FOR THE DISTRICT OF COLUMBIA
First Data Corporation's proposed acquisition of Concord EFS, Inc. would combine two of the three largest PIN debit networks, further concentrating an already highly concentrated market. The merger would eliminate the existing competition between First Data's NYCE network and Concord's STAR network, leaving only two dominant firms providing PIN debit network services to merchants. Thousands of merchants, and millions of American consumers, will bear the costs of this near-duopoly in the form of higher prices for an important method of payment to the American economy, and ultimately in higher prices for all goods and services.
Through the testimony of a diverse group of leading American retailers, the United States will demonstrate that the provision of PIN debit network services is a relevant product market under the antitrust laws, sound economic theory, and business reality. Because the transaction would combine two of the three largest firms in a highly concentrated market, the transaction is presumed illegal as a matter of law. The United States' merchant witnesses will demonstrate through their testimony about the business realities of the PIN debit market that the defendants cannot overcome this presumption. If completed, this transaction would significantly reduce competition. Expansion by small networks, or entry by new firms, will not overcome the substantial anticompetitive effects this transaction would produce. Indeed, there has been no meaningful entry into the PIN debit market for years.
Defendants' purported efficiencies do not, as a matter of law, overcome the transaction's substantial anticompetitive effects. The efficiencies, and their alleged benefits, are speculative, unsubstantiated, and would not prevent the anticompetitive harm this transaction will produce.
For these reasons, the transaction violates § 7 of the Clayton Act, 15 U.S.C. § 18. To prevent this substantial threat to competition from the merger, the United States seeks injunctive relief under 15 U.S.C. § 25 and Fed. R. Civ. P. 65.
First Data is a leading provider of payment services, generating $7.6 billion in worldwide revenues. Headquartered in Greenwood Village, Colorado, First Data acquired a controlling 64 percent interest in NYCE in 2001. First Data's other businesses include processing PIN debit, signature debit, and credit card transactions for merchants, as well as processing credit and signature debit card transactions for banks. First Data also owns Western Union, the world's largest provider of person-to-person money transfer services.
Concord is headquartered in Memphis, Tennessee. The company generated nearly $2 billion in revenues last year. Concord acquired a series of large PIN debit networks over the last several years. Concord bought MAC in 1999, Cash Station in 2000, and then STAR in 2001, merging it with the MAC network. Shortly before STAR was acquired by Concord, STAR bought the HONOR network in 1999, which had recently acquired MOST. Concord also is a leading merchant processor and provides an array of services to debit card issuers and ATM owners.
First Data and Concord executed a merger agreement on April 1, 2003. Under that agreement, First Data will acquire Concord through an all-stock transaction valued at approximately $7 billion. The merger combines two large firms that compete for the provision of PIN debit network services to merchant customers.
Electronic funds transfer ("EFT") networks were widely introduced in the late 1970s, when bank consortiums formed numerous regional networks to enable their customers to withdraw money at ATMs owned by other banks. The use of EFT networks for PIN debit purchases began on a small scale in the early 1980s. It was not until the mid-1990s, however, that EFT networks were widely used for PIN debit transactions.
Many EFT networks, including those operated by First Data and Concord, route both ATM and PIN debit transactions. Some companies, however, operate separate ATM and PIN debit networks. For example, Visa's Interlink is a PIN debit network, while Visa's ATM transactions are routed over its Plus ATM network.
A PIN debit network is the critical electronic switch connecting a network's participating financial institutions with merchants that accept the network. PIN debit networks also perform a number of related functions necessary for the efficient operation of the networks, including: (1) promoting the network brand names to financial institutions, merchants and consumers; (2) establishing rules and standards to govern their networks; and (3) setting fees and assessments for use of the networks' products and services.
To make a PIN debit purchase at a merchant, a customer swipes a debit card at a point-of-sale ("POS") terminal and enters a Personal Identification Number, or PIN, on a numeric keypad. After the PIN is entered, the POS terminal transmits the transaction and debit card information to a "merchant processor," which acts as a conduit between the merchant and the various PIN debit networks. The merchant processor sends the information to the appropriate PIN debit network, which switches the transaction to the "card processor" of the customer's bank. The card processor accesses the bank's account database to verify the PIN and ensure that the customer has sufficient funds to pay for the purchase. The card processor sends an electronic message to the PIN debit network accepting or rejecting the transaction. The PIN debit network switches this reply back to the merchant through the merchant processor to complete the transaction. The entire authorization process takes place electronically in moments.
A transaction can be routed over a particular PIN debit network only if the debit card issued by the customer's bank participates in that network. To provide customers with seamless access to a wide array of merchants, many banks place the "bug" of more than one PIN debit network on their debit cards. Cards that can access more than one PIN debit network are often referred to as "double-bugged" or "multi-bugged."
Most PIN debit networks, including NYCE and STAR, charge both the merchant and the cardholder's bank a "switch" fee for the network switching services provided by the network. This fee typically ranges from 2 cents to 4 cents per transaction. The PIN debit networks also set an "interchange" fee, which is a fee paid by the merchant to the PIN debit network. The PIN debit network then passes on the interchange fee to the card-holder's bank, in part as payment for permitting access to the cardholder's demand deposit account. The interchange fee is typically at least four to five times as large as the switch fee, depending on the network, the merchant, and the size of the transaction. The merchant's total charge for each PIN debit transaction is therefore the interchange fee plus the merchant switch fee.
Some networks have "priority routing" rules that specify which network routes a transaction when a customer uses a multi-bugged card at a merchant that accepts more than one of the networks. NYCE's routing rule allows the cardholder's bank to designate the network that carries the transaction. STAR's routing rule, by contrast, requires most transactions on cards bearing the STAR bug to be routed over the STAR network, regardless of whether there are other bugs on the card.
The PIN debit networks' routing rules substantially limit merchants' ability to engage in least-cost routing, whereby the transaction is routed to the least expensive network whose bug is on the card. For example, if a customer uses a card that is double-bugged with STAR and PULSE, and PULSE's combined switch fee and interchange fee is lower than STAR's, in most circumstances STAR's rule prevents the merchant from routing to PULSE. STAR and NYCE aggressively monitor and enforce merchant compliance with their routing rules. Both STAR's and NYCE's network routing rules provide that the networks can fine a merchant (through the merchant's bank sponsor) for violations of the rules.
There are instances, however, when two networks' routing rules conflict. For example, an issuer of cards double-bugged with STAR and NYCE may designate NYCE as its priority network, while at the same time the STAR rule requires that STAR be designated as the priority network. In these instances -- a conflict in routing rules -- some merchants (through their merchant processors) will route the transaction to the least expensive network. Combining STAR and NYCE will eliminate the substantial number of routing conflicts that currently exist between them, reducing opportunities for merchants to reduce their costs through least-cost routing.
PIN debit has a number of price and quality advantages that allow PIN debit networks to raise prices without fear of loss of transactions to signature debit or other payment mechanisms. As the defendants explained in one of their white papers to the Justice Department during the investigation:
PIN debit transactions cost significantly less than signature debit transactions for most merchants, including the largest mass-merchandisers and supermarkets that account for the large majority of PIN debit transactions. Typically, PIN debit is roughly 35 percent to 60 percent less expensive than signature debit.
This significant price gap for many merchants is likely to persist for the foreseeable future. Visa and MasterCard, the only two providers of signature debit network services, [Redacted] signature debit interchange rates for 2004 [Redacted] the rates temporarily imposed by the Wal-Mart settlement.(1) [Redacted] MasterCard has announced that it intends to maintain its current signature debit rates. [Redacted] There is no evidence, other than unsupported speculation, that the gap will decrease within any time frame.
Because signature debit prices are primarily calculated as a percentage of the transaction value, while PIN debit prices have largely been based on a flat, per-transaction fee, prices for signature debit and PIN debit can be closer together than average for merchants with low average transaction values. Conversely, merchants with an average transaction size larger than about $40 face an even greater disparity between PIN debit and signature debit prices. The PIN debit networks have proved adept at implementing targeted discounts towards certain types of merchants or transactions with a low average transaction value, without lowering prices to other captive merchant classes. For example, STAR and NYCE maintain separate interchange and switch tiers for quick service restaurants (e.g., McDonald's), which have a low average transaction value. Other merchants for whom PIN debit transactions are substantially less expensive than signature debit transactions do not benefit from these targeted lower rates. Similarly, this fall, when STAR announced increased interchange rates, STAR introduced a new "petroleum" category that offered lower interchange rates for gas stations (which also have a relatively low average transaction value), while increasing rates for supermarkets and mass merchandisers, such as Safeway and Target.
Many merchants also prefer PIN over signature transactions because PIN debit offers a more secure form of payment. It is harder to obtain someone's PIN than to forge a signature. In addition, PIN debit transactions settle in seconds, rather than a day or two, ensuring that the customer's account still has the funds available to pay for the transaction. [Redacted]
Some merchants also prefer PIN debit to signature debit because PIN debit can reduce the amount of time each customer spends at check-out. For some signature debit transaction, a merchant must scan the customer's merchandise and total the bill before printing out the receipt and obtaining and verifying the customer's signature. With a PIN debit transaction, by contrast, the customer can enter a PIN as soon as the first product is scanned, and the receipt prints as soon as the bill is totaled.
A crucial consideration for many merchant customers of PIN debit services is that there is substantial consumer demand to make PIN debit network transactions. Many customers value the "cash back" feature, which is not available with signature debit transactions. Others value the speed and security of the product. [Redacted] Consequently, many merchants are reluctant to stop accepting or to discourage customers from using it for fear of incurring higher payment costs and risking alienation of many of their consumers.
The three largest PIN debit networks are STAR, NYCE, and Interlink. As of March 2003, STAR routed 56 percent of all PIN debit transactions, while Interlink and NYCE accounted for approximately 15 percent and 10 percent of the PIN debit market, respectively.(2) While recent contract losses could reduce STAR's share, STAR will likely remain the largest network with at least a 35 percent market share. If the transaction is completed, the combined STAR/NYCE will be the largest network. Even assuming STAR cannot make up for the volume lost as a result of recent bank defections, the combined STAR/NYCE would have at least a 45 percent market share. Together, the combined STAR/NYCE network and Interlink will form an effective duopoly, accounting for approximately 80 percent of all PIN debit transactions.(3)
The PIN debit network services market is two-sided. PIN debit networks provide interdependent services to two sets of customers: financial institutions and merchants. The services are interdependent because financial institutions attach greater value to networks that are accepted by many merchants and merchants assign more value to networks with financial institutions that issue many debit cards.
PIN debit networks enter into membership agreements with financial institutions. A network's members issue debit cards with the network's electronic "bug" embedded in the cards' magnetic strip. Account holders can use their bank cards at any ATM or merchant that accepts the networks for which their cards are bugged.
PIN debit networks compete for issuing financial institutions by offering: (1) lower issuer switch fees for the financial institution; (2) higher PIN debit interchange fees; (3) broad merchant coverage; and (4) network reliability and speed. The networks offer superior terms for large banks because large bank contracts are particularly valuable. In addition to producing more volume, the participation of large banks makes the network more attractive to other issuers and merchants due to the substantial network effects in the industry.
PIN debit networks compete for merchant customers by attempting to convince merchants to accept and retain their networks and to route transactions to their networks when there is a choice of routing options. The primary terms of competition are switch and interchange fees, which together constitute the fee that the merchant pays for each PIN debit transaction. The fact that the card issuer, and not the network, ultimately receives the interchange fee does not remove the interchange fee as a crucial element of PIN debit network competition for merchants. To the contrary, PIN debit networks compete to offer merchants lower effective interchange rates, by limiting interchange increases, slotting merchants in interchange tiers with lower rates than the merchant would otherwise qualify for, or providing direct cash payments to offset rate increases.
For the PIN debit market, the particular circumstances that characterize that line of business mean that merchant customers have two principle methods to obtain better pricing: (1) threatening to drop or actually dropping a network; and (2) routing around a network on multi-bugged cards.
Because most issuers participate in at most a few PIN debit networks, merchants cannot seamlessly switch from one competitor to another, as they can for many of the goods and services they purchase. The increasing prevalence of debit cards with only one network bug makes it likely that dropping a network will cause rejected transactions, use of more expensive payment forms such as signature debit, delays at check-out lines, and dissatisfied and lost customers. Notwithstanding these risks, the threat to drop a network remains a potent, if drastic, competitive tool merchants can wield in the already concentrated PIN debit network services market. [Redacted]
The substantial consolidation of PIN debit networks over the past five years has made it increasingly difficult for merchants to counteract price increases by the PIN debit networks. For example, STAR (even after its recent bank losses) switches [Redacted] percentage of many merchants' transactions, and thus would be difficult to drop. NYCE, however, is a smaller network and, [Redacted], is a more realistic recipient of a dropping threat. For merchants in segments of the economy in which consumers have come to expect PIN debit services (e.g., supermarkets, mass merchandisers, drug stores), extracting pricing concessions is difficult. Nevertheless, because merchants can stop accepting a single PIN debit network, and yet continue accepting many or most PIN debit transactions on competing networks, merchants can constrain PIN debit network pricing in some circumstances.
Merchants are also able to constrain PIN debit network pricing by engaging in least cost routing. Some of the very largest merchants always least-cost route and do not adhere to STAR and NYCE network routing rules. Merchants such as [Redacted] when evaluating networks, engage in the type of decision-making that characterizes less concentrated, more competitive markets -- they select the least expensive alternative. For these merchants, the transaction will substantially reduce least-cost routing opportunities between STAR and NYCE. Dr. Ordover calculates that about [Redacted] percent of NYCE transactions occur on debit cards that are bugged with both.
This significant number of debit cards bugged with STAR and NYCE provides a real competitive constraint against the two networks' ability to increase switch and interchange fees:
Because of the substantial number of cards that are double-bugged with STAR and NYCE, the transaction will significantly limit the ability of merchants to hold down price increases as a result of their ability to engage in least-cost routing.
Moreover, some merchants and processors that otherwise adhere to the network routing rules least-cost route when there are conflicts in network routing rules. Conflicts occur when two networks both claim "priority" status for a particular debit card that contains each network's bug. In these instances, some merchant processors route PIN debit transactions to the least cost network. The number of routing conflicts that occur between networks, including STAR and NYCE, is [Redacted]
The most dramatic example of the three-way competition between STAR, NYCE, and Interlink occurred from mid-2001 through mid-2002. Interlink announced approximately a 70 percent increase in PIN debit interchange in July 2001. The increase was scheduled to take effect on October 13, 2001.(4) An internal STAR analysis undertaken in response concluded that [Redacted] On August 30, 2001, STAR announced a fee increase that resulted in STAR's new rates falling just below those of Interlink.(5) In deciding not to increase interchange to the same level as Interlink, STAR concluded that [Redacted]
On September 4, 2001, Wal-Mart announced that it would drop Interlink because it would not tolerate such "irrational pricing." Other major merchants, including Walgreens and RaceTrac, also threatened to drop Interlink.(6) In response to the merchant pressure, Visa announced a delay in the effective date of the Interlink fee increase until March 1, 2002. Shortly thereafter, STAR followed Interlink, announcing a similar delay.
At the same time, NYCE carefully considered its competitors' decisions to decide how to react. A September 2001 NYCE Board presentation explained that [Redacted] The presentation analyzed NYCE's pricing options and reasoned that:
Shortly thereafter, NYCE announced fees [Redacted] to STAR's, [Redacted]
NYCE's and the other PIN debit networks' concerns were partially motivated by retailer threats to route to competing networks on multi-bugged cards. [Redacted]
Against this back-drop, Interlink decided to reduce its already delayed price increases for large merchants by instituting a three-tiered pricing system. The tiered system offered lower interchange rates for larger merchants, and a separate category of even lower rates for supermarkets. STAR then effectively matched Interlink's partial retreat by introducing a similar tiered price schedule. [Redacted] NYCE dramatically scaled back its price increase, [Redacted]
After the PIN debit network pricing situation finally stabilized, [Redacted] calculated that the tiered interchange discounts adopted by STAR, NYCE and Interlink produced [Redacted] million in annual reductions in the total fees merchants would pay compared to the fees originally announced.(7)
These lower interchange rates, generated in large measure by STAR, NYCE and Interlink's three-way competition, translated into tens of millions in annual dollar savings for consumers. Merchant customers of PIN debit networks typically need to pass on some or all of the PIN debit networks' switch and interchange fee to consumers. [Redacted]
First Data's acquisition of Concord would destroy the three-way competitive dynamics that produced the tiered switch and interchange fee discounts in 2001. The merger of STAR and NYCE would make it harder for merchants to credibly threaten to drop a network and reduce opportunities to spur competition through least-cost routing techniques. Merchants' prices for PIN debit network services will rise, increases that the merchants will "inevitably" pass on to consumers [Redacted]
First Data Corporation's proposed acquisition of Concord EFS, Inc. would combine two of the three largest PIN debit networks, further concentrating an already highly concentrated market. Section 7 of the Clayton Act prohibits acquisitions that may substantially lessen competition. California v. American Stores Co., 495 U.S. 271, 284 (1990) (citing 15 U.S.C. § 18). By proscribing transactions that "may" substantially lessen competition, Congress "indicate[d] that its concern was with probabilities, not certainties." FTC v. Heinz Co., 246 F.3d 708, 713 (D.C. Cir. 2001) (quoting Brown Shoe Co. v. United States, 370 U.S. 294, 323 (1962)). Section 7 does not require proof that higher, anticompetitive prices will occur in the affected market. Hospital Corp. of America v. FTC, 807 F.2d 1381, 1389 (7th Cir. 1986) (Posner, J). "All that is necessary is that the merger create an appreciable danger of such consequences in the future. A predictive judgment, necessarily probabilistic and judgmental rather than demonstrable is called for." Heinz, 246 F.3d at 719 (quoting Hospital Corp. of America, 807 F.2d at 1389). "[D]oubts are to be resolved against the transaction." FTC v. Elders Grain, 868 F.2d 901, 906 (7th Cir. 1989) (Posner, J.) (citing Philadelphia Nat'l Bank, 374 U.S. at 362-63).
Courts generally follow three basic steps in analyzing a merger. A court must determine (1) the "line of commerce" -- usually referred to as a product market -- for assessing the transaction; (2) the "section of the country" -- or geographic market -- for assessing the transaction; and (3) whether the proposed transaction may substantially lessen competition in that defined product and geographic market. See United States v. Marine Bancorp., 418 U.S. 602, 618-23 (1974); FTC v. Swedish Match, 131 F. Supp.2d 151, 156 (D.D.C. 2000); FTC v. Cardinal Health, Inc., 12 F. Supp.2d 34, 45 (D.D.C. 1998); FTC v. Staples, Inc., 970 F. Supp. 1066, 1072-73 (D.D.C. 1997).
This three-step process is typically applied through a burden-shifting framework. Heinz, 246 F.3d at 715 (citing Philadelphia Nat'l Bank, 374 U.S. at 363). See also United States v. Baker Hughes Inc., 908 F.2d 981, 982-83 (D.C. Cir. 1990). First, the United States must define the relevant product and geographic market, and establish the market concentration level. If the United States shows that the post-merger firm would have a significant market share in a sufficiently concentrated market, the transaction is presumed illegal. Heinz, 246 F.3d at 715 (citing Philadelphia Nat'l Bank, 374 U.S. at 363); Baker Hughes Inc., 908 F.2d at 982. A merger that "produces a firm controlling an undue percentage share of the relevant market, and results in a significant increase in the concentration of firms in that market, is so inherently likely to lessen competition substantially that it must be enjoined in the absence of evidence clearly showing that the merger is not likely to have such anticompetitive effects." General Dynamics, 415 U.S. at 497 (quoting Philadelphia Nat'l Bank, 374 U.S. at 363).
The presumption is rebuttable, however. Market shares and concentration levels, "while of great significance," are not "conclusive indicators of anticompetitive effects." Heinz, 246 F.3d at 717 n.12 (citing United States v. General Dynamics Corp., 415 U.S. 486, 498 (1974); Brown Shoe, 370 U.S. at 322 n.38). If the United States establishes the presumption of illegality, the defendants "must produce evidence that 'shows that the market-share statistics give an inaccurate account of the merger's probable effects on competition' in the relevant market." Heinz, 246 F.3d at 715 (quoting United States v. Citizens & S. Nat'l Bank, 422 U.S. 86, 120 (1975) (brackets omitted)). Factors often invoked in an attempt to rebut the presumption of illegality include low barriers to entry and efficiencies. Baker Hughes, 908 F.2d at 985. If the defendants rebut the presumption of illegality, the United States must produce additional evidence demonstrating that the merger will produce anticompetitive effects. Heinz, 246 F.3d at 715; Baker Hughes, 908 F.2d at 983. The United States retains the ultimate burden of proof at all times. Id.
The relevant product market in which to assess the competitive effects of First Data's proposed acquisition of Concord is the provision of PIN debit network services. Courts define a product market by determining "the reasonable interchangeability of use [by consumers] or the cross-elasticity of demand between the product itself and substitutes for it." Brown Shoe, 370 U.S. at 325. See also Eastman Kodak Co. v. Image Tech'l Servs., Inc., 504 U.S. 451, 482 (1992); E.I. du Pont de Nemours & Co., 351 U.S. at 404. The product market "must be drawn narrowly to exclude any other product to which, within reasonable price variations, only a limited number of buyers will turn." Times-Picayune Publ'g Co. v. United States, 345 U.S. 594, 612 n. 31 (1953) (emphasis added).
Reasonable interchangeability and cross elasticity of demand are usually determined by considering how buyers of the good would respond to a price increase. Cf. United States v. Visa, 344 F.3d 229, 239 (2d Cir. 2003) ("[I]f prices for general purpose payment cards were to rise significantly, cardholders would likely pay the increased fees, rather than abandon their cards in favor of other forms of payment. Thus, general purpose payment cards constitute a distinct market, separate from the market for such other payment alternatives."). To analyze buyers' responses to price increases, courts usually apply what is known as the "hypothetical-monopolist test." Under the test, "a market is properly defined when a hypothetical profit-maximizing firm selling all of the product in that market could charge significantly more than a competitive price, i.e., without losing too many sales to other products to make its price unprofitable." Visa, 163 F. Supp. 2d at 335.
The hypothetical monopolist test is employed by, and detailed in, the Horizontal Merger Guidelines issued by the U.S. Department of Justice and the Federal Trade Commission.(8) The Guidelines define a product market by taking the smallest possible group of competing products and asking whether a hypothetical monopolist that sells those products would profitably impose a significant price increase. See U.S. Dept. of Justice & Federal Trade Commission, 1992 Horizontal Merger Guidelines § 1.11 (rev. 1997). This test is sometimes referred to as the "SSNIP test." If in response to a Small but Significant and Nontransitory Increase in Price (or "SSNIP") for the given product, enough buyers would turn to another product, making the price increase unprofitable, then the product market must be expanded by including additional products until a hypothetical monopolist controlling the expanded grouping of products would profitably impose at least a SSNIP. Id.
Markets are defined using the hypothetical-monopolist test because the test directly addresses the central question in product market analysis whether a candidate market is one in which market power can be exercised. Market power is the ability "to increase price above the competitive level without losing so much business to other suppliers as to make the price increase unprofitable." United States v. Rockford Mem'l Corp., 898 F.2d 1278, 1283 (7th Cir. 1990). See also Jefferson Parish Hospital Dist. No. 2 v. Hyde, 466 U.S. 2, 27 n.46 (1984) ("[M]arket power exists whenever prices can be raised above levels that would be charged in a competitive market."). If a company holding 100 percent of the defined product market could not exercise significant market power, then the candidate market would be useless in predicting the harm that the antitrust laws are designed to prevent that is, increased prices and lower output. "The lawfulness of an acquisition turns on the purchaser's potential for creating, enhancing, or facilitating the exercise of market power the ability of one or more firms to raise prices above competitive levels for a significant period of time." United States v. Archer-Daniels-Midland Co., 866 F.2d 242, 246 (8th Cir. 1988) (citing E.I. duPont de Nemours & Co., 351 U.S. at 391, 393).(9)
The D.C. Circuit has long applied the hypothetical monopolist test in defining antitrust markets. See, e.g., United States v. Microsoft, 253 F.3d 34, 81 (D.C. Cir. 2001) (concluding that "plaintiffs must as a threshold matter show that the browser market can be monopolized, i.e., that a hypothetical monopolist in that market could enjoy market power"); Heinz, 246 F.3d at 716 & n.9, 718 (applying Merger Guidelines in determining market concentration and noting that in defining the product market "the district court concluded that . . . consumers will switch between [the merging parties' goods] in response to a 'small but significant and nontransitory increase in price (SSNIP)"); see also Rothery Storage Van Co. v. Atlas Van Lines, Inc., 792 F.2d 210, 218 (D.C. Cir. 1986) ("To define a market in product and geographic terms is to say that if prices were appreciably raised or volume appreciably curtailed for the product within a given area, while demand held constant, supply from other sources could not be expected to enter promptly enough and in large enough amounts to restore the old price and volume."). (10)
The District Courts in this district have likewise applied the hypothetical monopolist test on multiple occasions to define product markets in merger cases. E.g., United States v. Sungard Data Systems, Inc., 172 F. Supp. 2d 172, 182, 186-92 (D.D.C. 2001); Swedish Match, 131 F. Supp. 2d at 160-01; Staples, 970 F. Supp. at 1076-77.
In addition, in the Visa case, the district court, specifically referencing the Merger Guidelines, held that the provision of general purpose credit and charge card network services (a two-sided market) was a relevant product market because "merchant consumers exhibit little price sensitivity and the networks provide core services that cannot reasonably be replaced by other sources." Visa, 163 F. Supp. 2d at 338. Notably, the court relied heavily on the hypothetical monopolist analysis presented by Dr. Michael Katz, one of the defendants' economic experts in this case:
Professor Katz also used the Merger Guidelines price sensitivity analysis to confirm the existence of a network services market . . . I adopt Prof[essor] Katz's opinion that there would be no loss to network transaction volume in the face of even a 10% increase in price for network services.
Visa, 163 F. Supp. 2d at 339.(11)
A corollary to the Guidelines' hypothetical monopolist product-market methodology is that functional substitutability between two products or services is not, by itself, sufficient to place the products or services in the same product market. For example, in Swedish Match, the court found that, while some switching between the two products occurs, "the limited amount of price-based substitution" was insufficient to show that the two products were in the same market. 131 F. Supp. 2d at 164. Similarly, in United States v. Archer-Daniels-Midland Co., 866 F.2d 242, 246 (8th Cir. 1988), the court found that sugar and high-fructose corn syrup, while functionally interchangeable, were not in the same product market because of the substantial price difference between the two products: "[A] small change in the price of HFCS would have little or no effect on the demand for sugar." Id. at 248; see also Staples 970 F. Supp. at 1074 (finding that office supplies sold through office superstores constituted the relevant product market, even though the functional interchangeability of those office supplies sold by other outlets was undisputed); U.S. Anchor Mfg. v. Rule Indus., Inc., 7 F.3d 986, 995-96 (11th Cir. 1993) (different types of boat anchors, though functionally perfectly interchangeable, were not in the same product market).(12)
The antitrust laws, Merger Guidelines and sound economic analysis all compel the same conclusion: the properly defined product market to assess the effect on competition of the proposed merger is the provision of PIN debit network services. Other payment methods would not act to constrain a 5-10 percent increase by a hypothetical PIN debit monopolist because a significant number of merchants find PIN debit to be a superior payment method with a lower cost than its closest substitutes. The factors that make PIN debit superior for the merchant include a significantly lower price, less fraud, less time to process at the register, quicker settlement, and no charge-backs. Moreover, a significant portion of the merchants' customers prefer to use PIN debit, which creates an incentive for the merchant to offer the service. Thus, PIN debit has grown substantially over the last decade and has become a very popular payment method in many sectors of the retail economy, including supermarkets, mass merchandisers, and drug stores.
A 5-10 percent increase in the fees the merchants pay for PIN debit would not change any of the above; PIN debit would still be a superior product at lower cost for the merchant.(13) Consequently, the overwhelming majority of merchants would not reject or discourage customers from executing PIN debit transactions in response to a moderate increase in the price of the product.
At trial, the United States will present testimony from large merchants from the sectors of the economy that generate the vast majority of PIN debit transactions. These merchants will explain that because of PIN debits' superior prices and/or features, and because of the strong demand from their customers for the product, they would not defeat a small but significant price increase by rejecting or discouraging consumers from making PIN debit transactions. The depositions taken to date indicate the Defendants will not be able to present any testimony at conflicting merchant testimony at trial.
There is no legal or economic support for the notion that the hypothetical monopolist test should be discarded simply because the PIN debit market is two-sided in nature -- as Dr. Katz himself recognized in the Visa case, where he applied the hypothetical monopolist test to credit cards, another two-sided market. The SSNIP test gauges the ability of a hypothetical monopolist to profitably raise prices -- exactly the appropriate question in assessing the area of competition likely to be effected by a merger.
In any event, consideration of the bank side of the market does not in any way suggest that the hypothetical monopolist test leads to the wrong result. First, most banks do not perceive PIN and signature debit networks as particularly close substitutes. [Redacted]
Furthermore, [Redacted], while some financial institutions have incentives to favor signature debit over PIN debit, these incentives would not prevent a hypothetical PIN debit monopolist from profitably raising merchant fees by a modest amount. If a network raised merchant switch fees by a moderate amount, holding interchange constant, the incentives of financial institutions would not change because they would not bear the costs of the increase. They would receive the same interchange fee per transaction and, as discussed above, merchants would continue to accept PIN debit. Even if, acting conservatively, one were to consider an increase in switch fees equal to 5-10 percent of the sum of switch plus interchange -- i.e., an increase of approximately 2 cents -- financial institutions would have no incentive to take any action to depress PIN debit volume.In short, the two-sided nature of the market does not limit the ability of a network to profitably impose a moderate price increase on merchants.
Notably, defendants' expert, Dr. Katz, does not even contest that, under the hypothetical monopolist test, the provision of PIN debit network services is a distinct relevant product market. Rather, unable to defend defendants' arguments by using what he previously termed the "now-standard procedure for defining relevant product markets," Dr. Katz embarks on a 40-page journey during which he offers a composite of his views on at least twelve different reasons why PIN debit is not an antitrust product market.(14) The length and breadth of Dr. Katz's arguments are indicative of their problems. Most glaring is that Dr. Katz's discussion sheds no light on the central market definition question that the antitrust law asks -- whether customers would defeat a moderate price increase in PIN debit network services by switching to alternative products. Indeed, Dr. Katz does not present any specific quantitative analysis of the impact on price changes of PIN debit network services of other forms of payment. Rather, he relies heavily on the opinions of financial analysts, as well as reports from the industry's equivalent of the morning newspaper, to support his core argument -- which skirts the actual issue -- that PIN and signature debit perform similar functions. This utterly unreliable evidence is simply not a substitute for the sworn testimony of the sophisticated financial institutions and merchants who participate in this market on a daily basis that Plaintiffs will present at trial.
Remarkably, although defendants count thousands of actual customers of PIN debit services as their customers (banks and merchants), Dr. Katz does not cite interviews, declarations, or testimony from any customer to support his arguments. Dr. Katz's report cites only a single merchant declaration [Redacted] obtained by the defendants for the unremarkable fact that merchants' preference for a payment product might decline in some circumstances. Notably, when the United States followed up [Redacted], the company provided a sworn statement supporting the existence of separate PIN and signature debit product markets:
Dr. Katz also implies (without actually committing himself) that paper checks are in the same product market as PIN debit network services. Missing again is any specific example of merchants using paper checks to defeat or limit PIN debit price increases or a specific hypothesis as to when and how merchants could possibly use checks in the future to prevent price increases for PIN debit network services. [Redacted]
As a final fallback argument, Dr. Katz implies (again, without committing himself) that electronic ACH (or Automated Clearing House) payment methods are in the same market as PIN debit network services. He offers no empirical evidence or customer views to support this speculation. In fact, ACH debit cards constitute far less than 1 percent of all debit transactions. Electronic debit cards that utilize the ACH system also do not provide meaningful competition to PIN debit networks. The most prominent provider of these cards is Debitman, [Redacted] Debitman cards are issued by merchants, rather than banks. For a merchant location to issue the Debitman card, the merchants' locations must have a complete "card activation system," [Redacted] Customers are unwilling to use the card until it is accepted by many merchants, and merchants are unwilling to devote the resources to issuing the product until their customers demonstrate a willingness to add yet one more card to their wallets and purses. Yet Dr. Katz incredibly relies on Debitman and similar firms to act as a competitive constraint on PIN debit networks -- which collectively handle more than 500 million transactions each month.
The United States and defendants agree that the relevant geographic market in this matter is the United States.
A transaction is presumed illegal under Section 7 of the Clayton Act if the combined entity would have a significant market share in a sufficiently concentrated market. Philadelphia Nat'l Bank, 374 U.S. at 363. "By showing that a transaction will lead to undue concentration in the market for a particular product in a particular geographic area, the government establishes a presumption that the transaction will substantially lessen competition." United States v. Baker Hughes, 908 F.2d 981, 982 (D.C. Cir. 1990) (footnote omitted). In other words, the government establishes a prima facie case of a Section 7 violation by demonstrating "that the merger would produce 'a firm controlling an undue percentage share of the relevant market, and [would] result [ ] in a significant increase in the concentration of firms in that market.'" Heinz, 246 F.3d at 715 (quoting Philadelphia Nat'l Bank, 374 U.S. at 363) (alterations in original).
A growing number of courts, including those in the D.C. Circuit, apply the Merger Guidelines' approach for assessing pre- and post-merger concentration through use of the Herfindahl-Hirschman Index ("HHI"). See, e.g., Heinz, 246 F.3d at 716; PPG Indus., 798 F.2d at 1506. The HHI for a market is calculated by summing the squares of the individual market shares of all firms participating in the market. Merger Guidelines ¶ 1.5.
The Merger Guidelines state that markets with an HHI below 1000 are "unconcentrated;" those with an HHI between 1000 and 1800 are "moderately concentrated;" and those with an HHI above 1800 are "highly concentrated." Id. ¶ 1.51. When the post-merger market is "highly concentrated," and an acquisition would result in an increase of more than 50 points in the HHI, the acquisition is presumed to be "likely to create or enhance market power or facilitate its exercise." Id. ¶ 1.51(c). The D.C. Circuit has followed this approach, finding that "[s]ufficiently large HHI figures establish [a] . . . prima facie case that a merger is anti-competitive." Heinz, 246 F.3d at 716.
First Data's acquisition of Concord would produce a PIN debit network services market that is significantly above the Merger Guidelines' threshold for presuming that the transaction would substantially reduce competition.
March 2003 industry data,(15) the most recent market share information available, does not account for some recent STAR losses of bank contracts to Interlink. Even using defendants' conservative estimates, the combined STAR/NYCE would still account for approximately 45 percent of all PIN debit transactions (35 percent from STAR, 10 percent from NYCE). Moreover, two networks -- STAR/NYCE and Interlink -- would dominate the market, routing approximately 80 percent of all PIN debit transactions. The post-merger market would be highly concentrated, with an HHI of approximately 3400, and the merger would result in an increase in the HHI of approximately 700.
Thus, regardless of the impact on STAR's market share of its actual and projected contract losses, the First Data/Concord transaction will produce a market structure that easily exceeds the standards for presumptive illegality. See, e.g., Philadelphia Nat'l Bank, 374 U.S. at 364-65 (holding that merger producing firm with 30 percent market share in market where four firms would have 78 percent of sales was presumptively illegal); United States v. Continental Can Co., 378 U.S. 441, 461 (1964) (finding merger presumptively illegal where combined firm's market share was 28 percent and the leading four firms accounting for 67 percent of the market post-merger); PPG Indus., 798 F.2d at 1502-03, 1506 (concluding that 53 percent post-merger market share and HHI of 3295 left "no doubt" that FTC was entitled "to some preliminary relief").(16)
Defendants can only rebut this presumption of illegality by showing that other market characteristics would preclude the merger from substantially lessening competition. Philadelphia Nat'l Bank, 374 U.S. at 363 (concluding that defendants must clearly demonstrate that the acquisition will not substantially lessen competition); General Dynamics, 415 U.S. at 497-98. The presumption of illegality can be overcome only if the defendants show that "the market-share statistics gave an inaccurate account of the acquisition['s] probable effects on competition." Citizens & S. Nat'l Bank, 422 U.S. at 120; accord Heinz, 246 F.3d at 715; Rockford Mem'l, 898 F.2d at 1285-86 (noting that once government established merger would create firm with monopoly market share, "it behooved the defendants to present evidence that the normal inference to be drawn from such a market share would mislead").
The recent experiences of participants in the marketplace, as well as economic analysis of data, squarely support the legal presumption that the First Data/Concord transaction violates Section 7 because it is likely to substantially reduce competition. PIN debit networks -- principally STAR, NYCE, and Interlink -- compete for merchant business based on the fees they charge merchants (a combination of switch fees and interchange fees) and their quality of service. A PIN debit network that sets merchant switch and interchange fees too high relative to other PIN debit networks, or that provides inferior service, takes the risk that merchants will decline to accept the network ("drop the network"), or route around the network to competing PIN debit networks when presented with double-bugged cards. The consolidation of STAR and NYCE would significantly increase the combined firms' ability to increase merchant fees because it would make it more costly for merchants to credibly threaten to drop, or actually drop, the combined network, and it would substantially reduce merchants' opportunities for least-cost routing.
Merchants will testify to the importance of network size in limiting their ability to drop the network and the impact of routing rules on their ability to least-cost route. Merchants reasonably generalize, based on their imperfect information, that the larger the share of a given merchants' PIN debit transactions routed over a single network, the greater the likelihood of adverse consequences if the merchant stops accepting the network. [Redacted]
The likely impact of the proposed merger on merchants' ability to drop a combined STAR/NYCE network is dramatic. Today, NYCE switches a relatively small percentage of transactions for many merchants [Redacted] This smaller scale makes it viable for these merchants to consider dropping NYCE. (For some networks in particular regions of the country, STAR may also currently account for a relatively modest share of the merchant's PIN debit transactions.) Combining the two networks would produce a network that switches approximately 45 percent of PIN debit transactions on a national basis, while in many regions of the country the combined network would account for a significantly higher percentage. Many merchants' ability to credibly threaten to drop such a large network will be limited.
Dr. Ordover analyzes the effect of the merger on the risk of dropping the network. His conclusions confirm the views of the marketplace participants. His economic analysis centers on data reflecting bugging patterns on transactions using PIN debit cards. Dr. Ordover concludes that the greater the proportion of a network's transactions that are on single-bugged cards, the greater the cost to a merchant of dropping that network. Dropping a network with a relatively large proportion of transaction on single-bugged cards would lead to: (1) more transactions denied at the register, which can cause the customer to leave;(17) (2) alienation and embarrassment of customers even if they do not walk out; (3) delay in the check-out lines; and (4) transactions that cannot be processed as a PIN debit transaction likely being completed using a more costly alternative such as signature debit. The data(18) establishes that the merged network would have a larger number of exclusively bugged cards. Today [Redacted] percent and [Redacted] percent of STAR's and NYCE's transactions, respectively, are on single-bugged cards; post-merger, [Redacted] percent of the transactions over the combined STAR/NYCE would go over single-bugged cards. Now [Redacted] percent and [Redacted] percent of all transactions that are on single-bugged cards are from NYCE and STAR, respectively. After the transaction, the combined STAR/NYCE would account for [Redacted] percent of all single-bugged cards (approximately [Redacted] percent if STAR's recent bank losses are projected forward). In other words, for many merchants, dropping the combined network would put at risk approximately 1 in [Redacted] current PIN debit transactions. Thus, the merchants' fear that dropping the merged network would be more costly is well-founded.(19)
The transaction would also limit competition by reducing least-cost routing opportunities. The merchant witnesses and Dr. Ordover reach the same conclusions. During the 2001-02 price increase period, merchants reduced the size of network price increases from STAR, NYCE and Interlink by threatening to route to lower cost networks. Dr. Ordover analyzed why this threat was effective, particularly as to STAR and NYCE. Dr. Ordover calculated that [Redacted] percent of NYCE transactions are made on cards that also have the STAR bug, and that [Redacted] percent of STAR transactions are made on cards that also have the NYCE bug.(20)
Post-merger, the option to route between NYCE and STAR disappears as a practical matter. Although network routing rules limit the ability of merchants to route, some large merchants, [Redacted], nonetheless route to the least-cost alternative. Other merchants, through their processors, route to the least-cost alternative only when there is a conflict in routing rules. Dr. Ordover calculated that over [Redacted] percent of transactions on cards with both STAR and NYCE bugs have such conflicts, and thus would be susceptible to least-cost routing but for the merger. From the combined firm's perspective, the merger means it would retain any transaction that would have otherwise have been lost by NYCE to STAR, or by STAR to NYCE, when the merchant could route via either of them. This, in turn, creates an incentive for the merged firm to increase merchant fees.
Interlink is currently the primary competitor to STAR and NYCE, and would be the only significant competitor to the merged firm. The proposed merger would therefore result in a duopoly accounting for over 80 percent of the PIN debit market. The critical question in assessing the effect of Interlink's market presence on the effect of the merger on competition is whether Interlink would act to undermine higher merchant fees imposed by the merged firm. The clear answer is "no."
Interlink is owned by Visa, which is a bank-controlled association. Visa's objective is to increase the profits of its member banks. To further this objective, Interlink attempted to increase interchange by over 70 percent in 2001.(21) Those efforts had to be scaled back and restructured when some of the large merchants credibly threatened to drop Interlink.(22)
In addition, Visa owns other payment mechanisms, including Visa Check Card. Visa and MasterCard are the only two providers of signature debit, with Visa controlling almost 80 percent. Banks would receive more revenue per transaction from signature debit than from PIN debit. [Redacted] Interlink, however is currently constrained in its ability to raise prices by competition in the marketplace; in 2001, these same competitive forces limited the amount of Interlink's proposed price increase. As detailed above, Defendants' proposed merger would reduce merchants' ability to constrain pricing. Visa will have no incentive to drive prices back down.
The presumption of illegality can potentially be overcome if entry in the relevant market is so easy that the merged entity could not profitably maintain a price increase above pre-merger levels. See, e.g., Baker Hughes, 908 F.2d at 987 ("In the absence of significant [entry] barriers, a company probably cannot maintain supracompetitive pricing for any length of time."). Whether entry is sufficiently easy to eliminate the anticompetitive danger posed by a given transaction depends on whether such entry would be timely, likely and sufficient in its magnitude, character and scope to deter or counteract the loss of competition. See, e.g., Cardinal Health, 12 F. Supp. 2d at 56-58; Merger Guidelines ¶ 3.0.
In the PIN debit network services market, substantial entry is very unlikely to occur within a sufficient period of time to discipline the merged firm. Nor are small PIN debit networks currently on the fringe of the market likely to expand in a timely manner to prevent the combined First Data/Concord from increasing prices to merchants. Any firm attempting to enter or expand in the PIN debit network services market faces a formidable series of challenges to obtaining significant market share. Most importantly, the market is characterized by substantial network effects: the more bank members and merchants that accept a network, the more valuable the network is to other banks and merchants. It is thus very difficult for a network to increase either its bank members or merchant acceptance unless it has already achieved a critical mass of bank members and merchants. Banks are less likely to join a network that only a small number of merchants accept, and merchants are less likely to accept networks that have a small number of bank members. [Redacted]
Because of these substantial network effects, even if a company could develop a new network that was equal technologically to existing networks, the incumbent firms would retain a tremendous advantage. See, e.g., New York v. Microsoft, 231 F. Supp. 2d 203, 212-13 (D.D.C. 2002) (noting that Microsoft's monopoly share of operating systems was protected by "network effects" or a "chicken-and-egg" problem because "(1) most consumers prefer operating systems for which a large number of applications have already been written; and (2) most developers prefer to write for operating systems that already have a substantial consumer base"); IIA Areeda et al., Antitrust Law ¶ 421f, at 71 (2d ed. 2002) (hereinafter "Areeda") (observing that "network externalities -- that is, where the product becomes more valuable as the number of other users increases" -- can result in "incumbent monopoly pricing and the entrant's superior product [being] insufficient to convince a decisive number of customers to switch").
Not surprisingly -- given the PIN debit market's substantial network effects and unique structure -- there have been no new entrants of significant size into the PIN debit network services market in some time. To the contrary, the market has been characterized by a rapid consolidation of regional networks. Simultaneously, the price that merchants pay for PIN debit network services has soared without spurring entry sufficient to prevent such price increases. Accordingly, the presence of network effects makes it quite unlikely that entry or expansion will occur to prevent the combined First Data/Concord from taking advantage of its increased market power to increase the price of PIN debit network services to merchants.
The combination of another of First Data's and Concord's respective businesses -- their merchant processing businesses -- with their PIN debit networks also would impede entry by new networks. The combined First Data/Concord will process between 40 and 50 percent of all PIN debit transactions. Merchant processors route debit card and transaction information from the merchant to the PIN debit network after the customer swipes a debit card at the POS terminal. When a double-bugged card is used, merchant processors can determine over which network to route a transaction. However, merchant processors can and do route transactions to favored networks, to the detriment of competing networks.
[Redacted] In contrast, First Data, which owns the smaller NYCE network, purportedly routes transactions where there are conflicts to the lowest-cost network most of the time to benefit it customers. The merged firm's loyalities will be clear. Because the combined First Data/Concord firm will own both the dominant PIN debit merchant processor and the dominant PIN debit network, First Data will have a significant incentive to route PIN debit transactions executed with double-bugged cards to the combined STAR/NYCE network, to the substantial detriment of competing networks.(23)
The likely continued presence of STAR's PIN debit network routing rule would also significantly impede new network entrants and prevent fringe entrants from expanding. STAR's rule requires that most PIN debit transactions executed with debit cards bugged with STAR be routed over STAR's network. The rule has enormous impact on current and potential existing competitors. [Redacted]
NYCE officials are keenly aware of the effect of STAR's routing rule on their network's PIN debit volume. [Redacted]
[Redacted] The effect of STAR's rule in precluding other networks from entering or expanding is no less significant.
Pleading a defense of efficiencies, defendants hypothesize that the transaction will produce over [Redacted] million in relevant annual cost savings.(24) To be considered by the Court, efficiencies must be verifiable and of a nature and magnitude such that they can be expected to remediate the competitive harm resulting from the merger. Defendants' efficiencies claims are insufficient to overcome the prima facie illegality of the merger of two of the three largest PIN debit networks; they are not verifiable, and many of them can be achieved without the merger or without its anticompetitive effect in the PIN debit network services market.
Only "convincing proof" of efficiencies may rebut evidence that a merger is highly likely to lead to an anticompetitive effect in a relevant market. Heinz Co., 246 F.3d at 720 n.18 (quoting 4 Phillip Areeda & Donald Turner, Antitrust Law ¶ 941b, at 154 (1980)). In a market as highly concentrated as the PIN debit network services market, defendants' "convincing proof" must be of "extraordinary efficiencies" in order to rebut the presumption that the transaction will reduce competition. See Heinz, 246 F.3d at 721-22 (citing University Health, 938 F.2d at 1223); see also Staples, 970 F. Supp. at 1088-90; IVA Areeda ¶ 971f, at 44 (requiring "extraordinary" efficiencies when the "HHI is well above 1800 and the HHI increase is well above 100"), cited in Heinz, 246 F.3d at 721. Defendants cannot meet this demanding standard.
Efficiencies cannot be speculative, but rather must be verifiable. See Staples, 970 F. Supp. at 1089-90; see also Guidelines at §4. Defendants cannot "overcome a presumption of illegality based solely on speculative, self-serving assertions." University Health, 938 F.2d at 1223; see also Staples, 970 F. Supp. at 1089-99 (noting that defendants "failed to produce the necessary documentation" to verify their claims, and defendants' efficiencies witness "was unable to explain the methods used to calculate many of the cost savings"). Especially in highly concentrated markets, "the court must undertake a rigorous analysis of the kinds of efficiencies being urged by the parties in order to ensure those 'efficiencies' represent more than mere speculation and promises about post-merger behavior." Heinz, 246 F.3d at 720. This rigorous analysis at a minimum requires defendants to provide enough documentation of the alleged efficiencies so that the defendants' analysis can be replicated. Clearly, the burden of demonstrating the bases for the efficiencies claims in significant enough detail to verify them falls on defendants. Considering that defendants are in control of their efforts to estimate efficiencies and all of the documentation and analyses that were done, this is not an onerous requirement. Only by replicating defendants' analysis can defendants alleged efficiencies be truly substantiated and verified.
Before the parties signed a merger agreement, rough estimates of the potential total efficiencies (or synergies) from integrating the two companies ranged from [Redacted] to [Redacted] million.(25) [Redacted] When the merger was finally announced on April 2, 2003, defendants stated they would realize $230 million in annual synergies by the end of 2005 from the integration of the two companies. From that day forward, $230 million has been the target(26) for defendants' integration team and has become a self-fulfilling prophecy.(27) [Redacted]
While defendants have committed significant resources to their integration efforts [Redacted] and held series after series of meetings, additional resources and process cannot fix an opaque effort to search out efficiencies. Further, the results of the effort cannot be replicated and therefore are not verifiable. The continuing top-down effort resulted in an estimate of over [Redacted] million in total efficiencies to be realized by [Redacted]. This number was reached after innumerable meetings among members of the integration team to identify cost savings from the merger. At these meetings, the efficiencies, their underlying assumptions, and their components were discussed and calculated, but for the most part, only the numbers resulting from those meetings were reported without any documentation of how they were reached.(28) The components of each assumption were then collected, collated, and aggregated at different levels of detail with the high-level numbers reported on the Dashboard Summary.(29) As Prof. Zmijewski will testify, none of this information allows defendants' estimates to be replicated, which is essential to allowing them to be verified.(30) These same deficiencies also apply to other aspects of the efficiency estimates, [Redacted] Defendants are effectively asking the Court to rely upon their business judgement without any documentation or the ability to verify their efficiencies claims. Under defendants' approach, every efficiencies claim would likely succeed.
In addition to the lack of documentation to allow the efficiencies estimates to be replicated and thus verified, defendants' past history at achieving the projected efficiencies is probative of the reliability of any estimate for this merger. See Staples, Inc., 970 F. Supp. at 1090 (considering defendants' "track record of achieving cost savings through efficiencies"). One of defendants' own experts has remarked on this point: "Firms that make repeated acquisitions may be able to establish reputations for being accurate forecasters of efficiencies and/or for being successful at realizing efficiencies." Michael L. Katz, "The Role of Efficiency Considerations in Merger Control: What We Do in the U.S.," 217, in EC Merger Control: A Major Reform in Progress (G. Drauz and M. Reynolds eds. 2003). [Redacted] First Data failed to achieve the vast majority of the efficiencies projected from its partial acquisition of NYCE, and Concord has also experienced delays in achieving projected efficiencies. [Redacted] Thus, defendants' projections in relation to this transaction must be accordingly discounted.
In addition to being verifiable, defendants' alleged efficiencies must be "merger specific" to rebut any anticompetitive effect. See Heinz Co., 246 F.3d at 720-21 (citing the Guidelines at § 4). The merger-specificity requirement limits the cognizable efficiencies to those "that cannot be achieved by either company alone because, if they can, the merger's asserted benefits can be achieved without the concomitant loss of a competitor." Heinz, 246 F.3d at 722 (citing generally 4A Areeda, et al. at ¶ 973; see also Cardinal Health, 12 F. Supp. 2d at 62 ("efficiencies, no matter how great, should not be considered if they could also be accomplished without the merger"); Staples, Inc., 970 F. Supp. at 1090.
For similar reasons, cognizable merger-specific efficiencies must generally be in the relevant market so that they act to cure the actual competitive harm from the merger. See University Health, 938 F.2d at 1222 ("It is clear that whether an acquisition would yield significant efficiencies in the relevant market is an important consideration in predicting whether the acquisition would substantially lessen competition."); Rockford Mem'l, 717 F. Supp. at 1288 ("The court's exclusive role is to evaluate the merger's effect on competition for the relevant market and no more."); Guidelines at § 4 ("cognizable efficiencies are of a character and magnitude such that the merger is not likely to be anticompetitive in any relevant market"). Out-of-market efficiencies that are inextricably linked with the relevant market may be considered in rare circumstances, see Guidelines at § 4 n.36; Areeda et al., Antitrust Law ¶ 972a, 47-48, but not when, as is the case here, a partial divestiture could eliminate the anticompetitive effect without sacrificing any hypothesized out-of-market efficiencies, see Guidelines at § 4 nn.35-36.
Concretely, defendants' claimed processing efficiencies can be achieved by merging the firms' processing operations while divesting First Data's interest in NYCE (or STAR). Efficiencies that "could be preserved by practical alternatives that mitigate competitive concerns, such as divestiture or licensing" are not considered. Guidelines at § 4 n.35; see also IVA Areeda et al., Antitrust Law at ¶ 972a, at 47.(31) The competitive harm from the merger of First Data and Concord flows directly from the combination of the NYCE and STAR networks, and it would be avoided if this portion of the proposed merger were prevented, but many of the efficiencies asserted by defendants stem from consolidating other operations of First Data and Concord. Following the Guidelines' suggestion, if an efficiency could be achieved with the merger even if First Data spun off NYCE, it should not be counted.
Most of defendants' alleged efficiencies are not merger specific. To assist the Court in categorizing the efficiencies, Prof. Zmijewski looked at whether the specific efficiency "relates to the operations of NYCE and STAR, and whether there are alternatives to realizing this synergy absent the NYCE/STAR merger."(32) Using this analytical rule, Prof. Zmijewski determined that [Redacted] million alleged efficiencies were not merger-specific at all, and the rest were only merger-specific in part.(33) [Redacted] were not merger specific because they could be achieved without any affect on NYCE or STAR or on PIN debit network services.(34) In addition, the [Redacted] categories of alleged efficiencies are not merger specific and can be achieved without the merger of NYCE and STAR.(35) The remaining categories, including [Redacted] are only partially merger specific and could be achieved in part without the merger of NYCE and STAR.(36)
Defendants may argue that this Court should find the presumption of illegality rebutted because joining First Data and Concord will make the combined firm a better rival to Visa. The D.C. Circuit rejected the same argument in Heinz. In that case, the second and third largest companies in a highly concentrated market sought to merge, claiming, among other things, that they would become a better rival to the number one firm, which held a 65 percent market share. Treating the argument as a species of efficiencies defense, the D.C. Circuit said that defendants failed to meet the requirements imposed on that defense. See H.J. Heinz Co., 246 F.3d at 720-24. As the United States explained above, defendants' efficiency defense fails in this case.
Defendants' argument also would be flatly inconsistent with the Clayton Act. "The lawfulness of an acquisition turns on the purchaser's potential for creating, enhancing, or facilitating the exercise of market power the ability of one or more firms to raise prices above competitive levels for a significant period of time." Archer-Daniels-Midland Co., 866 F.2d at 246 (citing E.I. duPont de Nemours & Co., 351 U.S. at 393). If defendants' argument is that Visa has market power and that only the merger will allow Concord and First Data to counteract that market power, a requirement that must be satisfied to meet the standard of merger specificity, Heinz, 246 F.3d at 722, then the argument flies in the face of fundamental antitrust doctrine. Unless the merger creates substantial efficiencies that overcome any anticompetitive effects, reducing the number of significant competitors in a market from three to two will exacerbate, not alleviate, the problems of market power. Cf. PPG Indus., 798 F.2d at 1505-06 (rejecting defendants' argument that the rapid growth of a high-technology market showed that it was permissible to allow the merger of two of the "only three fully capable firms in the market"). Just because one company might have some market power that is, the ability to maintain high prices to consumers the solution is not creating a duopoly where the two companies share market power.
Finally, if defendants are simply raising a weakened competitor defense, namely that the loss of Concord from the market will make no competitive difference, then their claim fails.(37) An acquired firm's weakness "is probably the weakest ground of all for justifying a merger." Kaiser Aluminum & Chemical Corp. v. FTC, 652 F.2d 1324, 1339 (7th Cir. 1981). Moreover, while Concord has had some recent business setbacks, it retains relationships with over [Redacted] banks(38) and will retain a market share of about 35 percent.
Most important, as the D.C. Circuit has said, the "Supreme Court, echoed by the lower courts, has said repeatedly that the economic concept of competition, rather than any desire to preserve rivals as such, is the lodestar that shall guide the contemporary application of the antitrust laws, not excluding the Clayton Act." Baker Hughes, 908 F.2d at 990 n.12. See also Brunswick Corp. v. Riegel Textile Corp., 752 F.2d 261, 266 (7th Cir.1984) ("The purpose of the antitrust laws as it is understood in the modern cases is to preserve the health of the competitive process which means . . . to discourage practices that make it hard for consumers to buy at competitive prices rather than to promote the welfare of particular competitors.").
For the reasons stated above, the Court should enjoin the merger between First Data and Concord.
Dated: December 10, 2003
1. The Wal-Mart class-action suit involved a challenge by merchants to the "honor all cards" rules imposed by Visa and MasterCard that required merchants to accept Visa and MasterCard signature debit products if they wanted to accept Visa and MasterCard credit cards. The settlement of that suit required Visa and MasterCard to lower their signature debit rates by approximately 30 percent from August 1, 2003 to December 31, 2003. (The settlement also abolished the "honor all cards" rule.) The settlement does not prescribe the signature debit prices that Visa and MasterCard can charge after January 1, 2004.
2. The fourth-largest network is PULSE, a non-profit corporation with most of its members in the Southwest and the Midwest. While PULSE has many small member banks, and an overall market share just slightly below that of NYCE, PULSE does not compete effectively for large bank contracts due to its regional focus and operational structure. [Redacted] Consequently, Pulse's market share has remained essentially flat, excepting its acquisitions of certain smaller networks.
3. The March 2003 data is the most recent available.
4. See E-mail from Michael Caruana to Susan Zawodniak, September 20, 2001, EDD-CE05-00000669 (GX020).
5. NYCE Corp., Network Interchange Recommendations, Prepared by Benton International, November 7, 2001, FDCP01005696 (GX026); see also email from Michael Caruana to Susan Zawodniak, September 20, 2001, EDD-CE05-00000669 (GX020).
8. The hypothetical monopolist test is not dependent on the identity of the current competitors or their products. Therefore, the incentives of particular competitors, such as Visa, and its PIN debit, signature debit, and credit products, are of no relevance to defining the product market. However, they are highly relevant to the competitive effects analysis.
9. In the recent Wal-Mart litigation challenging Visa's "honor all cards policy," the district court applied the concepts of a price sensitivity test to find that credit cards were a distinct relevant product market: "That consumers might switch to another form of payment in the event of a surcharge on their credit card transactions does not alter the fact that there is no cross-elasticity of demand at the merchant level between defendants' products and all other forms of payment." In re Visa Check/MasterMoney Antitrust Litigation, 2003 WL 1712568, *3 (E.D.N.Y. April 1, 2003) (unpublished). Defendants will undoubtedly cite In re Visa Check in an effort to support their view that there is an "all debit card" market. This position is incorrect as the Visa Check court did not address that issue. The court only addressed whether a price increase imposed by a hypothetical credit-card monopolist would be defeated by switching to debit products.
10. See also Coastal Fuels of Puerto Rico, Inc. v. Caribbean Petroleum Corp., 79 F.3d 182, 198 (1st Cir. 1996) ("The touchstone of market definition is whether a hypothetical monopolist could raise prices." ); Rebel Oil Co. v. Atlantic Richfield Co., 51 F.3d 1421, 1434 (9th Cir. 1995) ("A 'market' is any grouping of sales whose sellers, if unified by a monopolist or a hypothetical cartel, would have market power in dealing with any group of buyers.") (citing Areeda, et al., ¶ 518.1b at 534); Gregg J. Werden, The 1982 Merger Guidelines and the Ascent of the Hypothetical Monopolist Paradigm, 71 Antitrust L.J. 253, 259-66 (2003) (documenting the adoption of the hypothetical monopolist paradigm by courts in the United States and enforcement agencies around the world).
11. Dr. Katz has also advocated the use of the hypothetical monopolist test to define product markets in his recent scholarly writings. See Michael L. Katz & Carl Shapiro, "Critical Loss: Let's Tell the Whole Story," Antitrust, Vol. 17, No. 2 (Spring 2003) ("The now-standard procedure for defining relevant product markets in horizontal merger cases asks whether a hypothetical monopolist controlling a group of products would find it profitable to raise the price of at least one product significantly above the prevailing level.")
12. In defendants' motion on Edward Hogan's testimony, they relied on the following passage from Cardinal Health to argue that the hypothetical-monopolist test should not be used to define a product market: "[T]he determination of the relevant market in the end is a 'matter of business reality  how the market is perceived by those who strive for profit in it." Cardinal Health, 12 F. Supp.2d at 46 (quoting FTC v. Coca-Cola Co., 641 F. Supp. 1128, 1132 (D.D.C. 1986), vacated as moot, 829 F.2d 191 (D.C. Cir. 1987)).
But business reality is of course exactly what courts must consider in determining buyers' responses to a price increase by a hypothetical monopolist. The opinion in Cardinal Health itself pointed out that a court defining a product market "must determine whether . . . there is reason to find that if the Defendants were to raise prices after the proposed mergers, their customers would switch to alternative sources of supply to defeat the price increase." 12 F. Supp.2d at 46. Moreover, the court rejected the defendants' argument that business realities supported their broader market definition: "Numerous customers testified at trial that they would not increase their direct purchases from manufacturers or consider self distribution in the event of anti-competitive practices." Id. at 48.
13. There is always a price increase which would be large enough to force consumers to look to alternate products. So, if the hypothetical PIN debit monopolist raised PIN debit prices to merchants by, say, 100 percent, enough merchants might drop PIN debit to make the increase unprofitable. Substitutes are not considered "close enough" if they would constrain a price increase only when that increase is of that magnitude.
15. Based on March 2003 data, the combined STAR/NYCE would have a 66 percent share of the PIN debit market, and the top three firms -- STAR/NYCE, Interlink, and PULSE -- would account for over 90 percent of all PIN debit transactions. First Data's acquisition of Concord would raise the HHI by about 1120, resulting in a post-merger HHI of about 4710.
16. See also FTC v. University Health, Inc., 938 F.2d 1206, 1211 n.12, 1225-26 (11th Cir. 1991) (upholding challenge to merger with an increase in the HHI of 630 from 2630 to 3260); Hospital Corp. of Am. v. FTC, 807 F.2d 1381 (7th Cir. 1986) (upholding challenge to merger with an increase in the HHI of 174, from 2242 to 2416); United States v. United Tote, Inc., 768 F. Supp. 1064, 1069-70, 1087 (D. Del. 1991) (upholding challenge to merger with an increase in the HHI of 700 from 3940 to 4640); FTC v. Illinois Cereal Mills, Inc., 691 F. Supp. 1131, 1138, 1144-45 (N.D. Ill. 1988) (upholding challenge to merger with an increase in the HHI of 492, from 2114 to 2606), aff'd sub nom. FTC v. Elders Grain. Inc., 868 F.2d 901 (7th Cir. 1989); FTC v. Bass Bros. Enters., 1984-1 Trade Cas. (CCH) ¶ 66,041, at 68, 620-22 (N.D. Ohio 1984) (upholding challenge to merger with an increase in the HHI from between 1700 and 1800 to between 1900 and 2100).
17. If the presented card has multiple bugs that the merchant accepts, the transaction would be processed over another PIN debit network on the card.
19. Economics also supports the merchants' focus on size in that, for a merchant that is concerned about down-side risk, making the wrong judgment with respect to a large network may have grave consequences for the bottom line, whereas the same is not true with respect to a small network.
21. Because Visa in a non-profit association of banks, there is no economic difference between switch fees and interchange fees. Revenue in excess of costs is distributed to its member banks.
29. See, e.g., Concord Integration Pre-Close Program Dashboard Summary, November 7, 2003, at 1 (GX174).
30. See Expert Report of Mark E. Zmijewski, December 3, 2003, ¶¶ 85-93 & App. E.
31. "Section 7 of the Clayton Act prohibits mergers that may substantially lessen competition "in any line of commerce . . . in any section of the country." Accordingly, the Agency normally assesses competition in each relevant market affected by a merger independently and normally will challenge the merger if it is likely to be anticompetitive in any relevant market. In some cases, however, the Agency in its prosecutorial discretion will consider efficiencies not strictly in the relevant market, but so inextricably linked with it that a partial divestiture or other remedy could not feasibly eliminate the anticompetitive effect in the relevant market without sacrificing the efficiencies in the other market(s). Inextricably linked efficiencies rarely are a significant factor in the Agency's determination not to challenge a merger. They are most likely to make a difference when they are great and the likely anticompetitive effect in the relevant market(s) is small." Guidelines at § 4 n.36; see IVA Areeda et al., Antitrust Law ¶ 972a, at 47-48.
32. Expert Report Mark E. Zmijewski, December 3, 2003, at ¶¶ 32, 94.
33. Id. ¶ 95.
34. See id. App. E, at 6-12; [Redacted]
35. See Expert Report Mark E. Zmijewski, December 3, 2003, App. E, at 13-27.
36. See id. at 28-43.
37. See also University Health, 938 F.2d at 1220-21; FTC v. Warner Communications Inc., 742 F.2d 1156, 1164 (9th Cir. 1984) (per curiam); United States v. Rockford Memorial Corp., 717 F. Supp. 1251, 1289 (N.D. Ill. 1989); IVA Areeda, et. al, Antitrust Law ¶ 963(a)(3), at 13 (An acquired firm's financial difficulties "are relevant only where they indicate that market shares would decline in the future and by enough to bring the merger below the threshold of presumptive illegality.").