Full opinion text
OPINION HIGGINBOTHAM, A. L., Jr., District Judge. I. INTRODUCTION The plaintiff, SmithKline Corporation (“SmithKline”), instituted this antitrust action against the defendant, Eli Lilly and Company (“Lilly”), for purported violations of sections one, two and three of the Sherman Act, as amended, 15 U.S.C. §§ l, 2, and 3, and for an alleged violation of section three of the Clayton Act, 15 U.S.C. § 14. The plaintiff and defendant corporations are major manufacturers of prescription pharmaceuticals, engaged in both interstate and foreign commerce. This complaint was occasioned by the defendant’s marketing practices in the sale of certain pharmaceutical products, cephalosporins. More specifically, plaintiff claims that a marketing scheme Lilly created in April, 1975, known as the Revised Cephalosporin Savings Plan (“Revised CSP”), violated the antitrust laws. Under the Revised CSP, participating hospitals were eligible for two rebates: (1) a rebate based on the total volume of a hospital’s purchases of Lilly cephalosporins, the “base dividend”; and (2) a 3% rebate conditioned on the purchase of certain minimum quantities of each of any three of Lilly’s five cephalosporin products, the “bonus rebate”. The minimum quantity which had to be purchased in order to qualify for Lilly’s bonus rebate was separately calculated for each hospital. SmithKline brought this private antitrust action alleging that the Revised .CSP is an unlawful tying device in violation of sections one and three of the Sherman Act and section three of the Clayton Act. Furthermore, the plaintiff contends: (1) that Lilly has monopoly power; and (2) that the Revised CSP is a device designed to unlawfully foreclose competition or exclude competitors from the United States nonprofit hospital market in cephalosporins and, thus, enables Lilly to commit the offense of monopolization in violation of section two of the Sherman Act. Finally, SmithKline avers that the Revised CSP is a technique for the abuse and misuse of certain Lilly cephalosporin patents, in abrogation of sections one, two and three of the Sherman Act. Plaintiff, on May 14, 1975, filed a motion for a preliminary injunction; pursuant to conferences with the Court and a stipulation by the parties it was determined that a hearing on a final injunction would be held promptly. [See Document Nos. 11, 12, 15 and 41.] After extensive discovery and numerous pre-trial conferences, a non-jury hearing on liability commenced on November 24, 1975 and ended on January 6, 1976. Counsel delivered their closing arguments on March 19, 1976. The pretrial conduct of this matter was a model of cooperation among counsel and, once again, was a reminder of the ease with which a complex case can be effectively presented without undue antagonism or histrionics among counsel. Cephalosporins are extraordinary semi-synthetic, antibacterial agents which on certain occasions can save the lives of the ill or reduce significantly extraordinary suffering. Neither plaintiff nor defendant disputes the significance of the pharmaceutical breakthrough caused by cephalosporins. Lilly persuaded the medical profession to purchase more than $519,730,000 of its cephalosporins from 1970 through the first quarter of 1975. [Finding of Fact ¶ 96.] At issue here is not solely the efficacy of the products, but also the appropriateness of Lilly’s merchandising scheme — the Revised CSP. While claiming to better the medical condition of the seriously ill, has Lilly impermissibly sought to mortally wound SmithKline, so that it would no longer be the only viable competitor in the cephalosporin field? In the injury which it imposes on SmithKline by the Revised CSP, Lilly has clearly overstepped the boundaries of restraint required by the antitrust laws. After a most careful consideration of the detailed record and the parties’ respective briefs and proposed findings of fact, I find, for the reasons noted below, that since April 1, 1975, the date of the institution of the Revised CSP, that Lilly has monopolized the nonprofit hospital market for cephalosporins, in violation of section two of the Sherman Act; consequently, the plaintiff is entitled to final injunctive relief. I further find that SmithKline has failed to prove that the Revised CSP constitutes an illegal tying arrangement and, thus, Lilly has violated neither sections one and three of the Sherman Act nor section three of the Clayton Act. There is no need to separately consider SmithKline’s averment of patent misuse. This entire opinion, including the legal discussion, constitutes my Findings of Fact and Conclusions of Law, and any proposed findings of fact and conclusions of law inconsistent with those here found are hereby rejected in accordance with Rule 52 of the Federal Rules of Civil Procedure. II. HISTORY OF THE CEPHALOSPO-RIN MARKET While the Findings of Fact note with greater specificity the issues and relevant data, the following history is a brief synopsis of the development of the cephalosporin industry and the relevant marketing practices of the parties. In 1964 cephalosporins first became available for use in United States hospitals when Lilly introduced cephalothin, under the Lilly brand name Keflin, into the United.States market. Subsequently Lilly marketed three additional cephalosporins: cephalexin (Keflex), cephaloridine (Loridine), and cephaloglycin (Kafoein), all of which, including Keflin, are covered by the United States patents owned by Lilly under which it has exclusive rights. Lilly was the sole United States supplier of cephalosporins until October, 1973 when SmithKline brought yet another cephalosporin, cefazolin (Ancef), into the drug market. SmithKline markets cefazolin under a non-exclusive license obtained from a United States patent owner, Fujisawa, a Japanese pharmaceutical company. In November, 1973, Lilly,' also under a non-exclusive license, began marketing cefazolin under the brand name Kefzol. Later, Bristol and Squibb also began selling cephalosporins. Cephalosporins are distributed by both parties through independent wholesalers, who, then, sell the products to the hospitals. Lilly and SmithKline promote their respective products through sales representatives (“detail men”), who consult with both physicians and pharmacists within the hospitals. Cephalosporins are available for administration by a physician in a given hospital, in most instances, only if they are included within that hospital’s formulary — a list of drugs approved and available for use in that institution. Drugs are listed on the formulary as a result of: (1) a recommendation that a drug be so included by a physician affiliated with the hospital; (2) the review and approval or rejection of that recommendation by a committee composed of representatives from a hospital’s staff of physicians, nurses and pharmacists — the Pharmacy and Therapeutics Committee (“P & T Committee”); and (3) the P & T Committee’s independent recommendation that a certain drug be included in or deleted from the hospital’s formulary. The P & T Committee ultimately determines which drugs shall be listed on the formulary. The hospital pharmacists generally purchase drugs for the hospital. On occasion, hospital pharmacists from several institutions work in a collective purchasing group in order to secure bids from drug manufacturers. Lilly, in selling its. cephalosporin products, has varied its marketing approach. In’ October, 1972 Lilly instituted a marketing program entitled the Cephalosporin Savings Plan (“CSP”), a volume rebate scheme available to participating hospitals. A participating hospital could receive a percentage rebate based on its total Lilly cephalosporin purchases, the rebate to be paid in the form of Lilly merchandise. After the introduction of SmithKline’s Ancef in October, 1973 and Lilly’s Kefzol in November of that same year, Lilly altered its marketing strategy! The CSP was expanded to include Kefzol within the volume rebate scheme. In April, 1975 Lilly instituted the aforementioned Revised CSP. SmithKline, also, has offered hospitals several different rebate programs over the past few years. Its initial approach, was the Pricing Insurance Plan (“PIP”), adopted in response to Lilly’s inclusion of Kefzol in the CSP, which provided that participating hospitals could receive up to a five percent rebate on Ancef purchases. Furthermore, hospitals were eligible for an additional five percent rebate on each individual order for five hundred vials or more of Ancef. Later, PIP, like the CSP, was revised to grant a third rebate equal to five percent of a hospital’s Anspor purchases, if the hospital’s combined volume of Ancef-Anspor purchases equaled or exceeded five hundred grams per quarter. SmithKline changed its marketing scheme again in April, 1975 after Lilly instituted the Revised CSP. The plaintiff eliminated its rebate on combined Ancef-Anspor purchases. Instead, hospitals qualified for rebates in the following manner: (1) a five percent rebate was returned on any Ancef purchases per quarter; (2) in addition, a five percent rebate was paid for any individual orders of Ancef of five hundred or more vials per quarter; and (3) a third five percent rebate was available for Anspor purchases of five hundred grams or more per quarter. III. FINDINGS OF FACT: Some record references are given to substantiate many of the findings of fact appearing herein. However, some of these findings are predicated on the cumulative facts and inferences from the testimony, and facts which are further documented on numerous other pages of the record which are not cited. I recognize that many, if not most, judges make no page references in support of their general findings. See, e. g., United States v. International Boxing Club of N. Y., 150 F.Supp. 397, 401-419 (S.D.N.Y.1957) aff’d 358 U.S. 242, 79 S.Ct. 245, 3 L.Ed.2d 270 (1959); United States v. Brown Shoe Company, 179 F.Supp. 721 (E.D.Mo.1959), aff’d 370 U.S. 294, 82 S.Ct. 1502, 8 L.Ed.2d 510 (1962). Thus, these, record references are supplemental, but not exclusive. A. THE PARTIES AND JURISDICTION 1. Plaintiff SmithKline Corporation (“SmithKline”) is a corporation organized and existing under the laws of the Commonwealth of Pennsylvania with' its principal place of business, in Philadelphia, Pennsylvania. [Stipulation 1.1.] 2. SmithKline manufacturers, among other products, human ethical pharmaceutical products which it sells in interstate and foreign commerce. SmithKline markets its human ethical pharmaceutical products through independent wholesalers, who in turn sell the products to hospitals. [Stipulation 1.2; van Roden, Tr. 14, 36.] 3. In 1974, worldwide sales of Smith-Kline and French Laboratories, the division of Smith-Kline that conducts its human ethical pharmaceutical business, exceeded $250,000,000, of which 65% to 70% were sales in the United States, [van Roden, Tr. 16.] 4. Eli Lilly and Company (“Lilly”) is a corporation organized and existing under the laws of the State of Indiana with its principal place of business in Indianapolis, Indiana. [Stipulation 1.4.] 5. Lilly manufactures, among other products, human ethical pharmaceutical products which it sells in interstate and foreign commerce. In 1974, Lilly’s worldwide sales of human ethical pharmaceutical products exceeded $500 million. [Stipulation 1.5; P-33.] 6. Lilly markets its human ethical pharmaceutical products through independent wholesalers to whom it sells the products, who in turn sell the products to others, including hospitals, [van Roden, Tr. 36; Lange, Tr. 1117-1118.] 7. U.S. general hospitals in 1973 and 1974 purchased more Lilly pharmaceutical products (in terms of dollars) than any other manufacturer’s and at least four times as much Lilly pharmaceuticals as SmithKline pharmaceuticals. [Exhibit P-1, P-30.] 8. Lilly was the largest supplier of pharmaceuticals to wholesalers in 1973 and 1974. [Exhibit P-2.] 9. Jurisdiction of this subject matter duly appears pursuant to 15 U.S.C. §§ 1, 2, 14, 15 and 26. [See Complaint in C.A. 75-1102, ¶¶ 1 and 15.] 10. This matter comes before the Court on a motion for final injunction submitted by plaintiff in this action. B. THE HUMAN ETHICAL PHARMACEUTICAL INDUSTRY 11. Human ethical pharmaceutical products are drugs that are promoted to the medical profession and, generally, can be utilized only on the prescription of a licensed physician and dispensed by a licensed physician or by a licensed pharmacist. [van Roden, Tr. 17-18.] 12. Competition in the human ethical pharmaceutical drug industry generally is intense and is manifested in many forms, including: price competition; innovation, i. e., the invention or discovery of new or improved drugs; marketing efforts, e. g., selling, promotion, and advertising of drugs; establishing and maintaining the reputation of the manufacturer; producing a product of consistent quality; and providing consistent and needed service to all members of the health care delivery team. This finding is in reference to the industry generally and does not precisely describe the operation of the submarket, cephalosporins, as it functions pursuant to Lilly’s Revised CSP. [Step, Tr. 1028-1029.] 13. Research and development leading to the invention and marketing of new or improved products is an important form of competition since a company that introduces a product to the market first, particularly a new anti-infective (antibiotic), is able to establish a position of reputation and loyalty in the medical community. [Step, Tr. 1028-2029.] 14. A major objective of SmithKline is the discovery and development of new and unique drugs. SmithKline spends about $35 million per year in the United States on research in human ethical pharmaceuticals, a majority of which is spent in research for discovery of new chemical entities, [van Roden, Tr. 17.] 15. A major objective of Lilly is the discovery and development of new and unique drugs. It invests over $100,000,000 per year, which is approximately 9 to 10% of its total annual sales revenue, in research and development. [Step, Tr. 1035-1037.] 16. The marketing of a new drug, a competitively important factor, consists of identification of the capabilities of the drug, and utilization of the marketing firm’s abilities to promote and sell the drug. [Step, Tr. 1031.] 17. The most effective form of marketing human ethical pharmaceutical drugs is personal promotion to the medical profession through sales representatives (“detail men”), [van Roden, Tr. 18-19, 30-32, 57-58; Step, Dep. 33-34.] 18. An important competitive asset in the human ethical pharmaceutical industry is the strength and experience of a manufacturer’s sales representatives, [van Roden, Tr. 18-19; Step., Tr. 1034.] 19. SmithKline has approximately 510 sales representatives, of whom 3% are registered pharmacists. [SmithKline response'to Lilly Interrogatory No. 65.] 20. Among U.S. pharmaceutical manufacturers, Lilly has the largest number of sales representatives, nearly 1,200, 75% of whom are graduate registered pharmacists. [Step, Tr. 1034-1035.] C. THE RELEVANT MARKET 21. The Relevant Market is the nonprofit hospital market for cephalosporin drugs. [Findings of Fact ¶ 25-51, infra.] 22. The Relevant Geographic Market is the United States. 23. Cross elasticity of demand and price sensitivity do not exist, to any significant degree, between the cephalosporins and other antibiotic or anti-infective drugs. [Findings of Fact, ¶¶ 33 — 48, infra.] 24. Without Lilly’s Revised Cephalosporin Savings Plan, cross elasticity of demand and price sensitivity would exist among.the cephalosporins. [Findings of Fact ¶¶ 35, 48a-5l, infra.] 25. There is a sufficient disparity between cephalosporins on the one hand and all antibiotics (anti-infectives) on the other to distinguish the former from the latter. [Findings of Fact, ¶¶ 35-37, infra.] 26. Human ethical pharmaceutical products include among their number antibiotic drugs (anti-infectives). Antibiotics are chemical substances which are produced by microorganisms and are active against other microorganisms (bacteria). The major properties of an antibiotic are its spectrum of activity (i. e. which kinds of bacteria does it inhibit and which are resistent to it); its pharmacologic properties (e. g. how it is absorbed and excreted); its toxicities (side effects); and its allergenicity (i. e. how common are allergic reactions). [Exhibit P-178.] 27. Antibiotics include, among others: ampicillins, carbenicillins, cephalosporins, chloramphenicol, erythromycins, aminoglycosides, gentamycins, nitrofurantoins, penicillins, semisynthetic penicillins and tetracyclines. Such drugs are available in both parenteral (injectable, whether intravenous or intramuscular) and oral forms. [D-2082; D-355; D-2116; van Roden, Tr. 17-18.] i. HUMAN ETHICAL PHARMACEUTICALS PURCHASED FOR USE IN HOSPITALS 28. Hospitals generally serve bed-ridden, seriously ill patients while retail pharmacies serve ambulatory, less seriously ill patients. [P-189 at 2-3.] 29. In treating hospitalized patients, who generally have serious or life-threatening infections, physicians usually prescribe an injectable antibiotic, i. e. one that is administered intravenously or intramuscularly, sometimes following up with an oral antibiotic when the infection subsides. For outpatients they usually prescribe oral antibiotics; by contrast, retail pharmacies deal almost exclusively in orally administered forms of drugs. The market in this case is limited solely to the sale of drugs for use in hospitals. [First two sentences admitted by Lilly in its proposed final pretrial order; Ex. P-178, Ex. P-15; P-188 at 3-4 and Figs. 1-4.] 30. The Pharmacy and Therapeutics Committee (“P & T Committee”) of a hospital, made up primarily of staff physicians and pharmacists as well as nurses, determines the drugs which will be used at the hospital and listed on its formulary, which is a list of drugs approved for use at that hospital. Drugs are listed on the formulary as a result of: (1) a recommendation that a drug be so included by a physician affiliated with the hospital; (2) the review and approval or rejection of that recommendation by the P & T Committee; and (3) the P & T Committee’s independent recommendation that a certain drug be included in or deleted from the hospital’s formulary. The hospital pharmacist generally purchases drugs for the hospital. He purchases on the basis of past usage (prescriptions written by that hospital’s physicians). In marketing a drug to a hospital, a manufacturer seeks first to have the drug included in the hospital formulary and, second, to have the drug prescribed by physicians. [DiMatteo, Tr. 292-293, 327-332; Nudelman, Tr. 641-643, 651-659; P-105 at 9,11-12; P-114 at 58, 61-62; P-118 at 7; P-129 at 8-9, 12-15; P-135 at 6; P-144 at 5-6; P-154 at 6-7, 10; P-164 at 6, 8,11; van Roden, Tr. 30-32; Step, Tr. 1032-1033.] 31. Generic equivalents are drugs that have identical chemical structures and can be used interchangeably to treat the same conditions in a patient. [Shotwell, Tr. 264; DiMatteo, Tr. 293.] 32. Therapeutic equivalents are drugs that do not have identical chemical structures but give the same clinical response in treating a particular illness in a patient. [DiMatteo, Tr. 293.] ii. LACK OP INTERCHANGEABILITY BETWEEN CEPHALOSPORINS AND OTHER ANTIBIOTICS 33. Cephalosporins are “. . . semisynthetic antibacterial agents that are closely related chemically to the penicillins and, like them contain a beta lactam ring as part of the nucleus. They are produced by the addition of substituent groups to 7-aminocephalo-sporanic acid, a chemical nucleus obtained from cephalosporin C, which is elaborated by the fungus Cephalosporium. The cephalosporins interfere with the synthesis of the bacterial cell wall by inactivating a transpeptidase, thereby preventing cross-linkage of peptidoglycan chains.” [A.M.A. Drug Evaluations 523 (2d Ed. 1973).] 34. Antibiotics such as cephalosporins can be lawfully dispensed to a hospitalized patient only on the prescription of a licensed physician. Assuming that a physician prescribes an antibiotic that is approved for use in the hospital (i. e., listed in the hospital formulary), the hospital pharmacist must fill the prescription with the prescribed drug or its generic equivalent, i. e., a drug having an identical chemical, structure. Thus, a prescription for cefazolin must be filled with either Ancef or Kefzol. A prescription for a cephalosporin cannot be filled with a non-cephalosporin, such as penicillin, ampicillin or tetracycline. Thus, the hospital physician population, in practice, does not view other antibiotics as reasonably interchangeable with the cephalosporins. [DiMatteo, Tr. 296-297; Step deposition at 48-49; Schiefe, Tr. 1011-1013; P-164 at 8-9, 14-15; P-144 at 8,16, 49-50; P-129 at 23-24; P-118 at 16-17; P-135 at 10-12; P-154 at 12; P-105 at 12; P-114 at 65-66.] iii. SPECIAL CHARACTERISTICS OF CEPHALOSPORINS 35. The cephalosporin family of antibiotics all have, for practical purposes, an identical spectrum of activity and, all except Loridine, are among the safest of antibiotics, with very little toxicity and allergenicity. They differ significantly only in pharmacological properties and toxicity and are therefore generally interchangeable for treatment of the same conditions. Given this interchangeability, Loridine should be included within the same family of drugs and the same product market. Despite its relatively higher toxicity, Loridine is still less toxic than some other antibiotic drugs —e. g. the aminoglycosides. [Ex. P-11; P-12; P-35; Nudelman, Tr. 707; Kass, Tr. 828-829; Holloway, Tr. 856-857; Schiefe, Tr. 1007-1008; Ex. D-2096; Ex. P-178.] 36. There is a certain degree of interchangeability among all antibiotic drugs. Cephalosporins, penicillins and the aminoglycosides kill rather than merely inhibit the growth of bacteria, contrary to the effect of the erythromycins, chlorampenicol and the tetracyclines. However, there are noticeable and acknowledged differences in the relative effectiveness of cephalosporins as compared with other antibiotics in the treatment of certain illnesses. The cephalosporins are far less toxic than the aminoglycosides; this reduced toxicity is a characteristic shared by the penicillins. However, the cephalosporins are: a. Effective in treating Klebsiella — no penicillin is capable of the same activity. b. Active against both staphylococci and gram negative bacilli. Most staphylococci are resistant to penicillin G, ampicillin and carbenicillin, which have a good gram negative bacillus spectrum, whereas methicillin-like penicillins are active against staphylococci but lack gram negative bacillus activity. Cephalosporins thus provide a broader spectrum of activity. c. Although there is some cross-allergenicity between penicillins and cephalosporins, it is by no means complete. Most experts in Infectious Diseases will use cephalosporins in serious infections in penicillin-allergic patients who require therapy with a penicillin or a cephalosporin. Therefore, while the cephalosporins and other antibiotics are equally effective in treating some illnesses, for other illnesses there is no equal interchangeability. Regardless of some overlapping, the cephalosporins have sufficient peculiar characteristics and uses to make them a distinguishable product market. [Ex. P-178, P-20; P-21; P-24; P-25; P-26; P-27; P-31; P-34; Holloway, Tr. 844-846.] 37. The plaintiff and defendant recognized the special attractiveness and peculiar qualities of a family of antibiotics effective in treating both gram positive and gram negative infections. Cephalosporins demonstrated the above capabilities, and were promoted as unique drugs. [Exhibit D-102A at 9; P-20.] iv. LACK OF PRICE SENSITIVITY BETWEEN CEPHALOSPORINS AND OTHER ANTIBIOTICS 38. The cost of drugs is a relatively small percent of the overall daily cost of therapy for a hospitalized patient. [Step deposition at 48; Kass, Tr. 826.] 39. In selecting an antibiotic to prescribe for a hospitalized patient, the two properties given most weight by physicians are efficacy and safety. In comparison with these, cost is an insignificant factor. [Exhibit P-38; Step Dep. at 47-48; Holloway, Tr. 846, 861; Schiefe, Tr. 989.] 40. Cephalosporins are very expensive compared to other antibiotics in terms of cost per patient for a daily dose. Injectable cephalosporins cost several times as much as injectable penicillin G, a widely used antibiotic which is often compared to the cephalosporins. [Last sentenced admitted by Lilly in its proposed pretrial order; Ex. P-170, Tables II and IV; Chappell, Tr. 111-114; Ex. P-38; Step deposition at 45-47; P-164 at 14; P-129 at 21-23; P-144 at 15-16; P-135 at 17; P-154 at 11-12; P-105 at 50-51; P-114 at 65; D-2129.] 41. The average cost per day of therapy for the cephalosporins, as compared with other antibiotic drugs, is as follows: GOST/DAY OF MAJOR ANTI-INFECTIVES Keflin PolyLoridine cillin-N Kefzol Ancef 4.0 gm 2.0 gm 2.5 gm 2.0 gm 2.0 gm /day /day /day /day ' /day 1965 $17.76 1966 12.96 11.40 1967 12.96 11.40 1968 12.96 8.40 1969 12.96 7.80 8.40 1970 12.96 7.80 7.25 1971 12.96 7.80 6.48 1972 11.52 7.28 6.48 1973 11.24 7.28 3.25 9.68 . 9.88 1974 11.24 7.14 3.25 9.60 9.68 Chloromycetin Kantrex Garamvcin Colymvcin Terramvcin Bi-Cillin ck 1.200 MU 4.0 gm 1.0 gm 240 mg 300 mg 2.0 gm 2.4 mu /day /day /day /day /day _¿day_ 1965 $11.12 4.08 $12.60 $ 8.16 $ 3.22 1966 11.12 4.08 12.60 8.16 3.22 1967 10.80 '4.08 12.60 8.16 3.58 1968 9.60 4.08 12.60 8.16 3.74 1969 10.56 4.08 12.60 8.16 3.74 1970 10.56 4.08 21.00 12.60 8.16 3.94 1971 11.64 4.37 17.28 12.60 8.16 3.94 1972 11.64 4.80 17.28 14.50 8.96 3.94 1973 12.68 4.80 13.77 14.50 11.12 4.18 1974 13.68 4.80 13.17 14.50 9.84 4.18 Exhibit D-2129 42. During the first quarter of 1975, more than 95% of the general hospitals purchased cephalosporins and more than 89% purchased injectable cephalosporins. There is probably no short-term general hospital, and certainly no significant number of short-term general hospitals, that does not purchase and stock cephalosporins. [Admitted by Lilly in its proposed final pretrial order; Ex. P-170, Tables la, lb, Via; Chappell, Tr. 109-110, 121-124; Step deposition at 64.] 43. There are about 6,000 non-profit general hospitals in the United States serving more than 100,000 prescribing physicians. During the twelve-month period ending August 1974, the total number of “patient starts” of cephalosporins in hospitals (meaning the number of times cephalosporin treatment was initiated) was more than 2 million. [Chappell, Tr. 105-106; Gootee, Tr. 1210-1211; Shotwell, Tr. 789, 794.] 44. There is a particular demand by doctors for the peculiar qualities of cephalosporins in contradistinction to other antibiotics; that demand is reflected in hospital purchases of cephalosporins. In 1974, U.S. hospitals purchased more cephalosporins (in terms of dollars) than any other antibiotic. Their total purchases exceeded $120 million of which more than 75% were injectables. [Second and third sentences admitted by Lilly in its proposed pretrial order; Exhibit P-4; Nudelman, Tr. 705-706.] 45. Hospital purchases of cephalosporins since 1970 have been as follows: 1970 1971 1972 Volume Share Volume Share Volume Share $ % $ % $ % Total Cephalosporins ** 67.325 100.0 81.239 100.0 98.520 100.0 Lilly 67.325 100.0 81.239 100.0 98,520 100,0 Keflin (9/64) 40,693 60.4 51,062 62.9 62,796 63.7 Keflex (2/71) — — 11,239 13.8 20,752 21.1 Kefzol (11/73) Keflin Neutral (5/75) Loridine (3/68) 25,622 38.1 17,916 22.1 ■ 14,607 14.8 Kafocin (7/70) 994 1.5 1,016 1.3 356 0.4 Cephaloridine (9/68) 16 — 6 — 9 — 1970 1970 1970 Volume Share Volume Share Volume Share Bristol Cefadyl (5/74) SmithKline Ancef (10/73) Angpor (10/74) Squibb Velosef.(8/74) 1973 1974 1975 Volume Share Volume Share Volume Share Total Cephalosporins 105,405 100,0 123.771 100,0 65.007 100.0 Lilly 103,858 98.5 111,177 89.8 57,611 88.6 Keflin (9/64) 68,233 64.7 67,854 54.8 30,630 47.1 Keflex (2/71) 22,945 21.8 25,346 20.4 13,834 21.3 Kefzol (11/73) 1,149 1.4 13,593 11.0 8,355 12.9 Keflin Neutral (5/75) — — — — 5.1 Loridine (3/68) 10,996 10.4 4,322 3.5 1,430 2.2 Kafocin (7/70) 191 0.2 61 0.1 21 Cephaloridine (9/68) 14 — 1 — 1 Bristol Cefadyl (5/74) — — 1,865 1.5 1,766 2.7 SmithKline 1,547 1,5 10,425 8.5 5.292 8.1 Ancef (10/73) 1,547 1.5 10,355 8.4 4,988 7.7 Anspor (10/74) — — 70 0.1 304 0.5 Squibb Velosef (8/74) — — 304 0.3 338 0.5 P-170, Table VIII, at 36. 46. Changes in the relative amounts of cephalosporins and non-cephalosporins purchased by hospitals are not directly related to the relative costs thereof. During the period from 1966 through 1974, hospital purchases of cephalosporins increased by nearly 700%. In the same period hospital purchases of penicillin G decreased by nearly 60%. [Ex. P-170, Table III; Chappell, Tr. 113; P-38; P-154 at 66.] 47. A small number of pharmacist and physicians are encouraging the decreased usage of cephalosporins, in favor of other antibiotics, for reasons of economy. The vast majority of hospitals do not purchase their antibiotic requirements in accordance with the efforts of these persons. [Kass, Tr. 819-820, 825; Nudelman, Tr. 704; Holloway, Tr. 857-858.] 48. There has been no erosion in the price of Keflin in the last ten years. [Step, Tr. 1062.] v. PRICE SENSITIVITY AND INTERCHANGEABILITY OF USE AMONG CEPHALOSPORINS IN THE ABSENCE OF LILLY’S REVISED CSP 48. Many hospitals purchase their drug requirements for specific periods by letting formal bids and buying from the lowest bidder for the entire period (usually from one calendar quarter to one year in length). This is particularly true of an increasing number of hospitals who are forming “Buying groups” for the purpose of making such bids. [DiMatteo, Tr. 293-294, 298-301, SOS-SO?, 356-358; P-129 at 11; P-135 at 7-8; P-144 at 5-6; P-164 at 7-8.] 49. When there are two reputable manufacturers of the same generic pharmaceutical, as in the case of cefazolin, most hospitals can be expected to fill most, if not all, of their requirements with the brand which costs the least. This is particularly true of hospitals that buy such products on a bid basis. About 50% of hospital purchases of cefazolin are made on this basis. [DiMatteo, Tr. 293-296,301, 356-358; Step, deposition at 60; (pharmacist depositions); van Roden, Tr. 38-39; P-57; P-71; Nudelman, Tr. 653-655, 660, 706; Schiefe, Tr. 1013-1015; P-164 at 7-12, 24-25; P-144 at 8-9, 11, 32, 35; P-129 at 13-15, 42; P-135 at 12- 13, 47; P-154 at 8-10, 43, 63; P-105 at 13- 14, 34-35, 47; P-114 at 62, 82.] 50. Ancef and Kefzol are equivalent drugs, so that most hospitals stock only one brand. Thus, of hospitals purchasing cefazolin, about 75% purchase only one brand (either Ancef or Kefzol but not both). Of those purchasing both brands, about half purchase 70% or more of one of them. [Exhibit P-170, Table V; Chappell, Tr. 115-121; Exhibits P-57, P-71; DiMatteo, Tr. 294, 295a; Exhibits P-105 at 15, P-114 at 66; P-129 at 24, P-135 at 17, P-144 at 16; P-154 at 12; P-164 at 15.] 51. Dr. Weston’s economic theory of price responsiveness and cross-elasticity of demand among several hypothetical current and future generations of antibiotic drugs is rejected; the evidence does not establish that the relevant product market is all antibiotic or all anti-infective drugs. [Weston, Tr. 1624-1628; Exhibit P-188.] D. HISTORY OF CEPHALOSPORINS 52. The first cephalosporin became available for use in the treatment of patients with infectious diseases in United States hospitals when Lilly introduced cephalothin into the United States market in 1964. Lilly markets cephalothin under the brand name Keflin. Keflin is administered parenterally. [D-2000.] 53. In 1967, Lilly introduced another cephalosporin, cephaloridine, under the brand name Loridine. Loridine is administered parenterally. [D-2002.] 54. In 1971, Lilly introduced another cephalosporin, cephaloglycin, under the brand name Kafocin. Kafocin is administered orally. [D-2003.] 55. In 1972, Lilly introduced another cephalosporin, cephalexin, under the brand name Keflex. ■ Keflex is administered orally. [D-2004.] 56. In October, 1973, SmithKline introduced a cephalosporin, cefazolin, under the brand name Ancef. Ancef is the generic equivalent of Lilly’s Kefzol, and was introduced prior to Kefzol. Ancef is administered parenterally, both intramuscularly and intravenously, [van Roden, Tr. 23, 43; Stipulation 3.9.] 57. In November, 1973, Lilly began marketing another cephalosporin, cefazolin, under the brand name Kefzol. Kefzol is administered parenterally, both intramuscular and intravenously. [D-2001.] 58. In May, 1974, Bristol introduced cephapirin under the brand name Cefadyl. Cefadyl is a parenteral product. [D-1047.] 59. In August, 1974, Squibb introduced cephradine under the brand name Velocef. Velocef is marketed in oral and parenteral forms. [Exhibit D-2008.] 60. In November, 1974, SmithKline began marketing Anspor, the generic equivalent of Squibb’s Velocef, in oral form, [van Roden, Tr. 53-55.] 61. All Lilly cephalosporins except cefazolin are covered by U.S. patents under which Lilly has exclusive rights, and Lilly is the sole U.S. source of these products. Cefazolin is also covered by a U.S. patent or patents, under which both SmithKline and Lilly have non-exclusive licenses, as to each other, and these two companies are the only sources of cefazolin in the United States. [Admitted by Lilly in its proposed final pretrial order; Lilly’s Answer, para. 6-8; Lilly’s answers to plaintiff’s interrogatories 14-18; Hutchinson deposition at 21.] E. SMITHKLINE’S ENTRY INTO THE. MARKET 62. Historically, SmithKline’s prescription pharmaceuticals have not included antibiotics to any significant degree, [van Roden, Tr. 16.] 63. In the late 1950’s SmithKline made a substantial commitment of research and development resources toward antibiotic discovery, concentrating on the cephalosporin field. Accordingly, SmithKline in the late 1950’s and early 1960’s, did substantial research in semi-synthetic penicillins and cephalosporins. Its interest in cephalosporins goes back to 1959. [van Roden, Tr. 19, 22.] 64. In 1962 SmithKline negotiated with Lilly for the right to use Lilly’s patented process for producing a basic cephalosporin intermediate called 7ACA. Alternatively, SmithKline sought to buy 7ACA from Lilly. [Admitted by Lilly in its proposed final pretrial order.] 65. In the late 1960’s SmithKline’s interest revived. Management felt that cephalosporins would be the antibiotics of the 1970’s and that SmithKline, as a result of previous efforts, was in an excellent position to enter this market. It adopted a two-pronged approach: (1) a long-term research and development effort to develop new antibiotics with therapeutic advantages over those already on the market; (2) a licensing program to obtain immediate rights to products already developed by others.' [van Roden, Tr. 19, 22-23.] 66. SmithKline’s licensing program resulted in 1971 agreements with Fujisawa Pharmaceutical Company of Japan for cefazolin and with Squibb for cephradine. Squibb and SmithKline each had patents relating to certain aspects of cephradine. [Admitted by Lilly in its proposed final pretrial order; van Roden, Tr. 23.] 67. Cefazolin was originally synthesized by Fujisawa. Clearance for marketing in the United States resulted from Smith-Kline’s efforts with respect to Ancef. Lilly obtained clearance for Kefzol by “piggy backing” on SmithKline’s clinical studies. [First sentence admitted by Lilly in its proposed final pretrial order; van Roden, Tr. 23-28.] 68. SmithKline’s decision to enter the field of antibiotics with Ancef entailed a very substantial commitment. As of the present time, SmithKline’s investment in Ancef, including research, development, clinical testing, promotion, personnel and production facilities totals more than $20 million, [van Roden, Tr. 25-26, 66-67; Ex. P-5.] 69. In October 1973, when SmithKline began to market Ancef, SmithKline considered cefazolin (Ancef) a promising drug, much better than Keflin for IM, equally good for IV. It believed that the primary market for Ancef would be in hospitals. [Admitted by Lilly in its proposed final pretrial order; van Roden, Tr. 23-24, 27, 72.] 70. SmithKline knew that its cost of goods for Ancef would be substantially higher than Lilly’s cost for Kefzol. However, SmithKline management believed there would be a number of possibilities for lowering SmithKline's cost to a level competitive with Lilly’s, once SmithKline achieved a substantial level of sales. [Admitted by Lilly in its proposed final pretrial order.] 71. SmithKline would not have entered the hospital antibiotic market if its prospects had been limited to Ancef alone or even to Ancef and Anspor, because such a limited market entry would not have justified the required large investment in money and effort. In deciding to make antibiotics a major commitment, SmithKline viewed Ancef and Anspor as the first of a series of specialty antibiotics, in which SmithKline, through its research efforts, would play a substantial part as originator and/or developer. [van Roden, Tr. 22, 31-32.] 72. The names of the cephalosporins in current use in the United States are listed below. There is no generic equivalent for any of these cephalosporins except in the cases noted. The dates in parentheses, are the dates the particular cephalosporin was introduced in the market. Injectable Generic Name Brand Name cephalothin - (1964) Keflin (Lilly) cephalorldlne - (1967) Loridine (Lilly) cefazolin - (1973) Kefzol (Lilly) Ancef (SmithKIine) ** cephaplrin - (1974) Cefadyl (Bristol) cephradine - (1974) Velosef (Squibb) Oral cephalexin - (1972) Keflex (Lilly) cephaloglycin - (1971) Kafocln (Lilly) cephradine - (1974) Anspor (SmithKIine) ** Velosef (Squibb) F. COMPETITION BETWEEN SMITH-KLINE & LILLY — 1973 & 1974 73. When SmithKIine entered the market with Ancef, it felt that there was only a limited potential for overall expansion in injectable cephalosporins. Consequently, SmithKIine felt that Ancef would have to obtain its growth at the expense of Lilly’s established products, Loridine and Keflin. Because Keflin was used much more than Loridine (the use of which was declining), SmithKIine decided to position Ancef in competition with Keflin. Accordingly, SmithKIine decided to price Ancef about 5% below Keflin on a recommended daily dosage basis, [van Roden, Tr. 27-29, 34; Exhibits P-49, P-53.] 74. About one month after SmithKIine introduced Ancef, Lilly countered with Kefzol, which it priced 2% under Ancef. SmithKIine matched this price, increasing its spread below Keflin to 7%. [van Roden, Tr. 34, Lilly’s Answer to Plaintiff’s Interrogatory No. 7.] 75. SmithKline’s sales strategy was to emphasize that Ancef is equally efficacious and more convenient than Keflin as far as the doctor and the hospital staff are concerned, in addition to being lower in cost to the patient. Thus: —Since Ancef can be administered both IM and IV, whereas Keflin is rarely administered IM, there is no need to change drugs when the mode of administration is changed. —Because Ancef provides higher and more sustained blood levels, Ancef can be administered less often. —To the extent Ancef is administered IM rather than IV, there may be savings in hospital procedure costs. [van Roden, Tr. 23-24, 29; Exhibits P-13, P-14, P-44; Shotwell, Tr. 899-902, 905.] 76. Initially, SmithKIine concentrated its promotional efforts on large teaching hospitals which not only use cephalosporins in quantity, but also sometimes influence the doctors they train and, through their prestige, the medical profession generally, [van Roden, Tr. 29-30.] 77. Also, large hospitals strongly tend to use proportionately much more Keflin, so that Ancef’s positioning directly against Keflin would be directly related to 'such hospitals. [Lochridge, Tr. 387-390.] 78. Prior to the institution of the Revised CSP,- SmithKIine provided the only effective competition for Lilly in the cephalosporin market. By the end of 1974 plaintiff’s sales of Ancef were at an annual level in excess of $10 million and its share of cefazolin sales was in excess of 40%. However, at the end of 1974, SmithKIine commanded only 8.5% of the cephalosporin market; in contrast, Bristol had a 1.5% share of the cephalosporin market; Squibb enjoyed only 0.3% of all cephalosporin sales, [van Roden, Tr. 33-34; Exhibit P-170, Table VIb at 32, Table VIII at-36; Exhibit P-6; Exhibit P-53; Exhibit P-54.] 79. Initially, SmithKIine and Lilly set virtually the same net price to wholesalers. Lilly, in anticipation of competition from cephalosporins sold by others, had previously adopted, in October, 1972, a Cephalosporin Savings Plan (“CSP”), under which hospitals received cumulative graduated rebates (up to 12%) on their cumulative quarterly purchases of Lilly’s Keflin, Keflex, Kafocin and Loridine. As of November, 1973, Lilly expanded the CSP to include Kefzol. SmithKIine, in response to Lilly’s expanded CSP, countered with its Price Insurance Plan (“PIP”), a flat 5% rebate on total Ancef purchases plus an additional 5% for Ancef orders of 500 vials or more. After the introduction of Anspor into the cephalosporin market, and again in response to Lilly’s inclusion of Keflin, Keflex, Kafocin and Loridine for rebate purposes under the CSP, SmithKline expanded PIP to include Anspor. Hospitals were then eligible not only for the 5% rebate on their total volume of Ancef purchases, plus the additional 5% rebate on orders for 500 vials or more of Ancef, but also for a 5% rebate on Anspor purchases each quarter — provided that the hospital’s total purchases of SmithKline cephalosporins were equal to or greater than 500 grams per quarter. The rebates on Ancef were in no way conditioned on the purchase of Anspor, nor were hospitals required to make any minimum purchase of both Anspor and Ancef to qualify for the new rebate. Later, on bid business, SmithKline went to graduated rebates between 5% and 10%, and sometimes above 10%. SmithKline’s PIP rebates averaged about 7.5% of sales, [van Roden, Tr. 35-37; Exhibits P-28, P-39, P-40, P-41, P^8; Hutchinson deposition at 38-39; Step Deposition at 11-12; Lilly’s answers to plaintiff’s interrogatories No. 201, 202, 203; D-437; P-189.] 80. The original CSP provided for a rebate to be paid to participating not-for-profit hospitals, in the form of Lilly merchandise of the hospital’s choice (with certain very limited exceptions such as controlled substances), at an established rate based solely upon the total number of grams of Lilly cephalosporins purchased by the hospital. Rebates were payable quarterly, but the rate of rebate was also computed retroactively on an annualized basis in order to give each hospital the benefit of any higher rate of rebate it could earn on such a basis. [D-1010; P-76, Luedke deposition at 15-16; Southard deposition at 18-19.] 81. The rebate dividend levels under the original CSP were as follows: Dividend Quarterly Purchase Level (Total Grams') 2% 0-5,999 3% 6,000-8,999 4% 9,000-11,999, 5% 12,000-17,999 18.000- 23,999 7% 24.000- 29,999 8% 30.000- 38,999 9% 39.000- 47,999 10% 48.000- 71,999 11% 72.000- 95,999 12% 96.000- or more [D-1009.] 82. During 1974 SmithKline and Lilly reduced their prices in many instances to permit and encourage the wholesalers to offer lower prices to hospitals inviting bids or otherwise negotiating for the purchase of cefazolin. [Admitted by Lilly in its proposed final pretrial order; Ex. P-7; P-42; P-50.] 83. The competition between Smith-Kline and Lilly resulted in lower cefazolin’s cost to hospitals, particularly those purchasing on bids. There has been no comparable reduction in the price of Keflin. [First sentence admitted by Lilly in its proposed final pretrial order; Ex. P-7; P-50; Step deposition at 98-100; Lilly’s answers to plaintiff’s interrogatories No. 42 and .42(a); Step, Tr. 1062; Lange, Tr. 1122-1124.] G. LILLY’S ADOPTION OF THE REVISED CSP 84. Cephalosporins are particularly important to Lilly, accounting for 15% of the company’s total consolidated net sales in 1972 and 25% in 1974. The cephalosporins involved in this litigation (i. e., those sold to U.S. hospitals) accounted in 1974 for nearly 10% of Lilly’s sales and more than 15% of Lilly’s profits. [P-27; P-29; P-33; P-36; P-37; Responses No. 42, 43 and 44 to Lilly’s answers to SmithKline’s interrogatories; P-33 at 18; see “Memorandum to the Court explaining calculations made in Plaintiff’s Trial Memorandum,” at 2.] 85. When Lilly began to market Kefzol, it promoted Kefzol as a replacement for Loridine for IM use and continued to promote Keflin for IV use. This was in Lilly’s interest because Lilly had exclusive rights to Keflin (but not to Kefzol) and because Lilly’s profit margins were very much higher on Keflin. [Admitted by Lilly in its proposed final pretrial order, to the extent that Lilly promotes Kefzol for IM use and continues to promote Keflin for IV use; P-43; P-45; P-49; Step deposition at 11; Hutchinson deposition at 26-27, 34; Leudke deposition at 14-15; Lilly’s answers to plaintiff’s interrogatories No. 42, 43, 45, 60, 60(a); P-53; P-32; Step, Tr. 1056-1058, 1074.] 86. Lilly’s objective was to obtain 75% of cefazolin sales, which, when combined with sales of its exclusive cephalosporins, would maintain its share of the cephalosporin market at or above 90%. [Admitted by Lilly in its proposed final pretrial order to the extent that Lilly’s objective was to capture 75% of cefazolin sales; Ex. P-170, Tables VIb, VIII; Chappéll, Tr. 124-126, 131-133; P-29; P-45; P-50; Step deposition at 57; P-32; P-54.] 87. Despite Lilly’s earlier estimates that cefazolin would only replace Keflin for 15% of uses, early in 1974 Lilly found that Ancef and Kefzol were being used as a replacement for Keflin more than 60% of the time and that Lilly was- falling substantially short of its 75% goal in sales of cefazolin. [Admitted by Lilly in its proposed final pretrial order; P — 49; P-50; Leudke deposition at 11-12; P-43; P-52; P-53. 88. Thereafter, Lilly placed additional emphasis on efforts to avoid replacing Keflin with Kefzol and succeeded in cutting the percentage of cases in which Kefzol replaced Keflin to less than 50%. Meanwhile, the rate of replacement of Keflin by Ancef increased to more than 80%. [Admitted by Lilly in its proposed final pretrial order; Leudke deposition at 12-14; P-52; P-53.] 89. Also, in the second half of 1974, Lilly set up a Cephalosporin Task Force to devise a means of combatting SmithKline. The guidelines for 'the task force were that any new program must: (a) include as many as possible of the hospitals that were participating in the original CSP, (b) be more competitive price-wise than the original CSP, and (c) cost no more dollars than the original CSP to Lilly. [Second sentence admitted by Lilly in its proposed final pretrial order; Hutchinson deposition at 51; Southard deposition at 55; Leudke deposition at 31-32, 45-46; Lilly’s answer to plaintiff’s interrogatory No. 21(a); P-58; P-71.] 90. The task force recommended, and Lilly adopted, the Revised CSP which was put into effect by Lilly on April 1,1975, and provides for a rebate to be paid to participating hospitals, in the form of Lilly merchandise of the hospital’s choice (with the same limitations as under the original CSP), at an established rate (the “base dividend”) based upon the total number of grams of Lilly cephalosporins purchased by the hospital. The rates at which rebates are paid upon a given number of grams of Lilly cephalosporins purchased are somewhat lower, by_ approximately 3%, than the rates at which rebates were paid for the purchase of the same number óf grams under the original CSP. In addition, however, the Revised CSP provides for an additional 3% rebate (the “bonus dividend”) to be paid upon the hospital’s total cephalosporin purchases if the hospital buys established minimum quantities (separately established for each hospital) of each of any three of Lilly’s five cephalosporin products. [Admitted by Lilly in its proposed final pretrial order; P-73; P-78.] 91. The purchase levels to qualify for the base dividend, and the corresponding levels to qualify for the bonus dividend under the Revised CSP are as follows: Dividend Level Base Dividend Qtr. Purchase (Total Grams) Bonus Dividend Qtr. 3 Percent (Total Grams) 0% 0-7,999 150 1% 8,000-10,999 300 2% 11.000- 16,999 400 3% 17.000- 22,999 500 4% ■ 23.000- 29,999 750 5% 30.000- 38,999 1,000 39.000- 47,999 1,250 48.000- 71,999 1,500 72.000- 95,999 1,750 96.000- or more ■ 2,000 [Exhibit P-78.] 92. Lilly anticipated that virtually all hospitals would purcháse the specified minimum in Keflin and Keflex; that virtually none would purchase the minimum in Kafocin; and that only a handful would purchase the minimum in Loridine. This meant that the great bulk of hospitals, in accordance with the prescribing habits of their physicians and in order to qualify for the bonus rebate on Keflin and Keflex, would purchase Kefzol in the specified minimum amount. [See Finding of Fact ¶ 30; Lilly’s answer to plaintiff’s interrogatory No. 68; van Roden, Tr. 40; DiMatteo, Tr. 297-298,305; P-75; P-76; P-164 at 12-13, 20- 22, 27-28; P-144 at 22-23, 35, 47; P-129 at 29-33, 64; P-135 at 24, 36, 73-74; P-154 at 21-22, 30-31, 61-62; P-105 at 11, 21- 22; P-114 at 58-59, 70, 75, 88-89.] 93. Lilly knew, and instructed its salesmen to emphasize, that the entire bonus rebate (including the bonus rebate on Keflin and Keflex) should be considered as an inducement to buy Kefzol instead of Ancef. Lilly supplied its salesmen with calculations showing that to meet this inducement SmithKline would have to offer a rebate on Ancef in excess of 20%. [DiMatteo, Tr. 315; Ex. P-15, P-16, P-18, P-19, P-92; Leudke, deposition at 43; pharmacists’ depositions; P-72; P-95; P-96; P-97; P-98; P-99; P-100; P-164 at 39; P-144 at 29-31.] 94. Lilly hoped that the popularity of Keflin and Keflex would cause hospitals to opt to participate in the Revised CSP’s bonus rebate option. However, all of Lilly’s cephalosporin products, including its patented products Keflin and Keflex, are now and have been at all times separately available for purchase by any hospital in the United States, whether or not it participated in the original CSP or the Revised CSP. Both before and after the adoption of both the original CSP and the Revised CSP, some hospitals purchased Keflin and Keflex without participating in either plan. Hospitals always retained the option of losing the rebate and purchasing most of their cephalosporin requirements from other cephalosporin manufacturers. [See Plaintiff’s Exhibit P-90, at 2; Step, Tr. 1046-1047; Lange, Tr. 1114-1119; D-1143.] 95. The Revised CSP as adopted and as implemented does not contain any provision requiring that any hospital purchase any Lilly product in order to obtain any other Lilly product; nor does it contain any provision which requires any hospital to refrain from purchasing any product from any source in order to obtain any Lilly product, or which conditions the payment of any rebate on the purchase of any Lilly product, otherwise unavailable to the purchaser, upon an agreement that the hospital shall refrain from purchasing any product from any source. [Step, Tr. 1046; P-78; D-1100A.] 96. After Lilly instituted its Revised CSP, SmithKline, in an attempt to survive in the marketplace during the trial of this lawsuit, changed its PIP rebate program. Under this new arrangement the plaintiff eliminated its rebate on combined Ancef-Anspor purchases. Instead, hospitals qualified for rebates in the following manner: (1) a five percent rebate was returned on any Ancef purchases per quarter; (2) in addition, a five percent rebate was paid for any individual orders of Ancef of five hundred or more vials per quarter; and (3) a third five percent rebate was available for Anspor purchases of five hundred grams or more per quarter. Ancef and Anspor were not linked in any way through SmithKline’s revision of its PIP marketing scheme — rebates were available upon the separate, purchase of each of SmithKline’s products. [Plaintiff’s Exhibit P-189.] H. LILLY’S PRICING STRATEGY 97. Lilly has a unique position in the cephalosporin market. Until 1973, Lilly commanded 100% of the cephalosporin market. At that point in time, SmithKline introduced Ancef into the market, garnering 1.5% of the cephalosporin business; Lilly retained 98.5% of the market. As late as the end of 1974, Lilly continued to control 89.9% of the relevant product market, despite the fact that three other companies were marketing cephalosporin products. SmithKline, in that same year, secured only 8.5% of the gross sales. Lilly, from 1970 through the first quarter of 1975 has received $519,730,000 in gross sales of its cephalosporin products. By comparison, SmithKline has grossed $17,264,000 in sales of Ancef and Anspor, Squibb has received $642,000 on its sales of Velocef, and Bristol in marketing Cefadyl had gross sales of $3,631,000. [Exhibit P-170, Table VIII at 36; See Finding of Fact ¶ 56.] 98. Although the Revised CSP does not constitute an illegal tying arrangement, only 2% of Keflin purchases in the second quarter of 1975 were made outside of the Revised CSP. [Lange, Tr. 1118-1119.] 99. Lilly knew that cost is not an important factor in a physician’s choice of an antibiotic for a hospitalized patient. It calculated that even a 50% reduction in the price of Keflin would not greatly increase Keflin sales. [Ex. P-38; Step, deposition at 47-48.] 100. Lilly calculated that SmithKline’s profit margins on Ancef were as much as 25% lower than Lilly’s margin on Kefzol. [P-38; P-46; P-47.] 101. So long as its price on Kefzol was equal to or not much higher than Smith-Kline’s, Lilly counted on its reputation with physicians (particularly surgeons) and the reluctance of hospitals to suffer a decline in rebates (as compared with the rebates previously received under the CSP) because of their failure to participate in the Revised CSP, as a strategy to achieve its domination (goal of 75%) of the cefazolin market. [P-42; P-45; P-48; P-50; Step, Dep. at 57-64, 81; P-51; Step, Tr. 1029-1030.] 102. In many other cases, to keep its domination of the market, Lilly quoted a special price to its wholesalers where the hospital had received, or was about to receive, a bid on Ancef lower than the bid on Kefzol. One purpose of this procedure was to minimize the inroads that Ancef would otherwise make on Keflin. [Lilly’s answers to plaintiff’s intenogatories No. 19(d), -38, 58(a)(3); P-53; P-70.] I. COMPETITIVE EFFECTS OF THE REVISED CSP 103. Exclusive of the impact of the Revised CSP, the competitive factors applicable to the human pharmaceutical industry would be equally applicable to the broad antibiotic market. Thus, such competitive elements would necessarily be operative in the cephalosporin submarket. [See Step, Tr. 1042; See Finding of Fact ¶ 12.] 104. The barriers to entry and survival in the cephalosporin product market are substantial. It is difficult to dislodge the first company which introduces a line of products from its position of dominance. Absent the invention of a new cephalosporin which eclipses the performance of Lilly’s Keflin, a new market entrant must concentrate on marketing efforts to convince the hospital-based physician population of the merits of its cephalosporin product. The cost of marketing a cephalosporin product, absent the expense for salesmen (detail men) — the most costly and important aspect of the marketing effort — is set forth in Lilly’s answers to interrogatories. Lilly has spent, since 1973, more than $2,400,000 on Kefzol’s promotion (benefiting from SmithKline’s earlier research and development expenditures for cefazolin); since 1970, $9,929,007 on Keflin promotion; and $22,277,561 on Keflex promotion since 1970. SmithKline has spent more than $20,000,-000 on Ancef promotion and development, including its expenses for detail men; Lilly spent $2,361,000 on Kefzol’s development. Plaintiff’s total promotional expenditures for Ancef amount to $13,112,000 to date; from 1973 through 1975 alone, SmithKline spent $8,670,000 in detailing expenditures for Ancef. [Step, Tr. 1028-1035; See Step, Dep. 57; See Findings of Fact ¶¶ 16-18; Additional Responses of Eli Lilly & Company to Interrogatories of Plaintiff Smith-Kline Corporation — First Set, Nos. 49-50, 52, 54; • see Finding of Fact ¶ 68; Exhibit P-5, Attachment “B”.] 105. The Revised CSP calculates the rebate on one product in substantial part on the basis of sales of two other products. There is no way a competitor of Lilly can calculate a price on a product that competes with one Lilly product, unless the competitor knows the amount of the rebates the hospital will receive on the other two Lilly products. This is not public information. Lilly instructs its salesmen to be secretive about rebates. Often the hospital is uncertain what its Lilly rebate percentage will be until Lilly tells it, which does not occur until after the close of the quarter, when aggregate quarterly purchases are known, [van Roden, Tr. 40, 68-69; Exhibits P-171; Memorandum to the Court prepared by Charles H. Curl, Jr.; Lochridge, Tr. 417-419; Southard deposition at 19; P-78.] 106. If SmithKline could obtain the necessary information to calcúlate the rebate it would have to give to hospitals in order to match the rebate under the Revised CSP, it would find that such compensatory rebates would be in the order of 16% for average accounts and 35% or more for large accounts. [Lochridge, Tr. 394-413; P-171; P-172; P-173; Memorandum to the Court Prepared by Charles H. Curl, Jr.] 107. The rebates under the Revised CSP are payable in Lilly merchandise and are actually paid largely in Keflin and Keflex. By reason of the nature of the products, as well as Lilly’s integration of manufacturing processes, Lilly’s cost of goods is very substantially lower on Keflin and Keflex than on Kefzol, which means that any manufacturer with a product in competition with Kefzol, or with any other cephalosporin having a higher cost than Keflin and Keflex, will be at an automatic and substantial competitive disadvantage. [P-173; Lilly’s Answers to Plaintiff’s Interrogatories No. 42, 43, 45; Responses of Eli Lilly & Company to Interrogatories of Plaintiff Smith-Kline Corporation — Second Set 62(a), 76.] 108. The competitive strength of the Revised CSP is a function of Lilly’s dominance of the cephalosporin market. In the absence of the Revised CSP, and • using SmithKline’s cost of goods for its pharmaceutical division , and Lilly’s expense averages, the plaintiff has a pretax return on Ancef sales of 4.6%; the defendant’s return on sales of Kefzol is 17.6%. A 4.6% return on sales does not warrant continued marketing of cefazolin by SmithKline without the potential for significant improvement in profitability. SmithKline and Lilly have gross margins of 47.5% and 55% respectively on their sales of cefazolin. [Exhibit P-172, at 1.] 109. Lilly’s ability to offer the bonus and volume rebates under the Revised CSP is a function, not of its lower production costs, but, instead, the result of Lilly’s ability to recoup the resultant decline in Kefzol profits through its large profit margins on Keflin and Keflex. [See P-172, at 3; P-173 at 1-2.] ’ 110. If SmithKline remained in the cephalosporin market long enough to achieve Lilly’s cost ratio on cefazolin, the plaintiff’s return on sales would be 8.5% pretax, a sufficient return on sales to warrant remaining in the market. However, since the adoption of the Revised CSP, the payment of rebates constitutes a significa