National Bancard Corp. v. Visa U.S.A., Inc.

Full title: NATIONAL BANCARD CORPORATION (NaBANCO), a Florida corporation, Plaintiff…

Court: United States District Court, S.D. Florida, Fort Lauderdale Division

Date published: Sep 20, 1984

596 F. Supp. 1231 (S.D. Fla. 1984)

Opinion

No. 79-6355-CIV-WMH.

September 20, 1984.

Carlton, Fields, Ward, Emanuel, Smith Cutler, James M. Landis, Christopher L. Griffin, Chris S. Coutroulis, Robert L. Ciotti and Sylvia H. Talbot, Tampa, Fla., for plaintiff.

Heller, Ehrman, White McAuliffe, San Francisco, Cal., M. Laurence Popofsky, Stephen V. Bomse, Robin E. Neuman, Michael J. Shepard, Charles F. Robinson, Fleming Huck, P.A., Paul C. Huck, Miami, Fla., for defendant.

MEMORANDUM OPINION AND FINDINGS OF FACT AND CONCLUSIONS OF LAW

HOEVELER, District Judge.

INTRODUCTION

This case involves a claim by plaintiff, NaBanco that defendant, VISA has violated the Sherman Antitrust Act, causing plaintiff damages. Defendant issues the VISA card, used by many as a method of payment for goods and services, in connection with the transfer of transaction paper from a merchant through its bank to the card issuing bank. VISA regulations call for payment of an interchange fee ("IRF") if its exchange system is used. The interchange fee is set by the VISA Board of Directors. Plaintiff, among other things, asserts that the methods of setting and determining the interchange fee involves price fixing and are anti-competitive. Defendant asserts that the setting of the fee is reasonable, not only in method but in amount, and further that the VISA regulations encourage competition and are not in violation of the Act. The Court has determined that the plaintiff, NaBanco has failed to prove its case and that judgment shall be entered in favor of the defendant, VISA.

The last 150 years have witnessed the evolution of several distinct methods of payment for goods and services. Just as we once moved from an economy which relied heavily on barter as a primary means of exchange to an economy based on cash, so we now find ourselves increasingly becoming a "cashless" society. Today it is not unusual to find major sectors of the buying public foresaking cash in favor of credit cards¹ and, even more recently, debit cards.² Unlike consumer currency purchases, however, these newer payment forms leave a residue of paper credits and debits which must somehow be cleared in order to complete a particular consumer transaction.

1 The term "credit card" is actually a misnomer. Not all so-called credit cards authorize the issuance of "credit," i.e., an opportunity to pay off amounts owed over an extended period of time at some rate of interest. To the extent that any credit card does not require a cardholder to pay for a purchase at the point of sale, however, a "free" or "convenience use" period is endemic to all credit card systems.

2 A "debit" card typically accesses a cardholder's asset account at the card-issuing bank although a line of credit may be available as "overdraft" protection when the account balance reaches zero.

Like any major economic transition, the movement from cash to cashless payment systems is not without growing pains. This case is itself evidence of the fact that social or economic change often leaves in its wake those who feel aggrieved by the process. National BanCard Corporation ("NaBanco") brought this action against VISA U.S.A., Inc.³ and the members of the VISA Board of Directors (hereinafter referred to collectively as "VISA") for alleged violations of Section 1 of the Sherman Act, 15 U.S.C. § 1. In its complaint NaBanco alleged a continuing combination and conspiracy by VISA to fix and maintain the price paid for bank credit card "interchange" transactions. For this claimed offense NaBanco sought damages in excess of three million dollars, trebling of damages, injunctive relief, attorneys' fees, and costs. Jurisdiction of this court was invoked pursuant to Sections 4 and 16 of the Clayton Act, 15 U.S.C. § 15 and 26, and pursuant to 28 U.S.C. § 1337.

3 VISA is a member of VISA International, which administers the VISA system worldwide. Through VISA International, the VISA card is honored in 120 countries and has 100 million cardholders and 3 1/2 million merchants worldwide. Cards issued by members of VISA are accepted internationally, and cards issued by members of VISA International outside the U.S. are accepted by merchants in the U.S. With the consent of the parties, the issues and discovery in this case were limited to activities in the United States. Russell TR. 5-6 1/16/84.

The case was tried by the court without a jury. It began May 10, 1982 and, to accommodate the pressing criminal docket of the Southern District of Florida, continued in segments of one or more weeks until its conclusion in January of 1984. The total trial time was in excess of nine weeks during which the court reviewed thousands of pages of exhibits and heard in person or by deposition dozens of witnesses, experts and otherwise. Based upon this record, and for the reasons enunciated below, the court finds that plaintiff NaBanco has failed to prove any of the alleged antitrust violations with which it has charged VISA.

THE BANK CREDIT CARD INDUSTRY

The issues raised in this case can best be defined and understood in the context of the bank credit card industry and its history.

In the early 1800's, the two principal means of commercial transactions payment devices were bank notes issued by state banks and drafts.⁴ These payment forms satisfied the needs of commerce at a time in history when consumers and merchants would usually reside and do business in the same geographical area. Accordingly, payment media rarely had to be sent beyond the local area. Bank notes, issued by the local bank or banks, circulated through the immediate region and were used to a far greater degree than currency is used today. In the larger local transaction, and also in the relatively infrequent long-distance transaction, the draft was the typical medium used.⁵

4 Wm. F. Baxter, Bank Interchange of Transactional Paper: Legal and Economic Perspectives, The Journal of Law Economics, Vol. XXVI (3) Oct. 1983, p. 556.

5 Id. at p. 557. If a consumer became indebted to a distant merchant, the consumer would execute payment by purchasing a draft from his local bank made payable to the merchant as payee. The draft would instruct the merchant's bank to make payment to the merchant in the amount of the indebtedness. For this, the consumer paid a substantial fee. If the consumer maintained a payments balance with the bank, this sum would be debited from the account. If not, the consumer's would be obligated physically to transport to the merchant bank a mutually acceptable form of currency.

However, rapid technological changes in both transportation and communication in the mid-1800's, heightened the need for a medium of exchange which was acceptable to diverse and unknown persons across the country and which could travel easily and cheaply. The check was increasingly employed to meet these needs.⁶

6 Id. at p. 560. The consumer would send the check to the distant merchant who presented it at his bank for collection. When the bank accepted the instrument for collection, it might credit the merchant's account immediately or await payment by the consumer's bank. The check itself was sent via a number of correspondent banks to the consumer's bank, where the consumer's account was debited that amount upon arrival of the check.

Despite the check's utility, other payment devices arose to meet more specific consumer needs. About a century after the check gained common acceptance, the bank credit card was introduced. The chief antecedents of this card were the retail merchant's "open book" account⁷ and the somewhat later-developed but more closely-analogous "travel and entertainment" cards.⁸ Unlike either of its two predecessors, however, the bank credit card has come to play a much more versatile role in the universe of payment systems. The bank credit card provides many of the same services as the personal check, but, in addition, provides retailers of goods and services an extra measure of protection from the risk of default. Incident to the system, and in most cases, the banks, not the retailers, are responsible for seeking payment from the retailers' customers.

7 Id. at p. 572. A form of short-term, interest free credit to customers whose patronage is highly valued. These customers tend to be persons whose time costs are high, who shop regularly at a particular retail outlet in high volumes, often buying more higher priced items.

8 Id. at 572-73. Travel and Entertainment cards. ("T E Cards") Developed as an answer to an increasingly mobile post-WWII society. The issuing organization (notably American Express and Diners Club) signs up merchants across the country of the type frequently patronized by travelers: hotels, resorts, restaurants, and a small number of pretigious retail outlets. Annual fees are normally much higher than those charged for the VISA or MasterCard and the system as a whole tends to cater to more "upscale" clientele.

As in the case of the check, the bank credit card system is principally a four-party payment arrangement. It involves: (1) cardholders who use bank credit cards to purchase goods and services; (2) merchants who accept bank credit cards in exchange for goods and services; (3) financial institutions (issuer banks) which issue cards to, and contract with, cardholders; and (4) financial institutions (merchant banks) which contract with merchants to accept the bank credit card and thereafter manage the bank credit card accounts of these merchant clients.

A typical transaction can be most simply described as follows: Once a potential consumer has opened a bank credit card account with a particular issuing bank, he or she may use that bank credit card in lieu of cash to purchase goods and services from any merchant participating in that particular bank credit card system. The merchant, after a sale, then transmits the consumer/cardholder's draft evidencing this transaction (referred to in the parlance of the industry as "paper") to its merchant bank, this sum being immediately credited to the merchant's account minus a small charge agreed upon earlier by contract (called the "merchant discount"). If the merchant bank happens to be the same bank which issued the card, the consumer/cardholder's account in the bank will be processed "in-house" in what has been described as an "on-us" transaction. When the issuer bank differs from the merchant bank, the process becomes more complicated. First, the merchant bank sends the transactional paper to the issuer bank. The issuer bank then will either send the merchant bank the requisite sums due and owing from its cardholder, or will directly credit the merchant bank's account at the issuer bank, if the merchant bank has such an account. In either case, the issuer bank is ultimately responsible for the sums due and owing from its cardholders, and thus, absent a breach of agreed procedure by the merchant or merchant bank, the issuing bank bears the risk of default by the cardholder.

The process by which transactional paper is moved from the merchant bank to the issuer bank involves certain costs. In the system at issue here, the issuer bank withholds a small amount (called the "interchange fee") from the monies due and owing the merchant bank to cover the costs of this processing. Even more simply stated, the merchant bank (if not the issuer of the card used in a sale) must process the paper generated by the sale to realize both the small profit hopefully provided in the merchant's discount charge to the retailer and reimbursement for the sale amount credited to the account of the retail merchant. The paper then goes to the issuer bank which reimburses the merchant bank but only after the deduction of the interchange fee. This interchange fee is the subject of the dispute in this case.

THE PARTIES The VISA System: Its History and Its Product

One of the most successful three-party bank cards was the California-based Bank of America's ("BA's") local prototype credit card program, the BankAmericard. This program expanded in 1960 to a statewide system to take advantage of the ever-increasing pool of people amenable to credit card use within the state. This expansion increased the base over which the costs of the system could be spread. Still, the nationwide expansion of its principal competitors' operations (travel and entertainment cards) in conjunction with dramatic technological changes in data-processing and electronic communications soon led BA to develop its own national credit card program in 1966.

The BA network was achieved by means of a franchise program which permitted BA-licensed local banks across the country to issue credit cards bearing the BankAmericard name. The licensee banks actually owned the credit cards they issued, and created their own receivables by signing up local merchants who would accept the card. Licensees were also encouraged to engage other non-issuer banks as their agents to expand the merchant base still further.

With the advent of the BankAmericard licensing program came the concept of the interchange fee, referred to earlier as a means of paying for costs associated with transferring transactional paper. Included in the licensing agreement was a provision indicating the sum or "fee" licensees were entitled to receive when purchasing their cardholders' paper in interchange. The agreement provided that the merchant-servicing licensee had to send to the card-issuing licensee either the actual merchant discount earned in each interchange transaction, or the merchant bank's "average" merchant discount. (Emerson TR. at 54. 1/13/84); Russell TR. at 15. (1/16/84); Exh. 2007.

Between 1966 and early 1970, many of the BA licensees grew in size and expertise and accordingly moved for the creation of a system which could more accurately reflect their individual interests and concerns, as well as address some of the more pressing problems facing the licensee system. (Larkin, TR. 232-33; (1/11/83). These concerns led to the formation of NBI, a for-profit, non-stock-membership corporation which was established in 1970.

Along with the formation of NBI came a new conceptual model for the interchange fee. The new fee was adopted by the NBI Board of Directors in 1971, in response to serious problems of uncertainty and instability which the old fee system could not resolve since it was based upon each merchant bank's interpretation of how much was due and owing to issuer banks. The new, uniform fee was based upon a cost-reimbursement methodology developed in conjunction with Arthur Anderson and Co., the national accounting firm, and was called the "Issuer's Reimbursement Fee" (hereinafter referred to as "IRF"). This fee or charge reflected the concept of calculated reimbursement.

In 1977, NBI changed its name to VISA and exclusive rights to the name "BankAmericard" reverted to BA. To become a VISA member, of which there were approximately 13,400 in 1983, a financial institution must be eligible for federal deposit insurance. Upon proper application, any eligible institution can serve either as a proprietary member or as an agent of a proprietary member. There were 1866 proprietary members in 1983. Each can perform the function of issuing cards to cardholders and of signing merchants to participate in the VISA system. Proprietary members also elect VISA's governing Board of Directors which makes the rules under which the system operates. Agent members, 11,537 strong in 1983, elect only to contract with merchants on behalf of proprietary members; they are not eligible to vote for Board members and therefore do not play as significant a role as proprietary members in determining VISA policy.

It is important to understand that at all times material to the claims in this case, VISA did not require its participating members to interchange paper through the VISA system. Rather, members could elect to exchange transactional paper pursuant to separately negotiated agreements with other members without the imposition of an exchange fee by the VISA system. If, however, members choose instead to interchange their paper through VISA, the data on the paper will be sent in electronic form by the merchant bank (or its processing agent) through what is known as the VISA Base II computer system to the issuer bank (or its processing agent). It is only when members use BASE II to interchange their paper that interchange will be governed by the highly complex set of VISA Operating Regulations, which includes the IRF and other rules governing the conditions under which issuer banks are required to "purchase" transactional paper from merchant banks.

NaBanco

Plaintiff, NaBanco, formed under the name Southern Bankcard Corporation in 1974, is neither a proprietary nor an agent member of VISA. It is not eligible for VISA membership because it is not a financial institution eligible for federal deposit insurance. It participates in the VISA bank credit card business by contracting with various VISA members to serve as their processing agent, thereby obtaining access to the VISA Base II system.

In its capacity as an agent for a VISA member, NaBanco typically performs activities similar to those performed by merchant banks, such as providing authorization services and services relating to the processing of merchant transactions.⁹ The majority of NaBanco's business is with VISA members who have already negotiated and contracted with particular merchants. Bull TR. 670-72 (5/13/82). NaBanco then provides its services pursuant to the contract it has with the VISA member, ordinarily electing not to be paid a transaction fee for such services but to receive instead most or all of the merchant discount revenues that merchants would otherwise remit to the VISA member. In return for these merchant discount receipts, NaBanco agrees to discharge the VISA member's obligation to pay the IRF whenever sending data in interchange through VISA's Base II system. Bull TR. 336 (5/11/82) Exh. 6. Nevertheless, under VISA's rules, the VISA member continues to remain legally liable to VISA for payment of the interchange fee on all such transactions.

9 In a few instances, Nabanco has also served as an agent of a VISA member on the card issuing side. As a result, NaBanco "receives" IRF in the same sense that it sends it out; re, under the contract it negotiated with the VISA member. Exh. 515. Bull Tr. 752-56 (5/13/82).

In a minority of cases, NaBanco, on behalf of VISA members, negotiates directly with merchants regarding merchant discount rates and other terms of the services to be provided to them. In this capacity, NaBanco competes with nearly all of the VISA members, and testimony both from merchants and from VISA members indicates that NaBanco is one of the leading contenders for the business of major retailers nationwide.¹⁰

10 NaBanco describes itself as the "premier nonbank third party credit clearinghouse." Ex. 361. During 1976-1980, NaBanco's revenues went from $1 million to $9 million. In that same period, its earnings increased from $69,000 to $565,000. Among the major merchants for whom NaBanco is the processor are County Seat Stores, Eckerd Drugs, Waldenbooks, and Petrie Stores, and many major department stores, including Burdines' and Rich's. Bull TR. 1052-61 (9/13/82) Exh. 2199.

THE NATURE OF THIS ACTION

Among other bases for plaintiffs charges of antitrust violations by defendant, plaintiff contends that the interchange fee requirement (when processing through BASE II) diminishes its ability to compete for merchant business on an equal basis with VISA member banks which do both issuing and merchant business. The "on-us" transaction (where the card issuing bank and the merchant contracting bank are the same) does not present the need for interbank exchange. Thus, says NaBanco, these proprietary members of VISA — who have cooperated in mandating the interchange fee in non "on-us" transactions, have curtailed its ability to compete.

"On-us" transactions occur with the greatest frequency in the immediate geographic vicinity of a VISA member bank (especially where that member is a large card issuer) since it is in this area that a local merchant is most likely to have an account with the same bank as an area resident/cardholder.

NaBanco argues that in order to compete effectively with VISA member banks, it has had to develop relationships with merchants outside its immediate geographic market and find retailers with sufficient locations, or a broad enough customer base, so that the effects of "on-us" transactions are eliminated or at least minimized. Bull TR. 249 (5/11/82). NaBanco also claimed it has found it necessary to sign up merchants in the United States who cater to tourists whose cards have been issued outside the United States. Alternatively, NaBanco argued it has had to establish agency relationships with banks outside the United States since the interchange fee charged on international transactions has traditionally been less than that charged for domestic exchanges. Bull TR. 275-277 (5/11/82).

Nonetheless, NaBanco has claimed it can cite numerous examples of its inability to compete with VISA member banks for merchant contracts due to the competitive advantage these banks enjoy from "on us" transactions. Given the fact that both "on-us" and IRF transactions involve many of the same types of processing costs whether they occur "in-house" or not, NaBanco has questioned the intent of VISA in setting particular IRF levels. Specifically, NaBanco has contended that IRF is not cost-related and, therefore, is either unnecessary or set intentionally high so as to discourage competitors such as NaBanco.

In the NaBanco scenario, individual merchant banks compete with it to sign up and serve merchants but unfairly so when they are also in the position of issuer banks because in this latter sense they act like a group of competitors who have conspired to sell goods to one another (actually to themselves) at a lower rate than the one offered NaBanco as a non-issuer bank thus affecting the price at which NaBanco can offer its merchants services. Since NaBanco believes that individual issuer and merchant banks can and should negotiate with one another over the price to be paid for transactional paper, VISA member banks, says NaBanco, are nothing more than a group of competitors whose setting of IRF is akin to horizontal price-fixing for transactional paper and, therefore, such conduct is absolutely proscribed by Section 1 of the Sherman Act.

In terms of remedies, NaBanco has asked that this court enjoin VISA from setting IRF at any level other than "par" (zero) so that financial institutions with large numbers of "on-us" transactions will have no competitive edge over entities like NaBanco which have few or none. Essentially, NaBanco has asked the court to "fix" its own price for IRF but in accordance with terms NaBanco considers fair. Alternatively, NaBanco has asked the court to enjoin VISA from setting any particular IRF, thus permitting each merchant bank (or NaBanco) to establish individually-negotiated "IRFs" with issuer banks. In addition, NaBanco has sought over three million dollars in damages, trebled, to compensate it for the several years during which it paid IRF to its alleged detriment.

STANDING TO SUE

Before reaching any of NaBanco's anti-trust arguments, it is necessary to address defendant VISA's argument that NaBancoe lacks standing under the antitrust laws to bring its action. Since NaBanco brought this action seeking both damages and injunctive relief, and since the standards for standing differ as to each, two separate but related standing analyses follow.

Standing to Bring The Treble Damages Claim

Under Section 4 of the Clayton Act, "any person . . . injured in his business or property by reason of anything forbidden in the antitrust laws" may recover treble damages plus costs and attorney's fees. 15 U.S.C. § 15.¹¹ The Supreme Court has stated that § 4 on its face contains little in the way of restrictive language, but has read the "by reason of" language cited above as encompassing several separate and analytically distinct elements, each implying that the statute is not as broad as its words suggest.

11 There is also a "commerce clause" requirement under Section 4; however, the parties have already stipulated to the fact that the practices examined in this case affect commerce.

Neither courts nor parties to antitrust actions have been scrupulous in their attempts to keep the various restrictions on a plaintiff's right to a treble damage action separate and distinct. All seem to agree that a plaintiff must initially establish the existence of a causal connection between defendant's antitrust violation and plaintiff's injury. McClure v. Undersea Indus., Inc., 671 F.2d 1287, 1289 (11th Cir. 1982), cert. denied, 460 U.S. 1037, 103 S.Ct. 1427, 75 L.Ed.2d 788 (1983) ("The term `fact of damage' refers to causation and simply means `that the antitrust violation cause injury to the antitrust plaintiff.'" (quoting Alabama v. Blue Bird Body Co., Inc., 573 F.2d 309, 317 (5th Cir. 1978).)). Once the causal connection between plaintiff's injury and defendant's act is shown, however, the courts begin to differ about what additional elements a plaintiff must prove in order to successfully bring a treble damage action.

The cases suggest there are at least two, perhaps three, additional requirements for standing to maintain a treble damage action. The first of these requirements takes many forms, as most courts vacillate between the two alternative techniques of (a) measuring the directness of the injury complained about and (b) asking whether the plaintiff falls within the target area of the violation. The second requirement, which some courts do not distinguish from the standing requirement, is that an antitrust plaintiff must prove he has suffered "antitrust injury," as defined in Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 97 S.Ct. 690, 50 L.Ed.2d 701 (1977); i.e., injury of the type the antitrust laws were intended to prevent and that flows from that which makes defendants' acts unlawful.¹² Finally, some plaintiffs must also address the issue of defensive and offensive use of "pass-on" damages, considered by the Supreme Court in Hanover Shoe v. United Shoe Mach. Corp., 392 U.S. 481, 88 S.Ct. 2224, 20 L.Ed.2d 1231 (1968) and Illinois Brick Co. v. Illinois, 431 U.S. 720, 97 S.Ct. 2061, 52 L.Ed.2d 707 (1977). For the most part, courts deal with this last issue in conjunction with their discussions of the standing argument, perhaps because of a dearth of Supreme Court opinions delineating the standing requirement; however, it is important to note that neither Hanover Shoe nor Illinois Brick are "standing" cases per se. Therefore, to the extent the policies these cases promote are relevant to a discussion of a plaintiff's right to bring a treble damage action and are not addressed in a court's discussion of a plaintiff's "standing" to bring such an action, they must be separately considered.

12 "The injury should reflect the anticompetitive effect either of the violation or of anticompetitive acts made possible by the violation. It should, in short, be the `type of loss that the claimed violations . . . would be likely to cause.'" 429 U.S. at 489, 97 S.Ct. at 697 (emphasis in original) (quoting Zenith Radio Corp. v. Hazeltine Research, Inc., 395 U.S. 100, 125, 89 S.Ct. 1562, 1577, 23 L.Ed.2d 129 (1969)).

Tests for Standing

Lower federal courts have adopted various tests for antitrust "standing" to preclude suits by plaintiffs considered too remote from the alleged violation to be entitled to recover damages. Some cases have applied a "direct injury" test of standing which focuses on the relationship between the plaintiff and the alleged antitrust violator. Repp v. F.E.L. Publications, Ltd., 688 F.2d 441, 444-45 (7th Cir. 1982); Malamud v. Sinclair Oil Corp., 521 F.2d 1142, 1149 (6th Cir. 1975); In re Multidistrict Vehicle Air Pollution M.D.L. No. 31, 481 F.2d 122, 127 (9th Cir. 1973), cert. denied, 414 U.S. 1045, 94 S.Ct. 551, 38 L.Ed.2d 336 (1973). Generally, a plaintiff separated from the violator by one or more intermediate tiers of victims lacks standing under the direct injury test because, by analogy to tort law, the injury received is too remote to be fairly characterized as proximately resulting from the defendant's antitrust violation. Today, no circuit continues to apply a strict "direct injury" test. Instead, courts borrow from the language of that analysis, as well as from that of Hanover Shoe and Illinois Brick, to discourage or dismiss suits brought by plaintiffs whose injuries result from chains of events too speculative to justify treble damages. In essence, this is the position taken by VISA in arguing that NaBanco is not a "direct purchaser" of IRF and is therefore not eligible to recover overcharging damages. Only the merchant banks with which NaBanco contracts or, alternatively, the merchants themselves, are direct enough purchasers of IRF to be eligible to sue under Section 4 according to VISA. There is some support for VISA's position in cases which have held, based in part on the policies underlying Illinois Brick, that a plaintiff who purchased from a non-conspiring competitor of horizontal price-fixing defendants may not recover damages on the theory that the company from which it purchased charged an artificially inflated price under the "umbrella" of the defendant's price-fixing conspiracy. Mid-West Paper Prods., 596 F.2d 573, 587; In re Folding Carton Antitrust Litigation, 88 F.R.D. 211, 219-20 (N.D.Ill. 1980); In re Petroleum Prods. Antitrust Litigation, 497 F. Supp. 218, 227-228 (C.D.Cal. 1980), aff'd on other grounds, 691 F.2d 1335 (9th Cir. 1982); Liang v. Hunt, 477 F. Supp. 891, 896-97 (N.D.Ill. 1979); cf. In re Petroleum Prods. Antitrust Litigation, 691 F.2d 1335, 1341 (9th Cir. 1982) ( Illinois Brick bars umbrella claim in the context of a multi-tiered distribution chain).

The other test popular among courts and, as NaBanco is quick to point out, routinely employed in this Circuit, focuses on the nature of the antitrust violation and the plaintiff's relationship to the area of the economy claimed to have been injured by the defendant's conduct. Const. Aggregate Transport v. Florida Rock Ind., 710 F.2d 752 (11th Cir. 1983); Litton Sys., Inc. v. American Tel. Tel. Co., 700 F.2d 785, 821 (2d Cir. 1983); Aurora Enters., Inc. v. National Broadcasting Co., 688 F.2d 689, 692-93 (9th Cir. 1982); In re Municipal Bond Reporting Antitrust Litigation, 672 F.2d 433, 435-36 (5th Cir. 1982); Engine Specialties, Inc. v. Bombardier Ltd., 605 F.2d 1, 12-15 (1st Cir. 1979), cert. denied, 446 U.S. 983, 100 S.Ct. 2964, 64 L.Ed.2d 839 (1980); see also Cleary v. Chalk, 488 F.2d 1315, 1319 n. 17 (D.C. Cir. 1973), cert. denied, 416 U.S. 938, 94 S.Ct. 1940, 40 L.Ed.2d 289 (1974) (citing decisions under the target area standard without expressly adopting that test). Called the "target area" test, it requires the plaintiff to establish that he was "within that section of the economy . . . endangered by a break-down of competitive conditions in a particular industry" and that the illegal practices were aimed at him. Jeffrey v. Southwestern Bell, 518 F.2d 1129, 1131 (5th Cir. 1975). See also Yoder Bros., Inc. v. California-Florida Plant Corp., 537 F.2d 1347 (5th Cir. 1976); see also Southern Concrete Co. v. United States Steel Corp., 535 F.2d 313 (5th Cir. 1976); Buckley Towers Condominium, Inc. v. Buchwald, 533 F.2d 934 (5th Cir. 1976).

Unfortunately, courts which have used a target area analysis to determine antitrust standing have never agreed on the general requirements of the test, sometimes resulting in apparently inconsistent conclusions. This situation led the Court in Yoder Bros., supra, 1360, to comment: "A test focusing on the sector of the economy is more stated than applied." Similar kinds of problems in applying the "direct injury" test have led the Third Circuit, as well, (and to some extent the Eighth Circuit; Paschall v. Kansas City Star Corp., 605 F.2d 403, 409-10 (8th Cir. 1979)) to reject both tests in favor of a case-by-case approach called the "factual matrix" test. Cromar Co. v. Nuclear Materials Equipment Corp., 543 F.2d 501 (3d Cir. 1976). The Sixth Circuit has adopted for antitrust matters the test employed for determining standing in administrative law cases, i.e. requiring the plaintiff to show that its affected interests arguably are within the "zone of interests" protected by the antitrust statutes. Malamud v. Sinclair Oil Corp., 521 F.2d 1142, 1148 (6th Cir. 1975); see also Chrysler Corp. v. Fedders Corp., 643 F.2d 1229, 1234 (6th Cir. 1981), cert. denied, 454 U.S. 893, 102 S.Ct. 388, 70 L.Ed.2d 207 (1981).

At the heart of the Third Circuit's "factual matrix" test is a recognition of the reason for the existence of different Section 4 requirements; i.e., the multifaceted problem of determining when a person is sufficiently "injured in his business or property by reason of anything forbidden in the antitrust laws" so as to be eligible to sue for treble damages. Thus, the Mid-West Paper Prods. court concluded that only together do the separate Section 4 requirements "constitute the judicial gloss upon the words of § 4 by which the courts have patrolled the portals to a treble damage action." Mid-West Paper Prods., supra, 582. Citing the observation of the Supreme Court in Hawaii v. Standard Oil Co., 405 U.S. 251, 92 S.Ct. 885, 31 L.Ed.2d 184 (1972), that underlying the precepts that guard access to Section 4 is the recognition that the treble damage action represents a congressional determination that the antitrust laws be enforced in large measure by "private attorneys general" who are encouraged to sue by the prospect of recovering three times their actual damages, the Third Circuit went on to say:

Accordingly, whether the issue is the definition of injury in § 4, or the use of pass-on theories, or the proper scope of the standing doctrine, the line drawing that necessarily takes place is informed by an inquiry into what posture best effectuates the dual purposes of the treble damage remedy, namely, compensating victims of antitrust violations for their injuries and deterring violators by depriving them threefold of "the fruits of their illegality," while at the same time furthering the overriding goal of the antitrust laws — preserving competition. Mid-West Paper Prods. at 583 (quoting Hawaii v. Standard Oil Co., 405 U.S. 251, 262, 92 S.Ct. 885, 891, 31 L.Ed.2d 184 (1972)). Aside from these basic objectives of antitrust enforcement, ". . . additional concerns must be addressed in determining whether antitrust goals would be furthered by adopting a particular position. Among such concerns are whether a duplicative or ruinous recovery will result and whether resolution of the claim will unduly complicate the trial by necessitating the pursuing of complex and conjectural economic lines of causation and effect." (citations omitted).

This court could certainly do worse than to adopt these principles as its own in deciding whether or not NaBanco has standing to bring its treble damage action. This is not to say this court adopts the "factual matrix" test used by the Third Circuit. Neither, however, does the court feel bound by the dictates of the target area test as NaBanco urges. The Eleventh Circuit in Construction Aggregate Transport (supra) may have been correct in characterizing the Supreme Court's decision in Blue Shield of Virginia v. McCready, 457 U.S. 465, 102 S.Ct. 2540, 73 L.Ed.2d 149 (1982) as implicitly sanctioning continued, flexible use of the target area test. The two recent Supreme Court decisions which have even addressed the standing issue, Blue Shield and Associated Gen. Contractors, Inc. v. California State Council of Carpenters, 459 U.S. 519, 103 S.Ct. 897, 74 L.Ed.2d 723 (1983), while never going so far as to adopt any particular test, have acknowledged their existence as tools to be employed by courts to determine standing, as such tests are useful.

Nevertheless, the Associated Gen. Contractors case, following on the heels of Blue Shield, seems to have gone further than Blue Shield in not only refusing to adopt any of the particular tests used by various circuit courts but in actually delineating the skeleton of a new standing test for treble damage actions. The Supreme Court observed, in reference to different standing tests, that "these labels may lead to contradictory and inconsistent results." Associated Gen. Contractors, supra. Likening the task of federal judges to articulate a precise test to determine antitrust standing in treble damage actions to the struggle of common-law judges to announce a precise definition of the concept of "proximate cause," the Supreme Court emphasized the virtual impossibility of announcing a "black-letter rule that will dictate the result in every case," in light of the infinite variety of claims that can arise in the antitrust context. Id., 459 U.S. at 537, 103 S.Ct. at 908, 74 L.Ed.2d at 737. Instead, the Supreme Court suggested that courts "analyze each situation in light of the factors set forth" in its analysis of the plaintiff's standing position in the case before it. Id. 103 S.Ct. at 908 n. 33.

In light of its opinion in Associated Gen. Contractors, the Supreme Court vacated and remanded two appeals court decisions for further consideration,¹³ and, following that lead, several courts have concluded that the Associated Gen. Contractors opinion requires rejection of standing tests previously adopted by the courts of appeals. See Southaven Land Co. v. Malone Hyde, Inc., 715 F.2d 1079 (6th Cir. 1983); see also Meyer Goldberg, Inc. v. Goldberg, 717 F.2d 290 (6th Cir. 1983); Merican, Inc. v. Caterpillar Tractor Co., 713 F.2d 958 (3rd Cir. 1983); Indium Corp. of Am. v. Semi-Alloys, Inc., 566 F. Supp. 1344 (N.D.N.Y. 1983); Chelson v. Oregonian Publishing Co., 715 F.2d 1368 (9th Cir. 1983). The Eleventh Circuit in Construction Aggreg. (supra) while citing the Supreme Court's Associated Gen. Contractors decision, did not expressly consider the effect of that decision on the continued validity of the target area analysis. Thus, the scope of Construction Aggreg. as it relates to this question limits its application to the facts of this case. Associated General Contractors, supra, at 908 n. 33.

13 See, Industrial Inv. Dev. Corp. v. Mitsui Co., 671 F.2d 876 (5th Cir. 1982), vacated, 460 U.S. 1007, 103 S.Ct. 1244, 75 L.Ed.2d 475 (1983), on remand, 704 F.2d 785 (5th Cir. 1983); Ostrofe v. H.S. Crocker Co., 670 F.2d 1378 (9th Cir. 1982), vacated, 460 U.S. 1007, 103 S.Ct. 1244, 75 L.Ed.2d 475 (1983).

Whether or not Associated General Contractors truly represents a new standard for antitrust "standing," or is rather a purely semantical argument, this court adopts its language as the most thorough pronouncement of the Supreme Court to date on the standing issue in treble damage actions. At its core, Associated General Contractors admonishes us "to evaluate the plaintiff's harm, the alleged wrongdoing by the defendants, and the relationship between them" in the context of common-law damages litigation in 1890. 459 U.S. 519, 535, 103 S.Ct. 897, 907, 74 L.Ed.2d 723, 736. As guideposts for this evaluation the Court pointed to "previously decided cases [which] identify factors that circumscribe and guide the exercise of judgment in deciding whether the law affords a remedy in specific circumstances." 459 U.S. at 537, 103 S.Ct. at 908, 74 L.Ed.2d at 737. In light of the facts before it the Supreme Court found at least two factors favoring judicial recognition of the plaintiff's antitrust claim: (1) a causal connection and (2) intent to harm, although as to this latter factor the Supreme Court reiterated its Blue Shield holding that "[t]he availability of the § 4 remedy to some person who claims its benefit is not a question of the specific intent of the conspirators." 459 U.S. 537, 103 S.Ct. 908, 74 L.Ed.2d 738 (quoting Blue Shield, 457 U.S. 465, 102 S.Ct. 2540, 73 L.Ed.2d 149.) Factors favoring the defendants and which the Supreme Court concluded were controlling in Associated Gen. Contractors included: (1) "the nature of the plaintiff's alleged injury," (citing Brunswick v. Pueblo Bowl-O-Mat) 459 U.S. 538, 103 S.Ct. 908, 74 L.Ed.2d 738 and (2) "the directness or indirectness of their asserted injury." 459 U.S. 540, 103 S.Ct. 910, 73 L.Ed.2d 739. As to the issue of "directness" the Supreme Court first identified the chain of causation then highlighted the importance of finding an

"identifiable class of persons whose self-interest would normally motivate them to vindicate the public interest in antitrust enforcement . . . [and therefore] diminish the justification for allowing a more remote party to perform the office of private attorney general." ( 459 U.S. p. 542, 103 S.Ct. p. 911, 74 L.Ed.2d p. 741).

As part of its discussion of the "directness or indirectness" of the plaintiff's harm, the Supreme Court also raised its concerns about highly speculative damage claims, saying:

"The indirectness of the alleged injury also implicates the strong interest, identified in our prior cases, in keeping the scope of complex antitrust trials within practically manageable limits. These cases have stressed the importance of avoiding either the risk of duplicate recoveries on the one hand, or the danger of complex apportionment of damages on the other." 459 U.S. 543-544, 103 S.Ct. 911-912, 74 L.Ed.2d 741-42.

Turning now to the case at hand, the court first points out that a plaintiff's "standing to sue" is a preliminary matter to be evaluated in this circuit upon the pleadings alone. Const. Aggregate Transport at 763; Yoder Bros. at 1359 (standing to sue is a preliminary matter to be evaluated upon the allegations of the complaint). NaBanco's allegations that it has been unable to compete against VISA member banks for merchant contracts due to the competitive advantage these banks enjoy from "on-us" transactions seem sufficient on their face to suggest a harm causally related to VISA's setting of an IRF. VISA's argument that NaBanco's own conduct severs the chain of causation because it has had realistic alternatives available to it to avoid paying IRF misses the mark. The antitrust laws do not require a potential plaintiff to take extraordinary measures to avoid anticompetitive activity; in fact, this is one of the things the antitrust laws are designed to prevent. Simpson v. Union Oil Co., 377 U.S. 13, 17, 84 S.Ct. 1051, 1054, 12 L.Ed.2d 98 (1964); Columbia Broadcasting System, Inc., v. ASCAP, 562 F.2d 130, 136 (2d Cir. 1977), rev'd on other grounds, 441 U.S. 1, 99 S.Ct. 1551, 60 L.Ed.2d 1 (1979).

Further, the court finds that, under any characterization of antitrust standing currently urged upon us, NaBanco is a technically proper party to bring suit for treble damages. Using the methodology developed by the Supreme Court in Associated Gen Contractors, the court does not find that plaintiff has alleged or proved any specific intent on the part of VISA to harm NaBanco or those who stand in NaBanco's shoes but neither is such intent a necessary prerequisite to Section 4 standing. Plaintiff's alleged injuries are of the type Congress sought to remedy in providing a private remedy for violations of the antitrust laws. This is not a case, like Brunswick, in which plaintiff argues that the anticompetitive behavior by defendant (here, the setting of IRF) led to loss of profits due to increased competition in the market. Rather, NaBanco complains that it has been unable to compete effectively with some VISA merchant banks due to price-fixing by VISA. NaBanco not only argues the potential for current competitive losses, but argues that innovative technological changes are being discouraged by VISA's anticompetitive activities. To the extent that such changes also present the possibility for increased competition in the future, NaBanco seems to have more than sufficiently met the requirement of Section 4 standing to bring its challenge to VISA's Interchange Fee Arrangement.

Finally, the court does not find that NaBanco's alleged harm is either too remote or indirect to prohibit it from bringing its treble damage action. In this case, the "chain of causation" between plaintiff's claimed injury and the alleged restraint promoted by VISA seems "direct" enough to sustain plaintiff's standing. To argue, as VISA does, that the more appropriate plaintiffs are the merchant banks because they alone are in privity with VISA is to ignore the economic reality. While it is true that VISA merchant banks remain ultimately responsible for IRF, they are not primarily liable for its payment in all cases. Primary liability also lies with NaBanco as the merchant banks's functional equivalent in those cases in which VISA merchant banks (in contract with NaBanco) choose not to process transactional paper generated by VISA card purchases. In this way NaBanco aids the smooth functioning of the VISA system, a role which VISA should be reluctant to attentuate inasmuch as some of its participating merchant banks evidently do not find it cost-effective to perform the processing function themselves. Further NaBanco points to significant technological changes which it has made to accommodate changes in VISA's processing methodology. Bull TR. 593 (5/12/82) P.Ex. 1172; D.Ex. 1180. Insofar as these are changes which have not had to be made by VISA merchant banks which use NaBanco or NaBanco-like institutions to process transactional paper, VISA member banks are hardly in a position to say that they are more directly affected by the setting of IRF than NaBanco.

Similarly, VISA's argument that Illinois Brick outlaws treble damages for "functional equivalents" is misplaced. Associated Gen. Contractors demonstrates that, to the extent that Illinois Brick is relevant to antitrust standing, it is so in how the policies identified and promoted by that decision relate to the standing issues presented in this case. The Illinois Brick Court was primarily concerned with problems of dual recovery and the possibility of having to calculate the variable incidence of harm experienced by different antitrust victims where offensive use of pass-on damages was alleged. Arguably, these are problems which also arise in the context of this case. For instance, it is possible, as VISA points out, that some of the merchant banks with which NaBanco contracts actually absorb some of the costs and economic consequences associated with IRF in their individual contracts with NaBanco. This would, of course, cause problems for the court in calculating the extent of damage borne by NaBanco versus that borne by the merchant banks with which NaBanco contracts. But identifying the merchant banks as the more appropriate plaintiffs to bring this action does not eliminate our problems regarding the calculation of damages. In fact, it compounds them. At least when NaBanco pays IRF we can assume, even if not with complete accuracy, that it bears the full economic cost of that payment without having to examine its individual contracts with merchant banks. It becomes harder for us to make that assumption in the case of merchant banks which use NaBanco since these banks do not pay IRF directly but only by virtue of whatever amount their contracts with NaBanco suggest they pay. We have similar problems with VISA's alternative argument that merchants, not merchant banks, are more appropriate plaintiffs than NaBanco to bring this suit. If what VISA is really arguing is the defense of pass-on damages, the court notes that this defense is only applicable where there is an existing "cost-plus" contract or its functional equivalent; i.e., a contract in which the impact of the illegal overcharge is determined in advance of the actual overcharge without reference to the interactions of supply and demand and other market forces. Hanover Shoe, 392 U.S. at 494, 88 S.Ct. at 2232 (supra); In Re Plywood Antitrust Litigation, 655 F.2d 627 (5th Cir. 1981).

An argument that NaBanco's contracts were really cost plus or a legal equivalent is, at least, superficially persuasive. However, NaBanco's response to what it perceived to be the IRF problem was varied and thus, presents a set of facts in which the "cost plus" assertion appears wanting.

Further, VISA's standing agreement suffers from a consideration of the second of the dual concerns of the Illinois Brick Court, the impairment of anti-trust enforcement by a reduced incentive to sue. In Illinois Brick and Hanover Shoe the Supreme Court voiced this particular argument in terms of the diffusion of benefits to each plaintiff when potential recoveries had to be divided among a large group of plaintiffs down the line of economic causation, saying: "Added to the uncertainty of how much of an overcharge could be established at trial would be the uncertainty of how that overcharge would be apportioned among the various plaintiffs." Illinois Brick 431 U.S. at 745, 97 S.Ct. at 2074. By allowing enforcement only by those who actually pay IRF, we limit the pool of potential plaintiffs. Thus, the concern for reducing the effects of treble damage suits brought by indirect purchasers with a smaller stake in the outcome is alleviated. Illinois Brick at 724, 97 S.Ct. at 2064.

Even more importantly, however, we assure ourselves of a potential plaintiff at the very least. NaBanco does fall prey to some of the same conflicts of interest which beset many VISA member banks which both issue cards and serve merchants, raising questions about its ability to pursue vigorously any cause of action against VISA; however, in arguing that only merchant banks have standing to sue it, VISA essentially invites its members to sue themselves for antitrust violations from which they in fact may benefit. While the possibility of suit by VISA member banks does exist, it is also more reasonable to conclude that member banks will attempt to ameliorate any problems they have with the IRF via their more direct access to the VISA Board of Directors. This opportunity is less available to NaBanco as a VISA non-member.

Injunctive Relief

Aside from seeking treble damages, NaBanco asks the court to enjoin VISA from continuing to conspire to fix prices. This request is based on § 16 of the Clayton Act, 15 U.S.C. § 26, which states in relevant part:

Any person, firm, corporation, or association shall be entitled to sue for and have injunctive relief, in any court of the United States having jurisdiction over the parties, against threatened loss or damage by a violation of the antitrust laws, . . . when and under the same conditions and principles as injunctive relief against threatened conduct that will cause loss or damage is granted by courts of equity. . . .

Given the fact the court has already recognized NaBanco's standing under § 4 of the Clayton Act, an elaborate discussion of NaBanco's standing under § 16 is unnecessary. Most "courts take a less constrained view of standing in suits involving injunctive relief than in those demanding treble damages. . . ." Jeffrey v. Southwestern Bell, supra, at 1132. Even indirect purchasers can maintain actions for injunctive relief. In re Beef Indus. Anti-trust Litigation, 600 F.2d 1148, 1167 (5th Cir. 1979), cert. denied, 449 U.S. 905, 101 S.Ct. 280, 66 L.Ed.2d 137 (1980). This is because "the policy considerations underlying the pass-on rule are not implicated in claims for injunctive relief." Id. Neither the risk of exposure to multiple liability or the difficulty in tracing the allocation of the overcharging among different levels of purchasers is related to the issue whether a party should be entitled to sue for injunctive relief. Nor does the position taken in Illinois Brick that effective enforcement of the antitrust laws requires that only direct purchasers be permitted to sue for treble damages have validity in the context of § 16.

To achieve standing under § 16 then, all NaBanco must prove is that it is threatened with loss or injury cognizable in equity proximately resulting from the anti-trust violation. Mid-West Paper Products at 591. Some courts have interpreted this proximate cause standard as one "no more rigorous than the general rule of standing." Tugboat, Inc. v. Mobil Towing Co., 534 F.2d 1172, 1174 (5th Cir. 1976). Even if this were not the case, and the line drawing for proximate causation was necessarily to be drawn somewhere between the general rule of standing and standing for treble damage actions, that need not concern us here. This is not a case involving facts making it necessary to "measure the outer range of standing to sue under § 16. Mid-West Paper Products at 593. NaBanco is not like "potential plaintiffs . . . who may only be remotely affected by the ripples caused by the conspirators' tampering with the supply and demand curve." Id. NaBanco contends ". . . that under all circumstances prevalent in the real economic world, money is passing from [its] hands into the pockets of [VISA member banks] as a result of [IRF]. . . ." Mid-West Paper Products at 593. In order to participate in the system, NaBanco must pay IRF or it cannot access issuer banks to complete payment transactions. These facts seem sufficient in light of the policies behind § 16 to give NaBanco standing.

PER SE VERSUS RULE OF REASON

Section 1 of the Sherman Act reads in pertinent part as follows:

§ 1 Trusts, etc. in restraint of trade illegal; penalty

Every 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, is declared to be illegal. 15 U.S.C. § 1

In enacting Section 1, Congress condemned in broad language concerted conduct which restrains trade. With such guidance as might be found in the common law cases, it left to the courts the scope and meaning of this mandate. As has been the case with Section 16 of the Clayton Act, the very breadth of the condemnation unavoidably gives rise to ambiguity, leading courts long ago to conclude that Congress could not have literally meant to ban every contract which in any sense restrains trade. As Justice Brandeis pointed out in Board of Trade of the City of Chicago v. United States, 246 U.S. 231, 38 S.Ct. 242, 62 L.Ed. 683 (1918):

[T]he legality of an agreement or regulation cannot be determined by so simple a test, as whether it restrains competition. Every agreement concerning trade, every regulation of trade, restrains. To bind, to restrain, is of their very essence. The true test of legality is whether the restraint imposed is such as merely regulates and perhaps thereby promotes competition or whether it is such as may suppress or even destroy competition. Id. at 238, 38 S.Ct. at 244.

In the years since the adoption of the Sherman Act in 1890, a number of competing views have been put forward as to its meaning. Stated simply, however, the governing law has been that the Sherman Act bans all concerted arrangements which are adopted for the purpose of reducing competition, or which, regardless of purpose, have a significant tendency to reduce competition. However, arrangements which are adopted for and tend to achieve other legitimate purposes do not fall within the condemnation of the Act merely because of some incidental and inconsequential restraining effect on competition.

This general position has been reached, not directly, but through two subsidiary rules, the per se doctrine and the rule of reason, which taken together tend to condemn arrangements which have the purpose or effect of significantly restraining competition and to validate those which do not. The per se rule forecloses any analysis of the purpose of a restraint or the nature, extent and degree of its market effect, thus avoiding "the necessity for an incredibly complicated and prolonged economic investigation into the entire history of the industry involved, as well as related industries, in an effort to determine at large whether a particular restraint has been unreasonable." Northern Pacific Ry. v. United States, 356 U.S. 1, 5, 78 S.Ct. 514, 518, 2 L.Ed.2d 545 (1958). Thus the per se rule is generally limited to those categories of restraints for which no elaborate study of an industry is needed to establish that their nature and effect is "plainly" or "manifestly" anticompetitive. Broadcast Music, Inc. v. Columbia Broadcasting Sys., Inc., 441 U.S. 1, 8, 99 S.Ct. 1551, 1556, 60 L.Ed.2d 1 (1979); Catalano, Inc. v. Target Sales, Inc., 446 U.S. 643, 646, 100 S.Ct. 1925, 1927, 64 L.Ed.2d 580 (1980); National Society of Professional Engineers v. United States, 435 U.S. 679, 692, 98 S.Ct. 1355, 1365, 55 L.Ed.2d 637 (1978); Continental T.V., Inc. v. GTE Sylvania, Inc., 433 U.S. 36, 50, 97 S.Ct. 2549, 2557, 53 L.Ed.2d 568 (1977). To be manifestly anticompetitive, agreements or practices must have a "pernicious effect on competition and lack . . . any redeeming virtue." Northern Pac. Ry. v. United States, 356 U.S. 1, 5, 78 S.Ct. 514, 518, 2 L.Ed.2d 545 (quoted in Sylvania, 433 U.S. at 58, 97 S.Ct. at 2561).

The classic definition of the rule of reason was enunciated in Standard Oil, supra, if asserted in its embryonic form in United States v. Trans-Missouri Freight Ass'n, 166 U.S. 290, 17 S.Ct. 540, 41 L.Ed. 1007 (1897). As construed in Standard Oil, Section 1 of the Sherman Act was an "all-embracing enumeration to make sure that no form of contract or combination" by which undue restraint was achieved "could escape condemnation." 221 U.S. 1, 58-60, 31 S.Ct. 502, 515-516, 55 L.Ed. 619. The function of "reason" in the analysis is to discriminate between restraints which are "undue" and those which are not. In Professional Engineers the Supreme Court indicated that, contrary to its name and the suggestion of Mr. Justice White in United States v. Trans-Missouri Freight Ass'n, 166 U.S. 290, 17 S.Ct. 540, 41 L.Ed. 1007 (1897), the rule of reason "does not open the field of inquiry to any argument in favor of a challenged restraint that may fall within the realm of reason." Professional Engineers, 435 U.S. at 688, 98 S.Ct. at 1363. Instead, the basic inquiry is limited to whether the restraint in question "is one that promotes competition or one that suppresses competition." Id. at 691, 98 S.Ct. at 1365. The opinion stressed that such an inquiry "is confined to a consideration of impact on competitive conditions" and that the function of the court is "to form a judgment about the competitive significance of the restraint." Id. at 690, 692, 98 S.Ct. at 1364-1365. Because the inquiry is limited to determining the "market impact" on competition, defenses "based on the assumption that competition itself is unreasonable" or upon considerations unrelated to the effect of the restraint on competition are irrelevant. Id. at 696, 98 S.Ct. at 1368.

NaBanco's Per Se Argument

At least twice before¹⁴ this court has considered and rejected NaBanco's argument that the action of the VISA Board of Director's in establishing and setting a fixed fee for interchange transactions is a "per se" violation of 15 U.S.C. § 1. Despite this court's previous rulings on the matter, NaBanco has pressed the argument that the relevant market in this case is the market consisting of the five major general purpose credit cards. NaBanco argues further that the proper analytical framework within which to analyze whether VISA's activities have any competitive impact is one which consists of three separate and distinct "markets": one for card-issuing services; the second for merchant-servicing banks; the third for "IRF-like" interchange services. The card-issuing market consists of banks and other institutions which offer consumers various credit card services having a multiplicity of characteristics. Included among these characteristics are credit options and other money management capabilities, varying degrees of card recognition and/or acceptance rates, different convenience qualities, etc. The traits associated with the market for merchant banks are two-fold, including what NaBanco calls a "factoring service" and an "account carrying" or "clearinghouse" service. By "factoring service", NaBanco means the initial contract by which a merchant bank or a "NaBanco-like" institution agrees to "purchase" merchant receivable items created by consumer bank credit card use; NaBanco uses the second two terms to identify the mechanics of processing claims and paying accounts.

14 NaBanco has raised the issue of per se validity several times: (1) two motions for preliminary injunction; (2) two motions for partial summary judgment; (3) trial brief; and (4) at trial. By order dated September 15, 1980, this Court denied NaBanco's various summary judgment motions.

Each of these separate and distinct markets are national in scope according to NaBanco [Bull TR. 640-41 5/12/82], the competitors in the card-issuing market allegedly including not only VISA and Mastercard but also the traditional T E cards (American Express, Carte Blanche, Diner's Club), certain debit cards, and bank ATM cards. NaBanco's description of the competitive actors in the merchant-servicing market includes those banks which act as merchant-servicing banks, as well as other entities such as NaBanco which provide the same kinds of general services to merchants, with varying degrees of emphasis, facilities, and expertise.

As mentioned earlier, NaBanco has taken the position that merchant banks are selling receivables to issuer banks, and the issuer banks, through the control of the VISA Board of Directors, are directly fixing the discounted price of those sales by means of the Base II exchange fee imposed on those members which elect to use Base II.

The Facial Unreasonableness Test

Alternatively, NaBanco argues that should the court determine the IRF is not per se illegal the test the court should apply in determining IRF's legality under the antitrust laws is the "facial unreasonableness test." This test, first enunciated in United States v. Realty Multi-List, Inc., 629 F.2d 1351 (5th Cir. 1980) is actually a two-part test which, unlike either the per se or rule of reason tests, shifts the burden of proof from the plaintiff to the defendant; i.e., once the plaintiff has sustained its burden of proving that a defendant's activities are a "significant restriction" affecting competition and that the defendant has significant market power, the burden then shifts to the defendant to show that the particular practice in question is "justified by the competitive needs" of the defendant and also both "reasonably necessary to the accomplishment of the [defendant's] legitimate goals" as well as narrowly tailored to that end. Id. at 1370.

VISA's Answer to NaBanco's Per Se Argument

NaBanco buttresses its per se argument by pointing to a long line of cases, beginning with United States v. Trenton Potteries Co., 273 U.S. 392, 47 S.Ct. 377, 71 L.Ed. 700 (1927), in which price fixing, as a concerted activity among competitors, is treated as a "naked" price constraint. ( See also Trans-Missouri Freight Ass'n, supra, 17 S.Ct. 568.) This is true whether the price "fixed" is the ultimate price of a good or service or simply a component of the ultimate price. Catalano, Inc. v. Target Sales, Inc., 446 U.S. 643, 645, 100 S.Ct. 1925, 1926, 64 L.Ed.2d 580 (1980). Further, a price-fixing agreement that establishes maximum prices is considered to be as illegal as one establishing minimum prices. Arizona v. Maricopa County Medical Society, 457 U.S. 332, 102 S.Ct. 2466, 73 L.Ed.2d 48 (1982); see also Kiefer-Stewart Co. v. Joseph E. Seagram Sons, 340 U.S. 211, 71 S.Ct. 259, 95 L.Ed. 219 (1951).

This jurisprudence of per se price fixing validity exists because, for the most part, it has been difficult for courts to conceive of any example in which consequent enhancement of competition arises in the context of price fixing or, even if such enhancement occurred, that it could be so great as to render the restraint insignificant in accordance with the dictates of the rule of reason. Lawrence Sullivan Antitrust, W. Publishing Co., St. Paul, Minn. (1977) p. 196. Almost all the cases upon which NaBanco relies, however, display three unifying characteristics: (1) the practices examined were related to the market in such a way that if they were to have any effect at all it would be on price formation, (2) the practices lacked any significant degree of integration of functions among the competitors, and (3) the arrangements were not ones which helped to make or improve a market by facilitating trading, exchanging information, standardizing product or the like. Since price formation was affected, if there was any consequence at all, it is entirely fair and functionally accurate for courts to treat arrangements like these as naked restraints.

VISA, however, has successfully opposed NaBanco's arguments in prior proceedings in this case on the grounds that NaBanco's per se price fixing theory is simply inapposite to the economically complex practices at issue in this case. While acknowledging high levels of vigorous competition at both the card-issuing and merchant-servicing ends of its bank credit card system, VISA challenges NaBanco's "tri-market theory." First it contends that "issuer" banks and "merchant" banks are not two separately competing sets of VISA members. The terms are useful only as indicators of a member's role in any single interchange transaction. Nor is it accurate, says VISA, to speak of issuing and merchant banks as members with adverse interests. VISA further contends that it is meaningless to speak of a "product" offered at either the card-issuing or merchant-servicing end separated from its counterpart at the other. Bank credit cards, VISA argues, are useful only to the extent they provide the cardholder with purchasing options. All the characteristics which NaBanco asserts are so important to credit card differentiation, therefore, are only important to the extent they enhance any particular credit cardholder's purchasing options by increasing the numbers and kinds of participating merchants. Concomitantly, the numbers and kinds of participating merchants depends to a great extent on the numbers and kinds of consumer cardholders, as well as characteristics of individual merchant-servicing banks. In fact, (VISA asserts) the particular demand for merchant banks suggested by NaBanco would not exist but for the demand by consumers for purchasing options. Viewed from the perspective of the participants, for the VISA service to exist at all there must be members willing to issue cards and members willing to sign merchants.

Thus, argues VISA further, the relevant market in this case is the market consisting of all nationwide payment services used in retail sales, including VISA, Mastercard, T E cards, merchants' proprietary cards, merchants' open book credit, cash, travelers cheques, ATM cards, personal checks and check guarantee cards, rather than the market for the Interchange Fee which VISA contends is either non-existent or at most an example of market failure at its most extreme since only one buyer (the cardholder's issuer bank) reasonably exists for each piece of transactional paper. In making its argument then, VISA likens itself to a joint venture. The setting of IRF, far from being a naked restraint of trade as NaBanco asserts, rather becomes a device for sharing costs within the VISA system. This device, which, VISA argues is necessary in order to produce the VISA product, thereby enhances "interbrand" competition which VISA argues "is the primary concern of the antitrust law." Sylvania, 433 U.S. 36, 52 n. 19, 97 S.Ct. 2549, 2558 n. 197, 53 L.Ed.2d 568. Even if IRF were not necessary to market the VISA product, VISA argues it should be analyzed under the rule of reason because it is an agreement internal to a joint venture which yields efficiencies beneficial to competition, that its members, acting alone, could not offer, and which allows the venture to offer a product which is different from, and greater than, the sum of the individual products of its members. Broadcast Music Inc. v. C.B.S., 441 U.S. 1, 99 S.Ct. 1551, 60 L.Ed.2d 1 (1979); United States v. Multi-Realty (supra).

Rejection of the Per Se Rule

Previous discussions have consistently recognized the pro-competitive or efficiency creating potential of joint ventures and other cooperative organizations. While joint ventures are not immune from antitrust scrutiny, the anti-trust laws allow restraints necessary to accomplish pro-competitive objectives of joint ventures and similar organizations. See, e.g., United States v. Realty-Multi List, Inc., 629 F.2d 1351, 1368 (5th Cir. 1980); Central Iowa Power Federal Energy Regulatory Commission, 606 F.2d 1156, 1163 (D.C. Cir. 1979). When these restraints, sometimes called "ancillary restraints" are addressed, per se prohibitions are not applied because it cannot be said that in the aggregate, the gains from imposition of the per se rule will far outweigh the losses. Board of Regents of the University of Oklahoma v. National Collegiate Athletic Ass'n, 707 F.2d 1147, 1153 n. 6 (10th Cir. 1983), cert. granted, ___ U.S. ___, 104 S.Ct. 272, 78 L.Ed.2d 253 (1983). As the Supreme Court recognized in Broadcast Music Inc. v. C.B.S., 441 U.S. 1, 23, 99 S.Ct. 1551, 1564, 60 L.Ed.2d 1 (1979), joint ventures "are not usually unlawful, at least not as price fixing schemes, where the agreement on price is necessary to market the product at all." Accord, Arizona v. Maricopa County Medical Society, 457 U.S. 332, 102 S.Ct. 2466, 73 L.Ed.2d 48 (1982). Accordingly, such agreements are analyzed under the rule of reason so that courts can weigh their pro-competitive effects. See, e.g., North American Soccer League v. Nat'l Football League, 670 F.2d 1249, 1259 (2d Cir. 1982); Los Angeles Memorial Coliseum Commission v. National Football League, 468 F. Supp. 154 (C.D.Cal. 1979), reversed on other grounds 634 F.2d 1197 (9th Cir. 1980); Worthen Bank Trust Co. v. National Bankamericard, Inc., 485 F.2d 119 (8th Cir. 1973) cert. denied, 415 U.S. 918, 94 S.Ct. 1417, 39 L.Ed.2d 473 (1974); Smith v. Pro Football, Inc., 593 F.2d 1173 (D.C. Cir. 1978); National Bank of Canada v. Interbank Card Ass'n, 507 F. Supp. 1113, 1122 (S.D.N.Y. 1980) aff'd, 666 F.2d 6 (2d Cir. 1981); Broadcast Music, Inc. v. Moor-Law, Inc., 527 F. Supp. 758, 765 (D.Del. 1981) aff'd 691 F.2d 490 (3d Cir. 1982); Palmer v. Roosevelt Lake Log Owners Ass'n, 551 F. Supp. 486 (E.D.Wash. 1982).

Given this background of jurisprudence, the Court concludes that IRF should be analyzed under the rule of reason because it is an agreement on the terms of interchange necessary for VISA to market its product and be an effective competitor. Broadcast Music Inc. v. C.B.S., 441 U.S. 1, 23, 99 S.Ct. 1551, 1564, 60 L.Ed.2d 1 (1979). The Broadcast Music case teaches that in separating those price-fixing arrangements to which the per se rule is to be applied from those to be examined under a rule of reason analysis, the Court is to assess whether "facially" the practice "always or almost always" tends to restrict output, or whether "facially" it is likely to contribute to the creation of "economic efficiency." Other than the testimony of one of NaBanco's experts, Louis A. Guth, plaintiff has failed to produce any evidence suggesting the existence of separate and independent card-issuing and merchant-servicing markets. All the evidence presented at trial has pointed to a complex set of inter-relationships between cardholder and merchant demands for the VISA service suggestive of a joint venture — even if VISA cannot be properly characterized as one in the strictest sense of the word.

In VISA's case, profits and losses are not specifically shared among the various VISA members, nor is there any commingling of management functions. Furthermore, to the extent possible, each member engages as an independent unit in economic competition with other VISA members with respect to various aspects of their common venture. For example, merchant banks compete with issuer banks to sign up cardholders and, to the extent that certain banks perform both functions, VISA members compete with each and every one of their fellow affiliates.

Nevertheless, while considerations such as these may be important to courts considering joint venture labels, there is no requirement, for example, that a joint venture-like entity integrate or share costs in some prescribed fashion in order to survive per se scrutiny. See Louis, "Restraint Ancillary to Joint Ventures and Licensing Agreements: Do Early and Topco Logically Survive Sylvania and Broadcast Music?" 66 Va.L.Rev. 879, 896 n. 114 (1980); Los Angeles Memorial Coliseum Commission v. National Football League, 468 F. Supp. 154, 163 (C.D.Cal. 1979) (applies rule of reason even though "profits and losses are not shared.)"

The fact that VISA members have integrated to the extent of agreeing on the terms of interchange, but have not fully integrated and still compete for cardholders and merchants, is typical of pro-competitive joint ventures and serves to maximize VISA's competitive potential. The fact of the matter is that even NaBanco does not go so far as to challenge the whole plethora of ancillary regulations attendant to the VISA structure, including, for example, the obligation of all issuer banks to reimburse the merchant banks (and thus the "NaBancos" in the system) for processing the transactional paper generated by their cardholders' use of their VISA cards.¹⁵ It is, perhaps, appropriate to note parenthetically that so long as there are benefits to be parlayed into profits, NaBanco has elected to remain in VISA's beneficial ambit. Even if the Court were to conclude that VISA is not akin to a "joint venture;" we could not ignore NaBanco's reluctance to "throw the baby out with the bath water."

15 As a further example, merchants in the VISA system are required to honor proper customer requests to use the card, and, of course, to abide by other system regulations.

In any event, IRF serves to share substantial costs and risks between VISA members, and interchange transactions are an integration because in such transactions two different members combine their activities in order to offer a service to a cardholder and a merchant.

Thus, VISA is a joint venture in that terms most meaningful sense, i.e., whether or not the composite entities compete with one another in any meaningful sense in the marketplace under examination. Unwarranted emphasis upon formalistic aspects of the relationship of VISA and its member institutions ignores the subtle but more significant interdependency of the members and their indivisibility with VISA. Assuming that VISA cardholders want unplanned and rapid access to merchants anywhere, regardless of whether their own bank signed a particular merchant, and that VISA merchants want unplanned and rapid access to each cardholder who enters their shop, regardless of whether the merchant's bank signed the cardholder, then some before-the-fact agreements must be made. The principal purpose of these agreements with member banks does not appear to be to improperly fix prices as NaBanco asserts but rather to provide a service which each member bank could not alone provide, namely, universal payments service which ensures that a VISA card will be honored by any merchant regardless of which bank issued it so long as that merchant displays the VISA logo on its door. Given past and current interstate banking regulations which once flatly proscribed and continues to inhibit the growth of single bank entities which cross state lines, it is hard to conclude otherwise. Even assuming that each VISA member could afford to issue its own card, there would be no assurances of universality since issuer banks and merchant banks might still differ as to exchange rates. Cardholders would either have to carry a suitcase full of cards representing the different banks which have contracts with particular merchants or suffer the disappointment of being unable to transact certain purchases with merchants whose merchant banks refused for some reason to contract with the cardholders' bank; to say nothing of the inconvenience and dissatisfaction of the merchant who must check each card out with its merchant bank even if it may mean that he will lose a sale. Prohibiting IRF would thus undermine "interbrand" competition, "which is the primary concern of the antitrust law." Continental T.V. Inc. v. GTE Sylvania, Inc., 433 U.S. 36, 52 n. 19, 97 S.Ct. 2549, 2558 n. 19, 53 L.Ed.2d 568 (1977).

Even if IRF were not necessary to market a payment services product of the type in question, it should be analyzed under the rule of reason because it is an agreement internal to a joint venture which yields efficiencies beneficial to competition. VISA's members, acting alone, could not offer such efficiencies which indeed allow the venture to offer a product which is different from, and greater than, the sum of the individual products of its members. Broadcast Music, supra; Realty Multi List, Inc., supra. Broadcast Music and Maricopa do not require the court, as NaBanco contends, to make a subjective judgment as to whether a joint ventures product is "unique." Rather, they call for the court to compare the product of the venture to the individual offerings of the venture's members and determine whether the venture offers pro-competitive abilities which its members acting alone could not offer. Broadcast Music, 441 U.S. at 21-23, 99 S.Ct. at 1563-1564; United States v. Columbia Pictures Industries, 507 F. Supp. 412, 430 (S.D.N.Y. 1980). Thus the agreement to market compositions in Broadcast Music survived the per se test even though the individual composers could market their compositions on their own. 441 U.S. at 23-24, 99 S.Ct. at 1564-1565. The consolidation of thousands of what would otherwise be balkanized local cards into one internationally accepted card makes the VISA system a viable and competitive whole which is "truly greater than the sum of its parts." Broadcast Music, 441 U.S. at 21-22, 99 S.Ct. at 1563. A similar agreement (governing a cost sharing arrangement reimbursing one airline which provides credit card services to a passenger who flies on a different airline) accordingly was analyzed under the rule of reason and ultimately upheld by the Civil Aeronautics Board in In Re UATP — 1976 Agreements, 85-11 C.A.B. 2481 (1980).

Further, IRF is properly analyzed under the rule of reason because it is not mandatory. Acting on behalf of its principals, NaBanco is and always has been free to negotiate different terms of interchange, (not using Base II) and some VISA issuers have been willing to make alternate arrangements. A practice is not unlawful per se where as in this case, there is no legal, practical or conspiratorial impediment to making alternate arrangements. Broadcast Music, Inc. v. C.B.S., 441 U.S. 1, 24, 99 S.Ct. 1551, 1564, 60 L.Ed.2d 1 (1977); In Re ATP-1976 Agreements, 85-11 C.A.B. 2481 (1980).

The Supreme Court's decision in Arizona v. Maricopa County Medical Society, 457 U.S. 332, 102 S.Ct. 2466, 73 L.Ed.2d 48 (1982) confirms this analysis. The agreement in Maricopa was held to be illegal per se because, unlike IRF, it fixed the prices charged to third parties by individual doctors for their own services. 102 S.Ct. at 2480. Moreover, and also in contrast with the IRF arrangement, the Maricopa court granted summary judgment only after reporting — six separate times — that nothing in the record even arguably supported the defendants' assertion that agreement on a fee was necessary to market the product. 102 S.Ct. at 2477, 2477-78 n. 26, 2478, 2479 n. 33. At the same time, the court in Maricopa confirmed the approach set forth in BMI and emphasized that agreements on price which are inherent in the structure and successful operation of a bona fide joint venture can be "perfectly proper." 102 S.Ct. at 2480.

In Worthen Bank Trust Co. v. National BankAmericard, Inc., 485 F.2d 119 (8th Cir. 1973), cert. denied, 415 U.S. 918, 94 S.Ct. 1417, 39 L.Ed.2d 473 (1974), the Court refused to apply the per se rule because of lack of experience with the economics of a fledgling industry. In discussing a different by-law of VISA's predecessor NBI, the Court said:

The bank card industry is a relatively new one and, with over eight billion dollars in annual business, a very large and important segment of our economy. Considering the importance of the industry and the lack of definitive information relating to competition therein, it would be a mistake to determine this case of first impression on a per se basis. This is especially true where it may result in the dilution of competition or the elimination of one of the two competitors where this is not an aggregate of illegal restraints such as price fixing or exclusive territorial allocations and where the initial purpose in forming the joint venture, to produce a national credit card, is obviously not illegal. 485 F.2d 119, 129-130.

Courts have been reluctant to apply the per se rule in cases involving whole markets which do not readily fit into the antitrust context. Prime examples of this type of situation are those cases involving challenges to the rules of sports associations. There is no compelling reason why many of the characteristics the courts have found to be fundamental to those organizations should not be found to be equally important to the functioning of the VISA system. Like the sports association, the VISA system is an example of an association where cooperation is as much a necessary part of the "game" as competition. Just as there is a need to establish some "rules of the game," affecting such matters as the size of the field, the number of players on a team, and the extent to which physical violence is to be encouraged or proscribed,¹⁶ in order for there to be sports contests at all, so too there must be "rules" to enable the efficient coordination of the otherwise disparate operations of VISA members. Furthermore, just as the establishment of game rules by professional organizations does not mean that individual member clubs do not compete for players and fans or between each other on the playing field, the creation of VISA by-laws has not precluded competition among individual VISA members. In this way, cooperation and competition have made for the best of all possible worlds, a world which would probably not exist but for this particular combination. In NCAA v. Board of Regents (supra), the Supreme Court, while finding the television plan in question violative of the Sherman Act (because of its clear restraints), refused to apply the per se rule in a setting which (absent similar restraints) lends itself to comparison with the functional needs of the VISA system.

16 NCAA v. Board of Regents of Univ. of Okla., ___ U.S. ___, 104 S.Ct. 2948, 82 L.Ed.2d 70 (1984).

In choosing to apply the rule of reason analysis to this case, the court also rejects the applicability of the so-called "facial unreasonableness" theory set forth in Broadcast Music and applied in Realty Multi List. First, it should be noted that Realty Multi List is a group boycott case, and not a case of alleged horizontal price fixing as is this case. In any event, if we assume that Realty Multi List applies to the facts of this case, it presents only a poor variation of the rule of reason test appropriate here in light of the "joint venture" nature of the VISA system. The Realty Multi List court suggests as much when it says: "This argument is, of course, a version of the rule of reason." Id. at 1369.

The facial unreasonableness theory can properly be invoked only where the practice at issue constitutes a "significantly restrictive rule of a combination or trade association with significant market power, which lacks competitive justification or whose reach clearly exceeds the combination's legitimate needs." Id. at 1370. Membership rules, such as those at issue in Realty Multi List, present viable candidates for invalidation as "significantly restrictive" under this test since they presumptively preclude competitors from participating in an essential market. But cost sharing arrangements among partners or venturers are not patently "restrictive" for the reasons addressed by this court in rejecting per se treatment.

Moreover, even if IRF appeared to be sufficiently "restrictive" to invoke the facial unreasonableness theory, a rule of reason analysis could be avoided only if plaintiff demonstrated that VISA possessed "significant market power" since it is, of course, unlikely that serious anticompetitive effect will result from exclusion from membership [or restrictive practices]" if such power is lacking. Id. at 1372 at n. 40. In this respect, the contrast between the VISA situation and the multiple listing service in Realty Multi-List, Inc., is striking. Multiple listing services involving local markets and exclusionary membership rules almost invariably possess the requisite degree of market preclusion. Indeed, in Realty Multi List, the issue was hardly disputed, and there the issue was reserved for trial. Id. at 1372 n. 39. Here consideration of the relevant market reveals that VISA has no significant market power, notwithstanding plaintiff's suggestion that the court need not undertake a proper inquiry into demand cross-elasticities.

Furthermore, if we assume NaBanco could cross the "market power" threshold, the facts adduced at trial demonstrate sufficient pro-competitive justification to require that a full rule of reason inquiry be undertaken. The need for a cost-sharing mechanism within the VISA joint venture is itself sufficient justification as Broadcast Music indicates. This Court has already recognized, for example, that the question whether IRF could be replaced by "competition" at least presents an issue of fact, if we indulge the assumption that a negative answer to that question is not obvious.

In sum, care must be taken to avoid "easy labels," whether appearing in Latin (per se rules) or convenient short-hands ("facial unreasonableness"). The critical inquiry under the Sherman Act must be whether a challenged practice is anti-competitive in its economic effects. The explanation this court includes below of why IRF is highly competitive and thus lawful under the rule of reason confirms the conclusion that condemning the fee out of hand, as either per se unlawful or "facially unreasonable," would be a mistake.

RULE OF REASON ANALYSIS

Having, hopefully, illuminated the passage to a consideration of the reasonableness of VISA's fee plan, the court now turns to an analysis of the competitive effects of IRF as an agreement ancillary to a type of "joint venture." Under the rule of reason test it is plaintiff's burden to establish that the practice complained of has an adverse impact on competition. Graphic Products Distributors, Inc. v. Itek Corp., 717 F.2d 1560, 1573 (11th Cir. 1983). It should be stressed, however, that in making this a determination, the relevant question is not whether the challenged practice is the most competitive device that can be imagined, or the "least restrictive," but simply whether it is reasonable; i.e., not "unduly" restrictive of competition. GTE Sylvania, Inc., supra, 433 U.S. 49, 97 S.Ct. 2557; National Auto Brokers Corp. v. General Motors Corp., 572 F.2d 953, 960 (2d Cir. 1978), cert. denied, 439 U.S. 1072, 99 S.Ct. 844, 59 L.Ed.2d 38 (1979). So long as a practice is "fairly necessary" to achieve a legitimate purpose, it is not unlawful under the rule of reason. Broadcast Music, Inc. v. Moor-Law Inc., 527 F. Supp. 758, 769 (D.Del. 1981), quoting American Motors Inns v. Holiday Inns, Inc., 521 F.2d 1230, 1248-49 (3d Cir. 1975). In applying this standard, the competitive benefits of the challenged practice are weighed against its alleged harmful effects and a violation of the Sherman Act will be found only where the balance tips decidedly in favor of the latter. Columbia Broadcasting System, Inc. v. American Soc. of Composers, Authors and Producers, 620 F.2d 930, 934 (2d Cir. 1980), cert. denied, 450 U.S. 970, 101 S.Ct. 1491, 67 L.Ed.2d 621 (1981); National Bank of Canada v. Interbank Card Ass'n, 507 F. Supp. 1113, 1122 (S.D.N.Y. 1980).

As suggested above, plaintiff must shoulder the burden, in "rule of reason" presentation to demonstrate the balance tipping in plaintiff's favor; that is, demonstrating that alleged anticompetitive effects of the setting of an IRF decidedly outweigh the pro-competitive benefits of the practice. See, e.g., Red Diamond Supply Inc. v. Liquid Carbonic Corp., 637 F.2d 1001, 1005, 1007 (5th Cir.; cert. denied, 454 U.S. 827, 102 S.Ct. 119, 70 L.Ed.2d 102 (1981); Continental TV, Inc. v. GTE Sylvania Inc., 433 U.S. 36, 49, 97 S.Ct. 2549, 2557, 53 L.Ed.2d 568; Columbia Broadcasting System, Inc. v. A.S.C.A.P., 620 F.2d 930, 934 (2nd Cir., 1980), cert. denied, 450 U.S. 970, 101 S.Ct. 1491, 67 L.Ed.2d 621 (1981). Moreover, in meeting its burden, NaBanco must prove far more than that the practice it challenges has had an effect on its business. See, e.g., Oreck Corp. v. Whirlpool Corp., 579 F.2d 126, 133-34 (2nd Cir. 1978); Balogh's of Coral Gables, Inc. v. Getz, 510 F. Supp. 741, 748 (S.D.Fla. 1981); JBL Enterprises, Inc. v. Jhirmack Enterprises, Inc., 509 F. Supp. 357, 366 (N.D.Cal. 1981). The existence or non-existence of market power in a relevant market is a critical consideration. Spectrofuge Corp. v. Beckman Instruments, Inc., 575 F.2d 256, 277 (5th Cir. 1978) cert. denied, 440 U.S. 939, 99 S.Ct. 1289, 59 L.Ed.2d 499 (1979).

A relevant market, or "area of competition," consists of two elements — products and geography. Brown Shoe Co. v. United States, 370 U.S. 294, 324, 82 S.Ct. 1502, 1523, 8 L.Ed.2d 510 (1962). As for the former element, the Supreme Court has held that it is defined by reference to "the reasonable interchangeability of use or the cross-elasticity of demand between the product itself and the substitutes for it." Id. at 325, 82 S.Ct. at 1523-1524. See also United States v. E.I. DuPont DeNemours Co., 351 U.S. 377, 394-96, 76 S.Ct. 994, 1006-1008, 100 L.Ed. 1264 (1956).

There is a defect in NaBanco's argument which can only be successfully dealt with by accepting VISA's argument that the sale of receivables makes no sense except in its connection with a payment service system. It is, perhaps, appropriate to point out parenthetically that NaBanco would not exist but for the payment systems. The court finds that the relevant market in this case consists of the market for payment systems as VISA contends. Not only did the testimony at trial suggest that VISA developed as the result of Bank of America's determination to provide and promote a payment system of nationwide proportions to compete with other payment systems then in existence, the evidence also indicated considerable success. The cross-elasticity of both demand for and supply of VISA and other payment devices is quite high. Both cardholders and merchants who testified at trial considered the VISA services equivalent or sufficiently close to a variety of payment systems used in retail sales, including other credit cards, travelers cheques, cash, ATM cards, personal checks, and check guarantee cards. Edwards. TR. 110-124 (6/2/83); Baxter. TR. 24-35 (2/13/83); Phillips. TR. 144-151 (6/6/83); Emerson. TR. 20 (1/13/84). While each of these different payment service devices was not considered to be a close substitute for a VISA card for purchases of every possible product at every possible price, all payment services taken together were sufficient to provide, at the least, several close substitutes for a VISA card in any possible context. Cash, for example, might be a good substitute for face-to-face transactions involving small dollar amounts, while checks would be better for larger transactions involving long-distance exchanges. Phillips. TR. 144-151 (6/6/83); Baxter. TR. 24-35 (2/13/83).

Testimony by VISA experts also confirmed substitutability between VISA and other, technologically innovative, payment devices including but not limited to regional ATM cards (Cox. TR. 128-29 (1/13/83)), debit cards (Duffy. TR. 2870-72 (9/23/82)), as well as retail proprietary cards (Duffy. TR. 2667 (9/22/83)) and new multi-purpose cards like the Sears/Merrill Lynch cards now available. Cox. TR. 86-89, 128-29 (1/13/83).

To the extent that VISA also extends credit to consumers it can also be argued that it competes in the credit extension market along with commercial banks, savings and loan associations, consumer finance companies, credit unions, and certain competing credit cards such as Mastercard, department store proprietary cards, and oil company cards. NaBanco, however, has never alleged nor contended that VISA has any market power in this market, nor has it ever alleged that either VISA or its IRF requirement presented any anticompetitive effects therein. Therefore, the court does not include these credit extending devices within the relevant market except to the extent these devices also serve as payment system mechanisms (e.g. Mastercard, proprietary cards, oil company cards, etc.).

As for the geographic market which is relevant in the context of this case, there appears to be no issue: VISA competes nationally, indeed internationally, with competitive payment systems. Since NaBanco's business is similarly conducted on a nationwide scale, no relevant geographic sub-market is even arguably presented. NaBanco agrees with this. See NTB at 25.

Having already defined the relevant market as the nationwide market for payment systems, the court must still consider whether or not VISA's share of that market is such as to indicate that it wields power in that market. The Court concludes that it is not. VISA's share of the payment services market is probably less than five percent according to testimony heard at trial. Edwards 118-20 (6/3/83); Cox 139-40 (1/13/83). Exhibits 2089-2904. This share is far too small to allow VISA to have any legally pertinent anticompetitive effects. Cox TR. 139-40 (1/13/84); Phillips TR. 156-58 (6/6/83); Guth TR. 3-4 (1/18/84).

Furthermore, in the opinion of NaBanco's own expert, Dr. Cox, the new regional ATM debit card networks will one day force VISA out of business unless VISA promptly reduces IRF. Cox TR. 41-42 (1/13/83). The existence and strength of this type of competition, implying as it does ease of entry into the market, together with the absence of abnormally high profits earned from VISA operations, (as testified to at trial), seems to preclude any compelling argument that VISA has market power. If VISA had any choice in the matter it would certainly seem appropriate for it to reduce TRF in light of the evidence presented.

Building on the foregoing comments, the Court makes and enters these additional and somewhat more specific findings of fact:

A. VISA lacks power in the relevant market. In the absence of such market power, VISA lacks the ability to impose any restraint detrimental to competition.

1. The relevant market in this case is the nationwide market for payment systems.

(a) Both parties agree that the geographic market is nationwide.

(b) The relevant product market in this case is a market consisting of VISA and all payment services used in retail sales. This market includes VISA, MasterCard, T E cards, merchants' proprietary cards, merchants' open book credit, cash, travelers cheques, ATM cards, personal checks and check guarantee cards. None is a perfect substitute for a VISA card, but they are all quite good substitutes for each other and should be considered part of the same market. An examination of cross-elasticities of supply and demand demonstrates that the other payment services above are sufficiently close substitutes for the VISA card that they must be considered part of the same market.

B. VISA also competes in a market for the extension of credit to consumers. But NaBanco did not allege or contend that VISA had market power in the credit extension market, nor did it allege or contend that VISA or IRF had any anticompetitive effects in that market. That market includes other sources of credit available from commercial banks, savings and loans, consumer finance companies, credit unions and competing cards such as MasterCard, department store proprietary cards and oil company cards. In the market so defined, VISA's share is less than 10% and VISA does not possess sufficient market power to effect an anticompetitive restraint.

C. There is no market for the "sale" of cardholder transactions ("cardholder paper") from merchant banks to issuing banks, and it would be meaningless to use such a market for purposes of analysis in this case. Only the member which issued a card has any interest in acquiring from merchant banks transactions effected by that card.

D. VISA has a very insignificant share and no real power, in the payment service market.

(1) VISA's share of the payment service market is certainly less than 5%, and likely less than 3%. This share is far too small to allow VISA to have any anticompetitive effect.

(2) In the opinion of NaBanco's expert, Dr. Cox, the new regional ATM debit card networks will force VISA out of business if VISA does not reduce IRF soon. The existence and strength of this competition means that VISA does not have market power.

(3) Still further confirmation of the fact that VISA and its members lack significant market power is the absence of any abnormally high profits earned from VISA card operations.

(4) NaBanco's own arguments concede VISA's lack of market power. In its trial brief, it recognized that VISA and its members lack sufficient economic power over merchants to impose supra-competitive prices.

(5) Even if VISA currently had a large share of, and apparent power in, the relevant market, it still would not have market power because of the ease of entry into the market. There are no significant barriers to entry, particularly as a result of new technologies. The new ATM networks are an example of the easy entry into the marketplace. This ease of entry makes even more remote any possibility that VISA could impose an effective restraint.

(6) The fact that VISA and MasterCard set different IRF's during the early 1970's does not prove that either has market power. Rather, it confirms the existence of competition in the payment services market. On a transaction of the average size during that period, the difference was approximately only 10%.

The Interchange Fee is Pro-Competitive in Nature

IRF promotes the efficiency and competitiveness of VISA in two other related ways. The first, is of vital import to the day-to-day functioning of the system. As established, IRF eliminates the costly uncertainty and prohibitive time and expense of "price negotiations at the time of the exchange" between the thousands of VISA members. See Central Iowa Power Cooperative v. Federal Energy Regulatory Commission, 606 F.2d 1156, 1163 (D.C. Cir. 1979) (establishment of price is necessary to the functioning of the cooperative and "to ensure that costs and benefits of the power pool will be shared fairly and predictably.") In doing so, it guarantees the universal acceptability that is at the heart of the competitive success of the product.

The second way in which the fee promotes efficiency and competitiveness is in a more long-term, economic sense. Briefly, the theory is that the parties to a national bankcard system must distribute the costs of the system in relation to prospective benefits so as to encourage members to engage in the appropriate balance of card-issuing and merchant-servicing. Far from being presumptively anticompetitive, the economics of bankcard pricing suggest that a "transfer payment" such as the IRF is precisely what one would expect to find in a bankcard joint venture seeking to price competitively in the payments systems market.

In the VISA "joint venture," both issuer and merchant banks perform essential roles, and both perform necessary tasks for the benefit of each other, which must be performed if the basic payment service is to be offered. As stated earlier, the more cardholders in the system, the more attractive the system is to merchants. At the same time, the more merchants in the system, the more attractive the card is to cardholders. In short, there is a fundamental economic interdependence between card issuing and merchant signing in the system. Edwards. TR. 9-11 (6/3/83); Baxter (6/2/83) (dep. 2 at 18-21); Ashby. TR. 1621-23 (9/15/82); Guth TR. 8 (1/18/84).

This fundamental interdependence can be felt within the system in two ways. First, the more cardholders in the system, the more attractive the card is to merchants, and the more merchant discount revenue is generated for merchant banks. As more merchants enter the system, and more cardholders develop, more cardholder revenue is generated for issuers. Second, as to the expense of operation, costs that the issuer banks bear, such as fraud or float costs or the costs of issuing the plastic, benefit the merchant and merchant bank in that they facilitate and encourage use of the cards at merchant locations. Similarly, the processing costs borne by the merchant banks are valuable to issuer banks. Edwards. TR. 9-11 (6/3/83); Cox. TR. 4-10, 75-77 (1/14/83); Bull. TR. 971-74 (5/15/82); Ashby. TR. 1604-10 (9/15/82).

Viewed from the perspective of the members then, for the VISA service to exist at all there must be members willing to issue cards and members willing to sign merchants. This system of interrelationships can be represented diagramatically as in Appendix I.

As asserted at trial, however, at the same time that these interrelationships exist between the cardholder and merchant sides of the VISA business, there also exists a serious cost imbalance between the two sides. VISA has contended, and the evidence suggests, that the card-issuing side of the payments business bears the substantial majority of the costs of the business, up to 88% of total system costs. Edwards. TR. 13-15 (6/3/83); Cox. TR. 55-61 (1/14/83). Notably, the issuing side bears most of the float costs and most of the credit and fraud losses. Bull. TR. 907-17 (5/14/82); Edwards. TR. 13-15 (6/3/83). These costs are incurred on the issuing side by reason of VISA rules such as the rule which requires the issuer to pay the merchant bank for a transaction even if the cardholder never pays because such costs can be managed and controlled more efficiently by issuers. Edwards. TR. 13-15 (6/3/83).

There is nothing inherent in the system that suggests, much less guarantees, that the revenue stream that comes into the merchant bank from the merchant will correspond in its magnitude to the activity costs that must be incurred by the merchant bank. Similarly, there is nothing to suggest, much less guarantee, that the revenue stream that comes into the card issuing banks from the cardholder will be equal in magnitude to the activity costs that the issuer banks incur. IRF serves the function of re-distributing the costs of the VISA service more equitably between the merchant and card issuer sides, that is, it is a "transfer payment" of sorts. In the process, IRF "equillibrates" the supply and the demand for VISA services by bringing the costs of the system in line with the revenues for each participating VISA member bank regardless of the role it plays, either merchant or issuer, in the VISA system. The economics of this analysis is depicted in diagrammatical terms in Appendix II.

Re-distribution is perhaps not the only alternative theoretically available to VISA to achieve this equilibrium, but it is the most, if not the only, realistic alternative. For example, VISA might attempt to meet the cost imbalance problem by suggesting to issuer banks that, where they can legally, they further raise the price of issuing a card, or charge a fee for the cardholder's float period. The evidence produced at trial, however, suggests that cardholder demand for cards drops off dramatically when yearly fees are raised. In addition, many state banking regulations have prohibited and continue to prohibit charging cardholders for the float period.

NaBanco has offered another alternative; private negotiation of individual "IRFs" which, initially at least, presents an interesting alternative. Again, in theory, private contracts between issuer and merchant banks would enable these banks to set a charge for IRF which would reflect the actual, true "price" which an issuer would be willing to pay — a price which would hopefully reflect the true costs of interchange after subtracting the additional costs experienced by issuer banks alone but from which merchant banks benefit. Transaction costs in this type of system would no doubt be high and stultifying since each and every bank would have to somehow contract each and every other bank with which it might have some card business contact. Further, there is no evidence to suggest that any "market" at all exists at the level of (and between) issuer and merchant banks. If there is a "market," it is a monopsonistic one as there is normally only one issuer bank willing to "purchase" any particular merchant bank's receivable. While each and every issuer bank might not behave monopsonistically, the incentive is certainly there for opportunistic behavior by some issuer banks which would take advantage of the situation by trying to exact higher and higher fees from merchant banks. What VISA calls the "free rider" problem, would likely lead to the system's collapse as more and more issuer banks following this course of action led more and more merchant banks to drop out of the system.

VISA's own history provides the proof that is necessary to come to this conclusion. Prior to the IRF regulation, VISA members acted in their own individual self-interest, and contrary to the interests of the system as a whole, in their dealings relating to interchange fees. Banks acted in a similar fashion in charging interchange on checks in the late 19th and early 20th centuries. Baxter (6/2/83) (dep. 2 at 37-43). Even one of NaBanco's own witnesses, Ms. Ashby, could not deny that she might increase her bank's interchange fee if VISA no longer set IRF. Ashby. TR. 1501-03 (9/15/82).

Thus IRF enables VISA to exist as a viable competitor in the payment systems market. Even if it could be proved that a sub-market for receivables exists, IRF increases interbrand competition in line with the policies outlined in GTE and Sylvania.

IRF is Reasonably Cost-Related

The persuasive evidence at trial suggested that IRF developed as a cost-related mechanism. VISA has spent thousands of dollars and hundreds of hours choosing a methodology which attempts to "average" out those costs associated with interchange and card issuing. Further the methodology itself has undergone drastic changes from time to time.

Even from a purely theoretical standpoint, VISA has every incentive to set IRF at a level which establishes an equilibrium between the issuer and merchant sides of its operation. First, those members who are members of the VISA Board and therefore responsible for setting IRF invariably represent banks which both issue cards and sign up merchants.

Second, since the fee is only charged within the system, it is always a net "zero" to the VISA venture as a whole because dollar received by or on behalf of VISA member X is paid to on or behalf of member Y. Third, VISA members always have the option of altering the "mix" of their business between merchant and issuing activity. Finally, and perhaps most importantly, if consumer and merchant utilization must somehow be equalized reducing demand by increasing merchant costs will necessarily reduce issuer demand and consequently reduce revenue opportunities on the issuing side of business.

Nevertheless, NaBanco has attacked VISA's IRF methodology as inadequate and unscientific and, as a result, "overbroad." Specifically NaBanco has challenged the particular elements which go into the formula by which VISA calculates "average" issuer bank costs. For example, NaBanco has challenged the fact that neither merchant nor issuer banks revenues are included in analyzing the fee level set by VISA, and that merchant bank costs are also not considered. This, NaBanco has argued, necessarily puts the entire financial burden of providing the "convenience use" period, for example, on the merchant banks — unfairly so NaBanco has claimed because `revolving account' cardholders allegedly subsidize issuer banks for the costs associated with the float or convenience use period.

NaBanco has also attacked the "sampling" methodology used by VISA to determine which banks will be part of the larger, extended study by which VISA determines average costs. NaBanco has contended that VISA should use random sampling techniques to choose its sample banks, not the "purposeful" sampling engaged in by VISA.

The court finds that, contrary to NaBanco's arguments, the specific means used to calculate IRF — the theory used to set it and the implementation of that theory — while certainly not perfect, was and is careful, consistent, and within the bounds of sound business judgment. The cost reimbursement methodology in effect since 1973 was designed and recommended to VISA by Arthur Andersen after VISA solicited the advice of that accounting firm. Andersen was given free rein to design any appropriate methodology, both in 1973 and in its review of the VISA methodology in 1980. Friedman. TR. 15-18 (6/8/83); Stock. TR. 93-94 (1/5/83). Arthur Andersen carefully considered alternative methods for computing IRF, even some which have been suggested by NaBanco such as interchange at par. At the behest of VISA Arthur Andersen re-evaluated IRF methodology in 1977. Each time, Arthur Andersen concluded and VISA agreed that a cost-based methodology using systemwide average costs was preferrable to any other. Friedman. TR. 15-18 (6/8/83), TR. 7-8 (6/9/83).

IRF was designed to perform the equilibration function the court described earlier. Edwards TR. 73-76, 79-81 (6/3/83); Weil 31, 50-51 (1/9/84). It embodies the same analysis that a single firm such as American Express would conduct in order to allocate costs between its cardholder and merchant divisions were it to divide its business between the two. Weil. TR. 38-47 (1/9/84). These findings are true of both the 1973 and the 1981 methodology, even though the 1981 methodology contains refinements not present in the 1973 formula.

Similarly, the selection of the banks to be studied by Arthur Andersen is a careful process which yields banks that are, within limits, representative of the VISA system as a whole. VISA first examines the results of cost "questionnaires" made available to each member bank. Next it selects from these five or six banks which appear to be representative of all member banks based on a number of factors, most particularly whether their per-unit costs are typical of the system as a whole. Schmidt. TR. 176-78 (6/9/83); Wecker. TR. 73 (1/11/84). This selection process is then reviewed and approved by Arthur Andersen before the banks are actually visited by the accounting firm for further study. Schmidt. TR. 176-78 (6/9/83). While it may be that, in the best of all worlds, random sampling would yield the most "accurate" results, and study and evaluation of each and every bank in the VISA system even better ones, the sampling technique employed by VISA is certainly adequate in light of the costs associated with systemwide evaluation. In fact, VISA's sampling technique seems to produce a more accurate result than does random sampling. Wecker. TR. 80-83, 100-108 (1/10/84), 100 (1/11/84); Exhibit 2810. If anything, the banks selected by VISA are representative of banks with slightly lower than average costs resulting in an IRF which is lower in amount than one of each banks true costs. Wecker. TR. 109-110 (1/10/84).

CONCLUSION

Plaintiff, NaBanco, having been conceived in and nourished by the enterprise it presently attacks, now finds one of the stabilizing mechanisms of that enterprise legally offensive. NaBanco seeks injunctive and damage relief, contending that the interchange fee, and particularly the manner of its setting, are both anti-competitive and in violation of the Sherman Act. Indeed, the plaintiff contends that the fabric under which it became successful is so faulty that it merits per se treatment and swift demise.

Because of the findings of fact and related conclusions of law previously set forth — perhaps with prolixity justified only by the desire to make a complicated and important point evident — the court rejects plaintiff's position and will enter judgment for defendant; not however, before some additional conclusory remarks.

The focal point of the case, the VISA interchange fee, was the product of turmoil created by the absence of an efficient method of transferring (distributing) costs incident to card issuing and merchant servicing functions in the early days of what became the present VISA system.

Considerable evidence was presented concerning the etiology of the IRF, the nature of VISA, the complicated method of setting the fee, the selection of banks for testing, the operation of NaBanco, the experience of various banks and bankers involved with VISA, MasterCard and NaBanco, as well as on other relevant points in issue. Both parties were well prepared and effective in presenting their positions. However, as the trial progressed and upon consideration of the totality of the evidence, it became increasingly clear not only that the plaintiff had not presented the more convincing and weightier evidence, but rather that the greater weight — the more convincing evidence — had been provided by defendant.

The present VISA business arrangement is relatively young. While describing it as a "joint venture" may be overly ambitious in the strict business organization sense, the term seems fairly applied from the anti-trust standpoint. The venture soon became aware that the "universality" of acceptance so vital to the system could not be achieved without a uniformity of regulation which included a non-divisive method of distributing costs. The more convincing evidence was that the mandated interchange fee (for the use of Base II) was necessary to achieve stability and thus universality; that to require exchange at par or remove the fee to permit negotiations for interchange charges among issuer and merchant banks would result in loss of competitiveness and chaos with the eventual destruction of the enterprise.

Indeed, one questions how much expertise is necessary to envision the consequences of literally thousands of issuer and merchant banks negotiating exchange fee contracts or ad hoc agreements for the purchase of specific paper where there is no contract. The problem is stated: "A" — who has a card issued by a Miami bank visits a San Francisco store, whose bank has no exchange with the Miami bank. The Miami customer would simply be unable to use his card. Understandably, the store owner would accept only those cards authorized by his merchant bank. Universality is gone, and the enterprise is no longer competitive.

The argument that banks would eventually work out a par exchange is not borne out by VISA history. Different and more extreme methods of cost transfer (even if permitted in all states) would apparently impede the functioning of the system, rendering it less efficient, and thus less competitive.

The IRF serves to shave substantial costs and risks between VISA members and interchange transactions are an integration because in such transactions two different members combine their activities in order to offer a service to a cardholder and merchant. I conclude that to do violence to the method of setting the exchange fee would do violence to the functioning of the system.

IRF should be analyzed under the rule of reason because it is an agreement on the terms of interchange necessary for VISA to market its product and be an effective competitor. Prohibiting IRF would therefore undermine "interbrand" competition, "which is the primary concern of the antitrust law." Continental T. V., Inc. v. GTE Sylvania, Inc., 433 U.S. 36, 52 n. 19, 97 S.Ct. 2549, 2558 n. 19, 53 L.Ed.2d 568 (1977). Even if IRF were not necessary to market the product, it should be analyzed under the rule of reason because it is an agreement internal to a type of joint venture which yields efficiencies beneficial to competition, that its members, acting alone, could not offer, and which allows the venture to offer a product which is different from, and greater than, the sum of the individual products of its members.

VISA's establishment of an IRF, whether or not it is literally "price fixing," is not "price fixing" in any meaningful sense of the word or in a sense which makes it a per se violation of the Sherman Act. Price fixing generally refers to an agreement on prices which otherwise would be set by market forces. However, market forces could not set IRF for a number of reasons, most notably the likelihood of opportunistic, free rider behavior of the type identified by the Supreme Court when it eschewed per se analysis in Continental T. V. Inc. v. GTE Sylvania, Inc., 433 U.S. 36, 97 S.Ct. 2549, 53 L.Ed.2d 568 (1977).

In addition, IRF is properly analyzed under the rule of reason because it is not mandatory. Acting on behalf of its principals, NaBanco is and always has been free to negotiate different terms of interchange, and VISA issuers have been willing to make alternate arrangements. A practice is not unlawful per se where, as in this case, there is no legal, practical or conspiratorial impediment to making alternate arrangements.

BMI and Maricopa do not require the Court to make a subjective judgment as to whether a joint venture's product is "unique." Rather, they call for the Court to compare the product of the venture to the individual offerings of the venture's members and determine whether the venture offers pro-competitive utilities which its members, acting alone, could not offer.

The burden of proof in a rule of reason case rests with the plaintiff. The rule of reason requires "a systematic comparison" of the negative effects of the restraint on competition, if any, with any alleged positive effects on interbrand competition stemming from the restraints. In order to prevail, NaBanco has the burden of establishing that the anti-competitive effects of the setting of the IRF decidedly outweigh the pro-competitive benefits of the practice. It must show that the effects of the restraint were "substantially adverse," and that the adversity falls on market competition and not merely on its own business.

To establish anti-competitive effects, NaBanco must prove that VISA has power in a relevant market. In order to establish such power NaBanco must begin by offering proof of "a well defined relevant market upon which the challenged anti-competitive actions would have a substantial impact." ( Graphic Products, supra at 1569). NaBanco must then establish market power. (The ability to raise price significantly above the competitive level without losing all of one's business; Graphic Products, at 1570; the ability to raise prices above those which would be charged in a competitive market. Jefferson Parish Hosp. Dist. v. Hyde, ___ U.S. ___, n. 46, 104 S.Ct. 1551, 80 L.Ed.2d 2). Plaintiff has failed to establish its view of a relevant market and the court has found payment services to be the relevant market. In this market, VISA has little power. Even if this market definition were incorrect and VISA had short run power over price, it still would have no long-run monopoly power because there are few barriers to entry. Indeed, one of plaintiffs' experts predicted the demise of VISA due to competitive innovations, if VISA persists in maintaining the IRF.

Further, even if NaBanco had established that VISA had power in a relevant market and that IRF had substantial anti-competitive effects, VISA established that IRF is necessary to offer the VISA card — a pro-competitive benefit which offsets any anti-competitive effects. In a rule of reason inquiry, the Court must evaluate the alternatives to the challenged practice as part of its weighing of the practice's pro-competitive effects. Although some cases place the burden on the plaintiff to establish that a realistic, viable and more pro-competitive alternative to the challenged practice exists, this court need not address the burden of proof issue. Assuming that the burden were VISA's, VISA met that burden.

Can it be said, in the factual setting presented here, that the IRF is a "market restraint of trade with no purpose except stifling of competition." White Motor Co. v. U.S., 372 U.S. 253, 83 S.Ct. 696, 9 L.Ed.2d 738. I find, as a matter of fact and law that it is not, but rather, to borrow from BMI and Realty Multi-List, a method of integrating costs and risks which aids in the marketing of the product and permitting the public to utilize the service with such benefits as it may yield.

Judgment will this day be entered for defendant by separate order.

APPENDIX 1

(1) Figure 1 depicts the resulting demand-supply equilibriums when various demands by cardholders (dC), merchants (dM), and the system as a whole (dT) are superimposed on a graph representing the various marginal cost curves existing in the VISA system. The resulting price and quantity at the equilibrium (e) representing the intersection of dT and cT are equal to p* and q*.

The need for a transaction or interchange fee can be demonstrated by comparing the various line segments composing the dotted line q*e. In the process of producing an industry output of q*, merchant banks incur marginal costs in the amount of q*a and issuer banks incur marginal costs in the amount of q*d. The sum of these two sets of costs is q*e. In consideration for transactional services to facilitate q* exchanges, cardholders are willing to make expenditures in the amount of q*b and merchants are willing to make expenditures in the amount of q*c. The sum of these two revenue streams is q*e. What is important to note is that the marginal cost q*d of the activities performed by the issuer banks bears no necessary relation to the amount of revenue q*b forthcoming from the cardholders with whom these banks have contractual relationships. Similarly, the costs q*a associated with the activities performed by merchant banks have no necessary relation to the amount of revenue q*c forthcoming from the merchants with whom they have contractual relationships. The sum of the two revenue streams, however, must equal the sum of the two cost streams for the system to survive. Thus, whenever the revenue coming into a member bank does not match its costs, there must be some side or transfer payment from either the merchant or the issuer bank to bring the receipts of each bank into balance. In the diagram, this transfer payment is represented by the line segment ac which exactly equals the surplus coming into the merchant bank depicted by line segment ac and the issuer bank's deficit depicted by line segment bd.

It should be noted that the position or shape of these lines are not based upon any particular "empirical" evidence. The lines are simply drawn in accordance with testimony presented by VISA at trial suggesting that BM revenues are greater than BI's revenues in the VISA system and that dC drops off considerably as prices rise. The important point is that so long as revenues to either issuer or merchant bank do not equal the costs to each of these banks, there will be a need for some kind of transfer payment.

APPENDIX 2

(1) Figure 1 illustrates the derivation of aggregate demand (dT) for transactional services of a given type in a single-merchant, single-cardholder economy. The quantity (q) axis is calibrated in units which represent the bundle of services that must be provided by VISA member banks and their agents to both the cardholder (C) and the merchant (M) in order to facilitate the execution of one exchange of goods or services between cardholder and merchant. The price (p) axis represents the payments to member banks necessary to induce the requisite supply of services for any particular transaction.

Independently, neither cardholder or merchant necessarily confronts any particular price as the one he must pay in order to have his demand (dC) or (dM) fulfilled. Thus, the individual demand curves of cardholder and merchant are drawn separately, each clearly demonstrating a different price (PC) or (PM) which the cardholder or merchant is willing to pay for some level of services (q*); i.e., for any particular point on the combined demand curve, PC does not equal PM but their combined demand will be manifested by the price (p*) and quantity (q*) in this example.

(2) Figure 2 is similar to Figure 1 except that it is intended to represent total demand (dT) of all participating cardholders and merchants. While the individual demand schedules of particular cardholders must be aggregated vertically to reach a well-defined expression of aggregate demand for transactional services in a miniature economy, total cardholder demand and total merchant demand are first summed horizontally to get total aggregate demand in a multi-cardholder/multi-merchant economy. This is because cardholders only trade with merchants and vice versa. Cardholders and merchants do not trade with other members of the set they represent.

(3) Figure 3 illustrates that relationships corresponding to that of cardholder and merchant on the demand side exist on the supply side as well. The cardholder has his banking relationship with one institution (IB) and the merchant typically has his with another bank (MB). Different tasks are performed for the cardholder than for the merchant. These different tasks mean different marginal costs ((cIB) or cMB) for each bank for any particular transaction. Figure 3 depicts possible marginal cost curves for issuer banks and merchant banks, together with their vertical aggregation (cT) which corresponds to the total marginal cost per exchange facilitated by the two banks participating in that exchange. As was the case in Figure 2, Figure 3 represents the marginal cost curves one might expect in a multi-cardholder/multi-merchant economy.