By Makan Delrahim (Assistant Attorney General, Antitrust Division, U.S. Department of Justice)1
I. Introduction
The mission of the Antitrust Division of the U.S. Department of Justice is to promote and protect competition in the American economy. As mergers, collusive agreements, and exclusionary conduct cross national borders, the Antitrust Division must keep an increasingly close eye on developments in competition law around the world in order fully to discharge its responsibilities. Economic globalization has also led us to work more closely with other competition agencies and to deal more frequently with the competition-law regimes of other nations.
The importance of dialogue with other competition agencies and our shared interest in the appropriate enforcement of competition laws has led the Antitrust Division to engage internationally to promote cooperation and convergence in the development of competition law. We work bilaterally with our counterparts around the world and through multilateral organizations such as the Competition Committee of the Organization for Economic Cooperation and Development (“OECD”) and the International Competition Network (“ICN”) to facilitate international discussions of important issues common to competition law enforcement. Over the past few years, in addition to cooperating with our international counterparts on many of the cases, we have participated in an international effort to further common understandings on a variety of substantive issues, including competition analysis in digital industries, leniency programs, algorithms and collusion, and a number of questions associated with merger analysis and remedy selection. Our engagement with foreign competition law is not limited to substantive law, however; procedural devices are paramount to how competition-law regimes are implemented. For example, we are proud to be the impetus for the 2019 Framework on Competition Agency Procedures (“CAP”), designed to strengthen and promote due process in competition law proceedings globally. Participants commit to a set of substantive due process principles and, though not legally binding, CAP includes accountability and review mechanisms to ensure meaningful compliance.
Of course, developments in competition law, both substantive and procedural, can be driven as much by courts as by competition agencies, particularly in countries that allow for private antitrust enforcement in the form of class actions. A court’s decision on class certification can, for example, determine as a practical matter whether certain competition-law claims proceed at all.
The upcoming decision of the UK Supreme Court in Merricks v. Mastercard is of interest to competition enforcers around the world for this reason. The case involves novel questions on the proper approach to certification of an opt-out collective action — akin to a class action in the United States — brought by indirect purchasers. Whether the UK Supreme Court properly aligns the incentives for these new private collective-action proceedings2 can affect consumers everywhere; its decision may also have important effects on U.S. companies like Mastercard and their opportunities to compete globally.
The two central issues before the UK Supreme Court in Merricks are:
(a) The “legal test for certification” of a collective proceeding by indirect purchasers, including what the class representative is required to show to offer “a realistic prospect of establishing loss on a class-wide basis,” and
(b) The “approach to distribution” of damages among individual purchasers, including whether it was premature for the Competition Appeal Tribunal to consider distribution of damages.
This Essay aims to share the United States’ experiences confronting similar questions. The use of comparative material, which contextualizes experiences within each legal system, is not itself novel.3 Targeted insights from U.S. courts deciding whether to certify indirect purchasers, however, are not readily gleaned from existing commentary or scholarship. We therefore hope this Essay will be informative for the international antitrust community and the UK Supreme Court.
The Essay begins with a brief description of the procedural history of Merricks and acknowledges key differences between the U.S. and UK legal regimes. It then explores how U.S courts consider class-certification questions in cases brought by indirect purchasers, organized to mirror the two questions before the UK Supreme Court in Merricks: Part V discusses the standard of proof and Part VI the distribution of damages.
As described below, U.S. courts subject proposed methodologies to estimate pass-on proffered by indirect-purchaser plaintiffs to a “rigorous analysis” at the class-certification stage, which often requires some consideration of the merits of the plaintiffs’ underlying claims. After the U.S. Supreme Court’s holding in Comcast v. Behrend, 569 U.S. 27 (2013), some courts have engaged even more closely with the plaintiffs’ proposed expert methodologies, rejecting class certification if the plaintiffs cannot show they can offer common evidence that the defendant’s alleged anticompetitive conduct caused loss to all or almost all putative class members. Of relevance to the issues in Merricks, U.S. courts sometimes cite the unmanageability of determining and distributing individual damages as a reason to deny class certification, especially in cases involving a lack of reliable data and other complexities.
In making determinations on the standard of proof, U.S. courts recognize that improperly low standards for class certification create a “risk of ‘in terrorem’ settlements,” that is, settlements agreed to in order to avoid the “small chance of a devastating loss” from potentially large aggregated damages rather than due to the likelihood that plaintiffs will succeed in establishing liability. See AT&T Mobility LLC v. Concepcion, 563 U.S. 333, 350 (2011).
II. The Proceedings in Merricks
In 2016, Mr. Walter Hugh Merricks CBE applied before the Competition Appeal Tribunal to bring collective proceedings against Mastercard on behalf of 46 million consumers. He alleged Mastercard’s decision prior to 2008 to set an interchange fee applicable to cross-border payments — which the European Commission held in 2007 to be a violation of Article 101 of the Treaty on the Functioning of the European Union — caused merchants to pay higher fees to acquirers, i.e. banks or financial institutions that provide payment services to merchants. This overcharge, Mr. Merricks contends, was subsequently passed on by merchants to the consumers comprising the putative class.
The Tribunal rejected Mr. Merricks’ application after a hearing in July 2017. It held his claims were unsuitable to be brought in collective proceedings as two issues were not common: (1) the degree of pass-through from merchants to claimants and (2) the amount each claimant spent at each merchant.4 The Tribunal also found there was no “plausible way” to distribute damages to each individual in a way that estimates the loss s/he suffered.5 In reaching its conclusion, the Tribunal emphasized that although in theory “calculation of global loss through a weighted average pass-through … is methodologically sound,” applying that methodology “across virtually the entire UK retail sector over a period of 16 years is a hugely complex exercise requiring access to a wide range of data.”6 The Tribunal was unpersuaded that such data were available.7
Upon appeal, the Court of Appeal held the Tribunal had “demanded too much” from Mr. Merricks to show pass-on at the certification stage, and that the Tribunal’s decision to consider a proposed method of distribution was “both premature and wrong.”8 The Court of Appeal expressed it was not necessary at th[is] stage for the proposed representative to be able to produce all of that evidence [for assessing the level of pass-on to the represented class], still less to enter into a detailed debate about its probative value.”9 In the Court of Appeal’s view, the proposed methodology needs only be “capable of or [offer] a realistic prospect of establishing loss to the class as a whole.”10 The Court of Appeal then remitted the application in April 2019 to the Tribunal for reconsideration.
In July 2019, the UK Supreme Court granted Mastercard permission to appeal the Court of Appeal’s judgment.
III. Comparative Study of Procedures in the UK and the U.S.
As a threshold matter, we acknowledge the legal systems of the U.S. and the UK have both similarities and differences, including as regards their procedures for class actions. Recognizing parallels and material distinctions between the two systems provides a broader context for the often fact-specific cases and may render the analogy between class-certification law in the U.S. and in the UK more illuminating. For example, plaintiffs in the U.S. can seek discovery prior to the class-certification stage, which can inform the court’s analysis of whether a case is appropriate for class proceedings that would bind all putative plaintiffs at future stages of litigation, including summary judgment and trial.
The substantive law on whether indirect purchasers can sue for damages in antitrust suits differs as well. Unlike UK competition law, U.S. federal antitrust law generally does not permit recovery by indirect purchasers, following the Supreme Court’s decision in Illinois Brick Co. v. Illinois, 431 U.S. 720 (1977). Numerous U.S. states, however, have declined to follow Illinois Brick through legislation and by common law. Approximately two-thirds of U.S. states and the District of Columbia currently permit the use of pass-on analysis to apportion damages in federal or state courts based on violations of state antitrust law.11
An indirect purchaser suing under state antitrust law nonetheless may face certain hurdles: Several courts have held the indirect purchasers in the cases before them lacked “antitrust standing” to sue because their injury was too derivative or remote under the multifactor test from Associated General Contractors of California, Inc. v. California State Council of Carpenters, 459 U.S. 519, 535 n.31 (1983). The AGC factors include: “(1) whether the plaintiff is a consumer or competitor in the allegedly restrained market; (2) whether the injury alleged is a direct, firsthand impact of the restraint alleged; (3) whether there are more directly injured appellants with motivation to sue; (4) whether the damages claims are speculative; and (5) whether the plaintiff’s claims risk duplicative recoveries and would require a complex apportionment of damages.”12 Based on these concerns, courts have dismissed state-law antitrust claims involving tying of credit card and debit card services at the “motion to dismiss” stage — comparable to a strike-out in the UK — before the cases even reach the class-certification stage.13
IV. Legal Standards for Class Certification in the United States
The first issue before the UK Supreme Court in Merricks is what legal standard a court should apply in certifying a collective proceeding by indirect purchasers. The Court of Appeal held the Tribunal applied an unduly stringent test in evaluating the evidence and expert methodology before it; in the Court of Appeal’s view, Mr. Merricks needs only demonstrate “a real prospect of success” at the certification stage.14
Within the United States, courts evaluate questions on the standards for class certification by looking to Federal Rule of Civil Procedure 23 or the corresponding state rule. These rules set forth criteria to determine when the class-action device is appropriate, and a party seeking to bring a class action “‘must affirmatively demonstrate his compliance’ with [Federal Rule of Civil Procedure] 23.”15
Under Rule 23, a court cannot certify a class action unless it finds the plaintiffs have satisfied the four prerequisites of Rule 23(a) — often referred to as numerosity, commonality, typicality, and adequacy of representation — as well as one of the requirements of Rule 23(b). Of relevance here, Rule 23(b)(3) provides that in suits for damages actions, class certification is appropriate only if “the court finds that the questions of law or fact common to class members predominate over any questions affecting only individual members …” Certification of indirect-purchaser classes often turns on this “predominance” inquiry.
In deciding whether to certify a class, a U.S. court must conduct a “rigorous analysis” of requirements in Rule 23, which will inevitably “entail overlap with the merits of the plaintiff’s underlying claim,” Comcast, 569 U.S. at 33 (citation and internal quotation omitted). Questions on the merits, however, may be considered at this stage “only to the extent [] that they are relevant to determining whether Rule 23 prerequisites for class certification are satisfied.” Amgen Inc. v. Conn. Ret. Plans & Tr. Funds, 568 U.S. 455, 466 (2013).
To be sure, applying these legal standards in practice often turns on the specific facts and evidence of the case. Class-certification proceedings typically involve extensive evidentiary presentations by the parties, and courts assess whether plaintiffs meet their burden of proving each requirement for certification by a “preponderance of the evidence” — that is, whether “the evidence more likely than not establishes each fact necessary to meet the requirements of Rule 23.”16 To make this determination, U.S. courts analyze not only whether plaintiffs have made a sufficient evidentiary showing on their own, but also whether evidence and counter-arguments from defendants defeat certification.
V. Rigorous Analysis of Pass-on and Damages in the United States
In U.S. antitrust class-action suits brought by indirect purchasers, how plaintiffs establish the aggregate amount of pass-on damages — a question at the heart of Merricks — is secondary to a threshold question for certification on whether there is predominance of common proof over individual issues on the question of class-wide impact. Specifically, to assess whether plaintiffs meet the predominance standard, U.S. courts ask whether and how plaintiffs intend to show:
(a) Class-wide proof of “antitrust impact” or “antitrust injury” — that is, whether all or nearly all class members were in fact injured from the alleged anticompetitive conduct;
(b) Class-wide proof of damages; and
(c) For each of these inquiries, that class-wide proof of impact and damages predominate over individualized issues, such that class treatment of damages claims is appropriate.
Although there is no explicit requirement of predominance under UK law, the Tribunal and the Court of Appeal considered related inquiries in the proceedings in Merricks. The Canadian Supreme Court opinion on which they rely,17 Pro-Sys Consultants Ltd v. Microsoft Corp. [2013] SC 57, requires indirect-purchaser plaintiffs at the class-certification stage to show “a realistic prospect of establishing loss on a class-wide basis so that, if the overcharge is eventually established at the trial of the common issues, there is a means by which to demonstrate that it is common to the class….”18 As the Court of Appeal noted, Pro-Sys relies in turn on the U.S. conception of common impact:
“The requirement at the certification stage is not that the [expert] methodology quantify the damages in question; rather, the critical element that the methodology must establish is the ability to prove ‘common impact’, as described in the U.S. antitrust case of In Re: Linerboard Antitrust Litigation, 305 F.3d 145 (3rd Cir. 2002). That is, plaintiffs must demonstrate that ‘sufficient proof [is] available, for use at trial, to prove antitrust impact common to all the members of the class’ (ibid., at p. 155). It is not necessary at the certification stage that the methodology establish the actual loss to the class, as long as the plaintiff has demonstrated that there is a methodology capable of doing so.”19
This conception of common impact, including what a methodology capable of establishing the actual loss to the class must show, underpins the relevant reasoning in Pro-Sys and hence informs “the quality of the expert evidence needed” in Merricks “to establish loss on a class-wide basis as an issue common to the class.”20 The predominance inquiry under U.S. law incorporates these questions and also requires courts to conduct a rigorous analysis of whether common proof of impact predominates over individual issues among putative class members.
Moreover, following the U.S. Supreme Court’s instruction in Comcast, class-wide proof of damages must be “tied” to the class-wide proof of impact. In Comcast, the putative class of direct purchasers, comprising over two million current and former Comcast subscribers, brought antitrust claims against the cable-television services company for its “clustering scheme,” which concentrated operations within particular regions. In holding the plaintiffs could not meet the predominance standard of Rule 23, the Supreme Court first emphasized the importance of conducting a “rigorous analysis” which “will frequently entail overlap with the merits of the plaintiff’s underlying claim.”21 The district court and the court of appeals thus erred by refusing to decide whether the proposed “methodology [is] a just and reasonable inference or speculative” for fear of “inquir[ing] into the merits.”22 The Supreme Court then determined the plaintiffs did not meet the predominance standard for class certification because their expert’s calculation of damages — which assumed the validity of the plaintiffs’ four theories of antitrust impact rather than the validity of the single theory accepted by the district court — was not “tie[d]” or “attributable” to the one accepted theory of liability.23
A. Class-wide Proof of Impact
Any plaintiff bringing antitrust claims for damages in the United States must establish the fact of injury resulting from a violation of the antitrust laws.24 At the certification stage, then, U.S. courts conduct a “rigorous analysis” of whether this element of the plaintiffs’ claim — antitrust impact or injury — is common among class members, and whether such commonality predominates over individualized inquiries.
(i) Evidence of Pass-on of Injury
Problems with proving antitrust injury at the certification stage are typically more present in suits brought by indirect purchasers than in those brought by direct purchasers, because causation is more attenuated. Whether indirect purchasers suffered injury on a class-wide basis often depends on whether an alleged overcharge was “passed on” to the indirect purchasers through each chain of distribution by which the putative class members made purchases. As one court expressed:
“Courts have recognized that proof of injury in fact in indirect purchaser suits can be problematic since, notwithstanding economic theory, it cannot be presumed that intermediaries will, in fact, always pass through antitrust overcharges or that price increases by middlemen to ultimate consumers might not be attributable to upstream overcharges …” Ren v. Philip Morris Inc., 2002 WL 1839983, at *5 (Mich. Cir. Ct. 2002).
Consistent with the Tribunal’s view that Mr. Merricks cannot simply “address the problem of pass-through by submitting that the Tribunal can arrive at an aggregate award of damages, which would then be distributed to the class members,” U.S. courts typically require indirect-purchaser plaintiffs to show more than an estimate of aggregate damages to show pass-on and thus antitrust impact.25 The plaintiffs must introduce models from their economic expert demonstrating injury on a class-wide basis, accounting for variations among class members. The expert methodology must indicate class-wide that the indirect purchasers’ condition in a hypothetical “but-for” world — one without the alleged anticompetitive conduct — is better than their present condition.
Indeed, in deciding whether to certify an antitrust class action, U.S. courts have rejected methods of inferring class-wide injury from only aggregate damages and generalized statistical averaging; instead, direct proof is necessary.26 In other words, it is not sufficient for plaintiffs to proffer a methodology purporting to show that the average class member suffered injury based on an average overcharge from a statistical model and then assume that the average overcharge applied to all transactions made by indirect purchasers.
In addition, federal courts of appeal have generally endorsed the view that certification is inappropriate if a proposed class includes more than a de minimis number of uninjured plaintiffs because their presence defeats predominance.27
(ii) Rigorous Analysis of Class-wide Impact
The manners in which U.S. courts “rigorously analyze” common proof of impact, including pass-on, following the Comcast ruling can shed light on the present disagreement between the Tribunal and the Court of Appeal about the appropriate standard for reviewing Mr. Merricks’ proposed methodology. As described below, U.S. courts may also sometimes vary in what they require plaintiffs to show for class certification.
In some instances, courts have certified indirect-purchaser classes after conducting a rigorous analysis and determining the proffered expert methodology appropriately accounted for potential variations among class members. In re Nexium Antitrust Litig., 777 F.3d 9 (1st Cir. 2015) provides one example of the sort of rigor with which courts analyze plaintiffs’ proposed methodologies and engage with challenges from defendants. There, indirect purchasers brought an antitrust suit against drug manufacturers that allegedly delayed the launch of generic versions of heartburn medication.28 In considering defendants’ argument that more than a de minimis number of class members were uninjured, the court “conduct[ed] a detailed inquiry into the parties’ and experts’ economic analyses.”29 It determined the defendants’ challenges could not overcome the plaintiffs’ showing of predominance due to various logical and factual errors.30 The court then concluded the plaintiffs met their burden to show “the vast majority of class members were probably injured,” after reviewing whether prices to each of the allegedly uninjured groups would be lower in a but-for world absent the illegal conduct.31 The court’s inquiry in cases like Nexium parallels an important question in Merricks: As the Court of Appeal described, a “critical issue in deciding whether the proposed methodology is a suitable and effective means of calculating loss to the class is to determine whether it is necessary to prove at trial that each member of the proposed class has in fact suffered some loss due to the alleged infringement.”32
Courts in other instances have refused certification after conducting a rigorous analysis and finding the methodology for determining class-wide injury did not account for indirect purchasers that did not suffer injury — i.e. indirect purchasers to whom injury was not passed on. In In re Lithium Ion Batteries Antitrust Litig., 2018 WL 1156797 (N.D. Cal. Mar. 5, 2018), the district court refused to certify a class of indirect purchasers who alleged defendants engaged in a multi-year price-fixing conspiracy of lithium ion battery cells. The district court denied the request for class certification because the plaintiffs’ expert’s “estimate of 100% pass-through at each level of the supply chain does not adequately account for the effects of focal point pricing [set by sellers without overt communication], and therefore fails to yield reliable conclusions.”33 In the court’s view, by failing to address putative class members who purchased at focal point prices — and hence, according to defendants, suffered no overcharge attributable to the alleged anticompetitive conduct — the plaintiffs’ expert left “too much uncertainty as to whether pass-through can be estimated reliably at 100% as to retailers or distributors farther down the supply chain, and ultimately to the consumers who make up the proposed class.”34
Likewise, the court of appeals in In re Rail Freight Surcharge Antitrust Litigation, 725 F.3d 244, 253, 255 (D.C. Cir. 2013) characterized the “rigorous analysis” requirement in Comcast as requiring the court to undertake a “hard look” at the proffered expert methodology for class-wide antitrust impact. It then vacated class certification in a price-fixing antitrust case where, among other reasons, the plaintiffs’ expert methodology failed to filter out class members who suffered no damages.35
Certain district courts since Comcast have continued to apply a more relaxed standard of proof, however, creating some uncertainty about precisely how much a court should engage with an expert’s proposed methodologies on class-wide impact.36 There may be some tension between these rulings and the Supreme Court’s holding that a court must conduct a “rigorous analysis” of Rule 23(b)(3)’s predominance requirement, and we are aware of no federal court of appeals decision since Comcast that has adopted a more lenient certification test for analyzing methodologies on class-wide impact.
B. Class-wide Proof of Damages
In deciding whether indirect-purchaser plaintiffs should proceed as a class, a U.S. court also considers whether computation of damages for putative members of the class is susceptible to common proof. In certain circumstances, courts rely on plaintiff measures of aggregate damages to help meet this burden.37 Indeed, the class-action mechanism itself contemplates the use of aggregate damage calculations, which is well-established in federal courts.38
Still, proving aggregate damages alone is insufficient for class certification because courts must conduct a “rigorous analysis” into the question of whether common proof of such damages predominates over individualized questions of damages.39 To answer this question, courts analyze expert methodologies for calculating damages to individual class members, though such methodologies are often subject to a more relaxed standard than that required to show the fact of injury. “Calculations need not be exact,”40 and to make an initial showing, a “plausible” estimate of each class member’s damages will typically suffice as long as it does not rely on “speculation or guesswork.”41 The concern of U.S. courts with whether individual damages can be proven by common method is comparable to the Tribunal’s consideration in Merricks of whether there is a “plausible way” at the class-certification stage “of reaching even a very rough-and-ready approximation of the loss suffered by each individual claimant from the aggregate loss.”42
Further, although estimates of class-wide damages need not be proven with certainty, courts are becoming more willing to subject damages methodologies to greater scrutiny in order to confirm all or almost all class members are injured.43 In this regard, although class-wide impact and class-wide damages are two distinct inquiries, the analyses and supporting expert methodologies often overlap.
VI. Distribution of Damages
The second key issue before the Supreme Court in Merricks concerns the proper approach to distribution of damages: Specifically, whether the Tribunal was “both premature and wrong … to have refused certification by reference to the proposed method of distribution” because, as the Court of Appeal put it, “[d]istribution is a matter for the trial judge to consider following the making of an aggregate award.”44 In the Court of Appeal’s view, the Tribunal is not required at this stage to consider any “more than whether the claims are suitable for an aggregate award of damages which, by definition, does not include the assessment of individual loss.”45
Although U.S. courts do consider whether individual damages can be proven according to a common method across the whole class during the class-certification phase,46 they generally conceive of the “distribution” of aggregate damages as occurring in a later phase of litigation, such as after a determination of liability.47 In addition, courts rarely use “manageability” of the class action as a ground to deny certification under Rule 23.48
Some courts have nevertheless taken an approach more similar to the Tribunal’s and have been much more exacting in reviewing the plaintiffs’ proposed methods to distribute damages at the class-certification stage.49 Indeed, the availability of reliable data — which animated the Tribunal’s concerns in Merricks ([2017] CAT 16, [77]-[78]) — may be pertinent both to calculating individual damages and to distribution of those damages.50
The opinion by the Circuit Court of Michigan in Ren v. Philip Morris Inc., 2002 WL 1839983 (Mich. Cir. Ct. 2002) interprets a state-law rule on class certification and provides a helpful illustration of this principle. There, a proposed class of smokers sought to recover damages for the inflated cost of cigarettes due to price-fixing by cigarette manufacturers. Although the court found the plaintiffs’ expert had a valid methodology to establish class-wide impact, it held the plaintiffs failed to provide a method to determine damages for each consumer on a “formulaic basis.”51 The court noted that in some instances, “ascertainment of damages for each individual plaintiff … [is] amenable to formulaic treatment,” such as where “the number of products and number of purchases by any one plaintiff … is very limited, and the range of retail price a finite quantity” or “where the number of purchasers … are readily identifiable and reliable records for individual purchases exist.”52 It then compared cases which involved “computerized records of individual transactions” to the case before it, which implicated a “dizzying number of products, retail prices, individual transactions and purchasers over a multi-year period.”53 The court thus held the plaintiffs could not show predominance as it was unlikely that “any sort of ascertainment of individual damages can be made under some systematic or formulaic basis that avoids the necessity of individualized proofs…”54 The court further noted that these “considerations also would lead to the finding that the class action would not be manageable.”55
In contrast, the Court of Appeals of New Mexico certified a class of smokers on similar facts in Romero v. Philip Morris Inc., 109 P.3d 768 (N.M. Ct. App. 2005). In so doing, the court placed its faith in the jury to make a “just and reasonable estimate of the damage based on relevant data.”56 In its view, although “some minimum amount of individualized proof will be at the very least required for class members to receive any amount of damages,” the court “do[es] not see problems of such an intolerable or insurmountable character to cause [it] to pause at this stage and prevent certification on manageability grounds.”57 Instead, the court of appeals emphasized the district court has “numerous management tools at hand” to assist later in the “management and processing and or adjudication of the claims of individual class members.”58
To be sure, difficulties involved in distribution are heightened in complex antitrust cases, and courts tend to look more suspiciously upon proposed methods of distribution that would “require individual trials.” See In re Rail Freight, 934 F.3d at 625 (explaining the plaintiffs did not have a “winnowing mechanism” “short of full-blown, individual trials” to separate the uninjured from the injured, and noting that any questions presented in later individual challenges “would be … complex.”).59
Some jurisdictions therefore remain skeptical that distributing damages for individual class members can truly be separated from determining common antitrust injury during the class-certification phase. These courts reason similarly to the Tribunal in Merricks, which required at the certification stage a “practicable means for estimating the individual loss which can be used as the basis for distribution.”60
VII. Conclusion
Whether to certify classes of plaintiffs in antitrust suits can be a challenging question to resolve in any jurisdiction, and those challenges can be heightened where the harmed purchasers are indirect. In complex markets where transactions are affected by numerous factors, tracing and measuring injury after it has occurred is often a central point of dispute between the parties in class certification. Of relevance to the issues in Merricks, two trends in the United States may be of particular note to the UK Supreme Court:
(a) After the U.S. Supreme Court’s holding in Comcast, U.S. courts are engaging more deeply with the plaintiffs’ expert methodologies and, just as importantly, with the defendants’ challenges to these methodologies. Although the effects of Comcast continue to percolate through lower courts, courts have been willing to deny class certification to indirect purchasers if their proffered methodologies are unsound or fail to offer common evidence of class-wide antitrust harms that predominates over individual issues of injury. Of course, as noted in the cases cited above in which courts have certified indirect purchaser classes, the evidentiary complexities associated with pass-on analysis have been and can be surmounted in U.S. litigation.61
(b) Calculation of individual damages and administrability in distribution may also present a hurdle for some indirect plaintiffs at the class certification stage. Although many U.S. courts continue to conceive of distribution for a later stage of trial, some have denied class certification precisely because of a lack of reliable data for calculating individual overcharges and of complexities in distribution. For these skeptical courts, bifurcating impact and distribution simply delays the need eventually to adjudicate these individualized claims.
The UK Supreme Court’s decision in Merricks will affect not only the UK, but all nations that share the UK’s goal in promoting competition across the globe. We at the Antitrust Division therefore hope the new procedural option is implemented with care, balancing the many interests at stake in class proceedings and benefiting from the experience of nations that have considered similar questions.
Click here for a PDF version of the article
1 Assistant Attorney General, Antitrust Division, Department of Justice. I would like to thank Rene Augustine (Deputy Assistant Attorney General), Michael Murray (Deputy Assistant Attorney General), William Rinner (Chief of Staff), Elyse Dorsey (Counsel to the Assistant Attorney General), Lynda Marshall (Chief of the International Section), Daniel Haar (Chief of the Appellate Section), Caldwell Harrop (Assistant Chief of the International Section), and Matthew Mandelberg (Attorney-Advisor), for their contributions to this Essay. I also would like to especially thank Yixi (Cecilia) Cheng, James F. Rill Fellow (2019-2021), for her contributions and efforts on this project.
2 See Consumer Rights Act 2015, c. 15 (Eng.).
3 The U.S. Supreme Court has accepted briefs directly filed by foreign governments since as early as 1919. Kristen Eichensehr, Foreign Sovereigns as Friends of the Court, 102 Va. L. Rev. 289, 291 (2016). These briefs — referred to as “amicus briefs” in the United States or briefs by a “friend of the court” — can seek to clarify points of foreign law, challenge the parties’ assumptions or understanding of foreign law, or share a new perspective. From 1978-2013, 46 foreign countries, in addition to the European Union and the Council of Europe, filed or signed a total of 68 amicus briefs in the United States. The United Kingdom filed most frequently during that time period, with 13 briefs. Id. at 292. Foreign sovereigns have weighed in on topics such as foreign-relations law or the constitutionality of capital punishment. Id.
4 [2017] CAT 16, [63].
5 [2017] CAT 16, [84].
6 [2017] CAT 16, [77].
7 [2017] CAT 16, [78].
8 [2019] EWCA Civ 674, [48] and [62].
9 [2019] EWCA Civ 674, [44].
10 [2019] EWCA Civ 674, [50].
11 An indirect-purchaser class action typically is filed in a single federal court, on behalf of putative plaintiffs that reside in all U.S. states that permit such suits.
12 Kanne v. Visa U.S.A. Inc., 723 N.W.2d 293, 298 (Neb. 2006).
13 Id. at 295-96. Whether the AGC factors apply to state antitrust laws “has recently become murky,” however. In re Capacitors Antitrust Litig., 106 F.Supp.3d 1051, 1072 (N.D. Cal. 2015). See also Los Gatos Mercantile, Inc. v. E.I. DuPont De Nemours and Company, 2015 WL 4755335, *19 (N.D. Cal. Aug. 11, 2015).
14 [2019] EWCA Civ 674, [54].
15 Comcast, 569 U.S. at 33 (quoting Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 350 (2011)). Many states have class certification rules that are substantially similar to Federal Rule of Civil Procedure 23 (Rule 23), but several states follow class-action procedures that differ significantly. For the purpose of this Essay, we focus on state claims brought in U.S. federal courts, which follow Rule 23, or on state courts that follow similar rules.
16 In re Hydrogen Peroxide Antitrust Litigation, 552 F.3d 305, 320 (3d Cir. 2008). See also Teamsters Local 445 Freight Div. Pension Fund v. Bombardier Inc., 546 F.3d 196, 201-204 (2d Cir. 2008); Alaska Elec. Pension Fund v. Flowserve Corp., 572 F.3d 221, 228-29 (5th Cir. 2009); Bell v. PNC Bank, Nat. Ass’n, 800 F.3d 360, 376 (7th Cir. 2015). But see Gooch v. Life Investors Ins. Co. of Am., 672 F.3d 402, 418 n.8 (6th Cir. 2012) (declining to adopt a preponderance-of-the-evidence standard to avoid “superimpos[ing] a more specific standard” on the rigorous analysis standard from the Supreme Court).
17 [2019] EWCA Civ 674, [40]; [2017] CAT 16, [59].
18 [2017] CAT 16, [39] and [58] (quoting Pro-Sys, [2013] SC 57, [118]).
19 [2019] EWCA Civ 674, [43] (quoting Pro-Sys, [2013] SC 57, [115]).
20 Id.
21 Comcast, 569 U.S. at 33-34 (internal quotation omitted).
22 Id. at 35-36.
23 Id. at 36-38.
24 See Zenith Radio Corp. v. Hazeltine Research, Inc., 395 U.S. 100, 114 n.9 (1969).
25 [2017] CAT 16, [67].
26 See, e.g. In re Lamictal Direct Purchaser Antitrust Litig., No. 19-1655, 2020 WL 1933260, at *6 (3d Cir. 2020) (holding the district court abused its discretion by assuming, “absent a rigorous analysis, that averages are acceptable” to find injury for a putative class of direct purchasers; instead, “the acceptability of averages depends largely on the answer to several factual predicates,” which requires “weigh[ing] the competing evidence and mak[ing] a prediction as to how they would play out at trial.”); Food Lion, LLC v. Dean Foods Co., 312 F.R.D. 472, 489-490 (E.D. Tenn. 2016).
27 See In re Rail Freight Fuel Surcharge Antitrust Litig., 934 F.3d 619, 625 (D.C. Cir. 2019) (rejecting certification because the damages model indicated 12.7 percent of the class members were uninjured); In re Asacol Antitrust Litig., 907 F.3d 42, 57 (1st Cir. 2018) (citing multiple courts of appeal that have reached similar conclusions). Conversely, some courts have held “the inability to show injury as to a few does not defeat class certification where the plaintiffs can show widespread injury to the class.” In re Flonase Antitrust Litig., 284 F.R.D. 207, 227 (E.D. Pa. 2012) (internal quotation marks omitted).
28 777 F.3d 9 at 13-14.
29 Id. at 25-26.
30 Id. at 27-28.
31 Id. at 31; id. at 28-31.
32 [2019] EWCA Civ 674, [45].
33 2018 WL 1156797 at *3.
34 Id. at *3-*5.
35 On the other hand, some courts analyzing whether there is predominance on the question of damages interpret Comcast’s damages analysis as applying only to situations where the plaintiffs advance multiple theories of impact. See, e.g. In re Lidoderm, 2017 WL 679367, at *24 (N.D. Cal. Feb. 21, 2017) (explaining “Comcast presents no problem to plaintiffs” where “[t]hey have one theory of injury and one consistent theory of damages”); Gomez v. Tyson Foods, 2013 WL 5516189, at *3 (D. Neb. 2013) (explaining the “Comcast holding is not applicable to this case because the plaintiffs proceeded on only one theory of recovery and damages were attributable to that theory”).
36 See, e.g. In re Cathode Ray Tube (CRT) Antitrust Litig., 2013 WL 5391159, at *5 (N.D. Cal. 2013) (characterizing defendants’ challenges to the plaintiffs’ expert methodology as an impermissible push “toward a full-blown merits analysis.” Instead, “[t]he Court’s job at this stage is simple: determine whether [indirect-purchaser plaintiffs] showed that there is a reasonable method for determining, on a class-wide basis, the antitrust impact’s effects on the class members.”).
37 See, e.g. In re EpiPen (Epinephrine Injection, USP) Mktg., Sales Practices & Antitrust Litig., 2020 WL 1180550, at *25 (D. Kan. Mar. 10, 2020) (rejecting defendants’ criticisms against the plaintiffs’ methodologies to calculate aggregate damages because their challenges “don’t undermine [the] plausibility for establishing class-wide damages.”).
38 See In re Pharm. Indus. Average Wholesale Price Litig., 582 F.3d 156, 197 (1st Cir. 2009) (“The use of aggregate damages calculations is … implied by the very existence of the class action mechanism itself.”).
39 Comcast, 569 U.S. at 34.
40 Id. at 35.
41 Bigelow v. RKO Radio Pictures, 327 U.S. 251, 264 (1946). See In re Cardizem CD Antitrust Litigation, 200 F.R.D. 326,348 (E.D. Mich. 2001) (holding plaintiffs “have proffered several reasonable damage methodologies for measuring class-wide damages on an aggregate basis and for calculating damages for individual class members”) (emphasis added). See also In re Domestic Air Transp. Antitrust Litigation, 137 F.R.D. 677, 693 (N.D. Ga. 1991) (explaining the court is convinced “that an adequate method exists for calculation of individual damages.”).
42 [2017] CAT 16, [84].
43 See Section V(A)(ii) above.
44 [2019] EWCA Civ 674, [62].
45 Id.
46 See Section V(B) above.
47 Cf. Tyson Foods, Inc. v. Bouaphakeo, 136 S. Ct. 1036, 1050 (2016) (expressing that although the question “whether uninjured class members may recover is one of great importance,” it is not “a question yet fairly presented by this case, because the damages award has not yet been disbursed, nor does the record indicate how it will be disbursed.”); In re Nexium, 777 F.3d at 20-21 (certifying a class even when a “proper mechanism for exclusion of [uninjured] consumers has not yet been proposed”); In re NASDAQ Mkt.-Makers Antitrust Litig., 169 F.R.D. 493, 522 (S.D.N.Y. 1996) (citations omitted) (“Courts have allowed the plaintiffs to establish the measure of damages at trial, and this measure is then applied to the individual transactions (typically in a second bifurcated proceeding following trial on the common issues.”); William Rubenstein, Alba Conte & Herbert Newberg, Newberg on Class Actions (5th ed., West 2011) § 20:62.
48 William Rubenstein, Alba Conte & Herbert Newberg, Newberg on Class Actions (5th ed., West 2011) § 4:72.
49 Certain requirements under Rule 23 are concerned with administrability and require consideration of the practicalities of distribution. For example, Federal Rule of Civil Procedure 23(b)(4)(D) asks courts to consider the “likely difficulties in managing a class action” as part of determining whether a “class action is superior to other methods for fairly and efficiently adjudicating the controversy.” This manageability requirement has been interpreted leniently. See William Rubenstein, Alba Conte & Herbert Newberg, Newberg on Class Actions (5th ed., West 2011) § 4:72. Some courts of appeal also consider the administrative feasibility of separating the uninjured from injured class members in deciding whether to certify a class. See Byrd v. Aaron’s Inc., 784 F.3d 154, 163 (3d Cir. 2015); In re Asacol, 907 F.3d at 52; William Rubenstein, Alba Conte & Herbert Newberg, Newberg on Class Actions (5th ed., West 2011) § 20:33 (referring to ascertainability of the class as an “implicit requirement” of Rule 23). But see In re Petrobras Sec., 862 F.3d 250, 264 (2d Cir. 2017) (“We conclude that a freestanding administrative feasibility requirement is neither compelled by precedent nor consistent with Rule 23, joining four of our sister circuits in declining to adopt such a requirement.”). In general, U.S. courts may be less receptive to class certification if there is no comprehensive record of relevant transactions and if purchasers would find it difficult to determine on their own whether they are in the proposed class. See William Rubenstein, Alba Conte & Herbert Newberg, Newberg on Class Actions (5th ed., West 2011) § 20:33.
50 See, e.g. In re Cardizem CD Antitrust Litig., 200 F.R.D. at 349 (holding the plaintiffs have proffered damage methodologies common to the class after noting that “apportionment [of the aggregate damage amount among class members] will be possible here using computerized records of the relevant drug purchases, the proposed damage methodology and formula, and detailed claims forms.”) See also In re Potash Antitrust Litig., 159 F.R.D. 682, 698 (D. Minn. 1995) (describing the possibility of using transactional records to compute individualized damages and noting “various judicial methods are available to resolve individual damage issues without precluding class certification,” including “using the defendants’ transactional records to compute individual damages.”).
51 2002 WL 1839983 at *13-14.
52 Id. at *15 (emphasis added).
53 Id.
54 Id. at *16.
55 Id. at *18.
56 Romero, 109 P.3d. at 794 (quoting Bigelow, 327 U.S. 251, 264-265).
57 Id. at 795.
58 Id.
59 See also In re Asacol, 907 F.3d at 58 (“[T]o determine whether a class certified for litigation will be manageable, the district court must at the time of certification offer a reasonable and workable plan for how [the defendant’s opportunity to press genuine challenges to allegations of injury-in-fact] will be provided in a manner that is protective of the defendant’s constitutional rights and does not cause individual inquiries to overwhelm common issues.”).
60 [2017] CAT 16, [67].
61 See, e.g. Phillip E. Areeda & Herbert Hovenkamp, Antitrust Law: An Analysis of Antitrust Principles and Their Application (4th ed., Wolters Kluwer 2013-2018), para. 346k.
