Opinion
Docket 20-339(L) 20-340(CON) 20-341(CON) 20-342(CON) 20-343(CON) 20-344(CON)
03-15-2023
NATHANIEL A. TARNOR, Hagens Berman Sobol Shapiro LLP, New York, NY (Steve W. Berman, Hagens Berman Sobol Shapiro LLP, Seattle, WA, on the brief), for Objectors-Appellants Fikes Wholesale, Inc., et al. N. ALBERT BACHARACH, JR. (Paul S. Rothstein, on the brief), Gainesville, FL, for Objectors-Appellants Jack Rabbit, LLC, et al. KENDRICK JAN, San Diego, CA, for Objectors-Appellants Gnarlywood LLC, et al. JOHN J. PENTZ, Sudbury, MA, for Objectors-Appellants Pets USA LLC, et al. C. Benjamin Nutley, Pasadena, CA, and John W. Davis, Tampa, FL, for ObjectorAppellant Kevan McLaughlin. PATRICK J. COUGHLIN, (Joseph D. Daley, Alexandra S. Bernay, and Carmen A. Medici, on the brief) Robbins Geller Rudman &Dowd LLP, San Diego, CA; K. Craig Wildfang, Thomas J. Undlin, Ryan W. Marth, Robins Kaplan LLP, Minneapolis, MN; H. Laddie Montague, Jr., Merrill G. Davidoff, Michael J. Kane, Berger Montague PC, Philadelphia, PA, for Plaintiffs-Appellees. KANNON K. SHANMUGAM (Kenneth A. Gallo, Jessica Anne Morton, Stacie M. Fahsel, on the brief) Paul, Weiss, Rifkind, Wharton &Garrison LLP, Washington, DC; Gary R. Carney, Elyssa E. Abuhoff, Paul, Weiss, Rifkind, Wharton &Garrison LLP, New York, NY, for Defendants-Appellees.
Argued: March 16, 2022
A putative class of over 12 million merchants brought this antitrust action under the Sherman Act against Visa U.S.A. Inc., MasterCard International Inc., and numerous banks that serve as payment-card issuers for those networks. Plaintiffs alleged that Visa and MasterCard adopted and enforced rules and practices relating to payment cards that had the combined effect of injuring merchants by allowing Visa and MasterCard to charge supracompetitive fees (known as "interchange fees") on each payment card transaction. After nearly fifteen years of litigation, the parties agreed to a settlement of roughly $5.6 billion, which was approved by the United States District Court for the Eastern District of New York (Brodie, C.J.) over numerous objections. In so doing, $900,000 in service awards was granted to lead plaintiffs, and roughly $523 million was granted in attorneys' fees.
The appellants are various objectors who argue that the district court erred when it certified the class, approved the settlement, granted service awards and computed attorneys' fees.
We conclude that these arguments are without merit and therefore AFFIRM the district court's orders in all respects, except as noted below.
The opinion of the Court is unanimous. Judge Jacobs and Judge Leval each concur in separate opinions.
NATHANIEL A. TARNOR, Hagens Berman Sobol Shapiro LLP, New York, NY (Steve W. Berman, Hagens Berman Sobol Shapiro LLP, Seattle, WA, on the brief), for Objectors-Appellants Fikes Wholesale, Inc., et al.
N. ALBERT BACHARACH, JR. (Paul S. Rothstein, on the brief), Gainesville, FL, for Objectors-Appellants Jack Rabbit, LLC, et al.
KENDRICK JAN, San Diego, CA, for Objectors-Appellants Gnarlywood LLC, et al.
JOHN J. PENTZ, Sudbury, MA, for Objectors-Appellants Pets USA LLC, et al. C. Benjamin Nutley, Pasadena, CA, and John W. Davis, Tampa, FL, for ObjectorAppellant Kevan McLaughlin.
PATRICK J. COUGHLIN, (Joseph D. Daley, Alexandra S. Bernay, and Carmen A. Medici, on the brief) Robbins Geller Rudman &Dowd LLP, San Diego, CA; K. Craig Wildfang, Thomas J. Undlin, Ryan W. Marth, Robins Kaplan LLP, Minneapolis, MN; H. Laddie Montague, Jr., Merrill G. Davidoff, Michael J. Kane, Berger Montague PC, Philadelphia, PA, for Plaintiffs-Appellees.
KANNON K. SHANMUGAM (Kenneth A. Gallo, Jessica Anne Morton, Stacie M. Fahsel, on the brief) Paul, Weiss, Rifkind, Wharton &Garrison LLP, Washington, DC; Gary R. Carney, Elyssa E. Abuhoff, Paul, Weiss, Rifkind, Wharton &Garrison LLP, New York, NY, for Defendants-Appellees.
Before: JACOBS, LEVAL, and PARK, Circuit Judges.
DENNIS JACOBS, Circuit Judge
A putative class of over 12 million merchants brought this antitrust action under the sherman Act against Visa U.s.A. Inc. ("Visa"), MasterCard
International Inc. ("MasterCard"), and numerous banks that serve as paymentcard issuers for those networks. Plaintiffs alleged that Visa and MasterCard adopted and enforced rules and practices relating to payment cards that had the combined effect of injuring merchants by allowing Visa and MasterCard to charge supracompetitive fees (known as "interchange fees") on each payment card transaction. After nearly fifteen years of litigation, the parties agreed to a settlement of roughly $5.6 billion, which was approved by the United States District Court for the Eastern District of New York (Brodie, C.J.) over numerous objections. In so doing, $900,000 in service awards was granted to lead plaintiffs, and roughly $523 million was granted in attorneys' fees.
The appellants are various objectors who argue that the district court erred when it certified the class, approved the settlement, granted service awards and computed attorneys' fees.
We refer to "Appellants" plural, regardless of whether a point was advanced by one objector-appellant or more.
We conclude that these arguments are without merit and therefore AFFIRM the district court's orders in all respects, except as noted below.
BACKGROUND I
It is well known that Visa and MasterCard operate two of the world's largest payment-card networks. An understanding of this case requires some knowledge of how those networks operate. In brief: the customer presents a payment card to the merchant; the merchant relays the card information to its bank (the acquiring bank); the acquiring bank forwards that information to the appropriate network (Visa or MasterCard); the network relays the information to the bank that issued the customer's card (the issuing bank); and the issuing bank confirms that the customer has sufficient credit or funds to cover the purchase. . When these steps are completed, the issuing bank transmits its approval back through the chain to the acquiring bank, which relays it to the merchant at the point of sale.
Next, the issuing bank provides the funds via the appropriate network to the acquiring bank, less the "interchange fee" (the focus of this litigation) which the issuing bank - a member of the Visa or MasterCard network - keeps. While the issuing and acquiring banks are contractually free to agree on an applicable interchange fee, Visa and MasterCard each set a default fee in the absence of such an agreement. The acquiring bank, in turn, pays the merchant while charging what is known as a "merchant discount fee." That fee covers the interchange fee as well as an additional amount that compensates the acquiring bank for processing the transaction. The applicable default fee varies depending on factors that include the type of payment card.
Merchants who accept Visa- and MasterCard-branded cards are bound by the issuers' network rules, which in effect are identical to one another. For example, the "honor-all-cards" rules require any merchant that accepts a Visa- or MasterCard-branded credit card to accept all credit cards of that brand, regardless of the differences in interchange fees. Further, multiple rules prohibit merchants from influencing customers to use one type of payment over another--such as a credit card with a lower interchange fee, or cash rather than credit. These "anti-steering" rules include the "no-surcharge" and "no-discount" rules, which prohibit merchants from charging different prices at the point of sale depending on the means of payment.
II
Numerous antitrust lawsuits were filed against Visa and MasterCard beginning in 2005. The first consolidated complaint in this action (which was followed by several amended complaints) was filed the following year. The then-operative complaints alleged that the Visa and MasterCard interchange fees, and associated rules, were anticompetitive and violated the Sherman Act, 15 U.S.C. §§ 1 and 2, and California's Cartwright Act, Bus. &Prof. Code § 16700 et seq.
Prolonged negotiation resulted in a "2012 Settlement Agreement." Consistent with the then-operative complaint, the proposed settlement divided the plaintiffs into two classes. The first--the Rule 23(b)(3) damages class--covered merchants that accepted Visa and/or MasterCard from January 1, 2004 to November 27, 2012. The second--the Rule 23(b)(2) injunction class--covered merchants that accepted (or would accept) Visa and/or MasterCard on or after November 27, 2012. The 23(b)(3) class was set to receive roughly $5.3 billion, after opt-out payments. The 23(b)(2) class would receive injunctive relief in the form of changes to Visa's and MasterCard's network rules. While members of the (b)(3) damages class could opt out, members of the (b)(2) injunction class could not. The same counsel represented both classes.
The settlement agreement was given preliminary approval in late 2012. Final approval by the district court came roughly a year later.
III
On June 30, 2016, we held that the members of the (b)(2) injunction class received inadequate representation, in violation of the Due Process Clause and Rule 23(a)(4) of the Federal Rules of Civil Procedure, because, despite their divergent interests, they were represented by the same counsel and lead plaintiffs as the (b)(3) damages class. See In re Payment Card Interchange Fee &Merch. Disc. Antitrust Litig., 827 F.3d 223, 233-34 (2d Cir. 2016). Specifically, we explained that the interest of the (b)(3) class members was in "maximiz[ing] cash compensation for past harm," while the interest of the (b)(2) class members was in "maximizing] restraints on network rules to prevent harm in the future." Id. at 233. This conflict disadvantaged the (b)(2) injunction class, we concluded, because "[t]he class counsel and class representatives who negotiated and entered into the [2012] Settlement Agreement were in the position to trade diminution of (b)(2) relief for increase of (b)(3) relief." Id. at 234. Accordingly, we vacated the district court's certification of the settlement class, reversed approval of the 2012 Settlement Agreement, and remanded for further proceedings. Id. at 240.
IV
On remand, the district court appointed the three law firms that had previously served as class counsel (together, "Class Counsel") to serve as interim counsel only for the Rule 23(b)(3) damages class. About two years later, after additional discovery and renegotiation, the parties executed this new settlement agreement (the "Settlement Agreement"), which provides for a collective award of $5.6 billion (as reduced in the amount of $700 million to reflect opt-outs). The district court granted preliminary approval on January 24, 2019, and final approval on December 13, 2019.
The district court also appointed separate counsel to represent a proposed injunctive-relief class seeking certification under Rule 23(b)(2). The (b)(2) action is now proceeding as DDMB, Inc. v. Visa, Inc., No. 05-md-1720.
The Settlement Agreement defines the class as "all persons, businesses, and other entities that have accepted any Visa-Branded Cards and/or Mastercard-Branded Cards in the United States at any time from January 1, 2004 to the Settlement Preliminary Approval Date [January 24, 2019]." App'x at 3324 ¶ 4. The definition expressly excludes, among others, the "Dismissed Plaintiffs," a defined term referring to roughly 200 merchants (and their related entities) that had opted out of the initial settlement and separately settled their claims. Id. at 3316 ¶ 3(t), 3324-25 ¶ 4. The Settlement Agreement provides that each claimant will receive a pro rata share of the monetary fund "in accordance with the relative economic interests of the Claimants as measured by the Interchange Fee amounts attributable to their Visa- and Mastercard-Branded Card transactions during the Class Period." Id. at 3567.
In exchange for this cash, each class member released claims "arising out of or relating to" any conduct or acts that were or could have been alleged in the litigation. Id. at 3337 ¶ 31(a). That release covers all claims that have accrued during the class period, or that will "accrue no later than five years after the Settlement Final Date," id., which is defined as the date on which appeals from the Settlement Agreement become final, id. at 3321-22 ¶ 3(ss).
The release "extend[s] to, but only to, the fullest extent permitted by federal law." Id. at 3337 ¶ 31(a). The parties have stated, in the district court and on appeal, that this language is intended to comport with the "identical factual predicate" test, i.e., that a settlement may validly release claims "that would have to be based on the identical factual predicate as that underlying the claims in the settled class action." TBK Partners, Ltd. v. Western Union Corp., 675 F.2d 456, 460 (2d Cir. 1982). See Defendants-Appellees Br. at 52; Plaintiffs-Appellees Final Approval Br. at 63; App'x at 4424. The district court so construed the Settlement Agreement in granting approval. In re Payment Card Interchange Fee &Merch. Disc. Antitrust Litig., No. 05-md-1720, 2019 WL 6875472, at *23-25 (E.D.N.Y. Dec. 16, 2019) ("Final Approval").
Because of the dispute described below between integrated oil companies and their franchised service stations operating under the trademark of the franchisor, contesting which should receive the settlement funds allocated to customers using payment cards to purchase primarily gasoline, the district court determined to appoint a special master to study and determine the question, subject to de novo review by the district court.
After granting final approval of the Settlement Agreement, the district court awarded Class Counsel 9.31% of the settlement fund in attorneys' fees, approximately $523 million. See In re Payment Card Interchange Fee &Merch. Disc. Antitrust Litig., No. 05-md-1720, 2019 WL 6888488, at *25 (E.D.N.Y. Dec. 16, 2019). This award was meant to reflect, among other things, the "enormous number of hours and many years" devoted to the case, id. at *10, and the "substantively risky" nature of the litigation, id. at *14. The award for expenses was $39 million. See id. at *25.
Separately, the district court granted the class representatives $900,000 in service awards, in addition to out-of-pocket expenses. The district court recognized that "[t]here is no indication that any Class Representative assumed the position or undertook the time and labor they did in anticipation of being compensated," but still granted the award because of the "enormous amount of time and resources" they spent on the case. App'x at 7456-57.
Appellants argue that the district court abused its discretion when it certified the class, approved the settlement, granted service awards and computed attorneys' fees.
DISCUSSION I
The settlement-only class is defined to include (with limited exceptions) "all persons, businesses, and other entities that have accepted any Visa-Branded Cards and/or Mastercard-Branded Cards in the United States at any time from January 1, 2004 to [January 24, 2019]." App'x at 3324 ¶ 4. A controversy over the class definition exists between integrated oil companies (e.g., Shell, Chevron, and Valero) and their branded service stations--i.e., between franchisors and their franchisees--both of which claim to have been injured by the interchange fees imposed on gas sales and claim to "have accepted" Visa-Branded and MasterCard-Branded cards in payment for sales. Appellants are representatives of the service stations, and raise challenges regarding: (A) ascertainability, (B) adequacy of representation, (C) claims administration, and (D) notice.
A
The ascertainability challenge hinges on the interpretation of a single word: "accepted." The word "accepted" creates ambiguity. Both the franchisors and the franchisees claim to have "accepted" the payment card from the customers at the service stations. The franchisees claim to have "accepted" Visa and/or MasterCard from consumers paying for the gasoline dispensed at their stations, while the franchisors claim to have "accepted" the payment cards by, inter alia, providing operating and processing services on the same set of transactions. The question of which ones accepted the payment cards was not resolved by the district court but was instead set aside for determination by a special master, subject to de novo review by the district court. Appellants contend that this failure of the settlement agreement renders the class unascertainable, with dire consequences for the settlement. This is mistaken.
There was a question as to whether this unresolved issue might impact appellate jurisdiction. To obviate any such complications, we remanded pursuant to the procedure set out in United States v. Jacobson, 15 F.3d 19 (2d Cir. 1994), for the district court to consider whether, in the event its disposition of the dispute between service stations and oil companies was deemed not a final judgment, there was no just reason for delay in the appeal of all other issues. On July 18, 2022, the district court so concluded and certified that the judgment previously entered and appealed from should be considered a partial final judgment in accordance with Rule 54(b) of the Federal Rules of Civil Procedure. In re Payment Card Interchange Fee & Merch. Disc. Antitrust Litig., No. 05-md-1720, 2022 WL 2803352 (E.D.N.Y. July 18, 2022).
Ascertainability requires only that "a proposed class is defined using objective criteria that establish a membership with definite boundaries." In re Petrobras Sec., 862 F.3d 250, 269 (2d Cir. 2017). This definition meets that requirement by providing the timeframe (January 1, 2004 to January 24, 2019) and place (the United States) in which a particular group (entities that have accepted Visa and/or MasterCard) was allegedly harmed. That is all that is needed.
True, the word "accepted" lends itself to ambiguity. But the district court properly concluded that "the class definition is . . . objectively guided by federal antitrust standards." Final Approval, No. 05-md-1720, 2019 WL 6875472, at *31; see also In re Motorola Sec. Litig., 644 F.3d 511, 517 (7th Cir. 2011) ("[A]s a general rule, a class definition is interpreted according to the substantive law that provides the basis for the class action."). In view of those standards, the district court accepted Class Counsel's claim that "they represent only . . . the direct purchaser, and not every entity in the payment chain." Final Approval, No. 05-md-1720, 2019 WL 6875472, at *31 (internal quotation marks omitted). Put differently, the district court found that federal antitrust law clarified that the only entities that could fall within the class definition were those deemed to be direct payors of the challenged fees.
Appellants argue that it is not clear that federal antitrust law should guide the class definition, as the operative complaint also asserts claims under California's Cartwright Act, which expressly allows recovery for indirect payors. Cal. Bus. & Prof. Code § 16750(a). But Appellants raised this argument for the first time in their reply brief, thereby waiving it. See JP Morgan Chase Bank v. Altos Hornos de Mexico, S.A. de C.V., 412 F.3d 418, 428 (2d Cir. 2005) ("[A]rguments not made in an appellant's opening brief are waived even if the appellant pursued those arguments in the district court or raised them in a reply brief.").
No party to this appeal disputes that "for any given transaction, only one interchange fee was paid and only one claimant is entitled to recover based on that fee." Defendants' Br. at 31; see also Jack Rabbit Br. at 38, 55 ("Pursuant to Illinois Brick, there can only be one claim per antitrust injury."). We therefore accept that interpretation of federal antitrust law for the purposes of this appeal, without opining on its soundness.
Appellants argue that it is not enough for the definition to be discernible: the criteria for class membership must be "administratively feasible," such that identifying the class members "would not require a mini-hearing on the merits of each case." Brecher v. Republic of Arg., 806 F.3d 22, 24-25 (2d Cir. 2015) (internal quotation marks omitted). Appellants contend that the administrative feasibility requirement is not satisfied in this case, since the dispute over which entity was the direct payor for any given transaction will lead to "hundreds of thousands of mini trials." Jack Rabbit Br. at 64.
This argument rests on a faulty premise. Requiring administrative feasibility is "neither compelled by precedent nor consistent with Rule 23." Petrobras, 862 F.3d at 264. The only relevant inquiry is whether determinations as to class membership are "objectively possible." Id. at 270. Appellants do not contend that identifying the direct payor for each transaction is impossible. So, their ascertainability argument must fail.
B
Appellants next contend that the class definition gave rise to an intra-class conflict that left the franchisees without adequate representation. We are unpersuaded.
"Class actions are an exception to the rule that only the named parties conduct and are bound by litigation." In re Payment Card Interchange Fee &Merch. Disc. Antitrust Litig., 827 F.3d at 231. "In order to justify a departure from that rule, a class representative must be part of the class and possess the same interest and suffer the same injury as the class members." Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 348-49 (2011) (internal quotation marks omitted). That principle is secured in part by Rule 23(a)(4), which requires that "the representative parties . . . fairly and adequately protect the interests of the class." Fed.R.Civ.P. 23(a)(4).
Appellants argue that the franchisees have been inadequately represented because no class representative has a shared interest in the franchisees' dispute with their franchisors. Such shared interests are necessary to protect the franchisees, Appellants contend, since the loser of the intra-class conflict will be entitled to no recovery, but will still be bound by the settlement's release.
Appellants are confused. The dispute is not over which group of class members will get the recovery; the dispute is over which claimants are within the class. Whoever "accepted" the payment cards is by definition in the class, gets compensation, and is bound by the release; an entity that did not accept the payment is by definition excluded from the class and is not bound by the settlement. See Rothstein v. Am. Int'l Grp., 837 F.3d 195, 204 (2d Cir. 2016) (noting that non-class members "cannot be bound by any orders or judgments entered in respect to the settlement" (internal quotation marks omitted)). The class representatives owe nothing to the losing entities; outsiders are irrelevant.
The district court emphasized that point: "[B]ecause the conflict is not between class members, but between entities disputing who has the right to claim class status, there is no intra-class conflict or inadequate representation; Class Counsel and Class Plaintiffs are not responsible to, and do not represent, the entity that loses the dispute over the right to claim settlement funds." In re Payment Card Interchange Fee &Merch. Disc. Antitrust Litig., 330 F.R.D. 11, 33 (E.D.N.Y. 2019) ("Preliminary Approval"). As a result, the district court concluded, there "is a dispute that needs to be resolved," but it "need not be resolved through creation of subclasses or appointing new class representatives or counsel." Final Approval, No. 05-md-1720, 2019 WL 6875472, at *18.
We agree with the district court that the arguments advanced by the franchisees present no reason why the settlement affecting all claimants other than the franchisors and franchisees of integrated oil companies should not be consummated.
Until now, there has been no conflict between franchisors and franchisees. To the contrary, they have shared an interest in maximizing the recovery for all class members, which each claims to be. Going forward, however, franchisors and franchisees will become adverse to each other for resolution of the questions regarding which should be deemed to have accepted the cards and therefore to come within the class and receive settlement funds, and how those funds should be allocated,. It appears that class counsel will be conflicted on those questions. The district court should ensure that franchisors and franchisees will be represented by counsel, enabling each side to present arguments to the special master and the district court as to why they should be favored over the adversary, and enabling them to negotiate toward settlements of the disputes affecting each adverse pair.
Whether this will require designations of subclasses and appointment of further counsel to represent them or whether their existing privately retained counsel will adequately serve the needs, we leave to the good judgment of the district court.
C
Appellants next challenge the district court's plan to refer the dispute over class membership to a special master, subject to its de novo review. Appellants raise three arguments on this score.
First: they contend this dispute should have been resolved before class certification because "no class may be certified that contains members lacking Article III standing," and "[t]he class must therefore be defined in such a way that anyone within it would have standing." Denney v. Deutsche Bank AG, 443 F.3d 253, 264 (2d Cir. 2006). This argument stems from the same mistaken belief that this is a dispute between class members rather than a dispute over class membership. And there is nothing unusual about a special master identifying that entity after settlement approval. See Rothstein, 837 F.3d at 199, 201-02 (noting that class membership was determined after settlement approval); Motorola, 644 F.3d at 515 (same).
Second: they contend that the claims process is fundamentally unfair because it will necessarily result in one group receiving nothing and yet releasing all its claims. As discussed above, however, that is just not so. Excluded entities would not be bound by the settlement.
Third: they contend that the contemplated special master process is too vague, complicated, and adversarial to be appropriate under Rule 23. This vagueness argument rests on a series of open questions as to how the special master will go about identifying direct payors--e.g., whether the special master will allow legal briefs or expert testimony. But logistics can be worked out in time. The district court provided the only marching order that matters: that the special master identify the direct payor or payors in any given transaction. That is enough for now.
This fact distinguishes this case from the one that Appellants rely upon, in which the Fifth Circuit held that that a mandatory limited-fund settlement of tort claims required decertification because the district court punted, without any guidance, "the difficult question of equitable distribution . . . to the special master." In re Katrina Canal Breaches Litig., 628 F.3d 185, 194 (5th Cir. 2010).
As to the complicated and adversarial nature of the impending special master proceedings, special masters have been frequently tasked with deciding antitrust standing issues at the summary judgment stage. These are tough and hotly disputed issues. But district courts with loaded dockets may rely on special masters to decide thorny things. With respect to the restrictions imposed by Rule 53(a)(1)(C) on circumstances in which special masters may be appointed, limiting appointment to matters that "cannot be effectively and timely addressed by an available district judge or magistrate judge," we have no reason to doubt that the district court complied with its obligations. Fed.R.Civ.P. 53(a)(1)(C).[
Appellants contend that special masters are most useful at determining quantitative issues, but "[w]hen the issues referred to a master go beyond hard-to-measure damages or an accounting . . . the waters grow more turbid." Stauble v. Warrob, Inc., 977 F.2d 690, 694 (1st Cir. 1992). That observation was made in reference to a prior version of Rule 53, which provided that "in actions to be tried without a jury, save in matters of account and of difficult computation of damages, a reference [to a special master] shall be made only upon a showing that some exceptional condition requires it." Fed.R.Civ.P. 53(b) (as amended Apr. 30, 1991, effective Dec. 1, 1991) (emphasis added). Under Rule 53, as revised, special masters may now address pretrial and post-trial matters without a showing of exceptional conditions, so long as the circumstances of the appointment respect the conditions imposed by subsection (a)(1)(C). See Fed.R.Civ.P. 53(c).
In affirming the district court's decision, of course, we are not granting blanket approval to any future decision made by the special master. Decisions by the special master will be reviewable de novo by the district court and, in turn, reviewable by us on appeal.
D
The final objection is raised on behalf of franchisees who had been designated "Dismissed Plaintiffs," a term defined in the Settlement Agreement to include entities that opted out via a separate agreement. The problem first raised at the preliminary settlement hearing was that some or all of the franchisors that settled separately had included their franchisees in the release. The franchisees contested their dismissed status on the grounds that the franchisors lacked authority to release their claims and that, in any event, some of them dealt with more than one franchisor, not all of which had settled separately. The district court did not address the first objection, but did direct issuance of a "notice of exclusion," informing the franchisees that they were excluded from the class, and that, because their claims had been settled, they would not be eligible for class settlement funds unless they had also accepted payment in a capacity other than as a Dismissed Plaintiff.
At the final approval stage, exclusion was again contested. In response to the claim that separate settlements were entered without the consent of the franchisees, the district court instructed Class Counsel to send supplemental notice informing all entities that had received a "notice of exclusion" that that they could "make a claim for class settlement funds if [the settling entity] did not have authority to settle and release [their] claims." App'x at 7220. Appellants contend that this supplemental notice violated the Due Process Clause because it was sent only after final approval of the Settlement Agreement, thereby preventing the franchisees from opting out of or objecting to the settlement. We disagree.
"The standard for the adequacy of a settlement notice in a class action under either the Due Process Clause or the Federal Rules is measured by reasonableness." Wal-Mart Stores, Inc. v. Visa U.S.A., Inc., 396 F.3d 96, 113 (2d Cir. 2005). "There are no rigid rules to determine whether a settlement notice to the class satisfies constitutional or Rule 23(e) requirements; the settlement notice must fairly apprise the prospective members of the class of the terms of the proposed settlement and of the options that are open to them in connection with the proceedings." Id. at 114 (internal quotation marks omitted).
There was nothing unreasonable about the supplemental notice. The district court ordered these notices out of caution, after an objection raised at the final approval stage. There was no need then to provide an opt-out because all entities receiving the notice were already presumed to be excluded from the class. Nor was the notice needed to provide these entities with the opportunity to object; after all, it was the objection of a "Dismissed Plaintiff" that caused the supplemental notices to be sent in the first place.
All entities identified as Dismissed Plaintiffs would have presumably been bound by the release contained in separate settlement agreements; but the validity of those agreements could always be challenged in a separate action.
The district court acted (or reacted) reasonably in a sprawling case with many interested parties, in which neither the district court nor class counsel can be expected to predict and preempt every issue that might arise.
II
In exchange for the $5.6 billion, the settlement releases all claims that "accrue no later than five years after" the date on which appeals from the settlement become final. App'x at 3337 ¶ 31(a); see also id. at 3321-22 ¶ 3(ss). This feature of the release is said to violate two related rules.
Rule 23(a)(4). Newer merchants, who started accepting payment cards only toward the end of the class period, are said to be inadequately represented in violation of Rule 23(a)(4) because their interests diverged from those of the class representatives, all of which had accepted Visa and/or MasterCard for the full fifteen-year class period, and therefore had incentive to forgo five years (or more) of future claims in exchange for the cash that would afford newer merchants no more than a marginal recovery.
Rule 23(e)(2)(D). Relatedly, it is contended that the future release results in inequitable treatment among the class members in violation of Rule 23(e)(2)(D) because it requires class members to forgo relief for the same length of time, regardless of whether the class member was in business for fifteen years of the class period, or for the final month only.
We have no occasion to decide these questions. That is because the release of claims provision contains a de-facto severability clause, providing that the release "extend[s] to, but only to, the fullest extent permitted by federal law." App'x at 3337 at ¶ 31(a). This language ensures that the Settlement Agreement will stand even if certain aspects of the release were to fall. A holding as to the proper scope of the release--i.e., whether the future release violates federal law-can await a case in which the issue would directly affect the proceedings.
III
The district court granted $900,000 in service awards to the lead plaintiffs. Appellants contend that such awards are prohibited by Supreme Court precedent and that, even if they were generally permissible, the amount awarded in this case was excessive. We will address these two arguments in turn.
A
Service awards are likely impermissible under Supreme Court precedent. The Supreme Court has held that it was "decidedly objectionable" for cash allowances to be "made for the personal services and private expenses" of a creditor who brought suit on behalf of himself and other similarly situated bondholders. Trustees v. Greenough, 105 U.S. 527, 537 (1881). Such allowances, the Court reasoned, "would present too great a temptation to parties to intermeddle in the management of valuable property or funds in which they have only the interest of creditors, and that perhaps only to a small amount." Id. at 538.
Appellants argue that Greenough precludes the granting of service awards in this case, and in virtually all other cases--as the Eleventh Circuit held in Johnson v. NPAS Solutions, LLC, 975 F.3d 1244 (11th Cir. 2020). True, Greenough's holding was not about persons designated as class representatives under the later-formulated class action rules; but it "involved an analogous litigation actor." Id. at 1259. Finally, appellants argue that although Rule 23 post-dates Greenough, it makes no reference to service awards, and is thus "irrelevant." Id.
Service awards are expressly allowed by statute for class representatives in private securities litigation. See 15 U.S.C. § 78u-4(a)(4).
But practice and usage seem to have superseded Greenough (if that is possible). See Melito v. Experian Mktg. Sols. Inc., 923 F.3d 85, 96 (2d Cir. 2019); Hyland v. Navient Co., 48 F.4th 110, 123-24 (2d Cir. 2022). And even if (as we think) practice and usage cannot undo a Supreme Court holding, Melito and Navient are precedents that we must follow.
B
Appellants contend that the service awards here are excessive, even if service awards are otherwise deemed permissible, and must be reduced.
We have articulated no standard by which district courts can consider the grant of service awards. District courts within this Circuit have generally relied upon the following factors:
[1] the existence of special circumstances including the personal risk (if any) incurred by the plaintiff-applicant in becoming and continuing as a litigant, [2] the time and effort expended by that plaintiff in assisting in the prosecution of the litigation or in bringing to bear added value (e.g., factual expertise), [3] any other burdens
sustained by that plaintiff . . . and, of course, [4] the ultimate recovery.Dial Corp. v. News Corp., 317 F.R.D. 426, 439 (S.D.N.Y. 2016) (ellipsis in original) (quoting Roberts v. Texaco, 979 F.Supp. 185, 200 (S.D.N.Y. 1997)). District courts also "often look to the sums awarded in similar cases, and compare the named plaintiff's requested award to each class member's estimated pro rata share of the monetary judgment or settlement." In re AOL Time Warner ERISA Litig., No. 02-cv-8853, 2007 WL 3145111, at *2 (S.D.N.Y. Oct. 26, 2007) (citations omitted). Given the Supreme Court's precedent in apparent opposition to the practice, and the standardless use of it nevertheless, it is unsurprising that, as has been said, "the decision to grant the [service] award, and the amount thereof, rests solely within the discretion of the Court." Dial, 317 F.R.D. at 439.
Here, the district court bestowed a total of $900,000 in service awards to the eight lead plaintiffs, with the two highest awards in the amount of $200,000. The district court justified these awards as follows:
• "Courts have approved similarly-sized service awards, even in cases with smaller settlement fund sizes and for cases that proceeded for a shorter duration of time."
• "Class Representatives spent an enormous amount of time and resources in serving as named representatives in this complex litigation that has spanned well over a decade."
• "There is no indication that any Class Representative assumed the position or undertook the time and labor they did in anticipation of being compensated."
• "Class Counsel does not seek service awards in excess of the service awards requested and previously granted pursuant to the now-vacated 201[2] Settlement Agreement, despite significant additional efforts and time spent by the Class Plaintiffs in the six years since that time, including time spent as a result of another round of discovery."App'x at 7455-56 (citations omitted). Appellants challenge the amount of these awards on four grounds.
First: Appellants argue that the lead plaintiffs cannot adequately represent the class because the service awards dwarf the recovery for the lead plaintiffs and the average class member. As Appellants point out, certain lead plaintiffs will receive service awards that are about 100 times as large as their expected settlement recovery. Thus, Photos Etc., with an estimated claim of about $2,000, gets a service award of $200,000--more than 400 times the average recovery, which is less than $500. We recognize that certain district courts have relied on these sorts of comparisons; but we have never required their use, and we decline to do so now.
In 2014, Judge Gleeson questioned the requested service awards based on the reasons that Appellants now articulate. See In re Payment Card Interchange Fee & Merch. Disc. Antitrust Litig., 991 F.Supp.2d 437, 448-49 (E.D.N.Y. 2014). But Judge Gleeson's reservations are not law of the case, and Chief Judge Brodie, to whom the case was reassigned after Judge Gleeson's retirement, had no obligation to follow suit.
Second: The district court credited each lead plaintiff for the full time it spent on the litigation. Appellants contend that, to avoid rewarding inefficiency, the district court should have limited its consideration to the average number of hours expended by the lead plaintiffs with the largest claims--CHS and Payless. This idea is more creative than it is compelling, or workable. We cannot say that the district court's approach was an abuse of discretion.
Third: Payless reported that participation in this case cost it $70,000 in salary and wages; and CHS reported $39,000. Appellants contend that service awards should have been directly tied to those losses, and that the district court's failure to do so amounted to an abuse of discretion. Service awards have been limited to lost wages and out-of-pocket expenses in the context of private federal securities litigation. See 15 U.S.C. § 78u-4(a)(4) (limiting service awards to "reasonable costs and expenses (including lost wages) directly relating to the representation of the class"). The district court in this antitrust case was not bound by the limitation.
Fourth: Appellants contend that much of the work that the lead plaintiffs performed was geared toward obtaining legislative reform for the injunction class, and that those efforts should have been excluded in arriving at a service award for the damages class. Given that the basis for any service award in a class action is at best dubious under Greenough, and that, unsurprisingly, calculation of such an award is standardless, it is difficult to find traction for a ruling that this award is an abuse of discretion. In any event, among the variety of factors considered, the district court made no more than passing reference to time that lead plaintiffs spent on legislative activities. But the point here is valid. The class should not pay for time spent lobbying for changes in law that do not benefit the class. We direct the district court to reduce the award to the extent its size was increased because of time spent lobbying.
IV
The award of attorneys' fees exceeds half a billion dollars. The district court acknowledged that the "sheer size" of this award resulted in "apparent unfairness," but found that "public policy interests weigh in favor of a substantial award." In re Payment Card Interchange Fee &Merch. Disc. Antitrust Litig., No. 05-md-1720, 2019 WL 6888488, at *21 (E.D.N.Y. Dec. 16, 2019) ("Fee Opinion"). True, the amount is breathtaking; but the only question for us is whether the district court abused its discretion in awarding such fees. We cannot say that it did.
A
It is well-established under the common fund doctrine that "attorneys who create a fund for the benefit of a class of plaintiffs are entitled to reasonable compensation from that fund." Victor v. Argent Classic Convertible Arbitrage Fund L.P., 623 F.3d 82, 84 (2d Cir. 2010). This reasonableness principle has deep roots in Supreme Court jurisprudence, stemming from the decisions in Greenough and Central Railroad &Banking Co. v. Pettus, 113 U.S. 116 (1885). While this Circuit's approach to calculating fees has "evolved in a somewhat circuitous fashion," Goldberger v. Integrated Res., Inc., 209 F.3d 43, 48 (2d Cir. 2000), we established at the turn of this century that district courts could calculate fees using either the lodestar amount, which is the reasonable hourly rate multiplied by the hours reasonably expended, or a percentage of the fund, see id. at 50.
Regardless of which method is chosen, the analysis is effectively the same. In calculating a reasonable common fund fee, district courts are to be guided by the following factors: "(1) the time and labor expended by counsel; (2) the magnitude and complexities of the litigation; (3) the risk of the litigation . . .; (4) the quality of representation; (5) the requested fee in relation to the settlement; and (6) public policy considerations." Id. (alteration in original) (citation omitted). What is more, district courts that use the percentage method routinely keep the lodestar in sight, heeding Goldberger's advice to "requir[e] documentation of hours as a cross check on the reasonableness of the requested percentage." Id. (internal quotation marks omitted).
The district court here followed the trend in this Circuit by applying the percentage method. See Fee Opinion, No. 05-md-1720, 2019 WL 6888488, at *9, *24. Class Counsel had requested 9.56% of the settlement fund, which amounted to about $537 million. See id. at *3. The district court found that the requested award was supported by each of the Goldberger factors, except for the quality of representation. See id. at *10-21. While the district court found that Class Counsel did "a tremendous job in litigating this case," id. at *15, it explained that "the low percentage of recovery in comparison to the total alleged damages figure weighs slightly against granting the full amount of Class Counsel's requested fee," id. at *17.
The district court noted that the $5.6 billion settlement figure is less than the interchange fees that one retailer--Wal-Mart Stores, Inc.--paid in a single decade. See id. at *16.
The district court then conducted a lodestar cross-check, dividing the percentage fee requested ($537 million) by the lodestar ($215 million). It concluded that the resulting multiplier of 2.5 fell "well within a range of multipliers that have been deemed acceptable, especially in complex actions." Id. at *22 (citing Wal-Mart Stores, Inc. v. Visa U.S.A., Inc., 396 F.3d 96, 123 (2d Cir. 2005) (finding a multiplier of 3.5 to be reasonable in a megafund antitrust action)).
The district court ultimately adjusted the lodestar amount to about $213 million, after deducting time spent exclusively on injunctive relief after the remand. See id. at *24.
The district court then decided to reduce the requested percentage by .25% to account for the relatively small settlement figure, as a percentage of the amount of damages reasonably claimed. See id. at *24. This resulted in an award of 9.31% of the common fund, or roughly $523 million. See id. The final lodestar multiplier was 2.45. See id.
B
Appellants first challenge the roughly 630,000 hours that Class Counsel is found to have expended on the case. Specifically, Appellants contend that the district court abused its discretion by failing to either: (1) exclude from its analysis all pre-remand time--about 500,000 hours--because of Class Counsel's conflicting representation of the damages and injunction class; or (2) limit preremand time to exclude time that benefitted the injunction class only. Appellants argue that the district court's failure to adjust the pre-remand time in either of these ways allowed inflated hours to distort its Goldberger review (specifically, the time and labor factor) and its lodestar cross-check (by juicing up the lodestar).
1
Appellants' first argument seeks a categorical exclusion. That is, Appellants claim that Class Counsel should not have received credit for any preremand work, since attorneys cannot be compensated for their simultaneous representation of competing parties. We are unpersuaded.
Appellants also summarily state that even if the nature of the conflict alone does not preclude recovery for pre-remand work, "[o]f course, when an attorney creates a situation in which it is impossible to accurately bill separate clients . . . the remedy is to preclude the recovery of any attorney's fees from either client." Gnarlywood Br. at 43. But Appellants cite no authority for this proposition.
Appellants' argument hinges on Silbiger v. Prudence Bonds Corp., 180 F.2d 917 (2d Cir. 1950) (L. Hand, J.), in which the district court awarded attorney's fees from a bankruptcy reorganization in which the attorney represented two classes of bondholders with directly opposing interests, such that one class would be "certain to lose by the success of the [other]." Id. at 920. Judge Hand noted that the "usual consequence" of this sort of conflicting representation was that the offending attorney would be "debarred from receiving any fee from either [client], no matter how successful his labors." Id.
It is unclear whether that rule applies in the class action context. See In re Austrian &German Bank Holocaust Litig., 317 F.3d 91, 104 n.17 (2d Cir. 2003) ("We do not believe that the strict approach taken in Silbiger, which was not a class action, requiring forfeiture of at least one-third of the attorney's fee, is applicable in the context of the pending litigation." (citation omitted)). In any event, Silbiger noted an "exception" to the general rule: "it is reasonable not to impose an entire forfeiture of the allowance," when the award "comes in no part out of any group that [could] have been prejudiced by the attorney's divided allegiance." 180 F.2d at 921. This language provides the "salient distinguishing feature relied upon in Silbiger." New York, N.H. &H.R. Co. v. Iannotti, 567 F.2d 166, 176 (2d Cir. 1977).
Silbiger concluded that there needed to be some "penalty" for the attorney's failure to notify the court of the conflict, and the case was remanded with instructions to reduce the fee by at least one-third. 180 F.2d at 919, 921. Appellants do not argue for a similar partial forfeiture--at least not as a penalty for conflicted representation.
The exception applies with full force to this case. All fees will come out of the common fund created for the damages class--i.e., the class which always stood to benefit from Class Counsel's conflicted representation. The injunction class will pay none of it. So total forfeiture is neither required nor appropriate.
This analysis is unaffected by Appellants' attempt to cast Class Counsel's representation as an "ethical violation." Gnarlywood Br. at 48. Class Counsel never violated the New York Rules of Professional Conduct, see Fee Opinion, No. 05-md-1720, 2019 WL 6888488, at *18, and we expressly declined to "impugn the motives or acts of class counsel" on the last appeal, In re Payment Card Interchange Fee & Merch. Disc. Antitrust Litig., 827 F.3d at 234. Indeed, it was only in the context of a settlement (where relief for one party could be used as a bargaining chip to enhance relief for another) that a conflict between the classes arose. Class Counsel could have represented both classes without any conflict had the case proceeded to trial and to a final disposition.
2
Alternatively, Appellants contend that if Class Counsel is compensated for time spent prior to remand, there should be exclusion of hours spent pursuing injunctive-related relief--e.g., time pursuing extra-judicial legislative changes, and supporting third-party investigations and litigations. This exclusion, we are told, is not intended to be a penalty for conflicted representation; instead, it seeks to ensure that the damages class does not pay more than its fair share. This argument has some appeal; but we conclude that the district court acted within its discretion when it decided to include the pre-remand time in its entirety.
Appellants rely on Hensley v. Eckerhart, 461 U.S. 424 (1983), which considered whether prevailing counsel can be compensated for time spent on unsuccessful claims under a fee-shifting statute providing that "the court, in its discretion, may allow the prevailing party . . . a reasonable attorney's fee." Id. at 426 (quoting 42 U.S.C. § 1988) (emphasis added). Hensley observed that when a plaintiff "present[s] in one lawsuit distinctly different claims for relief that are based on different facts and legal theories," id. at 434, "counsel's work on one claim will be unrelated to his work on another claim" and, thus, "work on an unsuccessful claim cannot be deemed to have been expended in pursuit of the ultimate result achieved," id. at 435 (internal quotation marks omitted). Appellants latch on to this statement.
But the Supreme Court went on to consider cases in which "the plaintiff's claims for relief will involve a common core of facts or will be based on related legal theories." Id. In those cases, the Court reasoned, "[m]uch of counsel's time will be devoted generally to the litigation as a whole, making it difficult to divide the hours expended on a claim-by-claim basis." Id. As a result, attorneys in such cases may still "recover a fully compensatory fee." Id.
Hensley's reasoning is instructive here. In considering the time and labor expended on litigation (the first Goldberger factor), it makes little sense to credit counsel for work performed on behalf of a class that it no longer represents. So if the work done on behalf of the injunction class could all be teased apart from the work done for the damages class, the district court should have done so. But that was not an option. The injunctive and monetary claims were interwoven, such that "the majority of Class Counsel's work leading up to the 201[2] Settlement Agreement would have been aimed generally at proving antitrust violation, regardless of the particular remedy sought or class represented." Fee Opinion, No. 05-md-1720, 2019 WL 6888488, at *11.
Under the circumstances, "[t]here was simply no need for the district court to engage in an artificial distribution of attorney time between [monetary] and [injunctive] claims." Dague v. City of Burlington, 935 F.2d 1343, 1359 (2d Cir. 1991) (internal quotation marks omitted), rev'd in part, 505 U.S. 557 (1992).
C
Finally, Appellants raise three challenges to the lodestar multiplier, which was pegged at 2.45.
First: Appellants contend that Class Counsel artificially inflated the lodestar by failing to weed out the time spent working for injunctive-related relief. Had this time been excluded, and the lodestar lowered, the multiplier would have exceeded 2.45 and might have fallen out of a range deemed reasonable. As explained above, however, the district court acted within its discretion when it declined to exclude hours spent solely on behalf of the injunction class, which was in any event small in number relative to the 630,000 total hours.
Second: Appellants challenged the district court's comparison of the lodestar multiplier in this case to multipliers found reasonable in other similar actions. See Fee Opinion, No. 05-md-1720, 2019 WL 6888488, at *22. But in Wal-Mart, we endorsed the use of lodestar comparators, see 396 F.3d at 123, which would seem necessary to avoid picking numbers arbitrarily.
According to Appellants, such comparisons go beyond the factors laid out in Goldberger, which characterized the practice of fixing a "benchmark" as an "all too tempting substitute for the searching assessment that should properly be performed in each case." Goldberger, 209 F.3d at 52. But Goldberger eschewed benchmarks only for the percentage of the common fund--i.e., it counseled against pegging the percentage at 25%. It said nothing about applying benchmarks to the lodestar multiplier.
Third: The district court justified the lodestar multiplier by reference to the "significant litigation risk." Fee Opinion, No. 05-md-1720, 2019 WL 6888488, at *22. According to Appellants, that was an abuse of discretion because the risks identified--e.g., the challenge of proving antitrust standing--are not of the sort that we deemed applicable in Fresno County Employees' Retirement Association v. Isaacson/Weaver Family Trust, 925 F.3d 63 (2d Cir. 2019). That case, Appellants posit, relied exclusively on the risk generated from counsel's devotion of resources to a specific case while forgoing other paying work.
Appellants' reading of Fresno County is too narrow. The true risk we identified in Fresno County was the risk that a case taken on contingency will fail:
The plaintiff class is . . . appropriately charged for contingency risk where such risk is appreciable because the class has benefited from class counsel's decision to devote resources to the class's cause at the expense of taking other cases. That is, because class counsel has decided to represent the plaintiff class, class counsel's ability to freely represent other clients is limited by the risk she has assumed that the class's cause will be unsuccessful.925 F.3d at 70 (emphasis added). The district court reasonably concluded that the significant litigation risk present in this case meant that class counsel had taken on a venture with a high risk of failure, and that the risk should be compensated.
Finally: Appellants argue that the straight lodestar--i.e., the figure without any multiplier--is the presumptively reasonable fee in cases (such as this one) initiated under fee-shifting statutes. But we rejected that same argument in Fresno County, explaining that "an attorney seeking a fee after establishing a common fund will receive a fee calculated using either the lodestar method or a percentage-of-the-fund method, which can yield a fee that is less than, equal to, or greater than the lodestar fee." Id. at 68. We further explained that where, as here, "statutory fees and the common-fund doctrine collide, the common-fund doctrine operates autonomously from fee-shifting principles." Id. at 69. The district court therefore acted within its discretion when it granted Class Counsel a fee award that more than doubled the lodestar.
The case was filed under the Sherman Act, which provides that persons "injured in [their] business or property by reason of anything forbidden in the antitrust laws . . . shall recover threefold the damages . . . sustained, and the cost of suit, including a reasonable attorney's fee." 15 U.S.C. § 15(a).
CONCLUSION
For the foregoing reasons, we AFFIRM in all respects the district court's orders, to the extent they constituted a final judgment, with the exception that we direct the district court to reduce the service award to class representatives to the extent that its size was increased by time spent in lobbying efforts that would not increase the recovery of damages. We make no ruling as to how damages should be allocated as between branded oil companies and their branded service station franchisees, the reasonableness of the special master's ultimate findings, or the legality of releasing an as-of-yet hypothetical future claim; such issues must wait for a developed factual record and a final judgment, and, as to the breadth of the releases, the assertion by the defendants in a future litigation of a claim of release which is challenged on the ground that interpreting the release so broadly violates federal law.
This panel will retain jurisdiction over any further appeals in this case (though not any appeal from the grant or denial of injunctive relief).
DENNIS JACOBS, Circuit Judge, concurring:
Visa and MasterCard agreed to pay more than $5.6 billion to a class of roughly 12 million merchants for alleged antitrust violations built into contractual arrangements governing credit and debit cards. In return, the plaintiffs agreed to release any claims that were or could have been brought during the class period of fifteen years, as well as any related claims that accrued for an indeterminate number of years after the class period closed.
As the Opinion of the Court demonstrates, approval of the settlement in this case did not constitute an abuse of discretion under controlling principles. Nevertheless, several results produced by application of those principles in this case are so remarkable as to prompt disquiet.
I
1. Lawyers for the plaintiffs billed 630,000 hours. That is over 300 lawyer years. One of plaintiffs' lawyers billed over 14,000 hours; three others billed between 10,800 and 12,000 hours; and no fewer than 38 lawyers lavished over 2,000 hours on this case. They are to be richly rewarded: compensation to Plaintiffs' counsel will be half a billion dollars. The average hourly take-home rate for all the lawyers (partners and rookies alike) exceeds $1,000. The average hourly compensation for all those who billed in the plaintiffs' cause, lawyers as well as paralegals and such, is about $800.
2. The substantive issue in the case is whether a handful of standard-form contracts between the card issuers and the merchants (and their banks) violated the antitrust law. It would not seem that the question would entail heroic labors. Even so, that issue - the merits -- was preempted by settlement and was never decided.
3. As in many antitrust suits, there was here a preliminary question as to whether certain plaintiffs are the persons who can sue to vindicate the claim that is alleged. That was not settled either. As to that, Judge Leval's concurrence raises a pointed question.
4. Under this settlement, the card companies may continue the allegedly anticompetitive conduct, while members of the class agree to forbear from suing about it for five years. After five years, a new litigation over the same unresolved issues can be started; and why would it not be? A jaundiced observer might view the case as a delivery system for attorneys' fees.
5. If, as is likely, there is to be injunctive relief, it will be the subject of further litigation by other teams of lawyers who even now are beavering away, taking the 80 depositions they have noticed and reviewing the millions of documents they have demanded. Compensation for that work will be extra.
6. The average award, if every class member claims every dollar, will be $325. But no one knows how much of the $5 billion settlement amount will be claimed and actually paid out. Whatever is left over, the trial judge will be authorized to donate for unspecified good causes, defined as the "next best compensation use, e.g., for the aggregate, indirect, prospective benefit of the class." Masters v. Wilhelmina Model Agency, Inc., 473 F.3d 423, 436 (2d Cir. 2007) (quoting 2 HERBERT B. NEWBERG &ALBA CONTE, NEWBERG ON CLASS ACTIONS § 10:17 (4th ed.2002)). The judge may become one of that year's most munificent philanthropists. A lesser payment to the class itself will not, however, impair the lawyers' fees.
II
The half billion in legal fees is roughly ten percent of the class recovery, and thus appears plausible, prudent and restrained. But the principles for computing attorney's fees in a class action operate reciprocally to drive up the total.
The trend in this Circuit is to award fees based on the percentage-of-the-fund method, which nearly always results in an award that exceeds the lodestar method: i.e., the number of hours times the billing rate. Here, the percentage award exceeded the lodestar by more than $300 million. It is thought that the percentage method "directly aligns the interests of the class and its counsel and provides a powerful incentive for the efficient prosecution and early resolution of litigation." Wal-Mart Stores, Inc. v. Visa U.S.A., Inc., 396 F.3d 96, 121 (2d Cir. 2005) (citation omitted). As this case demonstrates, however, the benefit can be illusory.
The problem is that fee awards are usually computed at the end of a case, which prevents counsel from knowing which method of compensation will be used. This uncertainty can be an incentive for class counsel to maximize the time expended, as a hedge against the possibility that the lodestar method will be used. And the incentive to maximize the recovery for the class is diminished since counsel does not know what percentage (if any) of the increment will be awarded.
The lodestar cross-check is supposed to correct the incentive, but does not. If the percentage award exceeds the lodestar figure by too high a multiple, district courts will reduce the award to shrink the gap. See, e.g., In re Visa Check/Mastermoney Antitrust Litig., 297 F.Supp.2d 503, 522 (E.D.N.Y. 2003) (reducing multiplier to 3.5 after finding the requested multiple of 10 to be "absurd"). The cross-check is thus an incentive for counsel to prolong litigation and maximize billable hours to arrive at a lodestar that does not operate as a cap on a percentage award. It is only fair to add that I am not in a position to know whether these or any abuses occurred in this case.
III
The named plaintiffs in this case are deemed private attorneys general. Yet they are getting a total of $900,000 as a kind of tip. That is $900,000 more than permitted under Supreme Court authority. See Trustees v. Greenough, 105 U.S. 527 (1882). The Eleventh Circuit held as much in Johnson v. NPAS Solutions, LLC, 975 F.3d 1244 (11th Cir. 2020), and I am in accord with the views set forth in that thorough and well-reasoned opinion.
But this Court has twice come out the opposite way. First, Melito v. Experian Marketing Solutions, Inc., 923 F.3d 85 (2d Cir. 2019) affirmed the grant of so-called service awards, stating, without elaboration, that Greenough (and the Supreme Court's less relevant decision in Central Railroad &Banking Co. v. Pettus, 113 U.S. 116 (1885)) did not "provide factual settings akin to those here." Id. at 96. Then, recently, Hyland v. Navient Corp., 48 F.4th 110, 124 (2d Cir. 2022) over-read Melito to hold that "Rule 23 does not per se prohibit service awards like the one at issue here."
Melito can be best understood as an attempt to avoid a split with the Eleventh Circuit, which had held - in an opinion later vacated - that Greenough did not categorically prohibit service awards. See Muransky v. Godiva Chocolatier, Inc., 922 F.3d 1175, 1196 (11th Cir. 2019), reh'g en banc granted, opinion vacated, 939 F.3d 1278 (11th Cir. 2019), and on reh'g en banc, 979 F.3d 917 (11th Cir. 2020). But because the Eleventh Circuit changed course in Johnson, we now find ourselves on the wrong side of a circuit split.
Some fear that to do away with service awards would "do away with the incentive for people to bring class actions." Johnson, 975 F.3d at 1264 (Martin, J., concurring in part and dissenting in part). But the district court in this case found that "[t]here is no indication that any Class Representative assumed the position or undertook the time and labor they did in anticipation of being compensated." App'x at 7455-56. That may not always be the case. Perhaps class actions that plaintiffs lack incentive to bring are class actions that need not be brought.
IV
For all the attorneys' fees, one would hope that this case vindicated the public interest in antitrust enforcement. But it is just the opposite. Subject only to a possible future injunction, Defendants can continue their allegedly anticompetitive behavior, with Plaintiffs releasing all claims that will "accrue no later than five years" after all appeals from the settlement become final. App'x 3321, 3367-68.
For the class representatives, the release of claims accruing after the class period may well be a sacrifice worth making; all of them were in business for the full fifteen-year class period. But the same cannot be said for the newer merchants--i.e., those that started accepting Visa and/or MasterCard toward the end of that period. These newer merchants stood to gain only marginally from the settlement and are now releasing claims for a period that may far exceed their period of recovery.
It is arguable that the future release: (A) resulted in newer merchants receiving inadequate representation from the class representatives; (B) caused class members to be treated inequitably relative to each other; and therefore (C) should be stricken from the settlement agreement.
Ordinarily, we would reject a settlement which involved inadequate representation or inequitable treatment. That is what we did on the last appeal. But no such action need be considered this time around because the settlement agreement contains a safety valve: the release of claims extends "to, but only to, the fullest extent permitted by federal law." App'x 3336-37.
A holding as to the legality of the future release will therefore await the day that a class member brings suit based on claims that accrued after the close of the class period. (I suspect there will be a line of counsel willing to assist.) When such a case is brought, Visa and MasterCard will undoubtedly raise the future release as an affirmative defense. And the district court will have to make a ruling.
LEVAL, Circuit Judge, concurring:
I concur in Judge Jacobs's majority opinion and write separately to address an important question, which appears to have been overlooked in the litigation. In considering the dispute between the oil companies and their branded service stations over settlement funds paid by the defendants on account of customer payments by payment card for gas purchased at the service stations, the parties, and perhaps also the district court, appear to be making an unquestioned assumption that, under the rule of Illinois Brick, only one of the oil company or its branded service station may be deemed to have accepted a card payment. The participants appear to assume that only one of the two can have standing to win any part of the settlement funds attributable to a card payment, so that a division between them of those settlement funds would violate the Illinois Brick rule. See In re Payment Card Interchange Fee &Merchant Discount Antitrust Litig., No. 05-md-1720, 2019 WL 6875472, at *30-31 (E.D.N.Y. Dec. 16, 2019). In my opinion, this assumption is ill founded.
Illinois Brick holds only that an indirect purchaser from the antitrust violator, i.e., one whose purchase was downstream from the direct purchaser, may not recover for an antitrust injury. Illinois Brick Co. v. Illinois, 431 U.S. 720, 727-29 (1977). In Hanover Shoe, the Supreme Court had held that the direct purchaser from the antitrust violator, who buys at a price inflated by the antitrust violation, is deemed to suffer the full brunt of the antitrust injury, even if the direct purchaser raised his price to the next purchaser down the line (an "indirect" purchaser) so as to pass on the brunt of the injury. See Hanover Shoe, Inc. v. United Shoe Machinery Corp., 392 U.S. 481, 488-89 (1968). Hanover Shoe thus established that the antitrust violator cannot avoid liability to the direct purchaser by showing that the direct purchaser, by raising its price to the next purchaser in line, had avoided any loss. See id. In Illinois Brick, the Court espoused a corollary rule to the effect that, in the same scenario of a chain of purchases, it is only the direct purchaser from the antitrust violator (the one that is deemed to have borne the full brunt of the harm), and not indirect purchasers further down the line, who may sue the violator. Illinois Brick, 431 U.S. at 735-36. Accordingly, when illegal overcharges are passed down the "chain" of successive purchases, id. at 729, (absent specified exceptions) only the first purchaser may sue. Simply put, Illinois Brick provides that, where antitrust violator (A) has sold to direct purchaser (B) who then sold to another customer (C), B may sue A for the violation, but C may not. Apple Inc. v. Pepper, 139 S.Ct. 1514, 1521 (2019).
There are limited exceptions to the Illinois Brick rule. For example, an indirect purchaser may recover for its antitrust injury if it owns or controls the direct purchaser or if there was an explicit cost-plus contract between the direct and indirect purchaser, such that the full antitrust injury is clearly traceable to the indirect purchaser. See California v. ARC Am. Corp., 490 U.S. 93, 97 & n.2 (1989) (citing Illinois Brick, 431 U.S. at 732 n.12, 736 n.16).
Nowhere, however, does Illinois Brick state the wider proposition that there can be only one antitrust plaintiff. Nor does its logic support that conclusion. While Illinois Brick's prohibition of recovery by an indirect purchaser who purchased from the direct purchaser commonly results in there being only one purchaser who is qualified to sue for the injury inflicted in that sale, it does not follow that, in the unusual circumstance in which two entities share a direct purchase, the two may not share the recovery that would have gone to one of them if that one had been the sole purchaser.
Consider a circumstance in which farmers, A and B, jointly purchase an expensive crop-processing machine (too expensive for either of them to afford on their own), sharing the cost of the purchase and agreeing to divide equally between them the use of the machine and share equally the costs of its maintenance. If the seller of the machine violated the antitrust laws, inflicting an antitrust injury on its direct purchasers, then A and B, having purchased the machine jointly, have equal claim to being a direct purchaser. Neither one is more a direct purchaser from the violator than the other, and neither one is an indirect purchaser, who bought from the direct purchaser. It cannot logically be that only one of the two would be eligible to sue, and would win 100% of the damages, to the exclusion of the other. If Illinois Brick meant (despite having said no such thing) that only one entity may sue as the direct purchaser, a court would have no principled way of determining whether A or B is the one entitled to sue. The ruling of Illinois Brick, like the ruling of Hanover Shoe, which it extended, applied to successive purchasers in a sequential chain. That ruling did not address and has no application to the unusual, and very different, circumstance in which two entities act jointly in making a direct purchase. When this occurs, the rule of Illinois Brick, as I read it, is not an obstacle to both of the joint direct purchasers suing to vindicate their shared antitrust injury. That circumstance was outside the scope of the Illinois Brick ruling.
Although a recent Seventh Circuit case has some language suggesting that only one entity may sue, see Marion Healthcare, LLC v. Becton Dickinson & Co., 952 F.3d 832, 840 (7th Cir. 2020) ("The central point of Illinois Brick is to allocate the right to recover to one and only one entity in the market."), in making that statement, the Seventh Circuit was considering successive purchasers in a chain, and was not contemplating whether two or more entities might be considered to have acted jointly in making the direct purchase from the violator. That case concerned a typical chain of manufacture, where healthcare companies, the plaintiffs, used "Group Purchasing Organizations" to purchase medical devices from manufacturers. Id. at 836-37. The healthcare companies sued both the intermediary purchasing organizations and the more remote manufacturer, on the theory that those two groups of defendants had conspired together to jointly violate the antitrust laws, so that the plaintiffs purchased directly from the conspiracy. Id. The district court dismissed the suit, concluding that it violated the Illinois Brick rule. Id. at 837. The Seventh Circuit vacated that decision, concluding that if the complaint alleged a plausible conspiracy between the two groups of defendants, the plaintiffs' suit against both the intermediaries and the remote co-conspirators was permissible under Illinois Brick, because the plaintiffs had purchased directly from the conspiracy. Id. at 840-41, 843. Thus, while the Seventh Circuit has said that Illinois Brick allows only one plaintiff, it meant that in a chain of successive purchasers, each claiming to have suffered a part of the injury caused by the defendant's antitrust violation, only the direct purchaser from the defendant may sue. The court was not considering the question posed here-whether, consistent with Illinois Brick, two plaintiffs may share a recovery for an injury suffered by them on jointly making a direct purchase from the violator. The Seventh Circuit's statement that Illinois Brick allows "one and only one" plaintiff per antitrust violation appears to be another instance of misconstruing that opinion as going far beyond the question it was considering.
Furthermore, to the extent that the Supreme Court was concerned, in fashioning the Illinois Brick rule, about multiplication of claims for treble damages, see Illinois Brick, 431 U.S. at 730-31 ("[Allowing offensive but not defensive use of pass-on would create a serious risk of multiple liability for defendants.... [W]e are unwilling to open the door to duplicative recoveries . . . ." (internal quotation marks omitted)), allowing joint direct purchasers to sue for their shared injury would not undermine that concern. They would not be adding claims, but rather dividing between them the damages attributable to a single direct purchase.
As of yet, so far as we are aware, the district court has not addressed the relationships between the integrated oil companies and their branded service stations. The district court has stated an intention to appoint a special master to study that question and decide (subject to the district court's de novo review) how to distribute the settlement monies paid by the defendants on account of the retail sales at the service stations that were paid by payment card. Because this issue has not yet been explored in the district court, we do not have information that illuminates the relationship between the oil companies and their branded service stations. Furthermore, that relationship may vary from company to company, and perhaps also from service station to service station. Nonetheless, it appears likely that at least some of the relationships between the oil companies and their branded service stations (with respect to the question of who accepted the retail customers' payment cards, thereby purchasing the card-payment service rendered by the defendants) are not the sequential direct and indirect purchases addressed by Illinois Brick, but are more in the nature of a shared joint purchase of the defendants' payment card service. It appears at least plausible that, in buying gas at the pump with a payment card, the retail customers were buying a portion of the goods and services involved in the transaction from the oil company and a portion from the service station, so that, under the agreement between oil company and service station, a portion of the payment would ultimately go to each of them. If so, it may be that the oil company and service station have jointly sold gas to their retail customer, have jointly accepted the customer's payment card, and have jointly made a direct purchase from the defendant card company of the payment service it rendered. If that proves to be true, the two may stand in shared status as joint direct purchasers of the card payment service, each entitled to sue for its share of the antitrust injury.
Not knowing the facts that define each of the relationships between oil companies and their service stations with respect to gas sales at the pump (which have not yet been determined in the district court), I can express no view how the settlement payments on account of those sales should be distributed. I can only say that, in my view, it would be a mistake for the court or the parties to assume (as some appear to have done) that, under Illinois Brick, there can be no more than one direct purchaser per transaction. The parties, the special master, and the district court should carefully study the precedents to explore whether the Illinois Brick rule, which unquestionably denies standing to an indirect purchaser from the violator, similarly denies standing to one of two entities when the two have joined in making a direct purchase from the violator, notwithstanding the absence of any persuasive reason for choosing one over the other as the authorized plaintiff.
Judge Jacobs agrees that the inquiry suggested in this paragraph is appropriate.
Furthermore, should any oil company-service station pairs reach a settlement between them, agreeing to a formula for division between them of the settlement funds, I can see no reason why a court would not give effect to that settlement.