US Healthcare v. Healthsource

March 17, 1993 UNITED STATES COURT OF APPEALS FOR THE FIRST CIRCUIT No. 92-1270 U. S. HEALTHCARE, INC., ETC., ET AL., Plaintiffs, Appellants, v. HEALTHSOURCE, INC., ET AL., Defendants, Appellees ERRATA SHEET The opinion of this court issued on February 26, 1993 is amended as follows: In footnote 1, l. 2, replace "1992" with "1991". On page 7, l. 9, replace "mid-1992" with "mid-1991". March 12, 1993 UNITED STATES COURT OF APPEALS FOR THE FIRST CIRCUIT No. 92-1270 U. S. HEALTHCARE, INC., ETC., ET AL., Plaintiffs, Appellants, v. HEALTHSOURCE, INC., ET AL., Defendants, Appellees ERRATA SHEET The opinion of this court issued on February 26, 1993 is amended as follows: On page 7, three lines above section II, replace "1992" with "1991". February 26, 1993 UNITED STATES COURT OF APPEALS For The First Circuit No. 92-1270 U. S. HEALTHCARE, INC., ETC., et al., Plaintiffs, Appellants, v. HEALTHSOURCE, INC., ETC., et al., Defendants, Appellees. APPEAL FROM THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF NEW HAMPSHIRE [William H. Barry, Jr., Magistrate Judge] Before Torruella, Cyr and Boudin, Circuit Judges. Franklin Poul with whom Dana B. Klinges, Mark L. Heimlich, Wolf, Block, Schorr and Solis-Cohen, Andrew D. Dunn, Thomas Quarles, Jr. and Devine, Millimet & Branch were on brief for appellants. Thomas Campbell with whom Deborah H. Bornstein, James W. Teevans, Gardner, Carton & Douglas, William J. Donovan, Peter S. Cowan and Sheehan, Phinney, Bass & Green were on brief for appellees. February 26, 1993 BOUDIN, Circuit Judge. U.S. Healthcare and two related companies (collectively "U.S. Healthcare") brought this antitrust case in the district court against Healthsource, Inc., its founder and one of its subsidiaries. Both sides are engaged in providing medical services through health maintenance organizations ("HMOs") in New Hampshire. In its suit U.S. Healthcare challenged an exclusive dealing clause in the contracts between the Healthsource HMO and doctors who provide primary care for it in New Hampshire. After a trial in district court, the magistrate judge found no violation, and U.S. Healthcare appealed. We affirm. I. BACKGROUND Healthsource New Hampshire is an HMO founded in 1985 by Dr. Norman Payson and a group of doctors in Concord, N.H. Its parent company, Healthsource, Inc., is headed by Dr. Payson and it manages or has interests in HMOs in a number of states. We refer to both the parent company and its New Hampshire HMO as "Healthsource." In simpler days, health care comprised a doctor, a patient and sometimes a hospital, but the Norman Rockwell era of medicine has given way to a new world of diverse and complex insurance and provider arrangements. One of the more successful innovations is the HMO, which acts both as a health insurer and provider, charging employers a fixed premium for each employee who subscribes. To provide medical -2- care to subscribers, an HMO of Healthsource's type--sometimes called an individual practice association or "IPA" model HMO- -contracts with independent doctors. These doctors continue to treat other patients, in contrast to a "staff" model HMO whose doctors would normally be full-time employees of the HMO. HMOs often can provide health care at lower cost by stressing preventative care, controlling costs, and driving hard bargains with doctors or hospitals (who thereby obtain more patients in exchange for a reduced charge). Healthsource, like other HMOs, uses primary care physicians-- usually internists but sometimes pediatricians or others--as "gatekeepers" who direct the patients to specialists only when necessary and who monitor hospital stays. Typically, the contracting primary care physicians do not charge by the visit but are paid "capitations" by the HMO, a fixed amount per month for each patient who selects the doctor as the patient's primary care physician. Unlike a patient with ordinary health insurance, the HMO patient is limited to the panel of doctors who have contracted with the HMO. There are familiar alternatives to HMOs. At the "financing" end, these include traditional insurance company policies that reimburse patients for doctor or hospital bills without limiting the patient's choice of doctor, as well as Blue Cross/Blue Shield plans of various types and Medicare -3- and Medicaid programs. At the "provider" end, there is also diversity. Doctors may now form so-called preferred provider organizations, which may include peer review and other joint activities, and contract together to provide medical services to large buyers like Blue Cross or to "network" model HMOs. There are also ordinary group medical practices. And, of course, there are still doctors engaged solely in independent practice on a fee-for-service basis. Healthsource's HMO operations in New Hampshire were a success. At the time of suit, Healthsource was the only non- staff HMO in the state with 47,000 patients (some in nearby areas of Massachusetts), representing about 5 percent of New Hampshire's population. Stringent controls gave it low costs, including a low hospital utilization rate; and it sought and obtained favorable rates from hospitals and specialists. Giving doctors a further stake in Healthsource's success and incentive to contain costs, Dr. Payson apparently encouraged doctors to become stockholders as well, and at least 400 did so. By 1989 Dr. Payson was proposing to make Healthsource a publicly traded company, in part to permit greater liquidity for its doctor shareholders. U.S. Healthcare is also in the business of operating HMOs. U.S. Healthcare, Inc., the parent of the other two plaintiff companies--U.S. Healthcare, Inc. (Massachusetts) and U.S. Healthcare of New Hampshire, Inc.--may be the -4- largest publicly held provider of HMO services in the country, serving over one million patients and having total 1990 revenues of well over a billion dollars. Prior to 1990, its Massachusetts subsidiary had done some recruiting of New Hampshire doctors to act as primary care providers for border-area residents served by its Massachusetts HMO. In 1989, U.S. Healthcare had a substantial interest in expanding into New Hampshire. Dr. Payson was aware in the fall of 1989 that HMOs operating in other states were thinking about offering their services in New Hampshire. He was also concerned that, when Healthsource went public, many of its doctor-shareholders would sell their stock, decreasing their interest in Healthsource and their incentive to control its costs. After considering alternative incentives, Dr. Payson and the HMO's chief operating officer conceived the exclusivity clause that has prompted this litigation. Shortly after the Healthsource public offering in November 1989, Healthsource notified its panel doctors that they would receive greater compensation if they agreed not to serve any other HMO. The new contract term, effective January 26, 1990, provided for an increase in the standard monthly capitation paid to each primary care physician, for each Healthsource HMO patient cared for by that doctor, if the doctor agreed to -5- the following optional paragraph in the basic doctor- Healthsource agreement: 11.01 Exclusive Services of Physicians. Physician agrees during the term of this Agreement not to serve as a participating physician for any other HMO plan; this shall not, however, preclude Physician from providing professional courtesy coverage arrangements for brief periods of time or emergency services to members of other HMO plans. A doctor who adopted the option remained free to serve non- HMO patients under ordinary indemnity insurance policies, under Blue Cross\Blue Shield plans, or under preferred provider arrangements. A doctor who accepted the option could also return to non-exclusive status by giving notice.1 Although Healthsource capitation amounts varied, a doctor who accepted the exclusivity option generally increased his or her capitation payments by a little more than $1 per patient per month; the magistrate judge put the amount at $1.16 and said that it represented an average increase of about 14 percent as compared with non-exclusive status. The dollar benefit of exclusivity for an individual doctor obviously varies with the number of HMO patients handled by the doctor. Many of the doctors had less than 100 Healthsource patients while about 50 of them had 200 or more. 1The original notice period was 180 days. This was reduced to 30 days in March or April 1991. It appears, at least in practice, that a doctor could switch to non- exclusive status more rapidly by returning some of the extra compensation previously paid. -6- About 250 doctors, or 87 percent of Healthsource's primary care physicians, opted for exclusivity. U.S. Healthcare through its New Hampshire subsidiary applied for a New Hampshire state license in the spring of 1990, following an earlier application by its Massachusetts subsidiary. A cease and desist order was entered against it, limiting its marketing efforts, because of premature claims that it had approval to operate in the state. The cease and desist order was withdrawn on February 15, 1991, and the license issued on February 21, 1991, subject to later approval of marketing materials. The present suit was filed in district court by U.S. Healthcare against Healthsource and Dr. Payson on March 12, 1991. By mid-1991, U.S. Healthcare had only two New Hampshire "accounts" and only about 18 primary care physicians. In the district court, U.S. Healthcare challenged the exclusivity clause under sections 1 and 2 of the Sherman Act, 15 U.S.C. 1-2, and under state antitrust and tort law. The parties stipulated to trial before a magistrate judge. After discovery, two separate weeks of trial were conducted in August and September 1991. In a decision filed on January 30, 1992, the magistrate judge found for the defendants on all counts. This appeal followed. II. DISCUSSION -7- In this court, U.S. Healthcare attacks the exclusivity clause primarily as a per se or near per se violation of section 1; accordingly we begin by examining the case through the per se or "quick look" lenses urged by U.S. Healthcare. We then consider the claim recast in the more conventional framework of Tampa Electric Co. v. Nashville Coal Co., 365 U.S. 320 (1961), the Supreme Court's latest word on exclusivity contracts, appraising them under section 1's rule of reason. Finally, we address U.S. Healthcare's claims of section 2 violation and its attacks on the market-definition findings of the magistrate judge. The Per Se and "Quick Look" Claims. U.S. Healthcare's challenge to the exclusivity clause, calling it first a per se violation and later a monopolization offense, invokes a signal aspect of antitrust analysis: the same competitive practice may be reviewed under several different rubrics and a plaintiff may prevail by establishing a claim under any one of them. Thus, while an exclusivity arrangement is often considered under section 1's rule of reason, it might in theory play a role in a per se violation of section 1, cf. Eastern States Retail Lumber Dealers' Ass'n v. United States, 234 U.S. 600 (1914), or as an element in attempted or actual monopolization, United States v. United Shoe Machinery Corp., 110 F. Supp. 295 (D. Mass. 1953), aff'd per curiam, 347 U.S. -8- 521 (1954). But each rubric has its own conditions and requirements of proof. We begin, as U.S. Healthcare does, with the per se rules of section 1 of the Sherman Act. It is a familiar story that Congress left the development of the Sherman Act largely to the courts and they in turn responded by classifying certain practices as per se violations under section 1. Today, the only serious candidates for this label are price (or output) fixing agreements and certain group boycotts or concerted refusals to deal.2 The advantage to a plaintiff is that given a per se violation, proof of the defendant's power, of illicit purpose and of anticompetitive effect are all said to be irrelevant, see United States v. Socony-Vacuum Oil Co., 310 U.S. 150 (1940); the disadvantage is the difficulty of squeezing a practice into the ever narrowing per se nitch. U.S. Healthcare's main argument for per se treatment is to describe the exclusivity clause as a group boycott. To understand why the claim ultimately fails one must begin by recognizing that per se condemnation is not visited on every arrangement that might, as a matter of language, be called a group boycott or concerted refusal to deal. Rather, today that designation is principally reserved for cases in which 2Tying is sometimes also described as a per se offense but, since some element of power must be shown and defenses are effectively available, "quasi" per se might be a better label. See Eastman Kodak Co. v. Image Technical Services, Inc., 112 S. Ct. 2072 (1992). -9- competitors agree with each other not to deal with a supplier or distributor if it continues to serve a competitor whom they seek to injure. This is the "secondary boycott" device used in such classic boycott cases as Eastern States Retail Lumber Dealers' Ass'n, and Fashion Originators' Guild of America, Inc. v. FTC, 312 U.S. 457 (1941). We doubt that the modern Supreme Court would use the boycott label to describe, or the rubric to condemn, a joint venture among competitors in which participation was allowed to some but not all, compare Northwest Wholesale Stationers, Inc. v. Pacific Stationery & Printing Co., 472 U.S. 284 (1985), with Associated Press v. United States, 326 U.S. 1 (1945), although such a restriction might well fall after a more complete analysis under the rule of reason. What is even more clear is that a purely vertical arrangement, by which (for example) a supplier or dealer makes an agreement exclusively to supply or serve a manufacturer, is not a group boycott. See Klor's, Inc. v. Broadway-Hale Stores, 359 U.S. 207, 212 (1959); Corey v. Look, 641 F.2d 32, 35 (1st Cir. 1981). Were the law otherwise, every distributor or retailer who agreed with a manufacturer to handle only one brand of television or bicycle would be engaged in a group boycott of other manufacturers. There are multiple reasons why the law permits (or, more accurately, does not condemn per se) vertical exclusivity; it -10- is enough to say here that the incentives for and effects of such arrangements are usually more benign than a horizontal arrangement among competitors that none of them will supply a company that deals with one of their competitors. No one would think twice about a doctor agreeing to work full time for a staff HMO, an extreme case of vertical exclusivity. Imagine, by contrast, the motives and effects of a horizontal agreement by all of the doctors in a town not to work at a hospital that serves a staff HMO which competes with the doctors. In this case, the exclusivity arrangements challenged by U.S. Healthcare are vertical in form, that is, they comprise individual promises to Healthsource made by each doctor selecting the option not to offer his or her services to another HMO. The closest that U.S. Healthcare gets to a possible horizontal case is this: it suggests that the exclusivity clause in question, although vertical in form, is in substance an implicit horizontal agreement by the doctors involved. U.S. Healthcare appears to argue that stockholder- doctors dominate Healthsource and, in order to protect their individual interests (as stockholders in Healthsource), they agreed (in their capacity as doctors) not to deal with any other HMO that might compete with Healthsource. We agree that such a horizontal arrangement, if devoid of joint venture efficiencies, might warrant per se condemnation. -11- The difficulty is that there is no evidence of such a horizontal agreement in this case. Although U.S. Healthcare notes that doctor-stockholders predominate on the Healthsource board that adopted the option, there is nothing to show that the clause was devised or encouraged by the panel doctors. On the contrary, the record indicates that Dr. Payson and Healthsource's chief operating officer developed the option to serve Healthsource's own interests. Formally vertical arrangements used to disguise horizontal ones are not unknown, see Interstate Circuit, Inc. v. United States, 306 U.S. 208 (1939), but U.S. Healthcare has supplied us with no evidence of such a masquerade in this case. There is less to be said for U.S. Healthcare's alternative argument that, if per se treatment is not proper, then at least the exclusivity clause can be condemned almost as swiftly based on "a quick look." Citing FTC v. Indiana Federation of Dentists, 476 U.S. 447 (1986), and NCAA v. Board of Regents, 468 U.S. 85 (1984), U.S. Healthcare argues that the exclusivity clause is so patently bad that even a brief glance at its impact, lack of business benefit and anticompetitive intent suffice to condemn it. The cases relied on provide little help to U.S. Healthcare and, even on its own version of those cases, the facts would not conceivably justify a "quick look" condemnation of the clause. -12- In the cited cases, the Supreme Court actually contracted the per se rule by refusing to apply it to horizontal agreements that involved price and output fixing (television rights by NCAA members) or the setting of other terms of trade (refusal of dentists by agreement to provide x-rays to insurers). Given the unusual contexts (an interdependent sports league in one case; medical care in the other), the Court declined to condemn the arrangements per se, without at least weighing the alleged justifications. At the same time it required only the briefest inspection (the cited "quick look") for the Court to reject the excuses and strike down the agreements. Accord, National Society of Professional Engineers v. United States, 435 U.S. 679 (1978). In any event, no "quick look" would ever suffice to condemn the exclusivity clause at issue in this case. Exclusive dealing arrangements come with the imprimatur of two leading Supreme Court decisions describing the potential virtues of such arrangements. Tampa; Standard Oil Co. of California v. United States, 337 U.S. 293 (1949) (Standard Stations); see also Jefferson Parish Hospital District No. 2 v. Hyde, 466 U.S. 2, 46 (1984) (O'Connor, J., concurring). To condemn such arrangements after Tampa requires a detailed depiction of circumstances and the most careful weighing of alleged dangers and potential benefits, which is to say the -13- normal treatment afforded by the rule of reason. To that subject we now turn. Rule of Reason. Exclusive dealing arrangements, like information exchanges or standard settings, come in a variety of forms and serve a range of objectives. Many of the purposes are benign, such as assurance of supply or outlets, enhanced ability to plan, reduced transaction costs, creation of dealer loyalty, and the like. See Standard Stations, 337 U.S. at 307. But there is one common danger for competition: an exclusive arrangement may "foreclose" so much of the available supply or outlet capacity that existing competitors or new entrants may be limited or excluded and, under certain circumstances, this may reinforce market power and raise prices for consumers. Although the Supreme Court once said that a "substantial" percentage foreclosure of suppliers or outlets would violate section 1, Standard Stations, the Court's Tampa decision effectively replaced any such quantitative test by an open-ended inquiry into competitive impact. What is required under Tampa is to determine "the probable effect of the [exclusive] contract on the relevant area of effective competition, taking into account . . . . [various factors including] the probable immediate and future effects which pre-emption of that share of the market might have on effective competition therein." 365 U.S. at 329. The lower -14- courts have followed Tampa and under this standard judgments for plaintiffs are not easily obtained. See ABA Antitrust Section, Antitrust Law Developments 172-73, 176-78 (3d ed. 1992) (collecting cases). On this appeal we are handicapped in appraising the extent and impact of the foreclosure wrought by Healthsource because U.S. Healthcare has not chosen to present its argument in these traditional terms. Tampa is not even cited in the opening or reply briefs. Some useful facts pertaining to the extent of the foreclosure are adverted to in U.S. Healthcare's opening "statement of the case" but never seriously developed in the argument section of its brief. Since the brief itself also describes countervailing evidence of Healthsource, something more is assuredly needed. In the two paragraphs of its "quick look" formulation addressed to "anticompetitive impact," U.S. Healthcare simply asserts that competitive impact has already been discussed and that the exclusivity clause has completely foreclosed U.S. Healthcare and any other non-staff HMO from operation in New Hampshire. This is not a persuasive treatment of a difficult issue or, rather, a host of issues. First, the extent to which the clause operated economically to restrict doctors is a serious question.3 True, most doctors signed up for it; but who 3Even with no notice period, Healthsource's differential pricing policy--paying more to those who exclusively serve Healthsource--would disadvantage competing HMOs. Some courts -15- would not take the extra compensation when no competing non- staff HMO was yet operating? The extent of the financial incentive to remain in an exclusive status is unclear, since it varies with patient load, and the least loaded (and thus least constrained by the clause) doctors would normally be the best candidates for a competing HMO. Healthsource suggests that by relatively modest amounts, U.S. Healthcare could offset the exclusivity bonus for a substantial number of Healthsource doctors. U.S. Healthcare's reply brief offers no response. Second, along with the economic inducement is the issue of duration. Normally an exclusivity clause terminable on 30 days' notice would be close to a de minimus constraint (Tampa involved a 20-year contract, and one year is sometimes taken as the trigger for close scrutiny). On the other hand, it may be that the original 180-day clause did frustrate U.S. Healthcare's initial efforts to enlist panel doctors, without whom it would be hard to sign up employers. Perhaps even a 30-day clause would have this effect, especially if a reimbursement penalty were visited on doctors switching back to non-exclusive status. Once again, U.S. Healthcare's brief offers conclusions and a few record references, but neither hesitate to apply the exclusivity label to such arrangements because there is no continuing promise not to deal (see Antitrust Developments, supra, at 176), but the differential pricing plan is unquestionably part of a contract and so subject to section 1, whatever label may be applied. -16- the precise operation of the clause nor its effects on individual doctors are clearly settled. Third, even assuming that the financial incentive and duration of the exclusivity clause did remove many of the Healthsource doctors from the reach of new HMOs, it is unclear how much this foreclosure impairs the ability of new HMOs to operate. Certainly the number of primary care physicians tied to Healthsource was significant--one figure suggested is 25 percent or more of all such primary care physicians in New Hampshire--but this still leaves a much larger number not tied to Healthsource. It may be, as U.S. Healthcare urges, that many of the remaining "available" doctors cannot fairly be counted (e.g., those employed full time elsewhere, or reaching retirement, or unwilling to serve HMOs at all). But the dimensions of this limitation were disputed and, by the same token, new doctors are constantly entering the market with an immediate need for patients. U.S. Healthcare lays great stress upon claims, supported by some meeting notes of Healthsource staff members, that the latter was aware of new HMO entry and conscious that new HMOs like U.S. Healthcare could be adversely affected by the exclusivity clause.4 Healthsource in turn says that these 4Two examples of these staff notes give their flavor: "Looking at '90 rates - and a deterent [sic] to joining other HMOs (like Healthcare)"; and "amend contract (sending this or next week) based on exclusivity. HMOs only (careful about restraint of trade) will be sent to even those in Healthcare -17- were notes made in the absence of policy-making officers and that its real motivation for the clause was to bolster loyalty and cost-cutting incentives. Motive can, of course, be a guide to expected effects, but effects are still the central concern of the antitrust laws, and motive is mainly a clue, see Barry Wright Corp. v. ITT Grinnell Corp., 724 F.2d 227 (1st Cir. 1983). This case itself suggests how far motives in business arrangement may be mixed, ambiguous, and subject to dispute. In any event, under Tampa the ultimate issue in exclusivity cases remains the issue of foreclosure and its consequences. Absent a compelling showing of foreclosure of substantial dimensions, we think there is no need for us to pursue any inquiry into Healthsource's precise motives for the clause, the existence and measure of any claimed benefits from exclusivity, the balance between harms and benefits, or the possible existence and relevance of any less restrictive means of achieving the benefits. We are similarly spared the difficulty of assessing the fact that the clause is limited to HMOs, a fact from which more than one inference may be drawn. The point is that proof of substantial foreclosure and of "probable immediate and future effects" is the essential basis under Tampa for an attack on an exclusivity clause. U.S. Healthcare has not supplied that basis. already . . . ." -18- In formal terms U.S. Healthcare has preserved on appeal its claim that the exclusivity clause unreasonably restrains competition in violation of section 1. That concept is embraced by its complaint, and the limited depiction of evidence in its appellate briefs stirs curiosity, if not suspicion. But putting to one side its per se claims and alleged market definition errors, U.S. Healthcare's basic argument in this court must be that the evidence compelled the magistrate judge to find substantial foreclosure having an unreasonable adverse effect on competition. U.S. Healthcare, as plaintiff at trial and appellant in this court, had the burden of fully mustering the facts and applying the analysis to establish such a claim. It has not done so. In this discussion, we have placed little weight upon the formal finding of the magistrate judge that "the [exclusivity] restriction does not constitute an unreasonable restraint of trade under Section 1 of the Sherman Act." His finding rested primarily on the premise that whatever the impact of the clause on HMOs, ample competition remains in a properly defined market, which he found to be one embracing all health care offered throughout the state of New Hampshire.5 On this view of the antitrust laws, it does not 5On the other hand, we do not accept U.S. Healthcare's effort to salvage something from the decision by arguing the magistrate judge found substantial foreclosure in fact. For -19- matter whether substantial foreclosure of new entrants occurs so long as widespread competition prevails in the relevant market, thereby protecting consumers.6 Whether the law requires such a further showing of likely impact on consumers is open to debate. Our own case law is not crystal clear on this issue. Compare Interface Group, Inc. v. Mass Port Authority, 816 F.2d 9, 11 (1st Cir. 1981), with Corey v. Look, 641 F.2d at 36. Ultimately the issue turns upon antitrust policy, where a permanent tension prevails between the "no sparrow shall fall" concept of antitrust, see Klor's, 359 U.S. at 213 (violation "not to be tolerated merely because the victim is just one merchant whose business is so small that his destruction makes little difference to the economy"), and the ascendant view that antitrust protects "competition, not competitors". See Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 488 (1977). We need not confront this issue in a case where the cardinal requirement of a valid claim--significant foreclosure unreasonably restricting competitors--has not been demonstrated. the most part, the statements to which it points appear to us to be efforts by the magistrate judge to describe the allegations made by U.S. Healthcare. 6See, e.g., Dep't of Justice Merger Guidelines, 4.21, 4.213, June 14, 1984, 49 Fed. Reg. 26824, 26835-36 (1984), adopting this position. The 1992 DOJ-FTC guidelines are directed only to horizontal mergers and do not address the issue. 49 Fed. Reg. 26823 (1992). -20- Section 2. Exclusive contracts might in some situations constitute the wrongful act that is an ingredient in monopolization claims under section 2. See United Shoe Machinery Corp. The magistrate judge resolved these section 2 claims in favor of Healthsource primarily by defining the market broadly to include all health care financing in New Hampshire. So defined, Healthsource had a share of that market too small to support an attempt charge, let alone one of actual monopolization. U.S. Healthcare argues, however, that the market was misdefined. It may be unnecessary to consider this claim since, as we have already held, U.S. Healthcare has failed to show a substantial foreclosure effect from the exclusivity clause. After all, an act can be wrongful in the context of section 2 only where it has or threatens to have a significant exclusionary impact. But a lesser showing of likely effect might be required if the actor were a monopolist or one within striking distance. Compare Berkey Photo Inc. v. Eastman Kodak Co., 603 F.2d 263, 272 (2d Cir. 1979), cert. denied, 444 U.S. 1093 (1980). More important, the magistrate judge dismissed the section 2 claims based on market definition and, if his definition were shown to be wrong, a remand might be required unless we were certain that U.S. Healthcare could never prevail. -21- There is no subject in antitrust law more confusing than market definition. One reason is that the concept, even in the pristine formulation of economists, is deliberately an attempt to oversimplify--for working purposes--the very complex economic interactions between a number of differently situated buyers and sellers, each of whom in reality has different costs, needs, and substitutes. See United States v. E.I. du Pont De Nemours & Co, 351 U.S. 377 (1956). Further, when lawyers and judges take hold of the concept, they impose on it nuances and formulas that reflect administrative and antitrust policy goals. This adaption is legitimate (economists have no patent on the concept), but it means that normative and descriptive ideas become intertwined in the process of market definition. Nevertheless, rational treatment is assisted by remembering to ask, in defining the market, why we are doing so: that is, what is the antitrust question in this case that market definition aims to answer? This threshold inquiry helps resolve U.S. Healthcare's claim that the magistrate judge erred at the outset by directing his analysis to the issue whether HMOs or health care financing was the relevant product market. This approach, says U.S. Healthcare, mistakenly focuses on the sale of health care to buyers whereas its concern is Healthsource's buying power in tying up doctors needed by other HMOs in order to compete. -22- The magistrate judge's approach was correct. One can monopolize a product as either a seller or a buyer; but as a buyer of doctor services, Healthsource could never achieve a monopoly (monopsony is the technical term), because doctors have too many alternative buyers for their services.7 Rather, the only way to cast Healthsource as a monopolist is to argue, as U.S. Healthcare apparently did, that HMO services (or even IPA HMOs) are a separate health care product sold to consumers such as employers and employees. If so, it might become possible (depending on market share and other factors) to describe Healthsource as a monopolist or potential monopolist in the sale of HMO (or IPA HMO) services in New Hampshire, using the exclusionary clause to foster or reinforce the monopoly. Thus, the magistrate judge asked the right question. Even so U.S. Healthcare argues that he gave the wrong answer in finding that HMOs were not a separate market (it uses the phrase "submarket" but this does not alter the issue). This is a legitimate contention and U.S. Healthcare has at least 7U.S. Healthcare, of course, is not concerned with Healthsource's ability as a monopsonist to exploit doctors; it is concerned with its own ability to find doctors to serve it. The latter question--one of foreclosure--depends on the available supply of doctors, the constraint imposed by the exclusivity clause, the prospect for entry of new doctors into the market, and similar issues. Whether U.S. Healthcare is foreclosed, however, does not depend on whether consumers treat HMOs as a part of health care financing or as a unique and separate product. -23- some basis for it: HMOs are often cheaper than other care methods because they emphasize illness prevention and severe cost control. U.S. Healthcare also seeks to distinguish cases defining a broader "health care financing market"-- cases heavily relied on by the magistrate judge--as involving quite different types of antitrust claims. See, e.g., Ball Memorial Hosp., Inc. v. Mutual Hosp. Ins., Inc., 784 F.2d 1325 (7th Cir. 1986). Once again, we agree that the nature of the claim can affect the proper market definition. The problem with U.S. Healthcare's argument is that differences in cost and quality between products create the possibility of separate markets, not the certainty. A car with more features and a higher price is, within some range, in the same market as one with less features and a lower price. The issue is sometimes described as one of interchangeability of products or services, see duPont (discussing cross-elasticity of demand), although this formula is itself only an aid in trying to infer the shape of the invisible demand curve facing the accused monopolist. In practice, the frustrating but routine question how to define the product market is answered in antitrust cases by asking expert economists to testify. Here, the issue for an economist would be whether a sole supplier of HMO services (or IPA HMOs if that is U.S. Healthcare's proposed market) could raise price far enough over cost, and for a long enough -24- period, to enjoy monopoly profits. Usage patterns, customer surveys, actual profit levels, comparison of features, ease of entry, and many other facts are pertinent in answering the question. Once again, U.S. Healthcare has not made its case in this court. The (unquantified) cost advantage of HMOs is the only important fact supplied; consumers might, or might not, regard this benefit as just about offset by the limits placed on the patient's choice of doctors. To be sure, there was some expert testimony in the district court on both sides of the market definition issue. But if there is any case in which counsel has the obligation to cull the record, organize the facts, and present them in the framework of a persuasive legal argument, it is a sophisticated antitrust case like this one. Without such a showing on appeal, we have limited ability to reconstruct so complex a record ourselves and no basis for overturning the magistrate judge. Absent the showing of a properly defined product market in which Healthsource could approach monopoly size, we have no reason to consider the geographic dimension of the market. If health care financing is the product market, as the magistrate judge determined, plainly Healthsource has no monopoly or anything close to it, given the number of other providers in New Hampshire, such as insurers, staff HMOs, Blue Cross/Blue Shield and individual doctors. This is -25- equally so whether the geographic market is southern New Hampshire (as U.S. Healthcare claims) or the whole state (as the magistrate judge found). III. CONCLUSION Once the federal antitrust claims are found wanting, this appeal is resolved. U.S. Healthcare offers no authority to suggest that New Hampshire antitrust law diverges from federal law; indeed, the state statute encourages a uniform construction. N.H. Rev. Stat. Ann. 356:14. As for the state tort-law claims, primarily interference with potential contractual relationships, the magistrate judge dealt effectively with them, and U.S. Healthcare says little on appeal to undercut his dismissal of those counts. Given the arguments made and the record evidence arrayed in this court, affirmance of the magistrate judge's judgment on all counts is clearly in order. Nevertheless, we do not think that this case was inherently frivolous. The timing and original 180-day reach of the exclusivity clause could reasonably excite suspicion; the clause may have some impact though the extent of that impact remains unclear; and the motives of Healthsource in adopting the clause may well have been mixed. Competition remains an essential force in controlling costs and improving quality in health care. Courts are properly available to settle claims that one business device or another is -26- unlawfully suppressing competition in this vital industry. Although U.S. Healthcare's per se shortcut has taken it to a dead end, we have addressed the antitrust issues at such length precisely because of the importance of the subject. Affirmed. -27-