The plaintiffs, a Marathon dealer in Indiana and a company owned by him to whom he assigned his dealership contract, filed suit against Marathon under section 1 of the Sherman Act, 15 U.S.C. § 1, charging it with tying the processing of credit card sales to the Marathon franchise and also with conspiring with banks to fix the price of the processing service. The tying arrangement is challenged under section 1 of the Sherman Act rather than section 3 of the Clayton Act because the things alleged to be tied — the franchise and the processing service — are services rather than commodities. Though some old cases say otherwise, the standards for adjudicating tying under the two statutes are now recognized to be the same. E.g.,
Southern Card & Novelty, Inc. v. Lawson Mardon Label, Inc.,
The suit purports to be on behalf of all Marathon and Speedway dealers and so names Speedway as an additional defendant. But while Speedway is a wholly owned subsidiary of Marathon, the plaintiffs do not have a Speedway dealership and so we cannot see what Speedway is doing in the case or how these plaintiffs can represent Speedway dealers. But these are the lesser anomalies in the case, and need not detain us. The district court granted the defendants’ motion to dismiss the complaint for failure to state a claim, Fed.R.Civ.P. 12(b)(6), before a motion to certify the suit as a class action was filed.
The complaint alleges that as a condition of granting a dealer franchise Marathon requires the dealer to agree to process credit card “purchases of petroleum and other products, services provided and merchandise sold at or from the [dealer’s] Premises” through a processing service designated by Marathon. The terms of the dealership (set forth in a dealers’ handbook cited in the complaint) impose the requirement only with regard to sales paid for with Marathon’s proprietary credit card, which however the dealer is required to accept in payment. A dealer who wanted to process sales paid for with other credit cards by means of a different processing system would be contractually free to do so, but he would have to duplicate the processing equipment supplied by Marathon. We’ll assume that this would be so costly as to compel dealers to process all their credit card sales by means of Marathon’s designated system, since that system can process credit card sales whether or not they are made with Marathon’s credit card, thereby enabling the dealer to handle all such sales with one set of equipment. So Marathon might be said to have tied the processing of all credit card sales by its dealers to the Marathon franchise, and so we’ll assume — for the moment. The plaintiffs contend that such a tie-in is a per se violation of the Sherman Act.
In a tying agreement, a seller conditions the sale of a product or service on the buyer’s buying another product or service from or (as in this case) by direction of the seller. The traditional antitrust concern with such an agreement is that if the seller of the tying product is a monopolist, the tie-in will force anyone who wants the monopolized product to buy the tied product from him as well, and the result will be a second monopoly. This will happen, however, only if the tied product is used mainly with the tying product; if it has many other uses, the tie-in will not
*593
create a monopoly of the tied product. Suppose the tying product is a mimeograph machine and the tied product is the ink used with the machine, as in the old case of
Henry v. A.B. Dick Co.,
Only if all or most ink were used in conjunction with mimeograph machines might the manufacturer use the tie-in to repel competition. For then someone who wanted to challenge the mimeograph monopoly might have difficulty arranging for a supply of ink for his customers unless he entered the ink business. That might be hard for him to do. Entering two markets having unrelated production characteristics might both entail delay and increase the risk and hence cost of the new entrant.
Tying agreements can also be a method of price discrimination — the more ink the buyer of a mimeograph machine uses, and hence the more he uses the machine, the more valuable in all likelihood the machine is to him. In that event, by charging a high price for the ink and a low price for the machine, the manufacturer can extract more revenue from the higher-value (less elastic) users without losing too many of the low-value users, since they don’t use much ink and hence are not much affected by the high price of the ink but benefit from the low price of the machine. See
Eastman Kodak Co. v. Image Technical Services, Inc.,
The Supreme Court used to deem tying agreements illegal provided only that, as the language of section 3 of the Clayton Act seemed to require, the tying arrangement embraced a nontrivial amount of interstate commerce. E.g.,
Northern Pacific Ry. v. United States,
So “market power” is key, but its meaning requires elucidation.
Monopoly
power we know is a seller’s ability to charge a price above the competitive level (roughly speaking, above cost, including the cost of capital) without losing so many sales to existing competitors or new entrants as to make the price increase unprofitable. E. g.,
United States v. Microsoft Corp.,
As one moves from a market of one very large seller plus a fringe of small firms to a market of several large firms, monopoly power wanes. Now if one firm tries to charge a price above the competitive level, its competitors may have the productive capacity to be able to replace its reduction in output with an increase in their own output at no higher cost, and price will fall back to the competitive level. Eventually a point is reached at which there is no threat to competition unless sellers are able to agree, tacitly or explicitly, to limit output in order to drive price above the competitive level. The mere possibility of collusion cannot establish monopoly power, even in an attenuated sense to which the term “market power” might attach, because then every firm, no matter how small, would be deemed to have it, since successful collusion is always a possibility.
The plaintiffs in drafting their complaint were at least dimly aware that they would have to plead and prove that Marathon had significant unilateral power over the market price of gasoline and so could charge a supracompetitive price (folded into the price for gasoline that it charges its dealers) for credit card processing. But all that the complaint states on this score is that Marathon is “the fourth-largest United States-based integrated oil and gas company and the fifth-largest petroleum refiner in the United States” and sells “petroleum products to approximately 5,600 Marathon and Speedway branded direct-served retail outlets and approximately 3,700 jobber-served retail outlets.” *595 Marathon and Speedway’s alleged annual sales of six billion gallons of gasoline (improperly swollen by inclusion of Speedway’s sales) are only 4.3 percent of total U.S. gasoline sales per year (computed from “Official Energy Statistics from the United States Government,” www.eia.doe. gov/basics/quickoil.html, visited May 30, 2008). That is no one’s idea of market power.
Marathon does of course have a “monopoly” of Marathon franchises. But “Marathon” is not a market; it is a trademark; and a trademark does not confer a monopoly; all it does is prevent a competitor from attaching the same name to his product. “Not even the most zealous antitrust hawk has ever argued that Amoco gasoline, Mobil gasoline, and Shell gasoline” — or, we interject, Marathon gasoline — “are three [with Marathon, four] separate product markets.”
Generac Corp. v. Caterpillar, Inc.,
What is true is that a firm selling under conditions of “monopolistic competition”— the situation in which minor product differences (or the kind of locational advantage that a local store, such as a barber shop, might enjoy in competing for some customers) limit the substitutability of otherwise very similar products — will want to trademark its brand in order to distinguish it from its competitors’ brands. But the exploitation of the slight monopoly power thereby enabled does not do enough harm to the economy to warrant trundling out the heavy artillery of federal antitrust law. And anyway in this case monopolistic competition is not alleged either. So we are given no reason to doubt that if Marathon raises the price of using the Marathon name above the competitive level by raising the price of the credit card processing service that it offers, competing oil companies will nullify its price increase simply by keeping their own wholesale gasoline prices at the existing level. The complaint does not allege that Marathon is colluding with the other oil companies to raise the price of credit card processing. And under the pleading regime created by
Bell Atlantic Corp. v. Twombly,
— U.S. -,
There is more that is wrong with the plaintiffs’ charge of illegal tying. Earlier we assumed that Marathon had indeed tied credit card processing to the franchise, but that assumption will not withstand scrutiny. All it has done is require its franchisees to honor Marathon credit cards and to process sales with them through the system designated by Marathon so that customers of Marathon who use its card have the same purchasing experience no matter which Marathon gas station they buy from. The combination of card and card processing enables Marathon to offset in an economical and expeditious manner revenues from credit card sales against costs of *596 gasoline sold to the dealers. When a dealer makes a sale with a credit card, the Marathon processing system credits his Marathon account with the price of the sale and thus reduces the amount of money that the dealer owes Marathon for the gasoline that he buys from it.
The plaintiffs do not challenge Marathon’s right to offer this service. But once it is in place the dealer has a powerful incentive to route
all
his credit card transactions through the Marathon system, as otherwise he would have to duplicate the processing equipment that Marathon supplies and lose the benefit of being able to use his retail sales revenue to offset what he owes Marathon. The additional cost of using multiple card processing systems is not a penalty imposed by Marathon to force the use of its system, but an economy that flows directly from Marathon’s offering its own credit card and credit card processing service. To call this tying would be like saying that a manufacturer of automobiles who sells tires with his cars is engaged in tying because, although the buyer is free to buy tires from someone else, he is unlikely to do so, having paid for the tires supplied by the car’s manufacturer.
Jack Walters & Sons Corp. v. Morton Building, Inc.,
The plaintiffs’ other theory of antitrust liability is that in exchange for overcharging its dealers for credit card processing, Marathon is receiving kickbacks from the banks and other financial institutions that offer credit cards. This theory, as pleaded in the complaint and explained in the plaintiffs’ briefs and argument, makes no sense. If Marathon is forcing its dealers to pay an exorbitant fee for processing credit card sales, as the plaintiffs claim, this can only hurt firms that offer credit cards. Most of the fee will be passed on to the consumer in the form of a higher gasoline price, which reduces the demand for gasoline and hence the use of credit cards. Why would issuers of credit cards pay Marathon to reduce the demand for their product? If they are colluding among themselves, they will simply charge the Marathon dealers a supracompetitive price for processing credit card transactions. By doing this they may involuntarily induce dealers to switch to other franchisors, unless the price-fixing conspiracy targets them as well. In any event the complaint gives no hint of the role that Marathon might be hired to play in a conspiracy of the card companies. So this claim, too, must be dismissed under the rule of Bell Atlantic for failure to allege a plausible theory of antitrust illegality. And therefore the entire suit was rightly dismissed.
Affirmed.
