Hospital Corporation of America

Text (OCR of the scan at left; may contain errors)

IN THE MATTER 01' HOSPITAL CORPORATION OF AMERICA FINAL ORDER, ETC., IN REGARD TO ALLEGED VIOLATION OF SEC. 5 OF THE FEDERAL TRADE COMMISSION ACT AND SEC. 7 OF THE CLAYTON ACT Docket 9161. Complaint, July 30, 1982-Final Order, Oct. , 1985 This final order requires the nation s largest for-profit hospital chain bas in Nashville, Tenn., to divest North Park Hospital and Diagnostic Center hospital, both in Hamilton County, Tenn., and any medical offce buildings associated with the hospitals. The divestitures must be to different aequirers and obtain Commission approval. Respondent is also required to terminate its management contract with Downtown General Hospital, also in Hamilton County, and divest related real estate to a Commission-approved acquirer. Further, respondent is required to obtain FTC approval for any future acquisitions of certain hospitals in the Chattanooga, Tenn., area or any hospital meeting conditions specified in the order. Appearances For the Commission: M. Elizabeth Gee, Toby G. Singer, Garry R. Gibbs, Oscar M. Voss and Erika Wodinsky. For the respondents: Peter J. Nickles, William D. Iverson, K. Gregory Tucker and Michael A. Roth, Covington Burling, Washington C. and Margaret C. Mazzone, John W. Wade, Jr., Donald W. Fish and Jean L. Byassee in-house counsel, Nashville, Tenn. COMPLAINT The Federal Trade Commission having reason to believe that Hospital Corporation of America, a corporation subject to the jurisdiction of the Commission, has acquired the stock or assets of corporations subject to the jurisdiction of the Commission, hereinafter described in violation of Section 7 of the Clayton Act, as amended, 15 VB.C. 18 and having further reason to believe that respondent has engaged in unfair methods of competition in violation of Section 5 of the Federal Trade Commission Act, as amended, 15 V. C. 45, and that a proceeding in respect thereof would be in the public interest, hereby issues its complaint pursuant to the provisions of Section 11 of the aforesaid Clayton Act, as amended, 15 V. C. 21, and Section 5(b) ofthe Federal Trade Commission Act, as amended, 15 V. C. 45(b), stating its charges as follows:

Complaint 106 F I. DEFINITIONS 1. For the purposes of this complaint, the following definitions shall apply:

(a) Acute care hospital hereinafter sometimes referred to as hospital means an inpatient facility that furnishes care in connection with services of physicians for conditions for which nursing, medical or surgical services would be appropriate for care, diagnosis, or treatment, not including a facility specially intended for use in treatment of mental ilness, emotional disturbance or substance abuse. (b) The Health Service Area hereinafter referred to as HSA means the 13-county area encompassing the southeastern Tennessee counties of Hamilton, Bradley, (2) Marion, Sequatchie, Rhea, Meigs McMinn, Bledsoe, Grundy and Polk, and the northern Georgia counties of Catoosa, Dade and Walker.

(c) Hamilton County means Hamilton County, Tennessee, the county in which the city of Chattanooga is located. (d) The Chattanooga Standard Metropolitan Statistical Area hereinafter referred to as SMSA means the federally designated six-county area encompassing the southeastern Tennessee counties of Hamilton, Marion and Sequatchie, and the northern Georgia counties of Catoosa, Dade and Walker.

(e) A managed hospital means a hospital in which the owner (individual, corporate or public body) has contracted with a management company for that company to be responsible for the day-to-day operations of the hospital.

II. THE RESPONDENT 2. Respondent Hospital Corporation of America, hereinafter HCA" or !!respondent " is a corporation organized and existing under the laws of the State of Tennessee, with its principal executive offces at One Park Plaza, Nashvile, Tennessee. 3. HCA is primarily engaged in the operation and management of proprietary hospitals in the Vnited States and in foreign countries. It is the largest proprietary hospital chain in the united States and owns and operates acute care hospitals and psychiatric hospitals in 41 states throughout the country, including Tennessee. In HCA's Fiscal Year I981, it had total revenues of approximately $2.3 billon in connection with these hospitals.

4. At all times relevant herein, the respondent has been and is now engaged in or affecting commerce within the meaning of Section I of the Clayton Act, as amended, 15 VB. C. , and is a corporation whose business is in or affects commerce within the meaning of Section 5 of the Federal Trade Commission Act, as amended, 15 VB.C. 45. HCA 361 Complaint does business in a number of st"tes anti foreign countries. HCA and its hospitals in Hamilton County, the SMSA and the HSA, among other things: (3) (a) purchase substantial amounts of supplies, equipment and medicines in interstate commerce from sources outside of the State of Tennessee;

(b) receive substantial revenues in interstate commerce from private and governmental insurers located outside of the State of Tennessee; and (c) treat a substantial number of patients who travel from or reside outside the State of Tennessee.

5. Until the acquisitions described in Section III below, HCA owned only one acute care hospital, Parkridge Hospital, in Hamilton CO\lnty and the SMSA and only three acute care hospitals, Parkridge Hospital, Athens Community Hospital and Cleveland Community Hospital in the HSA. Until the acquisitions described in Section II below HCA' s psychiatric facilities in the HSA consisted only of one psychiatric unit in an acute care hospital, Cleveland Community Hospital. III. THE ACQUISITIONS Hospital Affiliates International, Inc.

6. On August 26, 1981, HCA purchased Hospital Affliates International, Inc. ("HAI"). In consideration thereof, HAl's parent corporation received approximately $425 milion in cash and approximately $225 milion in HCA voting stock.

7. Prior to its acquisition by HCA, HAI was a wholly owned subsidiary of INA Health Care Group, a wholly owned subsidiary of INA Corporation. HAI was a corporation organized and existing under the laws of Delaware, with its executive offces in Tennessee. HAI owned operated, and managed acute care hospitals and psychiatric hospitals in 33 states and several foreign countries. For its Fiscal Year 1980 HAl's revenues derived from its acute care and psychiatric hospitals were over $513 milion. HAI owned or managed three hospitals in Hamilton County: it owned Diagnostic Hospital; it managed Downtown General Hospital and Red Bank Community Hospital. HAI owned or managed five hospitals in the SMSA and the HSA: it owned Diagnostic Hospital and Sequatchie General Hospital; it managed Downtown General Hospital, Red Bank Community Hospital and South Pittsburg Municipal Hospital.

8. At all times relevant herein, HAI was engaged in or affecting commerce within the meaning of Section 1 of the Clayton Act, as amended, 15 U.sC. 12, and was a corporation whose business was in or affecting commerce within the meaning of Section 5 of the Federal 364 FEDERAL TRAme COMMISSION DECISIONS Complaint - 106 'f. Trade Commission Act, as amended, 15 (4) V. c. 45. HAI did business in a number of states and foreign countries. HAI and its hospitals in Hamilton County, the SMSA and the HSA, among other things: (a) purchased substantial amounts of supplies, equipment and medicines in interstate commerce from sources outside of the State Tennessee;

(b) received substantial revenues in interstate commerce from private and governmental insurers located outside of the State of Tennessee; and (c) treated a substantial number of patients who travel from or reside outside the State of Tennessee.

Health Care Corporation 9. On December 11, 1981, HCA purchased Health Care Corporation ("HCC"), for which it paid HCC's owners approximately $30 million including approximately $20 milion in HCA stock, and assumption of approximately $10 milion ofHCC liabilities. HCC became a wholly owned subsidiary of HCA.

10. Prior to its acquisition by HCA, HCC was a corporation organized and existing under the laws of Tennessee, with its executive offces in Chattanooga, Tennessee. HCC owned or managed acute care and psychiatric hospitals in three states. In Hamilton County, the SMSA and the HSA, HCC owned one acute care hospital, Medical Park Hospital, and one psychiatric facility, Valley Psychiatric Hospital. In 1980, HCC's gross revenues from its operations were approximately $9.8 milion.

11. At all times relevant herein, HCC was engaged in or affecting commerce within the meaning of Section 1 of the Clayton Act, as amended, 15 V. C. 12, and was a corporation whose business was in or affecting commerce within the meaning of Section 5 of the Federal Trade Commission Act, as amended, 15 V. C. 45. HCC did business in at least three states. HCC and its hospitals in Hamilton County, the SMSA and the HSA, among other things:

(a) purchased substantial amounts of supplies, equipment and medicines in interstate commerce from sources outside of the State of Tennessee;

(b) received substantial revenues in interstate commerce from private and governmental insurers located outside of the State of Tennessee; and (5) (c) treated a substantial number of patients who travel from or reside outside the State of Tennessee.

361 Complaint IV. COUNT I: ACUTE CARE HOSPITAL SERVICES MARKET Trade and Commerce I2. For purposes of this complaint and this count, the relevant product market is acute care hospital services, or any submarkets thereof, excluding psychiatric services 13. For purposes of this complaint and this count, the relevant geographic market is Hamilton County, the SMSA, or the HSA or any submarkets thereof.

14. Prior to the acquisitions ofHAI and HCC, the acute .care hospital services market in the geographic market was concentrated. Twofirm concentration was approximately 57-59% in Hamilton County, approximately 46-9% in the SMSA, and approximately 40-2% in the HSA. Four-firm concentration was approximately 84-88% in Hamilton County, approximately 75-78% in the SMSA, and approximately 65-8% in the HSA.

15. Barriers to entry are high in the acute care hospital services market in Hamilton County, in the SMSA and in the HSA. These harriers include, among others, substantial capital costs, and the health planning laws, especially given the numher of existing beds in the geographic market.

Effects of the Acquisitions 16. As a result of its acquisition ofHAI, HCA increased its market share of acute care hospital services in Hamilton County from approximately 16-18% to approximately 27-29%, in the SMSA from approximately 13-15% to approximately 28-30%, and in the HSA from approximately 16-17% to approximately 28-30%. Two-firm concentration increased in Hamilton County from approximately 57-59% to approximately 65-6%, in the SMSA from approximately 46-9% to approximately 59-61 %, and in the HSA from approximately 40-2% to approximately 53-54%. Four-firm concentration increased in Hamilton County from approximately 84-88% to approximately 91- 95%, in the SMSA from approximately 75-78% to approximately 85-90%, and in the HSA from approximately 65-8% to approximately 73-78%.

17. As a result of its acquisition ofHCC, HCA increased its market share of acute care hospital services in Hamilton County from approximately 27-29% to approximately 31-32%, in the SMSA from approximately 28-30% to approximately 32-33%, and in the HSAfrom approximately 28-30% to approximately 30-32%. Two-firm (6) concentration increased in Hamilton County from approximately 65- 66% to approximately 69-71%, in the SMSA from approximately 59-61 % to approximately 62-64%, and in the HSA from approxi- + , ,, ,,.. .. .... .... Complaint 106 F.

mately 53-54% to approximately 55-56%. Four-firm concentration increased in Hamilton County from approximately 91-95% to approximately 96-98%, in the SMSA from approximately 85-90% to approximately 88-93%, and in the HSA from approximately 73-78% to approximately 75-80%.

18. Through its acquisitions ofHAI and HCC, HCA acquired direct and actual competitors in the market for acute care hospital services in Hamilton County, in the SMSA and in the HSA. 19. The effects of the HAI and HCC acquisitions by HCA, individually and together, may be substantially to lessen competition or tend to create a monopoly in the relevant product and geographic market in the following ways, among others:

(a) actual and potential competition has been eliminated among some acute care hospitals;

(b) concentration in the market has been substantially increased; (c) patients and physicians may be denied the benefits of free and open competition based on price, quality, and service; (d) competition among hospitals for patients and physicians may be substantially impaired;

(e) competition among some hospitals for p'ltient referrals may be diminished or eliminated;

(f) Medicaid patients may be foreclosed from use of some hospitals now controlled by HCA; and (g) collusion or artificial price increases may be facilitated, and the risk of collusion will be aggravated.

Violation Charged 20. The acquisitions of HAI and HCC, individually and together violate Section 7 of the Clayton Act, as amended, 15 D . C. 18, and constitute unfair methods of competition in violation of Section 5 of the Federal Trade Commission Act, as amended, 15 D. C. 45. (7) v. COUNT II: PSYCHIATRIC SERVICES MARKET Trade and Commerce 21. For purposes of this complaint and this count, the relevant product market is inpatient psychiatric treatment services excluding substance abuse treatment services and long-term treatment of chronic mental ilness, hereinafter "psychiatric services " or submarkets thereof.

22. For purposes of this complaint and this count, the relevant geographic market is the HSA.

23. Prior to the HCC acquisition the market was highly concentrat- ;4-1. .. t-n." .f.... ......n.... .fO')(J ..1....... r..... HOSPITAL CORP. OF AMERICA 367 361 Complaint centration ratio of 100%. There were only two private firms in the market providing psychiatric services.

24. Barriers to entry are high in the market. These barriers include among others, substantial capital costs and the health planning laws, especially given the number of existing beds in the market. Effects of the HCC Acquisition 25. As a result ofHCA' s acquisition of HCC, HCA has increased its share of the market for psychiatric services in the HSA from approximately 7% to approximately 38%. Two-firm concentration is now 100%, and there is now only one private firm in the market providing psychiatric services.

26. Through its acquisition of HCC, HCA has acquired a direct and actual competitor in the market.

27. The effects ofthe acquisition of HCC by HCA may be substantially to lessen competition or tend to create a monopoly in the relevant product and geographic market in the following ways, among others:

(a) actual and potential competition between HCA and HCC in the provision of psychiatric services has been eliminated; (b) concentration in the market has been substantially increased; (c) patients and physicians may be denied the benefits of free and open competition among psychiatric facilities based on price, quality and service; and (8) (d) HCA may have secured the power to raise prices, and its incentives to provide high quality psychiatric treatment services may be reduced.

Violation Charged 28. The acquisition ofHCC constitutes a violation of Section 7 ofthe Clayton Act, as amended, 15 V. C. 18, and an unfair method of competition in violation of Section 5 ofthe Federal Trade Commission Act, as amended, 15 V. C. 45.

368 FEDERAL TRADE CgMMISSION _DECISIONS Initial Decision 106 F. INITIAL DECISION BY LEWIS F. PARKER, ADMINISTRATIVE LAW JUDGE OCTOBER 30, 1984 1. HISTORY OF THE PROCEEDING On July 30 1982, the Federal Trade Commission issued a complaint charging that Hospital Corporation of America ("HCA") had violated Section 7 of the Clayton Act, as amended, 15 U. C. I8, and Section 5 of the Federal Trade Commission Act, as amended, 15 U. C. 45, by purchasing two corporations, Hospital Affliates International, Inc. ("HAI") and Health Care Corporation ("HCC"

According to the complaint, HCA, a Tennessee corporation, is the largest operator and manager of acute care and psychiatric hospitals in the United States. Prior to their acquisitions by HCA, HAI and HCC also owned, operated and managed acute care and psychiatric hospitals in several states of the United States, but the complaint challenges only the acquisitions of hospitals or contracts to manage hospitals within: (1) the "Health Service Area " a 13-county area in southeastern Tennessee and northern Georgia; (2) Hamilton County, the county in which the city 0((2) Chattanooga, Tennessee is located; (3) the Chattanooga Standard Metropolitan Statistical Area ("SMSA"); and (4) any submarkets within these markets. Count I of the complaint defines the relevant product market as acute care hospital services, or any submarkets thereof, excluding psychiatric services. The alleged relevant geographic markets are described above.

Count II of the complaint defines the relevant product market as inpatient psychiatric treatment services excluding substance abuse treatment services and long-term treatment of chronic mental ilness or submarkets thereof. The relevant geographic market in Count II is claimed to be the HSA.

Count I alleges that as a result of the challenged acquisitions, HCA acquired direct and actual competitors in the acute care hospital services market in Hamilton County, the SMSA and the HSA or submarkets thereof and that the effect of these acquisitions in the relevant product and geographic markets may be substantially to lessen competition or tend to create a monopoly in the following ways: (a) actual and potential competition has been eliminated among some acute care hospitals;

HU::PlTAL CUltP. UJ;' ltlCA 361 Initial Decision (c) patients and physicians may be denied the benefits of free and open competition based on price, qualiy, and service; (d) competition among hospitals for patients and physicians may be substantially impaired;

(e) competition among some hospitals for patient referrals may be diminished or eliminated;

(f) Medicaid patients may be foreclosed from use of some hospitals now controlled by HCA; and (g) collusion or artificial price increases may be facilitated, and the risk of collusion wil be aggravated.

Count II alleges that as a result of its acquisition of HCC, HCA acquired a direct and actual competitor in the psychiatric services market and that the effect of the acquisition may be substantially to lessen competition or tend to create a monopoly in the relevant product and geographic market in the following ways: (3) (a) actual and potential competition between HCA and HCC in the provision of psychiatric services has been eliminated; (b) concentration in the market has been substantially increased; (c) patients and physicians may be denied the benefits of free and open competition among psychiatric facilities based on price, quality and service; and (d) HCA may have secured the power to raise prices, and its incentive to provide high quality psychiatric treatment may be reduced. The relief sought in the complaint is an order: (a) Divesting in whole or in part the assets and contracts acquired by HCA from the HAI and HCC acquisitions in the market, so as to form independent viable entities;

(b) Prohibiting HCA in some or all markets from making any future acquisition of any acute care hospital or psychiatric facility, or any contract for the management thereof, in competition with any other hospital it owns or manages, without prior Commission approval, for a period of years;

(c) Requiring HCA to fie compliance reports with the Commission and to give prior notice of any changes in corporate form or organization which would affect compliance obligations under the order entered; and (d) Containing other provisions reasonable and appropriate to correct or remedy the alleged anticompetitive practices engaged in by HCA.

On June 9, 1983, upon motion of complaint counsel who asserted that the possible relieffor the violation alleged in Count II would not justify the expenditure of resources necessary fully to litigate the 370 FEDERAL TRADE COMMISSION: DECISIONS Initial Decision 106 F. issues raised by it, I dismissed that count. Thus, this initial decision deals only with the factual and legal issues raised in Count 1. (4) After extensive discovery, hearings began in Washington, D.C. on November 28, 1983, continued in Chattanooga, Tennessee, and concluded in Washington on May 31, 1984. The record was not closed however, until July I3, 1984 because the parties offered some documents into evidence in subsequent written motions. The parties fied their proposed findings offact, conclusions oflaw and proposed orders on July 23, 1984. Answers were fied on August 6, 1984. At my request, the Commission granted me an extension of time to November , 1984 to fie this initial decision.

This decision is based on the transcript of testimony, the exhibits which I received in evidence, and the proposed findings of fact and answers thereto fied by the parties. I have adopted several of the proposed findings verbatim. Others have been adopted in substance. All other findings are rejected either because they are not supported by the record or because they are irrelevant. II. Findlngs OF FACT A. The Nature of HCA Business 1. HCA is a corporation organized and existing under the laws of the State of Tennessee, with its principal executive offces at One Park Plaza, Nashvile, Tennessee (Cplt. n 2; Ans. n 2).1 HCA is primarily engaged in the operation and management (5) of proprietary acute care and psychiatric hospitals2 in the United States and in several foreign countries, both for its own account and for other owners (Ans. n 3). It is the largest proprietary hospital chain in the United States (Cplt. n 3; Ans. n 3; CX's 427N, 13D).

2. HCA was founded in 1968 by Jack C. Massey, Thomas Frist, M. and Thomas Frist, Jr., M.D. with one health care facility in Nashvile Tennessee (CX 9A). By 1973, HCA owned 51 hospitals (CX 9A), and 1 The following abbreviations are used in this decision" ex - Commission exhibit RX - Respondent's exhibit CPF - Section number and finding in complaint counsel's proposed findings of fact CB - Complaint counsel's brief in support of their findingt offaet CAB - Complaint counsel's answering brief RPF - Respondent's proposed findings of fact RB - Respondent's brief in support of its findings of fact RAB - Respondent' s answering brief F. - Finding number in this decision Cplt. - Complaint Ans. - Answer 2 The complaint defines an acute care hospital as an inpatient facility that furnish B car in connection with services of physicians for conditions for which nursing, medical or surgical services would be appropriate for care, ..i","n . nr j,rp"t.mpn! nnt inf' ..inp'" r""ilit.v "n",,,i,,llv in!..nn..n rnr in tr.."tm""t nf,..."t,,1 mn.. "",,,t-inn,, 361 Initial Decision by 1978, it owned or managed under contract 100 hospitals and _had grown to 28 000 employees with more than 5 000 shareholders (CX 9B). From 1978 through I982, HCA increased the number of hospitals it owned and managed by more than 200%. Approximately 80% of this growth resulted from acquisitions (CX 13I). HCA operates 390 hospitals around the world (Tr. 3238), and it owns or leases approximately 200 hospitals and manages 170 in the United States (Tr. 3737; CX 13I).

3. Prior to HCA's acquisition ofHAI and HCC, as of December 30 1980, it owned 114 hospitals in the United States and provided management services to 56 hospitals in this country. In 1980, HCA had total assets worth $1 610 milion and had earned $81 milion on revenues of $1 429 millon (CX 8B, V).

4. Prior to the challenged acquisitions, HCA owned one hospital in Chattanooga, Parkridge Hospital, and two hospital facilities in outlying southeast Tennessee communities: Cleveland Community Hospital in Cleveland, Tennessee and Athens Community Hospital in Athens, Tennessee (Cplt. TI 5; Ans. TI 5; Stipulation, November 3, 1983). B. The Challenged Acquisitions 1. The HAI Acquisition 5. HAI was a proprietary hospital management company organized and existing under the laws of the State of Delaware and was a wholly-owned subsidiary ofINA Health Care Group, Inc. which was a wholly-owned subsidiary ofINA Corporation, whose primary line of business is insurance (CX 272B). (6) 6. At the time of the acquisition in August 1981, HAI owned or leased 57 hospitals and managed 78 hospitals nationwide (CX's 6E, 8K). In 1980, it had total assets worth $509 millon and had earned $29 milion on revenues of$704 milion (CX 272K, M). In August 1981 HCA acquired HAI for approximately $650 million in a stock transaction (Cplt. TI 6; Ans. TI 6; CX's 8Z-13, 13N, 535, p. 3). 2. HCA's Acquisition of HAl's Chattanooga Area Acute Care Hospitals 7. Prior to its acquisition by HCA, HAI owned or managed five acute care hospitals in the Chattanooga area, and HCA acquired ownership or management of these hospitals when it acquired HAI (Cplt. TI 7; Ans. TI 7). Three of the hospitals, Diagnostic (owned) Downtown (managed) and Red Bank (managed) are located in Chattanooga or its suburbs (CX's 50A, 51A, 56A, 27D). The other two, Sequatchie (owned) and South Pittsburg (managed) are located in counties adjacent to Hamilton County, the county in which Chattanooga is located. ,,_ .. ..__ _ Initial Decision 106 F. 3. The HCC Acquisition 8. At the time of the acquisition, HCC, a psychiatric hospital management company, owned three psychiatric hospitals located in Texas and Tennessee, and a single acute care hospital, Medical Park Hospital in Chattanooga (CX lop, Z-3). In a December 1981 stock transaction, HCA acquired HCC for approximately $30 million (Cplt. n 9; Ans. 4. HCA's Acquisition of HCC's Chattanooga Area Acute Care Hospital 9. Medical Park, which HCC owned, was an 83-bed general acute care hospital located in downtown Chattanooga (CX lop, Z-3). C. Acute Care Hospitals In Chattanooga And The Surrounding Area 1. The Chattanooga Area 10. The city of Chattanooga, Tennessee, situated in Hamilton County in southeast Tennessee on the state boundary with northwest Georgia (RX 1079 (3); CX 15, p. 14), has a population (7) of 170 000 according to the 1980 census (RX 920 (9)). Hamilton County has a population of approximately 288 000 (CX 15, p. 16). Chattanooga is the major city in two federally-designated geographic areas: the Metropolitan Statistical Area ("MSA") and the Georgia-Tennessee Health Service Area ("HSA"

11. An MSA is a large population nucleus, together with adjacent communities which have a high degree of economic and social integration with that nucleus (45 Fed. Reg. 956 (1980)). The Chattanooga MSA is a six-county area consisting of the Tennessee counties of Hamilton, Marion, and Sequatchie and the Georgia Counties of Walk- , Dade, and Catoosa (CX's 484L, 32Z-118). It has a population of approximately 426 540 (CX 15 , p. 16).

12. An HSA is the area designated by the Department of Health and Human Services as a region in which state and local health planners are to assess and identify the health needs of the population in that region (42 V. C. 3001(a) (1982)).

13. The HSA that includes Chattanooga is composed of thirteen counties: ten in southeastern Tennessee (Bledsoe, Bradley, Grundy, Hamilton, Marion, McMinn, Meigs, Polk, Rhea, and Sequatchie counties) and three in northwest Georgia (Catoosa, Dade, and Walker counties) (CX 15, p. 1). In 1980, it had a population of approximately 604,498 (CX 15 , p. 16).

14. Other cities in the area surrounding Chattanooga include Dal- . TTn ' ,p 11 1:__ .J ----- :"'..n. :"1. 361 Initial Decision a 1980 population of approximately 2l,OeO (RX 921 (7); Tr. 3291). Dalton is approximately 40 minutes southeast of downtown Chattanooga on Interstate 75 South (RX's 1030 (3); 1089). 15. Cleveland, situated in Bradley County, Tennessee, approximately 40 minutes northeast of downtown Chattanooga on Interstate 24 East (RX's 1030 (2-3), 1089), is an industrialized community with a 1980 population of approximately 26 000 (RX 920 (4); Tr. 3296). 16. South Pittsburg, in Marion County, Tennessee, is on the Tennessee-Alabama state boundary west of downtown Chattanooga across the Tennessee River (RX 1079 (3)). South Pittsburg is approximately 40 minutes from downtown Chattanooga (RX's 1030 (3), 1089). The 1980 population of South Pittsburg was approximately 3 600 (RX 920 (6)).

17. Dunlap, in Sequatchie County, Tennessee, is approximately 45 minutes from downtown Chattanooga over Signal Mountain and Walden Ridge (RX's 1030 (4), 1089; CX 15, p. 14). The population of Dunlap in 1980 was approximately 3 700 (RX 920 (8)) (8). 18. Whitwell, in Marion County, is approximately 40 minutes west of downtown Chattanooga over Suck Creek Mountain (RX's 1030 (4- 5), 1089, 1079 (3)). The 1980 census for Whitwell showed a population of approximately 1 800 (RX 920 (6)).

19. The largest community in the Tennessee counties of Grundy, Bledsoe, Rhea, Meigs, McMinn and Polk is Athens with a 1980 population of approximately 12 000 (RX 920 (4-8)). Athens is situated just off Interstate 24, north of Cleveland, approximately halfway between Chattanooga and Knoxvile (Tr. 282). Dayton, Tennessee in Rhea County had a 1980 population of approximately 6 000 (RX 920 (8)). 2. Hospitals In The Chattanooga HSA 20. The map on the following page shows the approximate locations of the hospitals in the Chattanooga HSA. (9) ;: :.....;:..: ...,...,,.. ,,:: ,., : . : ,,,, : : ..::: :. ! :...::..;/,. . :,.; ,::::.,,::.../;/\ , . ,( \ :. Initial Decision 106 F. ' I 1r:""i. M L. /I : ': -; G \.I ' f - H ,J 0 - w:., I- A-;nJ:' : I. '7 Far"',. G;';ao, Al ':1: :..7": c s.. Wt e':::. "I .. to 1 c2e":: HOSl':al ec . !lDvwr.. ., .:"."';0= o i"c; ng lio ;:.;:.a.. " ::JW't"... d G.c.gl..

r.r 'f..d.;c"" C.,:.. Jam :.. .!"t". c.."n 'fe.:".- ::r';-'':"wH.if0"''';':'' '1-=.;..0 Ilo"p It_ .,.s :,,=r..:,." ifosF:.o PU' ':ld,,", lO:rplU.": iiods e=l" "" if"s,, r...s;; ;;d... C"=.: Yo"p":: 5""4 P ,::..b=:; 'fu."' C";M': . o.': ugIo,, C.r.:.:- iio..pJ cow:=. 'f.d c...: C"' oe"- C."".:"" O:fpJ:. s.,..o=. :;Ct'""-."; :;o" Q.d .r.J c:;=.": =. lio-,p.;:41 C:Jppe., c .'fd'.:c", r:: P.xx iio"p.::.l .d-oe cou.:: .'O:1F":"" Jle-;opol;':UJllo:O;:J:..": Crucd", C" "''''i ;."..... 3o" :L. .c..,.C UJr.. .rl.Lii""'pl:.l w('..:-......: ilo"".:o.": C.".";oic C.:"""':., q""Pl:.L ':I!""od s.. , pp 361 Initial Decision (10)3- Th MSAa. Hamilton County, Tennessee 21. There are nine general acute care hospitals in Hamilton County, and HCA either owned, or, as a result of its acquisitions became the owner or manager of, five hospitals in this county. 22. Erlanger Medical Center located in downtown Chattanooga on East Third Street near the Tennessee River (RX 1079 (3); CX 66A), is a non-profit teaching hospital owned and operated by the Chattanooga Hamilton County Hospital Authority, a public agency created by state statute in 1976. It is controlled by a politically appointed board of trustees (RX 761 (3); Tr. 90, 109, 3289). Erlanger is a major medical complex with several different hospital buildings, including a general acute care hospital, Baroness Erlanger Hospital, and several other facilities (Tr. 96-97, 107, 489). It provides a wide range of health care services to the community, and it is the only tertiary care referral center in the HSA (Tr. 92, 94-95; CX's 15, p. 142; 18Z-4). While it has a certificate of need (F. 252) for 780 beds, only 754 are licensed. Ofthese 754 (11) licensed beds, only 714 are in actual use (Tr. 130, 132). The full number of certificate-of-need beds is expected to be put into use sometime in 1985 (Tr. 133).

23. Erlanger is required by law to accept all Hamilton County residents needing hospital care, regardless oftheir abilty to pay, and has the image ofa "public" hospital (Tr. 110-11; CX 408Z-17). It treats the vast majority of indigent patients in Hamilton County, and at least 100 beds are used for the treatment ofthese patients' at all times (Tr. 115, 134, 883, 138; CX's 18Z-4, 32Z-8, 38Z-17). Erlanger has several specialized intensive care units including a trauma unit, a neurosurgical intensive are unit, and a burn unit, which are unique and not available anywhere else in the area (Tr. 1374-75, 1408-9; CX , pp. 93-94).

24. The Erlanger complex includes a pediatric hospital, T. Thompson Children s Hospital (Tr. 96, 107). Children s Hospital is a 1 Hospitals are classified from most basic to most advanced as primary, secondary, or tertiary facilities. A primary hospital generally provides basic acute care services, such as obstetrics (unless such services are organized regionally), surgical services, x-ray, clinical laboratory, and blood services, a minimal level emergency room pharmacy and anesthesia services, and minimal intensive care capabilities. A secondary level facility generally has the primary services listed above and, in addition, more specialized capabilties such as EEG equipment diagnostic and therapeutic equipment for cancer patients and 24-hour physician coverage. A tertiary level hospital generally hils the same facilities 8S the other two levels, and is available as a primary level hospital for the majority of it.'! patients, but also has specialized services as are needed in the community, such as open heart - surgery capabilities, cardiovascular diagnostic lab, cr scanner, burn-care unit, and oncology (cancer) aervices (CX 15 141-42). A referral hospital is one that because ofthe level of sophistication of its services is able to attract patients from smaller facilities having more limited services and capahiHties (Tr. 123 442 615 747, 1507--8, 1967-68, 3277 1388) . Indigent patient.'! are economically-disadvantaged individuals who have no form of health insurance coverage and who do not qualify for Medicaid (Tr. 110). Initial Decision 106 F. 114-bed facility that is used exclusively for pediatric patients (Tr. 100-1 , 130).

25. Wilie D. Miler Eye Center, a 30-bed specialty eye center, is also a part ofErlanger. It is located adjacent to Baroness Erlanger Hospital and is used exclusively for the treatment of eye diseases (Tr. 107-D8, 131).

26. Erlanger also operates a 13-bed emergency psychiatric "holding unit" that is used exclusively to treat patients needing psychiatric care. It is the only psychiatric emergency facility in the area and many of the patients treated there are indigent. It is located in a maximum security wing of the hospital that is not easily accessible. The psychiatric unit is operated by a nursing staff specially trained in the care of psychiatric patients (Tr. 118-20). 27. Memorial Hospital is located on Citico Avenue in the downtown area of Chattnooga. It is a non-profit institution which is owned and operated by the Catholic Church (Tr. 1503; CX 53A; RX 1079 (3)). In 1981, Memorial Hospital was authorized by the Tennessee Health Facilities Commission to operate 349 general acute care hospital beds (RX 1092). In 1982, its authorized bed total was increased to 365 (RX 872).

28. Memorial Hospital has secondary capabilities but offers some tertiary services. It offers such sophisticated specialties as cardiovascular catheterization, open heart surgery, nuclear medicine, radiation therapy and neurological procedures. The (12) hospital's emergency room is staffed by physicians on a 24-hour basis (Tr. 1505 1507, 3291, 136; CX's 15, p. 142, 29Z-5).

29. Parkridge Hospital owned by HCA before the acquisitions, is located on McCallie Avenue in downtown Chattanooga (CX 36A; RX 1079 (3)). It is a medical/surgical hospital which offers diagnostic and therapeutic services usually found in major urban hospitals (Tr. 3276).

30. Parkridge Hospital was in 1981, and is today, authorized by the Tennessee Health Facilities Commission to operate 296 general acute care hospital beds (Tr. 3276; RX 1092). Parkridge is basically a secondary level hospital with some tertiary capabilities (Tr. 127 276 686; CX , p. 142).

31. East Ridge Community Hospital is located in the East Ridge community approximately six miles east of downtown Chattanooga adjacent to the intersection ofInterstate 75 and Interstate 24 (Tr. 678 750; RX 1079 (3)). It is a general acute care hospital authorized by the Tennessee Health Facilities Commission to operate 128 beds (Tr. 680; RX 1092). The hospital offers most ofthe medical and surgical services commonly found in suburban hospitals of its size, and has an active tptrl(, nr!'f't.;f'p ln urhi..h -it nrn" pc; c;p('rm rv lpvp rp. (1'1' 361 Initial Decision 680-81, 146, 3291) East Ridge -is owned and operated by Humana, Inc., a large, for-profit hospital chain (Tr. 682; CX 83A, F). 32. Diagnostic Center Hospital an 80-bed facility on McCallie Avenue in downtown Chattanooga, which HAI owned before its acquisition by HCA, specializes in the diagnosis and treatment of cardiopulmonary disease (Tr. 3592). It was established in the late 1950' s by the physicians who were the founding members of the Diagnostic Center Medical Group. All of the inpatient admissions to Diagnostic Center Hospital are made by the nine physicians in the Diagnostic Center Medical Group (Tr. 3592-93). 33. Diagnostic Center Hospital has no operating room or other surgical facilities, and the hospital' s services are strictly non-invasive (Tr. 3593).

34. Red Bank Community Hospital is a 57-bed general acute care hospital located in the Red Bank community north of the Tennessee river, and was managed by HAI prior to the acquisitions (Cplt. n 7; Ans. n 7; Tr. 1171; RX 1079 (3); CX's 38-2-42, 27D, 56A-F). Red Bank is a not-for-profit hospital owned by the Health and Educational Facilities Board ofthe City of Red Bank and is leased to the Red Bank Hospital Association (Tr. 1171; CX 628).

35. Medical Park Hospital (North Park) was in 1981 an 83-bed facilty on McCallie Avenue in downtown Chattanooga and was owned by HCC before its acquisition by HCA (Tr. 138; CX 75A; RX' 1092, 1079 (3)). (13) 36. Medical Park was granted a certificate of need by the Tennessee Health Facilities Commission in October 1980 authorizing its relocation to the suburban north Chattanooga community of Hixson (CX 19V; RX 858). North Park Hospital, owned by HCA (CX 895G) and located in the north Chattanooga suburbs, is a new facilty opened on November 15, 1982, which replaces Medical Park Hospital (Tr. 3291; CX 504E).

37. North Park does not provide neurological surgery, open heart surgery, major trauma treatment or obstetrics. The hospital administration expects to add obstetric services within the next several years if there is demand for the services and physician support. The new facility has the latest technology available for the medical and surgical specialties and services generally offered in suburban hospitals. It offers primary and limited secondary level services. Because of its suburban setting in the outlying community of Hixson, the hospital emphasizes outpatient procedures (Tr. 137, (Tr. 137, 545-46 585 3398; CX' s 15, p. 142, 20Z-52).

38. Downtown General is a 65-bed urban general acute care hospital which is located in central downtown Chattanooga and was established as a not-for-profit hospital in 1975 (Tr. 1417-23; CX 51A, F, 15 . __ Initial Decision 106 F. p. 179). HAI managed Downtown General before its acquisition by HCA (Cplt. n 7; Ans. n 7).

39. Metropolitan Hospital formerly Tepper Hospital, is a 64-bed proprietary hospital owned and operated by American Healthcorp, Inc., which purchased the facility in 1982 (Tr. 1045, 1089-90; CX 58A F). Tepper was a pediatric hospital formerly owned by a pediatrician whose four-person physician group practices at the hospital (Tr. 1089- 90). Metropolitan now provides some diabetes services to adults but stil focuses primarily on pediatrics (Tr. 454, 1090-91). Approximately 70% of its patients are pediatric (Tr. 1091), but since 1981 , its gynecology, orthopedics, family practice, internal medicine and general surgery services have been expanded (RX 731 (20)). b. Other Hospitals In The MSA 40. There are five hospitals outside of Hamilton County but in the MSA. Two of them are located near Chattanooga, just over the Georgia state line.

41. John L. Hutcheson Memorial Tri- County Hospital (Tri-County), is located in the Ft. Oglethorpe community in the Georgia suburbs just across the state line and about ten miles from Chattanooga, Tennessee (RX 1979 (3)). Tri-County is owned by the Hospital Authority of Walker, Dade and Catoosa Counties (RX 646 (10)), a body "corporate and politic" created pursuant to Georgia statute in 1947 (RX 646 (4-5)). The Authority is (14) controlled by a nine-member Board of Trustees nominated by the County Commissioners of the three Georgia counties (RX 679 (6)).

42. Tri-County is a medical! surgical hospital which provides primary and secondary services comparable in range and sophistication to those of Parkridge Hospital (Tr. 117, 3289). In addition to the services which are also available at Parkridge, Tri-County has a hospital unit dedicated exclusively to pediatric care including pediatric intensive care, and an obstetrics service (CX 106C). Tri-County was authorized by the State of Georgia on September 8, 1981, to operate 237 general acute care hospital beds (RX 911). It provides care for indigents and Medicaid recipients in northwest Georgia (Tr. 116 685; RX 361 (60)).

43. Wildwood Sanitarium Hospital is a 39-bed facility located in Dade County, Georgia, approximately 10 minutes from Chattanooga (Tr. 3290; CX's 92A- , 15, p. 142). It is owned by the Seventh Day Adventist Church (Tr. 3290, 692, 150). It provides non-traditional services such as alcohol, diet, and non-smoking programs (Tr. 1511). While its medical services are limited in scope (Tr. 692, 1511), and it . is not JCAH accredited,5 it does provide general medical services and L"TT_ '"'''.n''' ' .L_ 361 Initial Decision obstetrics, surgery, respiratory, therapy, physical therapy,audiology, diagnostic radiology, electrocardiography, cardiac rehabilitation emergency treatment, and psychiatric care (Tr. 3290; CX 92B, D-E). Wildwood is not considered a competitor by HCA facilities in the Chattanooga area, according to their planning documents (CX's 17Z- , 18Z-3, 20Z-11, 27I, 28Z-2, 32Z-5, 34E- , 82Z-11 , 235Z-23; see also Tr. 1512).

44. South Pitts burg Municipal Hospital a city-owned facility in South Pittsburg, Tennessee, is managed by HCA pursuant to a longterm contract (CX 21). The hospital was approved for 107 beqs in 1981 (RX 1093). It is comparable to the smaller Chattanooga hospitals in terms of quality and range of services (Tr. 690). 45. A drive from downtown Chattanooga to South Pittsburg Municipal Hospital on January 18, 1984 required 41 minutes and 17 (15) seconds by the most direct route under good-to-excellent, non-rush hour driving conditions (RX's 1089, 1030 (3)). 46. Sequatchie General Hospital is a small primary care facility in the rural community of Dunlap, Tennessee, situated in Sequatchie County northwest of Chattanooga over Signal Mountain and Walden Ridge (CX 15, p. 14; RX 1079 (3)).

47. Sequatchie General was acquired by HCA as part of the acquisition of HAI in 1981 (Tr. 3253-56; CX 874, pp. 18-19). In December 1982, HCA sold Sequatchie along with 18 other hospitals to Republic Health Corporation in exchange for cash, notes, stock and assumption of debt. As a result of this transaction, HCA obtained ownership of approximately one-third ofthe common stock of Republic (Tr. 544-45; CX 874, pp. 3, 18-19). A "voting agreement" between HCA and Republic gives HCA the right to elect a majority of Republic s board of directors if Republic fails to meet certain specified conditions (Tr. 545; CX 874, p. 16). HCA currently owns about 20% of Republic s stock (Tr. 3265).

48. Whitwell Community Hospital is a 25-bed facility located in Marion County, approximately 18 miles from Chattanooga (CX's 44A , 15, p. 142). It is a very old facility with a two-person medical staff (Tr. 617; CX 15, p. 138), and does not meet state codes and standards (Tr. 691) One witness testified that he didn t "think you would call it a general acute care hospital" (Tr. 691), whereas another witness stated that it is more like a "clinic" and is "really not a hospital at all" (Tr. 1511).

49. Whitwell was purchased by Rural Hospital Associates Inc. in for hospitals in order to improve and to maintain the basic quality of care of their medical staff and facilities (Th. 741). JCAH accreditation means that a hospital meets certain minimum standards of quality (Tr. 256, 1103-4 2284; ex' s 175, 250C). It is the "Good Housekeeping Seal of Approval" for hospitals. Tennessee accepts JCAH approval as suffcient to receive state licensing approval and hospitals with JCAH approval are deemed qualified to participate in the Medicare program (Tr. 257) Initial Decision 106 F. 1982 (CX 404B, S; RX 1093). Rural Hospital Associates is now building a new $5.5 milion facility to replace the existing Whitwell facility (CX 404Z-14, Z-20). The certificate-of:need for the 40-bed replacement facility was approved November 1, 1982 (RX 876). 50. A drive from Whitwell Community Hospital to downtown Chattanooga on January 19, 1984, required 39 minutes and 47 seconds by the most direct route under good-to-excellent, non-rush hour driving conditions (RX 1030 (4-5)).

4. Hospitals In The HSA Outside Of The MSA 51. Outside of the MSA, but in the HSA, are eight hospitals, two of which, Cleveland Community Hospital and Athens Community Hospital, are owned by HCA.

52. Cleveland Community Hospital formerly named Cherokee Park Hospital, is a 100-bed full-service community hospital located in Cleveland, Tennessee (CX's 71A, 15, p. 179 148m, a city which has experienced 50% growth since 1970. It is approximately 30 miles from Chattanooga (CX's 148Z-16, 235Z-(16J14). Cleveland Community Hospital has a 20-bed psychiatric unit located in a separate wing from the rest of the hospital (Tr. 3489-90; CX 15, p. 268; HCA's Response to Interrogatory 18, fied Apr. 20, 1983). This unit is used exclusively for psychiatric patients (Tr. 3490). HCA obtained Cleveland Community Hospital through its 1980 acquisition of General Care Corporation (CX's 546A, 566A, 487 A, llZ-17). The hospital is approximately 41 minutes from downtown Chattanooga by auto (RX's 1098 1030 (2-3)).

53. As an example ofthe isolation of the hospital from Chattanooga Jim Whitlock, administrator of Cleveland Community Hospital, testified that he drives to Chattanooga infrequently and then only to THA district meetings, has no business or personal interests in Chattanooga, is not familar with hospital facilities and locations in Chattanooga, and does not subscribe to either of the Chattanooga newspapers (Tr. 3476-77).

54. Athens Community Hospital is a 118-bed full-service hospital located in McMinn County (CX's 15, p. I79, 17Z-6, 54A, G). It is some 40 miles from Chattanooga in Athens, Tennessee, a small town with a population of approximately 12 000 (CX's 54A, 17Z-28). According to the former administrator of Erlanger, while Athens and other outlying hospitals compete with Erlanger "to a degree " patients who need general medical-surgical care are treated in their own communities (Tr. 148) and patients from the Athens Hospital needing special care would probably be referred to Knoxville, Tennessee (Tr. 282). This is also true of Woods Memorial Hospital (Tr. 283). Athens Hospil -i not. mpmhpr nft,hp thmnfll:T ATP:: nit, nidrlrt rtr ( , p.

----- ----- -----. -- _u,--- 361 Initial Decision 283). Two administrators from -urban -Chattanooga hospitals who testified in this case were not familiar with the services offered by Athens (Tr. 279, 1603).

55. Bradley County Memorial Hospital is in Bradley County. Bradley Memorial is a 251-bed city-owned hospital located in Cleveland Tennessee, approximately 30 miles north of Chattanooga and three miles from Cleveland Community Hospital (Tr. 136, 685, 3465; CX' , pp. 142, 179, 49A, F, 148Z-16). It is a full-service community hospital that provides primary and secondary care and some intensive care services (Tr. 136; CX 15, p. 142). It offers a range and quality of services comparable to many of the hospitals in Chattanooga (Tr. 685; CX 148Z-25). It also operates an 18-bed substance abuse treatment unit (CX 65F; HCA's Response to Interrogatory 18, fied Apr. 20 1983).

56. The other five hospitals in the HSA are relatively small facilities located in rural areas (see Tr. 3296). 57. Woods Memorial Hospital is a 72-bed county-owned community hospital located in Etowah, Tennessee, approximately 38 miles from Chattanooga (CX's 61A, 17Z-35). Any referrals from Woods Memorial in Etowah would probably go to Knoxvile, Tennessee hospitals (Tr. 283). Woods Memorial Hospital is not a (17) member of the Chattanooga area hospital council (Tr. 283), and hospital administrators in the Chattanooga area are not generally familiar with the Woods Memorial facility and services (e. Tr. 279). )' (RX 1081 (150). 58. Copper Basin is a 44-bed county-owned community hospital located east of Chattanooga near the North Carolina border in Polk County (CX 59A, F, 15, pp. 14 , 179; see Tr. 279). It is located in Cooperville, Tennessee approximately 65 miles from Chattanooga (CX 59A). Hospital administrators in the Chattanooga urban area are not familyiar with the facilities, location or services of Copper Basin Medical Center (e. Tr. 280).

59. The three remaining hospitals, aside from offering only the most basic treatment, are not JCAH-accredited. They are Bledsoe County Hospital, Grundy County Hospital, and Rhea County Medical Center. 60. Rhea is a 57-bed primary care hospital owned and operated by the county (CX's 45A, F, 15, p. 142). It is located in Rhea County approximately 25 miles from Chattanooga. (" ' ) (RX 1081). Hospital administrators in the Chattanooga urban area are generally unfamiliar with Rhea County Medical Center (e. Tr. 685). 61. Bledsoe is a 32-bed general acute care hospital located in Bledsoe County. It provides primary care services (Tr. 150; CX's 64A, F, I5 179), and is a county-owned facility approximately 35 miles from Chattanooga (CX 48A, F). ("' J (RX 1081 (160)). . Throughout this document, ('U ) refers tain camera maleria! that has been excised. ,. g., . , , pp. . Initial Decision 106 F. 62. Grundy County General Hospital is a 27-bed proprietary hospital located in Grundy County (CX's 651A, 15, p. 179). It is owned by Cumberland Heights Hospital, Inc. (CX 99A). Grundy is a small, primary care facility located approximately 36 miles northwest of Chattanooga (CX 15, p. 142). (*"J (RX 1081 (165)). Hospital administrators in the Chattanooga urban area are not familiar with the facilities or services of Grundy County General Hospital (e. Tr. 281, l603). (18J D. Other Health Care Facilities In The Chattanooga Area 63. There are 27 nursing homes in the HSA (CX 15, p. 208), nine of which are located in Hamilton County (CX 15, p. 225). 64. There are two psychiatric hospitals in Chattanooga (CX's 15 232- , 131DJ.

65. At the time of the acquisitions in 1981, there were no emergicenters or independent ambulatory surgicenters in Chattanooga (Tr. 1765-66, 162-64, 167-68, 554- , 3404, 3340, 468-69). 66. There are currently three free-standing emergicenters in Chattanooga not associated with a hospital (Tr. 469, 554, 3304, 3290, 168). E. Interstate Commerce 67. Prior to the acquisitions of HAI and HCC, HCA owned or managed over 170 acute care and psychiatric hospitals in more than 25 states (CX 535, pp. 18, 20) and in 1980, it had gross revenues of$1.4 billion from its interstate operations (CX's 8V, 13Y). 68. HCA's Parkridge Hospital in Chattanooga, Tennessee treated approximately 2 503 patients from outside of Tennessee, and this activity produced a gross revenue of around $5 230 733 (CX's 36E- 497E-G).

69. In 1981, Parkridge received federal Medicare funds of approximately $8.3 million from the United States Treasury in Washington C. (HCA' s Response to Interrogatory 28(q), fied Apr. 20, 1983), and in the same year, it purchased approximately $883 000 worth of drugs and supplies from out-of:state (HCA's Response to Interrogatory 28(0), fied Apr. 20, 1983).

70. Prior to being acquired by HCA, HAI owned or managed a total of 155 hospitals in 33 states (CX's 535 , p. 3, 272B; Ans. n 7). The headquarters for these interstate operations was Nashvile, Tennessee (CX 272A). In 1980, HAI had gross revenues of approximately $573 million from its interstate operations (CX llZ-17; Ans. n 7). 71. The hospitals owned and managed by HAI in the Chattanooga area, Diagnostic Center Hospital, Downtown General Hospital, Red Bank Community Hospital, Sequatchie General Hospital, and South Pittsburg Municipal Hospital treated approximately 2 304 out-of- , , 1 361 Initial Decision number of patients, produced gms& revenues of approximately 878 962 (CX's 16E, G, 23E, G, (19) 30E, G, 37I, M, 67E, G; HCA' Response to Interrogatory 28(m), fied Mar. 10, 1983 and Apr. 20 1983). In 198I, these hospitals received federal Medicare funds of well over $6 milion from the United States Treasury in Washington, D. (HCA' s Responses to Interrogatory 28(q), fied Mar. 10, 1983 and April , 1983). In 1981, Diagnostic, South Pittsburg, and Sequatchie purchased approximately $838 251 worth of drugs and supplies from outside Tennessee (HCA's Responses to Interrogatory 28(0), fied Mar. , 1983 and Apr. 20, 1983).

72. Prior to being acquired by HCA, HCC owned or managed two general acute care hospitals, three psychiatric hospitals, an outpatient psychiatric clinic, and psychiatric treatment units at two hospitals owned by others, in four states. These interstate operations were headquartered in Chattanooga, Tennessee (CX's 136D, M, lod, 136D). In 1980, HCC had revenues of approximately $12 million from these operations (CX 10M; Ans. n 10).

73. In 1981, Medical Park Hospital, HCC' s only acute care hospital in the Chattanooga area, treated approximately 409 patients from states other than Tennessee, which, based on the ratio of these patients to its total number of patients, produced gross revenues of approximately $1 027 209 (CX 75E, G), and it received federal Medicare funds of approximately $2.6 milion from the United States Treasury in Washington, D.C. (HCA's Response to Interrogatory 28(q), fied Apr. 20, 1983). In 1981, Medical Park purchased approximately $96 804 worth of drugs and supplies from out-of-state (HCA' Response to Interrogatory 28(0), fied Apr. 20, 1983). 74. In some cases, private insurers who provide health insurance for Chattanooga businesses are located in states other than Tennessee (RX' s 1001 (2), 1011 (1)), and funds from these insurers that reimburse hospitals in the Chattanooga area for patient care cross state lines. 75. HCA' , HAl's and HCC's total operations at the time of the acquisition were in interstate commerce as were their operations in the Chattanooga area and the acquisitions therefore occurred in, and affected, interstate commerce.

F. The Relevant Product Market 1. Acute Care Hospital Services a. Expert Opinion 76. Dr. David S. Salkever, complaint counsel' s expert witness, is a professor of health economics at John Hopkins (20) University (Tr. 2264). He testified that the relevant product market in this case should be defined as:

.... ....,. .. ,.,,.. ,. Initial Decision 106 F. (TJhe provision of acute inpatient hospital services and the provision of hospital services to critically ill emergency patients, typically who I suspect would be admitted a''I inpatients if indeed they survived at the critical episode (Tr. 2280). 77. Dr. Salkever s definition excludes from an acute care hospital' business its outpatient business, except for outpatients who are subsequently admitted (Tr. 2281). The reason for limiting the market to inpatient services provided by acute care hospitals is that these services are typically needed by and consumed by patients in combination (Tr. 2283) and can therefore be offered only by acute care hospitals (Tr. 2284).

78. Dr. Salkever s definition also excludes providers of inpatient services other than acute care hospitals, such as psychiatric hospitals and nursing homes because they provide a difierent type of service , longer-term care, and are not equipped to treat patients who need acute care (Tr. 2285). His definition also excludes free standing surgery facilities, emergency centers and doctor s offces, even though they offer some of the services provided by a hospital's outpatient clinics because, if they are included, one would be given "a very misleading picture of the structure of the market for the hospital's principal line of business, namely, inpatient services and care of critically il emergency patients" (Tr. 2287). Finally, he would exclude from the market the pediatric facilities at T.C. Thompson Children Hospital primarily because the beds in that facility could not easily be converted to use by adult patients (Tr. 2288). 79. Dr. Jeffrey E. Harris, HCA' s expert, is a physician on the staff of Massachusetts General Hospital and is an associate professor of economics at MIT (Tr. 3806). He disagreed with Dr. Salkever proposed product market because, in his view, it is too narrow (Tr. 3942--3). In his opinion, the product market should include outpatient care (Tr. 3943) because the portion of hospital business devoted to such care has increased over time as a result of advances in medical technology which permit outpatient treatment for conditions which were formerly treated on an inpatient basis (Tr. 3943--4). Another reason for inclusion of outpatient net revenues (his preferred unit of measurement) in the acute care hospital market is that outpatient facilities are frequently a feeder for inpatient facilities (Tr. 3945 3950-51). Dr. Harris also believes that outpatient providers such as emergicenters and multiple diagnostic services (21) and hospices are competing more and more with acute care hospitals (Tr. 3952-53). b. Inpatient Services Offered By Acute Care Hospitals 80. Representatives of health care providers in the Chattanooga "..no ............ 11u ......n"' ..1-.. 4- t-h",.... C' l ..".." r...._ , p.

.l.lVL..l.l.ln..u vV.l .l. VI l"UU.L.LUVn. 361 Initial Decision vided by acute care hospitals. These include medical and surgical beds, 24-hour observation, nursing services, laboratory and x-ray, intensive and coronary care, and ancilary support services (Tr. 93 463- , 552, 1396, 1401, 1514, 1762; CX 895E). 81. This range of services must be provided by all acute care hospitals because patient treatment often requires consumption of these services in combination (Tr. 2283), and because hospitalized patients are often unable to travel to another facility for treatment (Tr. 1763). Adding to the need for acute care hospitals to offer a range of services is the variability and uncertainty of ilness (Tr. 3856). c. State Requirements For Acute Care Hospitals 82. Under Tennessee law, a general hospital must be able to provide to its patients diagnosis, treatment and care of acute illness, injury, or infirmity for a period exceeding 24 hours, and it must provide an organized staff, a laboratory, x-ray facilities, surgery, obstetrics, an isolation unit, a kitchen, and an emergency department (Tenn. Admin. Compo ch. 1200-8-1- 02(3); see also Tenn. Code Ann. 68-11- 201(j) (1983). The state requires acute care hospitals to offer all the services enumerated in the regulations (Tenn. Admin. Compo ch. 1200 8-1- 02(3)); if any of these services are not provided, a waiver of the regulations must be obtained (see Tenn. Code Ann. 68-11-209 (1983); Tr. 3616). The 1981-1986 Health Systems Plan which describes the health care system in the Chattanooga area (see CX 15 , p. 1), also identifies basic acute care hospital services, including, among other things, obstetrics (unless offered elsewhere in the region), surgery, x-ray and laboratory services, blood services, pharmacy, pathology laboratory, respiratory therapy, and intensive care capabilities (CX' , p. 141, 169, pp. 181-82; see also 1983-1986 Tennessee State Health Plan ). To be eligible for accreditation as an acute care hospital by the Joint Commission on Accreditation of Hospitals, an establishment must have facilities, beds, and services available over a continuous period, 24 hours a day, seven days a week; it must have an organized medical staff and nursing service; its primary function must be the diagnosis, treatment, and/or rehabilitation (22) ofthe acutely il: and it must provide dietetic, emergency, nuclear medicine, pathology, laboratory, pharmacy, radiology, and respiratory services (CX 174 xviii).

83. Chattanooga area acute care hospitals provide all or almost all of the services identified above, as indicated in reports fied with the states of Tennessee and Georgia and the American Hospital Association (CX's 62B- , 63E, G, I, L (Memorial), 63B- , 654C.E (East Ridge), 65B- , 633C-F (Bradley), 66B- , 638C-F (Erlanger), 67B- , 24E, G , I (Red Bank), 68B- , 637E, G, I, K (Metropolitan), 69B-C, 634E, G, I, K , pp.

Initial Decision 106 F.'l. (Woods), 70B- , 632C-F (Copper Basin), 72B- , 635E, G , I, K (Rhea), 99B-C (Grundy), 496B- , 643C-F (Diagnostic),6 497B- , 644C-F (Parkridge), 498B- , 641E, G, I, K (Downtown), 499B-C (North Park), 500B- , 640C-F (Sequatchie), sole- , 642E, G, I, K (South Pittsburg), 502B- , 645B-D (Athens), 503B-C, 639E, G, I, K (Cleveland Community), 657F, H, 106C-F (Tri-County)).

d. Industry Recognition Of Acute Care Hospitals As Distinct From Other Providers 84. The state of Tennessee licenses "general hospitals" separately from specialty health care facilities such as psychiatric hospitals (Tenn. Admin. Compo ch. I200-8-1- 02(3)), as does the state of Georgia (Ga. Admin. Compo ch. 290-5-6-.03(3)). Acute care hospitals are treated separately from other health care facilities in the Health Systems Plan (CX 15, pp. 137-90), and the Tennessee and Georgia state health plans devote separate chapters to acute care hospitals (CX's 169 179-277 288 , pp. 414-47). General medical and surgical (i. acute care) hospitals have a separate Standard Industrial Classification SIC") code (8062) from other health care facilities (OMB Standard Industrial Classification Manual 1972 SIC Manual") (1972). 85. The Joint Commission on Accreditation of Hospitals has a separate set of eligibility criteria for acute care hospitals as distinguished from specialty or long-term facilities (Tr. 2284; CX 174, p. xviii) and the American Hospital Association has separate classifications for acute care psychiatric hospitals and nursing homes (Tr. 2285). 86. Hospital administrators also regard the services provided by acute care hospitals as distinct from those of other facilities (Tr. 93 1514, 3339). Health economists, as reflected in studies and other health economics literature, perceive acute (23) care hospitals as different from other health care providers such as psychiatric hospitals and nursing homes (Tr. 2285-86).

87. When they discuss competition in their planning documents Chattanooga area acute care hospitals list only other acute care hospitals (CX' s 17Z-35 through Z-37 (Athens), 18Z-3 through Z-4 38Z- 39 through Z-3 (Parkridge), 20Z-26 through Z-27 (North Park), 28Z- 15 (Downtown), 32Z-40 through Z-45, 29Z-5 through Z-7 (Diagnostic), 34E-F, 157Z-4 through Z-5 (Sequatchie) 82Z-19 (South Pittsburg), 235Z-23, 148Z-25 through Z-26 (Cleveland Community), 6HZ-55 (East Ridge).

88. Although some Chattanooga area hospital administrators perceive competition from other health care providers, it is limited (Tr. 168 769 1517 3287 3340 3402-03 3475 3492-94). Their only signifi- HOSPITAL CORP. OF AMERICA 387 361 Initial Decision cant competition comes from other acute care hospitals (Tr. 135 553 683- , 687, 1100, 1505).

89. Chattanooga area acute care hospitals, when they compare prices, do so only with those of other acute care hospitals (CX's 161A 162A (South Pittsburg), 179A, 180A, 181A, 182A, 183A, 283A, 284B (Diagnostic), 184A, 271A, 273A, 274A, 275B (Downtown), 239A- (HAl), 276A- , 277B-D (Parkridge), 279A, 280A, 281A, 282A (Sequatchie), 286A, 287 A (Red Bank), 315N, 316I, 317E (Erlanger)). The management contracts for Downtown and Red Bank require the rates at comparable hospitals nearby" to be considered when rates are set (CX' s 185E, 189D).

2. Other Health Care Providers And Services a. Hospital Outpatient Care 90. Nationally, hospital outpatient care represents 12% of their gross revenues (Tr. 2286) and outpatient care in Chattanooga area hospitals is being expanded by area administrators (Tr. 766). 91. Outpatient care such as day surgery, emergency care and diagnosis are Parkridge Hospital's most rapidly growing segment of services and it plans to increase its day surgery facilities (Tr. 3278); Tri-County has expanded its emergency care facility in Lafayette Georgia (RX 677 (1)) and opened a new day surgery unit in February 1983 (RX 682 (4)).

92. The installation of new CT scanning equipment at Tri-County in April 1982 and at Parkridge Hospital in 1981, substantially expanded the outpatient diagnostic capabilities of those two hospitals (Tr. 3276; RX's 683 (4), 682 (4), 1120). The outpatient capabilities at Parkridge Hospital have also been increased by the significant expansion of nuclear medicine and cardiology services since 1981 (Tr. 3276). (24J 93. The revenue contribution to East Ridge Community Hospital from outpatient services is significant and growing (Tr. 767). A renovation program in progress at East Ridge in 1983 involved the expansion and renovation of the emergency treatment area and all the ancilary departments of the hospital, and was prompted in part by the increasing volume of outpatient care (Tr. 768). 94. Outpatient care, a rapidly growing area for the new hospital accounts for approximately 12% of hospital revenues at North Park (Tr. 3400).

95. There are several reasons for increased outpatient treatment: Unlike the past, third-party payers in the Chattanooga area now provide incentives, such as lower or no deductible or co-payments (F. 243; Tr. 981- , 1509, 1809, 3949) which encourage providers to in- Initial Decision 106 F. crease ambulatory surgery and other outpatient care (Tr. 465). In addition, advances in medical technology have diminished the distinction between inpatient and outpatient care (Tr. 465, 501-03, 3943-44), and diagnostic testing and surgical work-ups formerly done after hospital admission are now being done on a pre-admission outpatient basis to shorten hospital stays (Tr. 504, 1571, 3948; CX 15, pp. 145-46). 96. Estimates in the professional literature suggest that 25% or more of hospital surgical procedures can be performed on an outpatient basis (Tr. 504), and a study commissioned by Erlanger Medical Center, where 12 to 15% of surgery is outpatient, claims that this figure could be increased to 25 to 30% (Tr. 166). Some area hospitals have experienced an increase in the percent of outpatient surgery which they perform (Tr. 1522, 1569, 3279, 3398, 3466-7, 3504). b. Other Providers (1) Inpatient Care 97. In Chattanooga, the only inpatient health care facilities other than acute care hospitals are nursing homes and psychiatric hospitals.

98. Nursing homes provide long-term services to "chronically il or seriously disabled persons over an extended period of time" (CX 169 p. 278; Tr. 3337). The average length of stay at nursing homes in the Chattanooga area is 374.8 days (CX 15, p. 226), as compared with a ay average length of stay at acute care hospitals (CX 15, p. 182). (25) 99. Tennessee and Georgia law treat nursing homes differently from acute care hospitals (Tenn. Code Ann. 68-11-201(1) (1983); Ga. Admin. Compo ch. 29G-5-B.21), and they are treated differently by the federal government for purposes of Medicare reimbursement (42 A. 1395d-1395f(West 1983). Both the Tennessee and Georgia state health plans (CX 169, pp. 278-91, 288, pp. 334-84) and the Health Systems Plan (CX 15, pp. 207-30) have separate sections discussing long-term facilities, the majority of which are nursing homes (CX 169, p. 282). When considering certificate-of:need applications nursing home beds are considered separately from acute care beds by the Tennessee Health Facilities Commission (Tr. 484) and the Georgia State Health Planning and Development Authority (Ga. Admin. Compo ch. 272- 09(8H9)). The health care industry also regards long-term care facilities like nursing homes as distinct from acute care hospitals (CX 174, p. xx; Tr. 159-60, 1514- , 2285). 100. The services provided by nursing homes are different from those provided by acute care hospitals. They do not have the facilities , p.

. -- _u._- 361 Initial Decision nursing services, to handle acutBly ill patients (Tr. 160, 471 , 553- 1400, 1761-62) and they cannot diagnose or treat ailments or follow up on their diagnosis and treatment (Tr. 1762). 101. Ifit decided to offer the same services as an acute care hospital a nursing home would have to purchase new equipment, hire a new staff and overcome the regulatory hurdles imposed by Tennessee and Georgia certificate-of-need laws (F. 252-77). 102. There are several inpatient mental health facilities in the Chattanooga area, including two free-standing hospitals (HCA's Valley Hospital and the state institute, Moccasion Bend) and four units in acute care hospitals (Erlanger (13 beds), Tri-County (15 beds), Cleveland Community (20 beds)) (CX 15, p. 268), and Bradley s substance abuse treatment unit (18 beds) (CX 15 , p. 256)). 103. Mental Health, or psychiatric, hospitals are generally recognized as distinct entities from acute care hospitals. Tennessee and Georgia have separate licensing classifications and requirements for psychiatric facilities (Tenn. Admin. Compo ch. 1200-8-1- 02(6); Ga. Admin. Compo ch. 290-5-6. 19). The Tennessee and Georgia state health plans have separate sections for mental health services (CX' 169, pp. 292 315, 288, pp. 164 204) as does the Health Systems Plan (CX 15, pp. 231-68). In addition, there is a separate Mental Health Systems Plan for the Georgia-Tennessee Regional Health Commission (CX 15, p. 231). Psychiatric beds are considered separately from acute care beds for purposes of decisions on certificate-of-need applications (Tr. 484; Ga. Admin. Compo ch. 272- 09(8)(b)). The health care industry also views psychiatric hospitals as distinct from acute (26) care hospitals (Tr. 2285-86). The Joint Commission on Accreditation of Hospitals has separate standards for psychiatric hospitals (CX 174 xx), and acute care hospital administrators do not view psychiatric hospitals as competitors (Tr. 159, 1505, 1964). Psychiatric hospitals have a separate SIC code (8063) from other health care facilities (1972 SIC Manual 104. Providers of psychiatric services offer different types of treatment from that which is available at acute care hospitals. Psychiatric treatment programs provide counseling and psychiatric therapy (for example, behavior modification, transactional analysis, and psychoanalysis), rather than treatment of physical ailments such as strokes or heart attacks (HCA's Response to Interrogatory 18(d), fied Apr. 20 1984; Tr. 159, 472, 1761, 3468). Psychiatric hospitals have specially trained personnel, such as social workers, psychiatric nurses, clinical psychologists, and psychiatric aides and technicians (HCA's Response to Interrogatory 18(e), fied Apr. 20, 1983; Tr. 3491), and they do not have the x-ray equipment, the laboratories, or the personnel neces- .

Initial Decision 106 F. sary for diagnosis and treatment of acutely il patients (Tr. 158, 472 1761 3470 3491).

105. Before they could offer the services of an acute care hospital psychiatric hospitals would have to purchase new equipment and hire qualified personnel. They would also have to meet state licensing and certificate-of-need requirements (Tenn. Code Ann. 68-11-103(5), (11), - 106(g) (Supp. 1983), as amended by THPRDA Amendments of 1984; Ga. Code Ann. 31-612), -40(b)). 106. While conversion of a psychiatric hospital to an acute care hospital would be a major undertaking, the psychiatric facilities in Chattanooga area acute care hospitals could easily be converted to other uses, although some personnel changes would also have to be made (Tr. 120 3469-71).

107. The beds in the psychiatric unit at Erlanger Medical Center are included within the hospital' s approved and licensed bed complement for medical/surgical services and are not used for long-term psychiatric care. Patients using these beds receive medical/ surgical and nursing care services (Tr. 345) and the beds could be used for non-psychiatric patients (Tr. 120).

108. A psychiatric treatment unit for short-term acute care patients was opened at Cleveland Community Hospital in Cleveland, Tennessee in October 1981 (Tr. 3468). Renovation ofa medical/surgical nursing unit for psychiatric care was completed at a cost of approximately $70 000 (Tr. 3469). The only difference between the psychiatric patient rooms and other patient rooms at Cleveland Community are the absence oftelevision and telephone (Tr. 3469), and conversion of the former to medical/surgical rooms would only require reassignment of staff (Tr. 3470). (27) 109. The psychiatric care offered at Erlanger, Tri-County, Cleveland Community and Bradley Hospitals is not comparable to that offered at the two psychiatric hospitals in Chattanooga (Tr. 159, 3469); while these services are specialized, they are similar to services which are provided, or which could be provided, by other acute care hospitals in the area. For example, Parkridge and East Ridge treat patients with psychiatric problems or disorders in rooms used for medical/ surgical care (Tr. 815 , 3284-85; see also Tr. 422). (2) Outpatient Care 110. Although physicians may perform some minor procedures in their offces which could also be done in an acute care hospital, their activities are completely different from those of hospitals (Tr. 4213). Recently, however, outpatient facilities have been opened which offer some of the same services as do the outpatient facilities of acute care 361 Initial Decision 111. For example, three emergicenters have been built in the Chattanooga area since the HAI and HCC acquisitions (Tr. 3290). An emergicenter is a convenience medical center, geared towards "medical problems normally treated by a family doctor " but which is open for longer hours than most physicians ' offces (CX 815; see Tr. 469, 554). Centra Care, which operates two of the Chattanooga centers, describes its centers as "a doctor s offce where you and your family can receive prompt medical treatment" (CX's 816B, 821). The types of ailments treated by emergicenters include, among other things, colds cuts, ear infections, pulled muscles, stomach aches, coughing, and childhood ilnesses (CX's 815, 816B, 821). Emergicenters also provide care for emergencies "which (do not) require back-up services from a hospital emergency room" (CX's 816B, 821).

112. Emergicenters are not equipped to handle truly il patients or serious emergencies. They do not have the back-up services of a hospital (CX's 816B, 821; Tr. 470, 1398), nor do they have on hand the kind of equipment usually available in hospital emergency rooms such as special monitors, life-saving equipment, equipment for taking care of major trauma, special catheters, and other items used to monitor very sick individuals (Tr. 470).

113. Since the HAI and HCC acquisitions, there have been proposals to construct an outpatient diagnostic center and a free-standing ambulatory surgicenter in Chattanooga (Tr. 469, 1766). Since these are only proposals, they have no present effect on the health care product mix in Chattanooga, but if one were to speculate on their future effect if they were built, one would have to conclude that they will not offer the same mix of(28) services as are offered by acute care hospitals (Tr. 1398, 1401--2, 1765-67).

c. Pediatric Services 114. There are four hospitals in the Chattanooga area that provide pediatric services. Two are acute care hospitals with pediatric units: Bradley Memorial and Tri-County Hospitals. The other two are Metropolitan Hospital and T.C. Thompson Children s Hospital at the Erlanger Medical Center (CX 15, p. 186).

115. The pediatric unit at Bradley has 14 beds and the unit at Tri-County has 20 (CX 15, p. 186), but the services provided at those units are not as specialized as those provided by pediatric hospitals. They are not equipped to handle more serious ilnesses or injuries (Tr. 455-56), and they do not have separate facilties such as operating rooms or laboratories (Tr. 107).

116. Metropolitan Hospital is a 64-bed hospital that treats primarily pediatric patients. It was formerly owned by the Tepper group, a group of physicians mostly pediatricians-who comprise the hospi- Initial Decision 106 F:r. tal' s medical staff. Although Metropolitan does treat some adult patients, primarily due to its developing diabetes program, it remains for the most part a pediatric hospital (Tr. 1089-91). 117. Metropolitan provides general pediatric services, but litle, if any, intensive care. Patients with serious problems are transferred to C. Thompson Children s Hospital (Tr. 105). Metropolitan does have some of the equipment for handling small children, but it is not as specialized as T.C. Thompson Children s Hospital (Tr. 456): 118. T.C. Thompson Children s Hospital, a 114-bed hospital located on the Erlanger Medical Center campus, is a highly specialized pediatric facility. Children s Hospital was an independent hospital until 1975 when it was moved to Erlanger (Tr. 96-97). Children s Hospital provides only pediatric services; it has a separate admitting offce emergency room, operating suites, laboratory services, radiology services, nursing care areas, and intensive care areas (Tr. 97, 104). Most of its patients are under six years old (Tr. 457). 119. Complaint counsel claim that the treatment of pediatric patients is different from that of adults (CPF VI 46-7) and that pediatric services are generally recognized as distinct from adult acute care services (CPF VI 48-50). Iftrue, such distinctions might support elimination of all acute care beds dedicated to pediatric services from consideration as part of the product market for health care in the Chattanooga area, but complaint counsel are much more selective; they argue that (29) while the pediatric beds at Bradley, Tri-County and Metropolitan should be considered equivalent to surgical-medical beds in acute care hospitals for purposes of computing market share those at T.C. Thompson should be excluded (CB, pp. 20-22). 120. T.C. Thompson pediatrics beds should be so treated because according to complaint counsel, they cannot easily be converted to adult use. As support for their claim, they cite the testimony of Mr. Lamb, Erlanger s former administrator, who stated that such conversion "would require major renovation, virtually gutting the entire area. . . ." (Tr. 103). They also argue that if Children s Hospital were to discontinue providing pediatric services, a certificate of need would be required (Tenn. Code Ann. 68-11-106(g)(1)(E) (Supp. 1983); Tenn. Admin. Compo ch. 0720-2-02(2) (c) (I3)).

121. With the use of additional instrumentation, the operating room at T.C. Thompson could be used for adult surgery (Tr. 101-02), but Mr. Lamb denied on cross-examination that adults could be accommodatedin themin the(Tr.patient399-400).rooms even if adult furniture were placed 122. Some of the pediatric rooms at T.e. Thompson are three-bed wards, and some are semi-private (two beds to a room), but well over half are private (Tr. 297-98). There is no reason why three-bed or 361 Initial Decision two-bed rooms could not be converted. to adult Lisage; however although he conceded that "square foot wise" (Tr. 400) an adult bed could be placed in a private room without violating state requirements 7 Mr. Lamb claimed that he would not "of choice" put a bed where it would have to be placed in the room with the patient facing the door (Tr. 400).

123. Mr. Lamb' s claim is based upon preference rather than physical limitations, and his preference seems not to be shared by other administrators since some of the patient rooms at Southern Hils Hospital, a new and very modern suburban hospital opened in Nashville, Tennessee in 1979, are designed so that the foot of the' patient' bed is pointed in the direction of the door to the room (Tr. 3397). 124. Furthermore, the recent renovation of pediatric facilities at Erlanger suggests that even if the existing pediatric facilities could not be used for adult care, the (30) conversion of these facilities to adult care could be accomplished without undue diffculty or expense. Following initial completion of the pediatric units at the Erlanger Medical Center, a pediatric intensive care unit was established by removing one partition between two-patient rooms and making a four-bed intensive care unit. Bathroom space in the original patient rooms was converted to storage space for the intensive care area (Tr. 309). Further pediatric intensive care unit renovation was in progress in 1983. This renovation involved expanding the area and adding to the number of intensive care beds (Tr. 303-04). The remodeling and expansion is expected to cost less than a million dollars (Tr. 333). The renovation includes changing the service areas in the center of the pediatric floor. Hallway walls ofthe patient rooms are being relocated as part of this renovation (Tr. 310-11).

125. If T.C. Thompson s facilities were unavailable, only a very small percentage of the patients hospitalized by Robert C. Coddington, M. , a Chattanooga area pediatrician practicing almost exclusively at Erlanger Medical Center (Tr. 489), would have to be hospitalized in specialty hospitals outside of the Chattanooga area rather than in another Chattanooga hospital (Tr. 507-08). 126. Most acute care hospitals could care for a younger patient needing traction for an upper leg fracture (Tr. 512). Most hospitals do tonsilectomies and most general acute care hospitals have facilities where an appendectomy could be performed on a six-year-old child (Tr. 495-96).

I27. Other area hospitals offer pediatric acute care. In 1981 , John L. Hutcheson, Tri-County Hospital, was operating a 20-bed pediatric 7The rooms measure 10 feet by 14 feet, or 140 square feet (excluding the private bathroom and adjoining alcove) (Tr. 315; RX 977). The State of Tennessee requires only 100 square feet for Ii private hospital room, with 3 feet of clearance around the sides and foot of the bed, which is administratively defined to be 3 feet by 7 feet 6 inches (Tr. 395). A bed of this size can easily be placed in a 10 foot by 14 foot room with the requisite clearance. Initial Decision 106 F. unit (CX 106HJ. Bradley Memorial Hospital was operating a 14-bed pediatric unit in 1981 (CX 65F). Tri-County and Bradley Memorial have operating rooms, and emergency and laboratory facilities dedicated to pediatrics (Tr. 107).

128. Parkridge Hospital does not have a specialized pediatric unit but has seven pediatricians on its medical staff with active admitting privileges and one with consulting privileges (Tr. 3343). There are also surgeons on the medical staff at Parkridge who received patient referrals from pediatricians, particularly in orthopedic surgery, general surgery, ear, nose and throat surgery and oral surgery (Tr. 3281). Parkridge has special equipment and supplies available for use with small children, including special instruments in surgery, special endoscopes in diagnostic departments, and special equipment in the emergency center (Tr. 3282). Parkridge would not need any additional facilities in order to accommodate the hospital work of an active group of pediatricians. In order to establish a pediatrics unit, Parkridge would make some cosmetic changes in some of the room decor and could be in a position to handle a substantial volume of pediatric work within a couple of months (Tr. 3283). Some of (31) the current nursing staff at Parkridge have training and experience in pediatrics and could be transferred to a specialized pediatric unit on relatively short notice (Tr. 3284).

129. North Park Hospital provides pediatric care including tonsillectomies, laryngotomies and general surgery such as appendectomies and hernia repairs (Tr. 3400).

130. East Ridge Community Hospital does pediatric work associated with its obstetrics practice and nursery, and also does tonsilectomies, appendectomies and accident-related pediatric care through the emergency room (Tr. 805). Twenty to thirty pediatricians are on the staff at East Ridge Community Hospital, but most pediatric admissions are through family practices or the newborn nursery (Tr. 806). 131. Cleveland Community Hospital in Cleveland, Tennessee does not have a pediatric unit but does treat children, and there are pediatricians on the staff of the hospital. Pediatric admissions are also made at the hospital by general practitioners or family practice physicians. The hospital provides medical devices and cribs for pediatric patients (Tr. 3467).

132. Metropolitan Hospital (formerly Tepper Hospital) provides specialized pediatric care. Most of the admissions by the Tepper Group, a major pediatric practice in Chattanooga, are made to Metropolitan Hospital (Tr. 146).

133. Pediatric discharges from some of the hospitals in the Chattanooga urban area without specialized pediatric units in 1981 included 44 discharges from Red Bank Communitv Hospital (CX 67GJ. 100 361 Initial Decision discharges from East Ridge CommunityHospitalCCX 63G), 314Irom Parkridge (CX 36G), and 538 discharges from Memorial (CX 62G). 3. Conclusion 134. I agree with HCA that non-hospital providers such as diagnostic and emergency care centers, and other providers such as doctors offces, over some of the same services as do acute care hospitals (RPF 287), and I agree that the relevant product market must encompass all services provided by acute care hospitals and those services offered by non-hospital providers which are an alternative to hospital care" (RPF 359); however, while individual services offered by an acute care hospital may be identical to ones offered by non-hospital providers, the latter are not an alternative to the kind of care which only acute care hospitals can provide: the unique combination of services which the acute care patient needs (F. 77, 80-1). For (32) that reason, I find that the relevant product market consists of the cluster of services offered by acute care hospitals, and that the best measure of the extent of those services includes outpatient as well as inpatient care, since acute care hospitals compete with each other in offering both kinds of care (F. 90-96) and since, as both Dr. Salkever and Dr. Harris agree, acute care outpatient facilities feed patients to the inpatient facilties (F. 76-79).

135. Including outpatient care in the acute care hospital market is not inconsistent with the position that outpatient facilities such as doctors' offces and emergicenters do not compete with acute care hospitals, for outpatient care offered by acute care hospitals is an inseparable part of the cluster of services which they offer. In any event, HCA concedes that, even if it could be quantified, the volume of services offered by non-hospital providers at the time of the acquisitions would not greatly affect the market positions of its owned and acquired hospitals (RB, p. 37).

136. Acute care hospitals cater to a different type of patient than do psychiatric hospitals (F. 104), but there is no sound reason why beds devoted to short-term psychiatric care in acute care hospitals should be viewed differently than regular medical/surgical beds in acute care hospitals (F. 106-9); therefore, revenues of psychiatric facilities in acute care hospitals should be counted along with the revenues generated by medical/surgical beds in such hospitals. 137. I also reject complaint counsel' s elaborate attempt to carve out an exception to their "cluster of services" concept so that the 114 beds at T.C. Thompson can be excluded from the acute care hospital market. The simplest rejoinder to their claim is that T.C. Thompson competes with the pediatric facilities of other area acute care hospitals (F. Initial Decision 106 F. 126-33).8 Since those facilities' beds are included in complaint counsel's acute care hospital market, so should T.C. Thompson s. Furthermore, I reject the claim that T.C. Thompson s pediatric beds could not be converted to adult beds, since T.C. Thompson s physical plant has been changed in the past (F. 124), and there is no legal impediment to the conversion of pediatric private rooms to adult rooms if pediatric services are not entirely eliminated (Tr. 305). (33) 138. Because psychiatric hospitals and nursing homes offer a different group of services to a different clientele, the long-term patient (F. 100, 104), they do not compete with acute care hospitals and should not be included in the acute care hospital market. G. The Relevant Geographic Market 1. Expert Opinion 139. Dr. Salkever defined the relevant geographic market in this case as the:

(A)rea within which patients view alternative providers as potential substitutes, and an area within which these alternative providers, that is, different hospitals in this case are competing with one another for the same groups of patients (Tr. 2295). 140. Dr. Salkever determined this area by applying the Elzinga- Hogarty test to patient flow data which reveals to which hospitals patients, as a practical matter, can turn for care. If patients in a particular area make substantial use of hospitals outside the area that implies that hospitals outside the area could act as a check on the exercise of market power by hospitals inside the area (Tr. 2518-19); if a substantial number of patients from outside the area travel into an area for hospitalization, that indicates that hospitals located in the areas where those patients reside could act as a check on the exercise of market power by hospitals inside the area (Tr. 2521-25). 141. The well-known and respected Elzinga-Hogarty test is based on LIFO ("little in from outside ) and LOFI ("ittle out from inside statistics.9 A LIFO statistic signifies the percentage of hospital patients from a particular area who remain in the area for hospital services (rather than use hospitals outside the area) (see CX 800A). A LOFI statistic signifies the percentage of patients in an area s hospitals who reside in the (34) area (rather than outside the area). Under the test as applied to hospital markets, if few patients leave an area T.C. Thompson does provide specialized services for it. patients (Tr- 100-2 448-9) and has II staff which is specially trained to treat children (Tr. 100, 452-4), but Erlanger (and presumably other hospitals) offers special cllre in an eye center hud a tumor center whose beds complaint coulls.l would not exclude from the acute care market (T. 107, 421, 490-91).

See Elzinga-Hogarty, The Problem o(Geogruphic Market Delineation in Antitrust Suits 18 Antitrust Bull. 45 (1973); Elzinga-Hogarty, The Problem o(Geographic Market Delineation Reuisited: The Case o(Coal 23 Antitrust 361 Initial Decision and few patients enter an area to obtain hospital services, that is strong evidence that the area constitutes a relevant geographic market. Dr. Salkever testified that to determine the boundaries of a market, one begins with the particular geographic area in which the principal facilities involved in the merger are located; if the LIFO and LOFI statistics for the area are not suffciently high, the area under consideration is expanded until the LIFO and LOFI statistics are high enough so that the percentages of "imports" and "exports" are no longer substantial (Tr. 2304--5, 2504--5). If both LIFO and LOFI statistics (or the average ofthe two) exceed 90%, the Elzinga-Hogarty test for a "strong market" is satisfied. Ifthe standards for a "strong market" are not satisfied, and both LIFO and LOFI statistics exceed 75%, the market is a "weak market" (Elzinga & Hogarty (1978) at 2). 142. Of the possible relevant geographic markets in the Chattanooga area, Dr. Salkever first considered Hamilton County, where most hospital beds, including those of HCA, HCC and HAI, are located; he observed, however, that while many patients living in the county use Hamilton County hospitals, hospitals in the county draw roughly one-third of their patients from outside the county, suggesting that it could not by itself be considered the relevant geographic market (Tr. 2300). Furthermore, while he recognized that the HSA was used by health authorities for planning purposes, patient flow data also suggested "that the HSA was not an appropriate area" (Tr 2300).

143. Dr. Salkever concluded that, in his judgment, the "best" relevant geographic market using the Elzinga-Hogarty test was the MSA plus Bradley County, although he stated that there might be some justification for viewing either the MSA or the HSA as relevant geographic markets if someone insisted on it (Tr. 2307). 144. It is readily apparent that deciding whether LOFI and LIFO percentages are acceptably high and what weight they are to be given is a matter of judgment (Tr. 2298), for complaint counsel propose different relevant geographic markets than the one suggested by Dr. Salkever.

145. Dr. Harris, HCA' s expert, rejected all of the markets proposed by Dr. Salkever and complaint counsel and testified that the correct market-the area in which physicians, the buyers with real decisionmaking power, have a choice of acute care hospitals to which to admit their patients-in this case is Hamilton County, Tennessee and Walk- , Dade, and Catoosa counties in Georgia ("the Chattanooga urban area ) (Tr. 3959).

146. Dr. Harris based his conclusion on an analysis of two factors physician utilization and patient origin. (35) Initial Decision 106 F. 2. Physician Utilization 147. Dr. Harris testified that whether hospitals occupy the same relevant geographic market is determined, to a great extent, by physician admitting practice, for physician preference, rather than patient choice, decides what hospitals wil be used because "for the great majority of people we re talking about in a health care market like Chattanooga, you pick your doctor and then your doctor is' the one who s going to decide where you re admitted" (Tr. 3965). 148. After reviewing RX 1081 (lH169), which lists, for each hospital in the Chattanooga area, the physicians by specialty who admitted to the hospital, and the number of inpatient days that each physician was responsible for in all of the hospitals in the area (Tr. 3961), Dr. Harris found that, with few exceptions, every physician who admitted to Erlanger admitted exclusively to the hospitals in the urban Chattanooga area, and concluded that this pattern was true of other downtown or urban Chattanooga area hospitals (Tr. 3961-63). 149. An appraisal of RX 1081 and testimony by Chattanooga-area physicians and administrators supports Dr. Harris' conclusion. There is general agreement that physicians playa primary role in determining where their patients are admitted (Tr. 366-67, 488, 785, 1147 1378- , 1629) and that physicians in Chattanooga limit their practice to one, two or at most three hospitals within a limited area (Tr. 499, 1549, 1755). The hospitals within a limited area (Tr. 499, 1549 1755). The medical staff afIliations held by a few Chattanooga physicians at hospitals outside of the Chattanooga urban area are limited to consulting or courtesy privileges (CX 892). As a rule, those few Chattanooga physicians who do have consulting or courtesy privileges at outlying hospitals do not actively practice at the outlying hospitals (RX 1114).

150. ("' J (RX 1081). ("') (RX 1081 (4, 7, I4, 15)). ("' ) (RX 1081 (7 14)).

151. ("'J (RX 1081 (27--2)). (36) 152. Approximately 350 physicians are on the staff at Parkridge Hospital (Tr. 3276). They have their offces in the Chattanooga urban area and admit and treat patients at one or more of the urban area hospitals in addition to Parkridge (Tr. 3277). l" ') (RX 1081 (43-57)). 153. (' " J (RX 1081 (58-63)).

154. (' " J 1981 (RX 1081 (68-72)).

155. Three internists and one surgeon who comprise the Newell Clinic Group account for 90 to 95% of the admissions at Downtown General Hospital (Tr. 3671). (" ' ) (RX 1081 (88-89)). 156. (" ' J (RX 1081 (94-95)).

11;7 r l fry inal (AA Q.11) -ity -ilo;..1u thd f1i';uTnru;:+i.. l'",nh"l' n."'Olln ( ( .L.L""".J .L.Lru... ""''''.LU. ''.L .H'''.. L.''''''-' 361 Initial Decision which founded Diagnostic Center Hospital adjacent to the medical group s offces, accounts for all admissions to Diagnostic (Tr. 3593; RX 1081 (83)).

158. (m) (RX 1081 (117-19)).

159. Physicians on the medical staff at Cleveland Community are not on the staff at any hospitals in the Chattanooga urban area (Tr. 3465). (mJ (37) (RX 1081 (128-29)). ) (RX 1081 (129)). 160. ("'J (CX 82Z-9, Z-18). J (RX 1081 (123)). 161. (" ' J (RX 1081 (134)).

162. The two physicians with admitting privileges at Whitwell Community Hospital in 1982 did not have privileges at any hospitals in the Chattanooga urban area (CX 892).

3. Patient Origin 163. Patients admitted to Chattanooga urban area hospitals who live outside of the Chattanooga urban area are, with few exceptions in need of specialized care and treatment not available in their own communities (Tr. 148, 1605, 3277, 3344, 3968). 164. Approximately 60% of the admissions to Erlanger Medical Center comes from Hamilton County with the remainder primarily from six ofthe several counties contiguous with Hamilton County (Tr. 120; RX 761 (63)).

165. A number of patients from northwest Georgia, northeast Alabama, western North Carolina and from Tennessee counties within a 50 to 75 mile radius of Chattanooga are referred to the Erlanger medical staff for tertiary or specialty care (Tr. 120-22). 166. (" ' J (RX 273 (9)). Patients admitted to Memorial who live outside of Hamilton County come mostly through referrals between physicians (Tr. 1605). Memorial Hospital admits cardiology referrals from Dalton, Georgia and neurological referrals from Bradley County, Tennessee (Tr. 1504-05).

167. The vast majority of patients admitted to Parkridge Hospital come from within the Chattanooga urban area, the practice area for the Parkridge medical staff. Some patients from outside the urban area are referred to physicians on the Parkridge medical staff for services not available in the outlying communities (Tr. 3277, 3344). (38) 168. Approximately 85% of the patients admitted to Tri-County reside within the three suburban Georgia counties near Chattanooga (RX 361 (17)). Over 50% of the people residing in the three suburban Georgia counties who are hospitalized are admitted to hospitals in Hamilton County, Tennessee (RX 361 (82)).

169. Most of the patients admitted to East Ridge reside in the communities situated in the southeast portion of Hamilton County ., ,.., .

Initial Decision 106 F. adjacent to I-24 and I-75 and from the Georgia communities across the state line along Interstate 75 (Tr. 75&-57). Patients admitted to East Ridge from areas beyond the Chattanooga urban area are usually coming to the hospital on a referral basis (Tr. 683). 170. The patients admitted to Downtown General come primarily from the downtown Chattanooga neighborhoods surrounding the hos-pital (Tr. 3673).14). 171. (*"J (CX 34E; see CX 15 , p. 172. The primary service area for Whitwell Community Hospital is comprised of limited portions of Marion and Grundy Counties in Tennessee (CX 404Z-70, Z-79, Z-80). Whitwell is the only hospital within a reasonable driving time for persons in this area and is not viewed by its patients as being in competition with any other hospital (CX 404Q).

173. The few residents of counties such as Marion and Sequatchie who are patients in Chattanooga hospitals, according to county-oforigin data, may simply live near the county line (Tr. 3969, 4197). 174. The Census Bureau reported that in 1980, 106 843 persons living in Hamilton County worked in Hamilton County, and 18 866 residents of the suburban Georgia counties of Dade, Walker and Catoosa worked in Hamilton County (CX 822L, T, Z-16; RX 1112 (3)). 175. In contrast, only 848 persons residing in Sequatchie County reported their place of work in Hamilton County, as did 832 residents of Rhea County, 2 493 residents of Marion County, and 1 750 residents of Bradley County (CX 882Z-3, Z-23, Z-29, Z-34; RX 1112 (3)). 4. Hospital Administrators' Testimony 176. Chattanooga urban area hospital administrators do not view outlying community hospitals as competitors (Tr. 3597- , (39) 3288), and administrators of the outlying rural and community hospitals do not view the Chattanooga urban area hospitals as competitors (Tr. 3475 617-18; CX's 82Z-19, 404Q, 17Z-35).

177. Cleveland Community Hospital identifies as its competitors only Bradley County Memorial Hospital, Athens Community Hospital, Woods Memorial Hospital and Rhea County Medical Center (Tr. 2532; CX 235Z-11; see also Tr. 3475). 178. ( ) (CX 82Z-14), ( ) (Tr. 2535; CX 82Z-19). 179. Sequatchie General Hospital identifies only Bledsoe County, Whitwell and South Pittsburg as competitors (Tr. 2529-30; CX 34Esee also Tr. 620).

180. Whitwell Hospital identifies no competitors, but states that patients needing more specialized care than is available in Whitwell ....r r11 jj rrt ,.......

361 Initial Decision 5. Conclusion 181. Although they prefer use of the Chattanooga MSA, complaint counsel suggest that market concentration can be measured in a smaller area, Hamilton County, and a much larger area, the HSA. They do not express much enthusiasm about the market chosen by their expert, Dr. Salkever (the MSA plus Bradley County) or by Dr. Harris (Hamilton County plus Dade, Walker, and Catoosa Counties in Georgia) (CPF VII 3).

182. Although Hamilton County has the largest population and the greatest concentration of hospitals in the area, neither Dr.Salkever nor Dr. Harris viewed it as a satisfactory geographic market. Hamilton County (or any other larger geographic area) does satisfy the LIFO standard of the Elzinga-Hogarty test but, with a LOFI value of some 68%, not that standard because about one-third of the patients who use Hamilton County hospitals come from outside that county (F. 142). The LIFO/LOFI calculations for the five geographic areas analyzed by complaint counsel and HCA are: (40) LIFO LOFI Hamilton County 98. 67. Chattanooga Urban Area 98. 78. Chattanooga MSA 98.4% 81.2% Chattanooga MSA and Bradley County 98. 82. Georgia-Tennessee Health Services Area 94. 88. (RX 1087).

183. The Chattanooga Urban Area, the MSA, the MSA plus Bradley County and the HSA all satisfy the Elzinga-Hogarty test. Neither of the experts viewed the HSA as a realistic market, and my analysis of physician referral patterns and testimony of knowledgeable Chattanooga area hospital offcials convinces me that physicians and patients do not, as a general rule, regard all acute care hospitals in the HSA as potential providers of health services to them. 184. With few exceptions, physicians in the Chattanooga urban area refer their patients to hospitals in the urban area (F. 148-2). Physicians in outlying counties do refer some patients to urban area hospitals, but usually only for services which are not available in the community hospitals (F. 163, 165--67). Regular medical treatment is generally speaking, sought in the area near where the patient lives and the doctor has privileges.

185. Generally, hospitals in the outlying counties ofthe HSA regard hospitals in the same or adjacent rural counties as competitors; they do not normally consider the urban area hospitals as competitors (F. 177-80). Supporting this view is testimony by urban area hospital Initial Decision 106 F. offcials that their competitors are the other urban area hospitals (F. 176).

186. The MSA, which contains Hamilton County plus Marion and Sequatchie counties in Tennessee and Dade, Catoosa and Walker counties in Georgia, was rejected by Dr. Harris as a relevant market and was not proposed by Dr. Salkever. I find that this proposed market, as well as the one suggested by Dr. Salkever, the M,SA plus Bradley County is too broad, and does not reflect competitive reality, as evidenced by physician referral patterns. These patterns disclose that physicians in the Chattanooga urban area do not usually refer patients to outlying communities in Marion, Sequatchie, Bradley, or other counties in the HSA (F. 148). (41) 187. All proposed relevant geographic markets except Hamilton County meet the LIFO/LOFI requirements of the Elzinga-Hogarty test, but the market proposed by Dr. Harris, the Chattanooga urban area, is the smallest to satisfy that test, and increases the LOFI value by 10 points over the Hamilton County area, whereas the market proposed by Dr. Salkever and those chosen by complaint counsel increase the LOFI value only marginally, suggesting that one need not seek beyond the confines of Dr. Harris' market to find the relevant geographic market.

188. This suggestion is borne out by analysis of referral patterns and industry opinion, which confirm that Chattanooga urban area hospitals compete with each other but not with hospitals in outlying area of the MSA or the HSA (F. 44-2). In conclusion, I find that the relevant geographic market is the Chattanooga urban area Hamilton County in Tennessee and Dade, Walker and Catoosa counties in Georgia.

H. HCA's Managed Hospitals 1. HCA's Hospital Management Program 189. HCA is the largest manager of hospitals for other owners in the country; with the acquisition of HAI, it increased the number of hospitals it managed from about 50 to 125-30 (Tr. 3755). HCA manages hospitals owned by others to earn a profit from management fees. Another purpose is to give HCA an opportunity to test local regulatory environments and health care market potential in parts of the country where it has no operating experience. Of the approximately 250 hospitals HCA has managed under contract, 10 or 12 were acquired later (Tr. 3737, 3740, 3742; CX's 13N, 895B). HCA's whollyowned subsidiary, Hospital Management Company, handles its management contracts (Tr. 3627, 3744-5).

1 Qn Mnct J.i:A hnc;n;t rno:n.:mo:n"'an t ronnf"'0;..tc ",..,a f'",.. ,.1,..",,, fro hU," 361 Initial Decision year terms with a fixed fee, and-policymaking control is vestedin the hospital owners. Corporate policy precludes variable fee contracts (Tr. 3746). The retention oflocal control is an important consideration for hospital owners in the communities they serve, and HCA management contracts expressly provide for this control (Tr. 3746-7). Some management contracts are cancelled before they expire, and some are not renewed; most, however, are renewed (Tr. 3751-52). 191. HCA recruits, employs and trains the hospital administrator and controller for most managed hospitals, provides day-to-day management services through the hospital administrator and c,Ontroller and provides specialized support services on an as-needed basis (Tr. 3749-50). (42) 192. The management objectives for an HCA-managed hospital are determined by the local governing body representing the owner. HCA does not determine the financial objectives for a managed hospital (Tr. 3752). Rate setting is a function of the financial objectives, and rate recommendations by HCA for managed hospitals suggest what rates wil accomplish those objectives (Tr. 3754, 3703). 193. There is no regular contact between corporate executives responsible for owned and managed hospitals or between owned and managed hospital administrators within HCA (Tr. 3743-45, 3625 3782- 3333); however, HCA employees do switch between positions with the HCA management subsidiary and the subsidiary responsible for owned hospitals (Tr. 3647).

194. HCA-managed hospitals are not required to prepare annual management plans as is required ofHCA-owned hospitals. The owner of a managed hospital decides what reporting mechanism wil be used by the hospital administration (Tr. 3485).

195. Contract-managed hospitals are not required to use the HCA national purchasing contracts, as are HCA hospitals unless they can justify a non-contract vendor because of a lower price (Tr. 3486 3751). 2. Chattanooga Area Hospitals Managed By HCA I96. HAI previously owned the two hospitals that became Downtown General and Red Bank Community Hospital (Tr. 1I 74 1425- 1040). Both hospitals were badly in need of modernization. Rather than build new hospitals itself, HAI in 1976 with respect to Downtown, and in 1977 with respect to Red Bank, arranged to have the new facilities built through tax-free bonds.lO Downtown and Red Bank were completed in 1976 and 1977 respectively (Tr. 1047-49; CX 27D). The cities of Chattanooga and Red Bank financed the building of the hospitals through their Health and Educational Facilities Boards to In the case of Red Bank, HAl had to arrange for the estahlishment ofa Health and Educationall"acilities Board to authorize issuance of the bonds. The members of that board were selected byHAI ('fr. 1076 , 1079). _ .. ,.,.\ .. . 404 FEDERAL TRADE CgMMISSlONDECISlONS Initial Decision 106 F. which then leased the hospitals to two newly organized corporations Downtown Hospital Association and Red Bank Hospital Association , 1076; see Tr. 1418-20; CX 212,with an option to purchase (Tr. 1047-48 628-29). HAI arranged for the formation of these nonprofit corporations with boards of directors it selected to run the new hospitals, and for itself to manage the two hospitals under long-term contracts. HAI in developing the bond indenture for Red Bank, made sure (43) that the language required the management company to be affliated with the hospital for the life of the bonds, which is 25 years (Tr. 2026; CX' 628Z-23, 211Z-24). HAI retained ownership of the land under which Downtown was built until it was acquired by HCA (Tr. 1419). HCA stil owns that land (Tr. 1419, 1463-B4, 3699, 3701). 197. From the beginning of the plans to build the new hospitals, it was understood that HAI would manage them (Tr. 1425). The typical management contract in the hospital industry at that time provided for a fixed fee and a term of 2-3 years (Tr. 1054). However, the management contracts for both Downtown and Red Bank were for 25-year terms with HAl's compensation set at 8% of the hospital' s gross revenues (CX's 185Z-1, 624F, 185Z-2, 624H). This resulted in much higher management fees than would otherwise have been paid (Tr. 1054). The fees HAI charged Red Bank were the highest in terms of 1973). In the case of revenues per bed in the entire company (Tr. Downtown, the Board agreed to pay the taxes on the land under the hospital even though HAI owned it (Tr. 1465, 3699-3700). Downtown pays these taxes in addition to the $60,000 a year paid to HAI (and now HCA) for leasing the land (Tr. 3699).

198. The 25-year contracts remained in effect for Downtown and Red Bank until the Department of Health and Human Services Medicare ) determined that it would not reimburse the hospitals for much of the management fees they paid, since it felt that HAI was a party related to both hospitals and that the fees were neither negotiated arms-length'J nor were allowable costs for fees paid to a separate entity (Tr. 1193, 1442, 1972). Downtown had to repay Medicare between $100,000 and $200 000, which put the hospital in a diffcult financial position. Because of the Medicare problem both management contracts were renegotiated in 1981 (Tr. 1192, I442; CX's 185A, 189A). The new contracts called for HAI to manage the hospitals for terms of four years (CX's 185I, 189G).

199. The renegotiated Downtown General contract was approved by Blue Cross after it decided that HAI and the hospital were not "related persons" under the 1981 contract (Tr. 1474, 1476; RX 545), and the administrative appeal ofthe Medicare disallowance ultimately determined that HAI and the Red Bank Hospital Association were not -,1 - - - 1 - fTV\7 1 ,. .... , . ,.. .u. .......u._ 361 Initial Decision 200. In 1981, HCA, as successor to HAI; began to manage three hospitals in the Chattanooga area; Downtown General Hospital and Red Bank Community Hospital (Tr. 3624-25; CX 8F), as well as South Pittsburg Hospital. HCA agrees that to the extent that it may be relevant, South Pittsburg may be treated as an HCA-owned hospital (RB, p. 51, n. 1). However, since this hospital is not located in the relevant geographic market-the Chattanooga urban area--nly the extent of HCA's control over Downtown General and Red Bank is at issue here. The testimony of (44) Mr. Arnold Stulce, a member of Red Bank Community Hospital's board of directors (Tr. 117H, of Mr. Chambliss, a member of the board of Downtown General (Tr. 1417) and of Mr. Bennet, a former administrator of the hospital (Tr. 3670) reveals that HCA had no control over the policy formulation of either hospital.

201. The current HCA contract for management of Downtown General provides for a fixed annual fee and wil expire in January 1985 (CX 189F-G).

202. The control of and responsibility for Downtown General Hospital is vested in its board of directors (Tr. 3678) while the day-to-day operations of the hospital are the responsibility of the management company under the supervision of the board of directors (Tr. 1432). 203. All matters of policy are determined by the Downtown General board (Tr. 1432). The board, frequently in response to medical staff requests or proposals, sets objectives with respect to occupancy rates new medical services, quality standards, accreditation, community and patient relations, employee relations, inventory control, accounts receivable, physician recruiting, pension plans, net revenues, and long-range planning (Tr. 3678-79, 3683- , 3690, 3695). 204. The Downtown General board establishes policies which determine generally what rates wil be charged (Tr. 3703), and the hospital administration advises the board as to what the rates are and suggests what rates will be needed in order to meet the hospital's objectives (Tr. 1436).

205. The Downtown General Board believes that the hospital should provide services at the lowest possible cost (Tr. I480), and does not want the hospital to build up any large cash reserves. Their objective is to have revenues cover expenses with a small amount for contingencies (Tr. 3686-87). If the administrator assigned to Downtown General by HCA were to recommend that the hospital pursue profit maximization, the board would request that HCA change administrators (Tr. 1481-82).

206. Although the management company is responsible for actually hiring employees, it must secure the approval of the board for staff employment. There have been occasions when the board has vetoed ___ __ _ ..

Initial Decision 106 F. management recommendations regarding new staff (Tr. 1432- 1437).

207. The board of Downtown General participates in the review vendor proposals to the hospital (Tr. 3683) and it also reviews proposed budgets and expects to hear analysis of the budget by the administration and management company representatives (Tr. 1483). Board meetings during which its members review and ask.questions about recommendations by management company representatives may be extremely contentious (Tr. 1439). (45) 208. The hospital administrator serves at the pleasure of the board, which would not hesitate to take issue with HCA if a new administrator proved to be unacceptable (Tr. 1477, 1479). Downtown General administrators have always had primary loyalty to the hospital' board and have not hesitated to criticize HCA practices ifthey did not agree with them (Tr. 1434).

209. The Red Bank Community Hospital board of directors has at all times been responsible for establishing the general principles by which the hospital operates. The board has the ultimate budget authority for the hospital, and must approve any unbudgeted expenditure in excess of $5 000. The board has authority for final review and approval of hospital rates and final authority for admitting physicians to the medical staff. Only the board can amend the corporate by-laws for the hospital (Tr. 1215-16).

210. The Red Bank board has authority to exclude representatives of the management company from a board meeting, and there have been occasions when such persons were excluded (Tr. 1217 , 1224; see CX' s 428A, 429B).

211. Red Bank hospital contracts, such as those with a physician group for particular services, are negotiated by a committee composed of board members appointed by the board chairman (Tr. 1217). Employee hiring by the management company must conform to guidelines established by the board (Tr. 2035-36). 212. The hospital board oversees and supervises the contract services provided by the hospital management company (Tr. 1209). The board members actively participate in board meetings and often question recommendations made by the hospital administrator (Tr. 1218). The Red Bank Board is composed of people of good judgment; the board is not a "rubber stamp" (Tr. 1210, 1219), although the boards of Red Bank as well as Downtown General rely heavily on the advice of their managers and often accept their recommendations (Tr. 1057 1059, 1080, 1190-92, 1435, 1438-39, 1954).

2I3. HAI, as contract manager, was responsible for the day-to-day operations of Red Bank prior to the 1981 acquisition of HAI by HCAaT! ....,.n\ TTnA l_- L__ TI_ J D__ 361 Initial Decision Community Hospital (Tr. 3628). The Red Bank hospital board req\!ested termination of the management contract in 1981, because they believed that HCA could not properly manage Red Bank while at the same time owning North Park Hospital (Tr. 1198, 2023). Initially, HCA was against termination of the contract and sought to persuade the Red Bank Board that HCA could avoid any conflicts arising from its ownership of a competing hospital (Tr. 1204). (46) 214. The termination agreement between Red Bank and HCA relieved the hospital of its obligations to pay certain past due management fees and obligated HCA to pursue a reimbursemen,t dispute which predated the termination agreement (CX 87). 215. Following Red Bank's termination of the management contract with HCA, the hospital contracted for management services from Carolinas Hospital and Health Services (CHHS) of North Carolina (Tr. 1205; CX 24C). The duties and responsibilities of the contract manager under the current management contract are generally the same as they were under the predecessor contract with HCA (Tr. 1206).

216. The current contract between Red Bank and CHHS requires that the administrator be acceptable to the hospital board. The same requirement was included in the predecessor contract with HCA (Tr. 1221), and Red Bank exercised its right in 1982, with the result that a new HHS administrator was appointed (Tr. 1222-23). 1. Competition Among Acute Care Hospitals 1. History 217. Because of the unique nature ofthe patient-doctor relationship and the existence of third-party providers, competition in the health care industry has, until very recently involved considerations other than price (Tr. 2332-35).

218. The health care consumer-the patient-relies upon his physician for basic information about the nature of his ilness and its treatment (Tr. 3857-58). The physician acts as the patient' s agent by arranging, if necessary, for hospital admission, diagnostic tests, treatment and release (Tr. 3848, 3858-59), and the patient therefore has no control over costs associated with the treatment mandated by his physician.

219. In addition, the patient usually need not concern himself about medical costs because ofthe prevalence ofthird-party reimbursement plans (Tr. 3839-40), both private and public. The private insurance area is dominated by the Blue Cross-Blue Shield plans and includes several other commercial insurance companies. The public reimbursement area is dominated by Medicare and Medicaid (Tr. 3834). ,.

Initial Decision 106 F. 220. Most ofthe private insurance programs reimburse on a charge basis. A charge-based payor reimburses health care providers in accordance with the provider s charge list for health care attributable to patients covered by that insurer. Medicare and Medicaid reimburse health care providers on a cost basis. A cost-based payor reimburses health care providers for (47) the proportion of the institution s costs attributable to patients covered by that insurer (Tr. 3835). Thirdparty reimbursement of health care costs has been pervasi e during the last two decades. By 1967, approximately 90% of hospital expenditures were covered by third-party reimbursement programs, and that figure has remained relatively constant to the present (Tr. 3833-34; RX 1105 (9)).

221. Because there is so little consumer sensitivity to price in the health care industry, there is a low price elasticity of demand (Tr. 2384 4237-39) and, as a result, price competition among hospitals was limited in the past; instead, hospitals competed primarily for physicians to admit their patients by offering them non-price inducements (Tr. 2333-35).

2. Competition Among Chattanooga Area Hospitals For Patients 222. Physicians in Chattanooga are usually members ofthe medical staff of more than one hospital (Tr. 172, 1375; CX 892A-Z-27), and because they can choose among them to admit their patients, area hospitals compete for physicians to use their services (Tr. 134, 547 697- 1524 2045 3287 3401 3474-75).

223. Chattanooga hospitals compete for physicians by offering new services, sophisticated equipment and qualified personnel. Equipment purchasing decisions are often influenced by equipment available in competing hospitals. For example, both Memorial and Erlanger provide heart catheterization equipment, and Erlanger has updated its equipment in response to the better quality of Memorial's (Tr. 174-75). Competition among Chattanooga hospitals by providing sophisticated equipment (Tr. 172- , 724, 1093, 1392- , 1539) is considered by area administrators as beneficial to patients; they do not believe that such competition leads to unnecessary duplication because equipment purchases are not made unless they are financially possible (Tr. 461-62, 1392- , 1542 3352- 3458) and will improve the quality of care (Tr. 174).

224. Chattanooga hospitals also compete in the recruitment qualified medical personnel (Tr. 17I, 698, 1546-7; CX's 18Z-7, 29L 235Z-31) and offer competitive salaries and benefits to retain them (Tr. 569-70, 1560, 2006; CX's 17Z-38, 143, p. 25, 163A- , 191C, 227 A 237 A, 519A-C, 522A- , 526A-C). Area hospitals also compete by offerloa !'Hr-;:u"'tlun ,.-iit;,::u;;: !H' T)pr li7Pr1 rl' tn HU),lr nti,:nt ('Tr 361 Initial Decision 171, 180, 562, 569, 1546). FinallYJ hospitals jn the Chattanooga aea have competed to obtain certificates of need for new beds, services and equipment (Tr. 547, 738-0). (48) 3. Price Competition Among Chattanooga Area Hospitals 225. Hospitals in the Chattanooga area keep abreast of each others prices by conducting rate surveys, both formal and informal (Tr. 633- , 1958-59; CX 276A-E); almost every hospital in the HSA has conducted such surveys at one time or another (CX's 145A-C (Athens), 147A-0 (South Pittsburg), 148Z-75 ( ) 179A-C (Diagnostic), 184A- (Downtown), 276A-E (Parkridge), 279A-C) Sequatchie), 286A-E (Red Bank), 316I (Erlanger), 326H (Tri-County), 845 (Medical Park)) or has participated in them (CX 145A-C (Woods, Bradley), 158A-C (Memorial), 161A-C (East Ridge), 279A-C (Bledsoe, Rhea, Whitwell). HCA administrators testified that the surveys which they conducted were never used for setting rates (Tr. 3306, 3599), but they have been used by area hospitals to determine whether the surveyor s prices were within the range of prices offered by competing hospitals (Tr. 564) to justify price increase to Blue Cross (Tr. 1965-66; CX's 839, 841, 845). 226. Hospital rates in Chattanooga have not been established without some reference to competitors' rates, for prior to the challenged acquisitions, hospital prices tended to fall into a pattern. Erlanger prices were always the highest because of its tertiary services, its teaching function, and its obligation to provide indigent care (Tr. 176 566 727-28). Memorial' s prices were perceived to be the lowest, with the other hospitals' prices somewhere in the middle (Tr. 208 , 566 727-28). Other hospitals in Hamilton County tried to keep their prices not too far above Memorial's and somewhat below Erlanger s (Tr. 1070; see CX 486G). Although denying that price plays a significant role in hospital competition, Dr. Harris acknowledged that hospital administrators in Chattanooga are not totally uninterested in their competitor s prices:

Well they don t want prices too low or they are going to look like they don t sell Tiffany lamps, and they don t want prices too high or maybe they wil show up in the newspaper as somehow gouging the public. . (Tr. 3892-93). (49J 227. The surveys in the record, many of them conducted by hospitals owned or managed by HAI l1 listed room rates at some other hospitals, as well as charges for a limited number of ancilary services Jl An HAT policy required administrators at owned and managed hospitals to survey three or four nearby hospitals concerning a limited number of hospital charges (Tr. 3712, 3599-600). The purpose of this HAT policy was to make certin that rates were not "out of line" with those of other hospitals because ofa fear that government regulation w01.ld freeze hospital rates at an ahnormally low level (Tr. 1060. 1062) Initial Decisio 106 F. (CX' s 179- 257 271 273- , 286-87). Also, on occasion, at a meeting of area administrators rate setting would be discussed (Tr. 577). 228. Hospital administrators also conducted rate surveys as a means of determining the effciency of the hospital, a process which Dr. Harris called yards ticking (Tr. 3893, 4148). For instance, Memorial Hospital conducted a rate survey that included Catholic hospitals in Nashvile and Memphis to determine how other Catholic hospitals were doing (RX's 244-45; Tr. 1658). A rate survey perform d by Erlanger Medical Center included large hospitals comparable to Erlanger located in other metropolitan areas in Tennessee to compare operating effciency (Tr. 196, 204; CX's 316H- , 317E). 229. While rate surveys were conducted in the Chattanooga area it does not appear that they were a necessary part of the procedure for setting rates because hospital rates are normally determined during the hospital's budgetary process on the basis of total projected costs (Tr. 1645-6, 370, 820-21 , 3599, 3301), which can differ substantially (Tr. 3879- , 1646). Total costs are allocated over the range of hospital services on a selective basis; rather than increasing each service by the same amount or factor, a process is used that requires subjective judgment (Tr. 1542, 1645, 372, 3621 , 631). 230. Chattanooga area hospitals typically adjust their charges once a year, normally at the beginning of their fiscal years (Tr. 1645, 1653 370, 629- , 3301), and since hospitals in the area use different fiscal years, price changes do not occur at the same time (Tr. 370-71, 629- 3301; RX's 270 (2), 638 (1)).

231. The budget cycle for the Chattanooga-Hamilon County Hospital Authority begins in November with preparation of budget data (Tr. 195). Pursuant to state law, the budget is completed and submitted to the Hospital Authority board of trustees for adoption by April 15 (Tr. 371). The medical center and hospital authority operate on a fiscal year ending June 30 (CX 408B). (50) 232. The budget process at Erlanger Medical Center includes determining the expense side of the budget and then determining what rate adjustments, if any, wil be required to cover expenses (Tr. 371). Rate increases when made are not uniform across all departments and services (Tr. 372). Rate changes are put into effect annually at the beginning of the fiscal year (Tr. 370).

233. The charge structure for Memorial Hospital is based on the hospital' s own costs, not on what other hospitals are charging for their services (Tr. 1646). Rate setting is part of the annual budget process at Memorial Hospital. Expenses and revenues are projected and expenses are reduced where possible; rate increases are then made selectively on an item-by-item basis as necessary to cover expenditures (Tr. I542-43, 1645-46). Rate changes become effective an- 361 Initial Decision nually at the beginning ofthe fiscal Year (Tr. 1653) whichends August 31 (Tr. 1542).

234. Parkridge Hospital' s budget process begins with an examination ofthe costs of providing the equipment, personnel, materials and services that are needed to support its operations for the coming year and the hospital determines whether rate adjustments are needed to support the expense side ofthe budget. Ifthe need for rate increases is indicated, then a rate request is prepared and submitted to Blue Cross for review and approval. The fiscal year for Parkridge Hospital is the calendar year, and rate or charge increases are implemented annually on the first day of the fiscal year (Tr. 3301-03). 'Rate increases for Parkridge Hospital are not uniformly applied across all services because of different utilization patterns and the impact of the various reimbursement programs, Individual rate adjustments are determined by an item-by-item review. The earnings projection for Parkridge differs from the earnings target for other HCA hospitals. The differences depend upon hospital mix, age of the facility, range of services, and categories ofpayors (Tr. 3303). Rate comparisons with other hospitals are not used in the Parkridge Hospital budgeting and rate setting activity. Detailed comparative data is not available, and would not be useful. Such comparisons would likely be misleading (Tr. 3306-7 3599, 1636-37, 1646, 1660).

235. The fiscal year for Tri-County is October 1 through September , and rate or charge increases are implemented annually on the first day of the fiscal year (RX 638 (1)).

236. Hospital rate-setting is part ofthe annual budgeting processing at East Ridge. Pricing at the hospital is determined by adjustments in line with the total patient costs per day and by such other factors as the consumer price index, area economic forecasts and budget targets. East Ridge does not use comparative price information from other hospitals and does not have access to such information (Tr. 728). Prices are revised on an annual basis in connection with the hospital's budgeting process (RX 84 (2)). The fiscal year for East Ridge begins September 1 (CX 654B). (51) 237. Diagnostic Center Hospital personnel do not use rate comparisons with other hospitals in developing rates for Diagnostic. Such comparisons have never been used for this purpose because there is no comparability represented by individual hospital prices (Tr. 3599). Rates at Diagnostic Center are put into effect annually on November , assuming Blue Cross approval (Tr. 3598). Rate surveys conducted by Diagnostic Center personnel in late 1980 and early 1981 were required by Hospital Affliates (Tr. 3599). This practice was discontinued in early 1981 because it was no longer required by HAI and was not useful to the hospital (Tr. 3600). When developing hospital rates Initial Decision 106 F. and charges on an annual basis, Diagnostic Hospital personnel review the rates and charges on an item-by-item basis in order to determine whether to make a proposed increase for each item (Tr. 3621). 238. Red Bank Community Hospital reviews and revises its rates on an annual basis (RX 308). The hospital's fiscal year begins on April 1 (CX 24C). Medical Park implemented rate increases once a year at the beginning of the fiscal year (Tr. 629). The fiscal year for Medical Park began on January 1 (CX 75E). New rates for Downtown General Hospital' are implemented annually on July 1, the start of the fiscal year for the hospital (Tr. 3674).

239. Chattanooga hospitals must secure the approval of Blue Cross which reviews proposed new rates on a selective, item-by-item basis before putting rate changes into effect (Tr. 1543, 1141 565 3598 3304 3405-06).

240. Although it is diffcult to compare the prices of different hospitals because they may have different names for different services, or because one hospital may include a professional charge as part of a charge for services while another may not, Dr. Harris undertook a comparison of hospital charges in the Chattanooga urban area for representative, high volume services (Tr. 3898-99; RX 1083). According to this analysis, different hospitals charged rates for similar services that were 50 to 200% higher than rates charged by other hospitals (RX 1083).

4. Attempts To Control Hospital Costs 241. Large increases in health care costs over the past several years have prompted purchasers of these services to seek ways to control such costs (Tr. 2348). After the Medicare and Medicaid programs went into effect in the mid-1960' , health care costs began to rise, and over a period of seven or eight years grew to become a much larger proportion of employers' total costs (Tr. 2342). Starting in about 1976 , the inflation rate for (52) health care costs became substantially greater than the general rate of inflation (CX 534E), and the growth rate is increasing, The average growth rate for health care expenditures was 13,9% from 1976 to 1981; for 1980 and 1981, however, it had grown to over 15% (CX 534A). The share of the Gross National Product accounted for by health care rose from 8.9% in 1979 to 9,8% in 1981 (CX' s 534A, 582, p. 41), and in 1982, it reached 10.5% (Tr, 3815). Hospital inpatient services comprise nearly half of health care expenditures (CX's 530C, 534G). Even though use of hospital services per person remained stable and average length of stay decreased, total expenditures on hospital care almost doubled from 1972 to 1977 (CX 530C). From 1980 to 1981 alone, there was a 17.5% increase in expenditures for hospital care (CX 534GJ.

361 Initial Decision 242. (" ' J (CX 209Z-8), and purchasers of health services have indeed become alarmed. Employers have realized that health care benefits provided to employees are a major element of their costs and have put pressure on insurance companies to come up with solutions. Employees have become concerned as well, because their premiums have been increasing (Tr. 2341-43). Employers in the Chattanooga area are experiencing these same cost increases and they have been forced to make significant changes in their employee health benefit plans (Tr. 976-77 1232- , 1806, 1809).

243. Provident Insurance Company s model insurance plqn, which it believes will give employees incentives to seek less costly health care (Tr. 851), is a good example of the type of plan that is becoming more prevalent. It calls for 20% coinsurance12 and a $100 to $150 deductible (Tr. 852).1 Other Chattanooga area employers are making similar changes in their employee health plans. For example, Siskin Steel Company changed from a plan with basic benefits for hospitalization with no deductible or coinsurance to a plan with a $100 deductible and 20% coinsurance for hospitalization (Tr. 1232, 1243). American Manufacturing Company went from a similar basic benefits plan (Tr. 1804) to a plan with a $200 deductible and a 20% coinsurance payment (Tr. 1802; see also Tr. 970, 972 (change to plan with $300 deductible and 20% coinsurance)). These changes are intended to encourage employees to become more cost conscious (Tr. 1234 978; see Tr. 1809), and therefore to be more sensitive to differences in the cost of different health care providers. Blue Cross also offers, in addition to basic coverage, a (53J comprehensive plan with a deductible ranging from $100 to $250 and a 20% coinsurance amount (Tr. 1295-96). This type of coverage has been available for the past three or four years, and Blue Cross now sells more ofthese plans than the traditional basic coverage (Tr. 1297).

244. In addition to encouraging cost consciousness through new health insurance plans, Chattanooga area employers have been counseling their employees to seek health care from low cost providers. In response to its encouragement (CX's 601, 605A-B), various companies have urged their employees to utilize Memorial because of its low costs (Tr. 1533). For example, Bristol Steel Corporation advises employees to go to Memorial, based on information about hospital prices obtained from insurance companies and hospitals (Tr. 1915) and its employees are encouraged to be cost conscious, and to let their physicians know that they are concerned about costs (Tr. 1914). Siskin Steel educates its employees on what they can do to save money (Tr. 1234) 12 Coin8uranc refers to the percentofthc hospital bil paid by the employee, rather than the insurer (T. 849-0). 13 A deductible is a dollar amount up front for which the insured is responsible before any benefits accrue (Th. 849).

(( Initial Decision 106 F. and tries to steer them to Memorial because it is "less expensive and as good" (Tr. 1236).

245. Recently, Chattanooga area hospitals have begun to make some efforts to compete on the basis of price. Memorial has taken advantage of its lower costs to encourage employers to steer their employees to it (Tr. 177). It has disseminated newsletters to Chattanooga industries encouraging price shopping for health se,:ices (CX 601) and preadmission counseling of employees (CX 605A-B). Memorial has urged employers "to identify the hospital which is giving your employees the best service for the least dollar amount (anda to take a look at the price differential among hospitals. . . ." (CX 601). 246. East Ridge has implemented an "acuity pricing" system for its emergency room that it advertises to the public (Tr. 703--4, 574). Under that system there are different levels of charges for different degrees of ilness or injury treated at East Ridge s emergency room (Tr. 703-4, 626). East Ridge s "Stork Club" publications urge prospective parents to inquire about hospital accommodations and rates (CX 255D) and explain the pricing system at the hospital (CX 265). ) (CX 38Z-96). ("') (CX's 148Z-75, 235Z-71). Sequatchie s administrater testified that it has provided discounts to its patients by waiving deductibles and coinsurance (Tr. 574) and has advertised these terms to the community (Tr. 629). Even Erlanger, which is in general unable to price compete (Tr. 176), has lowered its rates in response to competition (Tr. 178, 268).

247. HCA has acknowledged the increasing concern about runaway hospital costs. Its executive vice president has predicted a more price competitive environment for hospitals (54) because of increased pressure from private industry (CX 421Z-2; see also CX' s 100H, 111Z-16) and a 1981 HCA strategy document stated that ("'J (CX 100Z-27). J (CX 209Z-), ("' ) (CX 209H; see also CX 108Z-1), ("' J (CX 101Z-D). Another HCA document reflects the belief that (" ' J (CX 221Z-2; see also CX' s 357A- , 209Z-17, 1l0W, 209Z-18). 248. The growing importance of price competition is also reflected in HCA's policy that its hospitals should attempt to keep their charges at a competitive level. Its 1982 Form 10-K states that "(t)he rates charged by the Company s hospitals are intended to be competitive with those charged by other local hospitals for similar accommodations, supplies and services " (CX 13Q; see also CX 8N), and the 1983 Management Plan for HCA East, the division that includes Chattanooga hospitals, states that (" ' J (CX 1l0K). This strategy is in keeping with the sentiment expressed two years earlier by HCA' Eastern Operations Division, to ("' ) (CX 346I). Indeed, HCA believes it wil be able to ("' ) (CX 209Z-10), ("') (CX 209" see also Q1 Ql\/r f..;"ronH'If-;nft ."" ." "-'''IT\Of-;f-;HQ ", t..."tarnr , 361 Initial Dccision 249. Further pressures on hospital prices may be provided by health maintenance organizations (HMO's) and preferred provider organizations (PPO's) which operate as group purchasers of health services. 250. An HMO is a plan in which a subscriber prepays a fixed fee in return for comprehensive health care. HMO's generally have contractual arrangements with physicians and hospitals to provide care to their members, and its enrollees, studies suggest, may have lower health care expenditures than other patients (Tr. 50-51). According to HCA's President and Chief Executive Offcer (a) successful HMO wil help make other providers in a given market more responsive to consumers, as well as more cost conscious" (CX 123E). 251. A PPO is an arrangement whereby health care providers contract to provide services at a discount to volume purchasers of health care such as employers or other third-party payers (55) (CX 616I; Tr. 3853). There is generally a financial incentive for the group members or patients to use that provider (CX 309A-B). An example of such an arrangement is the PPO recently created by HCA in Florida, on a discounted fee-for-service basis (CX 616!). The PPO, which will be marketed to volume purchasers of health care services, is "designed to introduce a new competitive element into the comprehensive health care market" (CX 616J).

J. Barriers To En try 1. Certificate Of Need And Related Regulatory Programs 252. Both Tennessee and Georgia have statutes requiring certificates of need ("CON") for, among other things, the establishment of acute care hospitals, the expansion of their bed capacity, significant hospital capital expenditures and changes in the services they offer and similar actions in connection with non-bospital health facilities (Tennessee Health Planning & Resources Development Act of 1979 Tenn. Code Ann. 68-11-101 to - 107 (1983 & Supp. 1983), as amended by Act to Amend Tennessee Health Planning & Resources Development Act, Pub. Ch. No. 814 (May 25, 1984) (hereinafter "THPRDA Amendments of 1984"); Ga. Code Ann. 31-6-1 to -50 (Supp. 1983)). The stated purposes of these statutes include avoiding unnecessary duplication of health services and promoting their orderly and economical development (Tenn. Code Ann. 68-11-106(a) (Supp. 1983); Ga. Code Ann. 31-6I (Supp. 1983).

253. Acute care hospitals, psychiatric hospitals, nursing homes, ambulatory surgical centers and other health care providers are subject to CON regulation in Tennessee and Georgia (Tenn. Code Ann. 68-11-103(5), - 106(g)(1) (Supp. 1983), as amended by THPRDA Amendments of 1984; Tenn. Admin. Compo ch. 0720-2- 01; Ga. Code &:

Initial Decision 106 F. Ann. 31-6-2, (8), (12), -40(b) (Supp. 1983). A CON is needed for any project of a provider which involves:

(a) establishment, construction or relocation of the health facilty; (b) any increase or decrease in bed capacity; (c) any conversion of bed capacity from long-term to acute care, orvice versa; (d) the initiation or discontinuance of certain "health services" at a health facility, including, among other things, medical/(56) surgical, obstetrical, and psychiatric services; (e) acquisitions of medical equipment costing more than $500 000 in Tennessee, or more than $400 000 in Georgia; or (I) other capital expenditures in excess of $500 000 in Tennessee, or $600 000 in Georgia.

(Tenn. Code Ann. 66-11-103 (9), (11), - 106(g)(1) (Supp. 1983), amended by THPRDA Amendments of 1984; Tenn. Admin. Compo ch. 0720-2- 02; Ga. Code Ann. 31-62 (5), (14), -40(b) (Supp. 1983); Ga. Admin. Compo ch. 272- 01 (8), (11, (17)). 254. In Tennessee, a CON may be granted for a project only if it is necessary to provide required health care in the area to be served . . . and will contribute to the orderly development of adequate and efiective health care facilities and/or services" (Tenn. Code Ann. 66- 11- 106(h) (2), as amended by THPRDA Amendments of 1984) and a CON may be granted in Georgia only if the project is "needed" (Ga. Code Ann. 31-60(a) (Supp. 1983)).

255. The procedures for obtaining a CON in Tennessee (as well as the similar procedures in Georgia)!' involve several stages. The process for determining whether a CON is granted for a project begins with the initiation of the Hreview cycle " which commences after a CON application is submitted to the Tennessee Health Facilities Commission ("THFC"), and the THFC staff determines the application to be complete (Tenn. Admin. Compo ch. 0720-2- 03(5)). A public hearing is then held in the area where the proposed project is to be located if one is requested by an "interested party" (Tenn. Code Ann. 66-11- 106(i)(2), as amended by THPRDA Amendments of I984). The CON application is reviewed by the Tennessee state health planning and development agency ("SHPDA") (currently the Department of Health and Environment (Tenn. Code Ann. 66-11-104(a) (Supp. 1983)); it then reports the results of its review to the THFC (Tenn. Admin. Camp. ch. 0720-2- 03 (6), (8), (9)). After the public hearing and the SHPDA' s report on the application, the THFC makes its initial deci- J4 The Georgia CON process is fundamentally similar to that ofTennessce, except for minor djfferences in the time periods allowed for different stages afthe process, and the power of Georgia s health planning agency to make the initial decision on a CON application (subject to administrative appeal to an independent state(Seeboard)Ga. ('n .1n A 'I, A'I." AA IO:,, n 1GII'1\. l"n !o.. J. ...,') " IH\ 361 Initial Decision sion on the CON application. The-THFC'-sinitial decision may (57J be appealed by the applicant, the local Health Systems Agency (where there is one), certain persons who have previously participated in the proceeding, and (upon showing of good cause) any other person (Tenn. Code Ann. 68-11-106(1) (1) (Supp. 1983)). Upon such an appeal, the THFC holds a public hearing, which is to be commenced within 45 days (Tenn. Code Ann. 68-11-106(1) (2) (Supp. 1983)). Within 45 days of the hearing, the THFC decides the appeal (Tenn. Code Ann. 68-11- 106(1) (3)). The THFC' s decision is in turn subject to judicial review at the instance of any "aggrieved" person (fd. Tenn. Code Ann. 4-5- 322(a) (1983)).

256. Georgia not only regulates health facilities under its CON program, but also indirectly regulates them through another review of health facility capital expenditures, pursuant to an agreement with the Federal Government under Section 1122 of the Social Security Act (42 U. C.A 1320a-1 (West 1982); 42 C. R. 100. 101- 109 (1983); Ga. Code Ann. 31-6-50 (Supp. 1983); Ga. Admin. Compo ch. 272to - 03;HX 778 (46)).

2. HCA's Views On The Impact Of CON Programs 257. In an interview published in the Harvard Business Review 1981, HCA's President, Thomas J. Frist, Jr., stated: Federal and state health planning laws have erected formidable barriers to entry into the hospital industry by creating literal monopolies for physicians and hospitals. If the health planning laws state that a community can have only one cardiac surgery program, they might as well give the physician who performs that surgery an exclusive franchise. It' s the same for hospitals.

(CX 123D). Dr. Frist also said that these barriers to entry benefit HCA because they "protect our hospitals from competitors who might build new facilities and take away our market. We know what the market for a particular institution is going to be like five or ten years down the road" (CX 123C), and he claimed that "regulation severely restricts new hospitals from entering our markets" (CX 123F). 258. Other HCA statements and actions reflect its belief as to the restrictive effect of CON regulation on entry and expansion. In its 1982 Form lO- , fied with the Securities and Exchange Commission HCA states that "state certificate of need laws and Public Law 92-603 place limitations on the Company s and its competitors' ability to build new hospitals and to expand (58) existing hospitals. . . ." (CX 13S). HCA' s 1982 Management Plan for its Mergers and Acquisitions Division observes that (" ') (CX 125D) and its 1980 Corporate Strategy Statement noted that ("'J (CX 221Z-8). David G. Wiliamson, Jr. Executive Vice President of Domestic Development ofHCA (CX 6Z- 29), in a paper prepared for presentation at a 1981 conference, re- , Initial Decision 106 F. ferred to the "franchise value" and "franchise type protection" of hospitals that is created by certificate of need regulation (CX 124H J). This franchise value was quantified on one occasion by Wiliam G. White, HCA Senior Vice President of Acquisitions (CX 6Z-29), when he attached a value of$8 million, or $20 000 per bed, to the certificates of need for a total of 400 psychiatric hospital beds held by HCC (which HCA was seeking to acquire at the time) (CX 136A-B). A dollar value for a CON was also designated when HCA offered to purchase the bed complement of Medical Park Hospital-without the land, building or its contents-for about $400 000 in late 1976 or early 1977 so that HCA could add those beds to Parkridge s bed complement (Tr. 613). And Jonathan Grimes, administrator of HCA's North Park Hospital urged HCA to consider acquiring Red Bank Community Hospital in order to acquire the right to operate beds that could be transferred to North Park (CX 420A).

3. CON Experience In Chattanooga 259. Peaches G. Blank, former executive director of the Tennessee Health Facilities Commission (Tr. 2066), and a frequent attendee at THFC meetings after her departure from the Commission (Tr. 2086), believes it is "very diffcult to get the Commission to allow new beds to be constructed" (Tr. 2085), and that it is even more diffcult (and rarely possible) to persuade the THFC to grant a CON for a new hospital (rather than for new beds for an existing hospital (Tr. 2086). Other witnesses familiar with the Chattanooga area testified that it is diffcult to obtain CON approval for additions of new bed capacity (Tr. 482, 3429, 3431- , 3436). On the other hand, HCA employees denied that the CON process imposes any significant burden on a hospital seeking to expand its services or facilities (Tr. 3308, 3258 3482-83). Complaint counsel argue that this testimony is not credible since it is at odds with the statements of Mr. Frist (F. 257), but I believe that this conflict merely reflects the diffculty of deciding to what extent CON requirements are a barrier to eventual entry or expansion, for recent history reveals that hospital expansion has occurred in Chattanooga and expert studies suggest that entry or expansion may not be as diffcult as some industry members believe. 260. There is no doubt that the CON process can cause delay(Tr. 241-43, 3436), but there has been steady hospital entry and (59) expansion in the Chattanooga area in recent years, including the opening of Park ridge Hospital in 1971 (Tr. 3276), the opening of East Ridge Hospital in 1974 (Tr. 681), the construction of Downtown General Hospital as a new replacement facility for the outdated Newell Clinic in 1976 (Tr. I047-49), the construction of Red Bank Community Hos- ;t." l ,.. 1Q77 "''' '" 'tan l",,,c1'an t- fi-.. tho. nh",,,lat-a 'M....",.. T-,.",n;t", l r'T.. 361 Initial Decision 1172-76), and the construction of North Park Hospitalas a replacement for the obsolete Medical Park Hospital in 1982 (CX 504E). 261. A number of hospitals in Chattanooga have also expanded and made substantial improvements in facilities, including Memorial Hospital' s expansion of its facilities and bed complement from 245 beds to 365 beds between 1967 and 1983 (Tr. I503, 1670); Memorial Hospital' s current $12 to $13 milion expansion program (Tr. 1614- 15); Erlanger Medical Center s $90 millon expansion and renovation program (Tr. 129, 323-26; RX 761); East Ridge Hospital's $11 milion renovation and expansion program (Tr. 767-68); Parkridge Hospital's addition of73 beds in 1981 and the subsequent expansion of its intensive care unit and support service facilities (Tr. 3276, 2378); and Diagnostic Center Hospital's addition of 31 beds in 1981 (Tr. 3608). 262. Memorial Hospital has had no CONs denied (Tr. 1674) and Erlanger Medical Center has never been prevented from carrying out any of its plans by CON requirements (Tr. 346). The large majority of CON applications fied in Chattanooga have been approved, including a substantial majority of new inpatient bed applications (RX 1088).5 4. Expert Opinion 263. Dr. Salkever testified that CON regulation posed a "very substantial obstacle" to new entry or expansion of bed capacity in the Chattanooga area (Tr. 2321, 2325-32). Supporting this conclusion, in his opinion, is evidence that hospital administrators in the Chattanooga area had been deterred from even applying for CON approval of additional bed capacity they wanted (Tr. 2327); that there were significant costs and delays involved in the CON process (Tr. 2327); that CON regulation created a "franchise value" for existing hospitals by restricting opportunities to build a hospital or add hospital beds without (60) purchasing existing hospitals or their bed complement (Tr. 2327-28); that HCA offcials recognized that CON regulation is a barrier to acute care hospital entry and expansion (Tr. 2328); that attempts to add acute care hospital bed capacity in the Chattanooga area often failed, or were delayed for several years, because of CON regulation (Tr. 2328-30); and that local health planners considered the Chattanooga area to be "overbedded " and therefore were not inclined to approve additions to bed capacity (Tr. 2330-32). 264. Dr. Salkever s opinion was based on his analysis of approvals and denials, but both he and Dr. Harris agreed that any regulatory process is bound to result in some denials (Tr. 2554), and there is litte doubt that the use of crude approval-denial rates will produce false 15 There have been some denials. A recent application by North Park Hoapit.l to add obstetrical servces was denied, together with a parallel application fied by Bast Ridge Community Hospital. An application by a religious group to build a hospital in CoUegedale in eastern Hamilton County was denied and is on appeal (RX 10BB). Initial Decision 106 F. conclusions (Tr. 2554, 3930-31, 3941) because the motivations for gamesmanship created by the CON application process wil inevitably result in applications for projects that would very possibly not have been carried out even if there were no CON requirement (Tr. 2552- 53). For example, firms may submit inflated proposals, expecting a process of negotiation with the regulatory authorities to result in a scaled-down compromise consistent with what the applicant really wished to do in the first place (Tr. 2553). On other occasions, an application may be fied well in advance of any actual desire to begin work, in the hopes that an initial denial will pave the way for approval of a later reapplication (Tr. 2553). In other cases, a firm may submit an application simply to get its proposal on the record, before it has really decided itself whether the project is worthwhile. An eventual denial or withdrawal of the application may merely coincide with the applicant' s own determination that the project does not make sense (Tr. 2583).

265. Because of the problems associated with the use of approvaldenial evidence, statistical studies comparing hospital construction and investment levels in jurisdictions with and without CON regulations give a more reliable indication of the effect of the CON process on entry and expansion.

266. The only study which has found that CON's were a barrier to hospital bed construction was one conducted by Dr. Salkever and a collaborator. After examining data for 1968 to 1972 which compared hospital construction and investment in the five earliest states to adopt CON laws with comparable activities in other states, this study found a significant negative effect of CON laws on the addition of new hospital beds, which was counterbalanced by a significant positive effect on the amount of money invested per bed (Tr. 2557). That is, according to their data, hospitals added fewer beds, but spent more money on other equipment and facilities. Overall, CON laws in these early states had no effect on total hospital costs (Tr. 2558). 267. Dr. Salkever updated his study by analyzing data on hospital construction and investment from the period 1971 to 1974 (61) and in this case, found that CON laws had no significant effect on new hospital beds (Tr. 2569; RX 1127). A further analysis of the 1971-1974 data also found no significant effect on new hospital beds (Tr. 2570; RX 1128). Other studies have come to the same conclusion. 268. Professor Paul Joskow in his study of CON laws found that any effects of CON regulation were in reality attributable to the simultaneous existence of state rate regulation programs (Tr. 2578-0). Dr. Hellnger found that the only effect of CON regulation was the positive effect on the rate of investment in the first years of such regulation. which he attributed to preemptive investments to take ad- 361 Initial Decision vantage of grand fathering provisions (Tr. 2573). Drs. Sloan and Steinwald in their study of CON regulation found no evidence that CON laws curtailed bed growth, but rather that CON laws had a positive effect on bed growth. In particular, they found that bed growth occurred in anticipation of CON regulation (Tr. 2574). A study of CON regulation by the Policy Analysis Group found that CON laws had no significant effect on bed growth (Tr. 2575-77). A United States Department of Health and Human Services review of CON studies concluded that on average CON programs have not been effective in constraining hospital investment in beds, plant assets and \,ssets per bed (Tr. 2585, 3932).

269. Complaint counsel also argue that capital costs are a barrier to entry (CPF IX 35-37), but this is inconsistent with the reason for CON requirements, that there has been too much investment and duplication in the health care field (Tr. 3481; S. Rep. No. 1285, 93d Cong., 2d Sess. 39 (1974)).

5. Conclusion 270. While Tennessee s and Georgia s CON regulations would seem on first impression, to create barriers to new entry and expansion in the acute care hospital market, scholarly analysis of the actual effect of such regulations in other areas of the country strongly indicates that, if health care authorities believe that a need exists, new entry or expansion will eventually occur (F. 265-68), and the recent history of entry and expansion in the Chattanooga area supports this conclusion (F. 260-2).

271. However, HCA' s emphasis ofthese facts ignores an even more important consideration-that "need" in the acute care hospital market in CON states is not determined wholly by market forces, but by state administrators, and, because they are "interested parties " by hospitals whose market shares would be diluted by new entry and expansion. (62) 272. The result is that while needed entry or expansion wil most likely occur eventually,16 it can be delayed for several years by the CON procedure and by determined opposition using those laws. 273. Before the CON review process can begin, the applicant must complete its application. For major projects, completion usually takes between two and five months (Tr. 242, 735). Additional time may be needed to respond to inquiries of the THFC or Georgia State Health Planning Agency staff should there be some question as to whether the application is complete, or to clarify the application (Tr. 2088-9; see Tr. 242). Even after the application is deemed complete, there may 16 It is doubtful however, whether health planning authoriti!Js in Tennesse and Georgia wil find that any need for additional facilities in the HSA wil exist in the foreseeable future (Tr. 2331-32). ,. .... ....

422 FEDERAL TRADE COMMISSION DECISIQNS Initial Decision 106 F. be a further delay of up to two months before the review cycle begins for the proposed project (Tenn. Code Ann. 68-11- 106(h) (1), as amended by THPRDA Amendments of 1984).

274. Once it has begun, the length of the review cycle, from its beginning to the due date for the initial administrative decision as to whether or not to grant the requested CON, is at least 90 days; the THFC may defer action beyond the 90-day period for up to 40 days if necessary to clarify information concerning the application (Tenn. Code Ann. 68-11- 106(k)(1)(C), as amended by THPRDA Amendments of 1984).

275. The completion ofthe CON process can be further delayed by administrative appeals. These delays are of particular importance in Tennessee, where administrative appeals of THFC decisions have suffered from a backlog of about a year in recent years (Tr. 735-36; RX 898). Decisions on CON applications are also subject to judicial review which may create further delay. For example, review of the March 27 1979 THFC grants of CONs for additional beds to Parkridge Hospital and Diagnostic Hospital was not completed until January 6 1981-more than 21 months later. The delays occasioned by judicial review contributed to a total delay between the submission of CON applications and the affrmance of those CONs upon judicial review of more than three years (RX's 1051 (1)-12), 1057 (1)-12), 1088 (5)). 276. The CON process provides existing hospitals in the Chattanooga area ample opportunity to significantly delay the entry of a new hospital, or the expansion of an existing hospital in that area. Representatives of existing hospitals may (and frequently do) participate in the public hearings held prior to the initial decision on an application in order to express (63) opposition to the application (Tr. 233 826-27 2091-92; ex' s 115, 118, 127, 129-30). 277. This potential for delay was recognized by the board of directors of Red Bank Community Hospital in a meeting held after the Tennessee Health Facilities Commission s initial decision to approve a CON for the relocation of Medical Park Hospital to nearby Hixson (see CX 186A-B). An attorney who represented the board before the THFC in that CON proceeding reported to the board that if it appealed the THFC's decision, there would likely be a delay of about six months before the appeal could be heard, and that while the appeal was pending, Medical Park probably would be unable to obtain financing for the relocation project. Board chairman Arnold Stulce supported an appeal not only because it might succeed in blocking the relocation of Medical Park, but also because it would delay the relocation for several months or more-a delay that he recognized would h.. g4- 1)".. "0"..17 'f1... h",.... 4-1-",- .r +1-.. ......",..1 (r"V 1 Qt:1:' 361 Initial Decision K. The Competitive Effects Of The Acquisitions 1. Market Shares And Concentration 278. There are three ways to measure HCA's share ofthe acute care hospital market in the relevant geographic market, the Chattanooga urban area: licensed bed capacity and inpatient days, preferred by complaint counsel, and net revenues, which HCA claims is the most accurate measure of market presence.

279. Market share based on the number of beds measures the capacity of acute care hospitals in the market (Tr. 2368, 3951) whereas inpatient days disclose the current distribution of business among those hospitals (Tr. 2368). Since actual occupancy in Hamilton County varied over a wide range in 1981 (RX 986 (250)), use of bed capacity does not measure actual utilzation (Tr. 3952). Inpatient days, a measure which was preferred by the Commission s expert in American Medical International, Inc. Docket No. 9158 (FTC Opinion July 2 1984) (104 F. C. at 177) (Tr. 2818), ("AMI") does reveal current market performance, but only with respect to inpatients; it ignores outpatient care, a significant, and growing, segment of the business of acute care hospitals (F. 79).

280. Since outpatient care, which may result in subsequent inpatient treatment, is one of the cluster of services offered by acute care hospitals, and hospitals in Chattanooga do compete in offering this service, I find that net revenue, which combines revenues derived from inpatient and outpatient care, is the single best indicator of market performance (Tr. 3950-51). (64) 281. The following charts summarize the market share of acute care hospitals and the pre- and post-acquisition Herfindahl-Hirschman Index ("HHI") in the relevant geographic market-the Chattanooga urban area-using the preferred basis of measurement, and the alternatives proposed by complaint counse!.17 TOTAL NET REVENUES CHATTANOOGA URBAN AREA Year Ending September 30 1981 Total Net Revenues Hospital Number Percent Erlanger $ 81,720 182 41. Memorial 32,595 614 16. PARKRIDGE (HCA) 369 204 12. Tri-County 795, 127 11.1% East Ridge 689 926 11 The chart.s do not add Red Bank or Downtown General shares t.o HCA' s market share because it does not control these hospitals 424 FEDERAL TRADE COMMISSION DECISIQNS Initial Decision 106 F. 6. DIAGNOSTIC (HAl) 763 000 7. Red Bank 359 473 8. MEDICAL PARK (HCe) 706 140 9. Downtown General 548,826 10. Tepper 991,414 1.5%, 11. Wildwood 500 TOTAL $197 145 406 HCA (Parkridge) 12. HAl (Diagnostic) HCA + HAl 15. HCC (Medical Park) 2.4% HCA t- HAl + HCC 18. Herfindahl Index Before Acquisitions 2344 Change from HAl Acquisition Herfindahl Index After HAl Acquisition 2419 Change from HCC Acquisition Herfindahl Index After HAl & HCC Acquisitions 2495(65) APPROVED ACUTE CARE BEDS CHATTANOOGA URBAN AREA As of September 30, 1981 Approved Beds pital Number Percent 1. Erlanger 780 35. 2. Memorial 349 16. 3. PARKRIDGE (HCA) 296 13. 4. Tri-County 237 10. 5. East Ridge 128 5. 6. MEDICAL PARK (HCC) 83 3. 7. DIAGNOSTIC (HAl) 80 3. 8. Downtown General 65 3. 9. Tepper 64 2.

10. Red Bank 57 2. 11. Wildwood 1.8% 39 TOTAL 2178 HCA (Parkridge) 13. HAl (Diagnostic) HCA + HAl 17. HCC (Medical Park) HCA + HAl + HCC 21.

Herfindahllndex Before Acquisitions 1933 Change from HAl Acquisition 100 Herfindahllndex After HAl Acquisition 2032 Change from HCC Acquisition 132 Herfindahllndex After HAl + HCC Acquisitions 2164(66I 361 Initial Decision PATIENT DAY-S CHATTANOOGA URBAN AREA Year ending September 30, 1981 Approved Beds ospital Number Percent 1. Erlanger 201 098 35. 2. Memorial 108 215 19. 3. PARKRIDGE (HCA) 768 13. 4. Tri-County 62,494 11. 5. East Ridge 35,328 6. Downtown General 643 7. DIAGNOSTIC (HAl) 716 8. MEDICAL PARK (HCC) 535 9. Red Bank 15,521 10. Tepper 701 11. Wildwood lc320 TOTAL 564 339 HCA (Parkridge) 13. HAl (Diagnostic) HCA + HAl 16. HCC (Medical Park) HCA + HAl + HCC 19. Herfindahl Index Before Acquisitions 2028 Change from HAl Acquisition Herfindahl Index After HAl Acquisition 2114 Change from HCC Acquisition Herfindahl Index After HAl & HCC Acquisitions 2208 282. Using the preferred measure, HCA's acquisition raised the HHI from 2344 to 2495 (RX 1096 (IJ). Under the Justice Department Merger Guidelines CCH Trade Reg. Rep. No. 655 11(c) (June 14 1984), ("Justice Merger Guidelines ) relating to horizontal mergers where the post-merger HHI is above 1800, a market is considered to be "highly concentrated " and additional concentration (an increase in the HHI of over 50 points) resulting from a merger in that market is a matter of significant competitive concern; this is, however, not the only fact which must be considered before the competitive impact of the mergers can be assessed. Other considerations include the strength and weaknesses of the acute care hospitals in the market, barriers to entry, the probability of interdependent behavior, HCA' potential exercise of market power, and possible defenses. (67) 2. Competitive Position Of Chattanooga Area Hospitals a. Erlanger Medical Center 283. Erlanger is a public hospital and the major tertiary care provider for the Chattanooga area (Tr. 92, 94-95, 110-11, 1374-75, 140B- Initial Decision 106 F. 09). It is required by law to provide care to indigent Hamilton County patients (Tr. 111; CX 408Z-17) and virtually all indigent patients in the county are treated at Erlanger (Tr. 134). Erlanger s tertiary care costs cannot be recovered from patients using these services, and must be shifted to other patients (Tr. 883). 284. Other hospitals in Hamilton County have no obligation to treat indigent patients and routinely transfer them to Erlanger' (Tr. 115; CX 411C-D). Parkridge Hospital, an HCA-owned hospital, stated in its 1982 Management Plan that ("'J (CX 38Z-17). HCA acknowledges that when there are public hospitals in an area, its hospitals do not treat indigents (CX 411C-D; see also CX 123E-F), and it has devised a strategy to deal with public reaction to its position of not treating such patients (CX 412).

285. In 1981, Erlanger provided more than $12.6 milion worth of charity care, which accounted for approximately 12.3% of its gross revenues (CX's 66E, 638K). The other hospitals in Hamilton County together provided only $152 960 worth of charity care (CX's 16E, 30E 36E, 62E, 63E, 67E, 68E, 75E). For example, Parkridge Hospital in 1981 provided $23 237 worth of charity care, which amounted to approximately .07% of its gross revenues (CX 36E). In 1982, Erlanger cost of providing indigent care was approximately $15 milion, but only $3 million worth of funds for indigent care was provided by the county (See Tr. 112-13). This means that $12 milion of the cost of providing indigent care was shifted to paying patients (CX 597H; see CX 53E).

286. Mr. Lamb, its former administrator, estimated that Erlanger rates are $50 higher than they would be if it did not provide indigent care, and more than 100 beds are generally used for such care (Tr. 113-15). Erlanger also treats a disproportionate number of Medicare patients, for which it may not receive full reimbursement (Tr. 883). 287. Erlanger is a teaching hospital, and salaries of residents and support ofthe medical facility cost it about $3 milion a year (Tr. 109). 288. Erlanger s market share includes 30 beds at the Wilie D. Miler Eye Center, a facility located adjacent to the main (68J hospital which is used exclusively for the treatment of eye ailments. No other facility in the HSA provides the care offered at this center. There is virtually no overlap in use between the eye center and Baroness Erlanger Hospital. Furthermore, there is a contract between Erlanger and the Chattanooga Ophthalmological Foundation that provides that only patients receiving eye care can be admitted to the eye center 1'"\"1 f'0\ .LU"". J.. .L.Lr:Ll OJV.LH . OJ.Lr:UU-".LHOJr: 361 Initial Decision b. Tri-Ci5unty Hospital 289. Tri-County Hospital has an obligation to care for indigent residents of Georgia, and treats the majority of such patients in Walk- , Dade, and Catoosa counties. It also treats a sub tantial number of Medicaid patients (Tr. !l6, 685; CX's 597H, 302H, L). Approximately 13% of all admissions to Tri-County consist of Medicaid, free care, and bad debt patients (CX 302L; see CX 326Z-8). In 1981, approximately 8.4%, or more than $2.2 milion worth, ofTri-County s gross revenues was attributable to Medicaid and charity (CX 106K). In . I981, Tri- County Hospital provided approximately $564 424 worth of charity care as compared to Parkridge Hospital which provided approximately $23 237 worth of such care (CX's 36E, 106K). In 1982, Tri-County provided $600,000 worth of charity care and has budgeted $1.8 million in free care for 1983 (CX's 326Z-52, 597H). Parkridge, in 1982, provided only $17 203 worth of charity care (CX 505E). Tri-County also has substantial costs in connection with its nursing program which is operated in conjunction with the University of Tennessee at Chattanooga (see CX 326Z-52). The costs associated with providing care indigent and Medicaid patients and its teaching program result in a cost-shift from non-paying patients to paying patients ofapproximately $70 per day (CX !l8D).

c. Metropolitan Hospital 290. Metropolitan Hospital, formerly Tepper Hospital, is a 64-bed proprietary hospital owned and operated by American Healthcorp, Inc., a hospital management company (Tr. 1045, 1090; CX's 58A, F, 15 p. 142). Tepper was formerly a pediatric hospital owned by a pediatrician whose physician group practices at the hospital. Metropolitan now provides some diabetes services to adults but still focuses primarily on pediatrics (Tr. I090, 454). Approximately 70% of its patients are pediatric (Tr. 1091). However, it does not offer the level of intensive care in pediatrics offered by T.C. Thompson Children Hospital (Tr. 105-6 448). It is the only hospital in Hamilton County not accredited by the Joint Commission on Accreditation of Hospitals JCAH") (CX's 58A, 15, p. 142). The physicians practicing at Metropolitan are primarily pediatricians, with an otolarynogologist and two physicians who specialize in diabetes treatment (Tr. 1091) Because Metropolitan focuses on a very (69) small segment of the consumer market and is not currently equipped to deal with the broad range of medical problems which occur in the adult population, Dr. Salkever testified that its licensed beds and patient days slightly overstate its competitive significance in the general acute care hospital services market (Tr. 2370, 2290).

428 FEDERAL TRADE CQMMISSION DEC1SlONS Initial Decision 106 F. d. North Park Hospital 291. North Park, the successor to Medical Park is a new, full-service hospital (Tr. 1758 3398) whose location in the rapidly growing Hixson area makes it a much stronger competitor in the relevant market than its predecessor, Medical Park (Tr. 139-40; CX's 139Z-8 through Z-9, 29). North Park' s occupancy rate was 85% during its. first six months of operation (CX 119A). Medical Park's occupancy rate ranged only from 40-50% (Tr. 590; CX 15, p. 182). Thus, Dr. Salkever believes that North Park' s present competitive position is somewhat stronger than is revealed in the 1981 market share data (Tr. 2373). e. Diagnostic Center 292. Diagnostic Center does not compete with acute care hospitals in the relevant market for patients needing surgery (F. 33) but the services which it does offer are similar to those provided by Erlanger Memorial and Parkridge (Tr. 1081) 3. Interfirm Behavior a. CON Proceedings 293. Red Bank Community Hospital, located near Hixson, Tennessee, opposed Medical Park' s CON application for relocation to that area (Tr. 588, 1985). An HCC representative attended one of Red Bank' s board meetings and told the hospital that HCC was considering an antitrust lawsuit against Red Bank for its opposition to the CON but that if "you support us we can all work together and we can all share in the pie and everybody wil be happy thereafter" (Tr. 1985-86; CX 288B; see CX 233A-B). Red Bank decided to withdraw its appeal ofthe CON and refrain from challenging it in court in return for assurances that an antitrust action would not be fied and for certain other concessions from HCC, embodied in a formal agreement entered into in May 1981 (Tr. 1987; CX 96A-F). 294. The agreement provided that both parties would refrain from recruiting each other s medical staff or employees (CX 96D; (70) Tr. 1987 , 588; see CX 234B), and, further, that HCC would, for the next three years, support any CON application fied by Red Bank (CX 96C). Red Bank also agreed not to seek a CON for any psychiatric or nursing home facilities (CX 96C-D; Tr. 1987) and to support any HCC CON application related to the new hospital during the same time period (CX 96D).

295. The application of Medical Park to relocate to the Hixson area also generated joint conduct among opponents of the CON application. In response to a request by Mr. Smith, the HAI administrator . , 361 Initial Decision ridge, David Dunlap, initially agreed to oppose the move. He drafted an opposition letter to the Health Systems Agency (Tr. 1981 3326; CX 229A-B), but he decided not to send the letter because, he told Mr. Smith, HCA would probably end up owning North Park Hospital (Tr. 1981).

296. There are other examples of joint efforts regarding CON applications by hospitals in the Chattanooga area. One involved cooperative effort by Erlanger, Diagnostic and Parkridge to obtain approval of CON applications after individual applications by each ofthe firms were denied in 1979,18 The three hospital administrators met with a representative of the local Health Systems Agency and presented a united front as to the need for new beds (CX 238A). Another proceeding, the recent CON application for a new hospital to be built in Collegedale, was cause for more discussions among hospital administrators about possible joint action. The proposal was discussed at a meeting of the Chattanooga Area Hospital District, where various administrators expressed their opposition to the new hospital (Tr. 231-32, 576), as well as at meetings of the health care task force (Tr. 234; CX's 588B, 589B). The administrators decided not to take a joint position on the application, however, for fear of possible FTC action (Tr. 232, 576), and the task force also decided to take no offcial position on the application (Tr. 234; CX 589B). Joint support for or opposition to other CON applications has been discussed among hospital administrators (Tr. 248-9, 575, 1982; see also Tr. 3719). b. Exchanges Of Information 297. Several Chattanooga area hospitals have conducted rate surveys. HAI hospitals conducted semi-annual or quarterly surveys (Tr. 1060-61, I958-59, 3599) of room rates and high volume ancilary department charges (Tr. 1959); the information was gathered by phone calls to administrators and other personnel (71) at other hospitals (Tr. 1958; CX 239A). Non-HAI hospitals have conducted similar surveys (e.g., CX's 276A-E (Parkridge), 316I (Erlanger)). 298. Chattanooga area hospitals have also taken surveys to determine the wages that are being paid and benefits provided by their competitors (Tr. 210; CX's 163A- , 237C, 519A- , 520A-C, 521A- 237M The 522A- , 523A- , 524A- , 525A- , 526A-C; see CX' s 227A information for these surveys was generally obtained by communications between hospital personnel departments (Tr. 210-11 , 2006). Chattanooga area hospital administrators are members of the Chattanooga Area Hospital District, a subdivision of the Tennessee Hospital Association, which meets monthly (Tr. 230-31, 3325). Generally, 18 The applications at issue were Erlanger s request for 14 beds (CX 238A; RX 1088 (4)), Diagnostic's reql.eBt for 31 beds and Parkridge s request for 73 beds (CX 23AA; RX 1088 (5)). Initial Decision 106 F T:C. a representative from each area hospital attends each meeting (Tr. 231). Those meetings are an excellent setting for exchanging competitive information, and in fact they have at times served as a forum for discussions about hospital rates (Tr. 577). Some hospitals in the Chattanooga area have joint medical staff meetings (e. Memorial and Erlanger; Tr. 247), and some administrators testified that they have casually discussed price information over the phone or in informal meetings (Tr. 209, 570-71).

4. HCA's Market Power a. Pricing Behavior 299. The acquisitions give HCA control over three of the 11 hospitals in the relevant geographic market with a share of total net revenues of 18.2% (F. 281). While the increase in the HHI in this market is of concern (see Conclusions of Law), HCA' s share of the market can hardly be said to confer on it "suffcient market power to influence the level of hospital prices in Chattanooga even without collusion with other firms" as argued by complaint counsel (CPF X 88); III Areeda & Turner Antitrust Law n 804 (1978). b. Foreclosure Of Competition For Referral Patients 300. There are several types of patient referrals. "Outpatient" referrals are temporary visits of inpatients at one hospital to another hospital for outpatient testing (Tr. 615-16). In "inpatient" referrals an inpatient at one hospital (or a patient from the emergency room) is transferred for further inpatient treatment at another hospital (Tr. 123). For outpatient referrals, after the patient's physician orders a test, a nurse or other employee of the hospital usually decides which other hospital wil be used (Tr. 616, 747 1387-88 3317, (72) 3343). For inpatient referrals, the physician usually decides which hospital should receive the patient (Tr. 645-46, 761) 301. Outpatient referrals may be influenced to some extent by a hospital administrator instructing employees to send patients to particular hospitals for care (Tr. 616-17, 1967); however, in the final analysis, this decision is made by hospital nursing personnel on the basis of patient convenience, quality, reliabilty, and speed of com pletion of the test (Tr. 3317- 18; see also Tr. 747). 302. According to Dr. Salkever, administrators of hospitals, as demonstrated by research, have been able to influence patient length of stay, even though this has normally been the physician s decision to make, and he infers from this that administrators wil be able to lnfll1pnp.p. nhvsir.-i:m fp.rr:11 oHttprns in r.lm;p r.::SPS whprp. nnp. hmmi- 361 Initial Decision tal is not clearly preferable to OIie anothedTr. 2766-7, 2770 2775, 2777).

303. Up to now, administrators have sought to redirect physician referral patterns in Chattanooga primarily through providing opportunities for social and professional contacts (Tr. 1106 1148 1386), but complaint counsel argue that HCA intended to influence referral patterns after the HAI acquisition and the announcement of the HCC acquisition (see Tr. 596-00). For example, Parkridge and Diagnostic were selected to participate in a "sister hospital" relationship (Tr. 3375, 3613; CX 400B), created by HCA to integrate newly,acquired and new managed hospitals into its system (CX 206C-D). The Parkridge-Diagnostic relationship entailed joint department head meetings (Tr. 3375, 3613).

304. Even beyond the sister relationship between Diagnostic and Parkridge, HCA set up a series of Chattanooga area administrators meetings (Tr. 3321; CX 400A). At the first meeting, referrals were discussed, including the possibility of persuading physicians to utilize HCA hospitals (Tr. 600, 3380; CX 400B). Both Parkridge s open heart surgery plans and its CT scanner were mentioned (Tr. 1979; CX 400B). Several more administrators' meetings were held (Tr. 3321; CX' 400A, 625A-B), but referrals were no longer discussed, on advice of HCA counsel (Tr. 60O-1). HCA also had plans after the acquisitions to undertake an area wide study "to coordinate and maximize HCA' potential growth in the Chattanooga area" (Tr. 602-D3; CX 630). HCA Chattanooga area administrators attended a meeting to discuss the plan (CX 625A-B) and received a questionnaire to fill out to initiate the study (Tr. 610). The administrators were subsequently instructed not to complete the questionnaire, however, because of concerns about the FTC investigation (Tr. 612, 3797). 305. Inpatient referrals from Diagnostic to Memorial and Erlanger dropped off from 1981 to 1983, while referrals to Parkridge increased (CX 898A). Outpatient referrals from (73) Diagnostic followed the same pattern (CX 900A), and it is reasonable to conclude that the merger may have prompted physicians or hospital personnel to refer patients to affliated hospitals.

306. There is no reason, however, to believe that such referrals were done for other than sound medical reasons. HCA hospital administrators denied that they could influence patient referrals (Tr. 3316, 3403- 04), and an HCA memorandum on referrals recognized that physicians would stil determine where the majority of referrals would take place (Tr. 600). While complaint counsel argue that such denials are conclusory" (CPF X 107) I do not believe that HCA administrators would jeopardize their relationship with their medical staffs by insisting that they engage in a practice which might not be in the best Initial Decision 106 Ii. interest of their patients (Tr. 3316). Furthermore, I find no record support for complaint counsel's apparent claim that Chattanooga area medical personnel would subordinate their patients' needs to pressure by HCA administrators.

307. This is especially true in light of the paltry reward to HCA even if such a scheme were successfully adopted, for discharges by HCA hospitals in the relevant market for referral to other hospitals and referrals to HCA hospitals were not at all significant (RX 1086). 5. Expert Opinion 308. Dr. Salkever testified that, in the relevant market proposed by him, the MSA plus Bradley County (CX 646F, U), and using licensed beds and inpatient days as measures of market pressure, the postacquisition HHI is in the highly concentrated range, approximately 200, and that the change in the index from pre- to post-acquisition was "quite substantial " from 500-00 points (Tr. 2364-5). 309. The high, and substantially increased, post-acquisition concentration concerns Dr. Salkever because he believes that there are substantial entry and expansion barriers in the acute care hospital market in Chattanooga, and that this may lead to the independent exercise of market power by HCA hospitals in the market, or to collusive or interdependent behavior by Chattanooga area hospitals with respect to price or non-price competition (Tr. 2364-5, 2388-9). This fear is heightened by reference to past history in the market which, (74) Dr. Salkever believes, reveals a tendency to collusive or interdependent behavior (Tr. 2362).

310. Dr. Harris does not share Dr. Salkever s concern because he believes that post-acquisition concentration level in the Chattanooga MSA is comparable to that of other MSAs in the United States where competition is vigorous. In fact, Dr. Harris believes that most hospital markets in the United States are characterized by intense rivalry and no collusive behavior (Tr. 3905).

311. The degree of concentration in the Chattanooga MSA after the acquisition of HAI and HCC by HCA, as measured by the HHI, is lower than the average for comparable sized Metropolitan Statistical Areas (Tr. 3906; RX 1098). The HHI value for the Chattanooga Metropolitan Statistical Area after the acquisition of HAI and HCC by HCA was 1985, while the average HHI value for all MSA's at that time was 2530, more than 25% higher than the level for Chattanooga. Of the 70 metropolitan areas studied, 5 r almost 80% of the total had a HHI level of market concentration above 2 000, which is higher than the Chattanooga MSA's HHI (RX 1098). 19 What a physician, as opposed to an HCA administrator, would gain by referring a patient to an inadequate medical facility-aB complaint counsel fear-is diffcult to imagine. 361 Initial Dccision L. Defenses 1. Extensive Regulation 312. HCA cites as a "possible basis for immunity," the extensive government regulation of health care providers (RPF 642-02), but the Commission s recent decision in AMI at 12-17 so decisively rejected AMI's similar claim, that HCA's proposed findings on this point need not be considered.

2. Possible Benefits Of The Acquisitions 313. In the opinion of Mr. David G. Williamson, Jr., HCA's executive vice president, HCA has several strengths: Because of its widespread operations, it can attract highly qualified administrators who seek career advancement (Tr. 3249), as well as high quality personnel who enable HCA to develop system-wide approaches to hospital care which are unavailable to unaffliated hospitals. HCA also enjoys superior access to capital (Tr. 3247-48), has had extensive experience in designing and building hospitals, and its size allows it to realize purchasing economies (Tr. 3247-48). Furthermore, hospital administrators can look to in-house specialists for assistance in dealing with problems facing them (Tr. 3247-48, 3418-20). (75) 314. These advantages allowed HCA to make substantial improvements to HCC's original plans for North Park Hospital (Tr. 3414-17) and, in the opinion of HCA personnel, have made both North Park and Diagnostic better competitors (Tr. 3418, 3602; see also Tr. 141). 315. The opinions expressed by HCA personnel, while entitled to some weight, are subjective and it is impossible to determine whether or not HCA's structure and personnel confer any more advantage on hospitals it now owns than did HAI or HCC on those hospitals. For example, HCA administrator Isaac Coe testified only that HCA provided management specialists to Diagnostic who were of "a little better" quality than those provided by HAI (Trl 3617-18). 316. Furthermore, HAI, as does HCA, had its own in-house consultants (Tr. 1088-89; see also CX' s 825A-J, 826A- , 827 A-J), and HCA resorts, or can resort to outside consultants (and presumably so can other acute care hospitals) when necessary (Tr. 3436, 3321-22; CX 347B-C).

317. Since there are almost 70 group purchasing organizations in the United States (RX 778 (38) - (44)), which are available to area hospitals, including the Tennessee Hospital Association group purchasing program (Tr. 252- , 1566), there is no reason to believe that HCA is able to obtain any better prices from its suppliers than are available to individual hospitals which are members of such group buying organizations.

Initial Decision 106 F. 318. Whether or not HCA' s designers produced a better North Park Hospital than HCC would have is a matter of dispute. While Jonathan Grimes, the present administrator of North Park testified that HCA had improved its heating and air conditioning systems and the traffc flow patterns of hospital departments inter alia over what HCC had planned for the new facility (Tr. 3413 I 7), Marvin Stern, Mr. Grimes predecessor, believed that, while HCA did upgrade the air conditioning, this improvement reduced the space for ancillary services (Tr. 641-42). In any case, numerous independent firms that specialize in hospital design or construction are available to area hospitals (Tr. 3380-81). (76) III. CONCLUSlONS OF LAW A. Jurisdiction Since complaint counsel's narrowest proposed relevant geographic market included areas of both Tennessee and Georgia, HCA, prior to the hearings, withdrew its third defense claiming that the Commission was without jurisdiction because of the absence of interstate commerce (Order Amending HCA's Answer (Nov. 8, 1983). Nevertheless, a short discussion of jurisdiction is warranted (CB, p. 5 , n. 1). While HCA, HAI and HCC were headquartered in Tennessee, all of these companies owned and managed hospitals in other states (F. 67 , 72), and they were therefore doing business in interstate commerce. United States v. American Building Maintenance Industries 422 U.S. 271, 273, 275 (1975). These corporations were also engaged in interstate commerce through the receipt of federal Medicare and private insurance payments, the purchase of supplies and the treatment of patients from out of state (F. 68-69, 71, 73-74). United States v. Hospital Affiliates International, Inc. 1980-1 CCH Trade Cas. n 721 at 77 853 (E.D. La. Oct. 9, 1980).

Furthermore, because of the extensive involvement of HCA' HAl's and HCC' s Chattanooga area hospitals in interstate commerce, the acquisitions have affected and will affect interstate commerce Hospital Building Co. v. Trustees of Rex Hospital 425 U.S. 738, 744 (1976); thus, the jurisdictional requirements of Section 7 of the Clayton Act, 15 U. C. 18 (1982) and Section 5 of the Federal Trade Commission Act, 15 U. C. 45 (1982) are satisfied. B. The Relevant Product Market The "line of commerce 20 in which the impact of a merger or acqui- 2G Section 7 of the Clayton Act outlaws acquisitions whose effects may be substantially to lcs.cn competition or to tend to create a monopoly "in any !ine of commerce.. ." 15 V. C- 18 (1982), as does Section 5 of the FTC Act. AMTHt 17.

.L.L'-OJ.L ....,.... 'o'J.LU. 'J.LH""""UU 361 Initial Decision sition is measured should encompass products or services with "suffcient peculiar characteristics and uses to constitute them products suffciently distinct from all others. . . . United States v. E.I dupont de Nemours Co. 353 U.S. 586, 593-94 (1957). (77) The problem of identifying those products or services which, in this case, possess such unique characteristics that they should be viewed as a I'market" was also a central issue in the Commission s recent decision in a similar case AMI.

The Commission recognized AMI at 18, that the "controlling test" for defining the line of commerce is contained in Brown Shoe Co. v. United States 370 U.s. 294, 325 (1962):

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.

In the very broadest sense, hospitals and other providers of medical services offer one product to the consumer-health care-but that phrase encompasses such a variety of services and prices for those services that it does not define with enough particularity the area within which particular health care providers compete. On the other hand, analyzing the individual services offered by providers does not further the inquiry because these services are often offered in combination.

In such cases, the courts and the Commission have adopted a "cluster of services " as the relevant product market. AMI at 19; Grand Union Co., Docket No. 9121, slip op. at 19 (July 18, 1983) (102 F. at 1044). When a cluster of services has enough peculiar characteristics-for example, commercial banking-this fact disproves the theory that other services might be interchangeable with them. United States v. Philadelphia National Bank 201 F.Supp. 348, 363 (E.D. Pa. 1962), aff'd, 374 U.s. 321 (1963).

In AMI the Commission listed the facts which it believed called for the conclusion that the cluster of services offered by general acute care hospitals was the relevant product market in that case and which set such hospitals apart ! from other health care providers. AMI 20.

The facts are no different in this case; The health care industry recognizes general acute care hospitals as distinct competitive entities (F. 84-89); state statutes require general acute care hospitals to 21 Gompore Philadelphia National Rank 201 F.supp. at 363 It is the conglomeration of all the various services and functions that sets the commercial bank off from other financial institutions.

436 FEDERAL TRADE CO!\IMISSION DECISIONS Initial Decision 106 F. provide the same cluster of services (78) (F. 82-83); and many medical procedures are only available at acute care hospitals (F. 77, 80-81) The record in this case also establishes, as it did in AMI that while other health care providers, either inpatient or outpatient, may offer services which are also offered by acute care hospitals, they are not realistic alternatives for the physician who admits his patients to acute care hospitals and would not constrain the exercise of market power by a combination of acute care hospitals. Inpatient facilities such as nursing homes or providers oflong-term psychiatric care do provide overnight and extended care, but they do not have the same equipment or personnel as acute care hospitals and do not serve the same patient population (F. 78 98-105), and hospital administrators do not view them as competitors of acute care hospitals (F. 86-9). Finally, they could not, unless major renovations were made, begin to offer the same services as do acute care hospitals (F. 101 , 105).

Outpatient providers such as doctor s offces and ambulatory surgery centers often offer the same services as do hospital emergency rooms, but they are offered in a different context. Hospital emergency rooms probably deal with much more gravely il patients than do other surgical facilities (F. 76, 78) and, for that reason, much emergency care in acute care hospitals is continued in the hospital's medical-surgical rooms, an indication that emergency room care is part of the cluster of services offered by hospitals. Thus, the record in this case establishes, with no room for doubt that general acute care hospitals provide a unique cluster of services to the physician and his patients which sets these hospitals apart from other health care providers. See AMI at 21: Although each individual service that comprises the cluster of general acute care hospital services may well have outpatient substitutes, the benefit that accrues to patient and physician is derived from their complementarity. There is no readily available substitute supplier of the benefit that this complementarity confers on patient and physician.

Despite their argument on behalf of the cluster concept, complaint counsel claim that some services should not be included in the relevant product market inpatient psychiatric care offered by acute care hospitals, emergency room care, and the pediatric beds at Erlanger Medical Center. I disagree with them: pediatric and psychiatric care offered by general acute (79) care hospitals are part of the cluster of services which they offer, and the beds dedicated to those services, including the ones at T.C. Thompson could, in addition, be ), r1.vu1. 1.1..or. VV1.\.1. , V1.' reac.1.Hv.o 361 Initial Decision 106-9), 122-24).22 And, while acute care hospital emergency room service may differ from that offered by non-hospital outpatient facilities, hospitals compete with each other in offering this service, and it should not, therefore, be excluded from the acute care hospital mar. ket (F. 90-96), even though the Commission in AMI at 28 apparently concluded that inpatient revenue is the best measure of market presence.

Finally, while it is true, as HCA points out, that the types ofsurgical procedures which can be handled on an outpatient basis by surgicenters are increasing, this suggests only that the cluster of inpatient services offered by acute care hospitals is changing and does not indicate that hospitals are becoming head-to-head competitors with such outpatient providers.

C. The Relevant Geographic Market The relevant geographic market within which the competitive effects ofHCA' s acquisitions should be measured is the "section of the country 23 "where the effect of the merger on competition will be direct and immediate. United States v. Philadelphia National Bank, 374 V. C. 321 , 357 (1963).

The parties agree with authorities which have ruled on the question that acute care hospital markets are regional 2' but disagree as to the extent of the area surrounding Chattanooga to which the purchaser of medical services can practicably turn for health care. Tampa Electric Co. v. Nashville Coal Co. 365 V.S. 320, 327 (1961); AMI at 22-23. Complaint counsel argue that the best geographic market is probably the MSA, but they also propose Hamilton County and the HSA as alternative geographic markets while HCA argues that the Chattanooga urban area-to which complaint counsel pose no serious objection-is the proper region for analysis of (80) competitive interaction in Chattanooga. To further muddy the waters, Dr. Salkev- , complaint counsel' s expert, agreed with neither side and put forward his favorite-the MSA plus Bradley County (F. 143), giving reasons which complaint counsel call "persuasive" (CB , p. 35). The claim by the parties that several geographic markets for acute care hospital services exist in the Chattanooga area ilustrates Dr. Salkever s warning that choosing the correct market is a matter of judgment (F. 144).

In my judgment, the most appropriate geographic market in this Complaint counsel concede that t.his conclusion is true with respect to psychiatric facilties in acute care hospitals (Cn, p. 14 n. 3).

1 Section 7 of the Clayton Act, 15 C. C. 18 (1982) 24 United States u. HospitalAffiliotes In.ternational, Inc" 1980-1 CCH Trade Cas. n 63 721 (KD. La. Oct. 9 1980); Robin.son o. Magouern 521 F.Supp. 842 (W.D. Pa. 1981),affd 688 F.2d 824, 881-82 (3d Cir. ceft. denied, 459 U. 971(1982) , . , . Initial Decision 106 F. case is the one proposed by HCA-the Chattanooga urban area. I base my conclusion on two facts, the Elzinga-Hogarty test as applied to patient flow data and patient referral patterns. The Chattanooga urban area is the smallest area which satisfies the Elzinga-Hogarty test- , both LOFI and LIFO in excess of 75%. Hamilton County does not satisfy this requirement, whereas the other suggested areas-the MSA, the HSA and the MSA plus Bradley County do.

Since the Chattanooga urban area is the smallest area to satisfy the test, it is, presumptively, the one which most accurately encompasses those hospitals to which area physicians and their patients can practicably turn to for health care Tampa Electric Co. v. Nashville Coal Co. 365 U.S. 320, 327 (1961); AMI at 22- , for the other areas add little to the LIFO percentage, or with the exception of the HSA (which adds 10%), only marginally more to the LOFI percentages (F. 182). The Elzinga-Hogarty test is clearly useful in eliminating certain geographic areas from consideration as relevant markets, but it is not at least in this case, an infallble guide to the area which is that market. The most that can be said is that Hamilon County is too narrow an area, and that the MSA, the MSA plus Bradley County and the HSA areas are probably too broad. The conclusion that the remaining area-Hamilton County plus Dade, Walker, and Catoosa counties in Georgia-most accurately defines the area within which Chattanooga acute care hospitals compete is, I believe, confirmed by the perceptions of area hospital administrators and by physician admitting patterns.

Administrators from downtown Chattanooga hospital do not regard outlying hospitals as competitors, and administrators of outlying hospitals do not regard themselves as competing with downtown Chattanooga hospitals (F. 176-80). These opinions are confirmed by HCA' study of admitting practices which reveal that physicians, who determine at which hospital their patients receive care, confine their admissions to hospitals within a limited geographic area, and that in this case, Chattanooga urban area physicians admit patients to hospitals within the area, and (81) seldom admit patients to outlying hospitals in Marion, Sequatchie, Bradley and other counties in the HSA (F. 147-57).

By the same token, physicians in outlying hospitals in these counties admit their patients to a single hospital or hospitals in those counties (as in Bradley County where there are two hospitals), but seldom admit them to Chattanooga urban area hospitals (F. 61-62 158-2).

There are exceptions, of course. Patients from outlying areas who - 1 - - - 11 J 1 J. - r'l 361 Initial Decision tals, but this does not mean thafthose hospitals compete withoutlying hospitals; ifthat were the conclusion, then one would have to view hospitals from the entire Southeast who send gravely il burn victims to Erlanger s burn unit or to other tertiary hospitals in Chattanooga as competitors of those hospitals (F. 163, 165-167). Considering all of the above facts, my judgment is that the most appropriate relevant geographic market in the case is the one proposed by HCA the Chattanooga urban area.

D. The Effects Of The Acquisitions 1. Post-Acquisition Evidence Complaint counsel' s prediction of the probable competitive effects ofthe challenged acquisitions is based, in part, on developments in the health care market which occurred or which became more significant after the challenged acquisitions took place. Pointing to the general rule that only market conditions at the time acquisitions are made should be considered, HCA argues that I should ignore complaint counsel's claim that recent developments indicate that health care is becoming more price competitive (RAB, pp. 24-26). I reject HCA's argument for two reasons: First, assuming that the phenomenon is significant, the claimed increase in price competition in the Chattanooga area can be seen as the result of long-term preacquisition developments. Second, the ban on post-acquisition evidence is not absolute. It may not be considered when the evidence is a result of voluntary actions by the acquiring firm United States v. General Dynamics Corp. 415 U.S. 486, 504-5 (1976); FTC v. Procter & Gamble Co. 386 U.S. 568, 576 (1967), but it is appropriate to consider relevant post-acquisition evidence of industry-wide developments which the acquiring firm could not control. AMI at 42. (82) 2. Managed Hospitals Complaint counsel's argument that competition in the Chattanooga urban area may be adversely affected by the HCA acquisitions is based, in part, on concentration figures which include the market shares of the managed hospitals, Red Bank and Downtown General in their computation (CB, pp. 46-51) Complaint counsel concede that Section 7 acquisitions usually transfer ownership to the acquiring company, but they claim that HCA' s management of Red Bank and Downtown General resulted, in the words of the court in United States v. Columbia Pictures, 189 Supp. 153, 182 (S. Y. 1960), in a "transfer of a suffcient part of the bundle oflegal rights and privileges from the transferring person . ..

Initial Decision 106 F. to the acquiring person to give the transfer economic significance and the proscribed adverse ' effect.' "

The effect of the management contracts, according to complaint counsel, is equivalent to ownership because HCA obtained the ability to direct the operations of these hospitals (CB, p. 47). I do not agree. Complaint counsel rely, to some extent, on HAl's prior long- term contracts with the hospitals (F. 196-99); these contracts were, howev- , renegotiated before the acquisition of HAI and, under them, the boards of directors control the hospitals' activities. HCA has no present power to control the activities of Red Bank, for termination of the management agreement was initiated by its board when it learned that HCA planned to acquire HCC. After some initial reluctance, HCA agreed to terminate the management contract (F. 213-14).

The present contract with Downtown General is similar to other HCA contracts for the management of hospitals. Its term is for four years and calls for a fixed management fee, so that HCA has no interest--ther than to satisfy the board-in increasing revenues. Furthermore, Blue Cross had determined that (83) under the new contracts, HAI and Downtown General were not "related persons" (F. 198-99, 201)25 While the HCA administrator at Downtown General conducts its day-to-day activities (F. 202), its management contract and testimony at the hearings reveal that the board establishes the hospital's financial and health care objectives and expects management to carry them out (F. 202-08), and would not hesitate to dismiss an administrator who ignored its policies and, instead, attempted to further the objectives of HCA. In short, complaint counsel have not established that HCA dominates or controls the board of directors of Downtown General.

Complaint counsel suggest that even though the board of directors of the managed hospitals actually control them, they might be unaware of collusive agreements by administrators of HCA-owned and HCA-managed hospitals (CB, p. 51), but speculation of this sort, unsupported by record evidence, does not justify treating HCA managed hospitals the same as owned hospitals for purposes of analyzing the competitive impact of the challenged acquisitions. 1. Medicare regulations define an entity as "related to ltheJ provider" if the provider "to a significant extent is associated or affliated with or . is controlled by the organization furnishing the services. " CODtrol is defined as "the power, directly or indirectly, significantly to influence or direct the actions or policies of an organization. 42 C. R. 405.427(b)(1), (3) (1980).

.Ii Beyond making the claim, complaintcuunscl do not reveal how IlCA' s ownership of the land under Downtown General and the adjacent physicians' offce building give it such leverage that it could dictate policy to the board ( 361 Initial Decision 3. Concentration - HCA' s acquisitions in the relevant geographic market eliminated two competitors, Diagnostic and Medical Park and, using the measure which I believe most accurately reflects competitive activity in the market, total net revenues, they raised the HHI from 2344 to 2495. HCA' s share of total net revenues before the acquisition was 12.9%. The HAI acquisition raised its share by 2.9% to 15.8%. The HCC acquisition raised its share an additional 2.4% to 18.2% (F. 281). (84) HCA claims that these market shares are modest "by any standard" especially those existing in the acute care hospital.industry (RB, p. 58) where, according to Dr. Harris, vigorous competition exists in many MSA's with much higher HHI numbers (F. 310-11). Complaint counsel respond that "even under HCA's market calculations in its preferred geographic area using its preferred measure, total net revenues " the acquisitions are of concern under the Justice Merger Guidelines (CAB, pp. 5-6).

Dr. Harris' comments on the vigor of competition in other MSAs should not, in my opinion, be considered in deciding whether the acquisitions in the Chattanooga urban area are unlawful, for they are conclusions which are not based on facts of record in this case. Furthermore, his conclusion is contrary to the widely accepted theory that high market concentration leads to a lessening of competition; for example, the Justice Merger Guidelines emphasize the importance of increases in concentration caused by horizontal mergers: "Other things being equal, concentration affects the likelihood that one firm, or a small group of firms, could successfully exercise market power 1) As a guide to the prediction of anticompetitive consequences of such mergers, the Department relies on the HHI. The HCA acquisitions resulted in a post-merger HHI of well over 1800 and the increase in the index was over 100 points. According to the Justice Merger Guidelines 11(c):

Markets in this region (a Herfindahl Index over 1800) generally are considered to be highly concentrated. Additional concentration resulting from mergers is a matter of significant competitive concern. The Department is unlikely, however, to challenge mergers producing an increase in the HHI of less than 50 points. The Department is likely to challenge mergers in this region that produce an increase in the HHI of more than 50 points, unless the Department concludes, on the basis oft 1e post-merger HHI the increase in the HHI, and the presence or absence of the factors discussed in Sections , 3. , 3.4, and 3.5 that the merger is not likely substantially to lessen competition. However if the increase in the HHI exceeds 100 and the post-merger HHI substantially exceeds 1800, only in extraordinary cases wil such factors establish that the merger is not likely substantially to lessen competition. Although not as large as the HHI in the relevant market proposed , Initial Decision 106 F. by Dr. Salkever (F. 308), the HHI in the Chattanooga urban area substantially exceeded the 1800 standard, both before (85) and after the acquisitions, and this and the increase in concentration brought about by the acquisitions, is a persuasive indication that they may substantially lessen competition.27 The inquiry into probable competitive consequences cannot, however, be limited to consideration of the HHI numbers, for such qualitative factors as the market' s "structure history, and probable future " must also be considered. Brown Shoe Co. v. United States 370 U.S. 294, 322 n. 38 (1962). Indeed, the Commission insists that such factors are of more importance to merger analysis than market concentration figures:

Both the Justice Guidelines and Commission Statement28 reflect the importance of considering both quantitative and qualitative elements of the acquisition. Although the Commission has expressed an intent to give "considerable weight" to the Justice Guidelines, it has not endorsed either the analytical approach or the numerical thresholds and tests for analyzing mergers contained in the Justice Guidelines. AMI at 27. Although it rejected reliance on the numerical analysis of mergers the Commission in AMI noted that the acquisition under consideration increased the HHI from 3818 to 6025 in the county market and from 4370 to 7775 in the city market and stated that these figures tell a revealing story of the competitive conditions within those markets. AMI at 28. (86) The post-acquisitions HHI, and the increase in the HHI in this case is much smaller, but it is at a level which causes, according to the Justice Merger Guidelines great competitive concern" and this fact together with the qualitative considerations discussed below leads to believe that the HCA acquisitions may29 substantially lessen competition because they wil enhance the ability affirms in the market to collude, either expressly or tacitly. FTC Merger Statement at 2. 4. Health Planning Laws HCA argues that the market share figures present in this casewhich are "the primary index of market power Brown Shoe at 322 n. 38-should be ignored because Congress, by adopting the National Health Planning and Resources Development Act of 1974, has determined that acquisitions in the acute care hospital industry should not 27 AMI at 25; United States u. Phi!rzelphia National Bank 374 U.S. 321 , 363 (1963): Specifically, we think that II merger which prodl1ces a firm controlling an undue percentage share of the relevant market, and results in a significant increase in the concentration of firms in that market, is so inherently likely to lessen competition substantially that it must be enjoined in the absence of evidence clearly showing that the merger is not likely to have such anticompetitive effects. 28 Announcement of Policy: Federal Trade Commi&lion Announces Horizontal Merger-Enforcement Policy, reprinted inCCH Trade Reg. Rep. (No. 546), June 16, 1982 ("FTC Merger Statement" 29 In Brown Shoe at 323, the court stated that "Congress used the words may be substantially to lessen competition (emphasis suppliedL to indicate that its concern was with probabilities, not certinties. . . . . . .

361 Initial Decision be judged "by the singleminded standar:d of their effect or competition" (RB, p. 118), but the Commission has concluded that the health care industry operates like other industries and, for purposes of antitrust analysis, should not be treated any differently: the record clearly demonstrates that price constraints influence the decisions made by both buyers and sellers. Second, AMI's argument that market share evidence is valuable only in cases involving "manufacturing and related industries in which normal market forces can reasonably be assumed to operate" is contrary to both common sense and case law precedent. AMI at 26. 5. Collusion The increase in the HHI caused by HCA's acquisitions, which by themselves suggest the likelihood of collusive behavior, do not reveal the strengths and weaknesses of acute care hospitals in the Chattanooga urban area. A consideration of these factors indicates that HCA' s market share data does not fully measure its competitive strength, for North Park is thought by its competitors to be in an ideal location and its present occupancy rate of around 85% is much higher than that of its predecessor, (87) Medical Park, with the result that the 1981 market data understates North Park's competitive significance (F. 291) Conversely, the market shares of Erlanger and Tri- County somewhat overstate their competitive strength because of their obligation to treat indigent patients (F. 283-89). It might be argued that if competitive weaknesses of acute care non-HCA hospitals in the market are considered, then Diagnostic is at a significant disadvantage, for it has no surgical facilties, but this position is similar to the one complaint counsel have used to justify exclusion oft.C. Thompson s pediatric beds. The fact is that Diagnostic, like T.C. Thompson, is a highly-specialized, well-regarded hospital which is a significant competitor in the services which it does offer (F. 33).

The inference from market data that collusive behavior is more probable after the HCA acquisitions in the Chattanooga urban area is strengthened by consideration of the history of interfirm behavior in this market. FTC Merger Statement at 8. Dr. Salkever testified that hospitals have historically banded together to solve joint problems and that the risk of collusion is therefore high (F. 309), and the Chattanooga area has experienced a similar history. Some area hospital have exchanged price data for their rate surveys, as well as wage data (F. 297-98), and HCC and Red Bank Hospital entered into an anticompetitive market allocation agreement in 1981 (F. 293-94). Hospital administrators have also joined in . . .

Initial Decision 106 F:T. opposing CON applications which they perceive as competitive threats (F. 295-96).

New entry-or the fear of it -can dilute market power; conversely, high entry barriers can "exacerbate any market power conferred by the merger. AMI at 27. In AMI the Commission found that barriers to entry caused by CON requirements were "very high" and that because of excess capacity, which also exists in the Chattanooga area (F. 263), new entry was extremely unlikely. AMI at 29. Complaint counsel make the same argument here, but ifthey mean to suggest that barriers are so high that new entry wil not occur even if there is a need for an increase in capacity, I do not agree with them. The most reliable evidence indicates that if the planning authorities agree that a need exists, new facilities or additions to existing ones wil be approved (F. 265-68). However, it is apparent that CON requirements erect a time barrier to entry, and that it is high (F. 271- 77). Hospitals already in the market can oppose new entry and the entire process of approval could, conceivably, take many years. (88) This means, as a practical matter, that new entry or expansion would not be a significant constraint on the exercise of market power by existing acute care hospitals in the Chattanooga urban area. Collusive behavior by acute care hospitals in the Chattanooga urban area would impair the significant competition for patients and physicians which now exists (F. 222-24). There has been little price competition among acute care hospitals in the past (F. 221) and, despite the great public outcry about the enormous cost of health care I do not share complaint counsel's beliefthat price competition among acute care hospitals wil increase dramatically in the future; however there is some reason to believe that price competition wil be more intense in the future than it has been in the past (F. 241-51). Nevertheless, the attentuated price competition which does exist or which wil exist should be protected;

even assuming that the limited price competition that does exist in these markets may produce only marginal benefits in terms of overall consumer welfare, the antitrust laws wil endeavor to protect this price competition if, for nothing else, the hope that price competition will be enhanced. AMI at 32-33. 505 (2d Cir. 1972), cert. See also Stanley Works v. FTC 469 F.2d 498, denied 412 U.S. 928 (1973).

HCA sees little possibility of collusion among acute care hospitals in the Chattanooga area because of the enormous variety of services which they offer, the complexity of their prices, differences in their costs, the antipathy between non-profit and for-profit hospitals, and the presence of third party payors who must approve hospital charges to subscribers (RB, pp. 77-87).

361 Initial Decision These factors would undoubtedly make it-more diffcult for Chjlttanooga area hospitals to collude, but they could agree on pricing formulas, F. Scherer Industrial Market Structure and Market Performance 170 (2d ed. 1980), and could present Blue Cross with a concerted request for rate increases. In fact, because there is such great public pressure to reduce hospital costs, it would be to their mutual advantage-whether they are for-profit or non-profit ollectively to resist such pressures. Compare Michigan State Medical Society, 101 C. 191, 285-86 (1983). (89) 6. Effciencies The effciencies which the integration of North Park and Diagnostic into HCA wil bring to those hospitals-taking HCA's predictions at face value (F. 314-15)-are, it seems to me, attainable by other area hospitals. Assuming a need for expansion or new entry, financing should not be an insurmountable problem, lower prices can be obtained by affliation with bospital buying groups (F. 317), and management firms can supply the same level of expertise to single hospitals as HCA does to its owned hospitals (F. 316); in fact, hospital trustees can obtain the same benefits as HCA-owned hospitals by hiring HCA as a manager.

Assuming, however, for the moment, that HCA's acquisition of Medical Park and Diagnostic has given them a slight competitive edge over other hospitals, I find it impossible to determine whether this edge outweighs the universally-assumed competitive harm which a significant increase in concentration in a market which is highly concentrated wil probably cause. Indeed, it is arguable that past Supreme Court decisions have actually viewed as repugnant the concept that mergers may be effcient. Brown Shoe Co. v. United States 370 U. S. 294 (1962); United States v. Philadelphia National Bank, 374 U.s. 321 , 371 (1963); FTC v. Procter Gamble Co. 386 U.S. 568 (1978). After considering the "dicta" in these cases AMI at 47, the Commission rejected the concept that effciency should be viewed with suspicion and, along with other recent cases, recognized operating and scale effciencies as one factor to consider in analyzing the competitive impact of a merger. However, the Commission also emphasized that such effciencies must be established by substantial evidence. AMI at 51.

HCA has not presented such evidence. The effciencies which might result from the acquisitions of North Park and Diagnostic are, at best minimal and could conceivably be realized by their competitors. Most significant, however, is HCA' s failure to establish that the predicated effciencies would be of such a magnitude as to "warrant their consid- Initial Decision 106 F. eration as a procompetitive effect and to be balanced against the AMI at 53.anticompetitive impact" of the acquisitions. While HCA may believe that hospitals are better run on a decentralized basis and that the separate identities of North Park and Diagnostic wil be maintained (RB, p. 101), I assume that, if cooperation is more beneficial than confrontation, the managers of these hospitals wil cooperate. (90) 7. Conclusion The argument that HCA's hospital acquisitions should not be judged by their effect on competition was rejected by the Commission in AMI which applied the usual antitrust standards to acquisitions in the acute care hospital market. Those standards require, in this case, an order which divests the hospitals which HCA acquired in the Chattanooga urban area, for the acquisitions of Diagnostic and Medical Park (now North Park) hospitals eliminated two competitors in a highly concentrated" market, and increased the HHI by over 100 points. Consideration of other factors such as the quality of competitors in the market, the high time barrier to new entry or expansion and the past interdependent behavior of Chattanooga area hospitals lead to the conclusion that the acquisitions may contribute to a market structure in which collusive behavior with respect to service or price is a significant risk.

IV. SUMMARY A. The Federal Trade Commission has jurisdiction over the subject matter of this proceeding, and over HCA.

B. HCA was, at all times relevant herein, a corporation engaged in commerce, as "commerce" is defined in the Clayton Act, as amended. C. HAI was, at all times relevant herein, a corporation engaged in commerce, as !Ccommerce" is defined in the Clayton Act, as amended. D. HCC was, at all times relevant herein, a corporation engaged in commerce, as "commerce" is defined in the Clayton Act, as amended. E. The challenged acquisitions are in or affect commerce, as "commerce" is defined in the Federal Trade Commission Act. F. The appropriate product market within which to evaluate the competitive effects of the challenged acquisitions by respondent is inpatient and outpatient acute care hospital services. G. The appropriate geographic market within which to evaluate the competitive effects of the challenged acquisitions by respondent is the Chattanooga urban area, consisting of Hamilton County, Tennessee and Dade, Walker and Catoosa Counties, Georgia. H. The effect of the acauisition ofHAI bv HCA hhs hRRn or mHV hR , 361 Initial Decision substantially to lessen competition, or to-tend to create (91)a mOllJPoly, in the relevant product and geographic markets, in violation of Section 7 of the Clayton Act, as amended, and the acquisition is an unfair method of competition, in violation of Section 5 of the Federal Trade Commission Act, as amended.

1. The effect of the acquisition of HCC by HCA has been or may be substantially to lessen competition, or to tend to create a monopoly, in the relevant product and geographic markets, in violation of Section 7 ofthe Clayton Act, as amended, and the acquisition is an unfair method of competition, in violation of Section 5 of the Federal Trade Commission Act, as amended.

7. The order entered is appropriate and necessary to remedy the violations of law which have been found to exist. v. ORDER An appropriate order in this case must require, as a minimum, the divestiture of those hospitals whose acquisition I have found unlawful, Diagnostic Hospital and Medical Park (North Park). Also, since it is important for hospitals to own or control adjacent offce buildings to attract physicians who wil admit their patients to the nearby hospital, the order should require the divestiture of such facilities so that the hospitf\ls' new owners wil have the benefit of controllng the use of these important adjuncts.

Complaint counsel also urge that the order require HCA to obtain prior Commission approval before acquiring any acute care hospital (through purchase, lease, management contract, or otherwise) in the relevant geographic market (which I have decided is the Chattanooga urban area) and in other areas where horizontal acquisitions may create significant anticompetitive consequences (CE, pp. 111-18). They seek prior notice requirements in other instances (CE, pp. 118- 21) The Commission in AMI accepted a prior notice requirement which would enable it "to investigate an acquisition that appears to involve significant antitrust problems. AMI at 60. Complaint counsel' proposed order provision has the same purpose, and is not such a burdensome requirement that it would adversely affect HCA's business endeavors. It will be included in the order which I enter. The Commission in AMI rejected a prior approval requirement because, even though there appeared to be a trend toward increased consolidation, it could not, on the record before it assume that acquisitions in this industry, per se are anticompetitive." The Commission also found that AMI's presence as a potential purchaser oflocal hospitals has a substantial procompetitive impact, and that a prior approv- Initial Decision 106 F. al requirement might eliminate AMI as a potential competitor. AMI at 59-60. (92) Complaint counsel argue that their proposed order would require prior approval only in these markets where high concentration exists where the acquisition would increase concentration by more than 100 HHI points, where there are high barriers, and where the purchase price is more than one milion dollars (CB, p. 113). While this;proposal is in line with the recent Justice Department guidelines, it would not consider any ofthe qualitative factors which the Commission believes are as important as quantitative ones in analyzing the competitive impact of a merger. One of these qualitative factors is the increased competition which complaint counsel insist is occurring and will continue in the acute care hospital market. A prior approval notification would, therefore, not only impose a substantial burden on HCA 3o but it would also ignore the significant changes which, according to complaint counsel, are now occurring in the acute care hospital market. In addition to deleting the prior approval provision of complaint counsel's order, I have eliminated the words "or management contract" from the proposed order (I.E.; I.F.); changed the definition of the relevant geographic market from the Georgia-Tennessee HSA to the Chattanooga Urban Area (I. ); changed the basis for calculating the Herfindahl-Hirschman Index from state-licensed acute care hospital beds to total net revenue (I. ; V); and, eliminated all requirements with respect to South Pittsburgh Community Hospital and Downtown General Hospital since the former is not located in the relevant geographic market and the latter is not controlled by HCA. I have also eliminated the proposal with respect to Sequatchie General Hospital.

Therefore, the following order is appropriate: (93) ORDER Definitions For purposes of this order the following definitions shall apply: A. HCA means Hospital Corporation of America, a corporation organized under the laws of Tennessee, with its principal executive 30 Compar.. AMI at 60 Notwithstanding Complaint Counsel's arguments and Judge Barnes' findin1gs to the contrary, we believe that time is of the eSRcnce in negotiations for the purchase oflocal hospitals, and that the abilty to make II purchase commjtm nt wit.h same degree of certainty of obtaining the necessary regulatory approval, is an important element in this negotiating proce&!. The prior approval requirement would uniquely disable AMI i!1 theBe negotiations.

HOSPITAL CORP. OF AMERICA 449 361 InitiatDeci offce at One Park Plaza, Nashvile, Tennessee, and its directors offcers, agents, employees, and representatives, and its subsidiaries divisions, affliates, successors, and assigns. B. HAl means Hospital Affliates International, Inc. C. County also includes a county equivalent, such as a parish in Louisiana.

D. Acute care hospital herein referred to as tthospital " means a health facility, other than a federally owned facility, having a duly organized governing body with overall administrative and pfofessional responsibility and an organized professional staff that provides 24-hour inpatient care, as well as outpatient services, and which has as a primary function the provision of inpatient services for medical diagnosis, treatment, and care of physically injured or sick persons with short-term or episodic health problems or infirmities. E. Acquire any hospital means to directly or indirectly acquire all or any part of the stock or assets of any acute care (94) hospital, or enter into any arrangement by which HCA obtains direct or indirect ownership or control of any acute care hospital or any unit of such hospital, including a lease of any such hospital or unit of such hospital.

F. Operate a hospital means to own or lease an acute care hospital. G. MSA and PMSA mean, respectively, a Metropolitan Statistical Area, and a Primary Metropolitan Statistical Area, as defined as of July 1, 1983, by the Offce of Information and Regulatory Affairs, Offce of Management and Budget, Executive Offce of the President. H. The Chattanooga Urban Area means that geographic area comprising Hamilton County, Tennessee and Dade, Walker and Catoosa counties in Georgia.

I. Person means any natural person, partnership, corporation, company, association, trust, joint venture, or other business or legal entity, including any governmental agency.

J. HHI means the Herfindahl-Hirschman Index of market concentration as calculated in accordance with the procedures specified in Section 3 of the Merger Guidelines of the United States Department of Justice (revised June 14, 1984) for acute care hospitals in a defined geographic area, based on each hospital' s total net revenue. II.

A. It is ordered That, within twelve (12) months from the date this order becomes final, HCA shall divest, absolutely and in good faith at no minimum price: (95) (1) North Park Hospital in Hamilton County, Tennessee, and all ..

450 FEDERAL TRADE Commisslon DEC1SlONS Initial Decision 106 F. assets, properties, lands, licenses, leases, and other rights and privileges in connection with the hospital, both tangible and intangible. The divestiture required by this provision of this order specifically shall include any medical offce building owned by HCA that is adjacent to, affliated with, or operated in connection with, North Park Hospital, as well as the plot ofland on which each such medical offce building is situated. The purpose of this divestiture is to .:establish North Park Hospital as a viable competitor, and to restore competition in the area. The divestiture shall be subject to the prior approval of the Federal Trade Commission;

(2) Diagnostic Center Hospital in Hamilton County, Tennessee, and all assets, properties, lands, licenses, leases, and other rights and privileges in connection with the hospital, both tangible and intangible, that HCA acquired from HAI, together with any subsequent improvements in, or additions to, any such assets or properties. The divestiture required by this provision of this order (96) specifically shall include any medical offce building owned by HCA that is adjacent to, affliated with, or operated in connection with, Diagnostic Center Hospital, as well as the plot ofland on which each such medical offce building is situated. The purpose of this divestiture is to reestablish Diagnostic Center Hospital as a viable competitor, and to restore competition in the area. The divestiture shall be to a person other than the person to whom divestiture is made under Section II. paragraph A. (1) of this order, and shall be subject to the prior approval of the Federal Trade Commission.

Pending divestiture, HCA shall take all measures necessary to maintain North Park Hospital and Diagnostic Center Hospital in their present conditions and to prevent any deterioration, except for normal wear and tear, of any of the assets to be divested, so as not to impair the present operating abilties or market value of the hospitals or the other assets to be divested.

A. If HCA has not divested all of the properties, assets, or enterprises required to be divested pursuant to Section II of this order within the 12-month period provided therein, the Federal Trade Commission may select a trustee to effect any ordered divestitures yet to . be accomplished. The trustee shall (97) be a person with experience and expertise in acquisitions and divestitures. If the Federal Trade Commission should elect to appoint a trustee, it shall not be precluded from seeking civil penalties and other relief available to it for any failure by HCA to comply with this order. If the Federal Trade CommlRc:inn hn111n not, plpf't tf' !1nnn lnt '; t.."p+o.o n... +h .. Q",.. Tft.. 361 Initial Decision this order, it shall not be precluded from seeking civil penalties ihe appointment by the courts of a trustee to effect the divestitures, and other relief available to it, for any failure by HCA to comply with this order.

B. Any trustee appointed by the Federal Trade Commission pursuant to this Section shall have the following powers, authority, duties, and responsibilities:

1. The trustee shall have the exclusive power and authority to divest any properties, assets, or enterprises required to be divested pursuant to Section II of this order that have not been divested by HCA within the time period for the divestitures provided therein. The trustee shall have twelve (12) months from the date of appointment to accomplish the divestitures, which shall be subject to the prior approval of the Federal Trade Commission. If, however, at the end of the twelve-month period the trustee has submitted a plan (98) of divestiture or believes that divestiture can be achieved within a reasonable time, the divestiture period may be extended by the Federal Trade Commission. In addition, any delays in divestiture caused by HCA shall extend the time for divestiture in accordance with the delay caused.

2. The trustee shall have full and complete access to the personnel books, records and facilities of any property, asset, or enterprise that the trustee has the duty to divest, and HCA shall develop such financial or other information relevant to the properties, assets, or enterprises to be divested as such trustees may reasonably request. HCA shall cooperate with the trustee, and shall take no action to interfere with or impede the trustee s accomplishment of the divestitures. 3. The power and authority of the trustee to divest shall be at the most favorable price and terms available consistent with this order absolute and unconditional obligation to divest at no minimum price and the purposes of the divestitures as stated in Section II of this order. (99) 4. The trustee shall serve, without bond or other security, at the cost and expense of HCA on such reasonable and customary terms and conditions as the Federal Trade Commission may set. The trustee shah have authority to retain, at the cost and expense of HCA, such consultants, attorneys, investment bankers, business brokers, accountants, appraisers, and other representatives and assistants as are reasonably necessary to assist in the divestitures. The trustee shall account for all monies derived from the sale and all expenses incurred. After approval by the Federal Trade Commission of the account of the trustee, including fees for his or her services, all remaining monies shall be paid to HCA and the trustee s power shall Initial Decision 106be terminated. The trustee s compensation shall be based at least in significant part on a commission arrangement contingent on the trustee divesting the trust property.

5. HCA shall indemnify the trustee and hold the trustee harmless against any losses, claims, damages; or liabilities to which the trustee may become subject, arising in any manner out of, or in connection with, (100) the trustee s duties under this order, unless the Federal Trade Commission determines that such losses, claims, damages, or liabilities arose out of the misfeasance, gross negligence, or the wilful or wanton acts or bad faith of the trustee. 6. Promptly upon appointment of the trustee and subject to the approval of the Federal Trade Commission, HCA shall, subject to the Federal Trade Commission s prior approval and consistent with provisions of this Order, execute a trust agreement that transfer to the trustee all rights and powers necessary to permit the trustee to cause the divestitures.

7. If the trustee ceases to act or fails to act diligently, the Federal Trade Commission shall appoint a substitute trustee. 8. The trustee may ask the Federal Trade Commission to issue, and the Federal Trade Commission may issue, such additional orders or directions as may be necessary and appropriate to accomplish the divestitures required under this order. (101) 9. The trustee shall have no obligation or authority to operate or maintain any of the properties, assets, or enterprises required to be divested pursuant to Section II of this order. 10. The trustee shall report in writing to HCA and the Federal Trade Commission every sixty (60) days concerning the trustee s efforts to accomplish divestiture.

IV.

It is further ordered, That, for a period of ten (10) years from the date this order becomes final, HCA shall notify the Federal Trade Commission at least thirty (30) days prior to HCA's acquiring any hospital, as defined in this order, if:

A. The hospital to be acquired is within a Metropolitan Statistical Area ("MSA") or Primary Metropolitan Statistical Area ("PMSA") in which HCA already operates a hospital, or would operate two or more hospitals as a result of two or more simultaneous acquisitions from different persons, and: (1) the hospitals operated or to be operated by HCA after the acquisition(s) would have a combined share oftwenty (20) percent or more of the total net revenues of the state-licensed acute care hospitals in the MSA or PMSA: or (2) the HHT in t.n" MSA tiu;:YlTAL LUltY. U1-" A1VIE.N1L;A 'hJu 361 Initial Decision or PMSA would be between one thousand (1000) and eighteehhundred (1800) after the acquisition(s), and the acquisition(s) would raise the HHI in the MSA or PMSA by more than one hundred (100) points; or (3) the HHI in the MSA or PMSA would be greater than (102) eighteen hundred (1800) after the acquisition(s), and the acquisition(s) would raise the HHI in the MSA or PMSA by more than fifty (50) points; or B. The hospital to be acquired is not within an MSA or PMSA but is in a county in which HCA already operates a hospital, or would operate two or more hospitals as a result of two or more simultaneous acquisitions from different persons, and: (1) the hospitals operated or to be operated by HCA after the acquisition(s) would have a combined share oftwenty (20) percent or more of the total net revenues of the state-licensed acute care hospitals in the county; or (2) the HHI in the county would be between one thousand (1000) and eighteen hundred (1800) after the acquisition(s), and the acquisition(s) would raise the HHI in the county by more than one hundred (100) points; or (3) the HHI in the county would be greater than eighteen hundred (1800) after the acquisition(s), and the acquisition(s) would raise the HHI in the county by more than fifty (50) points; or C. The hospital to be acquired is not within an MSA or PMSA but is within thirty (30) miles of a hospital which HCA already operates in another county, or would operate in another county as a result of two or more simultaneous acquisitions from different persons, and: (1) the hospital to be acquired and any hospital(s) that HCA operates or will operate subsequent to the acquisition(s) have a combined share of twenty (20) percent or more of the total net revenues of the statelicensed acute care hospitals in the area within thirty (30) miles ofthe midpoint between any hospital to be acquired and any other hospital (103) operated or to be operated by HCA; or (2) the HHI in the area within thirty (30) miles of the midpoint between any hospital to be acquired by HCA and any other hospital operated or to be operated by HCA would be between one thousand (1000) and eighteen hundred (1800) ajler the acquisition(s), and the acquisition(s) would raise the HHI in that area by more than one hundred (100) points; or (3) the HHI in the area within thirty (30) miles of the midpoint between any hospital to be acquired by HCA and any other hospital operated or to be operated by HCA would be greater than eighteen hundred (1800) after the acquisition(s), and the acquisition(s) would raise the HHI in that area by more than fifty (50) points.

The notification required of HCA by this Section IV of this order shall not require additional notification by HCA to the Federal Trade Commission of any acquisition for which notification is required to be made, and has been made, pursuant to Section 7 A of the Clayton Act 454 FEDERAL TRADE C:OMMISSION DECISIONS Initial Decision 106 F. 15 U . C. 18a. The notification required of HCA by this Section IV of this order shall include such information and be in such form as is required ofthe acquirer for notification of an acquisition made pursuant to Section 7 A of the Clayton Act and any rules promulgated thereunder. Furthermore, any acquisition subject to this Section IV of this order, involving an arrangement to lease a hospital, shall be fully described in the notification regardless of whether tle acquisition involves acquisition of any stock or assets of a hospital. With respect to any acquisition of a hospital that is subject to this Section IV of this order, (104) provisions and requirements identical to those provisions and requirements of paragraphs (a), (b), and (e) of Section 7 A of the Clayton Act, and any rules promulgated thereunder, that relate to prohibition of an acquisition prior to expiration of the waiting period, waiting period duration, termination of waiting period extension of waiting period, and submission of additional information, shall apply, insofar as they relate to the Federal Trade Commission.

Provided, however that no acquisition of a hospital by purchase shall be subject to this Section IV of this order if the consideration to be paid for the acquisition of the hospital or any rights or interest therein, including assumption by HCA of any liabilities of its present owners, does not exceed one milion dollars ($1 000 000). It is further ordered That HCA shall, within sixty (60) days after the date this order becomes final and every sixty (60) days thereafter until it has fully complied with the provisions of Section II of this order, submit in writing to the Federal Trade Commission a report setting forth in detail the manner and form in which it intends to comply, is complying, and has complied with these provisions. Such compliance reports shall include, in addition to any other information that the staff of the Federal Trade Commission may request, a summary of all contacts and negotiations with potential purchasers ofthe stock, assets, or other rights or interests to be divested under this order, the identity and (105) address of all such potential purchasers, and copies of all written communications to and from such potential purchasers.

HCA shall submit such further written reports as the staff of the Federal Trade Commission may, from time to time, request in writing 361 Opinion VI. .

It is further ordered That HCA, upon written request of the Secretary of the Federal Trade Commission or the Director of the Bureau of Competition of the Federal Trade Commission made to HCA at its principal offce, for the purpose of securing compliance with this order, and for no other purpose, and subject to any legally recognized privilege, shall permit duly authorized representatives ofthe Federal Trade Commission or the Director of the Bureau of Competition: (1) reasonable access during the offce hours of HCA, which may have counsel present, to those books, ledgers, accounts, correspondence, memoranda, reports, and other records and documents in HCA' s possession or control that relate materially and substantially to any matter contained in this order; and (2) an opportunity, subject to the reasonable convenience of HCA to interview offcers or employees of HCA, who may have counsel present, regarding such matters. (106) VII.

It is further ordered That HCA shall notify the Federal Trade Commission at least thirty (30) days prior to any proposed corporate change, such as dissolution, assignment, or sale resulting in the emergence ofa successor corporation, the creation or dissolution of sub sid aries, or any other change in the corporation that may affect compliance with the obligations arising out of this order. OPINION OF THE COMMISSION By CALVANI Commissioner:

1. INTRODUCTION TO THE CASE A. The Acquisitions In August 1981, Respondent Hospital Corporation of America HCA"), the largest proprietary hospital chain in the United States acquired Hospital Affliates International C'HAI") in a stock transaction valued at approximately $650 milion. I.D.F. 1 6.1 At the time of I The fonowing abbreviations are UBed in this opinion: I.D. - InitialDecision page number I.D.F. - Initial Decision Findingof Fact number CX - Complaint CounBel'a Exhibit number RX - Respondent's Exhibit number CAB - Complaint Counsel's AppeaJ Brief RAB - Respondent' s Brief on Appeal from Initial Decision (footnote cont' Opinion 106 F.

the acquisition, HAI owned or leased 57 hospitals and managed 78 hospitals nationwide. LD.F. 6. Prior to its acquisition by HCA, HAI owned or managed five acute care (4) hospitals in the general area of Chattanooga, Tennessee, and HCA acquired ownership or management of these hospitals through the transaction. Some four months later HCA acquired yet another hospital corporation, Health Care Corporation ("HCC"), in a stock transaction valued at approximately $30 millon. LD.F. 8. At the time of the acquisition, HCC' owned a single acute care hospital in Chattanooga. These two transactions provide the genesis for the instant case.

As a result of the HCA-HAI acquisition, Respondent increased its hospital operations in Chattanooga and its suburbs from ownership of one acute care hospital to ownership or management of four of the area s eleven acute care hospitals. Within the six-county Chattanooga Metropolitan Statistical Area ("Chattanooga MSA"), HCA changed its position from owner of one hospital to owner or manager of six of fourteen acute care hospitals. With the acquisition of HCC, HCA obtained yet another acute care hospital in Chattanooga. Thus, HCA became owner or manager of five of the eleven acute care hospitals within the Chattanooga urban area and seven of the fourteen in the Chattanooga MSA.

On July 30 1982, the Commission issued a complaint charging that the effect of HCA's acquisitions of HAI and HCC, both together and separately, may be substantially to lessen competition or to tend to create a monopoly in the acute care hospital services market in the Chattanooga, Tennessee area in (5) violation of Section 7 ofthe Clayton Act, 15 V. C. 18 (1982), and Section 5 of the Federal Trade Commission Act, 15 VB. C. 45 (1982). Judge Parker issued his Initial Decision on October 30, 1984. He found that the acquisitions violated Section 7 of the Clayton Act and Section 5 of the Federal Trade Commission Act, and ordered HCA to divest two of the hospitals of which it had acquired ownership. Judge Parker also ordered that HCA provide prior notification to the Commission of certain of its further hospital acquisitions. HCA appeals the Initial Decision on several grounds; Complaint Counsel appeal certain of Judge Parker findings as well.

Cane - Complaint Counsel's Answer to Respondent' s Appeal Brief Hane - Respondent' s Brief in Response to Complaint Counsel' s Appeal Brief CRE - Complaint Counsel's Reply to Respondent' s Briefiri Response to Complaint Counsel' s Appeal Brief RRB - Respondent' s Reply Brief on Appeal from Initial Decision Transcript citations are given by the last name of the witnc88 and the transcript page on which the testimony appears.

2 Originally, the complaint alleged that the acquisition may substantially lessn competition or tend to create a monopoly in the inpatient psychiatric treatment services market as well CampI. para- 21-28. UPOD motion of Complaint COUDsel, the Administrative Law Judge ordered this count dismissed from the complaint. See LD. at 3. The count fOnTS no part of the instant nro ""dinl!. HOSPITAL CORP. OF AMERICA 457 361 Opini The Commission recently considered the legality ofa hospital merger in the matter of American Medical International, Inc. No. 9158 (July 2 1984) (104 F. C. 1) In that case, we confronted some diffcult questions concerning the application of the antitrust laws to mergers in the hospital industry. Here, we face some equally diffcult questions regarding the nature of competition and anticompetitive behavior among hospitals. We affrm Judge Parker s finding ofliability and modify his opinion only as stated below.

B. The Structure of Health Care Markets Both parties agree that the health care industry is unique in some respects. Before considering the merits of this case, it (6) is important to have a fundamental understanding of the role of physicians and third-party payors in the health care transaction. The role ofthe physician is a market response to the extremely high cost to consumers of health care information and expertise. As a result of the patient's grossly imperfect information concerning proper diagnosis and treatment, and the doctor s much greater knowledge the doctor decides what diagnoses, treatments, and so forth the patient will have. See, e. Harris 3832. The physician orders tests prescribes drugs and courses of treatment, and so forth, and most important for our analysis, decides whether and when a patient wil be admitted to and discharged from a hospital, along with the battery of tests and procedures he receives while there. See id. The patient simply cannot decide these things for himself; the doctor is his repository of information and expertise and thus plays the critical role in determining the nature and extent of hospital and other health services the patient wil receive.

In addition to a lack of information about how to diagnose and treat himself, the patient has perhaps even less perfect information about the occurrence and extent of future ilness and injury. For the most part, neither the doctor nor the patient can control frequency or intensity of disease or injury. For example, the typical patient cannot anticipate or prevent being in an automobile accident or developing cancer. Likewise, the doctor cannot determine the type or intensity of diagnosis and treatment until a problem develops, to the extent that he can determine the severity of a problem within a short period of time (7) at all. The uncertainty associated with the nature and extent of potential health problems is thus enormous, and the uncertainty about the cost associated with diagnosis and treatment of such contingent events is equally high. As a result, the patient cannot plan financially for the treatment of his health problems; he may be healthy for the rest of his life and have to spend no money on health ...

Opinion 106 F.

care whatsoever, or he may receive an injury so serious that he could not possibly hope to pay for his treatment with his annual salary. What is the logical market response to this dilemma? Health insurance.

Insurance is a response to uncertainty, and spreads the risk of financial loss occasioned by treatment of disease or injury over both the people who turn out to have little need for health care and those who turn out to have a great need. By paying an insurance premium in a world where the future need for health care is uncertain, a potential patient eliminates the risk of not having the money he needs to pay for diagnosis and treatment, particularly of serious illnesses or injuries, should health care and of particular interest to us, hospital care, be needed. The insurance mechanism is thus an integral part the market for hospital services.

Moreover, because health insurance is considered a nontaxable fringe benefit to employees, tremendous incentives are created for providing health insurance at the workplace, and many employers provide it. See, e. Salkever 2337-8. Health insurance premiums are a cost of doing business for employers. See, e. , id. Barth 844-5. Thus employers have an interest in (8) the cost of health care. Some employee groups are even self-insured. See, e. Henson 967-69. Moreover, the coverage of health insurance is likely more extensive than it would be if prospective patients had to pay premiums out oftaxable income, perhaps evoking more coverage for less financially catastrophic events than otherwise. See Salkever 2337- With respect to our analysis, there is one extremely important effect on the hospital services market of third-party payment: The extent to which a patient is insured determines the extent to which he is sensitive to the price of hospital care. Ifhe is fully insured, once he becomes ill his interest lies in receiving the best quality care possible, including the highest quality comforts and surroundings if he is in the hospital, no matter what the costs. Who, then, is concerned about price? We would expect third-party payors and their customers the world of potential patients and employers who pay insurance premiums, to be interested in minimizing the costs of insurance. Of course, the government and taxpayers, who insure many of the elderly and under-privileged through the Medicare and Medicaid programs, should be interested as well. There is one wrinkle, however. When hospital prices rise, the increased payments made by an insurance company are spread over all its subscribers, both patients and non-patients (i. prospective patients); premiums rise less than proportionally to the increase in hospital prices. Thus, not every significant increase in hospital prices will bring a significant market (+-i('n front\ ;nc.llr-;nfo f'OnCllmArc Unurt:nrAr if'inCI1.."n"'o. ...."".. ..

.u 0 .uo_-_u_-.

361 Opinion rise suffciently, (9) even after-the cost of-health' care isspreacLover so many people, then consumer reaction should reverberate into the health care market. To avoid losing business and to minimize their costs, insurance companies wil through the insurance mechanism take whatever actions they can to hold down the prices they pay for hospital and physician care.

We are thus confronted in this case with a very peculiar market indeed. Because of the uncertainty of ilness and injury and the grossly imperfect information available to consumers of hospital services patients generally rely on physicians to determine the nilture and extent ofthe medical care they receive and on third-party payors to provide the financial assurances that such care wil be paid for. Any analysis of hospital markets under Section 7 must bear in mind both the role that physicians play on behalf of patients and the role of the insurance market in financing hospital care. With this in mind, we now turn to the merits of the case before us. II. HOSPITALS IN THE CHATTANOOGA AREA A. Hospitals Within the Chattanooga Urban Area The city of Chattanooga is situated in Hamilton County in southeast Tennessee on the state boundary with northwest Georgia. LD. 10. Chattanooga is the major city in two federally-designated geographic areas, the Metropolitan Statistical Area ("MSA") and the Georgia-Tennessee Health Service Area ("HSA"). An HSA is an area designated by the Department 0((10) Health and Human Services as a region in which state and local health planners are to assess and identify the health needs ofthe population in the region. LD.F. 12. An MSA is a large population nucleus, together with adjacent communities which have a high degree of economic and social integration with that nucleus. LD.F. 11. The Chattanooga MSA is a six-county area consisting ofthe Tennessee counties of Hamilton, Marion and Sequatchie and the Georgia counties of Walker, Dade and Catoosa to the south of Chattanooga. ld. Marion and Sequatchie counties lie to the west and northwest of Chattanooga, adjacent to Hamilton County; adjacent Tennessee counties to the north and east of Hamilton County are not part ofthe Chattanooga MSA. (11) Opinion 106 F.

BLEDSOf RHEA lN. GA DADE CATnDSA hlt, Cules lbitas Ow WAlKfR or Ma by lD ' Coties Nu Sh th liti in Fo Hoita with an Ge: of th 'I 1. Erlm MO Cete ll. Wi. Sataun 2. U. Pltt Ibita J. phe ftJ 13. 1hita SetdSo Ge IiltaHoita 4. D 14. Hhbl H:ita Cete fbta s. tbto Ge Iblta 15. Brey Coty Hoita 6. t: lblta 16. Clla Caty fbit. 7. Eat Ri Caty !b1ta 17. H.teed Cete B. le Ba Hoita Clty Par !J. No' 10. Jci L. Hu tbita Tr-oty lbita 361 Opinion (12) There are eleven general acute care hospitals in Hamilton County, Tennessee and the Georgia suburbs of Chattanooga-the Chattanooga urban area. " After the acquisitions, HCA owned or managed five of these hospitals. The eleven hospitals include: 1) Erlanger Medical Center. Erlanger is located in downtown Chat" tanooga. It is a non-profit teaching hospital owned and operated by the Chattanooga Hamilton County Hospital Authority, a public agency created by state statute, and is governed by a politically appointed board of trustees. LD.F. 22. Erlanger is a 780-bed major medical complex which provides a wide range of health care services to the community, and it is the only tertiary care referral center in the HSA. Id. Erlanger is required by law to accept all Hamilton County residents needing hospital care, regardless oftheir abilty to pay, and as a result treats the vast majority of indigent patients in Hamilton County. Approximately 100 beds are used for (13) the treatment of these patients at all times. Lamb 110-11, 115, 134; LD.F. 23. 2) Memorial Hospital. Memorial is located in downtown Chattanooga. It is a non-profit institution which is owned and operated by the Catholic Church. In 1981, Memorial was authorized by the Tennessee Health Facilities Commission to operate 349 general acute care hospital beds. LD.F. 27. Memorial is a secondary care hospital which also offers some tertiary services, such as open heart surgery. LD. 28.

3) Parkridge Hospital. Parkridge is the only hospital HCA owned in Chattanooga before the acquisitions. It is located in downtown Chattanooga. LD.F. 29. Parkridge is a secondary level hospital with some tertiary capabilities. LD.F. 30. Parkridge is authorized to operate 296 general acute care hospital beds. Id. 4) Diagnostic Center Hospital. Diagnostic is an 80-bed facilty in downtown Chattanooga which specializes in the diagnosis and treatment of cardiopulmonary disease. LD.F. 32. HCA now owns this facility, which it purchased from HAL 5) Downtown General. Downtown General is a 65-bed urban general acute care hospital which is located in downtown Chattanooga. LD. 38. It is a non-profit hospital that was managed by HAI before the HCA acquisition, and is currently managed by HCA. 6) Metropolitan Hospital (formerly Tepper Hospital) Metropolitan :3 Hospitals are sometimes classified from most basic to most advanced as "primary, secondary" or "tertiary. A primary hospital generally provides basic acute care servccs, such as obstetrics, surgicalscrvcCB, x-ray, clinical laboratory and blood services, a minimal level emergency room, phannacy and ancsthcsia services, and minimal intensive care capabilities. Secondary care generally includes the primary services listed above along with more specialized capabilities such as EEG equipment, diagnostic and therapeutic equipment for cancer patients and 24-hciur physician coverage. A tertiary level hospital genemlly has the same facilities as hospitals on the other two levels, but also has specialized services such as open heart surgery cHpabilitics, cardiovascular diagnostic lab, CAT scanner, bum-care unit, and oncology services. LD. at 10 n. 3. A referral hospital is one that because of the level of sophistication of its services is able to attract patients from smaller facilities having more limited servces and capabilities. ld.

Opinion 106 F.

is a 64-bed proprietary hospital owned and operated by American Healthcorp, Inc., which purchased the facility in (14) 1982, subsequent to the HCA acquisitions. LD.F. 39. The hospital' s primary focus is on pediatrics, but it has recently expanded into other services. Id. 7) East Ridge Community Hospital. East Ridge is located in the community of East Ridge, approximately six miles east of downtown Chattanooga. LD.F. 31. It is a general acute care hospital authorized to operate 128 beds, and has an active obstetrical practice in which it provides secondary level care. East Ridge is owned and operated by Humana, Inc., a large, for-profit hospital chain. Id. 8) Red Bank Community Hospital. Red Bank is a 57-bed general acute care hospital located in the Red Bank community north of Chattanooga. LD.F. 34. Red Bank is a non-profit hospital owned by the Health and Educational Facilities Board of the City of Red Bank and is leased to the Red Bank Hospital Association. Id. HAI managed the facility prior to the acquisition, and HCA managed the facility for several months after the acquisition.

9) North Park Hospital (formerly Medical Park Hospital). Medical Park was in 1981 an 83-bed facility in downtown Chattanooga owned by HCC before its acquisition by HCA. LD.F. 35. The hospital was relocated to tbe suburban north Chattanooga community of Hixson and opened as HCA-owned North Park Hospital in 1982. LD.F. 36. North Park offers primary and limited secondary level services. It has the latest technology available for the medical and surgical specialties and services generally offered in suburban hospitals. LD. F. 37. (15) 10) John L. Hutcheson Memorial Tri-County Hospital. Tri-County is located in the Ft. Ogletborpe community in the Georgia suburbs just across the state line from Chattanooga. LD.F. 41. Tri-County is a public hospital owned by the Hospital Authority of Walker, Dade and Catoosa Counties in Georgia. Id. Tri-County provides primary and secondary services comparable to those of Parkridge Hospital. LD.F. 42. In 1981, Tri-County was authorized by the State of Georgia to operate 237 general acute care hospital beds. It provides care for indigents and Medicaid recipients in northwest Georgia. Id. 11) Wildwood Sanitarium Hospital. Wildwood is a 39-bed facility located in Dade County, Georgia, approximately 10 minutes from Chattanooga. I.D.F. 43. It is owned by the Seventh Day Adventist Church and provides non-traditional services such as alcohol, diet and non-smoking programs. It also provides some general acute care ",,,..r;rocoC! Tri 361 Opinion B. Hospitals Within a 45-minute Driving Radius of Chattanooga The closest population centers to Chattanooga are: Dalton, Georgia which lies approximately 40 minutes southeast of downtown Chattanooga on Interstate 75 (LD.F. 14); Cleveland, situated in Bradley County, Tennessee, approximately 40 minutes northeast of downtown Chattanooga on Interstate 24 (LD.F. 15); South Pittsburg, located in Marion County, Tennessee, approximately 40 minutes west of downtown Chattanooga across the Tennessee River (LD.F. 16); Dunlap, located in Sequatchie County, Tennessee, approximately 45 minutes northwest of downtown (16) Chattanooga (LD.F. 17); and Whitwell, in Marion County, Tennessee, approximately 40 minutes northwest of downtown Chattanooga. LD. F. 18.

At least one general acute care hospital is located in or near each of these communities. Hospitals within 45 minutes driving time of Chattanooga, numbered as shown on the map above, include: 12) South Pittsburg Municipal Hospital. South Pittsburg is a cityowned facility in South Pittsburg, Tennessee, which was approved for use of 107 beds in 1981. LD.F. 44. It is comparable to the smaller Chattanooga hospitals in terms of quality and range of services. ld. South Pittsburg was managed by HAI before the acquisition and is now managed by HCA pursuant to a long-term contract. ld. The hospital is approximately 42 minutes from downtown Chattanooga by car. LD.F. 45.

13) Sequatchie General Hospital. Sequatchie General is a small primary care facility located in Dunlap, Tennessee. LD.F. 46. In 1981 it was approved to operate 49 beds. RX 1093. Sequatchie was acquired by HCA from HAI in the 1981 acquisition. In December, 1982, HCA sold the hospital to Republic Health Corporation as part of a larger transaction in which it obtained stock ownership in Republic. I. 47.

14) Whitwell Community Hospital. Whitwell is a 25-bed facility located in Whitwell, Tennessee. LD.F. 48. It is an old, lower quality facility with a small medical staff ld. However, it was purchased by Rural Hospital Associates, Inc. in 1982, and a new 40-bed facility is currently being built. LD.F. (17) 49. Whitwell is approximately 40 minutes away from downtown Chattanooga by car. LD.F. 50. 15) Bradley County Memorial Hospital. Bradley Memorial is a 251bed city-owned hospital located in Cleveland, Tennessee. LD.F. 55. It is a full-service community hospital that provides primary and secondary care and some intensive care services. It offers a range and quality of services comparable to many of the hospitals in Chattanooga. ld.

16) Cleveland Community Hospital. Cleveland Community is a 100- , .,._ ,, 464 FEDERAL TRADE Commisslon DEC1SlONS Opinion 106 F.

bed full-service hospital owned by HCA and located in Cleveland, Tennessee. LD.F. 52. The hospital is approximately 41 minutes from downtown Chattanooga. fd.

17) Hamilton Medical Center. Hamilton is a publicly-owned facility located in Dalton, Georgia. It is a 297 -bed facility approximately 40 minutes from downtown Chattanooga on Interstate 75. RX 1030(3); RX 1093. Hamilton is a modern medical center providing s condary care comparable to that available from Tri-County Hospital and to a lesser extent Memorial and Parkridge. Lamb 152; see Petruzzi 757; Furrey 1605.

III. THE PRODUCT MARKET An acquisition violates Section 7 of the Clayton Act "where in any line of commerce in any section of the country, the effect of such acquisition may be substantially to lessen competition, or to tend to create a monopoly." 15 D. C. 18 (1982). Accordingly, we now turn to the definition of the relevant "line (18) of commerce" or "product market" in which to measure the likely competitive effects of these acquisitions. In measuring likely competitive effects, we seek to define a product or group of products suffciently distinct that buyers could not defeat an attempted exercise of market power on the part of sellers of those products by shifting purchases to still different products. Sellers might exercise market power by raising prices, limiting output or lowering quality. See Grand Union Co., 102 F. C. 812 1040 (1983).

Complaint Counsel argued below that the product market was properly defined as the provision of acute inpatient hospital services and emergency hospital services provided to the critically ill.' This definition would exclude non-hospital providers of outpatient services free standing emergency centers, as well as non-hospital providers of inpatient services nursing homes, from the product market. It would also exclude the outpatient business of hospitals except for that provided to the critically il in the emergency room. The rationale for excluding outpatient care is that inpatient (19) services are the reason for being of acute care hospitals; inpatient services are needed by and consumed by patients in combination and therefore can be offered only by acute care hospitals. Inpatients in almost all cases will purchase a range of services and not just one test or procedure; they wil typically consume a "cluster" of services in- 4 Since hospitals in the main offer services, the relevant line of commerce in this case could be deemed the service market." However we wil adhere to the term of art "product market" throughout this discussion. (; More precisely, Complaint Counsel's c1lpert witness!, Dr. David S. Salkever, professor of health economics at Johns Hopkins University, testified that the market should be defined as "the provision of acute inpatient hospital services and the provision of hospital services to critically il emergency patients, typically who I suspect would : - ':H_ :t':_ .L_ ..ld "co_ , HOSPITAL CORP. OF AMERICA 465 361 Opinion volving 24-hour nursing, the ;ervices. of specialized laboratory and X-ray equipment, the services of equipment needed to monitor vital functions or intervene in crises, and so forth. An acutely il patient must be in a setting in which all of these various services can be provided together. Salkever 2282-83. According to this reasoning, outpatient services are not an integral part of this "cluster of services" offered by acute care hospitals, and therefore must be excluded. Respondent, on the other hand, urged that the market be defined to include outpatient care as well as inpatient care. Respondent' expert witness, Dr. Jeffrey E. Harris, testified that outpatfent care is growing rapidly for hospitals, as well as for free-standing facilties such as emergency care and one-day surgery centers, which compete with hospitals for outpatients. Harris 3943-4. Moreover, because of substantial changes in medical technology, there are a growing number of procedures that can be provided on an outpatient basis that previously could have been done on only an inpatient basis. Harris 3944.

Judge Parker agreed that the market should include outpatient services provided by hospitals but excluded outpatient services provided by non-hospital providers, holding that only (20) hospitals can provide the "unique combination" of services which the acute care patient needs. LD.F. 134. He defined the relevant product market to be the cluster of services offered by acute care hospitals, including outpatient as well as inpatient care since acute care hospitals compete with each other in offering both kinds of care and since. . . acute care outpatient facilities feed patients to the inpatient facilities. Id. Neither HCA nor Complaint Counsel appeal Judge Parker product market definition. See Commission Rule of Practice 3.52(b). Accordingly, for purposes of this proceeding only we accept Judge Parker s finding on this issue.

However, we do note that Judge Parker s definition does not necessarily provide a very happy medium between the two competing positions; the evidence in this case tended to show both that free-standing outpatient facilities compete with hospitals for many outpatients and that hospitals offer and inpatients consume a cluster of services that bears little relation to outpatient care. See, e. Coddington 463 Harris 3943-45; Parkhurst 1396; Kennedy 1760-3; Salkever 2279- 87. If so, it may be that defining the cluster of hospital inpatient services as a separate (21) market better reflects competitive reality Gin this case, the product market illue would unlikely be outcome-etermnative. Firat, even if the outpatient servces of non.hospital providers were included in the market, HCA concedes that their inclusion would not greatly affect hospital market ahares. See RAB at 16; tD.F. 135. Thus, it seems safe to conclude that adding such services to the relevant product market would add little to the analysis in this case. Second, both measures of inpatient market power and measures that include hospital outpatient services provide the SRme basis for liabiliy in this case. See disculIioninfra section V.C. Thus, excluding outpatient care from the market would add little to the analysis 466 FEDERAL TRADE CQMM1SS!QNDECISIONS Opinion 106 F.

in this case. In American Medical International, Inc. No. 9158, slip op. at 21 (July 2 , 1984) ("AMF') (104 F. C. 1), we defined the relevant product market as the "cluster of general acute care hospital services" to the exclusion of outpatient substitutes for individual services that comprise the cluster, since the "benefit that accrues to patient and physician" is derived from the complementarity ofthose services. It may well be that in this case the proper product market xcludes all outpatient care; perhaps outpatient care should be a separate relevant market or markets. In any case, it is clear from the evidence that the core and vast majority of an acute care hospital's business is acute inpatient care. See, e. Salkever 2286-87. Certainly, it is clear that anticompetitive behavior by hospital firms could significantly lessen competition for hospital inpatients that could not be defeated by competition from non-hospital outpatient providers. Our analysis wil hence proceed with primary reference to the cluster of services provided to inpatients.

IV. THE GEOGRAPHIC MARKET We now turn our attention to the relevant geographic market or section of the country" in which competition could be substantially lessened by these acquisitions. See AMI slip. op. at 21-22. Because we are concerned only with an area in which competition could be harmed, the relevant geographic market must be broad enough that buyers would be unable to switch to (22) alternative sellers in suffcient numbers to defeat an exercise of market power by firms in the area. Again, sellers may exercise market power by raising prices, reducing output or reducing qualiy. See Grand Union Co., 102 F. 812 1040 1047 (1983). Ifan exercise of market power could be defeated by the entry of products produced in another area, both areas should be considered part of the same geographic market for Section 7 purposes, since competition could not be harmed in the smaller area. That is, the geographic market should determine not only the firms that constrain competitors' actions by currently sellng to the same customers, but also those that would be a constraint because of their ability to sell to those customers should price or quality in the area change. See AMI slip. op. at 22. Looking at a "static" snapshot of a market is thus insuffcient in itself, since that picture might not reflect a likelihood of future anticompetitive market behavior suspect under Section 7. Rather, evidence of current market behavior must be viewed in a "dynamic framework that considers the possible competitive responses of firms outside the current market area to anticompetitive behavior of firms within. Of course, a static picture of the market is a logical starting _. _._ 361 Opinion ture is apparent can the potential for .competition from other areas in the event of an exercise of market power in the current market be assessed. Second, the type of evidence utilized in a dynamic analysis may often be the same evidence (23) used in a static analysis. For example, evidence of shipment patterns (or of concern to us here patient flow" patterns) may reveal not only the firms that currently sell substantial amounts to customers in an area but also those that could substantially increase shipments to the area in response to theanticompetitive behavior of firms within the area. In any case, evidence must be looked at in the dynamic light of potential harm to competition.

HCA would have us adopt Hamilton County, Tennessee, together with Walker, Dade and Catoosa counties in Georgia, the "Chattanooga urban area," as the relevant geographic market. HCA predicates its conclusion largely on an analysis of evidence concerning physician admitting patterns.

Dr. Harris, HCA's expert, testified that the relevant geographic market is determined to a great extent by physician admitting pracpatient choice, decides tice, because physician preference, rather than the greatwhat hospitals wil be utilized. He contended that "for majority of people we re talking about in a health care market like Chattanooga, you pick your doctor and then your doctor is the one who s going to decide where you re admitted." Harris 3965. HCA introduced a study that lists, for each hospital in the Chattanooga area, the physicians by specialty who admitted to the hospital, and the number of inpatient days for which each physician was responsible in all of the hospitals in the area. RX 1081; Harris 3961. With few exceptions, every physician who admitted to Chattanooga urban area hospitals admitted exclusively to other hospitals in the (24) Chattanooga urban area. Harris 3961-62; I.D.F. 148, 150-57. Conversely, physicians admitting and treating patients at hospitals outside the Chattanooga urban area rarely admitted and treated patients at hospitals in the Chattanooga urban area. Harris 3963-64; I.D.F. 150 158-62.

Indeed, testimony made clear that physicians playa primary role in determining where their patients are admitted. E.g. Lamb 366-67; Coddington 488; Kantanie 1147; Parkhurst 137&-0. Moreover, Chattanooga physicians try to limit their practices to a local area because of the time and inconvenience involved in traveling between hospitals. Coddington 499; Furrey 1550; Kennedy 1755; see AMI slip. op. at 24. The few Chattanooga physicians who do have consulting or courtesy privileges at hospitals outside the Chattanooga urban area 149; do not actively practice at the outlying hospitals. RX 1114; I.D.F. see CX 892.

, 468 FEDERAL TRADE COMMISSION PECISIONS Opinion 106 F.

Additionally, the weight of the evidence concerning patient origin suggests that patients admitted to Chattanooga urban area hospitals who live outside of the Chattanooga urban area are, with few exceptions, in need of specialized care and treatment unavailable in their own communities. Lamb 148; Dunlap 3344; Harris 3968; LD.F. 163 165-67, 169; see AMI slip. op. at 24. Hospitals in outlying communities do not always provide quite the same product that the urban area hospitals provide such patients, and therefore patient inflows are not necessarily indicative of the wilingness of patients to leave their home areas for services that are available in those areas. In fact, most admissions to urban area hospitals from outlying counties appear (25) to be through physician referrals. Dunlap 3344; LD.F. 16&-7 , 169. Judge Parker agreed with HCA that the Chattanooga urban area is the relevant geographic market in this case.7 (26) On appeal, Complaint Counsel agree that the Chattanooga urban area is an appropriate geographic area in which to assess the competitive effects of these acquisitions. However, they claim that a much more appropriate geographic market is the federally designated Metropolitan Statistical Area that includes Chattanooga. In effect, Complaint Counsel would have us add the Tennessee counties of Marion and Sequatchie to the market proffered by HCA and adopted by Judge Parker. By adding this area, three additional hospitals-South Pittsburg Municipal Hospital, Sequatchie General Hospital, and Whitwell 7 Judge Parker 11180 found that the Chattanooga urban area was the smallest geograptlic area at issue to satisfy the "Elzillga-Hogarty test" ("E-H test") 8S applied to patient flow data. This geographic market treat, named for the economists who developed it, measures the amount of commerce that leaves the market in queation and the amount that enters that aame area. See Elzinga and Hogarty,The Problem of Geogrophic Market Delineation in. Antitrust Suits 18 Antitrust Bull. 45 (1973); Elzinga and Hogarty,The Problem of Geographic Market Delineation Re!1i8ited: The Cuse arCual 23 Antitrust Bull. 1 (1978). More technically, the E-H test is based on LIFO ("little in from outside ) and LOFI ("little out from inside statistics. A LIFO statistic as applied to hospital geographic markets signfies the percentage of hospital patients from a particulsr area who remain in the area for hospital servcea rather than use hoapitals outside the area. When the great majority of patients residing in a specified geographic area use hospitals within that area, then only a small proportion of hospital services are "imported" to local residents from hospitals without the area (even though "importtion" of servces from outside hospitals would actually require local residents to drive to those outside hospitals). A LOFI statistic as applied to patient flow data signifies the percentage of patients of an area hospitals who reside in the area rather than outside the area. If very few patients of the hospitals in question are residents of other areas, the amount of commerce "exported" is very low (though "export" of local hospiw.l servces would require outside residents to drive into the local area). Evidence that few patients leavc an area (LIFOJ and few patients enter an area to obtain hospital services (LOFIj, strongly support the conclusion that the area conatitutes a relevant geographic market, according to the analysis. SeeSalkever 229fH, 2299-2305, 2504-6; LD.F. 141. IfLOFI and LIFO percentagcs are low, then the area in question is obviously too small and mult be expanded to include other hospitals to which patients can and do turn. See id. Judge Parker found that utilzing the Chattanooga urban area increases the LOFI value by 10 percentage points over the Hamilton County area, whereas the Chattanooga MSA and other broader potential markets increase the LOFI value only margina!1y over the Chattanooga urban area suggesting that one need not seek beyond the confines of Dr. Harris' market to find the relevant geographic market. " LD.F. 187;Bce 1087. Presumably, the smallest area to satisfy the E-H test is appropriate since areas Jarger than that do not necessrily reflect the abilty of hospitals in any sub-area to exercise market power. For example, the United States would likely satisfy the Elzinga-Hogarty test in this case, but it is obvious that a national market would be absurd; doctors would not refer patients to California for general acute care if anticompetitive behavior occurred in Chattanooga. See Salkever 2505. Conversely, signifcant market power could not be exercised in an area too small to satisfy the FrH test, since the current significant infux of patients into the area could obviously cease and the 361 Opinion Community Hospital-would be included - in the relevant marJcet. Both South Pittsburg and Sequatchie were acquired by HCA from HAI, and Complaint Counsel seek divestiture by HCA of its long-term lease arrangement with South Pittsburg.

The Chattanooga MSA is the better geographic market, Complaint Counsel argue, because it takes into account not only current competitive conditions but also likely dynamic responses (27) to market changes by potential competitors. The analysis offered by HCA and adopted by Judge Parker is purely static, they argue. Complaint Counsel contend that the hospitale in Marion and Sequatchie. counties could react to the exercise of market power by hospitals in the Chattanooga urban area by drawing away patients should prices rise or quality drop. They assert that the evidence relied upon by HCA and by Judge Parker does not "take into account changes that could occur as a result of the exercise of market power by firms in the area, and it was therefore an error for the judge not to find as a market an area that does take such changes into account." CAB at 22-23. One must wonder, however, why only hospitals in the two MSA counties to the west and northwest of Chattanooga would deter an exercise of market power by Chattanooga firms when there are high quality hospitals of equal distance to the northeast and southeast of the city. Bradley County Memorial Hospital and Cleveland Community Hospital in Cleveland, Tennessee, as well as Hamilton Medical Center in Dalton, Georgia, are no further from Chattanooga than South Pittsburg, Sequatchie or Whitwell hospitals. See supra section ILB. and map. Moreover, Bradley, Cleveland and Hamilton have some 648 beds between them while the three outlying MSA hospitals house merely 181 beds. Complaint Counsel would essentially have us expand the market from the Chattanooga urban area by drawing a half-circle around Chattanooga, ignoring the potential competitive pressures of the three hospitals to the northeast and southeast of the city. (28) Complaint Counsel nevertheless contend that the MSA is an "integrated area" within which patients have the ability to choose between hospitals. They make several arguments in support ofthis contention. First, because an MSA reflects general trade and commuting patterns, the potential competition among hospitals located in the Chattonooga MSA for MSA residents can be inferred. Second, Complaint Counsel point to federal government use of MSA's to distinguish between urban and rural areas, a distinction which is one factor in determining reimbursement levels under the new Medicare prospective payment system. They claim that "MSA' s have been determined by the federal government, as a purchaser of health care services, to 8 RCA no longer owns Sequatchie. See su.prasection n. Opinion 106 F.

define reasonable boundaries for pricing of hospital services." CAB at 24. Moreover, the full implementation of the Medicare prospective payment system wil allegedly stimulate "further integration" of MSAs into distinct hospital markets because hospitals outside an MSA will be disadvantaged in their capacity to compete with nearby MSA hospitals; the smaller "rural" reimbursement rate they receive will leave them less able to expend funds on quality competition and service improvements. At the same time, at least with respect to the substantial number of patients enrolled in the Medicare program, the MSA wil become to an even greater extent an area within which pricing patterns wil be linked and price movements will be similar according to Complaint Counsel. CAB at 25. (29J We reject Complaint Counsel's analysis. Although their criticism of HCA' s argument as a purely static analysis is valid, their own "dynamic" analyis is flawed. Although some of the market forces that are reflected in designation of an MSA may be relevant to the determination of the relevant geographic market in a Section 7 case see Grand Union Co. 102 F. C. 812, 1049 (1983), any inferences to be drawn from federal designation of an MSA in this case are insuffcient to establish the Chattanooga MSA as the relevant geographic market. Complaint Counsel's argument does not convince us that the hospitals to the northeast and southeast of Chattanooga would be less able to exert a restraining influence on urban area hospitals than the hospitals to the west of the city. First, we can find no reason why Chattanooga urban area doctors would be more wiling to obtain admitting privileges at outlying MSA hospitals than at the more modern and better-equipped Bradley or Hamilton hospitals. Nor do we find any reason why it would be more diffcult for Chattanooga doctors to obtain admitting privileges at the hospitals to the east of the city. Likewise, to the extent that patients influence where they are admitted, we find no explanation for why they would prefer the smaller, more rural hospitals to the west of Chattanooga. The considerations that go into designating an MSA, and the ways in which the federal government puts such designations into use, do not answer these crucial questions. The fact is that use ofthe Chattanooga MSA as the relevant geographic market would exclude a large number of hospital beds that are equally as accessible to (30J physicians and patients as the beds it would include. Certainly, the greater number of beds to the northeast and southeast would provide a greater constraint on the exercise of market power in the Chattanooga urban area, if at all. We therefore find Complaint Counsel's argument to be economically artificial.9 Patient inflow into the Chattanoogll urban area from outlying MSA counties cannot save Complaint Counsel' nrnnM"rl m rk"t. Fir discussed above. the weight of the evidence suggests that the great part of patient flow 361 Opinion We do not here contend that the proper market lnthis case includes all hospitals within a 45-minute radius of downtown Chattanooga; indeed, that question is not before us. Rather, we onclude that if the market were to be larger than the Chattanooga urban area, under a dynamic analysis it would have to include all equally likely sources of potential competition.

Additionally, we note the weakness of the evidence Complaint Counsel does offer in view of the presence of the hospitals to the east of Chattanooga. Geopolitical designations such as "MSA" may reflect a host of considerations that do not concern the issue of competition between hospitals. Certainly, the Chattanooga MSA was not designated in a manner that would explain why hospitals to the east of Chattanooga are less likely to deter (31) an exercise of market power in Chattanooga than are MSA hospitals of equal accessibilty to the west of Chattanooga. Nor do we find any evidence that MSA designations were ever intended to reflect an economic market for purposes of Section 7. We do not here conclude that an MSA will never accurately reflect the relevant geographic market in a hospital merger case. But where, as here, the MSA designation excludes important sources of potential competition, it must be rejected. Likewise, the argument respecting the new Medicare reimbursement system, though intriguing, is weak. Complaint Counsel ask us to rely upon pure speculation as to the effect of the new Medicare prospective payment system on competition between hospitals within an MSA and hospitals without for Medicare or other patients. This we decline to do. We find no basis in the record for concluding that the new Medicare payment system somehow will create a separate geographic market for section 7 purposes.

Since Complaint Counsel concede that the Chattanooga urban area is an appropriate market within which to assess the competitive effects of these acquisitions, we conclude that the Chattanooga urban area is the relevant "section of the country" for purposes of this case. However, we note that Complaint Counsel's criticism of the evidence offered by HCA is a valid one; HCA offered a static picture of the market without offering evidence or argument considering the likelihood or unlikelihood that physicians and their patients in Chattanooga would travel to outlying hospitals in the event of an exercise of market power by Chattanooga urban area hospitals. (32) A proper dynamic analysis might have considered some of the eviinto Chattanooga hospitals from outlying areas is for specialized treatment not available in outlying hospitals. Thus, such inflow does not reflect we!! the ability of outlying hospitals to compete away those patients should Chattanooga hospitills behave antieompetitively. Second, the evidence shows in any case substantial inflow from Bradley County which Lies to the east of Chattanooga.See Salkever 2301-02. Indeed, Complaint Counsel's own expert, Dr. Salkever, rejected the ChattanoogaMSA as the relevant geographic market in this case partly on that basis- Id. The Chattanooga MSA thus cannot be distinguished en the basis of inflow datil _ .._ _ .. +. . Opinion 106 F.

dence in this case as follows: "The closest hospitals to the Chattanooga urban area are about 45 minutes driving time away. See supra section ILB. Chattanooga doctors try as much as possible to avoid travel because it is time consuming and inconvenient. See supra this section. It is unlikely that doctors would be wiling to make rounds that far away from home on a daily basis in response to a small but significant reduction in the quality of hospital services in Chattanooga. They therefore would be very unlikely to admit patients to outlying hospitals. It is also unlikely that patients themselves would seek hospitalization that far away from home even if they recognized a small but significant change in the quality or price of services in Chattanooga. The evidence suggests that family and friends do not like to commute far to visit patients. See, e. Parkhurst 1393. Proximity to family and friends is therefore very important to the hospital inpatient. Thus it is highly unlikely that many patients and their employers would agree to insurance coverage that required extensive travel for health care, even if insurance carriers had to increase premiums because of a small but significant exercise of market power by Chattanooga hospitals. "

It is clear that the analysis offered by HCA and adopted by Judge Parker is incomplete. A review of patient flow data, physician admitting patterns, and other facts integral to a static analysis may all be important to a proper dynamic analysis, since a picture of current competition must be drawn (33) before competitive responses to changes in that competitive pattern can properly be considered. But without looking at those facts in a framework considering potential competitive responses to the current market picture, a relevant geographic area in which competition may be substantially harmed wil be extremely diffcult to define. In any event, the Chattanooga urban area is the area within which we wil assess the competitive effects of these acquisitions.

v. THE EFFECT ON COMPETITON A. The Effect of HCA-Managed Hospitals One of the major dimensions of HCA's purchase of HAI was the acquisition of some 75 to 80 hospital management contracts. Main 3756, 3773 (President of HCA Management Company) ("It was our opinion that the total price we were paying to INA to acquire the assets of Hospital Affliates from them included a certain amount of money designated as the value of the management contracts and the future revenues to be derived from them that HAI had at that _u - ,'\ rT r-..L n____- -- -- - J. ..- LT Ii T 1. ""1. g., -- ---,--_u_-- 361 Opinion hospitals in the Chattanooga ur15an area-'Downtown General Hospital and Red Bank Community Hospital. The terms of both managenent contracts were four years. CX 185I; CX 189G. The question arises as to what degree HCA's assumption of management of these two hospitals enhances its market position, and how best to characterize HCA' s management role in assessing the likely effect on competition of these acquisitions.!o (34) HCA argues, and Judge Parker agreed that Downtown General (35) and Red Bank hospitals should be treated an entities completely separate from HCA, incapable of being significantly influenced by HCA in its role as administrator. Complaint Counsel argue that treating hospitals that HCA manages as entities completely independent ofHCA is contrary to the facts in this case and to common sense. They would have us include the market shares of HCA-managed hospitals within the HCA' s market share or at the very least consider HCA management as a significant factor 10 RCA' s management relationship with Red Bank was tenninated by agreement in July 1982, almost ooe year aftr HCA purchased the management contract from HAl and prior to the contract' s expiration date. See CX 27M; CX 87; CX 185A, 1. We are thus faced with the diffcult question of whether RCA' s voluntary post-acquisition termination of the management agreement (see CX 87) should be considered in IlSBing the likely competitive effects of these acquisitions.AMIIn we examined the controllng case law and determined that consideration of post-acquisition evidence is appropriate when such evidence is relevant to the effects of a merger and is beyond the control of the merging finns.AMI slip. op. at 38-4; see United Stutes u. General Dynamics Corp. 415 U. 486 504-6 (1974); FTC u. Procter &Gamble Co. , 386 U,S. 568, 576 (1967); FTC v. ConsoliMted Food Corp., 380 S. 592, 598 (1965);United States v. Continental Can Co. 378 U.S, 441, 463 (1964). The genesis of the termination was a dispute between the Red Bank hospital board and HCA. The Red Bank board requested termination of the management contract because board members believed that HCA could not properly manage Red Bank while at the Bame time owning neighboring North Park Hospital; the new North Park which HCA acquired from HCC while it was stil "Medical Park" in downtown Chattanooga, was expected to he Red Bank's prime competitor in its new location. See Stulce 1198-99; Pust 3661; CX 27M; I.D.F. 213. Nevertheless, HCA voluntarily relinquished its rights under the management contract when it could have aB8rtd those rights. Indeed, HCA initially opposed the termnation and did not agree to the termination until March, 1982, aftr HCA was apprised in December, 1981 (see Hand at 14) that FTC staff was continuing its investigation of these acquisitions. Stu1ce 1204; XC 87; Pust 3634; I.D.F. 213;see Pust 3661-63; ex 214B; RX 954. Termnation occured in July, 1982 shortly before. the complaint in this suit issued. HCA rarely agrees to early termination of a management contract. Main 3757-59. In fact, RCA sued several hospitals that had management contracts with HAl and attempted to sever their relationship with HCA aftr HCA purchased those contracts. Main 3757-58;see exs 216-219. HCA obviously had a good measure of control over the future ofits relationship with Hed Bank; certinly, the Red Bank termination was not a matter completely beyond HCA' s infuence. See AMI slip. op. at 38-4. Thus, no matter what the natt1re of the dispute between HCA and Red Bank, we will not consider a volt!1tary termination of that relationship in asseSBing the likely competitive effects of these acquisitions. Otherwise, an acquiring firm could make anticompetitive acquisitions and then, when confronted with Liability, dispose ofasats in a manner that best retains its market position while reduci!1g statistical measures of concentration to more favorable levels.See Dean Foods Co., 70 F, C. 1146, 1269 (1966) ("We do not believe that post-acquisition market shares can be adjusted for actions which the acquiring company itslf undertok. To allow such self-serving adjustments to affect the legality of a merger would be to invite companies to merge and then to exercise the resultant power to restructure the market according to their whims and desires provided some optimum market share was reached or other favorable restructurng achieved which might by itslf be regarded as inoffensive to the competitive dynamics of the market if it had been achieved as a result of unmanipulated market forces American General Insurance Co. 97 F, C. 339, 340-1 (1981). We do not believe that respondents should be encouraged to make ilegal acquisitions and then cure them with self-appointed remedies. Rather, we think it is up to the Commillion and court to detennine the proper restoration of pre-acquisition levels of competitiveness once an ilegal acquisition is made. Since the original management contract with Red Bank could have beenin effect at the close of evidence in this case had HCA successfully opposed the tennination, we wil treat the Red Bank contract no differently from that RCA retained with Downtown General for purposes of liability. See ex 185. In any case, the exclusion of Red Bank from HCA's market share would not affect the outcome in this case, Seeinfra n. 18. , , , Opinion 106 F.

increasing the likelihood of anticompetitive behavior in this market. (36) We conclude that treating the two managed hospitals as entities completely independent of HCA is contrary to the overwhelming weight of the evidence in this case. As manager, HCA controls the competitive variables needed for successful coordination with the activities ofHCA-owned hospitals in Chattanooga. Moreover, I;S manager it knows the competitive posture of managed hospitals so well that the likelihood of any anticompetitive behavior HCA wished to engage in is greatly increased.

We note first the role of market shares in assessing the likelihood of anticompetitive effects under Section 7. Market share figures are merely a convenient way to depict the structure of a market; they are all artificial to the extent that they deviate from an accurate repreassigned to. See. sentation of the market power of firms they are Echlin Manufacturing Co., No. 9157, slip. op. at 8-9 (June 28, 1985) (105 F. C. 410); Grand Union Co., 102 F. C. 8I2, 1038-1 (1983). Including the market share assigned to a managed hospital within the market share of its manager is artificial only to the extent that it is not an accurate representation of the manager s market position. We hold that including the market shares of Downtown and Red Bank within HCA's share presents a much more accurate picture of HCA' market power than does ignoring HCA's management position or considering its relevance in some other manner. Both the chief executive offcer (the hospital administrator) and the chief financial offcer (the controller) are provided by HCA to Downtown General and Red Bank. CX 185G; (37) CX 189E; Chambliss 1434. Although the boards of the managed hosl',tals retain ultimate policymaking authority, the management contracts provide that HCA as manager is responsible for the day-to-day operation of the facilties and is charged with making recommendations to the boards regarding virtually all aspects ofthe institutional operation. CX 185; CX 189; see Main 3746, 3749-50. This responsibility includes preparing the budget and proposed hospital rates for the boards. CX 185D-E; CX 189C-D; Chambliss 1436-37; Stulee 1190-91. Thus, HCA is intimately involved in the rate setting process. HCA is also responsible for other important aspects of the way its managed hospitals compete, including the quality control ofthe hospitals, hiring and discharging personnel, advising the hospital boards concerning charges and services offered by the hospitals, and compliance with government regulations. CX 185; CX 189; see Chambliss 1437.

Indeed, the very reason that a management firm is hired, as reflect- ..1-- ed in the management contracts,J 1 rrOlis to direct. 1the- competitive opera- 361 Opinion management recommendations,- includjIlg _proposed rate increases are almost invariably followed by the boards of directors of Downtown General and Red Bank. Chambliss I435-40; Stulce 1191-92; Smith 1954; Kantanie 1057, 1059 , 1080; see generally CX 310; CX 311; CX 390; CX 391; CXs 428-80. Long-time board members of both hospitals acknowledge that they and their colleagues do not have the expertise necessary to administer hospitals, and for that reason rely heavily on management. (38) Stulce 1178, 1190; Chambliss 1425, 1432 , 1435; see also CX 185A.

Thus, the question is not whether HCA "dominates" the boards of the respective institutions, but whether the management' arrangements enhance the ability to coordinate behavior between HCAowned and managed hospitals so that any collusion in the market in which HCA desired to participate is more likely. We think it clear that the management relationships greatly enhance HCA's ability to coordinate behavior, since HCA personnel run the hospitals' competitive mechanisms and the hospital boards have hired managers for the very reason that the boards have neither the time nor the expertise to manage the variables of hospital competition themselves. We can see no reason why HCA recommendations to the Downtown General and Red Bank boards to raise prices a certain amount or cut back on the employment of certain personnel, for example, would be either detected or discouraged by board members given their substantial reliance on HCA. Nor can we see why coordination with owned hospitals would be diffcult for HCA to establish in view of its influence over competitive variables, access to information, and substantial control of the flow of information to the boards. Indeed, the evidence in this case indicates more than a symbiotic relationship between the two types of HCA hospitals. HCA's sister relationship program for coordination between HCA hospitals in the same local area includes both owned and managed hospitals. CX 206C-D; CX 214A; Pust 3664; Bennett 3724-25. In Chattanooga, HCA held meetings for both owned and managed hospital administrators "to initiate (39) them into the HCA philosophy." Pust 3649-50; see also CX 400A. Moreover, though the evidence is not clear on the specifics involved it is clear that HCA was in the preliminary stages of an area-wide plan to coordinate the activities of its Chattanooga hospitals-both owned and managed- but abandoned the plan when this litigation began. CX 625; Colton 3796-97; Stern 602-03, 611-12. In the case of Downtown General, HCA owns assets important to the continuing existence of the institution. HCA owns not only the land on which the hospital is located, but also the physician offce building situated adjacent to the facility. Chambliss 1419, 1463-64; Bennett 3699-701. Physicians with offces in the adjacent building opinion - 106 F.

account for approximately 95% ofthe hospital's admissions. CX 334J. Moreover, the hospital board acknowledges that the hospital would not survive without the medical offcer building. Chambliss 1423-24. Interestingly, Judge Parker recognized that hospital control over adjacent offce buildings is important to attract admitting physicians when he ordered that HCA divest such buildings in connection with the ordered divestiture of its hospitals. LD. at 91. As long as HCA has the power to control the medical offce building or divert its use, it appears unlikely that the Downtown General board will disrupt the current management relationship under any circumstances. ll (40) Certainly, since HCA has access to all competitively important data, even ifit had no input into deciding the competitive strategy of a managed hospital, it could use information about the hospital in connection with planning the competitive strategy of its neighboring owned hospitals. Having access to essentially all information about the managed hospital' s ability to compete or to react to market forces or to collusive conduct by other firms, HCA is in a much better position to confidently make competitive or anticompetitive moves itself; such information would allow HCA and other conspirators to better assess the likelihood of successful collusive behavior and the manner in which it should be carried out. It is noteworthy that the administrators of managed hospitals are required by HCA to prepare annual management plans for their hospitals. CX 427V; Pust 3650, 3653-54; see CX 28; CX 191.

Finally, whatever role managers in this market may play, it has not discouraged them from exchanging sensitive information about prices and wages with other administrators nor from attempting to organize concerted opposition to certificate of need proposals. E.g. Smith 1961 1981-82; Bennett 3719. Indeed, the Downtown General and Red Bank hospital boards might well encourage profiable anticompetitive behavior, since it would be advantageous to the managed hospitals as well as to HCA-owned hospitals. As discussed in greater detail below the Red Bank board members themselves in fact signed a market allocation agreement with a competing firm. See infra section V. In short, we find that considering managed hospitals to be (4I) entities independent ofHCA in examining an increased likelihood of competitive behavior in this market strains credulity. Doing so would seriously understate the likelihood of competitive harm in the Chattanooga urban area from HCA's acquisition of HAL The evidence compels us to consider the market shares of Downtown General and Red Bank as part of HCA's market share in considering the effect on 11 We also note that RCA maintains a renewal rate for its management contracts of approximately 95%. Main 3751; Pust 3660.

HOSPITAL CORP. OF AMERICA 477 361 Opinion competition in this case.!2 Even were the evidence not as compellng, we would consider HCA's management of the two hospitals to greatly enhance the likelihood of collusion in this market.!3 (42) HCA' s arguments in response are without merit. HCA argues that because a fixed fee is paid to it for management services, regardless ofthe profitability ofthe managed hospitals, any incentive to collude is mitigated. We disagree. First, higher prices or lower quality services at the managed hospitals would protect owned hospitals and others involved in an anticompetitive scheme from being undercut or outdone by the managed hospitals. Second, HCA in any case could set its prices and other competitive variables at owned hospitals more effectively with full appreciation of its managed hospitals' pricing and other competition policies.

Furthermore, we do not believe that the boards demonstrated their lack of reliance on their managers by seeking, before these acquisitions, renegotiation ofthe original HAI management contracts when Medicare refused to allow the full management fee under those contracts as a proper COSt.4 We would expect (43) hospital boards that intend to and do rely heavily on a management company to determine how their hospitals wil be run to show a keen interest in obtaining the best possible management at the lowest cost.!5 12 We note that HCA believes that management of hospitals for other owners gives it more "control over the environment" in which hospitals operate add "expands (its) market share. " ex 208C; CX414A. 13 Although, as Complaint Counsel concede, no case has yet specifically addressed on the merits the status of hospital management contracts in a hospital merger case, inclusion of the two contracts in RCA's market share is certinly consistent with what law does exist on hospital mergers and the case law interpreting Section 7 ofthe Clayton ACt generally. SeeAmerican Medicorp, Inc. v. Humana, Inc. 445 F.Supp. 589, 605 (E.D. Pa. 1977) (in preliminary injunction proceeding, court found that plaintiff had demonstrated a likelihood of success OD the merits regarding its allegation that a hospital merger would likely substantially lessen competition or tend create a monopoly because, 3S a result of the acquisition, defendant would "own or manage" the only two hospitals in the area);United States v. Archer-Daniels-Midland Co., 584 F.Supp. 1134, 1139 (S.D. Iowa 1984) (court held that the lease of operations of one finn 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 procompetitive purposes of the Claytn Act). Indeed, Section 7 "is primarily concerned with the end result ofa transfer ofa suffcient part of the bundle of legal rights and privileges from the transferring person to the acquiring person to give the transfer economic significanceand the proscribed adverse ' effect.''' United States I), Columbia Pictures Corp. 189 F.Supp. 153, 182 (S. 1960). As we have seen, with its acquisition of the Downtown General and Red Bank management contracts, HCA obtained suffcieot influence over the managed hospitals to coordinate behavior and lessen competition. .. HAl previously owned the two hospitals that became Downtown and Red Bank. Stulee 1174; Chamhliss 1425-26; Kantanie 1040. When the hospitals were in need of replacement, HAl arranged for the two hospitals to be owned hy newly organized non-profit corporations, the boards of which were selected by HAl, so that new hospitals could be built with tax-free bonds- Kantanie 1047--9, 1051-52, 1075-76, 1079; Smith 2026; Chambliss 1425; Stulee 1178. HAl arranged to manage the two hospitals under long-tenn, 25-year contract with management fees set at 8% of the hospitals' gross revenues. Kantanie 1052 , 1075-76, 1079; ex 185A, Z-l- Z-2; CX 624F, IT; Stulee 1178; Chambliss 1425. These two management contracts were unusually flworable to HAl. with very high management fees, Kantanie 1054. The contracts were later renegotiated with fixed fees resulting in much lower manage. ment fees and shorter terms, after the Medicare intennediary refused to reimburse the hospitals for much of the management fees because HAl and the hospitals were found to be "related" parties. Chambliss 1442; StuJce 1193; Smith 1972.

lS Likewise, we attach no significance to the decisions made by Medicare that IlCA' s predecessor in interest, HAI and the managed hospitals were not "related persons" under the renegotiated contracts. See I.D.F. 198-99. The detenninations concerned the integrity of the reimbursement system-whether the managed hospitals could claim they had made payments to HAl that were reimbursable, See id. They do not reflect a lack of power to influence the hospitals' competitive strategies, Indeed, IlCA admits that the Blue Cross detenninations are "perhaps not directly controllng here," RAnS at 11.

(footnote cont'd) Opinion 106 F.

Though HCA maintains a separate corporate subsidiary for managed hospitals, we do not think that the separation of management services from the operations of owned hospitals within the HCA corporate structure precludes successful coordination in this market. Certainly, it has not precluded joint meetings in Chattanooga ofHCA administrators from both types of hospitals and the formulation of an areawide plan" to coordinate the activities of both owned and managed hospitals in Chattanooga. See supra this section. In any case company-wide effcient organizations should not be a bar to coordination among local (44) administrators where collusion is suffciently attractive to the administrators.

HCA does not contend as a legal matter that the acquisition ofthese management contracts falls outside the purview of Section 7. Nevertheless, because the potential harm to competition from HCA managing its competitors is so clear in this case, we hold that the acquisition of a management contract that may substantially lessen competition or tend to create a monopoly violates Section 5 of the Federal Trade Commission Act independently of a violation of Section 7 of the Clayton Act. See General Motors Corp. 103 F. C. 641 , 700 (1984), and cases cited therein; see generally Averitt The Meaning of "Unfair Methods of Competition " in Section of the Federal Trade Commission Act B.C.L. Rev. 227, 251, 271 (1980).

B. The Nature of Competition Among Chattanooga Hospitals Traditionally, hospitals have competed for patients in three general ways: first, by competing for physicians to admit their patients; second, by competing directly for patients on the basis of amenities and comfort of surroundings; and third, by competing to a limited degree on the basis of price. Salkever 2332-38. The first two constitute "nonprice" or "quality" competition, and by far have been in the past the most important of the three. Id. Non-price competition for physicians includes the provision of upto-date equipment, a qualified and reliable nursing staff (45) and other technically trained personnel, convenient offce space to make it easier for the physician to concentrate both his ambulatory and inpatient work within the same location, a nice doctors' lounge with a good selection ofjournals-Bverything that wil convince physicians that their patients are receiving the best care possible and make physicians ' lives more comfortable. Id. at 2333. Competition directed at patients themselves has traditionally been through the provision We also reject any suggestion that under the renegotiated contrOicts, the management company had less authori. ty. The evidence is conclusive that management under the old and new contracts did not differ and that HCA played the same role as manager that HAl did both before and after the original Downtown General and Red Bank contracts were renegotiated in response to the Medicare disallowance. Chambliss 1440-1 , 1449, 1451; Smith 1975-77: Pust 3649: Bennett 3708: Stulce 1197- ... .

n...... LLJ-U- U . "".L u.......'V 361 Opinion of amenities, such as pleasant surroundings, attractive rooms, televisions and telephones, high nurse-to-patient ratios, convenient parking--verything that wil make patients more comfortable. Id. 2334.

Price competition, on the other hand, was much more limited in the past than non-price competition. Hospital prices meant little to patients because by the mid-1970s they were almost completely insured. Id. at 2335-38. Patients were paying only a minimal amount out of their own pockets for hospital care. Moreover, health care fringe benefit costs were not a large portion of total costs for employers. Id. Employees desired the non-taxable income of health care benefis and, consequently, sought the best insurance plans possible in their benefits packages. Insurers weren t concerned about price because employers and employee groups, their primary customers, weren concerned about insurance costs. Id. The result of this price insensitivity by consumers, their employers and their third-party payors was that little incentive was given hospitals to bid patients away from other hospitals by maintaining lower prices (46) and informing thirdparty payors and customers of those lower prices. Id. at 2335. Instead a more "indirect" form of price competition took place. Afraid of losing patients because of a public perception that they were more expensive than their competitors, hospitals felt some constraint on raising prices. Id. However, because of the state of the third-party payment mechanism, hospitals did not compete vigorously on the basis of price. Id. at 2336.

Over the last decade, two major trends increasing competition developed. Id.among hospitals beyond its traditional limits have 2338-50. First, both non-price and price competition are now being directed much more toward patients themselves than in the past. Second, beginning in the late 1970s the hospital industry has seen the clear emergence of direct price competition. At the same time, traditional non-price competition for patients on the basis of amenities has intensified somewhat, through the provision of such amenities as private rooms. Non-price competition for physicians remains pervasive, since physicians stil largely determine the disposition and treatment of their patients. Id. at 2338-1, 2348. Increasing competitive efforts aimed directly at patients include health education programs, CPR classes, community activities, and direct mass media advertising, such as on bilboards and radio. Id. 2339-40 2348. The clear emergence of direct price competition is even more striking. Hospitals are now trying to attract the business of employers and insurers by (47) offering price discounts. Id. at 2342- 43. The reason for this increase in price competition is a reaction in the health insurance market to rising costs. ., , _. 480 FEDERAL TRADE C()MMISSIONDECISIONS Opinion 106 F.

Total health care and hospital costs in the United States have grown precipitously over the past several years. Id. at 2342, 2348-50; CX 812C. HCA itselfhas predicted that "total expenditures for health care services and products (area likely to continue to rise at a rate " CX 209Z-8 (1981); seewhich will alarm purchasers of health services. I.D.F. 242. The rapid rise in health care costs finally began to impact seriously on employers, employee groups and third-party payors. Salkever 2342-43. Health benefits, which were once a small element in employers' total payroll, suddenly became a major element in their cost structures, accounting for as much as 15 or 20% of total payroll cost. Id. at 2342; Barth 844-5. Moreover, where plans in which employees paid part of the health insurance premium out of their own pockets were operative, employees felt the same pressures because premiums were rising rapidly. Salkever 2342. Even where employees weren t paying part of the premiums themselves, employee organizations such as unions realized that the other benefits in their overall benefits packages were dropping because of the increasing cost of health care benefits. Id. at 2343. The result of rising insurance costs has been a change in employer and employee concern that in turn has changed insurer behavior. Insurers realized that there was a market for cost containment and have reacted by providing new insurance packages (48) and by marketing cost containment services. Id. at 2343-4. New insurance packages include alternative forms of health care delivery, such as Health Maintenance Organizations ("HMOs ) and Preferred Provider Organizations C'PPOs ), and higher first dollar deductibles and coinsurance in health care policies. Id. Barth 849-66. HMOs and PPOs operate as group purchasers of health care services. LD.F. 249. An HMO is a plan in which a subscriber prepays a fixed fee in return for comprehensive health care. Harris 3850; LD. F. 250. HMOs typically maintain contractual arrangements with physicians and hospitals to provide care to subscribers who as a result generally have lower health care expenditures than other patients. Harris 3850 51; LD.F. 250. Studies have shown that HMOs influence hospital competition and have been increasingly successful in obtaining discounts from hospitals. See, e. RX 1126(1),(3),(6). According to HCA' President and ChiefExercutive Offcer (aJ successful HMO wil help make other providers in a given market more responsive to consumers, as well as more cost conscious." CX 123E. HMOs thus act as additional buyers for which hospitals must compete, on price as well as quality terms.

A PPO is an arrangement whereby health care providers contract to provide services at a discount to volume purchasers of health care 361 Opinion There is generally a financial incentive for the group members or patients to use the "preferred" providers, such as a waiver of deductibles or coinsurance. CX 309; LD.F. 251. Industry now commonly negotiates with health (49) care providers for discounts. Wiliamson 3269. A good example of such an arrangement is the PPO recently created by HCA in Florida, on a discounted fee-for-service basis. CX 616. The HCA PPO is "designed to introduce a new competitive element into the comprehensive health care market." CX 616J. Higher first dollar deductibles and coinsurance in health care policies mean that patients are now paying more of the costs of)lOspitalization out of their own pockets. See Barth 851, 855; CX 534X; RX 1105(9). Recent studies have shown that health care expenditures vary depending upon how much of the cost of health care is borne by the consumer and that price has some effect on consumers' decisions to seek health care in the first place. CX 885 at v; CX 894D. Any time patients pay more out of their own pockets, we would expect some increased sensitivity to prices. "Stop-loss" provisions that limit the total out-of-pocket expenditures of patients under their insurance policies are common, however, and we therefore would not expect patient out-of-pocket expenditures in themselves to tremendously increase price competition among hospitals for inpatients. See Lamsey 1802-04 ($1500 stop-loss). If a major ilness or injury occurs the patient will have to pay the same amount out-of-pocket wherever he goes, unless his third-party payor is wiling to waive the deductible and coinsurance if he uses a lower-priced provider. Indeed, this is precisely the manner in which attempts to control costs through greater deductibles and coinsurance foster significant price competition among hospitals. See Stern 574 (price discounting by hospitals can take the form of (50) not charging for co-payments). For example, HCA itself has created its own health insurance plan with financial incentives, such as lower deductibles and coinsurance, designed to encourage subscribers to use HCA hospitals. CXs 809-13. In addition, insurers are not simply selling insurance but also marketing "cost containment" services to employers-the ability to assemble data on prices of services provided by alternative providers as well as the ability to monitor claims and to hold down costs by reducing utilization, a process known as utilization review . Salkever 2345; Barth 841-46. Finally, employers nationwide are trying to contain health care costs by forming coalitions to develop data bases on alternative prices for different providers, making comparison shopping easier. Some employers are even directly counseling employees to use cheaper providers that stil provide good quality care. Salkever 2346-7.

In sum, this increasing concern of employers and employees with g., Opinion 106 F.

the costs of insurance means that differences in prices between hospitals matter to them and their third-party payors, since insurance will cost less when hospital care costs less. Salk ever 2347; see supra section LB. The result is that hospitals are now far more likely to present themselves to insurers, employers and employee groups as less costly than their competitors as one method of attracting more business. ld. isPrice competition, fostered by these new insurance mechanisms, therefore growing in the hospital industry. See Salkever 2341 , 2347; Kantanie 1094; CX 309B; CX 616. (51) HCA' s claim that any increase in price competition among hospitals is insignificant is belied by its own records. HCA itselfhas predicted a "more price competitive environment" for hospitals because of increased pressure from private industry to reduce hospital expenses. CX 421Z-2; see, e. CX 100H. HCA planners have noted the sensitivity of major purchasers of health care to rising costs, which should stimulate considerable competition among health care providers. CX 209H; see also CX 108Z-1. Another HCA document reflects the belief that "increasing competition in the health care sector. . . will allow natural market forces to slow the price rise spira!." CX 221Z-2; see, e. CXs 357, 209Z-17 -18.

Moreover, the rise of price competition is reflected in HCA's policy that its hospitals should attempt to keep their charges at a competitive level. Its 1982 form 10-K states that "(t)he rates charged by the Company s hospitals are intended to be competitive with those charged by other local hospitals for similar accommodations, supplies and services" (CX 13Q); and the 1983 Management Plan for HCA East, the division that includes Chattanooga hospitals, states that (pJrices are budgeted to increase at the maximum competitive level in each local market." CX HOX. Because of the rise of health care costs and the resultant reaction by purchasers, HCA believes it has an opportunity to "captur(e) additional patients in areas where price competition becomes more important and we are the low cost provider." CX 209Z-1O. HCA is also concerned about price discounting by its competitors. CX 209" ; LD.F. 248. The LD.F. 248. The (52) record thus establishes the emergence and likelihood of increased price competition among hospitals because of pressures from buyers. Both the traditional forms of non-price and price competition are evident in Chattanooga, as well as the emerging trends, though changes such as the development of HMOs and PPOs are proceeding more slowly than elsewhere. E.g. Salkever 2350-53. The evidence is clear that Chattanooga hospitals compete for physician patronage (e. Lamb 134; Stern 547; Petruzzi 697-98), and that they do so in a variety of ways. They compete with respect to the range and quality rl tho .."'" lah thn.",-n t-.. 1 .."'..n,. ('f (.':ru i('p Pflnlnrru::\T"It fpr" Y1O''g., , _. A A u. 'uu 361 Opinion they provide. Lamb 171- Stern 556- i; Petruzzi 721-25; Kantanie 1093; Parkhurst 1390-92; Furrey 1539-40. For example, Erlanger recently updated its heart catheterization equipment in response to the better quality of Memorial's. Lamb 174-75. Chattanooga hospitals also compete for physician admissions by providing offce space close to the hospital, and other physician amenities. Stern 556-57; Petruzzi 723; Parkhurst 1390-91. They further compete for physicians by providing a pleasant environment and the amenities that keep the physicians ' patients satisfied. Parkhurst 1390-91. Hospitals compete for patients by offering personalized care attractive facilities, private rooms, parking, televisions, telephones, good meals and quality nurses. Lamb 171, 182; Stern 560- , 569; Furrey 1546.

Moreover, the testimony of hospital administrators is overwhelming that Chattanooga area hospitals compete in some (53J manner on the basis of price. Lamb 176-78; Stern 560; Petruzzi 727; Furrey 1525 1537; Kantanie 1060-61; Smith 1955. The price competition that has traditionally existed in Chattanooga is meaningful enough that competition could be harmed substantially ifit is restricted. See AMI slip. op. at 4- , 32-33. For example, the Blue Cross participating hospital system is a form of this price competition that has existed for years. Under the plan, hospitals become members of Blue Cross and Blue Shield of Tennessee and agree to charge only Blue Cross-approved prices to Blue Cross subscribers in order to attract them as patients. Long 1280, 1285. Blue Cross has a large number of subscribers who receive higher amounts of reimbursement ifthey are treated at member hospitals. Id. at 1291-92. The result is a strong incentive for hospitals to participate in the program; indeed, all ofthe Chattanooga urban area hospitals are member hospitals. Id. at 1304. This is "price competition" because if a hospital unilaterally refuses to deal with Blue Cross at the desired rates, it wil lose business to competing hospitals that are willng to charge lower rates for Blue Cross subscribers.

Moreover, the evidence shows that hospital rates in Chattanooga are established with at least some reference to the rates of other hospitals. See, e. CX 177B; CX 139Z-53; CX 239. Formal and informal rate surveys have been used by area hospitals to determine whether the surveyor s prices were within the range of prices offered by competing hospitals or to justify price increases to Blue Cross. Stern 564, 632-39; Smith 1958- , 1965-66; CX 845. Almost every hospital in the (54J Chattanooga urban area has conducted rate surveys at one time or another or participated in them. E.g. CX 179 (Diagnostic); CX 184 (Downtown General); CX 276 (Parkridge); CX 286 (Red Bank); CX 316I (Erlanger); CX 326H (Tri-County); CX 845 (Medi- );

Opinion 106 F.

cal Park); CX 158 (Memorial participated); CX 161 (East Ridge participated); LD.F. 225. In some instances, surveys and informal rate inquiries have concerned not only current prices but prospective prices. CX 184G; CX 239A; Kantanie 1134; Lamb 206. The surveys in the record list room rates at other hospitals and charges for a number of ancilary services. See, CX 179; CX 184; CX 286. In fact, those hospital rates tended to fall into a Pilttern in Chattanooga prior to the challenged acquisitions, with Erlanger prices being the highest, Memorial' s the lowest, and the other hospitals' rates somewhere in the middle. Lamb 176, 208; Stern 566; Kantanie 1070-71; CX 239; LD.F. 226. Other hospitals in Hamilton County tried to set their prices not too far above Memorial's and somewhat below Erlanger s. Kantanie 1070-71. Finally, the importance of rates to area administrators is reflected by the fact that rates were often discussed informally at meetings or over the telephone. Lamb 206-9; Stern 577.

HCA' s Parkridge Hospital has long been concerned that its rates were significantly higher than Memorial Hospital's. CX 18Z-97; CX 139Z-53; CX 301Z-50. In 1978, Parkridge increased recovery room and anesthesia rates, instead ofthe more visible operating room rates for competitive reasons." CX 301Z-50. A 1982 Parkridge management plan also discusses Memorial' s lower (55) prices (CX 18Z-65; CX 18Z-97), and concludes than Parkridge "must price itself more by looking at the local competition. " CX 18Z-7; see also CX 18Z-3, 64 66; CX 38Z-96; CX 139Z-53. The plan notes that "competing hospitals have chosen to hold down price and charge increases for 1983 significantly lower than Parkridge." CX 18J.1 Thus, it is obvious that price has been a competitively sensitive matter among Chattanooga hospitals. See Lamb 177-78. We do not here conclude that price has been the prime arena in which hospitals in Chattanooga compete. However, we do think it clear that even though rates are not constantly adjusted due to a changing price structure, they have been periodically set with some reference to what the market will bear in face of the prices of other hospitals. See , Kantanie 1069-70.

It is clear that Section 7 protects whatever price competition exists in a market, however limited. AMI slip. op. at 33 CTTJhe antitrust laws wil endeavor to protect this price competition, if, for nothing else, the hope that price competition will be enhanced" see United States v. Philadelphia National Bank 374 U.S. 321 , 368-69 (1963); Stanley Works v. FTC, 469 F.2d 498, 505 (2d Cir. 1972), cert. denied 16 Other HCA hospital management plan contain similar statemenl. reflecting I-CA's coocern with remaining competitive on prices and price increases. ex 20Z-5 (North Park); ex 28H (Downtown); see also Williamson 3269; 361 Opinion 412 U.s. 928 (1973). HCA, ofcollse, admits that this price competition should be protected ifit is endangered. RAB at 33. Indeed, the (56) fact that price competition is limited does not mean that consumers cannot be hurt substantially by its curtailment. For instance, any agreement among hospitals to refuse to accede to rates requested by Blue Cross, even under the traditional plan, would harm Blue Cross and its subscribers. See discussion infra section V. There is, however, even more reason to conclude that price competition could be harmed by these acquisitions. The evidence shows that the industry-wide growth in price competition has taken root as a market phenomenon in the Chattanooga area. Price competition is increasing and appears likely to further increase significantly among Chattanooga hospitals.

Because employers and employee groups in the Chattanooga area are experiencing severe cost increases, they have been forced to make significant changes in their employee health benefit plans. Lamb 179; Henson 967-78; Siskin 1232-33; Lamsey 1806-13. For example, Provident Insurance Company s model insurance plan, which it both markets and uses for its own employees, calls for 20% coinsurance and a 100 to 150 dollar deductible that employees must payout of their own pockets. Barth 852-53. It is designed both to save employers money and to give employees incentives to seek less costly health care. ld. 851. Siskin Steel Company changed from a plan providing basic benefits for hospitalization with no deductible or coinsurance to a plan with a 100 dollar deductible and 20% coinsurance for hospitalization. Siskin 1232, 1242-3. American Manufacturing Co. went from a similar basic benefits plan to a plan with a 200 (57) dollar deductible and a 20% coinsurance payment up to a $1500 stop-loss. Lamsey 1802-D4; LD.F. 243. Blue Cross also offers a comprehensive plan with a deductible ranging from 100 to 250 dollars and a 20% coinsurance amount up to some stop-loss level. Long 1295-96. Blue Cross now sells more of these plans than its traditional basic coverage. Long 1297; LD. 243. These changes are intended to encourage employees to become more cost conscious and therefore at least to some extent to be more sensitive to differences in the prices charged by different health care providers. Siskin 1234; Barth 851. In fact, the evidence shows some increased price sensitivity on the part of consumers. See, e. Lamb 179 ("Patients do comparisons. They shop around some. . . (A)s insurance coverages change and patients become more a part of paying the cost of their own hospital bils they are becoming more concerned about their cost. ); CX 605, Petruzzi 726. Following the industry trend, some Chattanooga area employers have been counseling their employees in an effort to persuade them to seek health care from low cost providers. Roddy 1913-14; Siskin . . .

Opinion 106 F.

1234; see Lamsey 1816. In response to Memorial's encouragement various companies have urged their employees to utilze Memorial Hospital because of its low costs. CX 601; CX 605; Furrey 1533; Roddy 1915; Siskin 1236, 1241-42. Moreover, employers are developing data bases so that the prices of services provided by different health care providers can be more effectively compared. See Siskin 1236; Lamsey 1816; Barth 865, 87 71; Salkever 2346. (58) As a result of these phenomena, Chattanooga hospitals are beginning to show the signs of more direct price competition. Memorial has encouraged employers to steer employees to it on the basis of its lower prices and concurrent high quality, and has disseminated newsletters to Chattanooga industries encouraging price shopping for health services and preadmission counseling of employees. CX 601; CX 605; Furrey 1533. Memorial has urged employers "to identify the hospital which is giving your employees the best service for the least dollar amount" and "to take a look at the price differential among hospitals. " CX 601. East Ridge has advertised emergency room price lists and its publications explain the pricing system at the hospital and urge prospective patients to inquire about hospital accommodations and rates when hospital-shopping. Petruzzi 703--4; CX 255D; CX 265. In addition to increasing price competition, the evidence clearly shows significant marketing directly to patients by Chattanooga area hospitals. East Ridge has advertised its emergency room prices on television. Petruzzi 703-704. Metropolitan also advertises on television. Kantanie 1102. Chattanooga hospitals advertise new procedures and equipment in newspapers. Coddington 487-88; Furrey 1537. In order to compete, Red Bank advertises on the radio and bilboards. Smith 2000; Stern 574. Chattanooga hospitals also make direct mailings and distribute promotional literature to individuals. Petruzzi (59) 699-705, 710; Parkhurst 1394. The evidence further shows that advertising is increasing in Chattanooga. Lamb 18 8l. !7 The importnce to our analysis of the emerging competition in the health care industry is eggentially acknowledged by HCA, we think, in ils attempt to exclude relevant post-acquisition evidence in this case. RCA argues that these acquisitions should bejudged by the facts as of the time the acquisitions occurred and that those factsindicate no likely lessening of competition.

We reject HCA'B attempt to exclude this evidence for two reasons. First, it presumes that this post-acquisition evidence does not "iluminate the validity of argument: relating to a possible lessening of competition based fundamentally upon market dynamics at the time ofthe acquisition." RAB at 52. IlCA is simply wrong- The record is conclusive that the trends toward both nationwide and in Chattanooga toward greater price and non.price competition, advertising and other marketing, patient awareness, and the cost pres. ures that spawned those trends, were well underway before the 1981 acquisitions at is. ue here. Salkever 2348-O; Lamb 180-1; Stern 561, 573, 649; Furrey 1547, 1553; Barth 845, 856, 894; Roddy 1912; Petruzzi 726. Thus, post-acquisition evidence relating to these trends does "iluminate the acquisitions' likely effects on future competition evident at the time of the transactions Second, and more fundamentally, considering this post-acquisition evidence is appropriate here. While exculpatory post-acquisition evidence of voluntary actions by the acquiring firm may be inappropriate for determining the legality of an acquisition, it is appropriate to consider changes in the market that are beyond the control of the acquiring firm and that are relevant to the effects of an acquisitiun.United Stales u. General Dynamics Corp. 415 U.S. 486, 504 (1974); seen. 10supra. Clearly, the changing market conditions nationally and in the Chattanooga hmmn Ifr' 1 Q lmm t 'n ,hD ", ffDdo nf'.hDQ", -.rn" ;';n o. nn"'_ ;t;n QnrD"' 361 Opinion Finally, we note that Chattanooga hospitals compete in the recruitment of qualified medical personnel, such as nurses and laboratory technicians. Lamb 171; Furrey 1546, 1560; Stern 569-70; LD. 224; see, e. CX 18Z-7. The evidence shows that they offer salaries and benefits competitive with other (60) hospitals to retain their personnel. Stern 569-70; Furrey 1560; Smith 2006; CX 143, p. 25; CX 163; LD.F. 224. Therefore, any analysis of the likely effects on competition from these acquisitions may also consider competition for inputs.

C. Respondent s Market Share and Concentration in the Chattanooga Urban Area Three ways to measure a hospital' s share of the acute care hospital services market are by using: (1) bed capacity; (2) inpatient days; and (3) net revenues. Bed capacity and inpatient days measure a hospital's position with regard to the cluster of inpatient services, the heart of hospital care. Net revenues, on the other hand, account for both inpatient and outpatient services.

Naturally, because oftheir proposed market definitions, Complaint Counsel advocate use of inpatient measures, while HCA urges net revenues as the preferable measure since it accounts for outpatient services. We conclude, however, that the three measures are so similar in this case that they yield the same result whatever measure is used.

The following table demonstrates clearly the significance of concentration in the Chattanooga urban area and the rise in concentration and HCA's market share, including managed hospitals, because ofthe acquisitions. (61) Concentration Statistics for the Chattanooga Urban Area Before After HAl After HCC Based on: quisitions guisition acquisi 1981 approved acute-care beds:

HCA market share 13. 22. 26. firm concentration ratio 51. 58. 62. firm concentration ratio 76. 85. 89.4% Herfindahl index 1932 2242 2416 1981 patient days:

HCA market share 13. 23. 25. firm concentration ratio 54. 58. 61.4firm concentration ratio 79. 88. 91. Herfindahl index 2028 2338 2467 1981 net patient revenues:

HCA market share 14. 22. 25. firm concentraton ratio 56. 62.4% 64. Opinion 106 F.

firm concentration ratio 80. 89. 92. Herfindahl index 2220 2515 2634 SOURCES: RX 1092, RX 1094: CAB a1 B-3: RX 986 (402); CX 612; CX 92J; CX 1 Q6K. (62I The Herfindahl-Hirschman Index ("HHI") of market concentration is calculated by summing the squares ofthe individual market shares of all the firms in the market. The HHI reflects the distribution of market shares between firms and gives proportionately' greater weight to the market shares ofthe larger firms, which likely accords with their relative importance in any anticompetitive interaction. U.s. Department of Justice Merger Guidelines (June 14 , 1984) CDOJ Guidelines 1; FTC v. Bass Brothers Enterprises 1984-1 Trade Cas. (CCH) n 66 041 , at 68 609 (N.D. Ohio 1984). As seen from the table using any measure of market power the Herfindahl index was above 1900 before the acquisitions. Thus, the acquisitions occurred in a market already highly concentrated. Bass Brothers at 68 620-21; DOJ Guidelines 1. Following HCA's acquisition of HAI, the HHI increased some 295 points using net patient revenues and over 300 using beds or patient days. With the acquisition of HCC, the HHI additionally increased well over 100 points using any measure. Again using any measure, the HHI at the very least rests at 2416 after the acquisitions. We consider such an increase in concentration in an already concentrated market to be of serious competitive concern, all other things being equal. See Bass Brothers at 68 620-21; DOJ Guidelines 11.1 More traditional measures of market share also support this conclusion. For example, using patient days HCA's market share (63) increased from 13.8% to 25.8% in the Chattanooga urban area, while four-firm concentration increased to almost 92% and two-firm concentration to 61 %. The figures for approved beds and net patient revenues are almost identical. These figures support an inference of harm to competition, all other things equal. See g., United States v. Philadelphia National Bank 374 U.S. 321, 364-65 (1963); Liggett & Meyers, Inc. v. FTC 576 F.2d 1273, 1275-76 (4th Cir. 1977); Marathon Oil Co. v. Mobil Corp. 530 F.Supp. 3I5, 323-24 (N. D. Ohio), aff'd, 669 2d 378 (6th Cir. 1981), cert. denied 455 U. S. 982 (1982); see also Scherer Industrial Market Structure and Economic Performance 280 (2d ed. 1980); R. Posner Antitrust Law: An Economic Perspective 55-56 (1976).9 lR Even excluding Red Bank from this calculus, the HUI jumps over 200 points, from 1932 to 2136, because of the HAr acquisition alone using beds, and almost 200 points using net patient revenues or patient days. '9 Additionally, with four-firm concentrlltion 130 high, the "fringe" finns should be able to "compete away" very litte business from the largest firms should the largest firms collude, if barriers to expansion are significant. With the significant barriers to entry and expansion in this market, discussed below, such firms could not likely expand to make conusion among tbe lan er firms unprofitable. Salkever 2364-66. g, 361 Opinion Moreover, all other things being eqUlI, an increase in marketconcentration through a reduction in the absolute number of competitive actors makes interdependent behavior more likely. Salkever 2366 67; Harris 4244-46; F. Scherer Industrial Market Structure and Economic Performance 199-200; 2 P. Areeda and D. Turner Antitrust Law n 404 (1978). These (64J acquisitions decreased the number independent firms in the market from 9 to 7. The costs of coordination or of policing any collusive agreement are less with fewer participants, and the elimination of competitive forces in this market facilitates joint anticompetitive behavior. See Salkever 2366--67. In sum, evidence of the increased concentration caused by these acquisitions points toward a finding oflikely harm to competition, all other things being equal!.2! HCA's acquisitions have made an already highly concentrated market more conducive to collusion by eliminating two of the healthiest sources of competition in the market and increasing concentration substantially. But all other things are not equal in this market, and statistical evidence is not the end of our inquiry. In the absence of barriers to entry, an exercise of market power can be defeated or deterred by the entry or potential entry of new firms regardless ofthe structure of the existing market. See, e. Echlin Manufacturing Co., No. 9157 (June 28, 1985). (105 F. C. 410J, We now turn to the issue of entry barriers and conclude that they confirm and even magnify the inference to be drawn from (65J the concentration evidence in this case.

D. Barriers to Entry Complaint Counsel urged below that state certificate of need legislation constitutes a barrier to entry into the Chattanooga acute care hospital market that must be considered in the assessment of the probable anticompetitive impact of these acquisitions. Not surprisingly, HCA argued that such laws do not constitute a barrier to entry in the Chattanooga market. Judge Parker sided with neither. Rather, he found that while the CON requirements do not prohibit entry, they significantly delay it. LD. at 87-88.

Some discussion of the state regulatory environment is necessary in order to evaluate this debate. Both Tennessee and Georgia require certificates of need ("CONs ) for, among other things, the establishment of acute care hospitals, expansion of bed capacity, significant capital expenditures, and changes in the services that they offer. 20 Even excluding HCA-managed hospitals from this calculus, the decrease in firms caused by the acquisitions is from 11 to 9-a drop of almost 20% in the absolute Dumber of competitors 2) Bacb acquisition occurred iD an already highly concentrated market and the increases in concentration from each acquisition were substantial. Our conclusion in this case therefore applies to thc two acquisitions viewcd separately as well a!l collectively. Since both parties generally treated the two acquisitions col1active!y throughout this proceeding, we wih for the !lake of simplicity refer to the acquisitions coliectively. Opinion 106 F.

Tennessee Health Planning & Resources Development Act of 1979 Tenn. Code Ann. Sec. 68-11-101 to -107 (1983 & Supp. 1983), amended by Act to Amend Tennessee Health Planning & Resources Development Act, Pub. Ch. No. 814 (May 25, 1984) ("THPRDA Amendments of1984"); Ga. Code Ann. Sec. 31-61 to -50 (Supp. 1983); see LD.F. 252. Under Tennessee law a CON may be granted for a project only if it is "necessary to provide required health care in the area to be served. . . and wil contribute to the orderly development of adequate and effective (66) health care facilities and/or services. Tenn. Code Ann. Sec. 68-11-106(h)(2), as amended by THPRDA Amendments of 1984. A CON may be granted in Georgia only if the project is "needed." Ga. Code Ann. Sec. 31-60(a) (Supp. 1983); see LD.F. 254. This much is clear--ne cannot simply enter the market. The procedures for obtaining a CON in Tennessee, as well as the similar procedures in Georgia, involve several stages. An application for a CON is first submitted to the Tennessee Health Facilities Commission ("THFC"). After the THFC staff determines that the application is complete, a "review cycle" begins. An Hinterested party, a competitor, has a right to demand a public hearing in the geographic area in which the project is to be located. Tenn. Code Ann. Sec. 68-11- 106(i)(2), as amended by THPRDA Amendments of 1984. The CON application is reviewed by the Tennessee state health planning and development agency, currently the Department of Health and Environment. Tenn Code. Ann. Sec. 68-11- 104(a) (Supp. 1983); see LD.F. 255. The Department of Health and Environment then reports the results of its review to the THFC. Tenn. Admin. Compo ch. 0720-2- (6), (8), and (9). The THFC then makes its initial decision on the CON application. That initial decision may then be appealed by the applicant, persons who have previously participated in the proceeding, such as a competitor, and any other person upon a showing of good cause. Tenn. Code Ann. Sec. 68-11-106(1(1) (Supp. 1983). Upon such appeal, the THFC holds a public hearing, which is supposed to be commenced within 45 days after the appeal is fied. Tenn. Code Ann. Sec. 68-11-106(1)(2) (Supp. 1983). (67) The THFC then decides the appeal within 45 days ofthe hearing. Tenn. Code. Ann. Sees. 68-11- 106(1)(3). The THFC decision on appeal is then subject to judicial review at the instance of any "aggrieved" person. Id. Tenn. Code Ann. Sees. 4- 322(a) (1983); see LD.F. 255. The Georgia CON process is fundamentally similar to that Tennessee. See Ga. Code Ann. Sees. 31-6-3 to -44 (Supp. 1983); Ga. Admin. Compo ch. 272- 04; LD. at 56 n. 14. However, Georgia not only regulates health facilities under its CON program, but also indirectly regulates them through another review of health facility (,:'nit. l PYnpnr1it.lIrp nllr ll::nt, n ::n H"rppmpnt urith thp H'pr1pr g., 361 Opinion Government under Section 1122-of the -Scial Security Act. 42 U. C.A Sec. 1320a-l (West I982); 42 C.F.R 100.101- 109 (1983); Ga. Code Ann. Sec. 31-6-50 (Supp. 1983); Ga. Admin. Compo ch. 272- 01 to 03; I.D.F. 256.

As the foregoing amply demonstrates, there is hardly free entry into the acute care hospital industry in either Tennessee or Georgia. Indeed, the CON laws at issue here create a classic "barrier to entry under every definition ofthat term. In Echlin Manufacturing Co. defined a "barrier to entry" to include "additional long-run costs that must be incurred by an entrant relative to the long-run costsJaced by incumbent firms. " Slip op. at 12; see G. Stigler The Organization of Industry 67 (1968); 4 E. Kintner Federal Antitrust Law 37.4 (1984); 2 P. Areeda & D. Turner Antitrust Law TI409a (1978); R. Bork The Antitrust Paradox 196 (1978); R. Posner Antitrust Law: An Economic Perspective 59 (1976); Baumol & Wilig, Fixed (68) Costs, Sunk Costs Entry Barriers and the Sustainability of Monopoly, 96 Q.J. Econ. 405 408 (1981). We explained that "(tJhe rationale underlying this definition is that low-cost incumbent firms can keep prices above the competitive level as long as those prices remain below the level that would provide an incentive to higher-cost potential entrants. " Slip. op. at 12. If a potential entrant desires to build a new hospital in Chattanooga, he must incur all the costs in time and money associated with obtaining a CON. The cost of starting a new hospital includes not only the start-up costs that any firm would incur to enter the market but also the costs of surviving the administrative process. Incumbents in this market, however, did not incur such costs during initial construction. They have only had to incur those costs for additions made to bed capacity since the enactment of the CON laws a decade ago. Incumbents thus (69) have a long run cost advantage over potential entrants. The result is that market power could be exercised by incumbents without attracting attempts at entry as long as supracompetitive profits are not high enough for a potential entrant to justify incurring all the ordinary costs of starting a hospital plus the significant costs of obtaining a CON.

22 The barrier to entry we are chiefly concerned with here is a barrer to adding n.ew beds to the market; obviously, any firm seeking to enter the market by constructing a new hospital seeks to add bed capacity to the market. It is true that incumbent hospitals must obtain CONs for adding expensive new eqilpment or constructing lOubstantial new facilities. But the evidence is clear that the costs of obtaining a CON for new beds, especially for a new hospital, are substantially higher than the costs of obtaining a CON for other purposes.See, e. Blank 2083 2086 2107 2112; Stern 587, 591-92. For example, Mr. Stern, former administrator of Medical Park hospital, testified that when HCC applied for a COI\' to replace the older facility with tbe new North Park hospitaUt did not apply for additional beds; though he expected the replacement facility to be approved, he did not apply for new beds becfll1.SC he fully expected they would not be approved. Stern 587, 591-92. The situation that incumbent hospitals thus face with respect to their already existing bed capacity is much like a cab driver who was issued a medallion to operate his taxi--themedallion, or franchise right, travels with him between autos. Before any other driver can operate at all, however, he must have a medallion- Likewise, it is the high cost of adding new beds to the market that we arc concerned with here, since new beds would be neces. ary to compete away patients from incumbents engaging in profitable anticompetitive behavior g., Opinion 106 F.

The evidence is clear that those costs are significant in this market. We agree with Judge Parker that because incumbent hospitals can oppose new entry, even an unsuccessful opposition to a CON application may delay its disposition by several years. I.D. at 87-88; I.D. 271-77. Several months are required, as is evident from the Tennessee statute, for preparation of a CON application and for the review process prior to the issuance of the initial decision by the Tennessee Health Facilities Commission. Lamb 241-42; Petruzzi 734-36; I.D. 273-74; see Blank 2088-89. As is equally evident, the CON process can be further delayed by administrative appeals. In Tennessee, administrative appeals of THFC decisions have suffered from a backlog of about a year. Blank 2100; Petruzzi 735-36; RX 898; I.D.F. 275. Obviously, judicial review ofTHFC decisions can create further significant delays. For example, judicial review of the March 27, 1979, THFC grants of CONs for additional beds to Parkridge Hospital and Diagnostic Hospital was not completed until January 6, 1981 more than 21 months later. I.D.F. (70) 275. The delays contributed to a total delay between the submission of CON applications and their final affrmance of more than three years. RX 1051(1)-12); RX 1057(1). (12); RX 1088(5); I.D. F. 275.

Thus the CON process provides existing hospitals in the Chattanooga urban area ample opportunity to significantly forestall the entry of a new hospital or the expansion of an existing hospital within the area. Indeed, the evidence shows that existing hospitals frequently oppose CON applications when they feel competitively threatened. Lamb 231-34; Petruzzi 828-7; CX lI5; CX lI8; CX 127; CXs 129-30; see Blank 2091-93. The potential for delay by competitor opposition and costs attendant to such a delay were recognized, for example, by the board of directors of Red Bank Community Hospital in a meeting held after the initial decision to approve a CON for the relocation of Medical Park Hospital to nearby Hixson. See CX 186. An attorney who represented the Red Bank board before the THFC in that CON proceeding reported to the board that if it appealed the THFC' s decision, there would likely be a delay of about six months before the appeal could be heard, and that while the appeal was pending, Medical Park probably would be unable to obtain financing for the relocation project. Board Chairman Arnold Stulce supported an appeal not only on the hope that it might succeed in blocking the relocation of Medical Park, but also because it would delay the relocation for several months or more-a delay that he recognized would benefit Red Bank. The board authorized the appeal. Id. The potential for delay of a (71) new entrant, where opposition may be over- . _.

361 Opinion whelming, is abundantly apparent.24 .

Even this analysis presumes that a CON can be obtained if the costly process is followed through in this market. In fact, the costs of entry include not only the time and expense involved but also the risk that entry will be denied altogether-a risk incumbents did not have to face upon initial entry. Obviously, if entry is being denied, the costs of entry are sometimes prohibitive. Market participants, including some formerly involved in the CON approval process, overwhelming. ly testified that it is extremely diffcult to obtain a CON for additional beds. E.g., Coddington 482-83; Grimes 3429, 3431- , 3436; Blank 2085 6; Kantanie 1111-12. For example, Ms. Blank, former executive di. rector of the THFC, observed that it is "very diffcult to get the Commission to allow new beds to be (72) constructed" and even more diffcult to convince the THFC to approve new hospitals. Blank 2083 2085-86. Indeed, the last three applications for new bed capacity in the Chattanooga area have all been denied, and the only application for a new hospital in the Chattanooga area since 1974 was denied. 890F. Only one CON for new beds in the Georgia portion of the Chattanooga health planning area has ever been approved. CX 302; 890. In fact, Judge Parker concluded that "(i)t is doubtful. . . .whether health planning authorities in Tennessee and Georgia wil find that any need for additional facilities in the HSA will exist in the forseeable future. " LD.F. 272 n. 16; see Salkever 2330-32. In sum, it is not merely the costs of obtaining a CON that a potential entrant faces, but the significant risk of being denied entry once those costs have been incurred. This risk, which incumbents did not have to face when building their hospitals, in (73) effect raises the costs of entry a significantly greater amount. As a result, many potential entrants may decide not to even attempt entry. See Salkever 2330. Indeed, the evidence shows that CON regulation has had a deterrent 2. HCA offers calculations purporting to show that the average time it takes to obtain a CON for additional beds througb the administrative process" is less than five months. RAB at 43-4. IICA ignorcs tbe time that it takes for judicial review when projects are of suffcient competitive importance to warrant an appeal by competitors HCA furthennore does not distinguish between those approved applications that were of suffcient competitive significance to warrant opposition and those that we;e not. None of the proposals by Chattanooga urban area hospitals that involved increases of more than 16 beds, except those automatically "grandfathered" under the Tennessee statute, were approved without judicial review. CX 890B-F. It appears that proposals for additional bed capacity in the Chattanooga urban area may be promptly approved if they are competitively insignificant, but any proposal that would threaten the profitability of incumbent hospitals would likely face years of delay. 25 HCA claims that since the introduction of the Tennessee CON law, 622 "new " heds have been approved by the THFC for the general Chattanooga area. RAnB at 24. Of those beds, more than two-thirds (455) were "grandfa. thered"; that is, the applications were required to be approved by the CON statute because their development had already begun before the CON laws went into effect.See Tennessee Health Facilities Act of1973, Pub. Ch. No. 257 , 1973 Tenn. Pub. Acts 996, 1003-4 (repealed 1979); CX 890B.D; RX 1088(1)-(5). Of course, grandfathered beds arc irrelevant to our analysis, since they do not bear on the likelihood of future entry. (Moreover, approval of 121 of the remaining 167 beds was substantially delayed by appeals and subsequent litigation.See discussion supra this section; RX 1088(5)-(6).) When the "grandfathercd" beds are disregarded, the statistics show that there has been no new entry into the Chattanooga market since the CO:: laws were enacted, more applications for new beds by existinl; hospitals have been denied than approved, and there has been very litte growth in capacity in the Chattanooga area. CX 890; RX 1088; CX 646A. Opinion 106 F.

effect in the Chattanooga market. See, e. Stern 587, 591- , discussed supra n. 22; Blank 2101; Salkever 2320, 2326-27. In response to this evidence, HCA argues that the empirical studies in the health economics field do not, in general, support a finding that CON regulation has had a "significant" efiect on the addition of new hospital beds. RAB at 41.26 We find the cited studies ambiguous. First of all, at least two studies have concluded that CON laws are" barrier to hospital bed construction.2 Second, the studies cited by HCA generally attribute expansions in investment and bed capacity to efforts to secure "grandfathering" to escape CON requirements, or (74) construction in anticipation of CON. Salkever 2573- , 2581; Harris 3931, 3934-5. Rather than support a finding that CON has had no negative impact on additional bed capacity, the conclusions are predicated on a barrier to entry that is anticipated to flow from CON regulation. Indeed, our concern lies in the likelihood of future entry and expansion in response to a joint exercise of market power by Chattanooga hospitals; such studies shed no light on this issue. Moreover, two ofthe studies relied upon by HCA focus on the efiect of CON regulation on overall hospital investment or costs, and do not examine whether CON laws restrict the growth of hospital bed capacity. Studies aside, it is important to note that the last three applications for new bed capacity and the only application for a new hospital in the Chattanooga area since 1974 were all denied. CX 890F. While Judge Parker points to several examples of where a CON has been granted for hospital construction (LD. F. 260; see RAB at 40), they either predated CON regulation or were (75) merely replacements for existing hospitals with no increase in bed capacity. Interestingly, HCA's executive leadership recognizes the barrier to See P Joskow Controllng Hospital Costs (1981); Hellnger The Effect urCertificate of Need Legislation on Hospital Investment 13 Inquiry 187 (1976); Sloan & Steinwald Effects of Regulatioll on Hospital Costs and Input Use 23 Journal of Law and Economics 81 (1980); Policy Analysis, Inc. Evaluation of the Effects of Certificate Need Prugrams (1980).

, Salkever & Rice, The Impact of Certificate of Need Controls on Hospital Investment 54 Milbank Memoria! Fund Quarterly 185 (1976); Joskow The Effects of Competition and Regulation on Hospital Bed Sllpply and the Reservation Qu.ality of the Hospital 11 Dell J. Econ. 421 , 440, 446 (H180);see Salkever 2318- , 2554-7, 2560 2576-77 28 See Hellinger supra D 26; Sloan & SteiDwald su.pra n. 26; Policy ADalysis, Inc. supra n. 26; I.D.F. 268. The ODe study that arguably does not, the 1981 Joskow study,supra n. 26, did Dot examine effect.' OD hed growth of CON laws. In fact, a.' noted, a 1980 study by Professor Joskow concluded that CON programs have the effect of restricting hospital bed capacity Joskow The Effects of Competition and Regulation on Hospital Bed Supply and the Reservation Quality of the Hospital Ii Bell J. Econ. 421 , 440, 446 (1980); see Salkever 2319, 2576-77 :t HelJnger supra n. 26; P. Joskow Controlling HospitalCosts (1981); see Salkever 2578-9; Harris 4163-6. CON laws can limit the growth ofhed capacity even if they do not affect overall levels ofiDvestment.See Salkever 2557 30 l'arkridge Hospital opened before the CON laws went into effect in 1973(see LD.F. 260), development of East Ridge Community Hospital was begun before the CON laws went into effect and was therefore "grandfatherI'd, see Tennessee Health Facilities Actofl973, Pub. Ch. No. 257 , 1973 Tenn. Pub. Acts 996, 1003-4 (repealed 1979); ex 890B; RX 1088(2), and Downtown General Hospital, Red Bank Community Hospital, and North Park Hospital were replacements for existing hospitals with no additional beds, thus not adding to the existing capacity in the market ex 890B, C, E; RX 1088(2). (7). Indeed, both Downtown and Red Bank sought additional beds, which were . . .

361 Opinion entry quality of CON regulation In an interview published in a 1981 issue ofthe Harvard Business Review Dr. Thomas J. Frist, Jr., HCA' president and chief executive offcer, observed: Federal and state health planning laws have erected formidable barriers to entry into the hospital industry by creating literal monopolies for physicians and hospitals, If the health planning Jaws state that a community can have only one cardiac surgery program, they might as well give the physician who performs that surgery an exclusive franchise. It's the same for hospitals.

CX 123D; see also LD.F. 258. Dr. Frist also observed that such barriers to entry benefit HCA because they "protect our hospitals from competitors who might build new facilities and take away our market. We know what the market for a particular institution is going to be like 5 or 10 years down the road. " CX 123C. He noted further that "regulation severely restricts new hospitals (76) from entering our markets. CX 123 F.

In its 1982 Form lO- , fied with the Securities and Exchange Commission, HCA states that "state certificate of need laws and Public Law 92-603 place limitations on the Company s and its competitors' ability to build new hospitals and to expand existing hospitals. " CX 13S. HCA's 1982 Management Plan for its Mergers and Acquisitions Division observed that "the HSA concept (certificate of need process) continues to preclude or slow down most development of new facilities. . . ." CX 125D. Its 1980 Corporate Strategy Statement noted that CON laws restrict new entry and expansion of existing facilities. CX 221Z-8.

Finally, HCA executives have in the past ascribed a "franchise value" to hospitals as a result of CON regulation. David G. Williamson, Jr., Executive Vice President of Domestic Development for HCA has referred to the ICfranchise value" and franchise type protection of hospitals that is created by certificate of need regulation. CX 124H The very existence of a "franchise value" is a tacit admission of the (77) existence of barriers to entry.

Accordingly, we agree with Dr. Salk ever that CON laws pose a very substantial obstacle to both new entry and expansion of bed capacity in the Chattanooga market. See Salkever 2321- , 2325-30. Indeed the very purpose of the CON laws is to restrict entry. Existing Chattanooga area hospitals appear virtually insulated from new competi- 1L This franchise value was quantified on one occasion by William G, White, RCA Senior Vice President of Acquisitions, when he attached a value of$8 milion, or $20 000 per bed, to the certificates of need for a total of 400 psychiatric hospital beds held by HCC which HCA was seeking to acquire at the time. ex 136; I.D.F. 258. A dollar value for il eon was also designated when HCA otTered to purchase the bed complement of Medical Park Hospital-without the land, building or its contents-for about$400 OOO in late 1976or early 1977 so that HCA could add those beds to Parkridge s bed complement. Stern 613; LD.F. 258. And Jonathan Grimes, administrator ofHCA' s Xorth Park Hospital, urged HCA to consider acquiring Red Bank Community Hospital in order to acquire the rigllt to operate beds that could be transferred to North Park. CX 420A; LD. F. 258 Opinion 106 If.

tion in the short term, and have an absolute cost advantage that extends into the long term. Therefore, any harm to competition that could be generated by profitable collusion or interdependent behavior in this market wil unlikely be deterred by threatened or actual new entry.

E. The Nature and Likelihood of Anticompetitive Behavior in the Chattanooga Hospital Market 1. The Nature of Anticompetitive Behavior Because HCA denies that anticompetitive behavior among Chattanooga urban area hospitals is likely, it is useful to consider the likely forms that any anticompetitive behavior would take. Profitable collusion could take a number of different forms and restrict price or non-price competition or both.

Some of the most likely forms of collusion between hospitals would involve collective resistance to emerging cost containment pressures from third-party payors and alternative providers. For example, joint refusals to deal with HMOs or PPOs may occur, or perhaps joint refusals to deal on the most favorable terms. Conspiracies to boycott certain insurance companies . that are generating price competition may occur. Utilization review (78) programs may be also be resisted. Hospitals could concertedly refuse to provide the information desired by third-party payors-information that would otherwise be provided as hospitals vie to attract the business of tbose payors and their subscribers. The result of any such boycott would be to raise prices reduce quality of services or both. (79) Collusion among Chattanooga hospitals could also inhibit advertis- J2 Without Bpeculating about the merits of the following cases, we find the allegations instrnctive with respect to the array ofboycott. of particular third.party payor!! that are possible in this industry-Ball Memoria! Hospital Inc. v. Mutual! Hospital Insurance, Inc. 603 F.Supp. 1077 (S.D. Ind. 1985) (court notes allegation in countersuit that hospitals conspired to boycott a PPO);Ohio ex rel. Celebrezze v. Greater Cleveland Hasp. A. 1983-2 Trade Cas. (CCH) 685 (N,D. Ohio 1983) (consent decree).summarized at (July-Dec.) Antitrust & Trade Reg. Rep. (ENA) No. 45, at 733 (Nov. 3, 1983) (allegation that trade association organized 90% of area s general hospitals to refuse to deal with non-Blue Cross reimbursers such as health maintenance organizations except lit prices higher than Blue Cross rate); United States Ii. North Dakota Hasp. Ass Civ. No. 82-131 (D- D. fied Aug. 25, 1983), complaint summarized at4 Trade Reg. Rep. (CCH) 083, at 53 628 (allegation that trade association organized hospitals to refuse to accept contracts of Indian Health Service unless contracts maintained level of charges at rates equal to hospitals' usual private rates);United States I). Montana Nllrsing Home Ass 1982-2 Trade Cas. (CCH) 64,852 (D- Mont. 1982) (consent decree),complaint summarized at 4 Trade Reg. Rep. (CCI-I) 45,080, at 53 421 (allegation of conspiracy to raisc prices by refusing to enter into Medicaid contracts with state except on Association s terms); United States IJ- South Carolina Health Care Ass 1980-2 Trade Cas. (CCH) 11 63 316 (D. C. 1980) (consent decree),complaint summarized at (July-Dee) Antitrust & Trade IWg. Rep. (BNA) No. 929, at A-lO (Aug. , 1979) (allegation of conspiracy to raise rates paid under Medicaid program by jointly negotiating with state); Maine IJ. Anesthesia Prof Ass No. CV 84-250 (Super. Ct. Kennebec City_, consent decree fied June 28, 1984), summarized at Trade Reg. Rep. (CCH) No. 658 (July 10, 1984) (allegation of collective refusal to accept Medicare reimbursement as payment in full and collective refuslIl to participate in Blue Shield progrllm);Pennsyluania u. Pennsyluania Dental Ass Cill- No. 81- 1187 (M.D. Pa. consent decree fied Sept. 3 , 1982), complaint summarized at (July-Dec-J Antitrust & Trade Reg. Rep. (BNA) No. 1037, at D-l (Oct- 29, 1981) (aHegation ofconspirllcy to boycott certllin third-party payors, particularly Blue Shield);Texa. Dental Association 100 F. C. 536(1982) (consent order) (allci!atinn nfnI1Rd.ivp rp,.i j;Jnr,,, t.n IJt.;1iz ;nn rpv;pw) . _._ _u_-- 361 Opinion ing and other forms of markehng of hospital prices and services. Chattanooga hospitals have become more openly competitive in the area of advertising both prices and range and quality of services. See supra section V.B. Hospitals could agree, for example, not to advertise or not to advertise in certain media, agreements which may be facilitated by the health care industry s traditional disapproval of advertising. See Harris 4052. Patients, their employers and their insurance companies would be substantially harmed by the denial of information useful (and necessary) for choosing among hospitals should advertising or other forms of marketing be restricted. See American Medical Association 94 F. C. 701 (1979), aff'd as modified 638 F. 443 (2d. Cir. 1980), aff'd by an equally divided Court 455 U.S. 676 (1982); United States v. Gasoline Retailers Association 285 F.2d 688 (7th Cir. 1961) (agreement not to advertise prices except by sign on gas pump found unlawful).

Quality competition itself might also be restricted. For example, the group of hospitals in a relevant market might agree to staff their wards with fewer nurses yet continue to maintain current rates for inpatient services. Patients would be harmed by the resulting drop in quality of services without any compensating reduction in price of services. Colluding hospitals (80) in the market, however, would profit from their agreement by cutting costs without cutting revenues. Again, hospitals could accomplish anticompetitive ends not only by fixing staff-patient ratios but by agreeing on wages or benefits to be paid certain personnel-for example, laboratory technicians. Indeed wage and salary surveys are common in this market. Lamb 210. The result would be the sam&-to hold the cost of inputs down with probable harm to the quality of output of health care services.33 Hospitals could also agree not to compete for each other s personnel or medical staff. Indeed, some Chattanooga urban area hospital firms have already engaged in such behavior. See discussion infra section 2 of the Red Bank-HCC agreement.

Moreover, under certificate of need legislation, the addition of new services and purchases of certain kinds of new equipment require a demonstration of need for the expenditure, and the existence of need is determined in part by the facilities already provided in the community. See supra section V. ; LD.F. 254. It would thus be to the advantage of competing hospitals to enter into agreements among themselves as to which competitor will apply for which service or for which piece of equipment. New bed applications might also be allocated among existing hospitals. The process would be similar to bidrigging construction contracts-every firm is guaranteed part ofthe pie at the expense ofthe customer receiving the best service at the (81) J. We also note the harm to hospital personnel that would result from such agreements. . . .

Opinion 106 F.

lowest price. In return for agreements not to fie competing applications, administrators can support each other s CON applications thereby assisting the applicant to demonstrate need. See, e. 96C-D. In fact, there is evidence that Chattanooga area hospitals have recognized the advantage of cooperating on CON matters and that this type of allocation has occurred in Chattanooga. See discussion infra section V.E.2 of the Red Bank-HCC agreement. Sucn market division by private agreement would save hospitals the expense of applying for numerous CONs but may harm the quality of care that would be available to patients were CON approval sought independently by each hospital with reference to its own merits and expertise. Concerted opposition to the CON application of a potential new entrant is yet another manner in which Chattanooga hospitals could successfully collude. Attempts by hospital administrators to garner opposition to threatening CONs are common. E.g. Smith 1981-82; Stern 571, 575; LD. F. 296; see supra section V.D. If the new entrant would inject new price and quality competition into the market, consumers wil be harmed when joint opposition is successful (and uncoordinated opposition would not be successful) in barring or stallng the desirable entry.

Anticompetitive pricing behavior could also take several forms. For example, hospitals could work out agreements with respect to pricing formulas. See F. Scherer Industrial Market Structure and Economic Performance 170 (2d ed. 1980). Chattanooga hospitals commonly compute percentage increases when raising rates or proposing rate increases to Blue Cross (CX 172-(82)77; CX 382-96), including different percent increases for different departments of the hospital. See 182-96; CX 322-93; CX 315J-L; CX 316D-F; CX 326H. They could thus base collusion upon percent increases in their prices. An example of data hospitals have that could serve as a basis for collusion is provided by one HCA hospital planning document:

Ancilary increases are budgeted at a 7.7% increase for 1982. This percentage wil vary among departments depending on the competitiveness of our current price with others in our area. Each ancilary department gathered price information from area facilities before building a price increase into their revenue budgets. CX 1482-75.

Hospitals could also successfully collude with respect to price by agreeing not to give discounts to businesses, insurers and other group purchasers such as HMOs and PPOs. Salkever 2361--2. Moreover, an agreement on the percentage discount to be offered or the group or groups of purchasers to receive a discount can be reached, even without an agreement on the base price to which the discount is applied. 361 Opinion Tenn. 1956), aff'd per curiam 352 U.S. 991 (1957). Because in the hospital industry customers are grouped into separate insurance plans, collusive arrangements can be established for certain groups of customers. Acute care hospitals can collude as to a specific alternative provider or health insurance company or plan, while competing in their usual fashion with respect to the rest of their buyers. Salkever 2388-89 2671-75.

In sum, we conclude that hospitals compete in a myriad of(83) ways that could be restricted anti competitively through collusion.34 Thus it appears that a merger analysis in this case need be no dine ent than in any other case; market share and concentration figures, evidence of entry barriers and other market evidence taken together appear to yield as accurate a picture of competitive conditions as they do in other settings. Nevertheless, although HCA concedes that many of the above described forms of collusion could occur, the heart ofHCA' case is that collusion in this market is inherently unlikely, and to that contention we now turn.

2. The Likelihood of Anticompetitive Behavior Section 7 of the Clayton Act prohibits acquisitions that may have the effect of substantially lessening competition or tending to create a monopoly. Because Section 7 applies to "incipient" violations, actual anticompetitive effects need not be shown; an acquisition is unlawful if such an effect is reasonably probable. , Echlin Manufacturing Co. No. 9157, slip. op. at 8 (June 28, 1985) (105 F. C. 410); American Medical International, Inc. No. 9158, slip. op. at 17-18 (July 2, 1984) (104 F. C. 1J.

The small absolute number of competitors in this market, the high concentration and the extremely high entry barriers indicate a market in which anticompetitive behavior is reasonably probable after the acquisitions. The fact that industry members recognize (84) the enormity of entry barriers makes collusion even more probable. In addition, hospital markets have certain features that evidence a likelihood of collusion or other anticompetitive behavior when they become highly concentrated. See Salkever 2362, 2382. First, price elasticity of demand for hospital services is very low (Salkever 2384; Harris 4237-39), which makes anticompetitive behavior extremely profitable and hence attractive. See Salkever 2384; R. Posner Antitrust Law: An Economic Perspective 48 (1976). Second because consumers of hospital services cannot arbitrage or resell them as is often possible with goods, discrimination among different 34 We by no means intend the above analysis to be an exhal1stive listofthe ways in which hospitals might behave anticompetitively. We do intend it to be iil18trative in view of HCA's claim that omccessfl1l collusion is highly I1nlikely in this market Opinion 106 F.

groups of consumers is possible. That is, collusion may be directed at a certain group or certain groups of consumers, such as a particular insurance company, without the necessity of anticompetitive behavior toward other groups. See supra section V.E.!. Third, the traditions oflimited price competition and disapproval of advertising (see Harris 4052) provide an incentive for future anticompetitive restrictions of those activities. Fourth, and in the same vein, the advent of incentives to resist new cost containment pressures may create a substantial danger of hospital collusion to meet those pressures. Salkever 2393- 2705-06. Fifth, the hospital industry has a tradition of cooperative problem solving which makes collusive conduct in the future more likely. Hospitals have historically participated in voluntary health planning in a coordinated manner, and along with other professional organizations, such as medical societies, have (85) participated in developing joint solutions to industry problems. Salkever 2382-84. Moreover, the history of interaction among Chattanooga hospitals supports a conclusion that anticompetitive behavior, whether through interdependent behavior or express or tacit collusion, is reasonably probable in the highly concentrated market created by these acquisitions. The most convincing evidence of the facility with which such collusion could occur is a blatant market allocation agreement executed in 1981 between Red Bank Community Hospital and HCC. I.D.F. 293-94; CX 96; Smith 1987-88; Stern 588-89. The parties actually signed a contract under which Red Bank agreed that for a period of three years it would not "fie any application for a Certificate of Need for psychiatric facilities or nursing home facilities." CX 96. Moreover, the parties agreed that they would not compete for each other s personnel and medical staff during that time period, and that they would not oppose each other s CON applications in certain areas. CX 96C-D. Such an overt agreement to refrain from competition at the very least demonstrates the predisposition of some firms in the market to collude when it is in their interest; at worst it shows a callous disregard for the antitrust laws.35 (86) The Red Bank-HCC agreement is not the only evidence ofthe potentially harmful interdependent behavior of Chattanooga hospitals 35 HCA argues, howevcT, that the agreement between Red Bank and BCe can only be viewed in the context of the fact that it involved settlement ora dispute; Red Bank agreed to drop its opposition to lICC's effort to relocate Medical Park Hospital in exchange for RCC's withdrawal of its threat to bring.m antitrust suit against Red Bank. HCA therefore contends that the agreement docs not show a marketwide predisposition toward collusion. RAB at 26 We disagree with HCA's assessment and find the Red Bank-BCe agreement very disturbing. We sce no reason why such a dispute resolution had to imlude terms dividing markets. Even if such blatant anticompetitive ag-eements are uncommon, the fact that a contract like this one was drafted at all shows the predisposition of so me Chattanooga market participants to combine to reduce competition. We find nothingredeeming in an ag-eement to refrain from competing for productive inputs and an agreement not to enter the psychiatric or nursing home facilities market The fact that HCC and Red Bank (a non-profit hospital, we might add) would settle their dispute so openly in an anticompetitive manner is leas than comforting. In tact, we t.think it is an indication of the haste g. g., t1Uti.tlTAL L:Uttl . ur AlV1r ttll:A DU.l 361 Opinion however. Joint opposition to CON "applications is often discussed among hospital administrators. I.D. F. 296; see supra section V.D. In another example of cooperation, the administrators of Diagnostic, Erlanger, and Parkridge at one time cooperated to present a united front to health planning offcials with regard to the need for new beds. CX 238A; I.D. F. 296.

Furthermore, a basis for collusion is provided by the exchanges of rate, salary and other competitively sensitive information that occur in this market. In addition to the rate surveys detailed above, which were chiefly conducted for the purpose of setting the surveyor s prices (e. Kantanie 1069-70), more casual exchanges of price, wage and other information over the telephone, at social gatherings and, at meetings of hospital administrators commonly take place. See, e. Lamb 209, 230-32, 247-49; Stern 564, 577, 638-9; Smith 1958, 1961; CX 239A. For example, Mr. Smith, former administrator of Red Bank (87) hospital, testified that "(t)ypically what wil happen is that the manager of radiology at another hospital wil call the manager of our radiology department and ask what we are now charging for and list off seven or eight tests and, again, it's through these informal relationships that are out there in the community that this kind of in formation is generally obtained. " Smith 1961. In some instances, rate inquiries have concerned not only current but prospective prices. CX 184G; CX 239A; Kantanie 1134; Lamb 206. Such exchanges of information are evidence of a tendency for collusion in this market. See R. Posner Antitrust Law: An Economic Perspective 65-66 (1976). HCA argues, however, that collusion in this market is too diffcult or costly to ever arrange or sustain, or that if not impossible, it poses no substantial likelihood of succeeding. Because collusion is so unlikely, HCA contends, no substantial harm to competition is likely as a result of these mergers.

As a threshold contention, HCA argues that the Chattanooga market is less concentrated than other health care markets that are vigorously competitive " and that "the fear ofa lessening of com pet tion in Chattanooga is plausible only if its market structure is more concentrated than the communities where Complaint Counsel assert competition is active and intensifying." RAB at 18-19. Its chief evidence that such markets are vigorously competitive " however, is supposed admissions by Complaint Counsel to that effect. See id. 18-19. This is a mischaracterization of the evidence and Complaint Counsel' s argument. (88) The contentions of Complaint Counsel to which HCA refers reflect evidence that competitive forces in the health care industry are on the rise. See id. It is true that the undisputed evidence shows that more vigorous competition, including more direct price competition, is , Opinion 106 F.

emerging in the health care industry, but it is a fallacy to conclude that growing competition in health care markets means that these acquisitions pose no threat to that competition. In fact, it is just that emerging competition that must be protected from mergers that facilitate the suppression of such competition. See Salkever 2393-94. Nothing in the record supports an assertion that highly concentrated hospital markets are performing at (89) optimal competitive levels. We therefore reject HCA's assertion.

a. Non-profit Hospitals and the Likelihood of Collusion HCA contends that the most fundamental difference between hospitals in Chattanooga is that several ofthe hospitals are "non-profit" institutions. Economic theory presumes that businesses in an industry are profi-maximizers and that output wil be restricted in pursuit of profits. Non-profit hospitals, the argument goes, have no incentive to maximize profits; rather, they seek to maximize Houtput" or the number of patients treated. RAB at 23-24. HCA contends that nonprofit hospitals may have other goals as well, such as providing the most sophisticated and highest quality care possible, or pursuing religious or governmental goals. RAB at 24; Harris 3860. In short, HCA argues that collusion would not occur because the (90) "for-profit" and non-profit" competitors have no common goal. RAB at 24. We disagree that non-profit hospitals have no incentive to collude with each other or with proprietary hospitals to achieve anticompetitive ends. First, we note that non-profit status of market participants is no guarantee of competitive behavior. For example, the Supreme Court in the NCAA case recently found unlawful anticompetitive conduct, including output restriction, by a combination of non-profit entities. National Collegiate Athletic Association v. Board of Regents of University of Oklahoma 104 S.Ct. 2948 (1984). The Court recognized that non-profit institutions may seek to maximize revenues in order to further their objectives. 104 S.Ct. at 2960 n. 22. In the words of HCA's own expert all hospitals have to worry 3r. Dr. I-ranis, HCA's expert, te tified that the degree of concentration in the Chattanooga MSA is lower than the average for comparable MSAs. HCA concludes from thi8that "these acquisitions did not make the Chattanooga market structure more concentrated than other hospital markets which complaint counsel argue are vigorously competitive." RAB at 19. As we have stated, complaint counsel do not argue that such markets are vigorously competitive, but only that competition is growing rapidly in hospital markets. Since HCA's argument is expres.ly dependent upon thi mischaract.erization of the evidence, its reliance upon Dr. Harris' study must be rejected. Moreover, HCA may not daim that the Chattanooga MSA i an inappropriate "section of the country" ill which to measure the effects of these acquisitions and then credibly base its analysis of competitive effects upon a measure of concentrat.ion in the Chattanooga MSA. J7 HCA further argues that a philo30phicaJ gulfhetween the established non.profit hospitals in Chattanooga and the newer proprietary hospitals has created an air of hostility in their dealings that predudes any kind of collusion. RAn at 27. One ilustration of this hostility is to be found in effort made by tbe non-profit Memorial Hospital to encourage the FTC to challenge theae mergers, IlCA states.Id. We do not find this fact to negate the probability !- hnQn;bk Uf"" J,1 l1"iI.. ifit. i" t.hp;r i"tprP"t, t.o rio RO. Comoetitors mav vie for a e'eaterslice of market .

361 Opinion about making money, if only because it is a means to other ends. Harris 3861. Indeed, HCA recognizes that non-profit institutions wil normally wish to generate some "operating surplus" in order to purchase new equipment or replace old equipment. RAB at 24; see Salkever 2389-90. If their net revenue is too low, they wil seek ways to increase it. (91) Collusion is one such way and an incentive for collusion to raise prices, restrict output or reduce quality therefore exists. Because all hospitals must earn such "surpluses, their competitive behavior will be similar. See Lamb 251; CX 143 at 15; CX 528G. HCA' own documents recognize that nonprofit hospitals resemble proprietary hospitals in the way they do business. CX 527B C"(t)here s no such thing as a non-profit hospital. . . . ); CX 243C; CX 209Z-IO. Dr. Salkever testified that non-profit hospitals "can find attractive ways from their point of view to make use of the monopoly rents which would appear as returns to stockholders in the case of the for-profit hospitals." Salkever 2390. Anticompetitive behavior effecting monopoly rents may permit non-profits to maximize other goals such as quality of care, amount of un reimbursable care, experimentation, development of highly specialized services, prestige, and so forth. See Salkever 2563-67. Our concern under Section 7 is not what economic profits would be used for by the parties exercising market power but simply the fact that such an exercise of market power would be to the detriment of consumers.

In addition, administrators of non-profit hospitals may seek to maximize their personal benefits and comfort through what would otherwise be known as profi-seeking activity. For instance, though doubling an administrator s salary may not legally constitute the accumulation of "profit" for a non-profit hospital, it will be necessary to take that money from revenue net of other costs. Any action that is "profi-maximizing" would (92) increase the pool of funds from which such distributions could be made. Likewise, maximizing profits would be consistent with obtaining nicely furnished offces, lavish business trips and so forth. See Salkever 2563-B5. Finally, even assuming that the goals of non-profits sometimes depart from pure profit maximization, often those goals are suffciently divergent from the ends of competition to make anti competitive collusion desirable. There is no guarantee that non-profit hospitals seek competitive levels of price, quality and output for their customers. For example, as Judge Parker found, because there is such great public pressure to reduce hospital costs, it would be to the mutual advantage of both for-profit and non-profit hospitals collectively to resist such pressures any time their goals are inconsistent with cost containment. See J.D. at 88. HCA itself contends that government efforts at cost containment have come about because of the unique incentives .

Opinion 106 F.

in the hospital industry to maximize quality of treatment. RAB at 44-6. To the extent that hospitals are quality maximizers that desire to experiment, obtain the best equipment possible and so forth, effcient reductions in spending for hospital services, effected through the efforts of public or private third-party payors or otherwise, could hurt those goals. Nonprofit hospitals thus have an incentive to collude if by doing so they can resist such pressures. Collective resistance to third-party payors' utilization review programs and efforts to obtain detailed cost and quality (93) data from hospitals to use in "comparison shopping, for example, would constitute anti competitive behavior beneficial to non-profit hospitals and harmful to insurance companies and their subscribers. See supra section V. 1. With such information available, competing hospitals may be made to charge lower prices at an output level and mix that purchasers desire. Detailed hospital-specific cost and quality data can be used to seek out the best quality health care services for the dollar among competing hospitals. Both non-profit and for-profit hospitals however, have an incentive to resist the provision of such information, since it is inconsistent both with profit (or "surplus ) maximization and with continued control over output for "quality" or other reasons. Even non-profit hospitals wil not want to lose their control over competitive variables, such as the flow of information, that could change their preferred output mix.

Any cooperative resistance to such pressures, however, is inconsistent with the requirements of competition, even if the goal of the nonNprofit hospitals is couched in terms of nmaximizing output" or maximizing quality." Ifhospitals have a definition of output maximization or quality maximization that is different from the competitive levels of output and quality, then any concerted action is furtherance of their goals wil be anticompetitive. In short, any behavior which deviates from competition is harmful to consumers, whether the motives are profi-maximization or maximization of some other market variable. 38 (94) Moreover, specific characteristics of non-profit hospitals in this market make anticompetitive behavior a reasonable probability. First, two of the six non-profit hospitals in the Chattanooga urban area, Downtown General and Red Bank, are managed by HCA. See supra section V.A. Indeed, the non-profit hospitals in the Chattanooga :! For example, HCA relies upon the testimony orMr. Furrey, thesenioT lay administrator of Memorial Hospital which supposedly "demonstrates" the commitment of Memorial to providing hospital care at the lowest possible price. RAE at 24; Furrey 1525. Such a motivlItioD would not necessarily keep prices at a competitive level but only at the level that Mr. Furrey believes is the "!owest possible" within the constraints of the variables he seeks to maximize. If those constraints are not market demand constraints, then his well-intentioned motivation wil not serve the ends of competition. And to the extent those cons:traint. diverge from competition, incentives to collude to reach those anticompetitive ends are present. Oftn a price above a competitive level, therefore, will be in the 361 Opinion area managed by HCA are considered to be proprietary hospitals by their competitors and others in the industry. Lamb 249'-51; Smith 1951-52; Blank 2067-68.

Second, two major non-profit hospitals, Erlanger and Tri-County, have a tremendous incentive to participate in price collusion. Erlanger has sole responsibility lor unreimbursed indigent care in Hamilton County. Lamb 115. More than 100 of Erlanger s beds are generally in use at all times for indigent care. Id. at 114-115. Because it must subsidize unreimbursed care out of the rates charged to paying customers, Erlanger cannot compete effectively through price cutting. Lamb 176-77. Erlanger s rates are 50 dollars per day or 10% higher than they would be if such cross-subsidization between paying and non-(95)paying patients were not necessary. Lamb 113-14. Because it cannot price below a level that covers the direct costs it incurs lor indigent care, Erlanger would in fact benefit from a decrease in price competition through interdependent behavior. Salkever 2370-72. The same analysis applies to Tri-County, which must provide care for indigent residents of Walker, Dade and Catoosa counties in Georgia and shift costs from non-paying to paying patients. CX 118D. Because Erlanger and Tri-County would be hurt by price competition through such methods as large discounts to third-party payors, they have every incentive to join in a collusive scheme that diminishes price competition.

In addition, because oftheir inability to markedly cut prices, it may even be that successful anticompetitive collusion could be carried out by HCA hospitals and the two other proprietary hospital firms in the market without the participation ofthe independent non-profit hospitals. If HCA could convince the other proprietary hospitals in the market, Metropolitan and East Ridge, to agree to an anticompetitive scheme respecting prices, Erlanger and Tri-County would not have the ability to undercut it as long as the price was set below their price floors created by the cross-subsidy between paying patients and indigents. See Salkever 2370-72. And again, they would have little incentive to discourage collusion that raised prices up to that price floor since market prices below that level could draw business away from them.

In the event of such collusion, Memorial and the tiny, low quality Wi1dwood are left as the only non-profit hospitals not (96) managed by HCA to defeat the collusive scheme. It appears unlikely in the face of the expense and diffculty of obtaining a certificate of need for expansion of bed capacity that these two hospitals themselves could provide enough beds to successfully deter harmful collusion. See discussion supra section V.D. To expand capacity the hospitals would have to believe they could garner enough revenue to justify the cost 506 FEDERAL TRAm; COMMISSION DECISIONS Opinion 106F.

of seeking a CON-and do so in face of the opposition ofHCA and the other colluders.

Finally, non-profit status has not precluded hospitals in this market from engaging in anticompetitive behavior or from exchanging sensitive competitive information with other hospitals. See, e. Lamb 206-11. For example, the Red Bank Board of Directors itself agreed to the market allocation scheme with HCC discussed above. See supra this section.

In sum, we conclude that non-profit hospitals often have incentives to act as profi-maximizers in an effort to further goals that can only be realized by increasing net revenues. We also find that even when the goals of non-profit hospitals diverge from profit-maximization, the evidence shows that they are not necessarily commensurate with the ends of competition that Section 7 is intended to protect. Divergent maximands between competitors can often produce the same anticompetitive ends. And lastly, the evidence shows that non-profits in this market have an incentive to collude and have engaged in cooperative behavior with other hospitals. (97) b. Purported Obstacles to Successful Coordination Relying entirely upon the testimony of its expert, Dr. Harris, HCA argues that even if hospitals in Chattanooga were inclined to collude the administrators of those hospitals would find it diffcult to reach anticompetitive agreements or understandings, or to sustain them if they ever were reached. This is so because the ideal market circumstances for collusion are not present e. where manufacturers are sellng "some simple, relatively homogeneous good, well characterized by a single price. " Harris 3865-66. HCA contends that hospital services are heterogeneous and influenced by a variety of complicating factors. Hospitals provide a large number of varied medical tests and treatments and each patient receives unpredictable personalized service the extent of which is determined by physicians. RAB at 30. Moreover, HCA claims, costs and demand vary between hospitals. And because the dominant avenues of competition relate to the quality of medical care and patient amenities, hospitals would have to agree on a whole host of things to eliminate competition in a manner suffcient to earn monopoly returns, it is alleged. HCA argues that "(aJny attempts to reach explicit understandings would require a long agenda, and any hypothetical agreements would have to be lengthy and complicated (this hospital may buy such and such equipment, but no more, unless such and such happens, while that hospital may purchase the same or some other equipment, and so forth for all market participants; such and such inducements may be offered to f981 phvsicians. but no others; and so forth and so on). r1var 1111.. L,vn.r. VI' 11uJ.c.J.UL,n.

361 Opinion RAB at 32. Similarly, HCA cOficludes' with respect to restriction of price competition: "To achieve any price uniformity or pattern with regard to the ultimate variable of interest-the total cost of a hospital stay-hospital administrators would have to find some way to bring (individual service prices) under collusive control. Among other obstacles, hospital administrators would have to find some way to force or induce physicians to treat and test all patients similarly situated (however that might be defined) in the same way. . . ." RAB at 35 n.28. HCA' s analysis of the likelihood of collusion distorts competitive reality. HCA would have us believe that the world of possible collusion is limited to complicated formulae concerning every aspect of hospital competition-that market power can only be exercised with respect to the entire cluster of services that constitutes the acute care hospital market through a conspiracy fixing the overall quantity or quality of treatment running to each patient in the market. Rather than focus on the likely avenues of collusion among hospitals, HCA assumes into existence a world in which collusion is infeasible. We reject this analysis. Neither Dr. Harris nor HCA in its briefs offer any explanation whatsoever for why hospitals would have to fix every aspect of competition between them to collude profitably.39 We see no reason why hospitals would have more (99J diffculty reaching an understanding with respect to many of the different aspects of acute care hospital competition described above than competitors would in any other market that is equally concentrated. See Salkever 2789 2795-97. For example, boycotts of third-party payors or agreements not to advertise do not depend upon complex formulae. The flaw in Respondent' s analysis, and perhaps the reason why we are asked to simply assume that hospitals would have to collude on all fronts to reap any benefits, is that profits gained from colluding with respect to fewer than all the dimensions of hospital competition wil unlikely be eradicated even by continued competition in other areas. Once hospitals have deviated through collusion from the competitive actions they would pursue independently, even increased competition in other areas would not provide a price-quality mix desired by patients, physicians and third-party payors. There are as a result diminishing returns to investing monopoly rents in other areas. For example, hospitals may fix a percentage price increase for services of their radiology departments. They could try to attract patients by taking that money and improving patient accommodations. But a patient needs only one television and one telephone, and the value of potted plants in the window may diminish greatly as 39 Indeed, Dr. Harris seemed to have read everything relating to health care except the record in this case. So much of his expert testimony was flatly contradicted hy other credible evidence that we have discounted his testimony accordingly.

.. .

Opinion 106 F.

more are added. The point is that the value to hospitals of increased competition on dimensions other than those anticompetitively altered will dissipate greatly before all (100) or even many of the monopoly profits garnered from collusion are used up. Because it is profitable the incentive to collude remains. See G. Stigler Price and Nonprice Competition, in The Organization of Industry 23 (1968). In fact, case law demonstrates that anticompetitive collusi m with respect to only certain variables in the overall competitive landscape can and has occurred even when competition remains with respect to other variables. , Catalano, Inc. v. Target Sales, Inc. 446 U.S. 643 (1980) (agreement on credit terms); United States v. Gasoline Retailers Association 285 F.2d 688 (7th Cir. 1961) (advertising agreement). In the health care field, the Commission has likewise found unlawful agreements restricting only some areas of competition-advertising and certain contractual practices. American Medical.Association, 94 C. 701 (1979), aff'd as modified 638 F. 2d 443 (2d Cir. 1980), aff'd by an equally divided Court 455 U. S. 676 (1982); Michigan State Medical Society, 101 F. C. 191 (1983). It is thus clear that agreements as to all aspects of competition are not necessary for effective collusion to take place and to have a negative impact on competition. See Salkever 2789, 2795-97.

Moreover, the high barriers to entry and expansion into this market make it even less likely that monopoly rents would be competed away. The addition of expensive new equipment requires a CON in this market, making it more diffcult to invest monopoly rents in their purchase. See supra section IV.D. The barriers to adding new bed capacity are even greater. If a hospital cannot add beds to accommodate new patients, the importance of(101) all the other ways in which it could possibly compete away monopoly rents are greatly reduced. HCA offers an additional reason why the acquisitions allegedly create no risk that Chattanooga hospitals wil collude to eliminate price competition, arguing that price collusion is unlikely because of the role of Blue Cross in this market. RAB at 35. All hospitals in the Chattanooga urban area participate in the Blue Cross program and agree to charge Blue Cross subscribers only those rates approved in advance by Blue Cross. Blue Cross in its rate review could detect and deter any concerted request for rate increases, HCA argues, because Blue Cross reviews rates of Chattanooga hospitals with reference to those of other hospitals in the state. If hospitals in a local area such as Chattanooga attempted to raise rates collusively, the argument goes, Blue Cross would immediately discover that and, with its commanding market position, reject the inflated rates. RAB at 36. We cannot accept HCA's claims that Blue Cross has both the omni- 1 1. HOSPITAL COKP. m. AM"K1lca iJU;: 361 Opinion hospitals. First, under the current Blue Cross charge approval system collusion could be diffcult to detect. See Salkever 2387. If all the hospital firms in Chattanooga attempt to raise prices a similar amount in the review process, coordinated pricing could be overlooked; there is no a priori reason why Blue Cross would consider this to be the result of collusion rather than a rise in costs. The evidence is clear that Blue Cross looks at no single factor in determining whether (102) or not to approve a hospi\al's rate increase. Long 1309. Among the myriad of factors it does consider, Blue Cross often looks to the charges of other local hospitals in assessing a rate increase proposal as well as sometimes looking to other comparable hospitals in the state. fd. Indeed, hospitals in the Chattanooga area have offered the rates of other local hospitals to Blue Cross as justification for rate increase proposals. Smith 1965-66. We think that uniform or similar rates of increase among Chattanooga hospitals would therefore likely be approved as reasonable, and certainly reject as unsupported by the evidence HCA's contention that the state-wide comparisions Blue Cross sometimes makes are dispositive here. Salkever 2387.

Furthermore, even if detected, we do not think such collusion could be easily deterred by Blue Cross. HCA ignores the fact that Blue Cross has a contract not only with participating hospitals but also with its subscribers. Blue Cross must serve its subscribers in the Chattanooga area, and HCA does not explain how Blue Cross could reject a concerted effort by the hospitals there even if it wanted to; certainly, Blue Cross could not ask its subscribers to all go to Knoxvile for (103) hospital care if Chattanooga urban area hospitals colluded. We find HCA' s failure to even address this reality fatal to its arguments. In fact, Blue Cross depends upon competition among hospitals to ensure that subscribers' incentives to utilize participating hospitals force individual hospitals to enter into participation agreements with reasonable price terms. See Long 1291-92. A threat of de participation in the Blue Cross plan by a group of Chattanooga hospitals could be enough to pressure Blue Cross into approving higher prices. Salkever 2388-89.

Even if Blue Cross were to deter an exercise of market power directed at it, collusion could be aimed at other groups of patients. The Blue Cross contract does not restrict participating hospitals' charges to non-Blue Cross patients and their third-party payors. Long 1319-20. 40 Indeed, onc of RCA's arguents elsewhere is that rate surveys of area hospitals are used to persuade Blue CroBS of the reasonableness ofa hOBpital's charges. RAB at 28 0.21. We therefore find its argument that Blue Cross wil only consider the reasonablencss of charges acrOM a statewide area erroneous, In fact, we believe that such SUTveYB could facilitatecollusion with respect to Blue Cross by providing the basis for persuading Blue CraBB of the reasonableness of each hospital's supposedly independent request. 4' Dr. Salkever testified that in a similar situation in Philadelphia . a Blue CTOBS plan was forced into lengthy negotiations with area hospitals and was not in (I position to hold firm. Salkever 2389. 510 FEDERAL TRADE CQMMISSION _DECISIONS Opinion 106 F.

In fact, some Chattanooga hospitals have maintained different sets of prices for Blue Cross subscribers and for other patients. Long 1320, 1356; Petruzzi 732; Kantanie 1074, 1138, 1141-42. Finally, HCA argues that any hypothetical agreements or tacit understandings, even if reached, would be unlikely to endure in this market because of the impossibility of monitoring and policing such arrangements and the destabilizing effects of asserted external events that continually buffet the market. For example, there is no opportunity among hospitals to detect cheating from a pricing agreement since information about (104) hospital prices is unavailable through posting or advertising, HCA contends. RAB at 37. We disagree with the conclusion that anticompetitive agreements could not endure because of the diffculty of policing and enforcing such agreements. Cheating with respect to prices would likely be detected through negotiations with third-party payors. For example if a united front were not presented, Blue Cross would likely use the lower price increases offered by cheating hospitals as leverage during negotiations. Long 1309. Moreover, a third-party payor like Blue Cross could not refuse to deal with all the hospitals in Chattanooga; any simple refusal to deal with a non-cheating hospital would indicate a payor had extracted its desired price elsewhere. Rate surveys are another potential tool to detect cheating. Surveys are now used to discover rates for purposes of comparison, despite HCA' s argument to the contrary, and we see no reason why they could not be targeted at items upon which hospitals attempt to collude. Deviations from agreements on price increase formulas could be detected in this manner.

Certainly, cheating on agreements to boycott certain parties or not to advertise would be quickly detected. For example, if hospitals decided to boycott a particular HMO, cheating would be obvious if the HMO successfully began operation. Any agreements to allocate new equipment purchases that require a CON could be monitored through the public CON process. Likewise, an agreement not to advertise would be easily detected by observing the media. In short, we see no reason why detection of cheating (105) would be significantly more diffcult in this than in any service industry. HCA also argues that hospital care is undergoing rapid technological change that makes any collusion unlikely to endure. Such rapid change in the variety and sophistication of treatment of diseases would require constant revision of any hypothetical understandings among hospitals, which would be particularly diffcult because different hospitals would view the opportunities posed by new technology differently, argues HCA. RAB at 38.

361 Opinion would undermine most of the likelyanticompetitive agreementsdiscussed above. For instance, the desire for new equipment would not undermine an agreement to increase prices on the daily charge for rooms. Nor would it impact an agreement to refrain from advertising in certain media. HCA offers no explanation for how such simple agreements or group boycotts would be undermined by changes in technology respecting other aspects of hospital competition, except to make the unsupported assertion that collusion with respect to all elements of competition is necessary.

Moreover, the certificate of need process makes an anticompetitive allocation of technological improvements possible by restricting the allocation of certain expensive new equipment. If only one hospital wil be allowed to obtain the new equipment anyway, an incentive for bid-rigging" exists so that different hospitals are assured of getting at least some 0((106J those improvements at a minimum of effort and expense. See supra section V. 1. Once the certificate of need is grant- , cheating is much more diffcult, and cheating during the CON process would be easily detected because opposition is public. In any case, there are many aspects of competition among hospitals that are not subject to such technological innovation yet are of significant competitive concern.

In sum, we reject HCA' s argument that the increased concentration in the Chattanooga hospital market caused by these acquisitions does not increase the likelihood of successful anticompetitive behavior among Chattanooga hospitals. Rather, based on our review of all the evidence, we find a reasonable probability of anticompetitive behavior in this market as a result of these acquisitions. F. Conclusion We conclude that the effect of the acquisitions in question may be substantially to lessen competition in the Chattanooga urban area general acute care hospital market. The acquisitions have increased concentration substantially in an already concentrated market. The market is no less prone to many forms of collusion than markets in other industries, and perhaps more prone to some types of collusion. Moreover, the market is characterized by extremely high barriers to entry, virtually eliminating any threat to the successful joint exercise of market (107J power by incumbent Chattanooga hospitals. Respondent urges two "defenses" on its behalf, to which we now turn our attention.

Opinion 106 r:T.

VI. DEFENSES A. The Uniqueness of the Health Care Industry Respondent briefly argues that the regulatory environment imposed by Congress in the health planning laws ofthe 1970s evidences that the health care industry is unresponsive to "classic marketplace forces," and draws two conclusions therefrom. RAB at 44-7. First HCA claims that these acquisitions are impliedly immune from Section 7 scrutiny. We recently faced precisely this issue with respect to a hospital acquisition in the AMI case and rejected respondent' s claim of implied immunity. American Medical International, Inc. No. 9158 slip. op. at 12-17 (July 2 1984)(104 F. C. 1). HCA, in the few pages it devotes to implied repeal, points to no facts that would indicate specific prompting" of these acquisitions by operation of a mechanism established under the health planning laws. Id. at 16. The acquisitions therefore cannot even be considered eligible for implied immunity from the antitrust laws. Id. at 15-17. We thus reject HCA' defense for the same reasons we rejected respondent's assertion of implied immunity in AMI. Id. at 12-17. (108) Second, HCA concludes that to the extent Section 7 is applicable the frequency and force with which Congress has recognized the distinctive characteristics of health care markets simply mandates that these acquisitions be judged on the basis of a careful and thorough 'examination of the particular market-its structure, history and probable future. . . .''' RAB at 47 quoting United States v. General Dynamics Corp. 415 U.S. 486, 498 (1974). We have completed such an analysis above.

B. Efficiencies HCA argues that there are "vertical" benefits from these acquisitions of "quality management, sophisticated support services and access to capital." RAB at 54. HCA concedes that any effciencies resulting from the horizontal integration of Parkridge Hospital and the acquired hospitals were modest at best. RAB at 54 & n.51. Its effciencies argument is offered merely to tip any balance in its favor; indeed, HCA makes no attempt at quantifying these alleged benefits. Its brief states: "HCA does not ask the Commission to hold thatthe benefits of these acquisitions outweigh a lessening of competition. HCA' s position is rather that Complaint Counsel have presented no persuasive evidence that a lessening of competition is likely, and that particularly in light ofthe benefits associated with these acquisitions airy speculations of conceivable competitive harm should not suffce to strike down these acquisitions as violating section 7 of the Clayton A - " U Ii D ()O f11\Ol J.J.r:J. VVJ. J. . V-" '.L".L .L..L'-.

361 Opinion For the reasons set forth above, we find the evidence convincing that a lessening of competition is substantially likely in this market as a result of these acquisitions. We must therefore reject HCA' argument on its own terms; there is no "balance" to tip in HCA' favor. We note in addition that HCA has provided no more than speculation as its "evidence" of effciencies. Certainly, HCA has not met the standards for asserting an effciencies defense set forth in AMI which requires "substantial evidence" of effciencies. Slip. op. at 51-53. Moreover, HCA offers no reason to believe that any effciencies resulting from association with a hospital chain were not already substantially realized by association with HAI and HCC, the acquired hospital chains. We therefore reject HCA's case for effciencies flowing from these acquisitions.

Vll. CONCLUSlON We hold that HCA's acquisitions ofHAI and HCC may substantially lessen competition in the Chattanooga urban area acute care hospital market in violation of Section 7 of the Clayton Act and Section 5 ofthe Federal Trade Commission Act. The only remaining issue is the appropriate remedy to be accorded the public in this case. (110) VIII. REMEDY The appropriate remedy in this case is yet another hotly contested issue. Because we find that HCA violated section 7 of the Clayton Act first in purchasing HAI and second in purchasing HCC, we order the divestiture of hospitals purchased-Diagnostic Center Hospital and North Park Hospital-and their adjacent facilities. We also order the divestiture ofthe Downtown General management contract acquired by HCA from HAI, since it is integral to the HAI acquisition and its likely pernicious effect on competition. See supra section V.A.42 The dispute between the parties revolves around the appropriateness of prior approval and prior notification remedies. HCA maintains that neither is appropriate, particularly a prior approval requirement, and that a prior notification requirement in any case should not go beyond that imposed in AMI. Complaint Counsel offer a prior approval remedy that allegedly satisfies the requirements set forth in AMI and in addition advocate a prior notification remedy, which goes somewhat beyond that imposed in AMI for acquisitions not covered by the prior approval provisions. The Commission of course has the authority to impose prior approval requirements. See, AMI slip. op. at 57. We rejected a broad prior approval requirement in AMI for two (Ill) separate but sup- 42 HCA 110 longer manages Red Bank Community Hospital, and no divlOstiture order is required. See supra Bection Opinion - 1G6 F:T. porting reasons. First, it is industry market structure and market conditions, not whether a "knowing and deliberate violation" or a likelihood of repeated unlawful conduct" has been shown, that determine the appropriateness of imposing a prior approval requirement in a Section 7 case. Id. at 57-59. Secondly, we found "that AMI's presence in the hospital market as a potential purchaser of local hospitals that are put up for sale has a substantial potential procompetitive impact, and that the-proposed prior approval requirement wil uniquely debilitate or perhaps entirely eliminate AMI as a competitor in this market. Id. at 60. We concluded that "time is of the essence in negotiations for the purchase of local hospitals, and that the ability to make a purchase commitment with some degree of certainty of obtaining the necessary regulatory approvals is an important element in this negotiating process. Id. Complaint Counsel propose an asserted "narrowly focused" prior approval provision to satisfy our concerns in AMI. It would (112) require HCA for a period of ten years to obtain prior Commission approval before acquiring a hospital which is: 1) within the relevant geographic market; or 2) within an area in which the HHI would be greater than 1800 after the acquisition, ifthe acquisition would raise the HHI by more than 100 points and a certificate of need is required by state law for a new acute care hospital to begin operation. We agree with Complaint Counsel that a prior approval requirement for future HCA hospital purchases and assumption of hospital management contracts in the Chattanooga urban area is appropriate here. The record evidence shows that any horizontal acquisition or assumption of a management arrangement by HCA in the Chattanooga urban area, like those in this proceeding, poses such a potential for harm to competition that prior Commission approval is warranted. A Commission approval requirement does not amount to a ban on acquisitions; but in view ofthe substantial danger of competitive harm in the Chattanooga urban area we think it is the most effcient way to screen out those mergers and management contracts that are potentially anticompetitive. Moreover, we do not believe that imposing a prior approval requirement in one local market, the Chattanooga urban area, wil substantially harm HCA's competitive posi- (113) tion in the market for hospital acquisitions as a whole.44 .J We concluded inAMI: Although the record evidence clearly indicates that the hospital industry is undergoing II move towards increased consolidation 00. the basis of this evidence we are unable to assets. the effects of theBe changes on competitive condition.swithin the multitude oflocallind regional geographic markets that may exist for hospital services.OUT reading ofthe record does not indicate any busis for defining the parameters ofthesc markets, determining concentmtion levels Of changes in concentration levels in these markets lis II result of the acquisitions that are taking place in this industry, or H88e8!ing whether the acquisitions have had either the effect of entrenching monopolists or increasing competition between market participants. Id. at 59. 44 We did not consider prior approval solely within the relevant geographic market AMIin beclluse the issue was never properly before UB. See Amerimn Medical Intemational, Inc., No. 9158, slip. op. at 9-HJ (Nov. 9, 1984) (Order and Opinion of the Cornrni88ion Granting in Part and Denving in Part Comvlaint CmmBp.l'" P"t.it.i"n fn- 361 Opinion However, we cannot agree that the evidence in this case justifies imposing a prior approval reqmrenient for areas outside the Chattanooga urban area. Complaint Counsel's proposal has some appeal since prior approval would be triggered only in highly concentrated markets with applicable CON laws presumably restricting entry. Nevertheless, the evidence does not establish that all hospital markets can be judged by Complaint Counsel's three criteria to be so manifestly anticompetitive as to warrant a prior approval remedy. There was some evidence offered with respect to other hospital markets in support of the various arguments made in this case, but it is insuffcient to allow us "to assess the effect. . . on competitive conditions within the multitude of local and regional geographic markets that may exist for hospital services. AMI slip op. at 59. For example the nature of regulation can vary markedly from state to state. See Harris 3861, 3885-86. We note that while some markets are subject to price controls, the Tennessee market at issue in this case is not subject to price regulation. See id.; AMI slip op. at 29. Competitive environments can thus differ substantially between different local markets. As a result, we cannot conclude that mergers outside the Chattanooga urban area that meet complaint counsel' criteria are so inherently anticompetitive that they should be subject to preclusion without an adjudication. Moreover, we believe as in AMI that such a (114) nationwide prior approval requirement would severely handicap HCA in the highly competitive market for hospital acquisitions in such areas.

It may be that in an industry with numerous, very local output markets such as hospital care, the evidence in a case such as this wil rarely justify nationwide prior approval. However, we by no means foreclose the opportunity for Complaint Counsel to seek prior approval outside the relevant geographic market where evidence suffcient to justify such a broad remedy is available. Suffcient evidence relating to other local markets was not presented in this case. By declining to impose a nationwide prior approval remedy on HCA, however, we do not imply that HCA' s future acquisitions are of no competitive concern. We do believe that a prior notification requirement is necessary and appropriate and not unduly burdensome under the facts of this case, and reject HCA's argument to the contrary. Legitimate concerns about HCA's future acquisitions can be satisfied by requiring HCA to notify the Commission of its intention to make acquisitions that may pose competitive concern. This enables the Commission to investigate and bring an enforcement action if necessary, while at the same time preservng the procompetitive benefits attributable to HCA' s presence in the acquisition market. AMI slip op. at 60.

g., Opinion 106 F::C.

HCA argues that a prior notification requirement is a "punitive remedy justified only when a violation is "flagrant or egregious" or when a party has proceeded in "bad faith" or with a (115) "wilingness to flout the law." RAB at 57-58. In addition, HCA contends, the Hart-Scott-Rodino Act ("HSR") already provides for prior notification to federal authorities of important mergers, and therefore it is both unnecessary and unlawful to require more than HSR demands. RAB at 57-59. We reject these arguments. As noted above, a prior.notification provision allows the Commission to examine mergers which may be of substantial competitive concern without imposing the burden on Respondent that a prior approval remedy would. Moreover, HSR does not assure the Commission adequate opportunity for monitoring because many HCA hospital acquisitions are exempt from its reporting requirements. For example, purchase transactions involving less than 15 milion dollars are exempt. 16 R. 801.1(h), 802.20(a) (1984). In local markets, such acquisitions could be manifestly anticompetitive; AMI's acquisition of French Hospital was almost a merger to monopoly yet the total cost of the transaction was less than 11 million dollars. American Medical International, Inc. No. 9158, slip. op. at 8 (Initial Decision, July 27, 1983) (104 F. C. 1) Such acquisitions are common for HCA. See, e. 117Z- 105; CX 125E. Other types of exempt hospital acquisitions in which HCA is also interested, and which are particularly likely in hospital markets, include purchase acquisitions from governmental entities (15 UB. C. 18a(c)(4) (1982); CX 418E; CX 423; Wiliamson 3265), and acquisitions through leases and management contracts. 15 U. 18a(a)(3) (1982); 16 C. R. 80U(c)(1), 801.2 (1984); see, e. , supra section V.A. (116) We think a prior notification requirement for acquisitions not covered by HSR is therefore appropriate. HCA' s argument further misstates the standard for imposing prior notification relief. Whether such relief is appropriate depends not on whether the respondent has a history of law violations or otherwise deserves to be punished, but on whether, in view of the violation proven in this case, the relief is necessary to detect and investigate future acquisitions that may significantly endanger competition. See AMI slip. op. at 60; American Medical International, Inc. No. 9158 slip. op. at 18 (Nov. 9, 1984) (Order and Opinion of the Commission Granting in Part and Denying in Part Complaint Counsel's Petition for Reconsideration) ("AMI Rehearing ) (104 F. C. 617). In any case the Commission clearly has the authority to impose prior notification relief, and in view of the much lighter burden imposed by such reporting than by a prior approval requirement, we think it appropriate to do so here. See AMI Rehearing at 13-17. Finallv. we see no reason to 11; beyond the prior notification relief _.

- .u.

361 Statement granted in AMI. Our reasons for not imposing waiting periods a,.e set forth well in the consideration of the petition for reconsideration in that case, and wil not be discussed at length here. See id. at 17-18. Suffce it to say that imposing an inflexible waiting period on HCA would subject covered acquisitions to a time constraint that would accomplish little other than disabling HCA vis-a-vis its competitors. Id. at 17. A requirement that HCA notify the Commission of an (117) acquisition, lease or management contract when it becomes legally bound to the arrangement should provide suffcient time for Commission staff to review the acquisition or contract. Nor do we find suffcient reason to go beyond the threshold requirements for reporting set forth in AMI. We will therefore conform the prior notice portion of the order in this case to that imposed in AMI. STATEMENT OF COMMISSIONER PATRICIA P. BAILEY CONCURRING IN PART AND DISSENTING IN PART As in last year s decision in American Medical International agree with the Commission majority on product and geographic market definitions, and the finding of liability. Also as in AMI I must dissent from a remedy which stops short of requiring prior approval for a certain limited class of future mergers, imposing instead an almost purely symbolic duty of prior notification. Indeed, this dissent is so completely parallel in substance to my concerns in AMI that I wil not repeat them here. There is, however one important difference between the two cases which requires further discussion. In AMI it seemed at least possible that the Commission s standards for prior approval could produce a prior approval order in some future case. Unfortunately, it is now abundantly clear that the AMI test, as interpreted in HCA is a mirage, at least as to prior approval which extends beyond the geographic market at issue in the litigation. (2) Consider: here, as in AMI we have a company with an aggressive corporate policy of growth through acquisition2 which engendered a series of takeovers in a local market characterized by high concentration and high entry barriers. We have every reason to suppose that HCA wil carry out its policy in similarly situated markets in the future.3 We also know that many of these mergers, no matter what their actual impact on a local market, may be priced at a level which does not trigger Hart-Scott-Rodino reporting requirements; conse- American Medica/International, Inc. Docket No. 9158 (July 2, 1984) ("AMI"). (104 F. C. 1) See, e. CX 110Z-28; CX 117-J, Z-103, Z-114; WjJiamson Tr. 3256. J In Appendix C to their appeal brief Complaint Counsel list twenty markets where IlCA owns, leases or operates hospitals. The HHI's for these markets range from 1698 to 3246; four.firm concentration ratios range from 75. to 98.6%; HCA's current market share ranges from 3.7% to 45.4%. 518 J.'EDERAL TRADE COMMISSION DECISIONS Statement 106 F.

quently the Commission cannot call upon that Act's provisions to forestall the merger until adequate information for antitrust analysis is received.

, In these circumstances !tfencing- seems appropriate. The Commission should want to evaluate and be able to block future HCA mergers which could adversely affect a vulnerable market. The AMI opinion seemed to say that, if a screen could be devised which would fiter out safe markets, a prior approval order would (3) be acceptable.' In my opinion, Complaint Counsel have produced such an order. They do not ask for a prior approval clause which blankets the nation or even specific regions. Instead, prior approval would be required for an HCA hospital acquisition outside the Chattanooga area only if: (1) HCA is already operating a hospital in the market or would end up operating two or more hospitals as a result of two or more simultaneous acquisitions from different persons; (2) the acquisition would result in a Herfindahl Index in a local hospital market of over 1800;

(3) the acquisition would increase the Herfindahl Index by 100 or more points; and (4) barriers to entry are likely to be high due to state certificate of need laws (which require government approval for new entry into the market) (Complaint Counsel's Appeal Briefat 39 and Appendix A) (4) These standards parallel those of the Department of Justice Merger Guidelines. It might be argued that this test presents two practical diffculties for respondent: how to tell which is a local market, and how to get "sales" data from which to calculate the Herfindahl Index; but Complaint Counsel have anticipated these queries. Their proposed order specifies that market shares be calculated upon statelicensed acute care bed capacity, which is public information, and that the market be deemed to be the Metropolitan Statistical Area MSA") or Primary Metropolitan Statistical Area C"PMSA"), likewise public knowledge. These geographic and product market surrogates are used only to determine the necessity of fiing with the Commission; they need not be used in the substantive evaluation of the merger. To my mind this is an order which is sensitive to respondent's reporting burdens and yet serves the Commission s enforcement needs.

Nevertheless, the Commission rejects this approach, showing that all along the true parameters of the AMI test were contained in the 4 Io AM! the majority rejected fencing-in with the following language: OUT reading.ofthe record does not indicate any basis for defining the parameters of these markets, determining concentration levels or changes in concentration levels in these markets as a result of the acquisitions that are taking place in this industry, or assessing whether the acquisitions have had either the effect of entrench - 361 Final Order discussion of the procompetitive effects of AMI's presence as a potential purchaser in the market, and the a1Ieged unwarranted hobbling effect of a prior approval requirement on hospital purchase negotiations where time is of the essence. AMI slip op. at 60. In the HCA opinion, the Commission is so solicitous of respondent' reporting "burden" that only a full-blown proof of the vulnerability of other local markets, extraneous to the case at hand, wil justify asking respondent to provide the (5) information described above. See HCA slip op. at 113-116. In other words, respondents wil not be burdened" with the duty to provide a few pieces of publicly available data before they merge, but Commission staff are asked to assume the vastly greater burden of expanding their litigated proofs of geographic market definition, barriers to entry, concentration ratios, testimony as to the nature of competition-in short, all the factors which are necessary before a certain local market can be defined and determined to be at risk from a potential future merger. We have recently revised our procedural rules to encourage all parties to streamline and expedite administrative trials. 50 FR 41485 (Oct. 11, 1985). Why then are we now unnecessarily complicating a merger case by requiring explorations into extraneous markets? The burden on our resources, not to mention the risk to the case in chief is not justified by the increasingly faint possibility of obtaining a prior approval order. On the Commission s commitment to such prior approval remedies, at least where hospital mergers are concerned, the time has come to point out that the emperor is wearing no clothes. CONCURRING STATEMENT OF COMMISSIONER MARY L. AZCUENAGA I concur in the result regarding liability and, based upon the recent precedent in American Medical International, Inc. Docket No. 9158 (July 2, 1984), remedy. (104 F. C. 1) FINAL ORDER This matter has been heard by the Commission upon the appeals of Complaint Counsel and respondent Hospital Corporation of America from the Initial Decision and upon briefs and oral argument in support of and in opposition to the appeals. For the reasons stated in the accompanying Opinion, the Commission has determined to affrm in part and reverse in part the Initial Decision. Accordingly, the Commission enters the following order. (2) Firiaf Order - 106- Definitions It is ordered That for purposes of this order the following definitions shall apply:

A. HCA means Hospital Corporation of America, a corporation organized under the laws of Tennessee, with its principal executive offce at One Park Plaza, Nashville, Tennessee, and its directors offcers, agents, employees, and representatives, and its subsidiaries divisions, affliates, successors, and assigns. B. HAl means Hospital Affliates International, Inc. C. County also includes a county equivalent, such as a parish in Louisiana.

herein referred to as hospital means a D. Acute care hospital health facility, other than a federally owned facility, having a duly organized governing body with overall administrative and professional responsibilty and an organized professional staff that provides 24-hour inpatient care, as well as outpatient services, and which has as a primary function the provision of inpatient services for medical diagnosis, treatment, and care of physically injured or sick persons with short-term or episodic health problems or infirmities. E. Acquire any hospital means to directly or indirectly acquire all or any part of the stock or assets of any hospital, or enter into any arrangement by which HCA obtains direct or indirect ownership, management or control of any hospital or any unit of such hospital including a lease of or management contract for any such hospital or unit of such hospital. (3) F. Operate a hospital means to own, lease or manage an acute care hospital.

G. MSA and PMSA mean, respectively, a Metropolitan Statistical Area, and a Primary Metropolitan Statistical Area, as defined as of July 1, 1983, by the Offce of Information and Regulatory Affairs, Offce of Management and Budget, Executive Offce of the President. H. The Chattanooga Urban Area means that geographic area comprising Hamilton County, Tennessee and Dade, Walker and Catoosa counties in Georgia.

1. Person means any natural person, partnership, corporation, company, association, trust, joint venture, or other business or legal entity, including any governmental agency.

A. It is ordered That, within twelve (12) months from the date this order becomes final, HCA shall divest, absolutely and in good faith, at no minimum price:

361 Final Order (1) North Park Hospital in HamiltoJl County, Tennessee, and all assets, properties, lands, licenses, leases, and other rights and privileges in connection with the hospital, both tangible and intangible. The divestiture required by this provision of this order specifically shall include any medical offce building owned by HCA that is adjacent to, affliated with, or operated in connection with, North Park Hospital, as well as the plot ofland on which each such medical offce building is situated. The purpose of this divestiture is to establish North Park Hospital as a viable competitor, and to restore competition in the area. The divestiture shall be subject to the prior approval of the Federal Trade Commission; (4) (2) Diagnostic Center Hospital in Hamilton County, Tennessee, and all assets, properties, lands, licenses, leases, and other rights and privileges in connection with the hospital, both tangible and intangible, that HCA acquired from HAI, together with any subsequent improvements in, or additions to, any such assets or properties. The divestiture required by this provision of this order specifically shall include any medical offce building owned by HCA that is adjacent to affliated with, or operated in connection with, Diagnostic Center Hospital, as well as the plot ofland on which each such medical offce building is situated. The purpose of this divestiture is to reestablish Diagnostic Center Hospital as a viable competitor, and to restore competition in the area. The divestiture shall be to a person other than the person to whom divestiture is made under Section II., Paragraph A. (1) of this order, and shall be subject to the prior approval of the Federal Trade Commission.

Pending divestiture, HCA shall take all measures necessary to maintain North Park Hospital and Diagnostic Center Hospital in their present conditions and to prevent any deterioration, except for normal wear and tear, of any of the assets to be divested, so as not to impair the present operating abilties or market value ofthe hospitals or the other assets to be divested.

B. It is ordered That, within twelve (12) months from the date this order becomes final, HCA shall divest any and all interest in, and divest or terminate all contracts or arrangements whereby it manages, Downtown General Hospital in Hamilton County, Tennessee, together with all assets, properties, lands, licenses, leases, and other rights and privileges in connection with the hospital, both tangible and intangible, that HCA acquired from HAI, together with any subsequent improvements (5) in, or additions to, such assets or properties. The divestiture required by this provision ofthis order specifically shall include the plot ofland on which Downtown General Hospital is situated, as well as the medical offce building owned by HCA that is adjacent to Downtown General Hospital, and the plot of land on .. __ _ Final Order 106 F.

which such medical offce building is situated. The purpose of this provision is to reestablish Downtown General Hospital as a viable competitor, and to restore competition in the area. Any divestiture pursuant to this provision, other than simple and unconditional termination of management contracts, leases, or other similar arrangements, shall be subject to the prior approval of the Federal Trade Commission.

Pending divestiture or termination, HCA shall take all measures necessary, within its responsibilities and authority, to maintain Downtown General Hospital in its present condition and to prevent any deterioration, except for normal wear and tear, of any assets to be divested, so as not to impair the present operating abilities or market value of the hospital or the other assets to be divested. A. If HCA has not divested all of the properties, assets, contracts arrangements or enterprises required to be divested pursuant to Section II of this order within the 12-month period provided therein, the Federal Trade Commission may select a trustee to effect any ordered divestitures yet to be accomplished. The trustee shall be a person with experience and expertise in acquisitions and divestitures. If the Federal Trade (6) Commission should elect to appoint a trustee, it shall not be precluded from seeking civil penalties and other relief available to it for any failure by HCA to comply with this order. If the Federal Trade Commission should not elect to appoint a trustee under this Section III of this order, it shall not be precluded from seeking civil penalties, the appointment by the courts of a trustee to effect the divestitures, and other relief available to it, for any failure by HCA to comply with this order.

B. Any trustee appointed by the Federal Trade Commission pursuant to this Section shall have the following powers, authority, duties and responsibilities:

(1) The trustee shall have the exclusive power and authority to divest any properties, assets, or enterprises required to be divested pursuant to Section II of this order that have not been divested by HCA within the time period for the divestitures provided therein. The trustee shall have twelve (12) months from the date of appointment to accomplish the divestitures, which shall be subject to the prior approval of the Federal Trade Commission. If, however, at the end of the twelve-month period the trustee has submitted a plan of divestiture or believes that divestiture can be achieved within a reasonable LL- L=.I__u- --- -1 - --- J_- -- .J_- 1- 1 1 ......

u. -- .U'-- 361 Final Order Commission. In addition, any delays in divestiture caused by !ICA shall extend the time for divestiture in accordance with the delay caused.

(2) The trustee shall have full and complete access to the personnel books, records, and facilities of any property, asset, or enterprise that the trustee has the duty to divest, and HCA shall develop such financial or other information relevant to the properties, assets, or enterprises to be divested as such trustee may reasonably request. HCA (7) shall cooperate with the trustee, and shall take no action to interfere with or impede the trustee s accomplishment of the divestitures. (3) The power and authority ofthe trustee to divest shali be at the most favorable price and terms available consistent with this order absolute and unconditional obligation to divest at no minimum price and the purposes of the divestitures as stated in Section II of this order.

(4) The trustee shall serve, without bond or other security, at the cost and expense of HCA on such reasonable and customary terms and conditions as the Federal Trade Commission may set. The trustee shall have authority to retain, at the cost and expense of HCA, such consultants, attorneys, investment bankers, business brokers, accountants, appraisers, and other representatives and assistants as are reasonably necessary to assist in the divestitures. The trustee shall account for all monies derived from the sale and all expenses incurred. After approval by the Federal Trade Commission of the acallcount of the trustee, including fees for his or her services, remaining monies shall be paid to HCA and the trustee s power shall be terminated. The trustee s compensation shall be based at least in significant part on a commission arrangement contingent on the trustee divesting the trust property.

(5) HCA shall indemnify the trustee and hold the trustee harmless against any losses, claims, damages, or liabilities to which the trustee may become subject, arising in any manner out of, or in connection with, the trustee s duties under this order, unless the Federal Trade Commission determines that such losses, claims, damages, or liabilities arose out of the misfeasance, gross negligence, or the wilful or wanton acts or bad faith of the trustee. (8) (6) Promptly upon appointment of the trustee and subject to the approval of the Federal Trade Commission, HCA shall, subject to the Federal Trade Commission s prior approval and consistent with provisions of this order, execute a trust agreement that transfers to the trustee all rights and powers necessary to permit the trustee to cause the divestitures.

(7) If the trustee ceases to act or fails to act dilgently, the Federal Trade Commission shall appoint a substitute trustee. Final Order 106 F.

(8) The trustee may ask the Federal Trade Commission to issue, and the Federal Trade Commission may issue, such additional orders or directions as may be necessary and appropriate to accomplish the divestitures required under this order.

(9) The trustee shall have no obligation or authority to operate or maintain any of the properties, assets, contracts, arrangements or enterprises required to be divested pursuant to Section II of this order. (10) The trustee shall report in writing to HCA and the Federal Trade Commission every sixty (60) days concerning the trustee s efforts to accomplish divestiture.

It is further ordered That, for a period of ten (10) years from the date this order becomes final, HCA shall not, without the prior approval of the Federal Trade Commission, acquire any hospital, as defined in this order, if the hospital to be acquired is within the Chattanooga Urban Area, as defined in this order. (9) Provided, however that no acquisition of a hospital by purchase shall be subject to this Section IV of this order if the consideration to be paid for the purchase of the hospital or any rights or interest therein, including assumption by HCA of any liabilities of its present owners, does not exceed one milion dollars ($1 000 000), and provided that no lease or management arrangement shall be subject to this Section IV of this order if the fair market value of the hospital to be leased or managed does not exceed one milion dollars ($1 000 000). It is further ordered That for a period often (10) years from the date this order becomes final, HCA shall not, without providing advance notification to the Federal Trade Commission, acquire any hospital as defined in this order, if:

A. The hospital to be acquired is within an MSA or a PMSA in which HCA already operates a hospital and in which HCA, immediately after the acquisition, would operate hospitals that combined have a twenty (20) percent or more share of the licensed acute care hospital beds within that MSA or PMSA; or B. The hospital to be acquired is not within an MSA or a PMSA but is within a county in which HCA already operates a hospital and in which HCA, immediately after the acquisition, would operate hospitals that combined have a twenty (20) percent or more share of the ... .... ... , rrv..:u J.ru.. 'V....

361 Final Order C. The hospital to be acquired is (1) not within an MSA or a PMSA or a county in which HCA already operates a hospital, but is within thirty (30) miles of a hospital which HCA already operates in another MSA or PMSA or county, and (2) the hospital to be acquired and any hospital(s) that HCA operates combined (10) have a twenty (20) percent or more share of the licensed acute care hospital beds in the area within thirty (30) miles of the midpoint between the hospital to be acquired and any hospital operated by HCA.

The notification required of HCA by this Section V of this order shall not require additional notification by HCA to the Federal Trade Commission of any acquisition for which notification is required to be made, and has been made, pursuant to Section 7 A of the Clayton Act, 15 D. C. 18a, or for which prior approval by the Federal Trade Commission is required, and has been requested, pursuant to Section IV ofthis order. Such advance notification shall be provided when HCA' Board of Directors or Executive Committee, or any individual or entity that is authorized to act on HCA's behalf in such acquisitions authorizes issuance of a letter of intent or enters into an agreement to make such an acquisition, whichever is earlier. The notification required by this Section V of this order shall be the Notification and Report Form set forth in the Appendix to Part 803 of Title 16 of the Code of Federal Regulations, as amended, and shall be prepared and transmitted in accordance with the requirements of that part. The notification required by this Section V of this order shall apply to HCA and shall not apply to any party that HCA seeks to acquire. However, HCA shall provide at the same time of the fiing of the Notification and Report Form supplemental information, either in HCA's possession or reasonably available to HCA, relating to the hospital to be acquired, the HCA hospital in that geographic (11) area, and identification and assessment of the area hospital market. Such supplemental information should include, where available, patient flow data, annual management and strategic plans, hospital utilzation and revenue data, and documents relating to market share, formulation of hospital prices, competitive interaction among area hospitals, implementation of certificate of need standards in the area, planned effciencies, relations with third-party payors, and physician admitting patterns.

HCA shall comply with reasonable requests by the Commission staff for additional information within fifteen (15) days of service of such requests.

Any acquisition subject to this Section V of this order, involving an arrangement to lease, manage, or control a hospital, shall be fully described in the notification regardless of whether the acquisition involves the acquisition of any stock or assets of a hospital. Final Order - 106F.

Provided, however that no acquisition of a hospital by purchase shall be subject to this Section V ofthis order if the consideration to be paid for the purchase of the hospital or any rights or interest therein, including assumption by HCA of any liabilities of its present owners, does not exceed one milion dollars ($1 000 000), and provided that no lease or management arrangement shall be subject to this Section V of this order if the fair market value of the hospital to be leased or managed does not exceed one milion dollars ($1 000 000). (12) It is further ordered That HCA shall, within sixty (60) days after the date this order becomes final and every sixty (60) days thereafter until it has fully complied with the provisions of Section II of this order, submit in writing to the Federal Trade Commission a report setting forth in detail the manner and form in which it intends to comply, is complying, and has complied with these provisions. Such compliance reports shall include, in addition to any other information that the staff of the Federal Trade Commission may request, a summary of all contacts and negotiations with potential purchasers of the stock, assets, contracts, or other rights or interests to be divested under this order, the identity and address of all such potential purchasers, and copies of all written communications to and from such potential purchasers.

HCA shall submit such further written reports as the staff of the Federal Trade Commission may, from time to time, request in writing to assure compliance with this order.

VII It is further ordered That HCA, upon written request of the Secretary of the Federal Trade Commission or the Director of the Bureau of Competition of the Federal Trade Commission made to HCA at its principal offce, for the purpose of securing compliance with this order, and for no other purpose, and subject to any legally recognized privilege, shall permit duly authorized (13) representatives of the Federal Trade Commission or the Director of the Bureau of Com pet ition:

(1) reasonable access during the offce hours of HCA, which may have counsel present, to those books, ledgers, accounts, correspondence, memoranda, reports, and other records and documents in 361 Final Order HCA' s possession or control that relate materially and substantially to any matter contained in this-order; and- (2) an opportunity, subject to the reasonable convenience of HCA, to interview offcers or employees of HCA, who may have counsel present, regarding such matters.

VII It is further ordered That HCA shall notify the Federal Trade Commission at least thirty (30) days prior to any proposed corporate change, such as dissolution, assignment, or sale resulting in the emergence ofa successor corporation, the creation or dissolution of sub sid aries, or any other change in the corporation that may affect compliance with the obligations arising out of this order. Complaint 106 F.