The opinion of the court was delivered by: ROSZKOWSKI
MEMORANDUM OPINION AND ORDER
STANLEY J. ROSZKOWSKI, UNITED STATES DISTRICT JUDGE
On September 27, 1987, the defendants Rockford Memorial Corporation ("RMC") and SwedishAmerican Corporation ("SAC") executed a Memorandum of Understanding agreeing, in principle, to form a new corporation which would become the sole member of and control RMC and SAC and all their affiliate corporations including their respective principal subsidiaries, Rockford Memorial Hospital ("RMH") and SwedishAmerican Hospital ("SAH"). The consolidation was originally scheduled to take place June 1, 1988. On June 1, 1988, the United States of America ("government") filed a verified complaint, to prevent and restrain the merger of RMC and SAH and to adjudge the transaction to be in violation of Section 7 of the Clayton Act, 15 U.S.C. § 18, and Section 1 of the Sherman Act, 15 U.S.C. § 1. The government also moved for a preliminary injunction to enjoin RMC and SAC from taking any action to proceed with the consolidation.
Subsequent to the hearing on the government's motion for a preliminary injunction, the defendants agreed to voluntarily refrain from proceeding with the consolidation until a ruling on the motion for a preliminary injunction. Beginning on June 20, 1988 and concluding on July 14, 1988 the court heard the parties on the government's motion for a preliminary injunction. Subsequently, the parties have stipulated to a consolidation of the preliminary injunction hearing with a trial on the merits. The defendants have also moved to dismiss the government's Section 7 Clayton Act claim.
APPLICABILITY OF SECTION 7 OF THE CLAYTON ACT
As a preliminary matter, the court must decide if section 7 of the Clayton Act, 15 U.S.C. § 18, applies to the parties and transaction in question. The defendants have moved this court to dismiss the government's Section 7 claim for lack of subject matter jurisdiction. In essence, the defendants contend that Section 7 does not reach a non-stock consolidation between not-for-profit hospitals.
The government disputes the defendants' conclusion and points to the United States Supreme Court's decision in United States v. Philadelphia National Bank, 374 U.S. 321, 83 S. Ct. 1715, 10 L. Ed. 2d 915 (1963), to support their contention that Section 7 does indeed apply to the instant transaction. The defendants, on the other hand, do not view the Philadelphia decision as dispositive on the issue of Section 7 applicability and instead primarily rely on a reading of the language in Section 7, itself.
Section 7 of the Clayton Act states as follows:
No person shall acquire, directly or indirectly, the whole or any part of the stock or other share capital and no person subject to the jurisdiction of the Federal Trade Commission shall acquire the whole or any part of the assets of one or more persons engaged in commerce or in any activity affecting commerce, where in any line of commerce or in any activity affecting commerce in any section of the country, the effect of such acquisition of such stocks or assets, or of the use of such stock by the voting or granting of proxies or otherwise, may be substantially to lessen competition or to tend to create a monopoly.
At the risk of oversimplifying their position, the defendants read the first part of Section 7 as being made up of two provisions -- the first provision referring to an acquisition of "stock" or "share capital" and the second provision referring to an acquisition of "assets." This first provision, according to the defendants, applies to all "persons" whether or not they are under the Federal Trade Commission's ("FTC") jurisdiction, while the second "asset acquisition" provision applies only to persons under the jurisdiction of the FTC.
The defendants contend that the first provision applies only to consolidations accomplished through an acquisition of stock or share capital. Furthermore, the defendants argue that this first provision is dispositive in the instant case since the second provision only applies to persons under the FTC's jurisdiction, and not-for-profit companies such as the defendants in this case, are not under the aegis of the FTC. Accordingly, the defendants reason that since the consolidation of Rockford Memorial Corporation and SwedishAmerican Corporation does not involve a transfer of stock or share capital, and neither entity is under the FTC's jurisdiction, the transaction here is not reached by either the first or second provision of Section 7 and hence the Section is inapplicable to the defendants' consolidation.
The defendants continue that such an interpretation is not unreasonable even after a reading of the Philadelphia National Bank decision. The government contends that the case law in Philadelphia and subsequent decisions does not support the defendants' position.
In Philadelphia National Bank, supra, two banks, not under the jurisdiction of the FTC, intended to consolidate in a manner that had not been traditionally reached by the pre-1950 amended "stock" or "share" acquisition provision in Section 7. Thus, the Philadelphia court embarked on an analysis of the post-1950 amended section 7 and its applicability to a merger of two entities outside the FTC's jurisdiction. Philadelphia, 374 U.S. at 337, 83 S. Ct. at 1727.
The Philadelphia Court reviewed the history of Section 7 from the time of its enactment to the Supreme Court's decisions which effectively rendered it useless on all manner of mergers and consolidations.
The Court then proceeded to review the effect of the 1950 amendment that added the "asset acquisition" provision.
Based on this review, the Philadelphia Court concluded that through the 1950 amendment "congress primarily sought to bring mergers within § 7 and thereby close what it regarded as a loophole in the section." Philadelphia, 374 U.S. at 341, 83 S. Ct. at 1729. The Court continued that not only was Section 7 amended to reach mergers but, in some cases, also transactions which simply involve a purchase of assets unaccompanied by an exchange of shares or alteration of corporate identity. Philadelphia, 374 U.S. at 341-42, n. 20, 83 S. Ct. at 1729-30, n. 20.
After determining the intent of the amendment, the Court explained how this intent was carried out through the insertion of the "asset acquisition" provision. The Court noted that in order to bring the "entire range of corporate amalgamations from pure stock acquisitions to pure assets acquisitions, within the scope of § 7, the stock acquisition and asset acquisition provisions must be read together in order to reach mergers, which fit neither category perfectly but lie somewhere between the two ends of the spectrum." Philadelphia, 374 U.S. at 342, 83 S. Ct. at 1730. This interplay between the "asset acquisition" provision and the "stock acquisition" provision expands the meaning of § 7 in general and, importantly for our purposes, the "stock acquisition" provision in particular, to include acquisitions by merger or consolidation. Philadelphia, 374 U.S. at 346, 83 S. Ct. at 1732.
Besides providing the interplay with the "stock acquisition" provision, Congress also intended the "asset acquisition" provision to cover pure asset acquisitions such as the type of transaction found in the Columbia Steel case which involved a cash purchase by United States Steel Corporation of the physical assets of Consolidated Steel. There was no exchange of shares and no alteration of Consolidated's corporate identity. Philadelphia, 374 U.S. at 341-42, n. 20, 83 S. Ct at 1720-30, n. 20; see also United States v. Columbia Steel Lo, 334 U.S. 495, 68 S. Ct. 1107, 92 L. Ed. 1533 (1948). Of course, Section 7 qualifies its reach of simple asset acquisitions to "person[s] subject to the jurisdiction of the Federal Trade Commission." 15 U.S.C. § 18. "So construed, the specific exception for acquiring corporations not subject to the FTC's jurisdiction excludes from the coverage of § 7 only asset acquisitions by such corporations when not accomplished by merger." 374 U.S. at 343-44, 83 S. Ct. at 1730 (emphasis added).
In sum, the only difference in the reach of amended § 7 with regard to persons under, or not under, FTC jurisdiction is that Section 7 does not reach non-merger asset acquisitions accomplished by persons not under the FTC's jurisdiction.
What is covered by amended § 7 for all "persons" is the entire amalgamation of corporate mergers, from pure stock acquisitions to everything up to, but not including, a pure asset acquisition not accomplished by merger. Philadelphia, 374 U.S. at 343-44, 83 S. Ct. at 1730-31.
The Philadelphia Court's interpretation of § 7 essentially refutes the defendants' argument that the "first provision" of § 7 only reaches consolidations accomplished through an acquisition of "stock" or "share capital." Aware that the court may interpret the Philadelphia case to hold that corporate mergers accomplished by persons outside the FTC jurisdiction are subject to § 7, the defendants attempt to limit the reach of the Philadelphia Court's decision.
First, the defendants assert that the Philadelphia decision only applies to bank mergers, presumably since that is the type of merger involved in the case. The defendants cite several references to the terms "bank merger" in the text of the decision as proof of the exclusivity of the decision. Even a cursory review of that portion of the Philadelphia Court's opinion relevant to the defendants' motion to dismiss, however, evidences the broad scope of the Court's decision.
The relevant discussion is entitled "III. The Applicability of Section 7 of the Clayton Act to Bank Mergers; A. The Original Section and the 1950 Amendment." While this headline directly refers to "Bank Mergers," it in no way limits the breadth of the discussion that follows to bank mergers. Indeed, for purposes of the Court's analysis of § 7 as amended, the fact the defendants were banks was little more than fortuity. Of primary importance for purposes of the discussion was the defendants' status as a "person not under the jurisdiction of the FTC." This fact is confirmed over and over again in the text of the section.
For example, in summing up the question before it as one of first impression, the Court explains that there is no previous case in which a merger or consolidation was challenged under § 7 of the Clayton Act as amended, when the acquiring corporation was not subject to FTC jurisdiction. See, Philadelphia, 374 U.S. at 337, 83 S. Ct. at 1727. Then, after analyzing § 7 and the 1950 amendment, the Court held that "the specific exception for acquiring corporations [not merely banks] not subject to the FTC's jurisdiction excludes from the coverage of § 7 only assets acquisitions by such corporations when not accomplished by merger." 374 U.S. at 342, 83 S. Ct. at 1730 (emphasis added). Further, the Court did not limit to banks the following findings: "It is unquestioned that the stock-acquisition provision of § 7 embraces every corporation engaged in commerce, including banks." 374 U.S. at 343, 83 S. Ct. at 1730 (emphasis added). "If, therefore, mergers in other industries outside the FTC's jurisdiction were deemed beyond the reach of § 7, the result would be precisely that difference in treatment which Congress rejected." 374 U.S. at 333-34, 83 S. Ct. at 1730-31. Finally, the Court explicitly acknowledges that "we have not overlooked the fact that there are corporations in other industries not subject to FTC's jurisdiction." Philadelphia, 374 U.S. at 344 n. 22, 83 S. Ct. at 1731 n. 22. These references and others belie any argument that only mergers between banks were being considered in section III-A of the Philadelphia Court's decision.
Second, the defendants argue that the Philadelphia Court's analysis specifically focused on a transaction that involves exchanges of stock for stock. Thus, the defendants contend the Philadelphia Court did not intend by its holding to cover mergers between corporations which do not issue stock and are not subject to FTC jurisdiction. The defendants cling to an argument analogous to the argument of the defendants in the Philadelphia case itself; namely, that there is a distinction between an acquisition of corporate control through the exchanging of stock and an acquisition of corporate control in another manner [i.e. through agreement]. The Congress and the Philadelphia Court, however, specifically rejected this distinction and the unwarranted importance placed on the role of stock in acquiring control of a corporation. See e.g. Philadelphia, 374 U.S. at 343, 83 S. Ct. at 1730.
Indeed, in United States v. Chelsea Savings Bank, 300 F. Supp. 721 (D.Conn. 1969), the court found the merger of two non-stock banks to be subject to Section 7 of the Clayton Act. The Chelsea court, relying almost entirely on the Supreme Court's decision in Philadelphia, found no reason to support a ruling limiting the Supreme Court's rationale to amalgamations involving stock. Chelsea, 300 F. Supp. at 725. Instead, the Chelsea court interpreted the Philadelphia decision as a rejection of form over substance.
Chelsea, 300 F. Supp. at 723.
The Chelsea court then analyzed the non-stock consolidation of the two non-stock mutual banks and, based on this analysis, found that their consolidation was "tantamount in its effects to a merger" and thus should be tested by the standards set forth in § 7. Chelsea, 300 F. Supp. at 723-24, see Philadelphia, 374 U.S. at 344 n. 22, 83 S. Ct. at 1731 n. 22.
The defendants counter that Chelsea does not stand for the proposition that mergers effected without the exchanging of stock are subject to § 7, since the Chelsea court found that the depositors of a non-stock mutual savings bank stand in the same relation to the bank as the stockholders of an ordinary bank. Thus the depositors interest in a non-stock mutual savings bank is "share capital" under the provisions of § 7 of the Clayton Act.
The defendants miss the point. First, the Chelsea court had already found that the consolidation was subject to § 7 before it considered the "little practical significance" between the interests of bank shareholders and non-stock mutual bank depositers. Chelsea, 300 F. Supp. at 723-24. Second, and more importantly, Chelsea demonstrates the foolishness of requiring an exchange of stock to trigger § 7 of the Clayton Act. The Chelsea court's finding that ownership in a non-stock bank is similar to a stock bank's strengthens, rather than weakens, the argument that the ethereal manifestations of ownership are unimportant for anti-trust purposes. Indeed, there is "little practical significance" to a consumer if a combination of economic power which may lessen competition occurs through an exchange of stock or an agreement to merge. Consumers, not depositors and stockholders, are the class of persons to be protected by § 7. The method and manner of acquiring or exchanging ownership in an entity is not the inquiry that should be emphasized; rather, the inquiry is whether the resulting corporation(s) owns or controls, however that is manifested, the economic power of the prior corporations. See United States v. Columbia Pictures, 189 F. Supp. 153, 182 (S.D.N.Y. 1960) (found that Section 7 "is primarily concerned with the end result of a transfer of a sufficient part of the bundle of legal rights and privileges from the transferring person to the acquiring person to give the transfer economic significance and the proscribed adverse 'effect'").
Any other construction of § 7 or the Supreme Court's decision in Philadelphia leaves open a loophole for clever companies to exploit. Over and over again, the Philadelphia Court points out that the amended language in § 7 was intended to ensure against such blunting of the anti-merger thrust of the section by evasive actions. Philadelphia, 374 U.S. at 346, 83 S. Ct. at 1732. Indeed, the Philadelphia Court found that even a pure "assets acquisition" consummated by a person not under FTC jurisdiction would be subject to § 7 "if such an exchange . . . were tantamount . . . to a merger." Philadelphia, 374 U.S. at 344 n. 22, 83 S. Ct. at 1731 n. 22; see also United States v. Archer Daniels Midland Co., 584 F. Supp. 1134, 1139 (S.D. Iowa 1984) (held that lease of operations of one firm by another was an "acquisition" within the purview of Section 7 since to hold otherwise would permit the adroit use of leases to frustrate the anti-competitive purposes of the Clayton Act).
The bottom line to the Philadelphia decision is that Congress amended § 7 in order to reach mergers without qualification. The Philadelphia and Chelsea decisions teach that distinctions between methods of merging are meaningless for purposes of § 7. That lesson has not been lost on this court and it will not recognize cosmetic distinctions between mergers, including the existence of a stock transfer or the lack thereof.
The court finds the defendants' argument irrelevant for several reasons. First, the FTC's jurisdiction is not at issue and hence neither is its "co-extensiveness" with § 7's jurisdiction. All parties agree that the FTC has no jurisdiction over the defendants. Next, the fact that Congress did not want to eliminate the "asset acquisition" exception for persons not under the FTC jurisdiction may be because there already existed the ability to challenge mergers by not-for-profit corporations and thus no broadening was needed. Finally, and most importantly, the actions or non-actions of a subsequent Congress form a hazardous basis for inferring the intent of an earlier one. United States v. Price, 361 U.S. 304, 313, 80 S. Ct. 326, 332, 4 L. Ed. 2d 334 (1960). What Congress intended in 1950 by amending § 7 of the Clayton Act may not be revealed by congressional action in 1980. Accordingly, the defendants have failed to persuade the court to preclude the application of the Philadelphia court's interpretation of § 7 to the transaction at bar.
The question that remains, therefore, is whether the transaction at issue is a consolidation or merger within the entire amalgamation of corporate mergers. Given the wide range of corporate mergers, a simpler inquiry may be to determine what the transaction is not -- namely, a pure asset acquisition.
The Philadelphia court distinguishes a pure sale of assets from a typical merger in the following footnote:
A merger necessarily involves the complete disappearance of one of the merging corporations. A sale of assets, on the other hand, may involve no more than a substitution of cash for some part of the selling company's properties, with no change in corporate structure and no change in stockholder interests. Shareholders of merging corporations surrender their interest in those corporations in exchange for their very different rights in the resulting corporation. In an asset acquisition, however, the shareholders of the selling corporation obtain no interest in the purchasing corporation and retain no interest in the assets transferred. In a merger, unlike an asset acquisition, the resulting firm automatically acquires all the rights, powers, franchises, liabilities, and fiduciary rights and obligations of the merging firms. In a merger, but not in an asset acquisition, there is the likelihood of a continuity of management and other personnel. Finally, a merger, like a stock acquisition, necessarily involves the acquisition by one corporation of an immediate voice in the management of the business of another corporation; no voice in the decisions of another corporation is acquired by purchase of some part of its assets.
Philadelphia, 374 U.S. at 336 n. 13, 83 S. Ct. at 1727 n. 13.
To this end, the government points out that the defendants have never averred that the consolidation of Rockford Memorial Hospital and SwedishAmerican Hospital is an acquisition of assets. Rather, this transaction was characterized by the defendants, themselves, as follows:
The consolidation will be a union of equals, with a board of directors comprised of an equal number of representatives from RMC [Rockford Memorial Corporation] and SAC [SwedishAmerican Corporation]. The board will guide the integration and consolidation of the programs, services and management and governance structure of RMC and SAC and their affiliates.
(Defendants' Motion to Dismiss, p. 6)
(Defendants' Finding of Facts, para. 1)
Based on the merger and sale of assets examples supplied by the Philadelphia court, the consolidation outlined above is clearly a merger, not a pure asset acquisition. The two existing corporations are subsumed into a new resulting entity. The resulting corporation inherits the program, services, management, rights, liabilities, obligations and abilities of SwedishAmerican Corporation and Rockford Memorial Corporation. There will be continuity in management and personnel after the consolidation, since representation of both corporations will oversee the new resulting corporation. In short, no empty purchase or sale of assets has occurred in the instant case. Instead, what we have is the combining of the economic power of SwedishAmerican Corporation and Rockford Memorial Corporation. Accordingly, the consolidation proposed by the defendants, as evidenced by their Amended Memorandum of Understanding, is a merger subject to § 7 of the Clayton Act.
Accordingly, the court finds that § 7 of the Clayton Act reaches non-stock mergers accomplished by persons not under the jurisdiction of the FTC. Thus, the court finds that the consolidation of SwedishAmerican Corporation and Rockford Memorial Corporation is subject to § 7 of the Clayton Act.
An analysis of an alleged Section 7 violation requires a determination of the relevant "line of commerce", or product market, in which competition is to be allegedly lessened. The Supreme Court's decision in Brown Shoe v. United States, 370 U.S. 294, 82 S. Ct. 1502, 8 L. Ed. 2d 510 (1962) is instructive on how to define a relevant product market or in the provision of healthcare services, service market:
The outer boundaries of a product market are determined by the reasonable interchangeability of use or the cross-elasticity of demand between the product itself and substitutes for it. However, within this broad market well-defined submarkets may exist which, in themselves, constitute product markets for anti-trust purposes [citations omitted].
The boundaries of such a submarket may be determined by examining such practical indicia as industry or public recognition of the submarket as a separate economic entity, the product's peculiar characteristics and uses, unique production facilities, distinct customers, distinct prices, sensitivity to price changes, and specialized vendors. Because § 7 of the Clayton Act prohibits any merger which may substantially lessen competition" in any line of commerce " (emphasis supplied) it is necessary to examine the effects of a merger in each such economically significant submarket to determine if there is a reasonable probability that the merger will substantially lessen competition [in this submarket].
Brown Shoe, 370 U.S. at 325, 82 S. Ct. at 1523-24.
In sum, the service market in which the impact of a merger is measured should include services with sufficiently peculiar characteristics and uses to constitute them products sufficiently distinct from all others.
In the case at bar, the government proposes that inpatient care is the relevant product market to be examined, while the defendants proffer a broader product market that includes both inpatient and outpatient care provided by all health care providers.
In support of their market, the defendants point to the trend in health care toward outpatient services. Many procedures once in the exclusive domain of inpatient care, are now performed on an outpatient basis. A boom in medical technology has revolutionized diagnostic testing through the use of computer tomography scans ("CT"), Magnetic Resonance Imagery ("MRI") and surgical procedures through the use of non-invasive techniques such as extracorporeal shockwave lithotriphy ("ESWL") and laser surgery. The defendants point out that scores of tests and treatments are now being performed on either an inpatient or outpatient basis. (FF, para. 40-43).
The decision on which type of care to seek, the defendants continue, is in the hands and the discretion of the patient and his or her physician. Further, the defendants reason that third party payers are very price sensitive entities that will choose outpatient care over the relatively more expensive inpatient alternative. The defendants conclude that a product market excluding outpatient services is too narrow; outpatient services are (and will be even to a greater extent in the future) adequate substitutes for inpatient care. Any attempted effort of the merging hospitals to exercise market power by raising inpatient prices would be counteracted by the shifting of patients from inpatient to outpatient care.
The court agrees that there is interchangeability between outpatient and inpatient services. The court, however, does not agree that the relevant line of commerce includes outpatient care. While the court acknowledges that there is a substitutability between inpatient and outpatient care, this substitutability is over time and therefore not dispositive for purposes of defining the product market. (Tr. 1634). What is meant by "substitutability over time" is simple. Over time, some exclusively inpatient tests and procedures are performed on an outpatient basis through the implementation of new technology and medical knowledge. Once it is established, however, that a test or procedure may be performed on either an inpatient or outpatient basis, other forces come into play negating free substitutability between inpatient and outpatient care.
As mentioned above, patients who need inpatient care in a general acute care hospital are generally more gravely ill than their outpatient counterparts. Their therapy may require several types of diagnostic tests, twenty-four hour nursing, extensive pre or post operative observation, or any number or combination of other services offered by an acute care hospital. (FF, para. 56-57).
At the core of these peculiar characteristics is the hospital's ability to provide overnight care -- no outpatient program can match this trait. In tandem with overnight care, the hospital has assembled a variety of services "under one roof." This ability to perform a variety of tests and procedures in one place is also unmatched by a non-hospital health provider. While the hospitals may claim that outpatient care providers have raided the hospital's more lucrative services, this does not mean the outpatient providers can provide all the same services. The reason is that no outpatient provider possesses the combination of services at the disposal of a hospital. This concept that a combination or "cluster" of products or services could serve as a relevant product market was adopted in United States v. Philadelphia National Bank, supra.
In Philadelphia, the "cluster" of services known as "commercial banking" had no real substitute, even in light of the existence of a variety of other types of financial institutions. In U.S. v. Connecticut National Bank, 418 U.S. 656, 94 S. Ct. 2788, 41 L. Ed. 2d 1016 (1974), despite finding a good deal of competitive overlap between commercial and savings banks, the Court still found that commercial banks were part of a distinct line of commerce since they offered a cluster of services that savings banks could not match, particularly with regard to commercial customers. Connecticut, 418 U.S. at 663-666, 94 S. Ct. at 2793-95. This "cluster of services" concept was also applied to the health care industry and acute care hospitals, in particular, by the Federal Trade Commission in In the Matter of Hospital Corporation of America, 106 F.T.C. 361 (1985), aff'd, HCA v. FTC, 807 F.2d 1381 (7th Cir. 1986).
In the FTC's decision, the FTC found that the "healthcare" market encompassed such a variety of services offered by providers that it did not reflect the area within which specific health care providers compete. On the other hand, analyzing the individual services offered by providers was also unacceptable, since services were often offered in combination. Thus, the FTC decided that in such cases a "cluster of services", with enough peculiar characteristics such that other services or even combination of services, would not readily interchange with them, would suffice as a service market. In re HCA, 106 F.T.C. at 434-35, 465-66. The cluster of services adopted by the FTC for acute care hospitals was acute inpatient hospital care. In re HCA, 106 F.T.C. at 464-66.
The FTC reasoned that although outpatient providers (doctor's officers, freestanding clinics, urgent care and surgical centers) offer the same services as hospitals, they are offered in a different context. In re HCA, 106 F.T.C. at 436, 464-66. Patients in hospitals are more gravely ill, requiring a cluster of general acute inpatient hospital services. In re HCA, 106 F.T.C. at 436, 466. Thus, the FTC found that while hospital services may well have outpatient substitutes, the benefit that accrues to the patient is derived from the complimentary cluster of services and that there is no readily available substitute supplier of the benefit that this complementing or combination of services confers on the acute care patient. In re HCA, 106 F.T.C. at 436, 466.
The characteristics of acute inpatient hospital care are unique among healthcare services. It is an inescapable fact that certain patients cannot be treated on merely an outpatient basis. The obvious reason is that the patient requires greater care either in number or depth of services. Today, and in the foreseeable future, this care can only be effected through inpatient hospital care. There is no substitute for this cluster of care. (FF, para. 56-57).
The provision of outpatient care by all healthcare providers, including hospitals, is increasing and will continue to do so in the foreseeable future. In hospitals, the same facilities and personnel used to provide inpatient care is also used to provide outpatient care. In addition, many outpatients treated by hospitals are ultimately admitted as inpatients. Undoubtedly, there is direct competition between the individual outpatient services provided by hospitals and non-hospitals. (FF, para. 58-65). Despite the presence of this strain of competition, outpatient care is no substitute for acute inpatient hospital care. The outpatient provider represents a few procedures at most and cannot provide in any circumstance, an overnight stay. In providing patient care, the hospital may utilize a procedure that competes directly with outpatient providers when used alone for an outpatient. However, all competition and substitutability between the hospital and the outpatient provider ends when that same outpatient procedure is used in tandem with other services to treat an inpatient. The outpatient provider has nothing comparable to offer.
(FF, para. 66-67).
If the defendants could assert a small but significant and non-transitory price increase in inpatient care, the exercise of market power could not be ameliorated by outpatient care. Of course, there are some inpatients currently receiving inpatient care that could possibly utilize outpatient care, but not in numbers significant enough to break an exercise of market power in the inpatient care market.
It is true, as the defendants contend, that non-hospital providers such as out-patient clinics, emergency care centers, doctors offices and other providers, offer some of the same services as do acute care hospitals. Furthermore, it is also true that the relevant product market must include all services provided by acute care hospitals and those services offered by non-hospital providers which are an alternative to hospital care. However, only the acute care hospitals can provide the unique combinations of services which an acute care patient must have. Outpatient facilities routinely feed patients to the hospitals inpatient facilities. Those services include both inpatient and outpatient care, both of which are competed for by acute care hospitals. Accordingly, the court finds that the relevant product market consists of that cluster of services offered only by acute care hospitals.
The next step in analyzing the effects of the defendants' putative consolidation is to determine the "area of effective competition in the known line of commerce - acute hospital inpatient care." The purpose in determining the appropriate geographic market is to identify the relevant competitors who could constrain the merging firms from exercising market power. (Tr. 757). The geographic "market [should be in an] area in which the seller operates, and to which the purchaser can practicably turn for supplies." Philadelphia, 374 U.S. at 359, 83 S. Ct. at 1739 quoting, Tampa Elec. Co. v. Nashville Coal Co., 365 U.S. 320, 327, 81 S. Ct. 623, 628, 5 L. Ed. 2d 580 (1961) (emphasis supplied), (Tr. 757-58). The Supreme Court refined the inquiry as follows:
Philadelphia, 374 U.S. at 357, 83 S. Ct. at 1738.
Above all, a "pragmatic factual approach" to the delineation of the appropriate geographic market and not a formal legalistic one should be enlisted. The relevant "section of the country" should "correspond to the commercial realities" of the industry and be economically significant. Brown Shoe, 370 U.S. at 336-37, 82 S. Ct. at 1530. Of course, the relevant geographical market need not encompass a recognized geopolitical boundary and may be as big as the entire country or as small as a metropolitan area. Brown Shoe, 370 U.S. at 337, 82 S. Ct. at 1530.
The relevant geographic areas proposed by the parties represent significantly disparate views of the market in which RMH and SAH operate. The defendants suggest that the appropriate section of the country consists of the following ten counties: Winnebago, Boone, Stephenson, Ogle, Lee, McHenry, Dekalb (all in Illinois) Walworth, Green and Rock (all in Wisconsin). (DX 120). This area has been labeled the "Winnebago area" and contains nineteen acute care hospitals. (DX 119). Conversely, the government proffers a much smaller area consisting of almost all of Winnebago County and the northeast portion of Ogle County. This area has been labeled the "Rockford area" and contains only the three Rockford hospitals. (GX 39).
It is no coincidence that the parties arrived at such distinct markets since their methodologies diverged soon after each recognized the importance of determining the "geographic structure of the supplier/customer relationship." The government attempted to apprehend this structure primarily through extensive interviews with local and not-so-local health providers and health purchasers. The defendants, on the other hand, placed primary reliance on an empirical study of patient origins and destinations in order to determine the geographic structure of supplier/customer relations or in terms used in the healthcare industry provider/purchaser relations. The court will not comment on the appropriateness of either approach in the abstract but only as necessary in examining the specific merits, or lack thereof, in the parties analyses.
The court will first examine the government's proposed geographic market. The government's market is supported by a relatively few simple propositions that, for the most part, are supported by the evidence taken at the preliminary injunction hearing.
The government explains that hospital admissions are determined, for the most part, not by putative inpatients but by their physicians. These physicians will invariably admit patients to hospitals where the physicians have admitting privileges. In turn, physicians seek admitting privileges at hospitals where they anticipate admitting a significant number of their patients. Usually, physicians want to admit patients to institutions that are convenient both to themselves and their patients. Doctors favor hospitals close enough to allow them to conduct their daily practice and still have time to check patients on daily rounds and if need be to respond quickly to an inpatient in need of immediate attention. In this same vein, patients generally desire to be admitted to a hospital near their family and friends and their admitting physician. In sum, physicians and patients generally choose to use relatively close hospitals. (FF, para. 68-73).
The government argues that this tendency for convenience in admitting significantly limits the number of competitors in the market. The government points to the fact that of the approximately 450 physicians in the "Rockford area" nearly all of them have admitting privileges, either active or courtesy,
with all three Rockford area hospitals (RMH, SAH and STA) but nowhere else. Thus, if a "Rockford area" physician's patient wanted to stay at a hospital other than RMH, STA or SAH, he or she would need to be referred to a non-Rockford physician to admit him or her. (Tr. 130, 517, 1268-69). The government surmises that these admitting patterns show that doctors in the "Rockford area" are unwilling to admit their regular patients to hospitals outside the "Rockford area," forcing them to travel longer distances to fulfill their duties. The government further concludes that these same doctors will be unwilling to give up the convenience of the nearby Rockford hospitals for others farther away, even in the event of a price increase in inpatient care at the defendants' hospitals.
Similarly, the government points out that another important group, third party payers, also prefer nearby hospitals. These payers are not amenable to purchasing healthcare outside the three Rockford hospitals for "Rockford area" patients. (FF, para. 80). Third party payers are employers, Medicare, Medicaid, and healthcare indemnity insurers such as Blue Cross and Aetna. These payers are the true purchasers of healthcare in most instances. In addition, managed health care plans such as Health Maintenance Organizations ("HMOs") and Preferred Provider Organizations ("PPOs") are involved in purchasing healthcare. A HMO contracts with a group of physicians to perform medical services for plan subscribers at a set pre-paid fee. A PPO is an agreement between a healthprovider such as a hospital or clinic and a purchaser. The agreement usually consists of the provider granting a discount to purchasers in return for the purchaser's promise to send patients to the healthprovider. (FF, para. 75-79). Based on the same doctor/patient sentiment for convenience outlined above, these third party payers are wary of attempting to force plan beneficiaries from their local Rockford hospitals to relatively distant hospitals. (FF, para. 75-80).
Nearby Community Hospitals
Next, the government surveyed the hospitals in nearby communities that could potentially compete for the defendants' inpatient business. The government sought to forge the image of small community hospitals that provide primary and secondary services and perhaps a few tertiary services, but do not compete in an overall sense with the three Rockford hospitals. The government was generally successful in demonstrating the relatively limited number and scope of services provided by, among others, Beloit Memorial, Freeport Memorial, Highland, St. Joseph, Sycamore, Rochelle Community, Kishwaukee and Harvard Community Memorial Hospital and the lack of staff overlap between the defendants and these same hospitals. (See FF, para. 74, 81-119).
The government ran into trouble, however, when it attempted to question the administrative heads of these hospitals on whether they "compete" with the three Rockford hospitals. (STA, SAH and RMH) Dr. Allen, the government's economist, queried the administrators as to the effect of a hypothetical price increase (usually 10 to 20%) in services offer by the three Rockford hospitals. (Tr. 878-79, 885). This approach was not without its difficulties. First, the pricing of the same services between the three Rockford hospitals often differed by more than ten percent, creating doubt as to what was meant by a 10% or 20% price increase in services offered by the Rockford hospitals. (DX 108, 109, Tr. 1152-56). Secondly, Dr. Allen did not employ a set routine in asking the administrators the hypothetical "price question", also creating doubt as to the consistency of the questions and hence the answers. (Tr. 876-77). In any event, the evidence clearly suggests that the aforementioned community hospitals lack size, services, skills and quality comparable to their Rockford counterparts.
Patient Origin and Destination
Stealing a page from the defendants' reliance on patient origin and destination data, the government contended that the patient origin and destination figures bear out the appropriateness of the "Rockford area" geographic market. First, the government points out that approximately 93% of the residents of the "Rockford area" who are hospitalized turn to one of the "Rockford area" hospitals. (Tr. 764, 944). Second, approximately 75% of the patients admitted as inpatients to the "Rockford area" hospitals reside in the "Rockford area." (Tr. 764, 944). While the government may look at these figures as an indication of the viability of the "Rockford area" market, the defendants cite the fact that twenty-five percent of the defendants' inpatients come from outside the "Rockford area" as "quite a bit of leakage" and proof that the government's proposed market is too small.
The theoretical significance of this "leakage" is simple when looked at in the context of a merger. Within the government's market there are no potential competitors other than the three Rockford hospitals. Assuming that the merged entity could exercise market power even in light of the presence of St. Anthony, then the patients inside the "Rockford area" (75%) could not turn to alternative hospitals with sufficient capacity to defeat the market power exercised and would be forced to suffer the consequences of an anti-competitive market. Meanwhile, the twenty-five percent of the defendants' hospitals' inpatients presently immigrating into the market could presumably turn elsewhere for inpatient care. As a consequence, the defendants could theoretically lose up to twenty-five percent of their patients. (Tr. 1124, 1147-1148). Although there are differing views as to the effect of losing these patients on the pricing of the defendants' services to their remaining patients, (see DX 112, GX 74), the more important and underlying question is whether the other hospitals that the immigrating patients turn to can effectively discipline or constrain the merging hospitals by offering a significant segment (25%) of the defendants' inpatient base as alternative. If this is the case, then the Rockford area market is too small.
The government counters that the defendants would not, in fact, lose twenty-five percent of their patients to surrounding community hospitals since they do not provide the type or quality of services offered by the "Rockford area" hospitals. In short, the government contends that these alternative hospitals do not provide a viable alternative to the immigrating patients and will not "win" these patients away from the defendants. This contention is one of the more hotly debated issues in the instant case. The defendants, through three graphic exhibits, have attempted to show that the reason patients travel to the Rockford hospitals is not because comparable services are not available in their home counties, but rather for the more nebulous reason of the perceived quality of bigger and better hospitals in Rockford.
The defendants attempt to show in two different ways that the average case severity of non-Winnebago County residents hospitalized in Rockford is similar to the average case severity of a patient residing in Winnebago County. The purpose behind these graphs (DX91 and DX130) are to illustrate the type of patient immigrating into the "Rockford area."
The averages revealed that of the ten counties, all but Boone, Rock and Green had higher average "case severity" than Winnebago County residents. The Rock and Green figures were based on only 881 and 25 cases, respectively, or a mere 1.54% of the total number of cases analyzed in the ten county area, rendering the results for Rock and Green Counties suspect since so few cases were observed. (DX 130). The average "case severity" of counties such as Stephenson, DeKalb, McHenry and Lee were quite a bit higher than Winnebago. (DX 91, 130). Ogle County patients also suffered higher "case severity" but not to any great extent. (DX 91, 130). Even conceding that the underlying assumptions made by the defendants are accurate, the conclusions to be drawn are mixed. For most of the ten county area, the patients who immigrate into Winnebago County are, on average, more seriously ill than Winnebago County residents, with Boone County patients the one conspicuous exception.
The defendants utilize a slightly different tack in DX 93. The defendants present information on how many different DRG's were utilized by patients residing in the ten county area (excluding Winnebago County) in Winnebago County and the number of different DRG's available in their respective home counties. The purpose behind this graph was to demonstrate that services comparable to the Winnebago County hospitals are available at the community hospitals in the ten county area. (Tr. 1161). Interestingly, the results consistently highlight the greater number of DRG's available in the Rockford hospitals as opposed to those offered by healthproviders in the other nine counties. (See DX 93). Of course, the point made by the defendants is not that all the DRG's available in Rockford hospitals are available in the other 9 counties, but that a majority of them are.
(Tr. 1162-64). In fact, the defendants believe these figures are understated since many DRG's that are available in the other nine counties may not have been utilized during the pendency of the study and thus were not figured into the study. (Tr. 1254). Of course, the same could be said for DRG's available at the Rockford hospitals. Moreover, there are no break downs as to severity within the DRG's offered; thus, there can be no assurance that more severely ill patients in a given DRG still do not travel to Rockford, while less severely ill patients in a given DRG stay closer to home. (Tr. 1257). The defendants figures again yield mixed results. It appears that more severely ill patients do travel to Rockford hospitals but, at the same time, some of the immigrating patients eschew comparable treatment in their home counties and travel to Rockford.
There are other facts and figures that are less equivocal in exposing the composition of the non-Winnebago County residents hospitalized in Rockford hospitals. A figure used in a defendants' exhibit reveals the nature of the immigrating patients In DX 112, assessing the "effect on [the] hospital inpatient price necessary to offset loss of non-Winnebago County admissions to Winnebago County hospitals," the defendants rely on the following fact: "For RMH, the proportion of non-Winnebago revenue-exceeds the proportion of non-Winnebago admissions by 36%." Using, as the defendants did in DX 91, and DX 130, the cost of treatment as a proxy for severity, the disproportionate amount of revenue generated by non-Winnebago County patients illustrates that the severity of the non-Winnebago County patients' illnesses treated at the Rockford hospitals are disproportionately greater than the illnesses of Winnebago County patients treated at the Rockford hospitals.
There are several referral agreements between the defendants and community hospitals in and around the ten county area. (FF. para. 81, 84, 97, 107, 115-16, 119). This reveals two things about the immigrating patient and the community hospitals. First, patients travel to Rockford because certain services are not available at their local community hospitals. Second, and to a much lesser extent, these community hospitals are not worried about losing their primary and secondary patients to the Rockford hospitals.
In sum, the court finds that a significant number of the 25% of patients migrating into the Rockford hospitals are seeking services not available in their local hospitals suggesting that three Rockford hospitals are the only relevant competition and the "Rockford area" is the relevant geographic market. Nevertheless, even if a number of immigrating patients travel to Rockford for special services, the very existence of such a large segment of immigrating patients indicates that there still may be competition and competitors outside the "Rockford area" which could exert market discipline on the merging entities. Thus, the preceding finding does not close the inquiry into the appropriate geographic market.
As mentioned above, the defendants argue that 25% is "quite a bit of leakage" and explain that the reason for the leakage is simply that the government's market is too small. They contend, in fact, that it is approximately nine counties too small. The defendants attempt to define the relevant geographic market by primarily relying on a study of patient origin and destination data known as the Elzinga-Hogarty test. The Elzinga-Hogarty test focuses on shipment patterns to identify geographically, pertinent competitors of merging firms. (Tr. 1120). In a hospital context, the test focuses on patient travel. (Tr. 1120).
The test consists of two separate measurements - Lifo and Lofi. In the hospital context, a Lofi or "little out from the inside" statistic signifies the percentage of patients in an area's hospitals who reside in the area (rather than outside the area). See In re HCA, 106 F.T.C. at 468 n. 7; (Tr. 1121-22). This statistic is useful in determining whether a substantial number of patients from outside the area travel into an area for hospitalization. Id. If so, that implies that hospitals located in the areas where those immigrating patients reside could act as a check on the exercise of market power by hospitals inside the area. Id. In sum, Lofi measures the immigration of patients into an area and Lifo measures the out migration of patients from an area. Id.
A "Lifo" or "little in from out" statistic signifies the percentage of hospital patients from a particular area who remain in that area for hospital services. Id.; (Tr. 1123). This statistic is useful in determining whether patients in a particular area make substantial use of hospitals outside the area. Id. If so, that implies that hospitals outside the area could act as a check on the exercise of market power by hospitals inside the area. Id.
Ideally, an area should be delineated where "few" patients leave an area and "few" patients enter an area to obtain hospital services. The defendants define "little" or "few" as no more than ten percent. (Tr. 1123-25). In other words, the Lofi and Lifo figures should ideally yield at least percentages of 90% or greater.
In order to implement the Elzinga-Hogarty test, a process must be followed. In DX 100, the defendants outline a step by step methodology to be employed as follows:
HOSPITAL GEOGRAPHIC MARKET DETERMINATION
STEP 1(a): DETERMINE the MERGING HOSPITALS' SERVICE AREA, I.E., THE AREA FROM WHICH THEY DRAW 90 PERCENT OF THEIR BUSINESS.
STEP 1(b): DETERMINE THE COLLECTIVE SERVICE AREA OF ALL HOSPITALS LOCATED WITHIN THE MERGING HOSPITALS' SERVICE AREA.
* * * THIS AREA SATISFIES THE LOFI TEST (i.e., LITTLE OUT FROM INSIDE).
STEP 2: DETERMINE THE AREA CONTAINING THOSE HOSPITALS THAT SUPPLY 90 PERCENT OF ALL THE BUSINESS THAT COMES FROM PATIENTS RESIDING IN THE COLLECTIVE SERVICE AREA.
* * * THIS AREA SATISFIES THE LIFO TEST (i.e., LITTLE IN FROM OUTSIDE).
* * * THE HOSPITALS IDENTIFIED BY THIS PROCEDURE ARE WITHIN THE GEOGRAPHIC MARKET RELEVANT TO THE PROPOSED MERGER.
Utilizing this methodology, the defendants arrived at their proposed ten county geographic market with a resulting Lofi of 91% and Lifo of 80.5% using patient admissions data and a Lofi of 90% and a Lifo of 74.7% using patient charges data. (DX 104, 105).
The court, however, is not convinced of the dispositiveness of the defendants results. While the court acknowledges the general efficacy of the Elzinga-Hogarty model, the court is wary of the defendants' application. In particular, the court perceives a rather result-oriented bent in the defendants' application of the Elzinga-Hogarty test; using the numbers to confirm a predetermined market rather than to help determine a market in the first place.
Using step one of DX 100, the defendants determined their service area or, in other words, the area in which the defendants receive 90% of their admissions. Utilizing 1985 patient origin data arranged by zip code, the defendants accomplish this task by tabulating the zip codes accounting for 90% of the hospital's admissions. Starting at the hospitals and building outward, the defendants constructed a service area encompassing an area equivalent to a circle with a twenty-five mile radius centered at the defendants' hospitals. The defendants however, curiously included within this "service" area three zip codes skirting the outside circle (60129, 53520, and 60111) that failed to contribute any patients to the defendants' admissions. (DX 118, 129).
Next, according to DX 100, is step 1(b). Step 1(b) is designed to determine the "collective service area" of all the hospitals located within the merging hospitals service area. Logically, step 1(b) would be completed by determining the service areas of hospitals located within the merging hospitals' service area and then aggregating all the service areas together to arrive at a "collective service area." Dr. Lynk, the defendants' expert, described the tasks he performed in preparing step 1(b) as follows:
starting with the service area, I have simply identified those hospitals located in zip codes that are literally adjacent to that service area as alternatives that at least those patients living in that part of the service area would turn to as alternatives.
identify other hospitals that although not literally continguous to the service area were about equally far from Rockford as other zip codes that had previously been identified by this principle of contiguity that I mentioned earlier
Through this method, Dr. Lynk categorized "Winnebago area" hospitals as an "A" if they were within the merging hospitals service area, as a "B" if they were contiguous to the service area, and as a "C" if although not contiguous the hospitals were almost as close as hospitals that are contiguous. (Tr. 1186; DX 119).
When pressed on cross-examination as to whether the hospitals categorized as a "B" or a "C" on the defendants' exhibit 118 were identified using Elzinga-Hogarty methodology, Dr. Lynk responded:
the delineation in Exhibit 118, the ancillary analysis of the outlying counties in Ex. 9. I mean, these are not -- I think these are useful inquires but these are not, strictly speaking, a part of the Elzinga-Hogarty test itself. So, they get at some of the dynamics of patient origin and destination apart from being part of the formal methodology of the list
. . . that the intuitive logic was that once you found the service area, they could opt out of the service area just as easily as opting into the service area, and identified hospitals that were literally contiguous to or adjacent to the service area and suggested them as obvious candidates.
I certainly looked at the geographic distribution of these hospitals; and based not formally on the test, but on observation of the distances, they appear to be potentially competitively significant.
Based on Dr. Lynk's testimony on cross examination, it is obvious that DX 118 does not represent a formally completed step 1, (a) and (b). DX 118 fails to present a consolidation of the merging hospitals' service area with the service areas of hospitals located within the merging hospitals' service area. In more concrete terms, DX 118 does not represent the combining of the service area of RMH and SAH with St. Anthony, Beloit Memorial, St. Joseph's, Highland, and Rochelle Community's service areas.
Moreover, even taking Dr. Lynk's version of step "1(b)" at face value, there are problems with the "intuitive logic" employed. First, as mentioned earlier, Dr. Lynk tacked on three zip codes into the RMH and SAH service area, even though none of them contribute to either RMH or SAH admissions. Dr. Lynk now uses these "empty" zip codes as the boundaries of the merging hospital's service area that are contiguous to three "B" category hospitals - Kishwaukee, Sycamore and Mercy Janesville. In effect, these hospitals are not contiguous since these zip codes should not have been included in the merging hospitals service area in the first place.
Second, Dr. Lynk estimates that "they [patients] could opt out of the service area as easily as opting into the service area." (Tr. 1208). Presumably, Dr. Lynk was referring to the patients in the zip codes skirting the edges of the service area ("at least those patients living in that part of the service area would turn to alternatives"); in that it would be just as easy, if not easier, for these patients to travel out of the service area as into the service area. (Tr. 1114-15, 1208). The zip codes bordering the edge of the 25 mile radius service area, however, account for only 3.8% of the 90% of patients admitted to RMH and SAH. This is a rather thin reed to justify adding 12 hospitals. (DX 118, 129).
Finally, Dr. Lynk described the hospitals contiguous to the service area as "obvious candidates." (Tr. 1208). Included among those obvious candidates is Sycamore Hospital. In DX. 118, Dr. Lynk is suggesting that Sycamore Hospital should be included in the geographic market because of its ability to constrain the merging hospitals' exercise of market power. Sycamore Hospital, however, is in the process of terminating its inpatient operations. (Tr. 1655-56). While the inclusion of Sycamore Hospital may simply be an oversight, the oversight exposes Dr. Lynk's determination of the collective service area as a "seat of the pants" estimate made under the guise of an empirical study.
When asked if a Lifo figure determined by step 1(a) and (b) of exhibit 100 was ever calculated, Dr. Lynk responded as follows:
You mean the Lofi (patient immigration) figures
That wasn't specifically tabulated or summarized in a table form. Its simply a list of zip codes with a cumulative percentage just hits 90.
My recollection is it is the yellow area [on DX 118] plus somewhere around the order of four to a half a dozen additional zip codes.
As I say, I haven't separately tabulated it. It would have been part of my workpapers.
The preceding testimony basically concedes that step 1(b) as officially set out in DX 100 was never actually tabulated. Dr. Lynk, however, indicates that DX 102 is an application of the Elzinga-Hogarty test and that "based on Exhibit 102, the test [Elzinga-Hogarty] itself suggests that indeed they [continguous and near-contiguous hospitals - categories "B" and "C"] are [competively significant], certainly on a charge basis." (Tr. 1210).
Dr. Lynk's assertion that DX 102 is an application of Elzinga-Hogarty is not entirely accurate since the test, as outlined in DX 100, is a cumulative test using the results of the first step (Lofi) in calculating the second step (Lifo) In this case, step 1 was not performed using the Elzinga-Hogarty methodology outlined, and if the "collective service area" was not determined via Elzinga-Hogarty then the second step is tainted with ad hoc data. Nor can DX 102 corroborate, as suggested by Dr. Lynk, the assumptions of DX 118 since the data in DX 102 is not independent of the calculations of DX 118.
DX 104 and DX 105
The defendants next turn to two exhibits (DX 104 and 105) which are based on patient charge and admission data, respectively, to demonstrate Elzinga-Hogarty calculations on their proposed ten county market. The ten county market corresponds to the ten counties that encompass the hospitals identified in DX 118. (Tr. 1135). The calculations for the ten county area are straightforward and yield the following results:
Admission data: Lofi ...