• 26 August 2009

Class Certification in the Age of Aggregate Proof

Richard A. Nagareda - Vanderbilt University Law School

Posted in , , , , , , ,

Since the emergence of the modern class action in the 1966 amendments to the Federal Rules of Civil Procedure, controversy has attended the certification of litigation to proceed on a class-wide basis.  The addition to the Rules in 1998 of express authorization for appeals of class certification determinations short of final judgments stands as recognition that certification is no mere preliminary, procedural ruling.  When individual claims are unmarketable, class certification effectively determines whether the aggregate unit will be something considerably more valuable than the individual claims that form the constituent parts of the class.  For defendants, the class certification determination can be equally momentous.  With vanishingly rare exception, class certification sets the litigation on a path toward resolution by way of settlement, not full-fledged testing of the plaintiffs’ case by trial.  In terms of their real-world impact, class settlements can be quite significant, potentially involving dollar sums in the hundreds of millions or substantial restructuring of the defendant’s operations.

With so much riding on the class certification determination, one would have thought that procedural law would have arrived quickly at a clear and broadly shared understanding of the nature of that determination and the permissible parameters for inquiry by the court.  That, however, has not been so.  For decades after the adoption of the modern Rule 23, procedural law found itself occupied with what one might describe as a “first generation” of questions concerning class certification.  These first-generation questions centered on the meaning of the Supreme Court’s cryptic 1974 statement in Eisen v. Carlisle & Jacquelin that “nothing” in Rule 23 “gives a court any authority to conduct a preliminary inquiry into the merits of a suit in order to determine whether it may be maintained as a class action.”1

In a series of decisions in recent years, the federal appellate courts largely resolved the first-generation questions about what had come to be known as the Eisen rule.  True enough, Rule 23 does not require plaintiffs seeking to certify a class to satisfy a preliminary injunction-like standard cast in terms of the likelihood of success on the merits.  Still, the court may not accept a bare assertion in the class complaint that class treatment is appropriate.  Class certification is not a matter of mere pleading but, rather, of affirmative proof that the requirements stated in Rule 23 have indeed been satisfied.  The court must make a “definitive assessment” that these requirements have been met, even if that assessment entails the resolution of conflicting proof and happens to overlap with an issue—even a critical one—on the merits.2

These first-generation answers constitute a helpful step forward, but they have yielded still harder and, as yet, underexplored second-generation questions.  The time has come to expose the nature of these second-generation questions surrounding class certification and to offer a normative account of the proper allocation of institutional authority to resolve them.  A deep and increasingly important trend in contemporary class certification disputes concerns the degree to which ostensible battles over conflicting proof on the certification question are the stalking horse for something else:  underlying disputes that often have little to do with the proof or the facts and everything to do with the proper meaning of governing law.  A world in which the class certification determination effectively gauges the on-the-ground impact of the class action device is a world in which institutional questions about who should make that determination, and how that determination should be made, loom large.

All of this may sound new.  There are, however, deeper roots for what has emerged today at the forefront of class certification.  Writing in the gendered language commonplace in 1897, Oliver Wendell Holmes famously ventured that, in legal studies, “the black-letter man may be the man of the present, but the man of the future is the man of statistics and the master of economics.”3  So it is, more than a century later, that the major unanswered questions surrounding class certification center upon its interaction with areas of what one might call “Holmesian law”—bodies of substantive law in which statistical or economic analysis is invoked to play a significant role in legal doctrine.

In keeping with Holmes’s prediction, the flashpoints today over class certification concern the role of aggregate proof of a statistical or economic nature.  By “aggregate proof,” I refer to evidence—characteristically, in the form of expert submissions involving sophisticated statistical or economic analysis—that presumes a view of the proposed class in the aggregate.  From that vantage point, aggregate proof then seeks to trigger the application of substantive doctrine in such a way as to suggest a common, class-wide wrong attributable to the defendant.  The hard question concerns the propriety of this use of aggregate proof to conceptualize all members of the proposed class as a cohesive unit—as the victims of the same wrong under governing law, rather than a series of individualized wrongs ill-suited for class treatment.

Appellate courts faced with motions for class certification in two of the most closely watched, high-stakes class actions in recent decades—one against the tobacco industry under the Racketeer Influenced and Corrupt Organizations Act (RICO),4 the other against Wal-Mart, Inc. under Title VII of the Civil Rights Act of 19645—diverged over the treatment of aggregate proof.  The proper role of aggregate proof in class certification determinations is far from a mere technicality.  The desired effect of aggregate proof is considerable—indeed, well-nigh decisive—on the class certification question.  If everyone in the proposed class is, in some sense, the victim of the same wrong (though, perhaps, to varying degrees), then it would seem straightforward for the court to recognize that commonality by way of class certification.  The impulse is for the scope of the lawsuit to conform to the scope of the aggregate proof—for the proposed RICO class to encompass all smokers of light cigarettes across the country and for the proposed class against Wal-Mart to encompass the company’s entire U.S. operations.

Cast in their strongest terms, arguments for class certification premised on aggregate proof seek to tap into the core justification for the class action as the crucial procedural tool to forge into a viable unit claims that would not be viable individually.  The implication invited by class counsel is that only by taking an aggregate, class-wide perspective does the wrong allegedly committed by the defendant come into focus.  The ultimate factfinder would be entitled to disbelieve the plaintiffs’ aggregate proof, just as it might disbelieve any other form of evidence.  But that recognition—so certification proponents contend—is all the more reason for a court not to abort consideration from a class-wide perspective by withholding class treatment in the first place.  In the parlance of Rule 23, aggregate proof inherently frames “questions” that are “common” across the posited aggregate unit, because such proof takes that unit as its starting point and then seeks to trigger substantive doctrine that characterizes all class members as the victims of a common wrong.

At the same time, arguments for class certification premised on aggregate proof exhibit a deeply troubling circularity.  On a more skeptical view, such arguments amount to the justification of aggregation by reference to evidence that—at least, as a matter of economic or statistical methodology—presupposes the aggregate unit whose legitimacy the court is to determine.  If a cohesive class can be created through such savvy crafting of the evidence, then there would seem to be little limit to class certification in our modern world of increasingly sophisticated aggregate proof.  The law would run a considerable risk of unleashing the settlement-inducing capacity of class certification based simply upon the say-so of one side.  Yet the status of class treatment as the exception, not the norm, for civil litigation suggests strongly that one side’s procedural preference alone cannot be determinative.

At this early second-generation stage, then, the law of class certification finds itself seemingly confronted with an all-or-nothing choice:  Either the scope of aggregate procedure must follow more or less automatically from the framing of admissible proof along the lines of the proposed aggregate unit or class certification must fail routinely on account of the circularity problem.  I resist this choice by explaining how the seemingly stark alternatives posited for class certification today stem from an understandable but mistaken premise.

The hard questions surrounding class certification today are—contrary to conventional wisdom—only superficially questions of fact, conflicting evidence, and dueling expert witnesses.  Properly understood, aggregate proof frequently offers not so much a contested account of the facts that bear on class certification but, more fundamentally, an implicit demand for a new and often controversial conception of the substantive law that governs the litigation at hand.  The real concern about aggregate proof in class certification lies in its threat “to conform the law to the proof.”6  The leap from aggregate proof to legal doctrine is precisely the point on which courts should focus today in the posture of class certification.

Three related points emerge from this fresh conceptualization of class certification:

  1. Courts should be more transparent about the precise nature of the dispute in contested class certifications. A major part of the problem today is that, oftentimes, courts seemingly do not even realize that contested class certifications center upon contested accounts of governing law—accounts that only superficially take the form of dueling expert submissions;

  3. Recognition of the interplay between aggregate proof and governing law informs the standard of review for class certification rulings. The analysis here highlights considerable room for appellate oversight of class certification determinations, with the appellate courts cast in their familiar role of de novo review to ascertain the proper account of governing law, and not in a deferential role to review discretionary, trial-level determinations as to factual or evidentiary matters; and

  5. This perspective reorients the class certification determination in institutional terms. This third point is the logical corollary of the first, which clarifies the legal character of many class certification disputes today. The institutional relationship that usually matters in contested class certifications, I argue, is not so much the one between court and jury but, rather, that between court and legislature. When aggregate proof offers not merely a contested account of the facts but, at bottom, a contested account of governing law, the court should be concerned not with intrusion upon the jury’s role in the event of trial but, instead, with the degree of lawmaking power that the court properly may wield relative to the legislature in the particular area of law at issue. This is not to suggest that class actions—any more or less than conventional, individual lawsuits—cannot serve as vehicles for change in legal doctrine. It is simply to say that the proposed class-wide nature of the litigation should exert no independent weight in arguments for such change. As in constitutional adjudication, one might say that it is “emphatically the province and duty” of the court “to say what the law is”7 when the answer to that question will determine whether the proposed class is cohesive.

Recognition of the law-declaring dimension of the certification inquiry does not point uniformly for or against class certification when aggregate proof is in play.  Rather, class certification appropriately admits of differences informed by the proper conceptualization of applicable law in a given case.  When social science has assumed the mantle of legal doctrine—in, say, antitrust or securities law—the resolution of competing expert submissions cast in social science terms will tend, quite properly, to gravitate in the direction of judge over jury and of law over fact.  Areas of law already infused with social science, however, are not the only terrain for class actions.  What the courts are now seeing in such contexts as RICO and employment discrimination amounts to an effort to invite a similar kind of infusion, one sometimes reliant not on economics alone but also on statistical analysis informed by such disciplines as sociology.  Here, courts are likely to encounter more difficulty in discerning when dueling expert submissions on class certification are really dueling over the meaning of governing law, precisely because the integration of legal doctrine and social science is still a tentative, contested enterprise in these areas.

The approach offered here does not authorize courts to reach out to decide legal questions unrelated to the application of Rule 23 requirements.  A question concerning the proper meaning of governing law matters in the class certification context only insofar as the answer has the potential to reveal fatal dissimilarities within the proposed class.  When the concern about the proposed class is not that it exhibits some fatal dissimilarity but, rather, a fatal similarity—a failure of proof as to an element of the plaintiffs’ cause of action—courts should engage that question as a matter of summary judgment, not class certification.  Here, too, second-generation case law exhibits confusion, with some courts overreaching in the class certification inquiry to the displacement of the domain for summary judgment—the motion that rightly polices the institutional relationship between the court and the factfinder at trial.

Law in a given area, moreover, is certainly not stuck forever with its present-day content.  But, as law reform efforts proceed apace, spurred by new insights from social science, courts in the posture of class certification must remain attentive to matters of institutional role—to the line between judicial interpretation and doctrinal changes appropriately left for legislative determination.  Interpretation of governing law should not occur in a manner oblivious to its implications for aggregate procedure.  But it should occur both transparently and self-consciously.

Looking back in procedural history, one might say that this analysis of aggregate proof in class certification offers a cautionary rejoinder to aspirations for the creation of a genuinely transsubstantive body of procedural rules.  What we are witnessing today, not merely in class certification but across the spectrum of major pretrial procedural rulings, is an emerging effort to bring the system of notice pleading embraced in the 1938 overhaul of the Federal Rules into line with the on-the-ground reality of civil litigation today:  a world dominated by settlement, in which the pre-trial phase effectively is the trial.  This reality underscores all the more the practical desirability of law clarification—and, if necessary, law correction via de novo appeal—before class certification precipitates settlement.  Before that occurs, it is incumbent on the courts to pinpoint and to resolve forthrightly open questions of law that bear upon the cohesiveness of the proposed class.dingbat


Copyright © 2009 by New York University Law Review

Richard A. Nagareda is Professor of Law and Director of the Cecil D. Branstetter Litigation & Dispute Resolution Program at Vanderbilt University Law School. 

This Editorial is based on the following full-length Article: Richard A. Nagareda, Class Certification in the Age of Aggregate Proof, 84 N.Y.U. L. REV. 97 (2009).

  1. 417 U.S. 156, 177 (1974).
  2. In re Initial Public Offerings Sec. Litig., 471 F.3d 24, 41 (2d Cir. 2006).
  3. O.W. Holmes, The Path of the Law, 10 HARV. L. REV. 457, 469 (1897).
  4. See McLaughlin v. Am. Tobacco Co., 522 F.3d 215 (2d Cir. 2008).
  5. See Dukes v. Wal-Mart, Inc., 509 F.3d 1168 (9th Cir. 2007), reh’g en banc granted, Nos. 04-16688, 04-16720, 2009 WL 365818 (9th Cir. Feb. 13, 2009).
  6. McLaughlin, 522 F.3d at 220.
  7. Marbury v. Madison, 5 U.S. (1 Cranch) 137, 177 (1803).

Post a Comment (all fields are required)

You must be logged in to post a comment.