Barr Laboratories, Inc. v. Abbott Laboratories

OPINION OF THE COURT HUTCHINSON, Circuit Judge. Barr Laboratories, Incorporated (Barr) appeals a judgment of the United States District Court for the District of New Jersey granting summary judgment to appel-lee Abbott Laboratories (Abbott). The district court exercised subject matter jurisdiction pursuant to 28 U.S.C.A. §§ 1331 1 and 1337 2 , and 15 U.S.C.A. §§ 15 3 and 26 4 . *101 This Court exercises jurisdiction over the appeal from the district court’s final judgment in favor of Abbott pursuant to 28 U.S.C.A. § 1291 (West Supp.1992). Ultimately, the district court granted Abbott summary judgment on all of Barr’s claims. We will affirm. I.' Procedural History Barr commenced this action in December 1987, claiming various violations of the antitrust laws in connection with Abbott’s marketing of erythromycin ethylsuccinate. Count I of the complaint alleged that certain contracts which Abbott had made with warehouse chain drugstores were unlawful exclusive dealing contracts in violation of section 3 of the Clayton Act, 15 U.S.C.A. § 14, and section 1 of the Sherman Act, 15 U.S.C.A. § 1 (the “exclusive dealing claim”). Count II alleged that Abbott had sold ethylsuccinate to these warehouse chains at discriminatory prices with predatory intent in violation of section 2(a) of the Clayton Act, as amended by the Robinson-Patman Price Discrimination Act, 15 U.S.C.A. § 13(a) (the “price discrimination claim”). Count III alleged that Abbott had monopolized or attempted to monopolize the erythromycin market in violation of section 2 of the Sherman Act, 15 U.S.C.A. § 2 (the “attempted monopolization claim” 5 ), by (1) entering into exclusive dealing contracts with warehouse chains as set forth in Count I, (2) discriminating in the price of ethylsuccinate 400-milligram tablets as set forth in Count II, and (3) refusing to sell on reasonable terms, the raw materials necessary to produce eryth-romycin (the “essential facilities claim”). Abbott first filed a motion to dismiss or, in the alternative, for summary judgment, on the price discrimination and essential facilities claims. The district court initially denied Abbott’s motion on April 29, 1988. On June 27, 1988, however, upon Abbott’s timely motion for reconsideration, the district court reversed its earlier decision and granted Abbott summary judgment on both claims. On the price discrimination claim, the district court found that Abbott had never sold ethylsuccinate 400-milligram tablets at or below the price charged by Barr for its generic equivalent. On this evidence, the district court recognized Abbott’s argument that Barr could not “substantiate a claim for unlawful price discrimination where Abbott has not undercut Barr’s own prices for the drug” and noted that Barr failed to produce evidence of such conduct. As an alternative ground for granting summary judgment, the court held that Abbott had established “an absolute defense on this claim” under the good faith meeting competition defense of section 2(b) of the Robinson-Patman Act because it was not charging a price below that of Barr. 6 On September 16, 1988, the district court denied Barr’s motion for reconsideration of the order granting Abbott summary judgment on the price discrimination claim. In February 1989, Abbott moved for summary judgment on the remaining counts. The district court denied that motion on June 1, 1989, because the parties had not yet completed discovery on the relevant market. When Abbott renewed its motion after completion of discovery, the district court again denied it because there remained disputed issues of fact regarding the definition of the appropriate relevant market. The district court entered a final pretrial order on May 10, 1991 and selected a trial date. Before the commencement of trial, Abbott moved in limine pursuant to Federal Rule of Civil Procedure 42(b) to bifurcate and try first the issue of definition of the relevant market. The district court granted the motion to bifurcate on July 29, 1991. On September 16, 1991, trial on the relevant market issue commenced. The parties had stipulated prior to trial that the entire United States constituted the relevant geographic market because both parties mar *102 ket their respective products on a national scale. On October 2, 1991 the jury found that the relevant product market is all adult oral erythromycin products. Abbott renewed its motion for summary judgment with regard to the exclusive dealing and remaining monopolization claims on that same date. On December 2, 1991, the district court entered an order granting summary judgment on these claims. With regard to the attempted monopolization claim, the district court held, despite Abbott’s approximate 50% market share and predatory intent, that Barr had not shown a dangerous probability of successful monopolization in light of the stability of the market structure, the entry of new manufacturers and products, and the stability of prices. The district court further concluded that the absence of evidence of anti-competitive effects on the market, coupled with Abbott’s legitimate business reasons for entering the contracts, required dismissal of the exclusive dealing claim. This appeal followed. 7 II. Factual History 1. Description of the Relevant Market Erythromycin is an antibiotic originally developed by Abbott in 1953. There are four major categories of adult oral erythro-mycin products: stearate, estolate, ethyl-succinate, and base. Abbott markets ethyl-succinate 400 mg. tablets under the brand-name “EES-400.” Abbott held patent rights and controlled production of EES-400 until January 3, 1978. In 1981 Barr obtained approval from the United States Food and Drug Administration (FDA) to manufacture and sell a generic 8 version of Abbott’s EES-400. During all times relevant to this case, a number of other major companies had FDA approval to manufacture one or more of the various types of adult oral erythromy-cin products in the relevant market. These companies included both “branded” or “pioneer” companies which manufacture name-brand drugs, 9 and generic companies like Barr. 10 Some branded companies, like Abbott, also manufacture generic drugs that compete with the products of other branded companies. In addition to the branded and generic companies, many pharmaceutical companies purchase erythromycin products from FDA-approved manufacturers and market them under their own name. 11 These companies need not seek independent FDA approval in order to market their products. *103 Manufacturers distribute their products, including erythromycin, to drugstores through three major channels: (1) sales directly to independent pharmacies; (2) sales to wholesalers who resell to pharmacies; and (3) sales to warehouse chain drugstores which supply their retail pharmacies from their own warehouses. Of these three, the warehouse chains are reputedly the most powerful and most price-sensitive buyers in the market. 12 The erythromycin market includes both single-source and multi-source products. Single-source products have no generic competition and are heavily promoted to physicians. During the relevant time period there were only two single-source eryth-romycin products on the market: Abbott’s PCE and Parke Davis’s ERYC. Multi-source products may or may not be promoted to physicians, but in any event they face fierce generic competition. Three erythro-mycin multisource products were on the market during the time relevant to this case: Abbott’s EES-400, Boots/Upjohn’s E-Mycin, and Parke Davis’s ERYC. A generic firm planning to enter the erythromycin market duplicates a branded drug chemically, then establishes the generic’s bioequivalence with that drug through studies on healthy volunteers. Barr’s Chief Executive Officer, Edwin Cohen, testified that there are no technical barriers to production of an EES generic, and that “literally a dozen” companies have the technical capability to make the drug. None of the major products in the erythro-mycin market benefitted from patent protection during times relevant to this case. FDA approval, however, was and remains a necessary prerequisite for market entry. Obtaining such approval generally took from six months to two years. However, Barr was informed in 1989 that the approval process would be lengthened because of criminal investigations concerning other generic firms. A pharmacist is legally prohibited from dispensing an erythromycin drug without a prescription from a physician. Additionally, a pharmacist also may not substitute one form of erythromycin for another. A prescription that specifies, for example, ethylsuccinate 400-milligram tablets may not be filled with any other form of eryth-romycin. Generally, the pharmacist decides whether to dispense the brand-name of the drug or its generic counterpart. There are limits, however, on this discretion. If the physician specifically states on the prescription “do not substitute” or similar language, then the pharmacist must fill the prescription with the brand-name drug specified. The number of times a physician chooses to do this, however, is relatively small. A number of states also have mandatory substitution laws requiring pharmacists to fill prescriptions with generic versions of name-brand drugs unless the prescribing physician explicitly directs otherwise. The overall structure of the erythromy-cin market has remained stable from 1984 to 1990, with the three leading manufacturers — Abbott, Boots/Upjohn, and Parke-Davis — holding a collective market share of 82.17% in 1984 and 82.69% in 1990. Abbott’s unit market share of all erythromy-cin products also remained relatively stable during this time period, increasing from 49.34% in 1984 to 51.19% in 1990. Its dollar market share increased 14%, rising from approximately 45% in 1984 to 59% in 1990. In 1990, Boots/Upjohn claimed a unit market share of approximately 17.67%, while Parke-Davis had a 13.68% unit market share. The number of manufacturers of erythromycin products increased by 23%, from twenty-six in 1984 to thirty-two in 1990. Although the variety of adult oral erythromycin products sold increased by 59%, from 111 in 1984 to 176 in 1990, 85 of the 111 products sold in 1984 represented products sold by pharmaceutical companies that repackaged and marketed products purchased from manufacturers under their own name. The FDA had approved the manufacture of twenty-six different eryth-romycin products as of 1984, and thirty-two as of 1990. Barr obtained four of those six *104 new approvals, each for a product Abbott sold under the global contracts. Prices in the erythromycin market have remained relatively stable in comparison to other antibiotics in the oral respiratory antibiotic market. Between 1984 and 1990, the average price of erythromycin increased approximately 62%, from $11.52 to $18.66 per 100 tablets. The price of other antibiotics per 100 tablets reflected a wide range of increases during the same time period: 13 Drug 1984 1990 % Change Broad Spectrum Penicillin $ 11.48 $ 24.35 + 112% Tetracycline $ 16.82 $ 26.74 + 59% Cephalasporine $ 74.51 $102.35 + 37% Cotrimoxazole $ 30.83 $ 20.67 - 32% Quinolones $178.00 $216.00 + 21% Penicillin $ 4.92 $ 4.22 - 14% 2. The Challenged Conduct . Barr claims that two groups of contracts Abbott entered with warehouse chain drugstores constituted impermissible price discrimination, exclusive dealing, and attempted monopolization of the national adult oral erythromycin market. The first group was composed of single-product contracts relating only to the product ethylsuccinate. Abbott entered these contracts with four warehouse chain drugstores for the period July 1, 1984 through June 30, 1985; the same four chains plus two more chains for the period July 1, 1985 through June 30, 1986; and the same six chains for the period July 1, 1986 through June 30, 1987. These contracts discounted the price for ethylsuccinate if the chain purchased a certain volume of the drug during a twelvemonth period. If the chain failed to purchase the specified volumes, then Abbott could charge back the difference between the discounted and undiseounted prices. The second group of contracts was made up of so-called “global” contracts covering five categories of Abbott multi-source adult oral erythromycin products: ethylsucci-nate, stearate, base film tab, and generic versions of the brand-name products E-Mycin and ERYC. Abbott entered into these contracts with ten of the largest warehouse chain drugstores for the one-year period from July 1, 1987 to June 30, 1988. The global contracts set volume targets for five specified products and provided discounts if those targets were met. Like the single-product contracts, these contracts gave Abbott the option to charge back to the chains the difference between the discounted contract price and the undis-counted price at which Abbott sold these products to buyers whose purchases did not meet the volume specified in the global contracts. The global contracts provided for a 1% rebate if the targets were met in any particular product category, and a 4% rebate if the targets were met in all five categories. A key provision of the global contracts for purposes of the present appeal provided: The. prices granted herein are given in consideration of the client’s agreement to use listed products wherever legally permissible to do so. Quantities in this contract reflect the best current estimate of that commitment. Joint Appendix (Jt.App.) at 1443. Evidence presented to the district court established that Abbott had designed the global contracts as “Package Deals for Warehouse Chains.” Jt.App. at 1475. Abbott’s Manager.of Pricing explained that the global contracts were constructed to provide an overall benefit to both parties by providing a total package rather than an individual line item or price. Advantages accruing to Abbott included fostering business goodwill by providing the chains with flexible, advantageous arrangements and minimizing transaction costs by dealing in large volumes. A memorandum prepared by Abbott’s Pharmaceutical Products Division president revealed that an additional objective was to *105 Jt.App. at 1314. A “Confidential White Paper” written for and adopted by Abbott’s president likewise stated that Abbott’s goal was to inhibit the entry of branded and generic competitors into the erythromycin market so that it could then raise prices and increase profitability. The “White Paper” further revealed that Abbott expected to attain a 75% market share by 1990. *104 —Sign Global Contracts with large chains to shut our competitors. —Establish 70-80% market share. Get marginal players out of the market. —Then raise prices. *105 None of the warehouse chains that entered into contracts with Abbott believed they were precluded from buying products from Abbott’s competitors. Rather, each deposed warehouse chain buyer testified that the chains chose to enter the global contracts because it was a wise business decision to do so. The chains cited numerous advantages stemming from the contracts, including overall price savings, quality products, service benefits, savings from stocking a drug from a single manufacturer, and the convenience of having several erythromycin products on one contract. Barr sold substantial quantities of generic ethylsuccinate to warehouse chains during the time periods in which Abbott prompted the single-product and global contracts. Of the six chains with which Abbott had entered single-product contracts, Barr sold considerable amounts of ethyl-succinate to four of those chains during that same period. 14 Barr also sold substantial units of ethylsuccinate to four of the chains during the time they had entered the global contracts with Abbott, and made documented sales during the same period to two others. In fact, Barr nearly doubled its unit sales to two chains, and increased sales to another, over the amount sold by Barr during the prior contract period. Barr reported aggregate net profits of more than $5,200,000.00 during the two-year period from July 1986 through June 1988. It is undisputed that Abbott did not sell EES-400 below cost at any relevant time. Abbott’s cost is $4.20 per 100 tablets, while its price was never less than $10.75 for 100 tablets. Moreover, Abbott’s prices did not cause Barr to sell below cost. Barr’s price for’ ethylsuccinate during the relevant time frame was at least 27% above its cost. In addition, Abbott’s price for EES-400 was, at all relevant times, above- Barr’s price for its generic equivalent. Barr argued that the single-product contracts foreclosed 5.30% of erythromycin sales during the 1986-87 period. In calculating the foreclosure percentage from the 1987-88 global contracts, Barr divided the maximum volumes set out in the contracts for all the adult products covered by the total adult units sold and arrived at a foreclosure percentage of 14.88%. Barr sought damages for sales of ethylsuccinate allegedly lost from July 1986 through June 1988 in the form of lost profits totalling less than $700,000.00. III. Analysis The standard of review on appeal from a grant of summary judgment is plenary. Mellon Bank Corp. v. First Union Real Estate Equity and Mortgage Inv., 951 F.2d 1399, 1404 (3d, Cir.1991). Although complex antitrust litigation infrequently merits summary judgment because of its fact-intensive nature, such disposition is certainly not forbidden in appropriate cases. Pennsylvania Dental Ass’n v. Medical Serv. Assoc. of Pa., 745 F.2d 248, 255 (3d Cir.1984) (citing Poller v. Columbia Broadcasting Sys., 368 U.S. 464, 473, 82 S.Ct. 486, 491, 7 L.Ed.2d 458 (1962)), cert. denied, 471 U.S. 1016, 105 S.Ct. 2021, 85 L.Ed.2d 303 (1985); Harold Friedman, Inc. v. Kroger Co., 581 F.2d 1068, 1071 (3d Cir.1978)). The standard of review on appeal from the granting of a'motion to bifurcate a trial pursuant to Federal Rule of Civil Procedure 42(b) is abuse of discretion. Gonzalez-Marin v. Equitable Life Assurance Soc’y of the United States, 845 F.2d 1140, 1145 (1st Cir.1988) (citing C. Wright & A. Miller, Federal Practice and Procedure § 2388 (1971)). The standard of review on appeal from a jury instruction is *106 “whether the charge, taken as a whole and viewed in the light of the evidence, fairly and adequately submits the issues in the case to the jury.” In re Merritt Logan, Inc., 901 F.2d 349, 359 (3d Cir.1990) (quoting Ayoub v. Spencer, 550 F.2d 164, 167 (3d Cir.), cert. denied, 432 U.S. 907, 97 S.Ct. 2952, 53 L.Ed.2d 1079 (1977)). 1. The Discriminatory Pricing Claim Barr claims that Abbott sold ethylsuccinate to warehouse chains in violation of section 2(a) of the Clayton Act,' as amended by the Robinson-Patman Price Discrimination Act. 15 There is no dispute in the present case that Abbott engaged in price discrimination by selling its erythromycin product to warehouse chains undér the single-product and global contracts at lower prices than it offered to its other customers. See O. Hommel Co. v. Ferro Corp., 659 F.2d 340, 346 (3d Cir.1981) (“The price discrimination referred to in Section 2(a) is merely a price difference.”) (citing FTC v. Anheuser-Busch, Inc., 363 U.S. 536, 549, 80 S.Ct. 1267, 1274, 4 L.Ed.2d 1385 (1960)), cert. denied, 455 U.S. 1017, 102 S.Ct. 1711, 72 L.Ed.2d 134 (1982). However, “[pjrice discrimination is not illegal per se. ” Id. At the primary level, 16 a plaintiff must show a “reasonable possibility that a price difference may harm competition.” Falls City Indus., Inc. v. Vanco Beverage, Inc., 460 U.S. 428, 434-35, 103 S.Ct. 1282, 1288, 75 L.Ed.2d 174 (1983) (citing Com Prod. Ref. Co. v. FTC, 324 U.S. 726, 742, 65 S.Ct. 961, 969, 89 L.Ed. 1320 (1945)). “ ‘A focus on detrimental effects on competition, rather than a concern with individual competitors, is fundamental to a reconciliation of the Robinson-Patman Act with overall antitrust policies.’ ” O. Hommel, 659 F.2d at 347 (quoting F. Rowe, Price Discrimination Under the Robinson-Patman Act 130 (1962)) (emphasis in original); but see J.F. Feeser, Inc. v. Serv-A-Portion, Inc, 909 F.2d 1524, 1535 (3d Cir.1990) (holding that evidence of injury to competitor may establish competitive injury necessary to show violation of RobinsonPatman Act in secondary-line cases), cert. denied, — U.S.-, 111 S.Ct. 1313, 113 L.Ed.2d 246 (1991). In either event, Barr must establish not only that Abbott engaged in price discrimination, but that this conduct substantially lessened competition or tended to create, a monopoly. O. Hommel, 659 F.2d at 346 (quoting Anheuser-Busch, 363 U.S. at 542-43, 80 S.Ct. at 1271) (other citations omitted). A plaintiff may establish competitive injury either directly from market analysis that illustrates actual competitive injury, or by inference from evidence of the defendant’s predatory intent. • Id. at 347. Barr has not asserted actual competitive injury to the relevant market. Its claim of price discrimination hinges on its showing of Abbott’s predatory intent from which it says competitive injury may be inferred. Predatory intent reflects a competitor’s willingness to sacrifice present revenues in order to achieve monopoly profits in the future. Id. at 348. In a manner similar to establishing competitive injury, a plaintiff may-show predatory intent either by express evidence or by inference from below-cost pricing. Id. at 347. *107 It is undisputed in the present case that Abbott did not price its EES-400 either below its own cost or below Barr’s price for its generic equivalent. Therefore, Barr relies exclusively on express evidence of Abbott’s predatory intent to establish competitive injury. In this connection, Barr has produced documents which unequivocally indicate that Abbott’s goal in extending the contracts at issue was to force its competitors out of the market and enable it to raise prices. Relying on 0. Hommel, Barr argues that this evidence is sufficient to establish competitive injury in absence of evidence of either below-cost pricing or actual competitive injury. An examination of 0. Hommel reveals that Barr’s reliance is misplaced. The 0. Hommel Company (Hommel) manufactured and sold porcelain enamel frit which, at the time, was the basic ingredient in the coating of such products as bathtubs, stoves, washing machines, and refrigerators. Id. at 342. It brought suit against the Ferro Corporation (“Ferro”), one of its competitors, alleging, inter alia, that Ferro had violated section 2 of the Clayton Act as amended by the Robinson-Patman Act by secretly selling to three customers at a price both below that charged to its other customers and below its average total cost, but above its average variable cost. 17 Id. at 342-43, 348. It was clear in 0. Hommel that the defendant had engaged in non-geographic, primary-line price discrimination, and that this discrimination had caused no actual competitive damage to the relevant market. Id. at 347. Thus, the primary issue was whether Hommel had presented evidence of predatory intent from which competitive harm could be inferred. Id. Finding no express evidence of predatory intent, this Court held that “in a non-geographic selective discriminatory pricing case when actual competitive harm has not been shown, Robinson-Patman primary line liability cannot be imposed merely upon evidence of selective below-average cost pricing.” Id. at 351. Contrary to Barr’s assertion, 0. Hommel does not stand for the proposition that a plaintiff may establish illegal price discrimination (a particular sub-class of predatory pricing, the perceived evil that RobinsonPatman was meant to correct) merely by producing express evidence of a competitor’s predatory intent. For the reasons that follow, we believe that such a broad interpretation of the dictum in 0. Hommel on which Barr relies would undermine the rationale of the case itself. Focusing on the legislative history of the Clayton Act, 0. Hommel noted that the primary evil that section 2 sought to eradicate was geographic price discrimination. The enactment of this section reflected Congress’s fear that a large, nationwide firm would lower costs in one region with the intent of destroying local competitors and acquiring a monopoly there, while financing its price cuts by higher prices in another region. Id. at 350-51 (citing S.Rep. No. 698, 63d Cong., 2d Sess. 2-4 (1914); H.R.Rep. No. 627, 63d Cong., 2d Sess. 8 (1914)). Non-geographic price discrimination, on the other hand, does not present a similar threat to competition because, in such a case, a competitor selectively offers lower prices to specific customers regardless of their geographic locality. The non-geographic nature of selective discrimination dissipates an otherwise concentrated threat to competition because it removes the danger of a national firm’s subsidizing price cuts in one region by correlative price hikes in a separate region, at the expense of local competitors. As O. Hommel observed, Professors Areeda and Turner would not *108 prohibit selective non-geographic price dis: crimination on the basis that If ... a monopolist should be permitted to make a general price reduction so long as his price equals or exceeds marginal cost, we are unable to see a persuasive case for prohibiting selective price-cutting to retain or obtain particular customers. Selective price-cutting cannot possibly be more harmful to small competitors than a general price reduction to the same level. Id. at 350 (quoting Areeda & Turner, Predatory Pricing and Related Practices Under Section 2 of the Sherman Act, 88 Harv.L.Rev. 697, 725-26 (1975)). Acknowledging that non-geographic price discrimination did not posé the same threat to competition as geographic price discrimination, this Court in 0. Hommel concluded that in a non-geographic selective discriminatory pricing case when actual competitive harm is absent, evidence of selective below-average cost pricing, without more, could not establish primary line liability under section 2 of the Clayton Act. Id. at 351. The same reasoning applies to the case before us. Barr claims that Abbott has engaged in non-geographic, primary line price discrimination that threatens competitive harm. The evidence it produced in support of this claim does show that Abbott’s strategy of offering selective price-cuts to warehouse chains was intended to squeeze its competitors out of business, thereby enabling it to raise prices. The evidence in this case does not show that Barr or other manufacturers were competitively impaired during the time Abbott entered the contracts at issue. To the contrary, the record affirmatively established that Barr’s share of the relevant market increased during this time. More significantly, Abbott’s prices for EES-400 were neither below its own cost nor below Barr’s price for its generic equivalent. Barr’s express evidence of Abbott’s bad faith in offering the contracts may have sufficed to establish predatory intent from which competitive harm could be inferred in the context of a geographic price discrimination claim. However, such evidence, in the absence of either below-cost pricing or actual market injury, does not establish competitive harm in a non-geographic context. 18 In this regard, it is important to remember that the relationship between competitive harm and predatory intent is inferential. In the absence of market data indicating actual competitive injury, a plaintiff may sometimes substitute evidence of a competitor’s predatory intent from which competitive injury can be inferred. The concern reflected in the legislative history of the Clayton Act, referred to in 0. Hom-mel, shows that this inference may be easily drawn in the context of geographic price discrimination. It is less readily drawn in the context of non-geographic price discrimination, where the harm to competition from subsidizing price cuts in one region by price increases in another region is, by definition, non-existent. It is theoretically possible that Abbott could have subsidized its price cuts to warehouse chains by increasing its prices to independent retail pharmacies and pharmaceutical companies. Realistically, however, were Abbott to do so in a competitive market, its competitors would step in with lower prices to defeat Abbott’s strategy. 19 *109 Abbott’s conduct does not fit the basic definition of predatory pricing. We recognize that in 0. Hommel we stated that predatory intent reflects a competitor’s willingness to sacrifice present revenues in order to achieve monopoly profits in the future. 0. Hommel, 659 F.2d at 348. Barr argues that the declarations by Abbott employees that the company’s goal was to squeeze out competition and then raise prices constitute evidence from which a loss of at least some revenue can be directly inferred. To that extent, Barr argues that Abbott did sacrifice present revenues when it reduced its price for the warehouse chains pursuant to the contracts even though the price was still above its own costs and also above Barr’s cost. This argument, at first glance, appears forceful. On analysis, however, it must fail. As Abbott pointedly notes, a choice not to maximize its profits is not per se impermissible under the antitrust laws. Indeed, basic economic theory teaches us that the chief benefit of competition is lower prices to consumers. Moreover, the advantages Abbott secured by its competitive pricing extended beyond direct monetary receipts. For example, when it entered the contracts, it also reaped business goodwill that cannot be measured by current accounting conventions. In addition, it saved transaction costs as a result of engaging in large-scale volume sales by virtue of at least the global contracts. We have previously noted that “[c]ourts have carefully scrutinized enforcement efforts by competitors because their interests are not necessarily congruent with the consumer’s stake in competition.” Alberta Gas Chems. Ltd. v. E.I. duPont de Nem-ours & Co., 826 F.2d 1235, 1239 (3d Cir. 1987), cert. denied, 486 U.S. 1059, 108 S.Ct. 2830, 100 L.Ed.2d 930 (1988). Actions that promote efficiency and lower prices in the marketplace, for example, may cause economic loss to competitors. Conduct that harms competitors may benefit consumers — a result the antitrust laws were not intended to penalize. “When the plaintiff is a poor champion of consumers, a court must be especially careful not to grant relief that may undercut the proper functions of antitrust.” Ball Memorial Hosp., Inc. v. Mutual Hosp. Ins., Inc. 784 F.2d 1325, 1334 (7th Cir. 1986); see Cargill, Inc. v. Monfort of Colo., Inc., 479 U.S. 104 [107 S.Ct. 484, 93 L.Ed.2d 427] (1986) (competitor’s loss of profits due to increased price competition following a merger does not constitute a threat of antitrust injury); see also Hovenkamp, Merger Actions for Damages, 35 Hastings L.Rev. 937, 956 (1984) (private plaintiffs’ injuries are as likely to be caused by the efficiency aspects of mergers as by their market power effects); Baumol & Ordover, Use Of Antitrust To Subvert Competition, 28 J.L. & Econ. 247 (1985). Id. (footnote omitted). We hold that, in a non-geographic, primary-line price discrimination case such as the present one, a plaintiff may not establish competitive market injury by inference from evidence of predatory intent standing alone unless the defendant’s motive for its price discrimination is coupled with evidence of below-cost pricing or some other evidence of actual competitive harm either to the plaintiff or to the relevant market. 20 We therefore affirm the district *110 court’s grant of summary judgment in favor of Abbott on the price discrimination claim. 21 2. The Exclusive Dealing Claim Barr alleges that the global contracts constituted unlawful exclusive dealing contracts in violation section 3 of the Clayton Act 22 and section 1 of the Sherman Act. 23 The district court assumed that the global contracts constituted exclusive dealing agreements, but ultimately concluded that they did not violate the antitrust laws because they had no anti-compe'titive effects, and the parties had legitimate business justifications for entering into them. 24 All exclusive dealing agreements must comply with section 1 of the Sherman Act. American Motor Inns, Inc. v. Holiday Inns, Inc., 521 F.2d 1230, 1239 (3d Cir. 1975). Contracts for the sale of goods, such as the global contracts at issue here, must also comply with the more rigorous standards of section 3 of the Clayton Act. Id. Thus, if the global contracts do not infringe upon the stiffer standards of anti-competitiveness under the Clayton Act, they will also be lawful under the less restrictive provisions of the Sherman Act. Id. (citing Standard Oil Co. v. United States, 337 U.S. 293, 312, 69 S.Ct. 1051, 1061, 93 L.Ed. 1371 (1949)). The legality of an exclusive dealing arrangement under the Clayton Act depends on whether the competition foreclosed constitutes a substantial share of the relevant market. Id. at 1250 (quoting Tampa Elec. Co. v. Nashville Coal Co., 365 U.S. 320, 327-28, 81 S.Ct. 623, 628, 5 L.Ed.2d 580 (1961)). Barr argues, under the quantitative substantiality test of Standard Oil, that the only relevant economic indicator is the percentage of the market the contracts foreclose. See Standard Oil, 337 U.S. at 299, 310-14, 69 S.Ct. at 1055 (exclusive dealing arrangement involving significant share of relevant market establishes section 3 violation). Barr thus asserts that the global contracts constitute impermissible exclusive dealing because they foreclose nearly 15% of the relevant *111 market. 25 Barr’s argument ignores the effect of Tampa Electric’s qualitative substantiality test, which “introduced greater flexibility and requires the courts actually to evaluate the restrictiveness and the economic usefulness of the challenged practice in relation to the business factors extant in the market.” American Motor Inns, 521 F.2d at 1251-52 n. 75 (citing Susser v. Carvel Corp., 332 F.2d 505, 516-17 (2d Cir.1964); Yon Kalinowski, Antitrust Laws and Trade Regulation, §§ 13.04, 13.05, 14.04 (1971); Bok, The Tampa Electric Case and the Problem of Exclusive Arrangements Under the Clayton Act, 1961 Sup.Ct.Rev. 267). Under this test: [I]t is necessary to weigh the probable effect of the contract on the relevant area of effective competition, taking into account the relative strength of the parties, the proportionate volume of commerce involved in relation to the total volume of commerce in the relevant market area, and the probable immediate and future effects which pre-emption of that share of the market might have on effective competition therein. Tampa Elec., 365 U.S. at 329, 81 S.Ct. at 629. Thus, under Tampa Electric’s qualitative substantiality test, the degree of market foreclosure is only one of the factors involved in determining the legality of an exclusive dealing arrangement. See American Motor Inns, 521 F.2d at 1252 (exclusive reliance on percentage of market foreclosed does not demonstrate consideration of sufficient factors to constitute type of economic analysis contemplated by Tampa Electric). Applying Tampa Electric, the district court here considered the impact of the global contracts on the adult oral erythro-mycin market as indicated by the following factors: —Abbott’s market share increased by less than 2% over the relevant time frame. —The level of concentration in the market increased by less than 1%. —New manufacturers and erythromycin products emerged, and prices remained stable. ' Jt.App. at 3188-89. Each of these factors supports the district court’s conclusion that the assumed 15% market preemption resulting from the contracts did not portend an immediate or future anti-competitive backlash. The market strength of the warehouse chains buttresses this conclusion. The existence of legitimate business justifications for the contracts also supports the legality of the global contracts. See Tampa Electric, 365 U.S. at 334-35, 81 S.Ct. at 632 (court must consider economic justification for exclusive dealing arrangement in deciding whether it violates section 3); Standard Oil, 337 U.S. at 306-07, 69 S.Ct. at 1058-59 (requirements contracts may assure supply, afford protection against rises in price, enable long-term planning on basis of known costs, and obviate expense and risk of storage in quantity necessary for commodity having fluctuating demand). As detailed above, the evidence established that the warehouse chains entered the contracts because of the inherent advantages they saw in them in price, convenience, and service. The contracts also proved advantageous from Abbott’s perspective in terms of reaping business goodwill, and as providing high volume, low transaction cost outlets for Abbott’s manufacturing capacity. The evidence presented supports the district court’s conclusion that the contracts at issue did not constitute impermissible exclusive dealing arrangements. 3. The Attempted Monopolization Claim Barr claims that Abbott attempted to monopolize the erythromycin market in violation of section 2 of the Sherman Act. 26 Monopoly power is the power *112 to control prices or to exclude competition. Pennsylvania Dental Ass’n, 745 F.2d at 260 (citing United States v. E.I. duPont de Nemours & Co., 351 U.S. 377, 76 S.Ct. 994, 100 L.Ed. 1264 (1956)). In order to establish a claim of attempted monopolization, a plaintiff must show that the defendant (1) had specific intent to monopolize the relevant market, (2) engaged in anti-competitive or exclusionary conduct, and (3) possessed sufficient market power to come dangerously close to success. E.g., International Distrib. Ctrs. Inc. v. Walsh Trucking Co., Inc., 812 F.2d 786, 790 (2d Cir.) (citing Northeastern Tel. Co. v. American Tel. & Tel. Co., 651 F.2d 76, 85 (2d Cir.1981), cert. denied, 455 U.S. 943, 102 S.Ct. 1438, 71 L.Ed.2d 654 (1982)), cert. denied, 482 U.S. 915, 107 S.Ct. 3188, 96 L.Ed.2d 676 (1987); Pennsylvania Dental Ass’n, 745 F.2d at 260 (citing Lorain Journal Co. v. United States, 342 U.S. 143, 72 S.Ct. 181, 96 L.Ed. 162 (1951); Swift & Co. v. United States, 196 U.S. 375, 25 S.Ct. 276, 49 L.Ed. 518 (1905)); Twin City Sport-service, Inc. v. Charles O. Finley & Co., Inc., 676 F.2d 1291, 1308 (9th Cir.) (citing William Inglis & Sons Baking Co. v. ITT Continental Baking Co., 668 F.2d 1014 (9th Cir.1981)), cert. denied, 459 U.S. 1009, 103 S.Ct. 364, 74 L.Ed.2d 400 (1982); Portland Retail v. Kaiser Found., 662 F.2d 641, 647 (9th Cir.1981)); see Twin City, 676 F.2d at 1308 (three elements of attempted monopolization claim are discrete but often interdependent so that proof of one may provide circumstantial evidence or permissible inferences of other elements). With regard to the third factor, “a dangerous probability of monopoly may exist where the defendant firm possesses a significant market share when it undertakes the challenged anticompetitive conduct.” International Distrib. Ctrs., 812 F.2d at 791. However, although the size of a defendant’s market share is a significant determinant of whether a defendant has a dangerous probability of successfully monopolizing the relevant market, it is not exclusive. Pennsylvania Dental Ass’n, 745 F.2d at 260 (citing United States v. Aluminum Co. of Am., 148 F.2d 416 (2d Cir.1945) (hereinafter “Alcoa”); see Indiana Grocery, Inc. v. Super Valu Stores, Inc., 864 F.2d 1409, 1414 (7th Cir.1989) (“market share indicates market power only when sales reflect control of the productive assets in the business ... [and thus] ability to curtail total market output”); see, e.g., United States v. Empire Gas Corp., 537 F.2d 296, 305 (8th Cir.1976) (market share of 47-50% alone not enough to establish dangerous probability of success), cert. denied, 429 U.S. 1122, 97 S.Ct. 1158, 51 L.Ed.2d 572 (1977); Alcoa, 148 F.2d at 424 (questioning whether 60-64% market share would confer monopoly power). Other factors to be considered include the strength of competition, probable development of the industry, the barriers to entry, the nature of the anti-competitive conduct, and the elasticity of consumer demand. International Distrib. Ctrs., 812 F.2d at 792. Applying these principles, the district court held that, despite Abbott’s approximate 50% market share, the other factors showed that Abbott did not have a dangerous probability of success in any attempt to monopolize the national adult oral erythro-mycin market. Specifically, the district court relied on the stability of the market structure, the entry of new manufacturers and products, and the stability of prices in the erythromycin market in making its determination. Barr argues that the district court’s reliance on these factors was misplaced in the present case. Although Abbott’s market share of all erythromycin products remained relatively stable at approximately 50% during the relevant time period, Barr observes that this market share was more than twice that of its nearest competitor, Boots/Upjohn, with 17.67% of the market, and over three times that of Parke-Davis, with its market share of 13.68%. Together, these three companies controlled 82.17% of the market in 1984 and 82.69% in 1990. Although these statistics reflect an in *113 crease in concentration of roughly only 0.5%, Barr asserts that the fact that three manufacturers controlled over 80% of the relevant market indicates a highly concentrated market. Despite the evidence of market concentration and Abbott’s sizeable market share, we agree with the district court’s conclusion that consideration of other relevant factors demonstrates that Abbott did not have a reasonable probability of successfully monopolizing the adult oral erythromy-cin market. See Empire Gas, 537 F.2d at 306 (proper to take evidence of dangerous probability of success as whole) (citing Sanitary Milk Producers v. Bergjans Farm Dairy, Inc., 368 F.2d 679, 691 (8th Cir.1966)). The variety of adult oral eryth-romycin products sold increased by 59%, from 111 in 1984 to 176 in 1990. We recognize Barr’s argument that these figures are misleading because 85 of the 111 products sold in 1984 represented products sold by pharmaceutical companies that repackaged and marketed products purchased from manufacturers under their own name. That argument, however, overlooks the fact that the pharmaceutical companies compete for sales with the manufacturers from which they buy their products. Consideration of the products promoted by pharmaceutical companies is thus proper in evaluating the competitive nature of the market. Even excluding the 85 products marketed to warehouse chains by pharmaceutical companies, the FDA had approved the manufacture of twenty-six different er-ythromycin products as of 1984, and thirty-two as of 1990, representing a 23% increase. With respect to barriers to entry, the parties are in disagreement as to whether the six month to two-year waiting period for FDA approval constitutes a significant barrier to entry. Barr relies on United States v. Marine Bancorporation, Inc., 418 U.S. 602, 629-30, 94 S.Ct. 2856, 2874, 41 L.Ed.2d 978 (1974), for the proposition that any form of government approval that is required for market entry automatically establishes a barrier to entry. Reply Brief for Appellant at 12. That case, however, discussed the unique nature of the commercial banking industry in terms of state and federal regulation. The Supreme Court of the United States noted that “this country’s bitter experience” with failed banks in the Great Depression established a historical predicate upon which both the federal and state governments had based voluminous regulations governing the requirements for entering the banking industry. Id. at 629, 94 S.Ct. at 2873. Contrasting United States v. Falstaff Brewing Corp., 410 U.S. 526, 93 S.Ct. 1096, 35 L.Ed.2d 475 (1973), the Court observed that while beer manufacturers are free to base their decisions regarding entry to a new market on nonregulatory considerations, including their financial capabilities, long-range goals, production and distribution costs, and profit projections, the entry of new competitors into the commercial banking field is “ ‘wholly a matter of governmental grace ...’ and ‘far from easy.’ ” 418 U.S. at 628-29, 94 S.Ct. at 2873 (quoting United States v. Philadelphia Nat’l Bank, 374 U.S. 321, 367 & n. 44, 83 S.Ct. 1715, 1744 & n. 44, 10 L.Ed.2d 915 (1963)). We think that the pharmaceutical drug market is more akin to the commercial beer market than to the commercial banking market in terms of barriers to entry. Barr’s reliance on Marine Bancorporation is thus misplaced. The fact that the number of manufacturers of erythromycin products increased from twenty-six in 1984 27 to thirty-two in 1990 supports our conclusion that Barr’s allegations of high barriers to entry pose no obstacle to the conclusion that a competitive market existed here. See, e.g., International Distrib. Ctrs. 812 F.2d at 792 (resolving dispute over barriers to entry at summary judgment stage by noting entry to market by one new competitor and expansion into market by several existing competitors). This increase in manufacturers is particularly significant when we con *114 sider the fact that Abbott held a legal monopoly by virtue of patent protection over a portion of this market until 1978. Although that fact could sometimes cut the other way on the question of a reasonable probability of success in an attempt to monopolize, we think the continued entry of competition, albeit with small initial market share shown on this record, indicates that Abbott’s position is subject to significant potential erosion. It thus tends to show that Abbott has no reasonable probability of success in any attempt to monopolize. Barr itself acknowledges that, “[qjuite simply, it is enormously difficult to dislodge the pioneer firm from its position of dominance after it has been selling the drug for such a long period of time under a patent and with its attendant lower costs of production.” Reply Brief for Appellant at 13 (citing SmithKline v. Eli Lilly & Co., 427 F.Supp. 1089, 1120-28 (E.D.Pa.1976), aff'd, 575 F.2d 1056 (3d Cir.), cert. denied, 439 U.S. 838, 99 S.Ct. 123, 58 L.Ed.2d 134 (1978)). Here, however, the evidence established that whatever barrier to entry FDA approval for new products poses, that barrier is not insurmountable. Barr also suffers from a lack of evidence showing that there was any significant reduction in the number of manufacturers in the market during the relevant time period because of Abbott’s allegedly anti-competitive contracts or otherwise. Evidence of a significant net reduction in the number of producers in the market could be evidence of some reasonable hope of success. Moreover, there is no evidence that any competitor was forced out because of Abbott’s conduct. See Empire Gas, 537 F.2d at 305 (finding no dangerous probability of success in absence of evidence that defendant’s competitors decided not to enter or to leave market because of defendant’s actions). The only evidence on this point is Barr’s observation that of thirty-two pharmaceutical companies that sold ethylsucci-nate products in 1990, ten had no sales, apparently indicating their exit from the market. Reply Brief for Appellant at 18 n. 6. We do not think that this evidence is sufficient in the face of the other evidence showing a continuing entry of competitors during the time that is at issue. Finally, the record shows that prices in the erythromycin market have remained relatively stable in comparison to other antibiotics in the oral antibiotic respiratory market. Between 1984 and 1990, the average price of erythromycin increased approximately 62%, from $11.52 to $18.66 per 100 tablets. The price of other antibiotics per 100 tablets reflected a wide range of increases during the same time period, from a 112% increase in the price of broad spectrum penicillin to a 14% decrease in the price of penicillin. In sum, the evidence presented establishes that during the relevant time period: (1) Abbott, the market leader, claimed a market share that increased from 49% to 51%; (2) the three leading manufacturers claimed roughly 80% of the market; - (3) six new manufacturers entered the market; (4) six new products received FDA approval, (5) the number of products competing for sales increased from 111 to 176, (6) barriers to entry remained low, and (7) prices have remained stable. Taken as a whole, this evidence reflects a competitive market and establishes the absence of, any dangerous probability that Abbott’s market share of approximately 50%, coupled with the market concentration present here might have suggested in other circumstances. Even if Abbott had achieved its goal of forcing Barr or other “marginal” players out of the market, the existing competition from other manufacturers, as well as the ability of new competitors to surmount the relatively low barriers to entry, would have prevented Abbott from raising prices for any lengthy period of time under the jury’s broad definition of the relevant market as the national adult oral erythromycin market. See Coleman Motor Co. v. Chrysler Corp., 525 F.2d 1338, 1348 (3d Cir.1975) (definition of relevant market critical in attempted monopolization claim). 28 We will affirm the district court’s grant of sum *115 mary judgment on the attempted monopolization claim because Abbott did not possess sufficient market power to pose a dangerous probability of successfully monopolizing the relevant market: ■ 4. The Bifurcated Trial Barr challenges the district court’s decision to grant Abbott’s motion to bifurcate and try first the issue of the proper definition of the relevant product market. Rule 42(b) governs a request by a party to bifurcate a trial and provides that: The court, in furtherance of convenience or to avoid prejudice, or when separate trials will be conducive to expedition and economy, may order a separate trial of any claim ... or of any separate is-sue____ Fed.R.Civ.P. 42(b); see Idzojtic v. Pennsylvania R.R. Co., 456 F.2d 1228, 1230 (3d Cir.1972) (decision to bifurcate is matter within discretion of trial judge to be decided on case-by-case basis). The district court expressly rejected Barr’s argument that the relevant product market issue was inextricably interwoven with the issues on the merits of Barr’s antitrust claims. The district court ultimately determined that bifurcation would prove beneficial by enhancing juror comprehension of the complex issues presented, and by excluding the presentation of lengthy evidence, some portion of which would inevitably become irrelevant upon definition of the relevant market. The district court further noted a lack of any unfair prejudice resulting from bifurcation, and recognized the possibility of resolving the issues on the merits by summary judgment after a factual determination of the relevant market. Barr cites Lis v. Robert Packer Hospital, 579 F.2d 819, 824 (3d Cir.), cert. denied, 439 U.S. 955, 99 S.Ct. 354, 58 L.Ed.2d 346 (1978), for the proposition that “a showing of prejudice is not necessary to obtain a new trial if the District Court failed to exercise ‘informed discretion.’ ” Brief for Appellant at 43. In that case, we determined that the appellants did not demonstrate prejudice from the bifurcation, and thus we refused to remand for a new trial, although we condemned the district court’s practice of bifurcating trials as a general rule. We nevertheless suggested that a showing of prejudice may not be necessary prospectively to justify a new trial when the trial court fails to support its decision to bifurcate with reasons demonstrating that it exercised .discretion informed by standards appropriate to the circumstances of the case. Lis, 579 F.2d at 825. Here, the district court properly set out its reasons for bifurcating the relevant market issue. Its recital of them indicates no abuse of informed discretion. None of the reasons given are inappropriate or contrary to law. Barr is not entitled to rely upon the Lis dicta to excuse its failure to demonstrate prejudice resulting from the bifurcation to which it objects. Since the district court did not abuse its discretion, and since Barr has failed to demonstrate any resulting improper prejudice from the bifurcation, we affirm the district court’s bifurcation of the issue of the relevant market. 5. The Jury Instruction Barr’s final objection is to the district court’s charge to the jury on the proper definition of the relevant market. The relevant product market includes those “commodities reasonably interchangeable by consumers for the same purposes.” United States v. E.I. duPont de Nemours & Co., 351 U.S. 377, 394, 76 S.Ct. 994, 1007, 100 L.Ed. 1264 (1956). Barr and Abbott strongly disagree as to whether the pharmacist or the physician should be considered the consumer. The district court instructed the jury on this issue as follows: In assessing whether products are reasonably interchangeable the question is whether the products are reasonably interchangeable in view of the person or persons who create the demand for the products at issue. Who creates the demand for the products at issue? You heard discussion with regard to the doctor. You heard further *116 discussion with regard to the pharmacist. What roles both play. In the case of prescription drugs like antibiotics the person who selects the particular product in the first instance is the prescribing physician. You should thus include in the relevant product market those antibiotic drugs which are reasonably interchangeable, price use and qualities considered, in meeting the antibiotic needs of the patients. The pharmacist can determine which generic manufacturer’s product is to be substituted for the drug prescribed by the doctor, assuming a generic substitute exists and also assuming the doctor has not prohibited the substitution of a generic. You should consider in the same relevant market those generic products which may be used to fill the prescription written by the doctor in the event he prescribes or authorizes the use of a generic. Jt.App. at 1149-50. Barr argues that this instruction imper-missibly removed from the jury the decision whether the physician or the pharmacist was the consumer from whose perspective the relevant market should be determined. According to Barr, By stating that it is the prescribing physician who in the first instance selects the product and “thus” the jury should include those “antibiotics [sic] drugs” that are reasonably interchangeable in treating the patient, the court instructed that it is the prescribing physician from whose perspective the relevant market should be determined. Brief for Appellant at 46-47. As Abbott pointed out in its brief, however, the instruction as worded accurately informed the jury, as the evidence established, that both the physician and the pharmacist play a role in selecting a particular drug for the patient. See Brief for Appellee at 47-48. From this information, the jury was left free to decide whether the pharmacist, the physician, or both should be considered the consumer for purposes of defining the relevant market. The instruction, taken as a whole, misstates neither law nor fact. IV. Conclusion We affirm ’the judgment of the district court on all five questions presented for appeal. First, Barr’s evidence of predatory intent, without more, is insufficient to support its non-geographic, primary line price discrimination claim. Second, Barr has not established that Abbott has the requisite dangerous probability of success to sustain its attempted monopolization claim. Third, lack of evidence of anti-competitive effects on the market, coupled with valid business justification for entering the global contracts, justifies the grant of summary judgment on the exclusive dealing claim. Fourth, the district court did not abuse its discretion in bifurcating and trying first the issue of the definition of the relevant product market. Finally, we find no error in the district court’s charge to the jury on the issue of whether the physician or the pharmacist should be considered the consumer from whose perspective the relevant product market should be considered. For these reasons, the judgment of the district court shall be affirmed. Footnotes [1] "The district courts shall have original jurisdiction of all civil actions arising under the Constitution, laws, or treaties of the United States.” 28 U.S.C.A. § 1331 (West Supp.1992). [2] "The district courts shall have original jurisdiction of any civil action or proceeding arising under any Act of Congress regulating commerce or protecting trade and commerce against restraints and monopolies____” 28 U.S.C.A. § 1337(a) (West Supp.1992). [3] "[A]ny person who shall be injured in his business or property by reason of anything forbidden in the antitrust laws may sue therefor in any district court of the United States....” 15 U.S.C.A. § 15(a) (West Supp.1992). [4] "Any person, firm, corporation, or association shall be entitled to sue for and have injunctive relief, in any court of the United States having jurisdiction over the parties, against threatened loss or damage by a violation of the antitrust laws....” 15 U.S.C.A. § 26 (West Supp.1992). [5] Barr limits its appeal on Count III to attempted monopolization. See Brief for Appellant at 24. [6] Section 13(b) provides, in pertinent part, that even if price discrimination exists, it is not unlawful if done "in good faith to meet an equally low price of a competitor.” 15 U.S.C.A. § 13(b) (West 1973). [7] Barr does not challenge the dismissal of the essential facilities claim, but it does seek review of the district court’s grant of summary judgment on its discriminatory pricing and exclusive dealing claims, as well as the attempted monopolization claim insofar as it is based on discrim: inatory pricing and exclusive dealing. Two months before the commencement of this action in December 1987, Barr filed an antitrust action against Abbott in the United States District Court for the Southern District of New York. Barr Lab., Inc. v. Abbott Lab., 86 Civ. 3164 (KTD). The parties voluntarily discontinued the New York action pursuant tó Rule 41(a)(2) of the Federal Rules of Civil Procedure. In February 1988, Abbott moved to dismiss Barr’s complaint in the present action based on res judicata or, alternatively, to transfer the case back to New York. The district .court denied both of those motions. In the pending case, Barr petitioned this Court for a writ of mandamus after the district court granted summary judgment on the price discrimination claim. Barr Lab., Inc. v. Abbott Lab. and The Honorable Anne E. Thompson, Docket No. 88-5877. After full briefing and oral argument, we denied the petition without opinion on February 14, 1989. [8] A generic drug is a drug which the FDA has determined, inter alia, to be “therapeutically equivalent" to its brand-name counterpart. [9] Major branded drug companies which had approval from the FDA to manufacture one or more of the various types of adult oral erythro-mycin products in the relevant market during all periods relevant to this case included the Parke Davis division of Warner Lambert, Upjohn (which later transferred its erythromycin product approvals to Boots Pharmaceutical), the Dista division of Eli Lilly & Co., the Wyeth Ayerst division of American Home Products, and Squibb (subsequently Bristol Myers Squibb). [10] Other major generic companies holding FDA approvals to manufacture erythromycin products during the relevant time period included Chelsea, Mylan, Zenith, and Purepac. [11] Such companies include the Lederle Laboratories division of American Cynamid, Rugby, Goldline, and Geneva Generics. [12] For example, Abbott lost its ethylsuccinate business with Oseo and Rite-Aid, both parties to global contracts the previous year, to aggressive pricing by Mylan. [13] These figures are computed by taking the amount of IMS Drugstore purchases for each class of drug appearing on Defendant’s Exhibit 302B and dividing that number by the amount of IMS Drugstore units on Defendant’s Exhibit 302C. [14] Specifically, between July 1986 and June 1987, Barr sold 36,673 units to Eckerd, 10,780 units to Peoples, 19,176 units to Reveo, and 2,172 units to Peyton. One unit is the equivalent of a one hundred tablet bottle. [15] This section provides in pertinent part that: It shall be unlawful for any person engaged in commerce, in the course of such commerce, either directly or indirectly, to discriminate in price between different purchasers of commodities of like grade and quality, where either or any of the purchases involved in such discrimination are in commerce, where such commodities are sold for use, consumption, or resale within the United States ..., and where the effect of such discrimination may be substantially to lessen competition or tend to create a monopoly in any line of commerce, or to injure, destroy, or prevent competition with any person who either grants or knowingly receives the benefit of such discrimination, or with customers of either of them____ 15 U.S.C.A. § 13(a) (West 1973). [16] The present case involves primary-line discrimination because Abbott’s allegedly wrongful behavior is said to pose a threat of injury to Barr, one of its competitors. See E. Thomas Sullivan, Understanding Antitrust and Its Economic Implications, at 308 (1988) (primary-line cases concern discrimination which may result in injury to seller’s competitors). Secondary-line cases, in contrast, involve discrimination which affects competition among customers of the discriminating seller. Id. [17] O. Hommel focused on whether below-cost pricing is to be measured by reference to either average total cost or average variable cost. " ‘Average cost is the’ sum of fixed cost and total variable cost divided by output.’ ” O. Hommel, 659 F.2d at 349 n. 11 (quoting Areeda and Turner, Predatory Pricing and Related Practices Under Section 2 of the Sherman Act, 88 Harv. L.Rev. 697, 700 (1975)). Variable costs include items which vary with changes in output, such as materials, fuel, labor directly used to produce the product, and repair and maintenance. Average variable cost is the sum of all variable costs divided by output. By definition, it is lower than average cost. Id. Here, the parties have stipulated that Abbott’s prices have not been below cost, whatever definition of cost is used. [18] We do not think this conclusion is inconsistent with Utah Pie Co. v. Continental Baking, 386 U.S. 685, 87 S.Ct. 1326, 18 L.Ed.2d 406 (1967), a case involving a geographic price discrimination claim in the frozen pie product market. The district court had found insufficient evidence to support the jury’s verdict of discriminatory pricing, and the court of appeals had affirmed. The Supreme Court reversed. Not only had the plaintiff presented evidence of predatory intent with respect to each respondent, but it had also shown undisputed evidence of below-cost pricing by all three respondents. The court also found that the drastically declining price structure constituted actual competitive harm. Id. at 702-03, 87 S.Ct. at 1336. Therefore, the jury could rationally find the requisite injury to competition. Id. [19] In a competitive market, unit prices tend to stabilize at the marginal cost of the most efficient producers. See generally A. Alchian & W. Allen, Exchange & Production: Competition, Coordination, & Control, at 141-49 (1983) (discussing achievement of production efficiency by equalization of marginal costs among competitors in specialized market). Marginal cost is the *109 increase in total cost between outputs differing by one unit of output. Id. at 139. Were Abbott to raise its prices to independent retail pharmacies and pharmaceutical companies in order to subsidize below-cost pricing to warehouse chains Abbott's competitors could respond by offering lower prices to those customers against whom Abbott had discriminated, provided only that those lower prices were less than Abbott’s competitors’ own marginal variable costs. [20] In Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 97 S.Ct. 690, 50 L.Ed.2d 701 (1977), the Supreme Court did, however, caution us that a plaintiff like Barr seeking treble damages pursuant to section 4 of the Clayton Act need not necessarily prove an actual lessening of competition in order to recover. This section provides in pertinent part: [A]ny person who shall be injured in his business or property by reason of anything forbidden in the antitrust laws may sue therefor ... and shall recover threefold the damages by him sustained, and the cost of suit, including a reasonable attorney’s fee. 15 U.S.C.A. § 15(a) (West Supp.1992). Although the short-term effect of certain anti-competitive behavior may be to stimulate price competition, *110 competitors may be able to prove antitrust injury before they are actually driven from the market and competition is thereby lessened. Id., 429 U.S. at 490 n. 14, 97 S.Ct. at 698 n. 14. Our conclusion in the present case is not at odds with this observation. It simply requires a plaintiff to prove more than just a competitor’s predatory intent in order to recover under section 2. This additional proof may be in the form of below-cost pricing, actual anti-competitive effects, or both. [21] Because we affirm the grant of summary judgment on this ground, we need not address the district court's alternative ground for granting summary judgment under the "good faith meeting competition defense” of section 2(b). [22] This section provides in pertinent part: It shall be unlawful for any person engaged in commerce, in the course of such commerce, to ... make a sale or contract for sale of goods ... for use, consumption, or resale within the United States ... or fix a price charged therefor, or discount from .., such price, on the condition, agreement, or understanding that the ... purchaser thereof shall not use or deal in the goods ... of a competitor or competitors of the ... seller, where the effect of such ... sale, or contract for sale or such condition, agreement, or understanding may be to substantially lessen competition or tend to create a monopoly in any line of commerce. 15 U.S.C.A. § 14 (West 1973). [23] This section provides in pertinent part: Every contract, combination in the form of trust or otherwise, or conspiracy, in restraint of trade or commerce among the several States, or with foreign nations, is declared to be illegal____ 15 U.S.C.A. § 1 (West Supp.1992). [24] As a preliminary matter, we note that it is not entirely clear that the global contracts were indeed exclusive dealing agreements. An agreement affecting less than all purchases does not amount to true exclusive dealing. Kellam Energy, Inc. v. Duncan, 668 F.Supp. 861, 883-84 (D.Del.1987). Although the global contracts required the participating chain “to use listed products wherever legally permissible to do so,” warehouse chains with outlets in mandatory substitution states had to buy from other manufacturers in order to comply with state law. Jt.App. at 1440. By necessity, the chains also carried products of other manufacturers in order to fill prescriptions precluding substitution. For purposes of analysis, we will, however, assume that the global contracts fall within the antitrust definition of exclusive dealing contracts. [25] Both Abbott and the district court assumed the accuracy of this foreclosure rate for purposes of evaluating Barr’s exclusive dealing claim. [26] This section provides in pertinent part that: Every person who shall monopolize, or attempt to monopolize, or combine or conspire with any other person or persons, to monopo *112 lize any part of the trade or commerce among the several States, or with foreign nations, shall be deemed guilty of a felony____ 15 U.S.C.A. § 2 (West Supp.1992). [27] Barr argues that in 1984, there were only 14 manufacturers in the relevant market. Brief for Appellant at 33. We fail to see, however, how the evidence to which Barr points in the record creates an issue of fact on this topic. See Jt.App. at 2005-10. [28] Barr does not contest the sufficiency of the evidence to support the, jury’s finding on relevant market.