flawed."This position may no longer be acceptable. As of April 2009, six of the 12 circuits have made class certification rulings requiring "rigorous" analysis. These include the Second Circuit (in 2006), the Third Circuit (in 2008), the Fourth Circuit (in 2004), the Fifth Circuit (in 2007), the Seventh Circuit (in 2001), and the Eleventh Circuit (in 2004). A growing consensus of federal appellate courts now requires a stricter legal standard to govern requests for class certification.

The class certification hearing is where the court examines the case facts in light of the requirements of Federal Rule of Civil Procedure 23. In the federal courts and in many state courts, this examination goes beyond the legal pleadings of the case; it includes a "rigorous" analysis that is increasingly likely to include the testimony of the opposing parties' experts as to whether the Rule 23 requirements can be met. While this rigorous analysis is intended not to pre-judge the ultimate merits of the case, the analysis is permitted to touch on the merits to the extent that is required to rule on the existence of a proper class.

Rules 23(a) and 23(b) set forth a total of seven criteria that must be met for the successful certification of a class. As a first-step Rule 23(a)'s four well-known criteria must be met. These consist of numerosity; common questions of law and fact; typicality; and adequacy of representation. If satisfied that the Rule 23(a) requirements are met, the plaintiff must also show that the lawsuit qualifies under Rule 23(b), owing to either: the risk of inconsistent or impaired adjudication; the defendant acted on grounds generally applicable to the class; or common questions of law or fact predominate and class resolution is superior to other available methods.

This final, two-legged criterion of predominance and superiority is often fact-dependant and therefore one of the issues likely to benefit from expert opinion. We will focus on this aspect in what follows.


One line of analysis related to superiority is the question as to whether, in principle, it will be possible to generate the information set necessary to adjudicate the matter as a class action through common proof. The lawyers rely upon experts to perform a rigorous analysis from the point of view of the class-wide aspect without straying too far into "merits."

Plaintiffs may seek to show that the requisite knowledge is easily obtained without undue discovery burdens. Plaintiffs may employ a sampling or statistical expert to show that a small sample will suffice to establish loss causation and class-wide damages. In short, plaintiffs may approach the "economy" aspect of certification by emphasizing a "small-sample" view of the world: the necessary facts are knowable, and available in a short time frame at reasonable cost by recourse to a small sample. Moreover, the information being acquired through the small sample consists of a few key facts that can be reliably obtained via the sampling process, whether that involves human subjects' recall of past events; billing, medical, personnel, or other official records; or other case-relevant objective information.

In response, the defendants may engage an expert to evaluate whether the optimistic small-sample view of the world put forward by plaintiffs and their experts is realistic. An expert may form an opinion that much more extensive scientific sampling of the population of potential class members is indicated. Plaintiff's proposed reliance on a small sample may have been driven by approaching the problem as one of so-called "Bernoulli trials" in which a homogeneous population is sampled with respective to a single, readily determined yes-or-no condition. The textbook analogy is one of drawing a sample of red and white balls from an urn, in sufficient quantity to estimate the proportion of balls that are red and the proportion that are white. Defendants are likely to argue that class certification is not about how many balls are white or red; certification hinges on whether it is possible to place the white and red balls into separate urns, without having to inspect each ball individually.

Defendants may envision a world in which there is not one homogeneous population, but many distinct sub-populations, each heterogeneous within itself and certainly different than other sub-populations. This hypothesis will demand a more elaborate and extensive sampling approach. And the problem would be compounded if the population naturally decomposes into multiple distinct populations, but there is no sampling frame available that identifies in advance which subject or case file belongs in which sub-population. Moreover, while it may be easy to distinguish a white ball from a red one in bright daylight, the information needed in real-world litigation may be much more problematical to extract and verify.


If and when the court authorizes discovery, the day will come when plaintiffs and defendants have developed data drawn from their samples. This information is used to determine whether each potential class member's situation is sufficiently similar to make it desirable to proceed collectively.

Perhaps the starkest outcome might be clear and convincing evidence that there is a conflict of interest within the proposed class.1The reader will appreciate how difficult it could be to persuade the court to proceed in such a case. Class actions do indeed arise where an "average" amount of harm disguises the fact that some putative class members were actually better off as a result of defendants' actions.

Another scenario arises in class certification challenges, when the evidence supports the conclusion that there is no homogeneous class, but rather multiple subclasses.In such cases a plaintiffs' expert might propose statistical techniques to differentiate the financial award or other intervention appropriate to each subclass. But too much heterogeneity or variation within the subclasses may render this remedy inadequate.

Finally, in some cases it is possible to formulate rigorous tests of the hypothesis that a representative class member (whether a named plaintiff or not) is sufficiently similar to other members of the proposed class. This should not be confused with the Rule 23(a) requirements, which address different issues about the representativeness (and commonality and typicality) of the case.

Evaluating The Expert Analysis

It was in 1966 when revisions to Rule 23 first accommodated monetary damages to unnamed members of the proposed class, and it was in 1966, too, that the Second Circuit issued its Eisen ruling, that henceforth was invoked to limit fact-finding and expert analysis at the certification stage. Over time, the courts spread out on a continuum between an Eisen -inspired no-merits stance, eschewing expert testimony and relying solely upon pleadings; to a middle ground in which the court considers primarily the plaintiffs' factual presentation and expert analyses (since the plaintiff carries the burden of proof); and finally to a rigorous view of both plaintiffs' and defendants' expert analyses, wherein the court will likely have to consider the appropriate weight to assign experts on both sides.

Whether or not the parties seek formally to exclude the other party's expert testimony, the reliability of all such testimony will be weighed by many courts today. Accordingly, there are pro's and con's to a formal Daubert -style motion to exclude. Of course, the original intent of Daubert and related rulings was to shield juries from so-called "junk science." There being no jury at the certification hearing, legal teams make expert challenges when they wish to drill-down more deeply into the methodological issues in the expert reports.

This may come about in part as a matter of timing, depending upon whether the experts had the opportunity to rebut opposing experts in either their affirmative or rebuttal report. Or the legal teams may seek additional, focused opportunities to rebut. This might arise when the expert analyses are extremely technical.

It might also arise in the exact opposite context. For example, some defendants will characterize a simple plaintiff expert report as a "trust-me" report. The plaintiff's expert may have submitted a report or declaration simply enumerating the expert's qualifications along with a brief discussion that a small sample addressing two or three key questions can easily and inexpensively be conducted. Defendants will seek to show that this is completely inadequate as a "rigorous" determination of certification issues.

Such Daubert challenges have been known to succeed, and where they have not led to the exclusion of all or part of an expert's opinion, they may still undermine the veracity of the expert. On the other hand, Daubert challenges are sometimes welcomed by a legal team, who believe their expert has done fine work. In such a case, the Daubert challenge may result in focusing judicial scrutiny on the work and confirming the rigor of the analysis. Even when the Daubert challenge exposes flaws in an expert's work, the Daubert challenge may convince the court that it will be possible to correct these errors after certification.2


In summary, there appears to be growing momentum toward more rigorous class certification hearings. The question for the practicing class action attorney, therefore, is whether a handful of new rulings will appreciably change either the manner or the success rate in achieving or contesting certification. Time will tell, but legal teams may feel compelled to litigate as though the trend is firmly established .

Given this trend, experts have an ever-increasing role to play in Rule 23 hearings, and many times a key role at that. The nature of the expert analysis will naturally vary with the fact pattern being analyzed. A white paper available from the author delves into some of the specific questions that may be suitable for expert analysis in greater detail than is possible here. 1 Rodriguez-Feliciano v. Puerto Rico Electric Power Authority, KDP 2005-1591 (P.R. Super. Ct. June 28, 2007).

2 In re Wal-Mart Stores, Inc. Wage and Hour Litig., 2008 WL 413749 (N. D. Cal.).

Dr. David Gulley is an economist with over 30 years of experience. He is a seasoned testifier and has served as a court-appointed expert. Dr. Gulley has been involved in many noteworthy and high profile assignments. He has been retained in varied causes of action including antitrust, breach-of-duty, employment, civil and criminal securities fraud, bank and financial markets investigations and litigation, and bankruptcy-related litigation, among others.

Please email the author at with questions about this article.