In re Visa Check/Mastermoney Antitrust Litigation , 192 F.R.D. 68 ( 2000 )

MEMORANDUM AND ORDER

GLEESON, District Judge.

In this antitrust action, several of the nation’s largest retailers, joined by a number of smaller merchants and three retail associations, challenge rules issued by defendants Visa and MasterCard that require stores accepting defendants’ credit cards to also accept their debit cards. Plaintiffs allege that this is a tying arrangement,.and that it is per se illegal under § 1 of the Sherman Antitrust Act, 15 U.S.C. § 1. They also contend that defendants have attempted and conspired to monopolize the debit card market in violation § 2 of the Sherman Act, 15 U.S.C. § 2.

Plaintiffs have moved for certification of their case as a class action pursuant to Rule 23 of the Federal Rules of Civil Procedure, with the proposed class comprising all individuals and businesses that have accepted Visa and/or MasterCard credit cards, and have therefore been required to accept the debit cards, within the statute of limitations period. Defendants oppose the motion and have themselves moved to strike the expert ’ opinion submitted by plaintiffs in support of class certification. For the reasons discussed below, the motion to certify the class is granted, and the motion to strike the expert is denied.

I. BACKGROUND

The plaintiffs in this action include WalMart Stores, Sears Roebuck, Safeway, Circuit City, the National Retail Federation, and a number of other merchants, large and small.¹ The defendants are Visa U.S.A. (“Visa”) and MasterCard International (“MasterCard”), associations owned and run by member financial institutions in order to license payment cards with the Visa and MasterCard logos and set some of the rules governing the handling of card transactions.

The following factual background is drawn from the plaintiffs’ Second Amended Consolidated Class Action Complaint (“Complaint”). Many of these assertions are of course vigorously disputed by the defendants.

A. Visa and MasterCard

Ail major American banks are members of both Visa and MasterCard, resulting in a 95 percent overlap in membership between the two associations. (Complaint 1143.) Likewise, nearly every merchant that accepts Visa also accepts MasterCard, and vice versa. (Id.) These facts lead to a phenomenon referred to as duality, in which Visa and MasterCard do not compete against each other and in fact coordinate many of their policies. (Id. 1144-45.)

B. Visa and MasterCard Transactions

Visa and MasterCard do not issue payment cards; they license member banks to do so. (Id. ¶8(m).) Those “card-issuing institutions” independently set cardholders’ interest rates and fees, although the policy of “duality” has “facilitated a high degree of uniformity” in those numbers. (Id. ¶18(m), 44.) Likewise, member financial institutions independently contract with retailers to accept the cards. (Id. ¶8(n).) Referred to in this capacity as “acquiring institutions,” they reimburse merchants for every Visa and MasterCard purchase, less a “discount fee.”² (Id. ¶¶18(n), 8(p).) The discount fee is largely based on the “interchange fee,” the fee that the acquiring institution pays the card-issuing institution every time it processes a payment by one of the card-issuing institution’s cardholders at one of the acquiring institution’s retailers. (Id. ¶8(o).) The “interchange fee” is set by Visa and MasterCard.³ (Id.)

C. Payment Cards

Payment cards come in a variety of forms. A credit card, such as a Visa or MasterCard credit card, allows a cardholder to make a purchase and pay off his card-issuing institution off over time. (Id. ¶8(d).) A “charge card” or “travel and entertainment card,” such as American Express’s green card, allows its holder to make a purchase on credit that extends only to the end of the monthly payment period. A debit card allows a holder to access his bank account directly. (Id. ¶8(f).) A debit card might be an ATM card, which enables the holder to withdraw cash from an ATM, or a “POS debit card,” which can be used to make a purchase at the “point of sale,” or both. (Id. ¶¶8(g), 8(h).)

POS debit card transactions are effectuated in one of two ways: “on-line” or “off-line.” In a debit transaction completed on-line, the cardholder enters his “personal identification number” or “PIN” into a PIN pad; the card-issuing institution verifies that there are sufficient funds present in the cardholder’s account; puts a “hold” on those funds; and transfers the money to the retailer’s account within a day. (Id. ¶8(k).) On-line debit networks include NYCE, MOST, Pulse, and Shazam. (Id.) In a debit transaction completed off-line, the cardholder does not enter a PIN but rather signs a slip as she would if paying by credit card. (Id. ¶8(1).) The card-issuing bank may or may not check the sufficiency of funds; may or may not put a hold on the purchase price; and will transfer the money from the cardholder’s account to the retailer’s in one to seven days. (Id.) Off-line debit transactions are less secure than those completed on-line, principally because fraud is more likely when a signature, as opposed to a PIN, is required. (Id. ¶56.) The VisaCheck and MasterMoney cards, which are at the center of this litigation, are off-line POS debit cards. (Id. ¶8(e).)

D. Visa and MasterCard Entry Into the Debit Card Market

Visa and MasterCard dominate the credit card market. By 1979, their share of the credit card market exceeded 90 percent; if the market is defined more broadly to in-elude charge cards, such as American Express, the share for Visa and MasterCard exceeded 70 percent. (Id. ¶47.) So prevalent were Visa and MasterCard that most retailers felt compelled to accept them. (Id.) Visa and MasterCard member institutions took advantage of the cards’ dominating position, and the members’ ability to coordinate action, to erect barriers to entry for potential competitors. (Id. ¶¶49-54.)

In 1979, Visa and MasterCard launched their off-line POS debit cards, VisaCheck and MasterMoney, respectively. (Id. ¶ 65.) Pursuant to the defendants’ “honor all cards” rule, any merchant accepting Visa or MasterCard credit cards was contractually obligated to accept VisaCheck and MasterMoney as well. (Id. ¶ 66.) Also linked were the interchange fees: Visa and MasterCard set them for the off-line POS debit cards at the same level as for credit cards. (Id. ¶ 67.) This equivalence occurred notwithstanding the fact that credit cards are much more expensive for banks (due to the risks inherent in extending credit) than are debit cards. (Id. 1168.) The fees associated with the Visa and MasterCard off-line POS debit cards far exceeded the fees for competing on-line POS debit networks. For example, in 1996 the interchange fee on a $100 retail transaction (at a non-supermarket) was $1.10 for a Visa-Cheek transaction or $1.31 for a MasterMoney transaction, but only $.05 for a MOST transaction. (Id. U 69.) Without the tie to credit cards, the plaintiffs allege that “[r]etailers would not pay these fixed, supra-competitive and extortionate VisaCheck and MasterMoney rates.” (Id. ¶ 73.)

In 1996, there were 1.2 billion retail transactions involving VisaCheck or MasterMoney, totalling $46 billion. (Id. ¶ 79.) These payment options are fungible with cash, checks, travelers checks, and on-line POS debit cards; therefore, retailers would have lost virtually none of those $46 billion in sales if they had not accepted VisaCheck or MasterMoney. The interchange fees on these transactions were $580 million; the costs incurred by retailers had they been able to refuse VisaCheck and MasterMoney and have customers pay through one of the other listed means would have been only $90 million. (Id. ¶¶ 82-83.)

Visa and MasterCard have undertaken a number of measures to “deceive” retailers about the off-line POS debit cards. (Id. ¶ 89.) They designed them to be “visually and electronically indistinguishable” from credit cards so that retailers would not even know they were accepting a different form of payment, and they set the interchange fees at the same level for debit and credit so that retailers would not notice the difference when they were billed.⁴ (Id. ¶¶ 92-93.) Since as early as 1982, numerous merchants discovered that they were accepting the cards, despite the “campaign of concealment,” and complained about it. But all were told that they must accept the cards if they wish to remain eligible for Visa and MasterCard credit cards. (Id. ¶¶ 98-110.)

In order to protect their positions in the growing market for debit payments, Visa and MasterCard set out to eliminate the competition offered by the “cheaper, faster and safer” on-line POS debit networks. (Id. ¶¶ 75, 113.) The efforts are succeeding. In 1995, the number of VisaCheck/MasterMoney transactions increased by 80 percent over 1994, while on-line POS debit transactions increased by only 40 percent. (Id. ¶ 119.) At the end of 1996, the defendants’ off-line POS debit cards had captured 65 percent of the POS debit card market. (Id. ¶ 120.)

E. This Action

The plaintiffs allege that the contractual requirement that they accept Visa and MasterCard debit cards along with the credit cards is an illegal tie pursuant to § 1 of the Sherman Act, 15 U.S.C. § 1. They also contend that the defendants have attempted and conspired to monopolize the debit card market in violation of § 2 of the Sherman Act, 15 U.S.C. § 2. The plaintiffs seek an injunction ending the alleged tie and treble damages.

Plaintiffs now move to certify a class of approximately four million merchants who have accepted Visa or MasterCard credit cards and who therefore have been required to accept VisaCheck or MasterMoney.

F. The Expert Reports

Both sides have introduced expert reports to support their respective positions in favor of and in opposition to class certification. Their analyses are summarized in brief here.

1. Dennis W. Carlton: Opening Report

The plaintiffs expert is Dennis W. Carlton, a professor of economics at the University of Chicago’s Graduate School of Business, who has published widely on industrial organization, antitrust topics, and the credit card industry. Assuming the allegations in the plaintiffs’ complaint are true, Carlton concludes that three central issues (whether credit and debit cards are distinct products; whether Visa and/or MasterCard have market power in credit cards; and whether the tying policy injures all class members) are appropriate for class-wide determination. (Declaration of Dennis W. Carlton, Ph.D. in Support of Plaintiffs’ Motion for Class Certification (“Carlton Declaration”), ¶ 4, 7.)

First, determining whether credit cards and debit cards are distinct products will turn on the objective characteristics of the products, which influence consumer attitudes, which in turn influence retailer attitudes. (Id. ¶ 10.) It would be “highly duplicative,” according to Carlton, to conduct this inquiry on a merehant-by-merchant basis. (Id.)

Second, determining whether the defendants, either individually or jointly, have market power in the credit card market would be best accomplished on a class-wide basis, according to Carlton. (Id. ¶ 16.) Market power can sometimes be demonstrated (at least in part) through market share; defendants’ market share by definition does not vary across members of the class. (Id. ¶ 20.) Another test of market power is the ability to engage in price discrimination; again, whether or not the defendants have that ability is a question subject to class-wide determination. (Id. ¶ 23.)

Third, Carlton concludes that the question of injury resulting from the complained-of tie is susceptible to class-wide determination. (Id. ¶ 24.) He starts with the plaintiffs’ assertion that, absent the tying arrangement, a large number of retailers would have refused to accept the Visa and MasterCard off-line POS debit cards. In that event, “Visa and MasterCard likely would have found it in their interest to reduce their interchange fees in order to maintain merchant acceptance of these cards.” (Id. ¶ 29.) As a result, all class members have suffered an injury as a result of the tie, namely, the payment of a higher interchange fee for VisaCheck and MasterMoney than they would have paid absent the tie. (Id. ¶ 16.) Moreover, Carlton asserts that the determination of the plaintiffs’ claim that retailers have been injured by defendants’ efforts to limit the volume of on-line debit transactions can be made on a class-wide basis. (Id. ¶ 30.)

Finally, Carlton concludes that damages could be calculated using a common formula based on his theory of injury: but for the tying arrangement, interchange fees for offline debit would be lower. (Id. ¶ 33.) Carlton estimates that in this “but-for” world the interchange fees for off-line debit would be comparable to that for on-line debit; an individual merchant’s damages could therefore be calculated by measuring the overcharge it had paid on all the off-line debit transactions it accepted. (Id. ¶ 34-35.)

2. Riehard L. Schmalensee

The defendants’ expert is Richard L. Schmalensee, the dean of the Sloan School of Management at the Massachusetts Institute of Technology and the Billard professor of economics and management. Like Carlton, Schmalensee has published widely on industrial organization, antitrust topics, and the credit card industry.

Schmalensee’s principal attack addresses Carlton’s theory of injury and damages. He concludes that Carlton’s “but-for” analysis is grossly over-simplified and does not account for a ripple effect of consequences that would have flowed from an end to the tie, resulting in different costs and benefits for different merchants. First, Schmalensee concludes that a dramatic reduction in off-line debit interchange fees would have resulted in many fewer banks issuing many fewer cards. (Declaration of Richard L. Schmalensee, Ph.D. in Support of Defendants’ Opposition to Class Certification and Defendants’ Motion to Strike the Declaration of Dennis W. Carlton, Ph.D. (“Schmalensee Declaration”), ¶ 60.) With the lower interchange fees, Visa also would likely have spent far less than the $100 million it has spent advertising Visa-Cheek since 1992, resulting in even less usage. (Id. ¶ 67.) With fewer cards in circulation, less promotion, and an end to the tie, fewer merchants would accept the off-line POS debit cards, notwithstanding the lower interchange fees. (Id. ¶ 70.)

Schmalensee suggests another consequence of Carlton’s “but-for” world: an increase in credit card interchange fees. “As a matter of elementary economics, changing the terms of a tied sale (e.g., untying the products and unbundling the prices) generally has to affect the prices of both the tied and the tying product if it is to have any effect on price.” (Id. ¶74.) This is because a merchant purchasing the two tied products must have concluded that it is willing to pay the aggregate price for the two products. If it is paying more than it would like for the tied product, it must be paying less than it would be willing to for the tying product, or else it would not complete the transaction. (Id.)

Based on these conclusions, Schmalensee contends that there would be “winners” and “losers” in the untied, “but-for” world. The factors determining whether a particular merchant would win or lose in the but-for world (and the measure of its win or loss) are its mix of credit and debit transactions; the level of incremental sales it generates by accepting off-line debit; its incremental profit margin on those sales; and the payment methods (and their costs) its customers would use instead of off-line debit. (Id. ¶¶ 90-107.) Because these factors will vary widely among merchants, the existence and extent of plaintiffs’ injuries are not amenable to class-wide determination. (Id. ¶ 112-35.) Similarly, the existence of these variables makes the accurate calculation of damages by a class-wide formula impossible. (Id. ¶ 136-42.)

Schmalensee also contends that the other issues Carlton addresses (whether debit and credit cards are distinct products and whether defendants possess market power) cannot be established on a class-wide basis. Cardholders may see debit and credit cards as substitutes for each other at certain classes of retailers, but not at others. (Id. ¶ 147-48.) Moreover, defendants’ market power may be greater or lesser when considered relative to merchants of different sizes or types. (Id. 11149-52.)credit cards are distinct products and whether defendants possess market power) cannot be established on a class-wide basis. Cardholders may see debit and credit cards as substitutes for each other at certain classes of retailers, but not at others. (Id. ¶ 147-48.) Moreover, defendants’ market power may be greater or lesser when considered relative to merchants of different sizes or types. (Id. ¶ 149-52.)

3. Carlton’s Reply

In his reply declaration, Carlton challenges Schmalensee’s contentions, using both empirical and theoretical arguments.

First, he disputes Schmalensee’s argument that, in the “but-for” world of lower debit card interchange fees, banks would have issued fewer of the cards and there would have been less promotion, resulting in an overall decrease in transaction volume. He points to Visa’s experience with Interlink, its on-line debit card.⁵ In April 1997, Visa dramatically lowered the interchange fees for Interlink transactions. Depending on the size of a given transaction, the reduction was as much as 75 percent. After this move, the number of Interlink transactions and merchants accepting the card rose, the opposite of what Schmalensee’s model would have predicted. (Reply Declaration of Dennis W. Carlton in Support of Plaintiffs’ Motion for Class Certification (“Carlton Reply”), ¶ 17.) Visa also cut its interchange fees for the offline debit card in 1992, yet usage has increased dramatically since that time. (Id. ¶ 18.) Carlton also notes that one of the original purposes of debit cards was to help banks reduce the high cost of processing checks; that incentive would remain, even if interchange fees fell. (Id. ¶ 20.) Finally, banks could make up for lower interchange fees by assessing cardholders annual fees or transaction fees. (Id. ¶ 22.)

Second, Carlton disputes Schmalensee’s assertion that, assuming market power in the credit card market, an end to the tie would lead to an increase in credit card interchange fees, which would in turn create “winners” and “losers” among merchants. Carlton first cites two empirical “benchmarks” in disputing Schmalensee’s conclusion. He points to Canada, where banks do not issue off-line debit cards and their use by consumers is rare. All else being equal, credit card interchange fees should be higher in Canada, according to Schmalensee’s model, because there effectively is no tie. Yet, credit card interchange fees in Canada are lower than in the United States. (Id. ¶ 27-30.) As a second “benchmark,” Carlton offers historical evidence from the United States. Between 1991 and 1998, off-line debit transactions increased many times over. All else being equal, credit card interchange fees should have declined during this period, according to Schmalensee’s model. Specifically, as the volume of off-line debit card transactions increased, merchants would- have to be compensated for the high interchange fees with a diminution in the interchange rates for the tying product. Yet the credit card interchange fees generally stayed the same or increased (when adjusted for inflation) during that period (Id. ¶ 31.)

Carlton also disputes Schmalensee’s assertions regarding credit card interchange fees as a matter of economic theory. Carlton contends that Schmalensee’s model was limited to situations in which the tying and tied products are sold in fixed proportions and are independently demanded. (Id. ¶ 34.) Models applicable in other settings, Carlton asserts, do not predict an increase in the price of the tying product when the tie is broken. In fact, one model, in which the seller possesses market power in both the tying and tied product, actually predicts a decline in price for both products when the tie is broken. (Id. ¶ 35.)

Finally, Carlton disputes Schmalensee’s conclusions that market power and definition are not susceptible to class-wide proof. (Id. ¶ 42-48.)

II. DISCUSSION

A. The Motion to Strike

The defendants have moved to strike Carlton’s expert opinion based on Federal Rule of Evidence 702, as construed by the Supreme Court in Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993), and Kumho Tire Co., Ltd. v. Carmichael, 526 U.S. 137, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999). Specifically, they contend that Carlton’s opinion is inadmissible because it is “legally irrelevant and not rationally based on the pertinent facts or on any analysis.” (Memorandum of Law in Support of Defendants’ Motion to Strike Opinions Offered by Dennis W. Carlton, at 1.)

Rule 702 provides:

If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise.

Fed.R.Evid. 702 (emphasis added). Daubert, on which the‘defendants rely, involved “the standard for admitting expert scientific testimony in a federal trial.” 509 U.S. at 582, 113 S.Ct. 2786 (emphasis added).

As the highlighted phrases indicate, the Daubert inquiry was designed to shield the fact-finder at trial from flawed evidence. We are obviously not at trial; the parties have not even filed dispositive motions yet. In this preliminary stage of the case, I am very far from the “trier of fact” contemplated in Rule 702. Indeed, I am expressly forbidden from engaging in “a preliminary inquiry into the merits” of the case. Eisen v. Carlisle & Jacquelin, 417 U.S. 156, 177, 94 S.Ct. 2140, 40 L.Ed.2d 732 (1974). In particular, a court at the class certification stage should not delve into the merits of an expert’s opinion, or indulge “dueling” between opposing experts. Caridad v. Metro-North Commuter R.R., 191 F.3d 283, 292 (2d Cir.1999) (quoting Krueger v. New York Tel. Co., 163 F.R.D. 433, 440—41 (S.D.N.Y.1995)), petition for cert. filed, (U.S. Dec. 27, 1999) (No. 99-1084).

Notwithstanding these distinctions, the defendants contend, and I agree, that there is a role for a Daubert inquiry at the class certification stage. It cannot be that a court could certify a class (in this case of approximately four million members) on the basis of an expert opinion so flawed that it is inadmissible as a matter of law. Although there is a role for a Daubert test here, it is a limited one, tailored to the purpose for which the expert opinion is offered. The question is not, therefore, whether a jury at trial should be permitted to rely on Carlton’s report to find facts as to liability, but rather whether I may utilize it in deciding whether the requisites of Rule 23 have been met.⁶

Under the Daubert standard, as it is typically applied at trial, “judges are charged with ensuring that expert testimony ‘both rests on a reliable foundation and is relevant to the task at hand.’ ” Zuchowicz v. United States, 140 F.3d 381, 386 (2d Cir. 1998) (quoting Daubert, 509 U.S. at 597, 113 5. Ct. 2786). In fulfilling this “gatekeeper” role, a judge is charged with making a “preliminary assessment” of the testimony’s “reasoning or methodology” and whether that reasoning or methodology “properly can be applied to the facts in issue.” Id. (quoting Daubert, 509 U.S. at 592-93, 113 S.Ct. 2786).

The Supreme Court has provided a non-exhaustive list of factors to consider in what should be a “flexible” inquiry by the district court:

(1) whether the theory can be (and has been) tested according to the scientific method; (2) whether the theory or technique has been subjected to peer review and publication; (3) in the case of a particular scientific technique, the known or potential error rate; and (4) whether the theory is generally accepted.

Id. (citing Daubert, 509 U.S. at 593-94, 113 S.Ct. 2786).

The Daubert gate-keeping function applies even in the case of a non-scientific expert. See Kumho Tire, 119 S.Ct. at 1175. In such eases, the Daubert factors may or may not be applicable; “the gatekeeping inquiry must be tied to the facts of a particular case.” Id. (internal quotation marks omitted).

The defendants attack Carlton’s opinion on several grounds: (1) it is based on supposition rather than empirical proof; (2) he fails to analyze what would happen to the price of the “tying” product (credit cards) if the tie was broken, instead focusing exclusively on his contention that the price of the “tied” product (debit cards) would decline; (3) he fails to account for the complexities of the interactions between different players in the plastic card market and how they would be affected by an end to the “honor all cards” rule; (4) he fails to account for the possibility that, if the tie is broken, the defendants might charge different debit interchange fees to different merchants; and (5) he fails to account for the possibility that if interchange fees decline for debit cards, the volume of transactions would decline.

These contentions might or might not have force at a later stage if the defendants challenge the admissibility of Carlton’s opinion. But the admissibility inquiry under Daubert and Kumho Tire must be adapted to the facts of the case at hand and to the stage of the proceedings. I find that Carlton’s opinion is admissible for the narrow purpose for which it is offered: to support the plaintiffs’ class certification motion. Carlton’s qualifications (which defendants do not challenge) are impeccable.⁷ The fact that his report is based principally on economic theory, not exhaustive empirical or quantitative analysis, is no bar to its admission at this stage; if it were, Schmalensee’s opinion would be inadmissible as well. Moreover, the defendants have not shown that Carlton has failed to “rely upon the type of methodology and data typically used and accepted” in cases such as this one. Iacobelli Constr., Inc. v. County of Monroe, 32 F.3d 19, 25 (2d Cir.1994). The gravamen of defendants’ motion is that they (and their expert) disagree with Carlton’s conclusions; that is not a basis for exclusion. In sum, defendants have fallen far short of showing that Carlton’s opinion is the kind of “junk science” that a Daubert inquiry at this preliminary stage ought to screen. See id. (quoting Tamarin v. Adam Caterers, Inc., 13 F.3d 51, 53 (2d Cir.1993)). The defendants’ motion to strike is denied.

B. The Class Certification Motion

1. Class Actions

a. General Standards

Class actions are “an exception to the usual rule that litigation is conducted by and on behalf of the individual named parties only.” General Tel. Co. v. Falcon, 457 U.S. 147, 155, 102 S.Ct. 2364, 72 L.Ed.2d 740 (1982) (quoting Califano v. Yamasaki, 442 U.S. 682, 700-01, 99 S.Ct. 2545, 61 L.Ed.2d 176 (1979)). The class device is designed for cases in which the “issues involved are common to the class as a whole and when they turn on questions of law applicable in the same manner to each member of the class.” Id. (quoting Califano, 442 U.S. at 701, 99 S.Ct. 2545) (internal quotation marks omitted). In those cases, a class action can conserve the resources of both the courts and the parties. See id.

Pursuant to Federal Rule of Civil Procedure 23(a), a class action may be initiated if four requirements are met:

(1) the class is so numerous that joinder of all members is impracticable, (2) there are questions of law or fact common to the class, (3) the claims or defenses of the representative parties are typical of the claims or defenses of the class, and (4) the representative parties will fairly and adequately protect the interests of the class.

Fed.R.Civ.P. 23(a).

Additionally, a class action may only be “maintained” if the four requirements of Rule 23(a) are met and one of the tests of Rule 23(b) is satisfied.⁸ Rule 23(b)(2) allows maintenance of a class action if “the party opposing the class has acted or refused to act on grounds generally applicable to the class, thereby making appropriate final injunctive relief or corresponding declaratory relief with respect to the class as a whole[.]” Fed. R.Civ.P. 23(b)(2). In the alternative, Rule 23(b)(3) allows maintenance of a class action if

the court finds that the questions of law or fact common to the members of the class predominate over any question affecting only individual members, and that a class action is superior to other available methods for the fair and efficient adjudication of the controversy. The matters pertinent to the findings include: (A) the interest of the members of the class in individually controlling the prosecution or defense of separate actions; (B) the extent and nature of any litigation concerning the controversy already commenced by or against members of the class; (C) the desirability or undesireability of concentrating the litigation of the claims in the particular forum; (D) the difficulties likely to be encountered in the management of a class action.

Fed.R.Civ.P. 23(b)(3).

The party seeking to certify a class bears the burden of establishing the prerequisites of Rule 23. See Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 614, 117 S.Ct. 2231, 138 L.Ed.2d 689 (1997); Caridad v. Metro-North Commuter RR, 191 F.3d 283, 291 (2d Cir.1999), petition for cert. filed, (U.S. Dec. 27, 1999) (No. 99-1084). A court may certify a class action only if it is “satisfied, after a rigorous analysis,” that the rule’s prerequisites have been established. Falcon, 457 U.S. at 161, 102 S.Ct. 2364. This analysis will inevitably be “enmeshed in the factual and legal issues comprising the plaintiffs cause of action.” Id. at 160, 102 S.Ct. 2364 (quoting Coopers & Lybrand v. Livesay, 437 U.S. 463, 469, 98 S.Ct. 2454, 57 L.Ed.2d 351 (1978)). In some cases, it may be possible to conduct the required inquiry by looking no further than the pleadings, but in other instances, the court may be required to “probe behind the pleadings before coming to rest on the certification question.” Id. In some cases in the latter group, it will be necessary to conduct discovery to establish a sufficient evidentiary record from which to make the class determination. See Sirota v. Solitron Devices, Inc., 673 F.2d 566, 571 (2d Cir.1982).

Notwithstanding the district court’s ability to look beyond the pleadings, its resolution of a class certification motion may not become “a preliminary inquiry into the merits” of the case. Eisen v. Carlisle & Jacquelin, 417 U.S. 156, 177, 94 S.Ct. 2140, 40 L.Ed.2d 732 (1974). “In determining the propriety of a class action, the question is not whether the plaintiff or plaintiffs have stated a cause of action or will prevail on the merits, but rather whether the requirements of Rule 23 are met.” Id. at 178, 94 S.Ct. 2140 (quoting Miller v. Mackey Int’l, 452 F.2d 424, 427 (5th Cir.1971)). In sum, a court considering a class certification motion “must look somewhere between the pleading and the fruit of discovery____ [E]nough must be laid bare to let the judge survey the factual scene on a kind of sketchy relief map, leaving for later view the myriad of details that cover the terrain.” Sirota, 673 F.2d at 571-72 (quoting Professional Adjusting Sys. of America, Inc. v. General Adjustment Bureau, Inc., 64 F.R.D. 35, 38 (S.D.N.Y.1974)).

b. Class Actions and Expert Opinions

In this circuit, a district court is not permitted to indulge “dueling” between opposing experts at the class certification stage. Caridad, 191 F.3d at 292 (quoting Krueger v. New York Tel. Co., 163 F.R.D. 433, 440-41 (S.D.N.Y.1995)). Because Caridad is the Second Circuit’s latest word on this question, extended discussion of that opinion, and the district court opinion that led to it, is warranted.

The plaintiffs in the ease were 25 present or former employees of Metro-North, each alleging a company-wide policy of discrimination against African-Americans. See Robinson v. Metro-North Commuter R.R. Co., 175 F.R.D. 46, 47 (S.D.N.Y.1997), rev’d sub. nom., Caridad, 191 F.3d 283. They sought to certify a class of all African-American employees at Metro-North from 1983 to 1996 who held union or lower- to middle-level management positions. See id.

In support of their motion, the plaintiffs submitted a statistical analysis showing that African-American employees faced discipline from Metro-North with more frequency than other employees and were promoted less often. See id. at 48. The district court rejected this analysis as a basis for establishing “commonality” under Rule 23(a) because, as shown by defendant’s expert, it failed to take into consideration the widely varying rates of discipline and promotion by position at Metro-North.⁹ See id. at 48. When those differences were added to the analysis, “no statistically significant racial disparities, either for discipline or promotion, can be established with respect to the great majority of the positions at Metro-North.” Id. Moreover, the court saw a more “fundamental” problem with class certification. The gravamen of the complaint was that Metro-North’s policy of “overdelegation” of promotion and discipline decisions allowed individual managers to discriminate. Without the statistical proof that the court had rejected, there was no evidence that this practice led to “generalized discrimination” at the company. Thus class treatment was inappropriate. Id. at 49.

In an opinion written by Judge Newman, the Second Circuit reversed. See Candad, 191 F.3d at 283. The court noted that it would likely be “extremely difficult” for the plaintiffs to prove that Metro-North’s grant of discretionary authority “affects one class of employees more harshly than others.” Id. at 291. Nonetheless, the court concluded that “the statistical report and anecdotal evidence submitted by the Class Plaintiffs are sufficient to demonstrate common questions of fact regarding the discriminatory implementation and effects of Metro-North’s company-wide policies regarding promotion and discipline.” Id. at 292. The court noted the critique of plaintiffs’ statistics credited by the district court, but cautioned that “this sort of ‘statistical dueling’ is not relevant to the certification determination.” Id. (quoting Krueger, 163 F.R.D. at 440-41). The critique “may prove fatal at the merits stage, but the Class Plaintiffs need not demonstrate at this stage that they will prevail on the merits.” Id.; see also id. at 293 (“More detailed statistics might be required to sustain the Plaintiffs burden of persuasion, but this report, in conjunction with the anecdotal evidence, satisfies the Class Plaintiffs’ burden of demonstrating commonality for purposes of class certification.” (internal citation omitted)).

The court ended its discussion with the following instructions to district courts:

In deciding a certification motion, district courts must not consider or resolve the merits of the claims of the purported class. Here, the District Court credited Metro-North’s expert evidence over that of the Class Plaintiffs. Such a weighing of the evidence is not appropriate at this stage in the litigation.

Id. (internal citation omitted).¹⁰

2. The Antitrust Claims

I will now briefly summarize the elements of the plaintiffs’ two causes of action.¹¹

a. Illegal Tying, Sherman Act § 1

Section One of the Sherman Act prohibits “[ejvery contract, combination in the form of trust or otherwise, or conspiracy, in restraint of trade or commerce among the several States, or with foreign nations.” 15 U.S.C. § 1 (1994). Against the backdrop of this general language, the courts have developed different doctrinal tests for different types of claims brought pursuant to § 1, including those challenging tying arrangements.

Tying is “an agreement by a party to sell one product but only on the condition that the buyer also purchase a different (or tied) product.” Yentsch v. Texaco, Inc., 630 F.2d 46, 56 (2d Cir.1980) (quoting Northern Pac. Ry. Co. v. United States, 356 U.S. 1, 5, 78 S.Ct. 514, 2 L.Ed.2d 545 (1958)).

A plaintiff must establish four elements to establish a defendant’s liability for tying under a “per se” theory:

(1) the alleged arrangement must affect a substantial amount of interstate commerce; (2) the two products ... must be distinct; (3) the defendants must have actually tied the sale of the two products; and (4) the seller must have appreciable economic power in the tying market.

United States v. International Bus. Machs. Corp., 163 F.3d 737, 741 (2d Cir.1998) (internal citations and quotation marks omitted); see also Northern Pac., 356 U.S. at 5, 78 S.Ct. 514 (explaining justification for per se illegality: “certain agreements or practices which because of their pernicious effect on competition and lack of any redeeming virtue are conclusively presumed to be unreasonable and therefore illegal without elaborate inquiry as to the precise harm they have caused or the business excuse for their use”).

If the plaintiff fails to establish the elements of a per se claim (for example, by failing to demonstrate market power), it still may prevail under the “rule of reason” if it can demonstrate that the challenged arrangements are unreasonable “in light of their actual effects on the market and their pro-competitive justifications.” Clorox Co. v. Sterling Winthrop, Inc., 117 F.3d 50, 56 (2d Cir.1997). Such an inquiry proceeds in three steps:

First, the [p]laintiff bears the initial burden of showing that the challenged action has had an actual adverse effect on competition as a whole in the relevant market ____ Then, [i]f the plaintiff succeeds, the burden shifts to the defendant to establish the pro-competitive redeeming virtues of the action. Should the defendant carry this burden, the plaintiff must then show that the same pro-competitive effect could be achieved through an alternative means that is less restrictive of competition.

Id. (internal citations and quotation marks omitted; alterations in original).

b. Attempt and Conspiracy to Monopolize, Sherman Act § 2

Section Two of the Sherman Act provides in relevant part that

[e]very person who shall monopolize, or attempt to monopolize, or combine or conspire with any other person or persons, to monopolize any part of the trade or commerce among the several States, or with foreign nations, shall be deemed guilty of a felony.

15 U.S.C. § 2 (1994). To win on an attempt to monopolize claim, a plaintiff must show “(1) that the defendant has engaged in predatory or anticompetitive conduct with (2) a specific intent to monopolize and (3) a dangerous probability of achieving monopoly power.” Spectrum Sports, Inc. v. McQuillan, 506 U.S. 447, 456, 113 S.Ct. 884, 122 L.Ed.2d 247 (1993). To win on a conspiracy to monopolize claim, a plaintiff must show “(1) proof of a concerted action deliberately entered into with the specific intent to achieve an unlawful monopoly, and (2) the commission of an overt act in furtherance of the conspiracy.” International Distribution Ctrs., Inc. v. Walsh Trucking Co., Inc., 812 F.2d 786, 795 (2d Cir.1987).

3. The Injury in Fact Requirement: Amenability to Class Determination

The defendants base their objection to class certification on Rule 23(a)(3)’s requirement that “the claims ... of the representative parties [be] typical of the claims ... of the class” and Rule 23(a)(4)’s requirement that “the representative parties will fairly and adequately protect the interests of the class.”¹² Additionally, they contend that Rule 23(b)(2) is inapplicable in a case such as this one, where damages are a prominent feature, and that the requirements of Rule 23(b)(3) are not met because individual questions will predominate over common ones.

The defendants’ principal attack on the class certification motion rests on the rule that even an antitrust plaintiff who has successfully proven illegal conduct on the part of the defendant must also show that it was “injured in fact” by the illegality. According to the defendants, some members of the proposed class will not be able to show injury, and the injuries of the others will vary in ways not susceptible to resolution by a class-wide formula. Therefore, no plaintiffs claims will be “typical” of any others; there will be conflicts of interest within the proposed class; and individual issues of injury will predominate.

Section Four of the Clayton Act provides a cause of action, with treble damages, for “any person who shall be injured in his business or property by reason of anything forbidden in the antitrust laws.” 15 U.S.C. § 15 (1994). Pursuant to this requirement, “a plaintiff must make some showing of actual injury attributable to something the antitrust laws were designed to prevent.” J. Truett Payne Co. v. Chrysler Motors Corp., 451 U.S. 557, 562, 101 S.Ct. 1923, 68 L.Ed.2d 442 (1981). This is true even if the elements of per se liability are established. See Atlantic Richfield Co. v. USA Petroleum Co., 495 U.S. 328, 341, 110 S.Ct. 1884, 109 L.Ed.2d 333 (1990); see generally II Phillip E. Areeda & Herbert Hovenkamp, Antitrust Law 11362e, at 216 (rev. ed.1995) (“No matter how clear the violation, the private plaintiff must still prove that he was actually injured in order to recover damages.”)

Injury in fact is required even when an antitrust case is litigated as a class action. See Alabama v. Blue Bird Body Co., Inc., 573 F.2d 309, 318 (5th Cir.1978) (citing the Rules Enabling Act’s dictate that a procedural rule cannot “abridge, enlarge, or modify any substantive right,” 28 U.S.C. § 2072; “Consequently, this court has no power to define differently the substantive right of individual plaintiffs as compared to class plaintiffs.”) “Whether or not fact of damage can be proven on a common basis ... depends upon the circumstances of each case.” Bogosian v. Gulf Oil Corp., 561 F.2d 434, 454 (3d Cir.1977). The fact of injury, which is required as an element of the plaintiffs claim, should not be confused with the extent of injury (as reflected by the amount of damages), which may not be amenable to establishment with great precision. See Kypta v. McDonald’s Corp., 671 F.2d 1282, 1285 (11th Cir.1982).

The defendants’ first challenge to injury in fact responds to the scenario described in the plaintiffs’ complaint, ie., absent the tie, virtually all members of the putative class would have refused the off-line debit cards and processed the same transactions using cash, checks, on-line debit, or other payment forms. Under this theory, defendants contend, there could be no class-wide injury in fact because (1) even some named plaintiffs have testified that they would have continued to accept off-line debit cards at current prices even absent the tie; (2) some merchants have PIN pads, while others do not (for widely varying reasons), resulting in variation in their ability to process transactions on-line; (3) transactions formerly processed through off-line debit would be replaced by other payment forms, whose mix and costs would vary merchant-by-merchant; and (4) the expense of those payment forms, when considered together with the incremental sales produced by accepting off-line debit and the merchant’s profit margin on those incremental sales, will determine whether a given merchant suffered any injury from the tie.

Although these arguments fully respond to the allegations in the complaint, they pass in the night with the plaintiffs’ current theory of the case, as espoused by Carlton. That theory takes the complaint’s scenario one step further. It posits that in the “but-for,” untied world, after large numbers of merchants refused off-line debit cards, Visa and MasterCard would have cut their interchange fees in order to preserve universal acceptance. That move, in turn, would have led merchants to accept the cards once again. Carlton’s scenario is a complete answer to the defendants’ attack on the theory of the complaint; it posits class-wide injury resulting from every single class member’s overpaying for off-line debit cards as a direct result of the tie. See Eisen v. Carlisle & Jacquelin, 391 F.2d 555, 562 (2d Cir.1968) (“[Defendants have argued that ... other class members may be satisfied with the present price policy. Nonetheless, all members of the class, including those who would otherwise prefer to abide by the status quo, will be helped if the rates are found to be excessive.” (internal footnote omitted)), vacated on other grounds, 417 U.S. 156, 94 S.Ct. 2140, 40 L.Ed.2d 732 (1974). Carlton’s theory does not require analysis of any of the individual questions listed above.

Defendants of course do not stop there; they contend that Carlton’s theory does not support a finding of class-wide injury. First, they contend that his theory of a price decline in the tied product is too speculative. However, they themselves admit that “Dr. Carlton’s ‘price’ theory of injury is not unknown, and defendants do not here contend that no plaintiff could ever recover under a ‘price’ theory.” (Defendants’ Memorandum of Law in Opposition to Plaintiffs’ Motion for Class Certification (“Defendants’ Memorandum”), at 29.); see also X Phillip E. Areeda et al., Antitrust Law ¶1769c, at 432 (1996) (“While uncommon ..., such injury-in-fact for buyers is not impossible.”). The difficulty of proof on a question such as this will not defeat class certification. See Caridad v. Metro-North Commuter RR, 191 F.3d 283, 291 (2d Cir.1999) (requiring class certification based on a theory “likely to be extremely difficult” to prove), petition for cert, filed, (U.S. Dec. 27, 1999) (No. 99-1084).

Second, the defendants contend that “[a]s a matter of elementary economic theory,” a defendant that has market power in the tying product and overcharges on the tied product will do two things if the tie is broken: decrease the price of the tied product and increase the price of the tying product. (Defendants’ Memorandum at 31-32.) If one assumes that credit card interchange fees would have been higher in the “but-for,” untied world and that this fact is relevant for purposes of assessing injury in a tying case, then the fact of injury would no longer be susceptible to class-wide proof. Specifically, merchants with relatively high amounts of credit card transactions vis a vis debit card transactions (e.g., jewelry stores) would have suffered little or no injury from the tie; indeed, they would have benefitted from it to the extent it kept credit card interchange fees down.

The defendants find support for their argument in a number of cases concluding that a tying plaintiff is injured only if it overpaid for the package of the tied and tying product. See Will v. Comprehensive Accounting Corp., 776 F.2d 665, 673 (7th Cir.1985) (“[U]nless the plaintiff shows that the package price was elevated the suit must be dismissed without further ado.”); Kypta, 671 F.2d at 1285 (“[I]njury resulting from a tie-in must be shown by establishing that payments for both the tied and tying products exceeded their combined fair market value. The rationale behind this requirement is apparent: A determination of the value of the tied products alone would not indicate whether the plaintiff indeed suffered any net economic harm, since a lower price might conceivably have been exacted by the [defendant] for the tying product.”); Siegel v. Chicken Delight, Inc., 448 F.2d 43, 52 (9th Cir.1971) (vacating a damages award based only on overcharge for tied product; “To ascertain whether an unlawful arrangement for the sale of products has caused injury to the purchaser, the cost or value of the products involved, free from the unlawful arrangement, must first be ascertained.”); L. Knife & Son, Inc. v. Banfi Prods. Corp., 118 F.R.D. 269, 271 (D.Mass. 1987).

The plaintiffs respond to this argument in several ways. First, they cite to a competing line of cases that support the proposition that purchasers challenging a tie are injured by paying an inflated price for the tied product (without any reference to the price of the tying product).¹³ See Kaiser Aluminum & Chem. Sales, Inc. v. Avondale Shipyards, Inc., 677 F.2d 1045, 1054 (5th Cir.1982) (“In a tying arrangement, the ordinary measure of damages would be the difference between the price actually paid for the tied product and the price at which the product could have been obtained on the open market.”); Bell v. Cherokee Aviation Corp., 660 F.2d 1123, 1133 (6th Cir.1981) (same); Northern v. McGraw-Edison Co., 542 F.2d 1336, 1347 (8th Cir. 1976) (same); Esposito v. Mister Softee, Inc., No. 75-CV-555, 1983 WL 1806, at *2 (E.D.N.Y. Apr. 11, 1983) (same). But see X Areeda et al., supra, It ¶769c, at 430 (calling these cases “quite wrong”).

Second, plaintiffs’ counsel contended at oral argument that the cases cited by defendants all involved “fixed proportion” ties (e.g., hot dogs and hot dog buns) and are therefore inapplicable to a case such as this one involving a “variable proportion” tie in which there is no necessary correlation in consumption of the two products. Counsel’s attempt at distinguishing this entire line of cases fails, however, since several did not involve fixed proportion ties. See Will, 776 F.2d at 669 (tying product: franchise; tied product: data processing services); Midwestern Waffles, Inc. v. Waffle House, Inc., 734 F.2d 705, 712 (11th Cir.1984) (tying product: franchise; tied product: equipment and vending services); Siegel, 448 F.2d at 46 (tying product: franchise; tied product: cooking equipment, food items, and packaging); see also X Areeda, supra, ¶ 1769c, at 431 (“Even when the products are not used exactly in fixed proportions, tied-product premiums are often offset by tying-product discounts below the price that would have prevailed without tying.”).

Finally, plaintiffs contend that under the particular circumstances of the market at issue in this case, credit card interchange fees would not have increased in the “but-for,” untied world. They rely principally on Carlton’s reply declaration. As summarized above (see supra § 1.F(3)) Carlton relied on both empirical evidence and economic theory to refute the defendants’ claims. He pointed out that in Canada, where banks do not issue off-line debit cards (and thus there is effectively no tie) credit card interchange fees are lower than in the United States. (Carlton Reply ¶¶ 27-30.) In addition, Carlton pointed out that the dramatic increase in off-line debit transactions between 1991 and 1998 did not result in a corresponding decline in credit card interchange fees, as defendants’ theory (all else being equal) would have predicted. Indeed, those fees generally stayed the same or increased (when adjusted for inflation). (Id. ¶ 31.) As for economic theory, Carlton contended, as plaintiffs’ counsel did at oral argument, that defendants’ proffered model was limited to situations in which the tying and tied products are sold in fixed proportions, and that models applicable in other settings predict either no increase in the price of the tying product when the tie is broken, or, in one model, a decline in price for both products when the tie is broken. (Id. ¶ 35.)

The plaintiffs also seek to undermine Schmalensee’s argument with his own testimony. Specifically, Schmalensee stated at his deposition that “while I think it is likely” that credit card interchange fees would have risen, “I haven’t pushed it far enough to have an opinion.” (Deposition of Richard Lee Schmalensee at 128, attached as Exhibit 1 to Reply Declaration of Lloyd Constantine.)

Finally, at oral argument, plaintiffs’ counsel suggested that, as part of plaintiffs’ proof on its § 2 attempt to monopolize claim, they would demonstrate that the defendants attempted to monopolize the debit card market partly for the purpose of keeping credit card interchange fees high.¹⁴ If that attempt were thwarted by breaking the tie, he continued, credit card interchange fees would actually decline.

I conclude that the plaintiffs have met their burden under Caridad of showing that injury in fact is susceptible to common proof and that class treatment is therefore appropriate. They have offered both empirical and theoretical support for their claim. Specifically, for- present purposes, taking into account that a “weighing of the evidence is not appropriate at this stage in the litigation,” Caridad, 191 F.3d at 293, plaintiffs have established to my satisfaction that, in the “but for,” untied world, defendants would have been compelled to lower their interchange fees for off-line debit cards and would not have raised the credit card interchange fees.

Plaintiffs have therefore made a sufficient showing that the package price, encompassing both credit and debit interchange fees, would decline absent the tie.¹⁵ Professor Areeda has noted that, although proof of such a theory is “elusive and seldom attempted,” antitrust plaintiffs “are clearly entitled to damages when the sum of tying and tied product prices would have been lower ... absent the tie.” X Areeda et al., supra, ¶1769b, at 429. There is nothing inconsistent between this assertion and class certification in this case; class certification is required even when the proffered common theory of injury is “likely to be extremely difficult to prove.” Caridad, 191 F.3d at 291; see also Krueger v. New York Tel. Co., 163 F.R.D. 433, 442 (S.D.N.Y.1995) (“The plaintiffs have laid a sufficient evidentiary foundation which, if believed, supports their allegations of common questions of law and fact.”); 7B Charles Alan Wright, Arthur R. Miller & Mary Kay Kane, Federal Practice and Procedure § 1781, at 7-8 (2d ed.1986) (“[I]f plaintiffs’ assertion of class status is at least colorable, the court should allow the action to proceed under Rule 23 and move the case forward to an adjudication on the merits.”).

4. The Defendants’ Remaining Injury/Causation Contentions

The defendants’ remaining contentions can be disposed of more expeditiously. First, they contend that in Carlton’s “but-for,” untied world (which assumes defendants’ market power), defendants would not lower offline debit interchange fees across the board, but would instead selectively reduce fees as necessary to keep individual merchants or groups of merchants on board. Yet neither defendant has ever engaged in merchant-specific pricing for any of its products, and it would be speculative to suppose that they would in the “but-for,” untied world. The defendants do have broad pricing categories for different merchant classes; Carlton’s damages methodology could easily be adjusted to account for that fact if necessary.

Next, defendants contend that Carlton’s theory fails to account for what they predict would be a decline in the volume of off-line debit transactions in the “but-for,” untied world with lower off-line debit interchange fees. That decline, considered in conjunction with defendants’ assertion that some merchants garner incremental sales from accepting the cards, would create “winners” and “losers” and render class-wide assessment of injury inappropriate. This argument is immaterial when an antitrust plaintiff proceeds on an “overcharge theory” of damages.¹⁶ See New York v. Hendrickson Bros., 840 F.2d 1065, 1079 (2d Cir.1988) (“[Antitrust treble-damage actions should not be complicated by a need to trace the effects of the overcharge with respect to such matters as prices, costs, and the potentially different behavior of all the pertinent variables in the absence of the overcharges.”). The defendants dispute the applicability of horizontal price-fixing cases such as Hendrickson Brothers to a tying case such as this one, but I have already concluded that the plaintiffs have made a sufficient showing to go forward with an overcharge theory on a class-wide basis. Therefore, Hendrickson Brothers is applicable for present purposes.

Defendants contend that causation may not be proven on a class-wide basis. See Argus, Inc. v. Eastman Kodak Co., 801 F.2d 38, 41 (2d Cir.1986) (“[L]ack of causation in fact is fatal to the merits of any antitrust claim. Consequently, an essential element in plaintiffs’ claim is that the injuries alleged would not have occurred but for [defendant’s] antitrust violation.” (internal citation omitted)). Specifically, defendants contend that any merchant that does not like off-line debit cards is capable of “steering” its customers to other forms of payment. Any merchant that chooses not to steer (and the record indicates there are many) cannot claim that any injury it suffers from off-line debit interchange fees is caused by the tie; instead, it is caused by the decision not to steer.

I -will assume for present purposes that defendants have correctly applied the causation requirement to the facts of this case; their objection to class certification on this basis still fails. Defendants contend that their policies allow merchants to attempt to “steer” off-line debit card holders to other forms of payment, or to prompt them to enter their PIN and therefore convert the sale into an on-line transaction. They also acknowledge, however, that if a customer refuses to be “steered” or “prompted,” the merchant must allow him to complete the transaction with his off-line debit card. That is the essence of defendants’ “honor all cards” rule. A merchant that steers will not succeed in every single transaction; there will always be an irreducible minimum of customers who will use their off-line debit cards.¹⁷ Therefore, even hypothetical Merchant A with the most aggressive steering program permissible under its contract would suffer some injury caused by the fact that it is forced to accept the unwanted offline debit card.¹⁸ Likewise, Merchant B, which engages in no steering, would suffer injury caused by the tie even under the defendants’ strict theory of injury and causation; if it implemented a steering program, it would end up in the same situation as Merchant A.

The presence of injury and causation is binary; it is either there or it is not. According to Carlton’s theory, injury and causation is present for every putative class member. The extent of injury is another question. See Zenith Radio Corp. v. Hazel-tine Research, Inc., 395 U.S. 100, 114 n. 9, 89 S.Ct. 1562, 23 L.Ed.2d 129 (1969) (“[Plaintiffs] burden of proving the fact of damage under § 4 of the Clayton Act is satisfied by its proof of some damage flowing from the unlawful conspiracy; inquiry beyond this minimum point goes only to the amount and not the fact of damage.”). It may be that defendants’ argument about steering will have force if repackaged as one addressing mitigation of damages. However, the presence of individualized defenses, such as mitigation, going only to damages are generally regarded as no barrier to class certification. See In re Sumitomo Copper Litig., 182 F.R.D. 85, 95 (S.D.N.Y.1998) (“‘It is well established that individual questions with respect to damages will not defeat class certification or render a proposed representative inadequate unless that issue creates a conflict which goes to the heart of the lawsuit.’ ”)(quoting In re AM Int’l, Inc. Securities Litig., 108 F.R.D. 190, 196 (S.D.N.Y. 1985)); see also Eisen v. Carlisle & Jacquelin, 391 F.2d 555, 566 (2d Cir.1968) (“All of these differences among the class members bear only on the computation of damages, a factor which, by itself, does not justify dismissal of the class action.”), vacated on other grounds, 417 U.S. 156, 94 S.Ct. 2140, 40 L.Ed.2d 732 (1974); Krueger, 163 F.R.D. at 441 (“To the extent that there may be specific defenses against specific members of the class, they would not justify denying class action certification.”); Uniondale Beer Co., Inc. v. Anheuser-Busch, Inc., 117 F.R.D. 340, 344 (E.D.N.Y.1987); 7B Charles Alan Wright, Arthur R. Miller & Mary Kay Kane, supra, § 1781, at 8-9 (“[I]t uniformly has been held that differences among the members as to the amount of damages incurred does not mean that a class action would be inappropriate.”).¹⁹

5. Class Treatment of the Substantive Elements of the Claims

In addition to their substantial arguments regarding injury, the defendants tack on several half-hearted contentions that some of the substantive elements of plaintiffs’ claims are not susceptible to class-wide treatment. Those contentions are meritless.

a. Whether Credit and Debit Cards Are Separate Products

In a citation-free portion of their brief, defendants argue that putative class members may differ on the question of whether, from their individual, subjective perspectives, credit and debit cards are separate products, and that this question is therefore unsuitable for resolution on a class-wide basis.

“The courts uniformly consider the issue whether the tying arrangement consists of separate tying and tied products to be common to all members of the class.” Herbert Hovenkamp, Tying Arrangements and Class Actions, 36 Vand. L.Rev. 213, 220 (1983); accord John H. Matheson, Class Action Tying Cases: A Framework for Certification Decisions, 76 Nw. U.L.Rev. 855, 861 (1982) (“The court’s determination whether the defendant offered two products will affect all prospective members of a class in a common manner.”). The Supreme Court has held that “the answer to the question of whether one or two products are involved turns not on the functional relation between them, but rather on the character of the demand for the two items.” Jefferson Parish Hosp. Dist. No. 2 v. Hyde, 466 U.S. 2, 19, 104 S.Ct. 1551, 80 L.Ed.2d 2 (1984). The operative question is whether “two separate product markets have been linked.” Id. at 21, 104 S.Ct. 1551. This is a question easily decided on an aggregate basis.

b. Whether the Defendants Possess Market Power in the Tying Product

The defendants also contend that the existence of their market power will depend on the size of the putative class member, thus making class-wide resolution of this element unwarranted. I reject this contention.

Market power is “the ability of a single seller to raise price and restrict output.” Eastman Kodak Co. v. Image Technical Servs., Inc., 504 U.S. 451, 464, 112 S.Ct. 2072, 119 L.Ed.2d 265 (1992) (quoting Fortner Enters., Inc. v. United States Steel Corp., 394 U.S. 495, 503, 89 S.Ct. 1252, 22 L.Ed.2d 495 (1969)). “The existence of such power ordinarily is inferred from the seller’s possession of a predominant share of the market.” Id. Because this inquiry focuses on the position of the seller, not that of individual buyers, -it “is readily susceptible to proof in a class action.” Hill v. A-T-O, Inc., 80 F.R.D. 68, 69 (E.D.N.Y.1978). One commentator has cautioned that class treatment of market power is appropriate only when plaintiffs rely on “objective evidence of either the defendant’s market dominance or the product’s uniqueness,” as opposed to “the attitude of the purchaser or ... the conduct of particular business relationships.” Matheson, supra, at 881-82. According to the Carlton Declaration, the plaintiffs here will rely on objective evidence of market share and price discrimination (by category, not by individual merchant) to prove market power.²⁰ (Carlton Declaration KH 15-23.) Class-wide determination of defendants’ market power is warranted.

Id. at 31 (quoting 2 Areeda & Turner, Antitrust Law § 332c, at 157 (1978)). It may well be appropriate to address before trial the merits of some of defendants’ damages-related arguments, and whether, upon resolution of those issues, there remains a "practical means” to litigate the damage claims of class members. See infra § II. B(7).

c. Whether There Was Coercion

The defendants contend that question of “actual coercion” is not susceptible to class-wide proof because, for example, some members of the putative class would opt to take Visa and MasterCard off-line debit cards even absent the tie. However, in cases such as this one, in which the alleged tie is the product of an undisputed contractual provision, the “requisite coercion” is established. See Hill v. A-T-O, Inc., 535 F.2d 1349, 1355 (2d Cir.1976); see also Hovenkamp, supra, at 221 (“[C]ircuits hold uniformly that if all members of a putative class can produce a written contractual provision requiring them to take the tied product, then the court will assume that the seller actually forced the arrangement upon them. Class certification then will be prima facie appropriate on this issue.”). Defendants respond that Hill is inapposite because in that case customers were forced to buy an unwanted product, whereas here plaintiffs are not forced to buy anything: each use of an off-line debit card is the product of a decision by a consumer. This argument may or may not be relevant to the questions of what the contractual provision means or whether the disputed arrangement is a tie at all or whether injury in fact can be established. But it does not serve to distinguish Hill on the issue of coercion.²¹

d. The Attempt and Conspiracy to Monopolize Claims

The defendants devote a mere six lines in their sixty-page brief to their contention that class treatment is not appropriate for the § 2 attempt and conspiracy claims. Their argument is the same as their principal argument against the tying claim, namely, that the fact of injury is not susceptible to common proof. I addressed that argument above. The defendants do not challenge the appropriateness of class treatment of the actual elements of these claims, all of which focus on the defendants’ conduct and its effects on the relevant markets, factors that will not vary from plaintiff to plaintiff.

* * * * * *

For the foregoing reasons, I conclude that plaintiffs have carried their burden of establishing the prerequisites of a class action under Rule 23(a). For substantially the same reasons, I find that the plaintiffs have carried their burden under Rule 23(b)(3) by establishing that the common questions of law and fact identified predominate over any individual questions and that a class action is superior to any other means of adjudication. Without class certification, there are likely to be numerous motions to intervene, and millions of small merchants will lose any practical means of obtaining damages for defendants’ allegedly illegal conduct.

6. Rule 23(b)(2)

I also conclude that the action may be maintained under Rule 23(b)(2), which states that a class action may be maintained if “the party opposing the class has acted or refused to act on grounds generally applicable to the class, thereby making appropriate final injunctive relief or corresponding declaratory relief with respect to the class as a whole.” Fed.R.Civ.P. 23(b)(2). The notes to the rule list a suit “to test the legality of [a] ‘tying’ condition” as among the types of cases suitable for maintenance under Rule 23(b)(2). Id. advisory committee’s note. The “honor all cards” rule is “generally applicable” to all members of the class, and the request for an injunction ending it is central to the plaintiffs’ suit.

To be sure, the plaintiffs seek a large damages award as well, but the presence of a damages claim will not defeat maintenance of a class action under Rule 23(b)(2) when “the requested ... injunctive relief is a significant component of the overall relief which plaintiffs seek.” In re Nasdaq Market-Makers Antitrust Litig., 169 F.R.D. 493, 517 (S.D.N.Y.1996). Because the highly significant injunctive relief sought here is as important as the damages claimed, the cases cited by defendants, such as Kaczmarek v. International Business Machines Corp., 186 F.R.D. 307 (S.D.N.Y.1999), are inapposite. In Kaczmarek the court held that 23(b)(2) was unavailable because plaintiffs sought an injunction only “[a]s an aside” in á case in which equitable relief was not even available. Id. at 313. Here, the preliminary damages estimate is approximately $8 billion. By the same methodology, the requested injunction would save the putative class $63 billion over the next decade. The requested injunction is clearly not window dressing.

Having found certification appropriate under both Rule 23(b)(2) and (3), I recognize that “major problems can arise” because the procedural consequences vary under the subsections. Chateau de Ville Prods., Inc. v. Tams-Witmark Music Library, Inc., 586 F.2d 962, 966 n. 14 (2d Cir.1978). My inclination at this point is to provide general notice to the class, but to provide opt-out rights only as to the damages action. See 7A Charles Alan Wright, Arthur R. Miller & Mary Kay Kane, Federal Practice and Procedure § 1775, at 470 (2d ed.1986) (“If the Rule 23(a) prerequisites have been met and injunctive or declaratory relief has been requested, the action usually should be allowed to proceed under subdivision (b)(2). Those aspects of the case not falling within Rule 23(b)(2) should be treated as incidental. Indeed, quite commonly they will fall within ... Rule 23(b)(3) and may be heard on a class basis under [that] subdivision.”). The defendants have indicated that they plan to seek appellate review of an adverse decision on class certification, and no notice would issue during the pendency of any such review in the court of appeals. Thus, there will be ample opportunity to solicit the parties’ views on the question.

7. Rule 28(c)(1)

Rule 23(e)(1) states that an order deeming a class action maintainable “may be altered or amended before the decision on the merits.” I am “required to reassess [my] class rulings as the case develops.” Boucher v. Syracuse Univ., 164 F.3d 113, 118 (2d Cir. 1999) (quoting Barnes v. American Tobacco Co., 161 F.3d 127, 140 (3d Cir.1998), cert. denied, 526 U.S. 1114, 119 S.Ct. 1760, 143

L.Ed.2d 791 (1999)); accord Parkinson v. April Indus., Inc., 520 F.2d 650, 653 (2d Cir.1975) (“The initial order is strictly provisional .... ”). If factual or legal underpinnings of the plaintiffs’ successful class certification motion are undermined once they are tested under a more stringent standard than Caridad allows, a modification of the order, or perhaps decertification, might then be appropriate. The same would be true if it appears that individual damages issues will come to overwhelm common questions in a class maintained under Rule 23(b)(3). See 7B Charles Alan Wright, Arthur R. Miller & Mary Kay Kane, supra, § 1781, at 8-9 (“[T]he question of damages can be severed from that of liability and tried on an individual basis.”).

8. Appeal

Rule 23(f), which became effective on December 1, 1998, provides that “[a] court of appeals may in its discretion permit an appeal from'an order of a district court granting or denying class action certification under this rule if application is made to it within ten days after entry of the order.” Fed.R.Civ.P. 23(f). According to the note following this provision, the rule’s drafters contemplated that the decision whether to accept such an appeal is within “the sole discretion” of the court of appeals and can be analogized to the Supreme Court’s certiorari power. Id. advisory committee’s note. The committee note also states that “[t]he district court, having worked through the certification decision, often will be able to provide cogent advice on the factors that bear on the decision whether to permit appeal.” Id.

To the extent my advice is useful to the Second Circuit, I recommend that it take an interlocutory appeal of the class certification order. This litigation poses enormous financial risks for the defendants, risks that are obviously increased drastically by certification of the class. Moreover, this certification motion raises substantial and novel questions involving the standards a district court should apply in evaluating a class motion and the interaction of those standards with antitrust principles.

CONCLUSION

I certify a class of all persons and business entities who have accepted Visa and/or MasterCard credit cards and therefore have been required to accept VisaCheck and/or Master-Money debit cards under the challenged tying arrangements during the fullest period permitted by the applicable statute of limitations.²² The Class does not include the named Defendants, their directors, officers or members of their families. Plaintiffs’ counsel are appointed class counsel.

So Ordered.

1 . The full list of plaintiffs is: Wal-Mart Stores, The Limited, Sears Roebuck, Safeway, Circuit City, the International Mass Retail Association, the National Retail Federation, the Food Marketing Institute, Bernie’s Army-Navy Store, Auto-Lab of Farmington Hills, Burlington Coat Factoiy Warehouse, Sportstop, Payless Shoesource, Shoes, Etc., the Coffee Stop, UCC Kwik Doc, Computer Supplies Unlimited, Denture Specialist, Inc./Geneva White D.M.D., Shark 3 Audio, 53, Inc., and Scrub Shop.

2 . A bank can function as both a card-issuing institution and an acquiring institution. (Id. $ 7.)

3 . The complaint includes the following illustration of the chain of transactions:

Bank A issues a Visa credit card to Consumer X, who purchases a garment for $100 at Store Y, which was "acquired” for Visa by Bank B. Visa rules mandate that Bank B must pay Bank A an interchange fee of 1.25% of the amount of the transaction, i.e., $1.25. Bank B will charge Store Y a "discount fee” higher than $1.25 in order to recover the mandated interchange fee and other fees that Visa rules mandate Bank B to pay Visa on each and every Visa credit card (and debit card) transaction and to earn a profit for itself. Thus, Bank B may charge a discount fee of 1.60% of the transaction amount (or $1.60) to Store Y. When Store Y presents Consumer X’s $100 Visa transaction to Bank B, the bank will credit Store Y’s account for $98.40, send the Visa mandated $1.25 interchange fee to Bank A and retain the $.35 balance of the "discount fee.”

(Id. ¶8(o).)

4 . In 1994, Visa adopted a slightly different interchange fee for debit and credit, while MasterCard has continued to charge the same rate. (Id. ¶¶ 95-97.)

5 . This card, along with MasterCard's on-line Maestro card, is not subject to the "honor all cards” rule. Only merchants with PIN pads are equipped to accept these cards.

6 . For these reasons, the defendants' reliance on Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209, 113 S.Ct. 2578, 125 L.Ed.2d 168 (1993), and Matsushita Electric Industrial Co., Ltd. v. Zenith Radio Corp., 475 U.S. 574, 106 S.Ct. 1348, 89 L.Ed.2d 538 (1986), is misplaced. The language cited by defendants did not involve questions of admissibility at all, and certainly not at the class certification stage. Instead, the Supreme Court held that the expert economic opinions provided by plaintiffs in those cases were insufficient to withstand judgment as a matter of law. See Brooke Group, 509 U.S. at 242, 113 S.Ct. 2578 (expert opinion not supported by facts in the record "cannot support a jury’s verdict”); Matsushita, 475 U.S. at 594 n. 19, 106 S.Ct. 1348 (expert opinion based on “implausible" assumptions inconsistent with record evidence insufficient to create genuine issue of material fact to withstand summary judgment). Another case relied upon by defendants, In re Aluminum Phosphide Antitrust Litigation, 893 F.Supp. 1497 (D.Kan.1995), was decided under Daubert, but it decided a motion in limine before trial, not at the class certification stage. In that relatively fact-rich context, the court held the proffered expert opinion inadmissible because it was offered only "to plug evidentiary holes in plaintiffs’ case, to speculate, and to surmise.” Id. at 1506.

Finally, defendants rely on In re Agricultural Chemicals Antitrust Litigation, No. 94-40216-MMP, 1995 WL 787538 (N.D.Fla. Oct. 23, 1995). In denying class certification, that court did not rely on Daubert, but instead engaged in a lengthy, substantive critique of the findings of plaintiffs’ expert. See id. at *5 — *11. That kind of inquiry is not permitted in this circuit on a motion for class certification. See Caridad, 191 F.3d at 292; see also In re Polypropylene Carpet Antitrust Litig., 178 F.R.D. 603, 621 (N.D.Ga. 1997) (criticizing Agricultural Chemicals for “performing a function that is ... best reserved for a fact-finding stage of the proceedings”).

7 . Carlton graduated summa cum laude with a degree in applied math and economics from Harvard College in 1972; he went on to earn a Ph.D. from the Massachusetts Institute of Technology ("MIT”) in economics in 1975. He has served on the faculty at MIT (economics department) and at the University of Chicago (economics department, law school and (from 1984 to present) graduate school of business). He serves as the co-editor of the Journal of Law and Economics, has written two books, and has published more than fifty scholarly articles. He and Schmalensee were the two outside economists retained to assist in the Department of Justice’s preparation of its 1992 Horizontal Merger Guidelines. Carlton was also a scholar in residence at the Federal Reserve, where he lectured on the credit card industry. (Carlton Declaration, Exhibit A; Schmalensee Declaration, V 8.)

8 . The plaintiffs do not assert Rule 23(b)(1) as a basis for maintaining this class action.

9 . In particular, 25 of the 37 departments at Metro-North had no disciplinary cases at all during the relevant time period. See Caridad, 191 F.3d at 288. In addition, the defendant’s expert contended that the plaintiffs’ expert failed to take into account "non-promotable positions.” If jobs for which the rate of promotion was under one percent were excluded, the difference in promotion rates between whites and blacks disappeared. See id. at 289.

10 . Judge Walker dissented from the panel’s opinion. He concluded that as a matter of law plaintiffs’ allegation of "over-delegation,” which led to a series of allegedly discriminatory individual decisions, could not satisfy Rule 23(a)’s commonality requirement. See Caridad, 191 F.3d at 296 (Walker, J., dissenting). He agreed with the majority, however, that the district court erred in crediting the defendant’s expert; instead, the court should have "deferred" consideration of both reports until it considered the merits of the case. Id. at 296-97 (Walker, J., dissenting).

11 . Although X find it necessary to state the elements of the substantive claims and discuss them in relationship to the requirements of Rule 23, none of this discussion should be taken as my resolution of any merits question for purposes other than the pending motion.

12 . The defendants do not dispute that the proposed class of meets the numerosity requirement of Rule 23(a)(1). Nor do they do challenge the adequacy of class counsel under Rule 23(a)(4). Although they do not explicitly dispute commonality under Rule 23(a)(2), commonality, typicality, and adequacy of representation (putting aside consideration of adequacy of counsel) “tend to merge” because they "serve as guideposts for determining whether under the particular circumstances maintenance of a class action is economical and whether the named plaintiff's claim and the class claims are so interrelated that the interests of the class members will be fairly and adequately protected in their absence.” General Tel. Co. v. Falcon, 457 U.S. 147, 158 n. 13, 102 S.Ct. 2364, 72 L.Ed.2d 740 (1982).

13 . The plaintiffs contend that this is the rule in the Second Circuit, but they cite only this tepid language from a 1953 case: "But if the Sherman Act was violated, the actual damages flowing from that violation may well have included the additional amount which the plaintiffs had to pay because of the tie-in purchases they were compelled to make.” Bascom Launder Corp. v. Telecoin Corp., 204 F.2d 331, 334 (2d Cir.1953). This language is equivocal, and it was dictum. I therefore do not consider it to have decided this question.

14 . In this regard, plaintiffs point to an internal Visa document from 1990 that termed Visa expansion into the on-line debit market a "[b]usiness [n]ecessity” because development of a "mature” regional on-line debit "alternative” could cost Visa members $813 million due to "downward pressure” on credit card interchange. (Memo from Ron Hodges, Visa Debit Market Development, November 29, 1990, Bates Stamp 0024939.)

15 . I thus need not choose between the competing line of cases described above: for plaintiffs have proffered a sufficient theory of class-wide injury under either the "package” or the "tied product” measure of injury.

16 . In addition, plaintiffs argue that, for several reasons, transaction volume would not decline in the "but-for,” untied world. After Visa dramatically cut the interchange fee for Interlink, its online debit card, transaction volume increased. Moreover, lower off-line debit fees would reduce merchants' incentive to "steer” customers (i.e., encourage them to use other payment forms), thus increasing transaction volume. Some of the banks' lost interchange revenue might be recovered in the form of cardholder fees.

17 . Consumer surveys submitted by defendants show that a sizeable minority of cardholders prefer off-line to on-line debit. (Declaration of Mark Kirsch, Exhibits 98-99.)

18 . For purposes of the present motion, I reject defendants’ contention that there is no causation when this irreducible minimum of hard-core offline debit card users refuses to be steered. According to this argument, the customer’s refusal, not the tying arrangement, is the cause in fact of any injury. See Zenith Radio Corp. v. Hazeltine Research, Inc., 395 U.S. 100, 114 n. 9, 89 S.Ct. 1562, 23 L.Ed.2d 129 (1969) (“It is enough that the illegality is shown to be a material cause of the injury; a plaintiff need not exhaust all possible alternative sources of injury in fulfilling his burden of proving compensable injury under § 4.’’).

19 . Most of the cases simply state the rule that individual damages questions will not bar class certification in conclusory fashion without assessing whether individual questions would predominate if each damages action had to be tried separately. But that is not always the case. In Abrams v. Interco, Inc., 719 F.2d 23 (1983), the Second Circuit quoted favorably this passage from Professor Areeda:

The damage issue turns out to be a major stumbling block for [antitrust] class actions. The evidence establishing damages usually varies from class member to class member.... Even the necessity of individual trials on damages may be fatal if the class numbers in the thousands or millions. Individual litigations in such cases would be clearly infeasible, but the courts will not permit class actions unless they can devise a practical means for their litigation.

20 . The defendants indicate that they will dispute Carlton's definition of the relevant market. Resolution of this definitional question is susceptible to class-wide resolution.

21 . The defendants do not challenge the propriety of class certification as to the remaining element of the § 1 claim, namely whether a substantial share of interstate commerce was affected by the disputed arrangement.

22 . The parties have not briefed the statute of limitations aspect of class definition.