Summary
approving "method of building the service area by including zip codes according to the number of patients each contributes. . . ."
Summary of this case from California v. Sutter Health SystemOpinion
No. 95-5015-CV-SW-1.
June 9, 1995.
Janice Charter, Denver, Colorado, for plaintiff.
Antonio deBlasio, Thomas Campbell, Chicago, IL, David L. Taylor and Paul G. Taylor, Webb City, MO, for defendants.
ORDER
Pending before the Court is Plaintiff's motion for preliminary injunction. Said motion together with Plaintiff's suggestions in support, Defendants' suggestions in opposition, Plaintiff's reply suggestions, and all proposed findings of fact and conclusions of law were reviewed by the Court. After due consideration of the above and all of the evidence submitted at the evidentiary hearing held on March 23 and 24, 1995, for the reasons set forth below, the motion is denied.
I. FACTUAL BACKGROUND A. Procedural History
The City of Joplin, Missouri, which has a population of approximately 40,000, is located in the extreme southwestern corner of Missouri, roughly five miles from the Kansas state line, twelve miles from the northeast corner of Oklahoma, and 40 miles from the northwest corner of Arkansas. On February 28, 1994, Freeman Hospital ("Freeman") and Tri-State Osteopathic Hospital Association, Inc. d/b/a Oak Hill Hospital ("Oak Hill"), two general acute care hospitals located within the Joplin city limits, entered into an agreement to consolidate their operations. Under the planned consolidation, all of the assets, liabilities, activities, and management of the two hospitals will be combined to form a new nonprofit corporation to be known as Health Southwest Alliance of Missouri, Inc.At present, Freeman is a nonprofit Missouri corporation that operates a 158-bed general acute care hospital and is the secondlargest of the three hospitals located within the Joplin city limits. Freeman also owns Freeman-Neosho Hospital in Neosho, Missouri ("Freeman-Neosho"), a 67-bed facility located approximately 15 miles southeast of Joplin. Both Freeman and Freeman-Neosho have historically been allopathic, as opposed to osteopathic, hospitals. Oak Hill, the smallest hospital in the city of Joplin, is a nonprofit Missouri corporation that operates a 96-bed general acute care facility staffed primarily by osteopathic physicians. Joplin's third and largest general acute care hospital is St. John's Regional Medical Center ("St. John's"), an allopathic hospital with 331 beds. There are a number of other hospitals nearby, however, the parties dispute which of them should be considered relevant in analyzing the competitive impact of this proposed consolidation. All other hospitals in the markets proposed by the parties are allopathic.
Defendants filed the necessary pre-merger notification form with the Federal Trade Commission ("FTC") as required by Section 7A of the Clayton Act, 15 U.S.C. § 18a, on July 21, 1994. This provision imposes a statutory waiting period of thirty days before any merger may be consummated to allow the FTC an opportunity to review the consequences of the transaction. On August 19, 1994, the FTC issued a second request for information to Defendants. After a response to the second request was submitted, the parties agreed to several extensions of the waiting period during which time the FTC continued to investigate the proposed transaction and its expected competitive effects. The last agreed extension expired at 11:59 p.m. on February 22, 1995. On February 24, 1995, the FTC filed the complaint in this action.
The FTC seeks injunctive relief to prohibit this consolidation pending final disposition of administrative proceedings that will determine whether this transaction violates section 7 of the Clayton Act. 15 U.S.C. § 18. Initially, the FTC sought a temporary restraining order ("TRO") to restore and maintain Oak Hill as an independent hospital. After conducting a TRO hearing on February 22, 1995, the Court denied the FTC's request for a temporary restraining order. The Court then held a preliminary injunction hearing on March 23 and 24, 1995.
On February 24, 1995, the FTC filed an emergency motion for injunction pending appeal with the United States Court of Appeals for the Eighth Circuit. On March 1, 1995, Defendants consummated their transaction when articles of consolidation were filed with the Missouri Secretary of State. The new entity's board of directors met and undertook various routine corporate actions such as appointing officers. Later that day, the Eighth circuit entered an order remanding the matter to this Court for an evidentiary hearing on the FTC's complaint for injuncitve relief. The Eighth Circiut retained jurisdiction and enjoined "the defendants from further consolidation of their hospitals or assets." After further motions and hearings before the Eighth circuit, the parties agreed to operate the hospitals on a "hold separate" basis pending this Court's decision on the preliminary injunction motion.
B. The Relevant Geographic Market
At the preliminary injunction hearing, the establishment of the relevant geographic market for the merging hospitals boiled down to a battle of the experts with the combatants as follows: Professor Keith Leffler for the FTC versus Dr. William Lynk for Defendants. In their respective analyses, both experts relied on the Elzinga-Hogarty test. The Elzinga-Hogarty test is comprised of two measures of historical service flows into or out of a proposed geographic market. The first measure, Little Out From Inside ("LOFI"), represents the percentage of patients served by area hospitals who reside within the area. The second measure, Little In From Outside ("LIFO"), represents the percentage of patients who remain in the area for hospital services rather than go outside the area.
While this test is often utilized to analyze geographic markets, it is not solely dispositive of the geographic market determination. Other evidence may be relevant.
Both experts spent considerable time debating the proper cut-off percentage to use in establishing an accurate geographic market. Generally, applying the Elzinga-Hogarty test at a 75 percent inclusion criterion results in a "weak" market definition. By contrast, an inclusion criterion of 90 percent is considered the minimum standard necessary to have a "strong" market definition that accounts for all important competitive influences. Although the authors of the Elzinga-Hogarty test initially suggested a 75 percent cutoff criterion, they now suggest that 90 percent is the more appropriate cutoff threshold. See Elzinga Hogarty, The Problem of Geographic Market Delineation Revisited: the Case of Coal, 23 Antitrust Bulletin 1 (1978); In re Adventist Health System/West, (available on WESTLAW, FATR-FTC database), 1994 FTC LEXIS 54, at *19. The FTC has not officially endorsed the use of either the 75 or 90 percent figure.
As the inclusion percentage decreases, the number of hospital patients who are ignored in the analysis increases. For example, a 90 percent inclusion standard ignores 10 percent of the hospitals' patients, whereas a 75 percent standard ignores 25 percent of the patients who use the hospitals.
1. Determining service area
The first step in evaluating the relevant geographic market is to determine where the patients of the Joplin hospitals come from. It is undisputed that the following hospitals in the immediate Joplin area are in competition with one another: Freeman, Oak Hill, St. John's, Freeman-Neosho, and McCune-Brooks Hospital. The area from which these hospitals get their patients is referred to as their "service area." Regardless of which inclusion percentage is used, the measurement of service area requires a choice of geographic units to construct and define the area. Such units can be comprised of, for example, cities, zip codes, or entire counties. Whatever the building blocks, the service area is constructed by adding successive geographic units until they cumulatively account for enough of the hospitals' patients to meet the specified inclusion percentage.
a. Professor Leffler's service area
Although Professor Leffler proposed several alternatives, he expressed his preference for a service area consisting of all zip codes within a 27-mile radius of Joplin. Plaintiff's Exhibit 111. In 1993, his proposed service area, consisting of 44 zip codes in Jasper, Newton, and McDonald Counties in Missouri, Cherokee County in Kansas, and Ottowa County in Oklahoma, collectively accounted for 80.8 percent of the Joplin area hospitals' 24,336 patient admissions. While this criterion exceeds the 75 percent "weak market" inclusion criterion, it is closer to a "weak" than a "strong" market standard. However, the accuracy of Professor Leffler's proposed service area is questionable for several reasons.
First, Professor Leffler's preferred service area ignores several cities and zip codes that contribute significant numbers of patients to the Joplin area hospitals. For example, Professor Leffler's service area excludes zip code 66762 in Crawford County, Kansas, which contributed 533 patients to the Joplin area hospitals in 1993 according to his own tabulation. Plaintiff's Exhibit 111. If ranked according to the number of patients contributed, this zip code, which encompasses the city of Pittsburg, Kansas, would have ranked ahead of 35 other zip codes that Professor Leffler did include. By excluding Pittsburg, Kansas, Professor Leffler excluded Mt. Carmel Medical Center, an 82-bed facility, from his proposed geographic market. Professor Leffler's service area also excluded zip code 64759 in Barton County, Missouri that contributed 317 patients to the Joplin area hospitals in 1993. In addition, Professor Leffler's service area includes zip codes that provided as few as seven patients to the Joplin area hospitals, Jasper County zip code 64833, and nine patients, Ottawa County zip code 74358. Plaintiff's Exhibit 111.
Second, the initial service area proposed by Professor Leffler was based on whole counties rather than specific zip codes. Plaintiff's Exhibit 108, at 8. However, after learning of Dr. Lynk's zip-code based service area, Professor Leffler altered his approach and acknowledged that a zip code-based analysis provides a greater degree of accuracy than a service area based solely on counties. Furthermore, in making his initial service area determination, Professor Leffler relied upon a data diskette supplied by the Hospital Industry Data Institute ("HIDI"), which initially indicated that the Joplin area hospitals had a total of 19,879 patient admissions in 1993. Plaintiff's Exhibit 108, at table 1. However, there is some question as to the accuracy of this number, as HIDI's published numbers report that the 1993 patient total was 24,336, not 19,879. Defense Exhibit 19. Professor Leffler's use of this erroneous data casts doubt on the reliability of his calculations.
The number of total admissions listed by the American Hospital Association and the Missouri Hospital Association are also consistent with a total of approximately 24,336. Defense Exhibit 19.
b. Dr. Lynk's service area
Dr. Lynk used a 90 percent standard to define his service area, which utilized zip codes and towns as geographic building blocks. Using data for 1993 published by HIDI, Dr. Lynk added towns to the service area in descending order of importance to the hospitals as sources of patients. A total of 45 towns were required to account for 90 percent of the Joplin area hospitals' admissions. Defense Exhibit 13. This service area accounted for 21,905 of the Joplin area hospitals' patients, whose total patient count from all areas was 24,336 in 1993. Dr. Lynk's service area includes cities and towns located within thirteen different counties in the states of Missouri (seven counties), Kansas (four counties) and Oklahoma (two counties). Defense Exhibits 14 and 15. The most important sources of the Joplin area hospitals' patients are the counties of Jasper and Newton in Missouri (15,667 patients from both counties combined), followed in order by the counties of Cherokee, Kansas (2,238 patients), Ottawa, Oklahoma (1,197 patients), McDonald, Missouri (1,018), and Crawford, Kansas (693 patients). Defense Exhibit 17. Of all thirteen counties containing towns within Dr. Lynk's service area, Bourbon County, Kansas was the smallest in terms of patients admitted to the Joplin area hospitals with 119 patients.
Dr. Lynk's service area includes the following seventeen acute-care hospitals, located in the thirteen counties noted above: Freeman, Newton County, MO; Oak Hill, Newton County, MO; St. John's, Newton County, MO; McCune-Brooks Hospital, Jasper County, MO; Freeman Neosho Hospital, Newton County, MO; Maude Norton Memorial City Hospital, Cherokee County, KS; Baptist Regional Health Center, Ottawa County, OK; Mt. Carmel Medical Center, Crawford County, KS; Barton County Memorial Hospital, Barton County, MO; Cox-Monett Hospital, Barry County, MO; Grove General Hospital, Delaware County, OK; Crawford County Hospital District One, Crawford County, KS; LaBette County Medical Center, Labette County, KS; South Barry County Memorial Hospital, Barry County, MO; Aurora Community Hospital, Lawrence County, KS; Mercy Hospital of Kansas, Bourbon County, KS; Nevada Regional Medical Center, Vernon County, MO. Defense Exhibit 18. All seventeen hospital campuses are located within 54 miles of Joplin. Twelve of these are within 40 miles of Joplin, and the remaining five are between 40 and 54 miles of Joplin.
2. Patient Outmigration
Once a service area has been defined, the second step in formulating the relevant geographic market is to identify other hospitals to which patients residing in the service area could turn if they were dissatisfied with the prices or services of the merging hospitals. Those hospitals, in conjunction with the Joplin area hospitals, constitute the competitors in the relevant geographic market. There are two methods to identify these competitively relevant alternative hospitals. One is to examine current usage — the actual hospital choices that the residents of the service area have historically made prior to the merger. This approach requires an examination of data on hospital usage patterns for all residents of the service area. If the service area residents currently use other hospitals, then those hospitals are considered alternatives to the Joplin area hospitals. However, because this method of identifying relevant competitors is limited to evaluating pre-merger usage patterns, it is incapable of identifying competitors who will become relevant if the prices of the Joplin area hospitals rise in the future.
The other method to determine competitively relevant alternative hospitals is by recourse to the principle of geographic proximity. Under this approach, a hospital is deemed to be competitively relevant to a service area town if that hospital is closer to the town in question than the Joplin hospitals. While the residents of a particular service area town may currently use the Joplin area hospitals, in the future they may choose to visit a closer hospital as a convenient alternative to the Joplin area hospitals. This geography-based approach identifies alternative sellers that hospital consumers can turn to without traveling any further than they currently are for hospital care.
a. Professor Leffler's analysis
Professor Leffler relied on the current usage method to identify the competitors in his relevant geographic market. To satisfy the second prong of the Elzinga-Hogarty test Professor Leffler attempted to determine where the patients in his service area go when they choose not to use the Joplin area hospitals. In forming his conclusions, Professor Leffler relied primarily on data provided by HIDI, supplemented by limited statistics from the Kansas Hospital Association. In quantifying the extent of outmigration, he determined that 78 percent of the patients from his service area do not leave the service area for hospitalization. Plaintiff's Exhibit 112.
However, Professor Leffler's attempt to analyze the flows of patients going to hospitals outside the service area is flawed because it is based on incomplete data that ignores significant patient outflows. Complete data on how local hospitals and state hospital associations conduct their work is not available due primarily to the fact that the area in question is located at the junction of four states. The HIDI data that Professor Leffler relied on does not include information for any hospitals in Oklahoma, Arkansas or any of the hospitals in the four counties in the southeast corner of Kansas. Presumably because the data is unavailable, the record contains no analysis of the volume of patient outflows from southwest Missouri to hospitals in southeast Kansas or northeast Oklahoma. Thus, it is impossible to perform a proper patient outmigration analysis. Because of the noted deficiencies in the data, it is not possible to accurately rely upon the current usage method to identify hospital alternatives for inclusion within the relevant geographic market.
The FTC also claims that its relevant geographic market is accurate because, among other things, the Joplin hospitals view themselves as their own principal competitors and outside hospitals do not perceive themselves to be in competition with the Joplin hospitals for most purposes. Moreover, the FTC asserts that Defendants' proposed thirteen county market is inconsistent with the perceptions of third-party hospital administrators, third-party payors, and the parties themselves. While such non-empirical data may have some probative value as a starting point to evaluate this market, such data will not carry the FTC's burden. Informal, off-the-cuff remarks and anecdotal evidence concerning the marketplace are no substitute for solid economic analysis.
b. Dr. Lynk's analysis
In contrast to Professor Leffler, Dr. Lynk used the principle of geographic proximity to determine the competitors in his relevant geographic market. Dr. Lynk first identified thirteen counties that contain the towns comprising the Joplin area hospitals' service area. He then determined that all of the acute-care hospitals within these counties were relevant competitors of the Joplin area hospitals. The basis for that determination was his finding that most residents of the service area towns were at least as close to the hospitals in their own home counties as they were to the Joplin area hospitals. Thus, if each service area town contains relevant consumers, and if the hospitals in those towns are more geographically convenient alternatives than the Joplin hospitals, it follows that the hospitals in those counties are relevant alternatives, and on that basis they should be included in the relevant geographic market. Dr. Lynk found that patients in each of the service area counties presently contributing patients to the Joplin area hospitals could just as easily seek care in a hospital in their home county. Because the service area extends into parts of thirteen counties, there is no reason to think that hospitals that are physically located in those counties are not equally good geographic alternatives to the hospitals in Joplin. In formulating his geographic market definition, Dr. Lynk did not attempt to analyze the second prong of the Elzinga-Hogarty analysis, which requires a patient outflow calculation. In order to complete this prong it is necessary to have accurate and complete information concerning what hospitals the patients in the service area actually visit. However, as noted above, in the present case no such reliable data is available.In considering the competing market definitions, the Court gives significantly greater weight to Dr. Lynk's relevant geographic market. First, his method of building the service area by including zip codes according to the number of patients each contributes is more economically sound than Professor Leffler's method of arraying zip codes out to an arbitrary cutoff 27 miles from Joplin. Second, Dr. Lynk's analysis of geographic proximity, in lieu of the second prong of the Elzinga-Hogarty test, directly evaluated where patients could practicably turn for alternative sources of acute care inpatient services. The Eighth Circuit requires such an inquiry. Morgenstern v. Wilson, 29 F.3d 1291, 1296 (8th Cir. 1994). Accordingly, the Court adopts Dr. Lynk's relevant geographic market.
C. Competitive Effects 1. Concentration
Once a relevant market has been established, the Court must next determine whether the proposed consolidation will have anti-competitive effects. Concentration analysis is a standard means of quantifying the likelihood that, with fewer competitors in the future, the firms in a market can collude or exercise market power to the detriment of consumers. See United States v. Philadelphia National Bank, 374 U.S. 321, 83 S.Ct. 1715, 10 L.Ed.2d 915 (1963); Federal Trade Commission v. National Tea Company, 603 F.2d 694 (8th Cir. 1979). Market concentration is one of the primary factors affecting the feasibility of successful collusion. As concentration in a particular market increases, the "greater is the likelihood that parallel policies of mutual advantage not competition, will emerge." United States v. Aluminum Co. of America, 377 U.S. 271, 280, 84 S.Ct. 1283, 1289, 12 L.Ed.2d 314 (1964). A common method of measuring market concentration is the Herfindahl-Hirschmann Index ("RHI"). The Department of Justice's Merger Guidelines define a market as "highly concentrated" if the post-merger HHI exceeds 1800 points. Moreover, economists generally maintain that collusion is likely when the HHI exceeds 2000. Transcript at 100-02, testimony of Professor Leffler.
An HHI of 2,000 is representative of a market with five equal-sized competitors. An HHI of 10,000 denotes a complete monopoly.
The post-merger levels of market concentration, based upon Professor Leffler's proposed geographic market all exceed 1800. Plaintiff's Exhibit 113. Professor Leffler applied his HHI analysis to several alternative geographic markets. His first market, comprised of Jasper, Newton, McDonald and Cherokee Counties, rendered an Hill of 3887 based on beds and 4356 based on admissions. His second market, based on the county equivalents of his 27 mile geographic market, yielded HHI's ranging from 3088 to 3417. Hill's for a third market, comprised of Jasper, Newton, McDonald, Cherokee, Crawford and Ottowa Counties, ranged from 2288 to 2578. According to Professor Leffler, the fluctuations in Hill caused by the merger of Freeman and Oak Hill range from a high of 986 to a low of 408, depending upon which measure of Hill and which proposed market are used.
The post-merger Hills for Dr. Lynk's proposed geographic market range from 1322 based upon the number of beds, to 1496 based on admissions, to 1624 based on shares of patient census. The corresponding merger-related fluctuations in the Hill are 189 based on beds, 251 for admissions, and 222 based on patient census. Defense Exhibit 21. None of these measurements exceed the 1800 benchmark for a "highly concentrated" market. Moreover, using Dr. Lynk's relevant geographic market, Defendants' post-merger market share would be approximately 24.4 percent based on hospital admissions, and 21.2 percent based on number of beds. Defense Exhibit 21. This combined market share for the consolidated entity is only marginally higher than Freeman's current market share. In addition, based on either expert's relevant market, the consolidation will reduce the number of hospitals by only one. In Dr. Lynk's geographic market, this will leave a total of fourteen unaffiliated hospitals post-consolidation. Defense Exhibit 18.
Any analysis of market concentration is necessarily wedded to a specifically defined relevant geographic market. Because the Court has adopted Dr. Lynk's relevant geographic market for the reasons noted above, the Court finds that the post-merger levels of Hill will not exceed 1800, thus, this market cannot be considered highly concentrated. Based on this concentration analysis, the FTC has failed to demonstrate that competition will be substantially lessened as a result of this consolidation. However, the Court notes that estimating market concentration is only the starting point in assessing the likelihood of anti-competitive effects. See United States v. Baker Hughes, Inc., 908 F.2d 981, 984 (D.C. Cir. 1990). Numerous other factors play a role.
2. Nonprofit Status
The FTC essentially ignores the relevance of the merging hospitals' status as nonprofit entities. However, there is an economic basis for seriously questioning the assumption that nonprofit status is a distinction that does not matter. This is not to say that nonprofit organizations are inherently more altruistic than their for-profit counterparts. To the contrary, nonprofits generally have an incentive to maximize profits whenever practical, just like private sector organizations. However, by simply doing what is in their own economic best interest, certain nonprofit organizations ensure a competitive outcome, regardless of market structure.
Arguably, a private nonprofit hospital that is sponsored and directed by the local community is similar to a consumer cooperative. It is highly unlikely that a cooperative will arbitrarily raise prices merely to earn higher profits because the owners of such an organization are also its consumers. See Henry B. Hansmann, The Role of Nonprofit Enterprise, 89 Yale L.J. 835, 889 (1980). Similarly, if a nonprofit organization is controlled by the very people who depend on it for service, there is no rational economic incentive for such an organization to raise its prices to the monopoly level even if it has the power to do so. William J. Lynk, Property Rights and the Presumptions of Merger Analysis, Antitrust Bulletin, 363, 377 (1994). In the hospital context, this rationale applies to nonprofit hospitals whose boards are effectively controlled by persons representing the interests of hospital consumers or other groups that desire competitively-priced hospital services. Id.
To determine if this theory is applicable in a given context, it is necessary to evaluate who controls the hospitals. Local businesses have an interest in competitively-priced hospital services, because they pay those prices through the cost of their employees' hospitalization. Physicians also have an interest in maintaining competitive prices at the hospitals where they practice because higher prices drive patients to alternative hospitals, thus causing a reduction in the physicians' patient volume.
As of March 1, 1995, the combined Freeman-Oak Hill Board of Trustees was composed of eighteen business and community leaders from the southwest Missouri area who serve without compensation. Defense Exhibit 32. Eight Trustees are owners or employees of local businesses and another four are retired owners or employees of local businesses. Five Trustees are local physicians who practice at Freeman and or Oak Hill and one Trustee is a Freeman administrator. Thus, the vast majority of the combined Board of Trustees is comprised of persons who indirectly represent the interests of hospital consumers. Consequently, it would not be in these individual Board member's best economic interest to permit prices to be raised beyond a normal competitive level. Interestingly, John Hale, Chairman of the Tri-State Health Care Coalition, a group of twenty Joplin area employers formed in 1993 to seek lower prices for health care, made a similar observation when he was asked about the future prices of the consolidated entity. Plaintiff's Exhibit 145, at 115-16.
3. Barriers to Market Entry
In assessing competitive effects, the Court must also evaluate the possibility and likelihood of new competitors entering the relevant geographic market. If barriers prevent new firms from entering a market, the existing firms can theoretically raise prices without fear of new competition. Some states, such as Missouri, have Certificate of Need ("CON") laws that require state approval before hospitals or other health care facilities may be built or expanded. The Missouri CON statute requires hospitals to obtain state approval before undertaking any capital projects exceeding $600,000 within a two year period, including opening a new hospital or adding beds at an existing hospital, or making any purchase of equipment exceeding $400,000. Plaintiff's Exhibit 2, at 43-44. The CON laws in Kansas, Oklahoma and Arkansas impose no restrictions on the construction of inpatient facilities.
Defendants point out that a would-be entrant into the Joplin area hospital market would not necessarily need to construct a new facility from scratch. There are currently two closed hospital campuses located in the immediate Joplin area, in Webb City, Missouri, and Baxter Springs, Kansas, although both contain less than fifty licensed beds. There is no financial impediment preventing a new entrant from purchasing either of these facilities and operating them as acute care inpatient hospitals. Further, there are five closed hospital facilities within 40 miles of Joplin in which new medical facilities could be constructed, or existing medical facilities could be expanded. Defense Exhibit 9. However, on balance, the Court finds that there are some substantial barriers facing potential entrants into the relevant geographic market. Among these are the high costs of constructing new or refurbished medical facilities and the detailed CON compliance procedures in Missouri.
4. Other Factors a. third-party payors
The market for hospital services in the Joplin area is dominated by large, sophisticated payers. Notably, no third-party payors or customers have expressly objected to this consolidation. This fact is significant because these payors will ultimately bear the brunt of any higher prices caused by the consolidation. James Stival, the Manager of Managed Care for Wal-Mart Stores, Inc. who is responsible for the health-care benefits of Wal-Mart employees throughout the United States, including approximately 23,700 covered lives in Missouri, was interviewed by the FTC on February 15, 1995, regarding his views on the proposed consolidation of Freeman and Oak Hill. He stated that the consolidation between Freeman and Oak Hill "is a great idea and will enhance competition in the Joplin region." Defense Exhibit 72, at 1. Curiously, the FTC did not ask Mr. Stival to submit a declaration in support of their investigation.
In addition, Sheila Maerz, the Personnel Manager for the City of Joplin who supervises health benefit plans for city employees and their dependents, stated in an affidavit dated March 13, 1995, that she did not believe this consolidation would reduce competition. She noted that "competition relating to the provision of health care services had never been so brutal as it had been in the past twelve to fourteen months" and there is a "great deal of animosity between Freeman Hospital and St. John's Regional Medical Center that increased competition and reduced the possibility of any collaboration between the hospitals." Defense Exhibit 71, at 1. She added that "new city employees choose Oak Hill . . . because of its osteopathic identity and a perception that it offers lower quality than the other two acute care hospitals in Joplin." Id.
Furthermore, John Hale — Vice-President of Personnel at Legget and Platt, a local manufacturing firm with 2,200 employees in southwest Missouri, and Chairman of the Tri-State Health Care Coalition — stated in his deposition that Oak Hill does not represent a viable competitor to Freeman or St. John's. For this reason, his firm did not award a contract to Oak Hill when it considered awarding contracts for various medical services. He also asserted that the merger would not detract from competition in the Joplin community. Defense Exhibit 73, at 1-2. Moreover, the FTC's expert witness, Professor Leffler, acknowledged that not a single third-party payor has voiced any objection to this consolidation. This fact weighs against a finding of anticompetitive effects. See In re Adventist Health System/West, (available on WESTLAW, FATR-FTC database), 1994 FTC LEXIS 54, at *50.
While the consolidation of Freeman and Oak Hill will cause a slight increase in concentration, the likelihood is great that it will not harm competition. After the consolidation, the new entity will be closer in size to St. John's, clearly the Joplin market leader, and will have equivalent resources and assets allowing it to offer comparable services. Forging a more formidable competitor for St. John's could have very beneficial effects. The consolidated Freeman-Oak Hill entity will be able to compete more effectively with St. John's because it will be better positioned to respond to requests for proposals from managed care plans on a comparable level to St. John's. See Defense Exhibit 71, at 2; Defense Exhibit 72, at 2. Furthermore, payors can play these two major competitors off each another if prices are considered too high. Plaintiff's Exhibit 128, at 46-47. The consolidation may also make possible the creation of significant economic efficiencies through less overhead expenses and less administrative duplication. The current plan of consolidation calls for moving most of the acute care inpatient activities to the Freeman campus, thereby freeing the Oak Hill campus to perform other services, including ambulatory and outpatient work.
The Court finds that the risk of successful collusion or coordinated behavior among the Joplin hospitals in the future is slight. This due in part to the consumer responsiveness of the Boards of Trustees. In addition, the evidence presented by the FTC failed to demonstrate a history of successful prior cooperation indicative of trust and forbearance that would support an inference of anticompetitive consequences in the future. The FTC's warnings of future collusion based on past meetings between certain hospital executives is necessarily speculative and also unpersuasive. The evidence does not support a finding that these hospitals are unusually disposed to cooperate. See Hospital Corporation of America v. FTC, 807 F.2d 1381, 1388-89 (7th Cir. 1986), cert. denied, 481 U.S. 1038, 107 S.Ct. 1975, 95 L.Ed.2d 815 (1987).
b. private-pay percentages
Oak Hill's competitive significance is diminished by the fact that among the five Joplin area hospitals it has the lowest percentage of privately insured ("private-pay") patients. Because reimbursement amounts for Medicare and Medicaid patients are dictated by the government, the exercise of monopoly power through higher prices has an adverse effect only on private-pay patients. Based on a 1993 patient census, only 18.3 percent of Oak Hill's patients were private-pay. Defense Exhibit 23. Inversely, 81.7 percent of Oak Hill's patients relied on Medicare and Medicaid for payment. The 1993 private-pay percentages for the other Joplin area hospitals were as follows: St. John's 29.0 percent; Freeman 38.8 percent; Freeman-Neosho 31.5 percent; and McCune-Brooks 43.7 percent.
According to internal monthly operating statistics regularly compiled by Freeman and Oak Hill, over the period from January 1989 through January 1995 Oak Hill's private-pay percentage of its patient census has steadily declined, standing at just 12.7 percent of Oak Hill's census as of January 1995. Defense Exhibit 26. The degree of success in attracting private-pay patients is a measure of competitive significance, at least as it relates to the potential for influencing prices for private-pay patients. This illustrates that in relation to the other Joplin area hospitals, Oak Hill is much less competitively significant.
c. declining operating statistics at Oak Hill
Oak Hill's total inpatient admissions began to decline in early 1992 and, apart from a short-lived recovery from January to March 1994, have continued to decline ever since. Defense Exhibit 26. More recent data attest to Oak Hill's continuing decline in patient volume, financial sustainability, and competitive significance. Id. Oak Hill's average daily patient census has been in decline since early 1992, as has its daily census of newborn infants which stood at only 2.57 newborns in January 1995.
Oak Hill's monthly net income, total revenues less total costs, began a steady decline in early 1992 that has continued with a loss of $199,012 for January 1995 alone. Defense Exhibit 26. Oak Hill's net operating income, which is similar to net income except that it is based only on revenue from actual hospital operations and excludes revenue from charitable contributions and interest, has also declined significantly. Id. Oak Hill's net operating income fluctuated around the break-even level from early 1989 through 1991, but has fallen into the red ever since. At the end of January 1995, Oak Hill had a monthly net operating loss of $404,495.
Unlike Dr. Lynk, Professor Leffler primarily used only 1993 figures to analyze the competitive significance of Oak Hill. See Plaintiff's Exhibits 108, 112.
Because of its identity as virtually an exclusively osteopathic hospital, John Hale, Chairman of the Tri-State Health Care Coalition ("the Coalition"), stated that the Coalition did not feel its members could steer enrollees to Oak Hill. Defense Exhibit 74, at 2. Mr. Hale asserted that because most people who seek medical care do not seek medical care from osteopaths, employee groups would resist a health plan that sought to steer enrollees to an institution such as Oak Hill. The evidence clearly demonstrates that osteopathic hospitals have more difficulty competing than do allopathic hospitals. Defense Exhibit 95.
Another potential measure of competitive significance is the degree to which one seller in a market is able to underprice its rivals. The lower a seller is able to price its product, the greater is its competitive significance; conversely, the higher the seller must set its price, the lesser is its competitive significance. See Defense Exhibits 24 and 25. Mr. Hale noted that the Coalition viewed Oak Hill as a historically higher-priced hospital than its competitors.
Moreover, whenever patients are steered away from Oak Hill by their covered health plan, Oak Hill physicians risk losing patients. Even before the Coalition awarded its contracts to St. John's and Freeman, physicians on Oak Hill's staff were increasingly concerned about the prospect of losing patients because of its disfavored position in managed care contracting. Tim Weiss, a witness called by the FTC who has done consulting work for Oak Hill, stated that as a freestanding independent hospital, Oak Hill has a limited future of only two to three years. While other companies have expressed some interest in acquiring Oak Hill, the decision to sell to an outside entity or merge with Freeman is strictly one for the Board of Trustees of Oak Hill. As long as consolidation with Freeman does not adversely affect competition, Oak Hill is free to pursue that course as a rational alternative.
II. LEGAL ANALYSIS
Section 13(b) of FTC Act, which provides the standard for granting injunctive relief in this action, provides as follows:Whenever the Commission has reason to believe —
(1) that any person, partnership or corporation is violating, or is about to violate, any provision of law enforced by the Federal Trade Commission, and (2) that the enjoining thereof pending the issuance of a complaint by the Commission and until such complaint is dismissed by the Commission or set aside by the court on review, or until the order of the Commission made thereon has become final, would he in the interest of the public
the Commission by any of its attorneys designated by it for such purpose may bring suit in a district court of the United States to enjoin any such act or practice. Upon a proper showing that, weighing the equities and considering the Commission's likelihood of ultimate success, such action would be in the public interest, and after notice to the defendant, a temporary restraining order or a preliminary injunction may be granted without bond.15 U.S.C. § 53 (b). Thus, § 13(b) allows the FTC to obtain injunctive relief to prevent a consolidation pending an administrative adjudication of its legality. An FTC administrative law judge will ultimately make a determination on the merits. Consequently, the only issue before this Court is whether the FTC has made the necessary showing to justify preliminary injunctive relief. To prevail on its motion for preliminary injunction, the FTC must prove: (1) that there is a likelihood that it will ultimately succeed on the merits, in that this consolidation will substantially lessen competition and (2) that the balance of equities and the public interest favor the granting of an injunction.
A. Likelihood of Success on the Merits
Pursuant to section 7 of the Clayton Act, the FTC must demonstrate a likelihood that the effect of the proposed merger "may be substantially to lessen competition, or to tend to create a monopoly" within the relevant product and geographic markets. 15 U.S.C. § 18. To establish a prima facie case, the FTC must show that this acquisition will produce "a firm controlling an undue percentage share of the relevant market, and [would] result in a significant increase in the concentration of firms in that market." United States v. Philadelphia National Bank, 374 U.S. 321, 363, 83 S.Ct. 1715, 1741, 10 L.Ed.2d 915 (1963). At bottom, the FTC's case must "satisfy some minimum threshold of persuasiveness and be better than the defendant's case." United States v. Rockford Memorial Corp., 898 F.2d 1278, 1286 (7th Cir.), cert. denied, 498 U.S. 920, 111 S.Ct. 295, 112 L.Ed.2d 249 (1990).
1. The Relevant Geographic Market
The question of whether a merger substantially lessens competition can only be answered in terms of a relevant antitrust market. United States v. E.I. du Pont de Nemours Co., 353 U.S. 586, 593, 77 S.Ct. 872, 877, 1 L.Ed.2d 1057 (1957). The "[d]etermination of the relevant product and geographic market[s] is a necessary predicate" to establishing a claim based on Section 7 of the Clayton Act. United States v. Marine Bancorporation, 418 U.S. 602, 618, 94 S.Ct. 2856, 2868, 41 L.Ed.2d 978 (1974) (citations omitted). The Court finds that the relevant product market at stake is acute care inpatient hospital services. As illustrated above, defining the relevant geographic market is a more difficult task. A relevant geographic market is defined as "the area of effective competition . . . in which the seller operates, and to which the purchaser can practicably turn for supplies." United States v. Philadelphia National Bank, 374 U.S. 321, 359, 83 S.Ct. 1715, 1739, 10 L.Ed.2d 915 (1963) (citations omitted). The burden of proving the relevant geographic market rests with the FTC. United States v. Connecticut National Bank, 418 U.S. 656, 669, 94 S.Ct. 2788, 2796-97, 41 L.Ed.2d 1016 (1974); United States v. Empire Gas Corp., 537 F.2d 296, 302 (8th Cir. 1976), cert. denied, 429 U.S. 1122, 97 S.Ct. 1158, 51 L.Ed.2d 572 (1977).
The Court finds that Professor Leffler's preferred market definition, offered on behalf of the FTC, does not identify the relevant geographic market in which to analyze the competitive effects of the Freeman-Oak Hill merger. First, Professor Leffler failed to consider a comprehensive service area from which the Joplin area hospitals draw their patients. In addition, Professor Leffler attempted to determine where residents of the greater Joplin area actually went, as opposed to where they could practically go, for acute care inpatient hospital services. Such evidence is insufficient because it ignores other geographically proximate hospitals to which patients in the greater Joplin area can practically turn for alternative sources of acute care inpatient hospital services if Joplin hospital prices are raised above competitive levels. Morgenstern v. Wilson, 29 F.3d 1291, 1296 (8th Cir. 1994). The FTC's evidence fails to establish that, to the exclusion of other outlying towns or counties, a 27-mile radius around Joplin is the relevant geographic market. Because the FTC has not met its burden of establishing a valid geographic market, its complaint must fail. In re Adventist Health System/West, (available on WESTLAW, FATR-FTC database), 1994 FTC LEXIS 54, at *10.
2. Competitive Effects
The Court must evaluate many factors in assessing the likelihood of anti-competitive effects. Based on the relevant geographic market proposed by Dr. Lynk, the Court finds that the post-merger levels of HHI will not exceed 1800, thus, this consolidation will not cause the Joplin market to become highly concentrated. However, in the FTC's favor, it does appear that entry into the relevant market would be somewhat difficult. Although, the merging hospitals' status as nonprofit entities must also be considered. Because the combined Freeman-Oak Hill Board of Trustees is composed of individuals who indirectly represent the interests of hospital consumers, it would not be in their best economic interest to permit prices to rise beyond a normal competitive level. See William J. Lynk, Property Rights and the Presumptions of Merger Analysis, Antitrust Bulletin, 363, 377 (1994).
In addition, Oak Hill's significance as a competitive force is diminished by several factors. First, Oak Hill has a very low percentage of privately insured patients. Furthermore, Oak Hill's osteopathic identity is a deterrent to its obtaining managed care contracts. Also, Oak Hill's steadily declining inpatient admissions, net income and net operating income, as well as its higher prices, all attest to its continuing financial problems. Moreover, the fact that no third-party payors have objected to this consolidation weighs against a finding of anticompetitive effects. In the aggregate, the Court finds that above factors demonstrate the Freeman-Oak Hill consolidation will not foster an anticompetitive environment. Consequently, the FTC has not shown a likelihood of success on the merits.
B. Balance of Equities/Public Interest
Section 13(b) requires the FTC to demonstrate that the harm to the parties and the public that would flow from an injunction is outweighed by the harm to competition that would occur in the period between the injunction's denial and a final judgment on the merits. FTC v. Great Lakes Chemical Corp., 528 F. Supp. 84, 86 (N.D.Ill. 1981). However, the likelihood of harm to the government's position diminishes as the government's likelihood of success on the merits decreases. United States v. Country Lake Foods, Inc., 754 F. Supp. 669, 680-81 (D.Minn. 1990). The Court has already determined that the FTC does not have a likelihood of success in showing anticompetitive effects.
The primary public interest in this case is the interest in maintaining competitive hospital prices. The FTC has failed to show that there is a credible threat of harm to competition during the time between the denial of this preliminary injunction and the final adjudication of this case. Moreover, "[i]t is well recognized that the issuance of a preliminary injunction prior to a full trial on the merits is an extraordinary and drastic remedy . . . [because] the issuance of a preliminary injunction blocking an acquisition or merger may prevent the transaction from ever being consummated." FTC v. Exxon Corp., 636 F.2d 1336, 1343 (D.C. Cir. 1980) (citations omitted). If the Court issues an injunction, it is likely that Oak Hill will no longer be in business by the time the FTC gets around to conducting a hearing on the merits of this dispute. This factor must be balanced against the FTC's desire to maintain the status quo, so as to avoid having to "unscramble the eggs" later. Giving due regard to the public equities favoring the FTC's enforcement of the antitrust laws, the Court finds that enjoining the consolidation of Freeman and Oak Hill is not in the public interest.
The average time from the issuance of a complaint by the FTC to an initial decision by an administrative law judge averaged nearly three years in 1988. Moveover, additional time will be required if that initial decision is appealed. FTC v. Owens-Illinois, Inc., 681 F. Supp. 27, 53 n. 69 (D.D.C. 1998).
III. CONCLUSION
It is therefore ORDERED that Plaintiff's motion for preliminary injunction is DENIED. It is further
ORDERED that the Court hereby CERTIFIES the above findings of fact and conclusions of law for review by the United States Court of Appeals for the Eighth Circuit.