457 F.3d 608 | 7th Cir. | 2006
Before E ASTERBROOK , M ANION , and S YKES , Circuit
Judges . E ASTERBROOK , Circuit Judge . People infected by the human immunodeficiency virus (HIV), a retrovirus that causes the acquired immune deficiency syndrome (AIDS), can slow the progress of the disease by taking protease inhibitors, which hamper HIV’s ability to copy itself into additional cells. Abbott Laboratories holds a patent on Norvir ® (ritonavir), one such drug. When used in doses high enough to work as a stand-alone protease inhibitor, how- ever, Norvir causes serious side effects. It serves better as a booster for other protease inhibitors, causing them to last longer in the bloodstream. Norvir has this effect because it inhibits Cytochrome P450-3A4, an enzyme in the liver that normally metabolizes away protease inhibitors. For exam- ple, a standard dose of Fortovase ® (saquinavir) is 1,200 mg three times a day; when combined with Norvir, however, Fortovase is effective in doses of 800 mg twice a day. Abbott offers its own combination under the brand name Kaletra ® , which includes ritonavir plus the protease inhibitor lopinavir. Abbott’s patents (Nos. 5,886,036 and 6,037,157) cover ritonavir taken alone and in combination with any other protease inhibitor.
Gary Schor, who proposes to represent a class of everyone who uses protease inhibitors, contends that Abbott charges too much for Norvir alone and too little for the Norvir component of Kaletra. (Stated otherwise, Schor’s contention is that Kaletra sells for less than a cocktail made by combining Abbott’s Norvir with a protease inhibitor from some other supplier.) According to Schor’s complaint, the disparity between the unduly high price of Norvir and the unduly low price of Kaletra is designed to monopolize the market in protease inhibitors, in violation of §2 of the Sherman Act, 15 U.S.C. §2. Schor calls the strategy “mo- nopoly leveraging”: Abbott is trying to use its patent to obtain a monopoly of all protease inhibitors by inducing HIV patients to buy Kaletra, which will lead other vendors to drop out of the market. Once rivals’ products have been vanquished, Abbott will be able to jack up the price of Kaletra as well as Norvir. The district court dismissed the complaint under Fed. R. Civ. P. 12(b)(6), however, after concluding that it does not state a claim on which relief may be granted. 378 F. Supp. 2d 850 (N.D. Ill. 2005). The court concluded that “monopoly leveraging” does not violate the antitrust laws unless it takes a particular form, such as a tie-in sale or refusal to deal.
Schor’s complaint does not allege any of the normal exclusionary practices—tie-in sales (or another form of bundling), group boycotts, exclusive dealing and selective refusal to deal, or predatory pricing. Abbott sells ritonavir as part of Kaletra, but this is not a tie-in because ritonavir is available separately as Norvir. Abbott will sell to anyone willing to pay its price: there is no refusal to deal. The price of Norvir cannot violate the Sherman Act: a patent holder is entitled to charge whatever the traffic will bear. This is true of both Norvir’s price, see Brunswick Corp. v. Riegel Textile Corp ., 752 F.2d 261, 265 (7th Cir. 1984), and of a claim that the patent holder has engaged in price discrimi- nation by cutting ritonavir’s price to people who buy it (through Kaletra) in combination with lopinavir. See In re Brand Name Prescription Drugs Antitrust Litigation , 186 F.3d 781 (7th Cir. 1999); Zenith Laboratories, Inc. v. Carter- Wallace, Inc ., 530 F.2d 508, 513 n.9 (3d Cir. 1976). And antitrust law does not require monopolists to cooperate with rivals by selling them products that would help the rivals to compete. See Verizon Communications Inc. v. Law Offices of Curtis V. Trinko, LLP , 540 U.S. 398 (2004). Cooperation is a problem in antitrust, not one of its obligations.
The (relatively) lower price of ritonavir in Kaletra summons up thoughts of the price-squeeze claim in United States v. Aluminum Co. of America , 148 F.2d 416, 436-38 (2d Cir. 1945) (L. Hand, J.), which held that Alcoa vio- lated the Sherman Act by selling processed aluminum sheets for less than the price it charged for the raw alumi- num required to make them. That necessarily excluded all rivalry in the sheet-metal market. Schor’s claim is no more than a faint echo of Alcoa , however, because Kaletra sells for more than its ritonavir component purchased as Norvir, and Kaletra therefore does not meet the legal standard articulated by Judge Hand. See also Mishawaka v. Ameri- can Electric Power Co ., 616 F.2d 976 (7th Cir. 1980); Concord v. Boston Edison Co ., 915 F.2d 17 (1st Cir. 1990) (Breyer, J.) (describing the very limited scope of a price- squeeze doctrine). We therefore need not decide whether Alcoa ’s holding about price squeezes is sound.
Schor does not contend that Kaletra is an instance of predatory pricing. Even if the ritonavir component of Kaletra were deemed to cost the same (per milligram) as ritonavir sold as Norvir, the imputed price of Kaletra’s lopinavir component would be above the average variable cost of its manufacture. None of Abbott’s rivals contends that, at Kaletra’s going price, it is unable to sell its own protease inhibitor profitably. If Abbott’s rivals continue to make money from their protease inhibitors, they cannot be knocked out of the market and Abbott will be unable to raise the price of Kaletra. And without any prospect of rivals’ exit, there is also no prospect of higher prices later (“recoupment,” in antitrust argot) and no antitrust worry. See Brooke Group Ltd. v. Brown & Wil- liamson Tobacco Corp ., 509 U.S. 209 (1993); Matsushita Electric Industrial Co. v. Zenith Radio Corp ., 475 U.S. 574 (1986). A (relatively) low price for ritonavir in Kaletra then is an unalloyed benefit for consumers. The antitrust laws condemn high prices, not low ones, and it would be wholly inappropriate to use the Sherman Act to oblige Abbott to raise its price for Kaletra. And if, as Schor seems to con- tend, Kaletra is not as beneficial for consumers as the combination of Norvir and a protease inhibitor other than lopinavir, then it is easy to understand why Kaletra is sold at a discount: there’s no antitrust rule against reducing the price of products that consumers desire less than competi- tive goods.
That leaves the question whether there is a free-standing theory of “monopoly leveraging.” The first subject would have to be whether Abbott enjoys a monopoly, which seems unlikely. A patent does not (necessarily) create market power. See Illinois Tool Works, Inc. v. Independent Ink, Inc ., 126 S. Ct. 1281 (2006). Although the complaint alleges (and we therefore must assume) that ritonavir is unique in its ability to inhibit Cytochrome P450-3A4, the only benefit of that effect is to reduce the quantity of protease inhibitor required for treatment. Many drugs act as protease inhibi- tors and are substitutes for Abbott’s products. In addition to lopinavir and saquinavir, which we have already men- tioned, amprenavir (Agenerase ® ), atazanavir (Reyataz ® ), fosamprenavir (Lexiva ® ), indinavir (Crixivan ® ), and (Viracept ® ) are widely used. See nelfinavir http://www.aidsmeds.com/PIs.htm. Nonetheless, because the complaint was dismissed under Rule 12(b)(6), we must assume that Abbott has market power. Likewise we must assume that some clever combination of prices for Norvir and Kaletra could induce one or more of Abbott’s rivals to withdraw their protease inhibitors from the market, or reduce the rate of new entry. Still, there is no antitrust concern unless Abbott could make a monopoly profit for itself by keeping other drugs off the market—and there is no good economic reason to think that it could do so.
The problem with “monopoly leveraging” as an antitrust theory is that the practice cannot increase a monopolist’s profits. Abbott has (we must assume) a monopoly, but a monopolist can take its monopoly profit just once. It can collect a monopoly profit for ritonavir and allow a competi- tive market to continue in other products. Or, by reduc- ing the price of ritonavir, it can induce customers to buy more from it. But it can’t do both. Suppose the competitive price of ritonavir would be $2 per 100 mg, and that the monopoly price is $7; suppose further that the competitive price of some other protease inhibitor such as saquinavir is $3 per 400 mg. Without ritonavir, the patient must take 3,600 mg of saquinavir daily, at a price of $27; take 100 mg of ritonavir with each 800 mg of saquinavir, how- ever, and the cost falls to $26 (1,600 mg of saquinavir plus 200 mg of ritonavir) even with ritonavir at the monopoly price. If Abbott offered Kaletra at $24 for a daily dose, that would knock saquinavir out of the market—but Abbott would make less money than if it had charged the monopoly price for ritonavir alone. If it then raised the price of Kaletra to $28 (say), the producer of saquinavir would bring that drug back to market—and Abbott would lose money from reduced sales even if it did not, for it would now be charging an (implicit) price of $8 per dose of ritonavir, or more than the profit-maximizing, monopoly price.
The basic point is that a firm that monopolizes some essential component of a treatment (or product or service) can extract the whole monopoly profit by charging a suitable price for the component alone. If the monopolist gets control of another component as well and tries to jack up the price of that item, the effect is the same as setting an excessive price for the monopolized component. The monopolist can take its profit just once; an effort to do more makes it worse off and is self-deterring. See Philip Areeda & Herbert Hovenkamp, 9 Antitrust Law ¶¶ 1706a, 1706b (2d ed. 2000).
The monopolist’s profit-maximizing strategy is not to take over the market in related products (ritonavir and other protease inhibitors are complements, not substitutes, given the bad side effects when ritonavir is used alone) but to promote competition among the other producers. The less the complements cost, the more the monopolist can charge for its own product. Thus Microsoft does not make computers but encourages vigorous competition among Dell, Hewlett-Packard, Sony, Lenovo, and other participants in that market; the less it costs to buy the hardware, the more sales of operating system software there will be and the more Microsoft can charge. Similarly Abbott hopes that competition among other drug manu- facturers will drive down the price of protease inhibitors; the less they cost, the more Abbott can charge for Norvir (or the ritonavir component in Kaletra). There’s no reason to think that Abbott would be better off if it took over the market in protease inhibitors and tried to charge a monopoly price for substances that complement ritonavir. And if a manufacturer cannot make itself better off by injuring consumers through lower output and higher prices, there is no role for antitrust law to play. See Menasha Corp. v. News America Marketing In-Store, Inc ., 354 F.3d 661, 663 (7th Cir. 2004); Ball Memorial Hospital, Inc. v. Mutual Hospital Insurance, Inc ., 784 F.2d 1325, 1333-34 (7th Cir. 1986).
We appreciate the potential reply that it is impossible to say that a given practice “never” could injure consumers. A creative economist could imagine unusual combinations of costs, elasticities, and barriers to entry that would cause injury in the rare situation. See Einer Elhauge, Defining Better Monopolization Standards , 56 Stan. L. Rev. 253, 282-93 (2003); Robin Cooper Feldman, Defensive Leveraging in Antitrust , 87 Geo. L.J. 2079 (1999); Michael H. Riordan & Steven C. Salop, Evaluating Vertical Mergers , 63 Antitrust L.J. 513, 516-19 (1995); Michael D. Whinston, Tying, Foreclosure & Exclusion , 80 Am. Econ. Rev. 837 (1990); Thomas G. Krattenmaker & Steven C. Salop, Anticompetitive Exclusion: Raising Rivals’ Costs To Achieve Power over Price , 96 Yale L.J. 209, 230-49 (1986). But just as rules of per se illegality condemn practices that almost always injure consumers, so antitrust law applies rules of per se legality to practices that almost never injure consum- ers.
Rules for predatory pricing are good examples. Lower prices almost always benefit consumers. Subjecting all low prices to litigation, and the inevitable risk of error in a search for the rare instances in which consumers could be made worse off in the long run by low prices today, would make it more risky for firms to reduce prices, and they would be less inclined to do so—to consumers’ considerable detriment. That’s why in Matsushita and Brooke Group the Supreme Court held that low prices are lawful, even if the seller has considerable market power, unless rivals have been driven out of the market and recoupment is either ongoing or imminent. It is why any firm’s unilateral conduct is almost always deemed lawful unless it creates a danger- ous probability of success in monopolizing. Spectrum Sports, Inc. v. McQuillan , 506 U.S. 447 (1993). Cf. Areeda & Hovenkamp, 9 Antitrust Law ¶1730 (recommending the greatest restraint in condemning any unilateral practice as “monopolization,” given the risk of forbidding a practice that benefits consumers in ways that judges do not appreci- ate); David S. Evans & A. Jorge Padilla, Designing Antitrust Rules for Assessing Unilateral Practices , 72 U. Chi. L. Rev. 73, 80-83 (2005).
Just so with arguments that low prices are designed to “leverage” a firm from one monopoly to another. As long as rivals continue to sell, and no second monopoly is in prospect, the search for the rare situation in which that second monopoly just might allow the firm to gain a profit by injuring consumers is not worth the candle. The search itself (and the risk of error in the judicial process) has much more chance of condemning a beneficial practice than of catching a detrimental one. A price high enough to avoid condemnation under predatory-pricing rules cannot be condemned under a “monopoly leveraging” theory that is just a predatory-pricing variant without the intellectual discipline of that doctrine. Schor does not contend that Kaletra’s pricing could be condemned under Matsushita or Brooke Group , so there is nothing to this case.
Having said this, we must acknowledge that one court of appeals has adopted just such an undisciplined monopoly-leveraging principle. See Image Technical Services, Inc. v. Eastman Kodak Co ., 125 F.3d 1195, 1208-13 (9th Cir. 1997). Perhaps some portions of Berkey Photo v. Eastman Kodak Co. , 603 F.2d 263 (2d Cir. 1980), should be included in the same category. At least one court of appeals has gone the other way, rejecting Image Technical by name. See In re Independent Services Organizations Antitrust Litigation , 203 F.3d 1322, 1327 (Fed. Cir. 2000). It would be possible to cabin Image Technical by observing that, despite the opinion’s language, the case arose from a refusal to deal, so it occupies one of the traditional antitrust categories rather than a claim of “naked” monopoly leveraging of the sort that Schor attempts to pursue. But we think it better to join the Federal Circuit in saying that Image Technical just got it wrong.
The ninth circuit did not give any reason for thinking that a monopolist’s acquisition of market power in a complemen- tary product injures consumers. Instead the court attrib- uted the principle to Eastman Kodak Co. v. Image Technical Services, Inc ., 504 U.S. 451 (1992). That’s a misunderstand- ing of the Supreme Court’s opinion. See Digital Equipment Corp. v. Uniq Digital Technologies, Inc. , 73 F.3d 756, 762-63 (7th Cir. 1996). What the Supreme Court held in Image Technical is not that firms with market power are forbidden to deal in complementary products, but that they can’t do this in ways that take advantage of customers’ sunk costs. Kodak sold copiers that customers could service themselves (or through independent service organizations). Having achieved substantial sales, Kodak then moved to claim all of the repair work for itself. That change had the potential to raise the total cost of copier-plus-service above the competitive level—and, we observed in Digital Equipment , above the price that Kodak could have charged had it followed a closed-service model from the outset. Schor does not accuse Abbott of any similar switch that would exploit customers’ sunk costs; none is possible in this market. Unless we generalize the Supreme Court’s decision in Image Technical to a rule against selling products that comple- ment those in which the defendant has market power—which Digital Equipment already has held would be inappropriate—Schor is left without a leg to stand on.
Schor pretty much concedes most of this analysis but maintains that patented products are different. That’s what the ninth circuit said in Image Technical . But why would a patent matter? A given patent may (or may not, see Illinois Tool Works ) create market power, but if a monopolist cannot gain by “leveraging” its way to dominance of a related product, the fact that the patent rather than something else supplies the market power can’t create an antitrust problem.
Abbott’s patents do more to support its position than to assist Schor. Recall that the patents cover not only ritonavir administered by itself but also ritonavir adminis- tered in combination with another protease inhibitor. Abbott therefore could take control of the market in combi- nation treatments until the patents expire. A patent does not permit its owner to condition use of the patented product on the surrender of a monopoly in some other unpatented product. See Ethyl Gasoline Corp. v. United States , 309 U.S. 436, 457-58 (1940); Motion Picture Patents Co. v. Universal Film Manufacturing Co ., 243 U.S. 502, 512 (1917). But the product “ritonavir in combination with another protease inhibitor” is patented to Abbott, which therefore is entitled to monopolize the combination. Yet it has not done so—doubtless because, as we have explained, Abbott’s profits are highest when the price of other protease inhibitors is lowest, and Abbott therefore has a powerful incentive to encourage competition among other producers rather than monopolize the market for all protease inhibi- tors. It would make little sense to use the antitrust laws to condemn Abbott for a strategy (a) that it has not in fact pursued; (b) that would disserve its own interests; and (c) that the patents entitle Abbott to pursue if it chooses.
One final topic and we are done. Schor maintains that he is entitled to prevail without regard to the merits—that issue preclusion (collateral estoppel) blocks Abbott from offering any legal defense. Three users of protease inhibi- tors have brought essentially identical suits against Abbott. Two were filed in the Northern District of California and the third, Schor’s, in the Northern District of Illinois. District Judge Wilkin denied Abbott’s motion to dismiss one of the California suits under Rule 12(b)(6). See Doe v. Abbott Laboratories , 2004 U.S. Dist. L EXIS 29129 (N.D. Cal. Oct. 21, 2004). Then she consolidated them and denied (as premature) Abbott’s motion for summary judgment, con- cluding that plaintiffs are entitled to conduct additional discovery. See In re Abbott Laboratories N ORVIR Anti-Trust Litigation , 2005 U.S. Dist. L EXIS 24238 (N.D. Cal. Sept. 12, 2005). According to Schor, these decisions conclusively establish that his complaint does state a claim on which relief may be granted; all that remains for decision, he insists, is whether the complaint’s allegations can be proved at trial.
There are two problems with Schor’s use of issue preclu- sion. The first is that the California decisions are not final. They do not resolve any issue in plaintiffs’ favor; they conclude only that more litigation is required. No judgment has been entered; Abbott has not had an opportunity to appeal. Federal law determines the preclusive effect of a federal court’s decision, and as a matter of federal law the denial of a motion (whether under Rule 12(b)(6) or Rule 56), so that a suit continues and the issue remains alive, has no preclusive effect. See Financial Acquisition Partners L.P. v. Blackwell , 440 F.3d 278, 284-85 (5th Cir. 2006). Although it is possible to imagine circumstances under which the denial of a motion to dismiss may conclusively resolve some concrete issue, see Gilldorn Savings Association v. Com- merce Savings Association , 804 F.2d 390, 393-96 (7th Cir. 1986), that’s not what happened in the Northern District of California. Nothing has been resolved there with finality.
Even if a point of law had been resolved against Abbott in the California suits, that would not be preclusive in Illinois. Schor is invoking a doctrine known as offensive non-mutual issue preclusion. (The preclusion is offensive because Schor is the plaintiff and non-mutual because he is not a party to the California cases. A decision favorable to Abbott in California would not have been conclusive against Schor in Illinois, unless the California court first certified a class and Schor failed to opt out.) Although federal law recognizes the possibility of offensive non-mutual issue preclusion, see Parklane Hosiery Co. v. Shore , 439 U.S. 322 (1979), the Supreme Court added in Parklane that circumstances may make its application inappropriate. One of those circum- stances is a difference in the governing law. A district court in California must apply the ninth circuit’s decision in Image Technical . We need not. Having concluded that Image Technical misunderstood the Sherman Act, we are unwilling to allow its effect to extend beyond the boundaries of that circuit. The district court in Illinois did not err in making an independent decision about the merit of Schor’s complaint.
A FFIRMED A true Copy: Teste: ________________________________ Clerk of the United States Court of
Appeals for the Seventh Circuit