United States: Class actions – litigation, policy and latest developments
This is an Insight article, written by a selected partner as part of GCR's co-published content. Read more on Insight
Class actions play a substantial role in the US legal system, particularly in antitrust litigation. It is ‘the longstanding policy’ of US law to ‘encourage [the] vigorous private enforcement of the antitrust laws’, and the class action device creates the economic incentives that allow plaintiffs to litigate complex claims that otherwise would carry only small individual damages. Even so, these litigations carry substantial financial risks for defendants; as the Supreme Court has recognised, antitrust litigation is a colossal undertaking with discovery that imposes heavy burdens on defendants. What is more, the point of ‘class certification’ – when a proposed class is converted into a certified class and all absent members are joined as plaintiffs to the suit – is often the financial tipping point in litigation. If a class is certified, the sheer magnitude of the defendant’s potential ‘liability at trial becomes enormous, maybe even catastrophic, forcing companies to settle even if they have meritorious defenses’.
Accordingly, US antitrust class actions are hotly litigated, from their outset through the point of class certification. This guide provides an overview of key issues in that process, with an eye towards trends in litigation and the coming developments that may shape the future of collective antitrust litigation.
Who May Sue?
The Illinois Brick rule
From its initiation, US antitrust litigation presents difficult legal issues that plaintiffs and defendants must carefully navigate. As a threshold matter, class antitrust litigation in the United States is constrained by the concept of ‘antitrust standing’ – that is, to bring an antitrust action, plaintiffs must show not only that they have suffered concrete injuries (a requirement in all US federal litigation), but also that they are proper parties ‘to bring a private antitrust action’.
In 1977, the Supreme Court held in Illinois Brick Co v Illinois that the proper party to bring an antitrust action, in almost all instances, is the ‘direct purchaser’. As the court later explained, ‘“the immediate buyers from the alleged antitrust violators” may maintain a suit against the antitrust violators,’ but otherwise ‘indirect purchasers who are two or more steps removed from the violator in a distribution chain may not sue.’ This limitation, the Supreme Court has stated, is justified by the policy goals of ‘(1) facilitating more effective enforcement of antitrust laws; (2) avoiding complicated damages calculations; and (3) eliminating duplicative damages against antitrust defendants’. The direct purchaser rule might, however, allow direct purchasers to reap a windfall, recovering damages that they in fact passed down the line.
But who qualifies as a direct purchaser? The Supreme Court addressed the question most recently in Apple v Pepper. There, the court considered a suit brought by users of Apple products who alleged that Apple unlawfully monopolised the market for iPhone applications by allowing those apps to be distributed only through its ‘App Store’, and raised the prices of those apps by taking supracompetitive commissions. Apple, for its part, argued that the app users were not ‘direct purchasers from Apple’, because the app developers, rather than Apple, were ultimately responsible for app prices.
The Supreme Court held that the app users were ‘direct purchasers’, and therefore entitled to sue because, at the most basic level, the consumers ‘bought the apps directly from Apple’, which then extracted its commission and passed the remainder on to the app developers. The court rejected the contrary proposed ‘“who sets the price” theory’, which would have held that the ‘seller’ in the ‘purchaser-seller’ relationship was the entity that sets the price of the good in suit, regardless of that entity’s role in the supply chain.
The Supreme Court in Apple v Pepper therefore took a bright-line approach to the direct purchaser inquiry: the first buyer from the entity with monopoly power, no matter that entity’s role, is the proper antitrust plaintiff. Courts following Apple v Pepper understand it the same way. For example, in one recent putative class action, the Northern District of Illinois held that a homebuyer was not the proper plaintiff to state a claim against a trade association responsible for developing real estate broker commission rules and the corporate defendants who enforce those rules because, although the commission is baked into a home’s sale price paid by the buyer to the seller, the seller is the direct purchaser of agency services and the seller agent is responsible for remitting the buyer agent’s commission. Similarly, in In re Xyrem (Sodium Oxybate) Antitrust Litigation, a class of health insurance companies and unions that ‘bought or reimbursed Xyrem’, a medication for the treatment of narcolepsy, ‘for their members’, brought suit against the medication’s manufacturer. Applying Apple v Pepper, the court held that these end payors lacked antitrust standing because, as a formal matter, the prescriptions in issue were distributed through a pharmacy (Express Scripts), even though that pharmacy was the exclusive, defendant-selected distributor of Xyrem.
Taken together, these authorities suggest that all parties to antitrust litigation – plaintiffs and defendants alike – must studiously scrutinise the chain of distribution for the products at issue; any interruption in the distribution chain, however minor or functionally inconsequential, will likely be enough to defeat a class’s right to proceed.
Changes to Illinois Brick
Because Illinois Brick is a court-made, non-constitutional doctrine, Congress has the authority to repeal it at any time. Efforts to that end have been ongoing for decades and have garnered support from voices across the political spectrum, including the Democratic Party-led Antitrust Task Force of the Committee on the Judiciary of the House of Representatives for the 110th Congress and Makan Delrahim (the assistant attorney general for the Department of Justice’s Antitrust Division during the Trump administration). In 2021, Republican senator Mike Lee of Utah introduced legislation that would, among other things, overrule Illinois Brick; as of the date of publication, the legislation is pending before the Senate committee on the judiciary.
Further, because each state has its own antitrust laws (which substantially follow the federal law), state legislatures are free to permit downstream purchasers to bring antitrust claims on their own behalf. The majority of states have passed such laws – generally called Illinois Brick repealer laws. By doing so, these states have expanded the set of plaintiffs who might recover for conduct that violates state antitrust laws, but have also introduced significant legal complexity to antitrust litigation and attempts to settle disputes. For example, in Stromberg v Qualcomm, the plaintiffs brought a nationwide antitrust class action in California court, arguing that Qualcomm’s cellular chip licensing practices led to higher consumer mobile phone prices. The trial court certified a nationwide class, which the Ninth Circuit recently reversed, reasoning that because the class included citizens of both states that have repealed Illinois Brick within their borders and those that have not, and because of ‘significant variations in the scope of the repealer laws […] [e]ven among the repealer states’, choice-of-law issues might overwhelm the common-questions analysis at the core of the US class certification framework.
Illinois Brick is not an absolute rule. While, as a general matter, ‘indirect purchasers who are two or more steps removed from the antitrust violator in a distribution chain may not sue,’ courts have recognised or suggested exceptions for: buyers who purchase from the direct purchaser under ‘a pre-existing cost-plus contract’; ‘where the direct purchaser is owned or controlled by its customer’; and where the ‘indirect purchaser’ transacts with a ‘co-conspirator’ in a price-fixing conspiracy. Because of the prevalence of these provisions, US antitrust class actions commonly involve suits brought on behalf of both direct and indirect purchasers.
The application of the Illinois Brick exceptions presents a substantial legal issue at the outset of any antitrust action, including those brought by indirect purchaser classes, and it offers an early opportunity to defendants to limit the scope of any indirect purchaser action.
Marion Diagnostic Center v Becton Dickinson, a Seventh Circuit decision from March 2022, is instructive. There, the Seventh Circuit considered the application of the co-conspirator exception to a suit brought by a putative class of purchasers of medical products, who alleged that the manufacturers of those products (Cardinal and McKesson) conspired with regional distributors, including defendant Becton Dickinson, to fix equipment prices. The class, the ‘Indirect Purchaser Plaintiffs’, bought only McKesson products through Becton Dickinson, but nevertheless sued both McKesson and Cardinal, arguing that they could proceed against both under the co-conspirator exception. The Seventh Circuit, however, held that plaintiffs could only avoid Illinois Brick as to McKesson – even though McKesson and Cardinal were alleged to have engaged in the same conduct. The court found that the claims against Cardinal, to the extent they were viable at all, belonged to a different class of indirect purchasers, and were therefore dismissed. Accordingly, Marion Diagnostic confirms that Illinois Brick is a meaningful limitation on the scope of class action litigation, even when industry participants are alleged to have otherwise engaged in the same wrongdoing such that ‘efficiency’ considerations might support resolving all claims, actual and potential, in a single proceeding.
The Federal Rule of Civil Procedure 23 framework
Assuming a proper plaintiff brings the putative antitrust action, and that the plaintiff can meet the pleading requirements that apply to all allegations in a US litigation, an antitrust action will eventually arrive at the ‘class certification’ stage. Class certification is governed by Federal Rule of Civil Procedure 23 (Rule 23). To certify a class, a plaintiff must establish each of the requirements of Rule 23(a):
- the class must be ‘so numerous that joinder of all members is impracticable’;
- there must be ‘questions of law or fact common to the class’;
- the named plaintiff must have ‘claims or defenses … [that] are typical of the claims or defenses of the class’; and
- the named plaintiff and class counsel must be ‘adequat[e]’.
Further, to seek damages, the plaintiff must establish that the action fits within Rule 23(b)(3), such that the common questions ‘predominate over any questions affecting only individual members, and that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy’. The primary Rule 23(b)(3) inquiry, ‘predominance’, overlaps with the Rule 23(a) ‘common questions’ inquiry, but it ‘is far more demanding’; courts often analyse the two together at class certification.
Expert testimony at class certification
It is the plaintiff’s burden to prove that each of the Rule 23 requirements are satisfied. Accordingly, and as would perhaps be expected, class certification motions (and defendants’ corresponding opposition papers) rely, almost as a rule, on expert opinion testimony.
Defendants will often challenge that expert testimony through their own expert, at which point there is a ‘battle of the experts’, and a court must undertake an analysis of the methodology and reliability of the competing experts’ testimonies under the standards laid out in Daubert v Merrill Dow Pharms. While the Daubert inquiry might overlap with the merits, that overlap is not a proper basis for a court to defer consideration of the expert’s opinion testimony. Instead, the court must weigh the expert testimony to evaluate whether it might be admissible and if, considering the conflicting testimony, it is convinced that plaintiffs pass the Rule 23 test. If the proffered expert testimony does not meet the Daubert standard, courts can and do exclude it at class certification.
The court’s analysis of expert testimony regarding class certification is a high-stakes moment for antitrust plaintiffs. If a plaintiff’s expert is excluded at class certification, it will be extremely difficult for the plaintiff to win class certification. For example, in In re Foreign Exchange Benchmark Rates Antitrust Litigation, the Southern District of New York declined to certify a class because the plaintiffs’ experts’ methodology had an ‘error rate’ that was ‘too great to accept the method as reliable under Daubert’, and without this method the court would have had to undertake ‘tens or hundreds of thousands of individual determinations’. However, the inverse is not necessarily true – a court’s decision to accept a plaintiff’s expert testimony is not dispositive of the class certification inquiry.
Class certification and predominance issues
Predominance: impact and uninjured class members
While a plaintiff must establish that each of the Rule 23(a) and Rule 23(b) factors are satisfied, the bulk of antitrust class certification litigation focuses on the Rule 23(b) ‘predominance’ inquiry. At this stage, defendants attempt to demonstrate that a number of issues will have to be adjudicated on an individualised, class member-by-class member basis, undermining any efficiency gains associated with class treatment and making common adjudication impossible.
Often times, defendants will focus on the individualised nature of ‘antitrust impact’ – that is, the harm resulting from the alleged antitrust violation, usually in the form of a price overcharge. While courts have held that the existence or non-existence of a price-fixing conspiracy may be susceptible to common proof, defendants have often successfully argued that individualised questions of antitrust impact will predominate over plaintiffs’ proposed common issues because individualised analysis will be required to determine whether, and to what extent, each member of the proposed class was harmed by the alleged conspiracy. Like the other elements of an antitrust claim, ‘[w]hether or not antitrust impact can be proven on a classwide basis often hinges on (often conflicting) expert testimony.’
Where the products-in-suit are commodity items priced according to lists or published indices, plaintiffs have had success in defeating challenges to their common impact showings. For example, in May 2022, the Northern District of Illinois certified a class of purchasers of broiler chickens relying on expert testimony showing an artificial increase in chicken list prices and ‘that “the commodity nature of broiler products” means that “market prices depend on total industry supply,” which in turn means that “the prices for every customer that is based on these market prices indices … increases.”’ On the other hand, where the transactions-in-suit involve individualised, context-specific negotiations, courts take a more sceptical view toward plaintiffs’ supposed common showings of antitrust impact and often deny certification on predominance grounds.
In recent years, the federal courts have been divided over whether a plaintiff has satisfied the Rule 23(b)(3) predominance requirement when the proposed class contains members who have not suffered an antitrust injury at all. The decisions run the gamut, from those holding that a class cannot contain any uninjured plaintiffs to those suggesting that common issues might still predominate even if over a quarter of the class did not actually pay higher prices or otherwise suffer antitrust injury. For example, in Olean Wholesale Grocery Cooperative v Bumble Bee Foods, the US District Court for the Southern District of California certified a class of the purchasers of tuna products, even though the expert evidence established that at least 5 per cent of the class (in the plaintiffs’ model) and perhaps more than 28 per cent (in the defendants’ model) could not have paid increased prices as a result of the conspiracy. A Ninth Circuit panel reversed the district court, but the circuit reheard the case en banc and affirmed the certification order. The Ninth Circuit held that the presence of even ‘more than a de minimis number of uninjured class members’ is no barrier to class certification, instead deferring the determination of the number of uninjured class members to the jury.
Unfortunately, clarity is not likely to come to the law any time soon. Olean petitioned for review of the Ninth Circuit’s en banc order, presenting an opportunity for the Court to harmonise the discord among the lower courts, but the Supreme Court denied review in an unexplained November 2022 order. As a result, plaintiffs and defendants can expect to vigorously litigate the issue of predominance and absent class members in the years to come.
Comcast, damages and disaggregation
In addition to the question of whether class plaintiffs can prove antitrust impact on a common basis, plaintiffs seeking class certification must present a model that measures only the damages ‘attributable to [their] theory’ of liability. In other words, pursuant to Comcast v Behrend, plaintiffs must map their damages to each of the theories of liability in suit, setting aside the question of whether that theory is accurate or workable.
In Comcast, the plaintiffs filed a suit challenging Comcast’s business practices in the Philadelphia, Pennsylvania metropolitan area. At class certification, plaintiffs presented four theories of how Comcast’s unlawful practices increased prices (causing antitrust impact), and then offered damages evidence that identified an alleged overcharge, but ‘did not isolate damages resulting from any one theory of antitrust impact’. The district court threw out three of the four theories of antitrust impact, but nevertheless certified the class. The Supreme Court ultimately reversed certification, reasoning that ‘a model purporting to serve as evidence of damages in [a] class action’ based on a specific theory of injury ‘must measure only those damages attributable to that theory’. Thus, because the expert’s damages model failed to measure the damages under the only remaining theory of antitrust impact, the court held that class treatment was improper.
Plaintiffs face a number of challenges under Comcast. They must carefully define their theory of liability and ensure that their expert’s damages model matches, and does not exceed, the bounds of that theory. What’s more, if circumstances change, Comcast will bite litigants who have not thought ahead. Often, plaintiffs argue that the defendants’ conduct, taken together, caused an antitrust injury. While this sort of aggregated-effects theory might be permissible, it is risky. Where the plaintiffs’ theory of injury is based on the effects of multiple forms of conduct, and the expert does not separate the harm attributable to each form, Comcast and the related principle of disaggregation may bar the expert’s testimony even if some of the underlying conduct is found lawful. For example, in Viamedia v Comcast, the plaintiff’s expert failed to separate the injury caused by one set of practices (refusal to deal) from alleged ‘tying, exclusive dealing, and other exclusionary conduct’, instead assuming that all of the defendant’s practices were unlawful and calculating a single injury. The trial court excluded the expert’s testimony, as a plaintiff ‘must be able to segregate the damages ... caused by lawful competition from those caused by anticompetitive acts’. While the appellate court later reversed the trial court’s conclusion that the alleged refusal-to-deal was necessarily lawful (after which point the plaintiffs dropped the theory), the case underscores the importance and potentially certification-dispositive nature of the disaggregation inquiry.
Settlement and objectors
The majority of antitrust class actions settle. This is due to many reasons, including that the federal antitrust statutes allow for treble damages and that, under section 1, defendants are subject to joint and several liability. In addition, discovery and expert testimony in antitrust cases are recognised to be complex and burdensome.
US law allows courts to certify classes for the sole purpose of settlement. While these settlement classes must satisfy the strictures of Rule 23 to be certified, they present significant practical benefits – as one court recently recognised, ‘the class action device’ was ‘the only feasible method of resolving … claims’ in a suit involving millions of class members.
Settlement classes also carry a finality benefit for defendants: because class settlements are binding on all class members, a class settlement buys a defendant peace not just from the claims at hand, but from potential future claims. However, defendants should be wary of class members bound by time – that is, all members who purchased a product between 2015 to 2020, as a settlement of that class would not preclude a class action from purchasers of that product from 2014 or 2021.
However, class members may choose to ‘opt out’ of the settlement before it is completed and to litigate on their own; these opt-out plaintiffs can substantially prolong a litigation, exposing settling defendants to additional defence costs and liability.
As class action settlements will bind absent class members, court approval is required before an antitrust class settlement can be final and aggrieved members of the putative class may object to the settlement, including to the fees to be paid to class counsel. This objection-and-court review can upset even the largest settlements – for example, in 2016, the Second Circuit overturned a settlement worth up to US$7.25 billion in antitrust litigation between payment card networks and their customers because, in the court’s view, the settlement improperly favoured some class members over others.
Class antitrust litigation is a difficult process, fraught with legal issues for both plaintiffs and defendants. From the outset of an antitrust litigation, both plaintiffs and defendants must litigate with Rule 23 and its attendant issues in mind, and should not lose sight of class-related issues following class certification or at the point of a negotiated resolution.
Illinois Brick Co v Illinois, 431 US 720, 745 (1977)
Bell Atl Corp v Twombly, 550 US 544, 557–59 (2007).
Olean Wholesale Grocery Cooperative, Inc v Bumble Bee Foods LLC, 31 F4th 651, 687 (9th Cir 2022) (en banc) (Lee, J, dissenting).
Associated General Contractors of California, Inc v California State Council of Carpenters, 459 US 519, 535 n31 (1983).
Illinois Brick Co v Illinois, 431 US 720, 745–46 (1977).
Apple Inc v Pepper, 139 S Ct 1514, 1520 (2019) (quoting, in the first passage, Kansas v UtiliCorp United Inc, 497 US 199, 207 (1990)).
 id at 1524.
 139 S Ct at 1519.
 id at 1520.
Leeder v Nat’l Ass’n of Realtors, 2022 WL 1307100, at *4 (ND Ill May 2, 2022).
 555 F Supp 3d 829, 846 (ND Cal August 13, 2021).
 id at 841, 881–882. This holding is in unacknowledged practical tension with Blue Cross & Blue Shield United of Wi v Marshfield Clinic, in which the Seventh Circuit held that health insurers are the direct purchasers of medical services from clinical providers for antitrust purposes (65 F3d 1406 (7th Cir 1995)).
 Findings and Recommendations of the Antitrust Modernization Commission.
 Section 501 overturning Illinois Brick and Hanover Shoe.
In re Dynamic Random Access Memory (DRAM) Antitrust Litigation, 2013 WL 12333442, at *13–15 (ND Cal 8 January 2013) (recounting negotiations).
Stromberg v Qualcomm Inc, 14 F4th 1059, 1063 (9th Cir 2021).
 id at 1069–1074.
Apple Inc v Pepper, 139 S Ct 1514, 1521 (2019).
Illinois Brick, 431 US at 736.
 id at 736 n16.
In re Nat’l Football League’s Sunday Ticket Antitrust Litig, 933 F3d 1136, 1157–1158 (9th Cir 2019).
 The existence of these actions – and the fact that they can reliably be litigated and resolved – undermines at least to some extent the rationale behind Illinois Brick. Nevertheless, it does not appear that the court is likely to overrule Illinois Brick any time soon.
 29 F4th 337, 341 (7th Cir 2022).
 id at 347–348.
 Ultimately, the court dismissed the McKesson-based claims on pleading grounds. Id at 351.
In re Warfarin Sodium Antitrust Litig, 391 F3d 516, 528 (3d Cir 2004). Rule 23(b)(3) also requires a plaintiff show that the class mechanism is ‘superior’ to other forms of adjudication; this prong is less frequently litigated.
In re Blood Reagents Antitrust Litig, 783 F3d 183, 187 (3d Cir 2015).
In re Blood Reagents Antitrust Litig, 783 F3d 183, 187 (3d Cir 2015); In re Processed Egg Prods Antitrust Litig, 81 F Supp 3d 412, 417 (ED Pa 2015).
In re Processed Egg Prods Antitrust Litig, 81 F Supp 3d 412, 415–416 (ED Pa 2015).
In re Restasis (Cyclosporine Ophthalmic Emulsion) Antitrust Litig, 2020 WL 2280144, at *2 (EDNY 5 May 2020) (excluding defendants’ experts at class certification).
 407 F Supp 3d 422, 434–435 (SDNY 2019).
In re Processed Egg Prods Antitrust Litig, 81 F Supp 3d 412, 417 (ED Pa 2015).
Blades v Monsanto Co, 400 F3d 562, 566 (8th Cir 2005): ‘The requirement of Rule 23(b)(3) that common questions predominate over individual questions tests whether proposed classes are sufficiently cohesive to warrant adjudication by representation. The nature of the evidence that will suffice to resolve a question determines whether the question is common or individual.’
Comcast Corp v Behrend, 569 US 27, 30 (2013).
 See, for example, In re Industrial Diamonds Antitrust Litig, 167 FRD 374, 379 (SDNY 1996).
In re Domestic Drywall Antitrust Litig, 322 FRD 188, 205 (ED Pa 2017): ‘Antitrust impact is critically important to the Rule 23(b)(3) predominance analysis, because it is an element of an antitrust claim that may call for individual, as opposed to common proof.’ In re New Motor Vehicles Canadian Export Antitrust Litig, 522 F3d 6, 20 (1st Cir 2008): ‘In antitrust actions, common issues do not predominate if the fact of antitrust violation and the fact of antitrust impact cannot be established through common proof.’
In re Broiler Chicken Antitrust Litig, 2022 WL 1720468, at *13 (ND Ill 27 May 2022).
Blades v Monsanto Co, 400 F3d 562, 570–571 (8th Cir 2005); In re Graphics Processing Units Antitrust Litigation, 253 FRD 478, 490–491 (ND Cal 2008).
In re Apple iPhone Antitrust Litig, 2022 WL 1284104, at *14–15 (ND Cal 29 March 2022) (collecting cases).
 332 FRD 308 (SD Cal 2019).
 993 F3d 774 (9th Cir 2021).
 31 F4th 651 (9th Cir 2022) (en banc).
 id at 676.
Comcast Corp v Behrend, 569 US 27, 34–35 (2013), rejecting a ‘model [that] failed to measure damages resulting from the particular antitrust injury on which petitioners’ liability in this action is premised’.
 id at 29–30.
 id at 31–32.
 id at 35.
In re Suboxone (Buprenophine Hydrochoride and Nalaxone) Antitrust Litigation, 421 F Supp 3d 12, 37–38 (ED Pa 2019).
Sumotext Corp v Zoove, Inc, 2020 WL 533006, at *7 (ND Cal 3 February 2020) (quoting Litton Sys, Inc v Honeywell, Inc, 1996 WL 634213, at *2 (CD Cal 24 July 1996)).
 335 F Supp 3d 1036, 1072 (ND Ill 2018).
Amchem Prods, Inc v Windsor, 521 US 591, 619–620 (1997).
In re Blue Cross Blue Shield Antitrust Litig, 2022 WL 4587618, at *13–14 (ND Ala 9 August 2022).
In re Payment Card Interchange Fee & Merchant Discount Antitrust Litig, 827 F3d 223, 233–234 (2d Cir 2016).