Opinion
No. 15 C 11473
03-16-2017
FEDERAL TRADE COMMISSION and STATE OF ILLINOIS, Plaintiffs, v. ADVOCATE HEALTH CARE, ADVOCATE HEALTH AND HOSPITALS CORPORATION, and NORTHSHORE UNIVERSITY HEALTHSYSTEM, Defendants.
REDACTED MEMORANDUM OPINION AND ORDER
Plaintiffs, the Federal Trade Commission ("FTC") and the State of Illinois, have sued defendants to enjoin them from consummating their proposed merger, pending completion of the FTC's administrative trial on the merits of plaintiffs' antitrust claims. This Court held a preliminary injunction hearing over nine days in April and May of 2016. The Court initially denied the motion for preliminary injunction, holding that plaintiffs had not met their burden of proving a relevant geographic market, but on appeal, the Seventh Circuit reversed and remanded for further proceedings, holding that this Court's decision was based on erroneous factual findings. The Court has received additional briefing from the parties, and it now reconsiders plaintiffs' motion for preliminary injunction in light of the guidance the Seventh Circuit has provided. This opinion sets forth the Court's findings of fact and conclusions of law pursuant to Federal Rules of Civil Procedure 52(a)(2). For the reasons set forth below, the Court grants the motion.
To the extent any findings of fact may be deemed conclusions of law, they shall also be considered conclusions of law. To the extent that any conclusions of law may be deemed findings of fact, they shall also be considered findings of fact. See McFadden v. Bd. of Educ. for Illinois Sch. Dist. U-46, 984 F. Supp. 2d 882, 886 n.1 (N.D. Ill. 2013).
BACKGROUND
Parties
Defendant Advocate Health Care Network, which is the parent of Advocate Health and Hospitals Corp. (collectively, "Advocate"), is a health care system that includes eleven hospitals: (1) BroMenn Medical Center; (2) Christ Medical Center; (3) Condell Medical Center; (4) Eureka Hospital; (5) Good Samaritan Hospital; (6) Good Shepherd Hospital; (7) Illinois Masonic Medical Center; (8) Lutheran General Hospital; (9) Sherman Hospital; (10) South Suburban Hospital; and (11) Trinity Hospital. See http://www.advocatehealth.com/hospital-locations (last visited January 26, 2017). Defendant NorthShore University HealthSystem ("NorthShore") is a health care system that includes four hospitals: (1) NorthShore Evanston Hospital; (2) NorthShore Glenbrook Hospital; (3) NorthShore Highland Park Hospital; and (4) NorthShore Skokie Hospital. See http://www.northshore.org/locations (last visited January 26, 2017). In September 2014, Advocate and NorthShore signed an affiliation agreement to merge and create Advocate NorthShore Health Partners ("ANHP"). (See DX3118, Affiliation Agreement.) "The combined entity would operate 15 GAC [general acute care] hospitals in Illinois and would generate approximately $7.0 billion in revenue." (Pls.' Proposed Findings of Fact ("PFF") ¶ 3, ECF No. 446.) Health Care Contracting
Commercial health insurers (also called payers) try to create networks of health care providers that are attractive to potential members. (Id. ¶ 12; Defs.' Proposed Findings of Fact ("DFF") ¶ 21, ECF No. 459; Preliminary Injunction Hr'g Tr. ("Tr.") 75:11-16 [Norton-CIGNA]; id. at 148:12-18 [Hamman-Blue Cross Blue Shield of Illinois ("BCBSIL")].) Among the factors insurers consider when determining whether to include a hospital in a network are "the attractiveness of that hospital, the quality, the reputation of that hospital, . . . its willingness to . . . meet certain price points," and its geographic coverage. (Tr. at 149:3-11 [Hamman-BCBSIL]; see id. at 74:18-75:7 [Norton-CIGNA].)
Hospitals compete to be included in insurers' networks and negotiate reimbursement rates and services with the insurers. (PFF ¶ 9; Tr. at 76:8-19 [Norton-CIGNA]; id. at 149:12-20 [Hamman-BCBSIL]; JX 9, Englehart Investigative Hr'g ("IH") Tr. at 142:2-9, ECF No. 453-9).) A hospital has more bargaining leverage if there are fewer substitutes for it that can be included in the insurer's network; the insurer has more leverage if there are more substitutes for the hospital. (Tr. at 106:23-107:3 [Norton-CIGNA]; id. at 150:22-51:22 [Hamman-BCBSIL]; [Redacted] .) The Chicago market is dominated by one commercial payer, BCBSIL, which has about 4 million members in the Chicago area. (Tr. at 145:9-11 [Hamman-BCBSIL]; id. at 1121:3-8 [Beck-United]; id. at 1175:13-22 [Nettesheim-Aetna]; id. at 1412:18-25 [Sacks-Advocate].) The other payers include United Health Group, Aetna, CIGNA, and Humana, which have about 1.5 million, [Redacted] , 350,000, and [Redacted] members, respectively, in the area. (Tr. at 72:2-4 [Norton-CIGNA]; id. at 1115:4-6 [Beck-United]; DX1515.0002, Carrier Market Share Calculation; DX1862.0005, ECF No. 460-8; Advocate/Aetna Collaboration Discussion Guide, ECF No. 460-12.)
Insurers pay health care providers under fee-for-service ("FFS") or risk-based contracts. Under FFS contracts, the payer pays a set fee for every service the provider gives to a patient. (Tr. at 85:16-18 [Norton-CIGNA].) Risk-based contracts "[are] a set of payment arrangements in which providers hold some degree of financial risk." (PX 6001, Jha Report ¶ 10, ECF No. 450-2.) These arrangements may take any of a number of different forms. (Id. ¶ 24.) For example, in their most extreme form, known as a full capitation or global risk arrangement, a provider is paid a set amount per patient per month for all of that patient's health care services, regardless of the extent of the care that patient ultimately requires or how much it costs. (Id.) Ninety percent of NorthShore's commercial revenues come from FFS contracts; less than a third of Advocate's commercial revenues come from FFS contracts. (DFF ¶ 50; Tr. at 785:10-13 [Golbus-NorthShore]; id. at 1410:18-20 [Sacks-Advocate].) Rationale for the Merger
Advocate's alleged rationale for the merger is "to create a new, low-cost, high performing network ("HPN") insurance product that can be sold . . . throughout Chicagoland," which it claims it cannot do "unless and until the merger with NorthShore is consummated due to [Advocate's] geographic gap east of Interstate 94." (DFF ¶¶ 38, 49.) NorthShore's alleged rationale for the merger is "[to] engage in large-scale full risk contracting," which it says it cannot do "absent a merger, because it lacks: (1) sufficient geographic coverage; and (2) utilization management tools, care management tools, physician workflows and experience, . . . which Advocate can provide." (Id. ¶ 52.)
DISCUSSION
Section 7 of the Clayton Act prohibits a merger "in any line of commerce or in any activity affecting commerce in any section of the country, the effect of [which] may be substantially to lessen competition, or tend to create a monopoly." 15 U.S.C. § 18. The Court may preliminarily enjoin a violation of § 7 "[u]pon a proper showing that, weighing the equities and considering the Commission's likelihood of ultimate success, such action would be in the public interest." 15 U.S.C. § 53(b). "Therefore, 'in determining whether to grant a preliminary injunction . . . , a district court must (1) determine the likelihood that the FTC will ultimately succeed on the merits and (2) balance the equities.'" FTC v. OSF Healthcare Sys., 852 F. Supp. 2d 1069, 1073 (N.D. Ill. 2012) (quoting FTC v. Univ. Health, Inc., 938 F.2d 1206, 1217 (11th Cir. 1991)). "[T]o demonstrate such a likelihood of ultimate success, the FTC must raise questions going to the merits so serious, substantial, difficult and doubtful as to make them fair ground for thorough investigation, study, deliberation and determination by the FTC in the first instance and ultimately by the Court of Appeals." FTC v. Tenet Health Care Corp., 186 F.3d 1045, 1051 (8th Cir. 1999) (quotations omitted). "Although the district court may not 'simply rubber-stamp an injunction whenever the FTC provides some threshold evidence,' the FTC 'does not need detailed evidence of anticompetitive effect at this preliminary phase.' Instead, 'at this preliminary phase it just has to raise substantial doubts about a transaction.'" OSF Healthcare, 852 F. Supp. 2d at 1074 (quoting FTC v. Whole Foods Mkt., Inc., 548 F.3d 1028, 1035-36 (D.C. Cir. 2008)) (internal citations omitted).
I. LIKELIHOOD OF SUCCESS ON THE MERITS
A. Geographic Market
"Determination of the relevant product and geographic markets is 'a necessary predicate' to deciding whether a merger contravenes the Clayton Act." United States v. Marine Bancorporation, Inc., 418 U.S. 602, 618 (1974) (quoting United States v. E.I. Du Pont De Nemours & Co., 353 U.S. 586, 593 (1957)); see Tenet Health Care, 186 F.3d at 1051 ("It is . . . essential that the FTC identify a credible relevant market before a preliminary injunction may properly issue."); OSF Healthcare, 852 F. Supp. 2d at 1075 ("In fact, '[a] monopolization claim often succeeds or fails strictly on the definition of the product or geographic market.'") (quoting Tenet Health Care, 186 F.3d at 1052).
The parties agree that the relevant product market in this case is inpatient general acute care services sold to commercial payers and their insured members ("GAC services"). (PFF ¶ 15; Tr. at 1270:3-6 (defense expert Dr. Thomas McCarthy conceding that the relevant product market is GAC services).) GAC services are a cluster of medical services that require a patient to be admitted to a hospital at least overnight. (PFF ¶ 16; Tr. at 78:18-19 [Norton-CIGNA]). See OSF Healthcare, 852 F. Supp. 2d at 1075 ("This is a 'cluster market' of services that courts have consistently found in hospital merger cases, even though the different types of inpatient services are not strict substitutes for one another. See FTC v. ProMedica Health Sys., Inc., No. 3:11 CV 47, 2011 WL 1219281, at *54 (N.D. Ohio Mar. 29, 2011) (collecting cases); see also United States v. Rockford Mem'l Corp., 898 F.2d 1278, 1284 (7th Cir. 1990) (upholding a similar GAC product market).").
The parties do not agree on the boundaries of the relevant geographic market. The relevant geographic market is "[the] area in which the seller operates, and to which the purchaser can practicably turn for supplies." United States v. Phila. Nat'l Bank, 374 U.S. 321, 359 (1963) (internal quotation omitted). There is no formula for determining the geographic market; rather, it should be identified in "a pragmatic [and] factual" way and should "correspond to the commercial realities of the industry." Brown Shoe Co. v. United States, 370 U.S. 294, 336-37 (1962) (quotation omitted). The geographic market "need not . . . be defined with scientific precision," United States v. Connecticut National Bank, 418 U.S. 656, 669 (1974), but it "must be sufficiently defined so that the Court understands in which part of the country competition is threatened," FTC v. Cardinal Health, Inc., 12 F. Supp. 2d 34, 49 (D.D.C. 1998).
1. Geographic market analysis of plaintiffs' expert Dr. Tenn
Plaintiffs contend that the relevant geographic market, which their expert Steven Tenn refers to as the "North Shore Area," includes six of the merging hospitals—Advocate Lutheran General Hospital, Advocate Condell Medical Center, NorthShore Evanston Hospital, NorthShore Skokie Hospital, Glenbrook Hospital, and Highland Park Hospitals—as well as Vista East Hospital, Northwest Community Hospital, Presence Resurrection Hospital, Northwestern Lake Forest Hospital, and Swedish Covenant Hospital, all of which are located in northern Cook or southern Lake Counties. (PX 6000, Tenn Report ¶¶ 9-11, 14-15, 18, 72, ECF No. 450-1.) Dr. Tenn explained in his report that this area of "overlap" between the four NorthShore hospitals and their two most significant Advocate competitors, Advocate Lutheran General and Advocate Condell, is the "primary area of competition between Advocate and NorthShore." (Id. ¶ 17.) In fact, Dr. Tenn explained that these six party hospitals alone constitute "a relevant geographic market in which it would be appropriate to assess the transaction." (Id. ¶ 76.)
However, in an effort to be "conservative," he "focus[ed] [his] analysis on a broader geographic market," defined to include the five additional competing hospitals. (Id. ¶ 79.) He selected these hospitals based on their location, including hospitals "with at least a two percent share in the area from which the relevant Advocate and NorthShore hospitals attract patients" and hospitals "that overlap with [, i.e., draw patients from the same area as] both Advocate and NorthShore." (Tr. at 453:22-23, 463:2-65:12.) He excluded a number of academic medical centers and specialized hospitals, which he called "destination" hospitals, i.e., Northwestern Memorial Hospital, Rush University Hospital, University of Chicago Hospital, Loyola University Hospital, Cancer Treatment Centers of America, and Lurie Children's Hospital, because these hospitals draw patients from not just the North Shore Area but from all over the Chicago metropolitan area. (Tenn Report ¶ 85 at n.175.) As Dr. Tenn recognized, and as the evidence showed, patients generally prefer to receive GAC services close to home. (PFF ¶¶ 26-27, 41.) See FTC v. Advocate Health Care Network, 841 F.3d 460, 474 (7th Cir. 2016) ("FTC"). Based on this preference, employers require—and insurers must offer—health plans that provide patients with access to in-network hospitals near where they live. (PFF ¶¶ 26-32.) See FTC, 841 F.3d at 473-75. Thus, although many patients travel from the North Shore Area to these destination hospitals, Dr. Tenn nevertheless excluded them from his analysis because these hospitals cannot fulfill the function of providing local care within the North Shore Area. (Tr. at 454:4-9 ("Here the competitive concern is that Advocate and NorthShore are substitutes for commercial payers when they're putting together provider networks in the northern Chicago suburbs. The destination hospitals . . . are not located in the northern Chicago suburbs and, therefore, do not fulfill this role for commercial payers.").)
After identifying the market, Dr. Tenn tested whether it passed the hypothetical monopolist test; that is, whether a hypothetical monopolist that owned all of the hospitals in the market could profitably impose a small but significant non-transitory increase in price ("SSNIP") (i.e., 5% or more) at one or more of the merging hospitals due to the hypothetical monopolist's "internalization of substitution" in the region. (PFF ¶ 33; Tenn Report ¶¶ 57, 71.) See DOJ/FTC Horizontal Merger Guidelines §§ 4, 4.1.1, 4.2, 4.2.1, available at https://www.ftc.gov/sites/default/files/attachments/merger-review/100819hmg.pdf (last visited Jan. 31, 2017).
Tenn measured the level of substitution by calculating diversion ratios, that is, the fraction of patients who use one hospital for GAC services that would switch to another hospital if their first-choice hospital were no longer available. (Tenn Report ¶¶ 95-98.) He determined that 48% of the patients admitted to one of the eleven hospitals in the North Shore Area would substitute to one of the other hospitals in the North Shore Area if their chosen hospital were no longer available. (Id. ¶ 99.)
This "level of intra-market diversion," Tenn opined, "is sufficiently high . . . to pass the hypothetical monopolist test." (Id. ¶ 100.) Dr. Tenn concluded that he had identified a relevant geographic market.
2. Geographic market: procedural history
In its prior opinion in this case, this Court rejected Dr. Tenn's analysis and denied the motion for preliminary injunction because it found that plaintiffs did not prove a relevant geographic market. In particular, the Court found that Dr. Tenn provided no compelling reason for excluding destination hospitals from the geographic market, considering that (a) the very diversion ratios he calculated show that patients consider some destination hospitals, particularly Northwestern Memorial, to be close substitutes for some of the merging hospitals, and (b) the evidence did not unequivocally support Dr. Tenn's assumption that patients prefer to receive hospital care near their homes. (See Am. Mem. Op. & Order, June 20, 2016, ECF No. 484.)
The Seventh Circuit reversed and remanded for reconsideration of the motion for preliminary injunction, explaining that "the geographic market question is . . . most directly about 'the likely response of insurers,' not patients, to a price increase," because "[i]nsured patients are usually not sensitive to retail hospital prices, while insurers respond to both prices and patient preferences." FTC, 841 F.3d at 471 (quoting Saint Alphonsus Med. Ctr.-Nampa Inc. v. St. Luke's Health Sys., Ltd., 778 F.3d 775, 784 (9th Cir. 2015)). As a result, "insurers are the most relevant buyers." FTC, 841 F.3d at 475. Although Dr. Tenn's diversion ratios showed that certain hospitals outside of the North Shore Area, particularly Northwestern Memorial, draw large numbers of patients from the North Shore Area, this fact did not fatally undermine Dr. Tenn's geographic market analysis because "measures of patient substitution like diversion ratios do not translate neatly into options for insurers." Id. Insurance executives "unanimous[ly]" testified that "an insurer's network must include either Advocate or NorthShore to offer a product marketable to employers." Id. at 474. This testimony was supported by "strong, not equivocal" evidence that patients generally prefer to receive hospital care locally. Id. Economists studying hospital markets have long recognized the "silent majority fallacy": even if evidence shows that some patients are willing to travel for care, it does not follow that more would do so to avoid a price increase; it may be that there is a "silent majority" that would pay supra-competitive prices to receive hospital services close to home rather than travel. Id. at 470. Thus, even if it is true that large numbers of patients who live in the North Shore Area travel outside the Area to hospitals such as Northwestern Memorial for GAC services, it is error "to focus on the patients who leave a proposed market instead of on hospitals' market power over the patients who remain, which means that the hospitals have market power over the insurers who need them to offer commercially viable products to customers who are reluctant to travel farther for general acute hospital care." Id. at 476.
On remand, the parties take dramatically different approaches to the geographic market issue. Plaintiffs' position is that the Seventh Circuit's opinion all but resolves the issue. The Seventh Circuit held that there is strong evidence that an insurer would not be able to sell a plan that does not include either Advocate or NorthShore. Insurers unanimously testified to that effect, and the record as a whole supports that testimony because "the overwhelming weight of the evidence" shows "(1) the large proportion of patients who prefer hospitals close to their homes and (2) the resulting need for insurers to offer networks that include community hospitals close to their customers' homes." Id. at 475 n. 4. It follows inevitably from the Seventh Circuit's opinion, plaintiffs argue, that because many patients in the proposed North Shore Area would have limited or no access to nearby hospitals if their insurance plan did not provide access to any of the eleven hospitals in the North Shore Area, insurers would surely pay a SSNIP in order to be able to offer patients in that region a health plan that includes access to local hospitals. According to plaintiffs, there is no need on remand to delve into the details of Dr. Tenn's analysis or reduce this case to a "battle of the experts" with respect to determining the relevant geographic market; the Seventh Circuit's decision is conclusive on that issue.
Defendants argue that, while insurer representatives may have testified that they cannot sell a plan that excludes both Advocate and NorthShore, they also offered some contrary testimony (see DFF ¶¶ 259-60; Tr. at 280:9-14 [Hamman-BCBSIL]), and it falls to this Court to make "credibility determinations about inconsistent testimony." (Defs.' Post-Remand Resp. Br. at 4.) According to defendants, some insurers, especially BCBSIL, which sees the defendants' proposed merger as a threat to its own business, may have their own competitive reasons for opposing the defendants' proposed merger, and therefore the Court should view the testimony of these insurers' representatives skeptically. (See DFF ¶¶ 211-17.) Because, defendants argue, the insurers' testimony that they cannot sell a plan that excludes both Advocate and NorthShore is inconsistent, biased, and ultimately not credible, this Court must rely on expert testimony in order to define the relevant geographic market. Further, defendants renew their attack on Dr. Tenn's analysis, arguing that the Seventh Circuit's opinion in this case reveals Dr. Tenn's reliance on diversion ratios to be a fatal flaw. Defendants argue that if it is error to rely on Dr. Tenn's patient-centric diversion ratios because insurers, not patients, are the "most relevant buyers," then it must be error to accept Dr. Tenn's conclusions with regard to the scope of the geographic market and the question of whether a hypothetical monopolist could impose a SSNIP because they, too, depend heavily on diversion ratios.
The Court shares some of defendants' concerns about the credibility of the insurers' testimony, which may indeed be self-serving, but even taking their testimony with a grain of salt, the record as a whole supports the view that insurers genuinely believe that a plan that excludes Advocate and NorthShore is not viable in the North Shore Area. Defendants point to BCBSIL's "Project Remedy," a recent attempt by BCBSIL to [Redacted] (See DFF ¶¶ 259-62.) But there is no inconsistency in BCBSIL's testimony on this point. True, in his testimony about Project Remedy, Mr. Steve Hamman, a BCBSIL executive, testified that [Redacted] but he went on to clarify that, [Redacted] (Tr. at 280:5-281:3 [Hamman-BCBSIL].) Defendants argue that the Seventh Circuit "did not examine" this testimony (Defs.' Post-Remand Resp. Br. at 4), but it would have added nothing to its analysis if it did; the Seventh Circuit specifically recognized that a plan that excluded NorthShore and Advocate might well be successful with patients outside the North Shore Area. See FTC, 841 F.3d at 474 ("One company offers a network in the Chicago area without either of the merging parties, but . . . fewer than two percent of those individual members live near NorthShore's hospitals."). (See also PFF ¶ 77.)
Defendants cannot undermine plaintiffs' proposed geographic market definition simply by showing that a plan that excludes both Advocate and NorthShore from its network might attract some enrollees in the North Shore Area because some—or even many—patients are willing to travel outside the market for hospital care. They also have to establish that "enough patients would buy a health plan . . . with no in-network hospital in the proposed geographic market" and instead "turn to hospitals outside the relevant market," FTC v. Penn State Hershey Medical Center, 838 F.3d 327, 343 (3d Cir. 2016)), that insurers are unlikely to agree to pay supra-competitive prices to hospitals in the Area in order to be able to offer attractive products to patients who live within the geographic market but might be "reluctant to travel [outside it] for general acute hospital care," FTC, 841 F.3d at 476. Testimony that an insurer has actually offered a commercially-successful healthcare plan that enrolled large numbers of patients within the North Shore Area but did not include Advocate or NorthShore in its network might have sufficed. Cf. Penn State Hershey, 838 F.3d at 343. But the defendants offered no such testimony in this case, nor did they offer any evidence to demonstrate that a healthcare plan that excluded both Advocate and NorthShore would be successful among patients living in the North Shore Area. In the absence of any such evidence, as the Seventh Circuit explained, the fact that some patients are willing to travel outside the North Shore Area for GAC services does not suggest that there is no "silent majority" in the North Shore Area that is reluctant to travel and that the hospitals in the Area can use as leverage to charge supra-competitive prices.
In any case, even if the Court indulges defendants and undertakes another close examination of Dr. Tenn's work, it must conclude, in light of the guidance the Seventh Circuit has provided, that Dr. Tenn has appropriately delineated the relevant geographic market.
3. Tenn's reliance on diversion ratios
Defendants' central argument with respect to Dr. Tenn's analysis is that the Seventh Circuit's opinion in this case reveals his reliance on diversion ratios to be a fatal flaw. If this Court erred by relying on Dr. Tenn's diversion ratios to reject his geographic market analysis because insurers, not patients, are the most relevant buyers, and "measures of patient substitution like diversion ratios do not translate neatly into options for insurers," FTC, 841 F.3d at 475, then, defendants reason, it must be error to rely on Dr. Tenn's analysis at all, considering that the entire analysis is based on diversion ratios.
But the Court agrees with plaintiffs that "the Seventh Circuit did not hold that it is inappropriate to consider patient-level diversions" (Pls.' Post-Remand Resp. Br., at 5, ECF No. 577); it merely criticized how defendants and this Court interpreted them. A relevant geographic market need not include every firm that competes for business in the vicinity; it need only include those competitors that would "substantially constrain" the merged firm's "price-increasing ability." FTC, 841 F.3d at 469 (citing AD/SAT, a Div. of Skylight, Inc. v. Associated Press, 181 F.3d 216, 228 (2d Cir. 1999)). The purpose of the diversion ratios is to show whether the level of substitution between hospitals in the North Shore Area is high enough that, should a merger occur, the merged entity could profitably impose a SSNIP. As Dr. Tenn explained, "[t]he predicted post-merger price increase is higher for larger diversions between the parties, since there is more substitution for the combined firm to internalize post-merger." (PX06020, Tenn Rebuttal Report ¶ 25 n.38, ECF No. 450-4.) In other words, if the diversion ratios between merging hospitals are higher, then a price increase at a given merging hospital, assuming steady prices at nearby hospitals, would theoretically drive fewer patients away from the merged entity's system because many patients would only flee to another merging hospital. (Tenn Report, ¶ 178; Tenn Rebuttal Report, ¶ 27.)
The Seventh Circuit nowhere stated or suggested that Dr. Tenn's use of diversion ratios for this purpose is improper. It merely cautioned that even if diversion ratios show that a proposed geographic market excludes significant competitors, it does not necessarily follow that the geographic market is defined too narrowly, as Dr. Tenn himself recognized. (See Tenn Report ¶ 100 n.194.)
4. Whether "closer substitutes" should have been included
Finally, defendants argue that Dr. Tenn failed to include any "closer substitutes" for the hospitals in the North Shore Area, particularly Rush, Lurie, Northwestern Memorial, and Presence St. Francis (a hospital very near NorthShore Evanston but excluded by Dr. Tenn because it did not compete significantly with any of the Advocate hospitals in the vicinity), before concluding that the market passed the hypothetical monopolist test. While there is some support in the record for inclusion of Northwestern Memorial, which the diversion ratios show to be a close substitute for some of the NorthShore hospitals, and Presence St. Francis, only three miles from NorthShore Evanston, defendants have admitted that even if those two hospitals are included in the geographic market, the post-merger market concentration would still be so high that the merger would be presumptively unlawful. (Tr. at 1890:5-1891:11.) Further, as the Seventh Circuit also recognized, Northwestern Memorial is a destination hospital that is not located in or near the North Shore Area, so it cannot fulfill commercial payers' need for a hospital providing local care in the North Shore Area. FTC, 841 F.3d at 475 n.5. This is also true of Lurie and Rush, and in any case, the Seventh Circuit did not find "comparable evidence about those centers as close substitutes for the hospitals of the merging parties," id., and neither does this Court. While they may draw patients from the North Shore Area, they remain "destination" hospitals unable to fulfill the need for local GAC services in the North Shore Area, and the data does not support the argument that they are "closer" substitutes for the defendant hospitals, from the standpoint of insurers, than other hospitals in the North Shore Area.
In light of the guidance of the Seventh Circuit and based on Dr. Tenn's persuasive analysis, the Court finds that plaintiffs have proven that the relevant geographic market is the North Shore Area.
B. Effect on Competition
After proving the relevant market, plaintiffs must answer the "ultimate question under [section 7 of the Clayton Act]: whether the effect of the merger 'may be substantially to lessen competition' in the relevant market." Phila. Nat'l Bank, 374 U.S. at 362.
1. Market Concentration
Plaintiffs may make a prima facie case of likely harm to competition by "showing that the proposed merger would result in 'a firm controlling an undue percentage share of the relevant market' as well as 'a significant increase in the concentration of firms in that market.'" OSF Healthcare, 852 F. Supp. 2d at 1074 (quoting Phila. Nat'l Bank, 374 U.S. at 363). If they can make this showing, plaintiffs establish that the merger is "inherently likely to lessen competition substantially," and it "must be enjoined in the absence of evidence clearly showing that the merger is not likely to have such anticompetitive effects." Phila. Nat'l Bank, 374 U.S. at 363. In other words, a "presumption of illegality arises." OSF Healthcare, 852 F. Supp. 2d at 1074 (citing Univ. Health, 938 F.2d at 1218).
Under the Department of Justice and FTC's Horizontal Merger Guidelines, a useful measure of market concentration is the Herfindahl-Hirschman Index ("HHI"), which is calculated by summing the squares of the individual firms' market shares. Merger Guidelines § 5.3. A market is "highly concentrated" if the HHI is above 2,500. Id. "Mergers resulting in highly concentrated markets that involve an increase in the HHI of more than 200 points will be presumed to be likely to enhance market power," although the presumption "may be rebutted by persuasive evidence showing that the merger is unlikely to enhance market power." Id.
NorthShore and Advocate have a 31% and 29% share of the North Shore Market, respectively. (PFF ¶ 44.) Merged, they would control 60% of the market, followed by Northwest Community (14%), Swedish Covenant, (8%), Vista East (6%), Northwestern Lake Forest (6%), and Presence Resurrection (6%). (Id.) Based on these market shares, the pre-merger HHI of the North Shore Area is 2,161. (Id. ¶ 45.) The merger would increase the HHI by 1,782 points to 3,943. (Id.)
Thus, the market concentration that would result from the merger is well beyond the level that the Merger Guidelines identify as presumptively likely to enhance market power. (Id.; Tenn Report ¶¶ 112-115.) Dr. Tenn's market concentration calculations, which defendants do not seriously challenge, are sufficient to establish the presumption of illegality. (Pls.' Proposed Conclusions of Law ("PCL") ¶ 3, ECF No. 446.)
Dr. Tenn performed additional alternative calculations of market concentration as a check on his results, and under all of the various approaches, the increase in market concentration was great enough to make the merger presumptively unlawful. (Tenn Report ¶¶ 116-122.)
2. Dr. Tenn's Analysis of Anticompetitive Effects
Although plaintiffs have established the presumption of illegality by establishing that the merger would cause high levels of market concentration, they do not rely solely on market concentration to establish that the merger would harm competition. See FTC v. Sysco Corp., 113 F. Supp. 3d 1, 61 (D.D.C. 2015) (citing United States v. Baker Hughes Inc., 908 F.2d 981, 992 (D.C. Cir. 1990)) ("The Herfindahl-Hirschman Index cannot guarantee litigation victories."). Among other evidence, they offer additional testimony of Dr. Tenn, who, after defining the relevant geographic market, proceeded to analyze the potential anticompetitive effects the merger might cause in the form of a rise in reimbursements to defendants due to a price increase.
Based on the diversion ratios and pricing data from the relevant hospitals, and assuming that the range of variable cost margins for commercial admissions (i.e., the difference between revenue and variable costs) at NorthShore hospitals was comparable to the range at Advocate hospitals, Dr. Tenn calculated that the merger would cause an average price increase of 8% across the six party hospitals in the North Shore Area, resulting in an annual increase of inpatient GAC reimbursement paid to those hospitals of about $45 million. (Tenn Report ¶¶ 178-84.)
Plaintiffs barely mention this anticompetitive effects evidence on remand, focusing instead on the Seventh Circuit's discussion of the geographic market and on the uncontroverted market concentration data, but defendants devote considerable space in their supplemental briefs to challenging Dr. Tenn's model. Because the Court has found that the market concentration evidence alone establishes the presumption of illegality, the burden of disproving anticompetitive effects falls on defendants, and they cannot logically carry this burden by attacking evidence that was unnecessary to establish the presumption in the first place. Additionally, defendants ignore the voluminous factual evidence, including "ordinary course" documents, showing that Advocate and NorthShore are close competitors who dominate the North Shore Area (PFF ¶¶ 47-52, 58-66) and whose merger would necessarily entail substantial harm to competition, see Sysco, 113 F. Supp. 3d at 61-62, 65-66, 69-70, which bolsters the presumption of anticompetitive effects and Dr. Tenn's conclusions. In light of all this other evidence, the Court need not address defendants' criticisms of Dr. Tenn's opinion of the merger's likely anticompetitive effects. Nevertheless, as the following discussion will demonstrate, even if plaintiffs were forced to rely on the anticompetitive effects portion of Dr. Tenn's opinion, Dr. Tenn's analysis is sound, and defendants' criticisms fail.
a. Inputs to Dr. Tenn's analysis
Dr. Tenn's model relied on three principal inputs: diversion ratios, contribution margins, and pricing data. The Court has already rejected defendants' argument that Dr. Tenn's use of diversion ratios impaired his analysis, but defendants take issue with the other two inputs as well.
Defendants argue that Dr. Tenn should not have used contribution margins, or as Dr. Tenn calls them, variable cost margins, as an input in his analysis because high contribution margins may actually be driven not by high prices and profits but by high fixed costs, and therefore should not be "conflat[ed] with market power." (Defs.' Post-Remand Br., at 6, ECF No. 557.) According to defendants, Dr. Tenn should instead have used operating margins, which account for fixed costs.
Dr. Tenn explained that his "model predicts that, when margins are higher [i.e., when the difference between revenue and variable costs is higher], there is greater incentive to raise price post-merger because the substitution that would be internalized would be more profitable." (Tenn Report ¶ 179.) If margins are zero, there is no incentive to increase price because "no additional profit would be internalized." (Id.) He did not use operating margin because the measure he used, "variable cost commercial margin for inpatient services," which excludes not only fixed costs but also "revenue and costs for non-commercial patients [such as Medicare, Medicaid, HMO or workers' compensation patients, all of whom fall outside the relevant product market] as well as other services (e.g. outpatient services), . . . captures the profit loss for those patients who respond to a price increase by switching to alternative hospitals." (Id. at n.300.) The Court agrees with plaintiffs and Dr. Tenn that this measure is appropriate because it is tailored to measuring effects related to the GAC services that this case concerns. But in any case, defense expert Dr. Thomas McCarthy admitted that the revised margin figures defendants advocate using would lead only to a "slightly lower estimate of the post-merger price increase of 6.9 percent," which is still above the 5% SSNIP threshold. (DX 5000, McCarthy Report, ¶ 104 n.159, ECF No. 452-5.)
Defendants also argue that Dr. Tenn erred by using relative pricing as the final input in his model, rather than actual claims data, as the defense experts did. They make much of this difference, but they do not adequately explain how the use of this data flawed or skewed Dr. Tenn's analysis. Dr. Tenn relied on relative pricing data that came from Advocate (Tenn Report ¶ 180), and defendants do not claim that the data was erroneous.
The Court finds no error in Dr. Tenn's use of any of the three inputs in his analysis.
b. Dr. Tenn's model generally
Defendants argue that, even if the inputs Dr. Tenn plugged into his model are not inappropriate individually, the model as a whole is flawed in a number of ways.
i. Price increase every time?
Defendants argue that Dr. Tenn's model has no predictive power because it will always produce a price increase. Dr. Tenn admitted that his model always predicts a price increase, if diversion ratios and contribution margins are positive. But that proviso makes all the difference, as Dr. Tenn demonstrated by applying his method to a hypothetical merger between NorthShore and Centegra, in McHenry County, far to the west of NorthShore. The diversion ratios between NorthShore and Centegra are low because those systems are not close substitutes due to the distance between their hospitals, so his method predicted a low price increase that was too small to be significant for antitrust purposes. (Tenn Rebuttal Report ¶ 52.) This shows that Dr. Tenn's method is useful because it reveals how strong the merged entity's profit-maximizing incentives to raise price will be based on their levels of substitution and potential profitability. The fact that the method predicts at least a small price increase whenever the inputs are positive does not represent a weakness.
This is because, as described above in Parts I.A.3 & I.B.2.a., (1) if diversion ratios are positive, then there is an incentive to raise prices because any lost volume at one party hospital is likely to be recaptured by another (i.e., substitution will be "internalized"), and (2) if contribution margins are positive (i.e. revenue is greater than the cost of providing services to the patient), then again there is an incentive to raise prices because it will be profitable to provide services to patients at one party hospital who are fleeing a hypothetical price increase at another merging hospital.
ii. Failure to address whether SSNIP would be profitable despite volume loss?
Defendants argue that Dr. Tenn's analysis is deficient because it does not address whether imposing a SSNIP would be profitable in light of the volume loss that would follow from any predicted price increase. But the Court has already explained, Dr. Tenn does exactly that by calculating, based on the levels of substitution between merging hospitals, that a price increase could be profitable in part because there would be little overall volume loss to the system. This criticism is without merit.
iii. Inconsistent with commercial realities?
Defendants argue that Dr. Tenn's analysis is inconsistent with commercial realities in the hospital industry. Hospitals do not unilaterally set prices for a given hospital in the way Dr. Tenn describes; rather, hospital systems settle on prices by way of bilateral bargaining with insurers on a system-wide basis. (See DFF ¶¶ 74-76; McCarthy Report ¶ 97.) Defendants argue that Dr. Tenn's price-setting model, geared toward predicting when a merged hospital system might have an incentive to raise prices at a particular hospital, is inconsistent with the commercial reality that hospital systems do not set prices for each hospital on an individualized basis. According to defendants, Dr. Tenn was unable to cite a case in which his model had accurately predicted a price increase following an actual, real-world merger, and in fact, defendants argue, there have been documented cases in which a hospital merger resulted in lower prices. (See Defs.' Post-Remand Br., at 9-10, ECF No. 557.)
Defendants may be correct that many hospital mergers actually result in lower prices, but it is defendants' burden to demonstrate that this particular merger will be one of those that has no anticompetitive effects. Defendants' experts focused more heavily on explaining their own bargaining-based models (which the Court will discuss in more detail below) than on demonstrating why Dr. Tenn's is inadequate. Their criticisms of Dr. Tenn's anticompetitive effects analysis were desultory and superficial, and therefore unconvincing. (See, e.g., Tr. at 1512:3-513:25 [Eisenstadt]; Tr. at 1224:7-8 [McCarthy].)
Dr. Tenn repeatedly explained that his analysis relied on a price-setting model because it is easier to understand than a bilateral bargaining-based model and certain economic literature, including some that specifically addressed the hospital industry, shows that the price-setting model and the bilateral bargaining model will "generate identical predicted post-merger price increases." (Tenn Rebuttal Report ¶ 16; see id. ¶¶ 14-15, 17; Tenn Report, ¶¶ 81 n.167, 177.) In fact, according to a recent article, "one way to think about the posted price model is that it is a reduced form way of capturing a complicated underlying bargaining relationship." (Tenn Rebuttal Report ¶ 15 (citing Martin S. Gaynor, A Structural Approach to Market Definition with an Application to the Hospital Industry, 61 J. Indus. Econ. 243, 261 (2013)).) Dr. Tenn used a price-setting model only "for ease of exposition," and to improve the "simplicity and transparency" of the analysis (Tenn Report ¶ 177), because "non-economists are (generally) more familiar with environments where suppliers set price (e.g., milk sold to consumers in the supermarket) compared to environments where suppliers negotiate price with their customers (e.g., hospital system contracting with MCOs)" (Tenn Rebuttal Report ¶ 17). Defendants do not directly dispute this explanation. (See id. ¶ 18; Tr. at 1347:20-25 [McCarthy].)
Defendants questioned the strength of the support Dr. Tenn cited for his position that a price-setting model and a bargaining model generate similar results (see, e.g., Tr. at 1512:3-513:25 [Eisenstadt]; see generally DFF ¶¶ 235-47), and Dr. Tenn's belief that his price-setting model somehow illustrates the anticompetitive effects of this transaction more clearly, although it conforms to the actual functioning of the market in question less closely, is certainly counterintuitive. But it is no more counterintuitive than defendants' experts' conclusion, discussed further below, that even if willingness to pay for access to defendants' systems increases, the merger will result in no price increase and possibly even a price decrease (see PFF ¶¶ 70, 74-75, 82-87). Defendants do not directly engage with and dispute Dr. Tenn's explanation for his use of the price-setting model, and the Court will not dismiss it based only on the relatively superficial criticisms defendants have made. The Court finds that Dr. Tenn has persuasively demonstrated that the merger is likely to cause a significant price increase resulting in a loss to consumers.
3. Defendants' Rebuttal of Anticompetitive Effects
Defendants attempt to rebut plaintiffs' evidence of anticompetitive effects by relying on (a) insurer testimony and (b) their experts' opinions.
a. Insurer testimony
Defendants contend that four of the Chicago area's six largest insurers—Aetna, United, Humana, and Land of Lincoln—testified that defendants' merger will strengthen competition and benefit consumers by allowing defendants to provide lower-cost, higher-quality care. (Defs.' Post-Remand Br., at 18, ECF No. 458.) Plaintiffs respond that most of the insurance executives who testified at the hearing expressed concern as well as support for the merger, and to the extent they expressed support, they admitted that they lacked a factual basis for it. (Pls.' Post-Remand Resp. Br., at 9, ECF No. 577.) Further, BCBSIL, the largest insurer in the area, is vehemently opposed to the merger. (PFF ¶ 74.)
The Court agrees with plaintiffs that the insurers' support for the merger was equivocal, unenthusiastic, and without a factual basis. In some cases, what lukewarm support the insurers mustered was expressed in terms of hope rather than expectation. (See, e.g., PX03004 ¶ 19 (Humana is "hopeful"); PX03005 ¶ 16 (Land of Lincoln is "hopeful").) Moreover, one insurer, [Redacted] (DFF ¶ 201.) The insurers are not in a position to know precisely how the merger will affect consumer welfare, and to the extent the insurers expressed support for the merger, for all the evidence shows, they, like [Redacted] , might have done so because they believed the merger would improve their own competitive positions. In short, the insurer testimony on this point is not credible and does not carry the weight defendants would give it.
b. Defense experts McCarthy and Eisenstadt
Defendants also contend that the merger will not cause a significant price increase, as shown by their economic experts, Drs. McCarthy and Eisenstadt, whose methodology, defendants argue, conforms more closely to the commercial realities of the hospital industry than Dr. Tenn's because it takes account of the parties' relative bargaining strength. Dr. McCarthy and Dr. Eisenstadt used the Hospital Merger Simulation Model, developed by FTC economists, which measures the relationship between actual prices negotiated by hospital systems and "willingness to pay" ("WTP"), a quantitative measure of a hospital system's desirability, or the "willingness to pay" for access to a system. (See DX6000, Eisenstadt Report ¶ 73, ECF No. 452-6; DFF ¶¶ 224-29; see generally Tenn Rebuttal Report ¶ 32.) The experts concluded that a significant price increase was unlikely; in fact, Dr. McCarthy calculated that, by one measure, the merger would lead to a price decrease of 3.3 percent. (DFF ¶¶ 230-31.) Dr. Eisenstadt concluded that the worst-case scenario was a moderate price increase leading to a gain to defendants of $11 million—approximately one-fifth the size of Dr. Tenn's predicted reimbursement increase of $45 million. (Id. ¶ 232.)
But in rebuttal, Dr. Tenn explained that these results are implausible and inconsistent with economic theory. The defense experts' work "impl[ies] a negative relationship between price and WTP, i.e., higher WTP is associated with lower prices"; an increase in a hospital system's bargaining leverage, as measured by WTP, should lead to an increase in price, or at a minimum, no change, but certainly not to a price decrease. (Tenn Rebuttal Report ¶ 35.) Dr. Tenn suggests that the defense experts likely failed to control for other important, unknown variables (an error he calls "endogeneity bias"), or they improperly measured WTP, inevitably skewing their analysis of its impact on price (an error he calls "measurement error bias"). (Id. ¶¶ 37-50.) To demonstrate the unsuitability of the defense experts' mode of analysis, Dr. Tenn simulated a merger of all 48 hospitals that Dr. McCarthy identifies as competing with at least one of defendants' hospitals. (See McCarthy Report ¶¶ 61-66.) Dr. Tenn calculated that, using the defense experts' mode of analysis, the post-merger price change would range from a 33% decrease, which is patently absurd, to a modest 6% increase, which is too small to be plausible for such a large area. (Tenn Rebuttal Report ¶¶ 51-54.)
Dr. Tenn's critique is convincing. The Seventh Circuit cited a laundry list of economic literature demonstrating that hospital mergers in concentrated markets tend to lead to significant price increases. FTC, 841 F.3d at 472-73. Defendants now ask this Court, faced with choosing between an expert economic analysis that is consistent with this literature and another that is inconsistent with it, to choose the one that is inconsistent with the literature the Seventh Circuit has cited, without adequately explaining why their merger is an outlier. "If anything, defendants' argument only reinforces the conclusion that there are serious and substantial questions requiring further determination by the FTC at the trial on the merits." OSF Healthcare Sys., 852 F. Supp. 2d at 1086 ("At this stage of the proceedings, the court is only determining whether there are 'questions going to the merits so serious, substantial, difficult and doubtful as to make them fair ground for thorough investigation, study, deliberation and determination by the FTC in the first instance and ultimately by the Court of Appeals.'") (quoting Univ. Health, 938 F.2d at 1218).
Defendants argue that, even if Dr. Tenn is correct that the merged entity would have an incentive to raise prices if the competitive landscape remained otherwise unchanged, his conclusion is of little usefulness because the competitive landscape would not remain unchanged. Competitors would reposition to attract customers within the North Shore Area, particularly by opening outpatient offices in the region. Northwestern and Presence, in particular, are already taking steps in that direction. (See DFF ¶¶ 248-58.) There was copious evidence that once patients establish a relationship with a physician at an outpatient office, that physician is able to influence where her patients will go for inpatient services. (See, e.g., Tr. at 345:19-46:10 (Dechene of Northwestern testifying that outpatient facilities and doctor's offices are "front doors" to the hospital); id. at 1116:14-18 (Beck of United testifying that "a member's physician relationship influence[s] where they seek hospital care"); JX 3, Bagnall Dep. at 37:2-8, ECF No. 453-3 (testifying on behalf of University of Chicago Medical Center that "patients don't shop for inpatient providers, they shop for physicians" and "it's the physician who makes the decision of what inpatient facility that patient goes to"); JX 19, Maxwell Dep. at 94:1-24, ECF No. 454-9 (testifying on behalf of Humana that hospitals "extend their geographic breadth" by opening outpatient centers and doctor's offices further from the hospital, and the doctor "plays a significant role [in determining] where [a] patient goes to seek care").)
The Court recognizes that defendants' repositioning argument is alluring. The merging hospitals in the North Shore Area are ringed by non-party competing hospitals, especially the five Dr. Tenn included in the North Shore Area; it is tempting to believe that these numerous hospitals could reposition to attract patients from the merging hospitals' service areas, or to present themselves to insurers as viable alternatives to the merging hospitals, if the merged entity were to attempt to impose a price increase. But this argument is merely another version of an argument that the Seventh Circuit rejected, the argument that there will be no anticompetitive price effects because the diversion ratios show that patients within the North Shore Area view hospitals outside the area as close substitutes. Regardless of this alleged repositioning, the Seventh Circuit explained that there is overwhelming evidence that insurers believe that their networks generally must include either Advocate or NorthShore in order to be successful because many patients are not willing to travel outside their immediate geographic area for inpatient hospital care, and insurers must be responsive to the preferences of these patients who are not willing to travel by offering them a local in-network option. FTC, 841 F.3d at 474-76. While hospitals outside or on the fringes of the North Shore Area may be able to draw increasing numbers of patients from within the Area via their repositioning efforts, the merging hospitals will nevertheless "have market power over the insurers who need them to offer commercially viable products to customers who are reluctant to travel farther for [GAC services]." Id. at 476. Viewing the evidence in light of this guidance from the Seventh Circuit, the Court cannot accept that the repositioning of competitors will offset or prevent the anticompetitive effects that Dr. Tenn has identified without stronger evidence than the generalized testimony defendants have offered.
4. Efficiencies
Although the defense has never been sanctioned by the Supreme Court, the Horizontal Merger Guidelines and some lower courts recognize that defendants in a horizontal merger case may rebut the government's prima facie case by presenting evidence of efficiencies offsetting the anticompetitive effects. See FTC v. H.J. Heinz Co., 246 F.3d 708, 720 (D.C. Cir. 2001). Where the merger would result in high market concentration levels, as in this case, the defendants must provide proof of "extraordinary efficiencies" based on a "rigorous analysis" that ensures that the proffered efficiencies represent more than "mere speculation and promises about post-merger behavior." See id. at 720-21. Further, the efficiencies must be "merger-specific," i.e., "they must be efficiencies that cannot be achieved by either company alone." Id. at 721-22. Defendants contend that the merger will result in significant efficiencies providing substantial benefits to consumers and offsetting the anticompetitive effects. Plaintiffs respond that defendants have not sufficiently proven the claimed efficiencies.
a. Physician reimbursement savings
As an initial matter, once the merger is consummated, NorthShore will immediately move to Advocate's lower reimbursement rates, as required by Advocate's contracts with insurers. This change will result in immediate savings of approximately $30 million to insurers (and by extension, to consumers).
Plaintiffs correctly explain that these savings are at best a temporary benefit. They do nothing to alleviate the problem of the merged firm's market power, which will allow it to raise prices when it renegotiates its reimbursement rates in the future; therefore, they do not offset the anticompetitive effects of the merger.
b. High-Performing Network
Defendants claim that, once the merger is consummated, the merged entity ANHP will be able to offer an improved version of Advocate's narrow HPN product. Advocate has experimented with Advocate-only networks, offering them to its own employees and to individuals and small employers as part of the "BlueCare Direct with Advocate" ("BCD") plan, in partnership with BCBSIL. These narrow-network plans differ from traditional health maintenance organization ("HMO") plans because they consist of one or very few providers, they do not rely on a primary care physician acting as a "gatekeeper" whom patients must see for referrals in order to see specialists, and they permit patients to seek care out of network at a fifty-percent-reimbursement rate. (Tr. at 1420:24-21:13, 1422:6-18 [Sacks-Advocate].) Defendants claim that Advocate has been informed by insurers and others in the industry that it cannot offer a commercially-viable Advocate-only narrow network product attractive to large employers in the Chicago area unless it fills its geographic gap in the north suburban area east of Interstate 94. Acquiring NorthShore will fill that gap in Advocate's system. For its part, NorthShore claims that it cannot offer an HPN or similar ultra-narrow network without a merger because it lacks sufficient geographic coverage and it lacks Advocate's expertise in risk-based contracting and population health management, which are critical to the success of any such product. (DFF ¶¶ 296-304.)
Defendants claim that, if the merger is consummated, the merged entity will offer its HPN product at the same price that its current BCD product commands, about 10% below the price of BCBSIL's HMO plan. (Id. ¶ 271.) Consumers who enroll in the HPN will receive a narrow network product with broader geographic coverage than Advocate had previously been able to offer, at the same low price. Defendants claim that Advocate's research shows that there is strong demand for an ANHP HPN, and they calculate that the savings of consumers who switch from a higher-priced plan to the HPN will fall somewhere between $210 million and $500 million in the aggregate. (Id. ¶¶ 274-95.) According to defendants, even if plaintiffs and Dr. Tenn are correct that the merged entity will be able to raise prices, costing consumers over $50 million in the aftermath of the merger, the savings the HPN will generate for consumers will greatly outweigh the price increase.
Plaintiffs argue that the HPN is not a cognizable, merger-specific efficiency for a number of reasons.
i. The HPN is already marketable to employers
The evidence that Advocate cannot sell an Advocate-only narrow network product to employers without expanding east of I-94 is thin. It primarily consists of insurers' conclusory statements to that effect. The record does not show the basis for this belief or whether it is the product of any serious analysis, and some of the insurers have offered conflicting testimony. (See Pls.' Post-Remand Resp. Br. at 9.) Additionally, these statements are of uncertain credibility. Defendants themselves have argued that BCBSIL, for example, views single-provider narrow networks as a threat to its own business because, if a healthcare provider can satisfy all of its patients' healthcare needs by itself, then the patients (or their employer) may as well pay the provider directly rather than the insurer middleman. (See DFF ¶¶ 211-17.) There is nothing in the way of expert testimony or economic analysis to support the position that Advocate cannot offer its Advocate-only narrow network product to employers without expanding east of I-94, and in fact one employer has quite successfully done so: Advocate itself. (Tr. at 1420:3-22 [Sacks-Advocate].)
ii. Defendants have not proven that the savings of consumers who switch to the HPN from another plan create an efficiency
In support of their position that the ANHP HPN will generate massive consumer savings, defendants offer the opinion of Dr. Eisenstadt, who makes three alternative estimates of these consumer savings, each calculated by a different method.
For his first estimate, he compares the price of the proposed HPN to the most similar plan currently available. Aetna currently offers a plan known as Aetna Whole Health ("AWH") that includes Advocate and NorthShore as the only two in-network providers. At present, AWH is only available as an exchange product for small employers; a broader version of the product is available to large employers, but this version includes Rush as well. (Eisenstadt Report ¶ 51 n.77.) Dr. Eisenstadt assumes that (a) Aetna will offer AWH to large employers at the same price it currently charges for the exchange version of the product, (b) the ANHP HPN price will be the same as the price of Advocate's current BCD product, as defendants claim it will (see DFF ¶ 271), and (c) all buyers of the ANHP HPN would otherwise have bought AWH, so each of them saves the difference in price between AWH and the ANHP HPN (Eisenstadt Report ¶¶ 52, 54). Based on all those assumptions, he calculates how many enrollees the HPN would need to attract in order for the consumer savings to offset the approximately $50 million in increased payments to the merging firms from commercial health plans that Dr. Tenn identified, concluding that the ANHP HPN need only attract about 1% of the Chicagoland employer-sponsored insurance market. (Id. ¶ 54, Table 4.)
The trouble with this approach is the assumption that all ANHP HPN enrollees would otherwise have purchased a large-employer version of AWH offered at the same price, which is essentially speculative. Aetna does not currently offer AWH as a large employer plan, and there is no reliable evidence showing how many enrollees the plan would have or where they might live if it did. There is also no reliable evidence as to how many enrollees an ANHP HPN would attract, nor is there even evidence of how many enrollees an Advocate-only HPN such as BCD would attract if it were offered to large employers, which Advocate insists, without explanation, it cannot do without explaining why. In short, there is no firm evidentiary ground for assuming that Dr. Eisenstadt's hypothetical Aetna plan, similar to AWH but offered to large employers, will ever exist and, if it did exist, that it would be the second choice of all ANHP HPN enrollees. (Tenn Rebuttal Report ¶ 122.)
For his second estimate, Dr. Eisenstadt assumes that all customers who enroll in the ANHP HPN will switch from some plan other than AWH. Because these customers are switching from a plan that offers access to more hospital systems, the net consumer benefit they get from switching to the HPN will necessarily be smaller than the pure difference in price; the benefit will have to be reduced to some degree because the customer is sacrificing hospital options by switching. For such customers, the true value of the consumer benefit derived from switching to the ANHP HPN could range anywhere from zero to the price difference between AWH and the HPN, depending on the breadth of the hospital options the consumers give up by switching to the narrower HPN. (PFF ¶ 111.) Dr. Eisenstadt assumes that "consumers are uniformly distributed between these two bounds, so that the average consumer benefit is one half the price difference" between AWH and the HPN. (Tenn Rebuttal Report ¶ 123.) By that measure, the ANHP HPN need only attract between 1.7% and 2.7% of the Chicagoland employer-sponsored insurance market to offset any anticompetitive effects. (Eisenstadt Report, Table 4.) But there is no basis for the assumption that the consumer benefit that accrues to customers who switch to the HPN will be uniformly distributed throughout the possible range of values. (Tenn Rebuttal Report ¶ 123.) Therefore, this estimate is essentially speculative.
For his third estimate, Dr. Eisenstadt assumes that the consumer benefit equals the price difference between the ANHP HPN and Blue Precision ("BP"), an HMO plan offered by BCBSIL. Using the BP plan as a benchmark, Dr. Eisenstadt calculates that the ANHP HPN need only attract between 2.8% and 4% of the Chicagoland employer-sponsored insurance market to offset any anticompetitive effects. But Dr. Tenn explains that Dr. Eisenstadt has not adjusted for the fact that BP is a more desirable plan because it offers wider provider options than just Advocate and NorthShore, unlike the ANHP HPN. (Id. ¶ 124.) The pure price difference between BP and the HPN does not alone capture the benefit consumers would realize by switching from BP to the HPN because the consumers making that switch are not merely paying less; they are also getting less. Dr. Tenn calculates that on an adjusted basis, the true "upper bound" of average savings per member per year is approximately $432, which would require ANHP to enroll at least 124,000 members (or about 3% of the market) in the HPN in order to offset the anticompetitive effects of the merger, and the true "lower bound" of average savings per member per year might well be near zero. (Id. ¶ 125.) This differs starkly from Dr. Eisenstadt's calculations of a lower bound of $298 and an upper bound of $1,426. (See Eisenstadt Report, Table 4.)
Dr. Tenn's critiques of Dr. Eisenstadt's opinion are persuasive. The upshot is that Dr. Eisenstadt's analysis sheds little light on what the true level of savings generated by the HPN might turn out to be, and Dr. Eisenstadt's calculations are essentially "uninformative regarding whether consumer benefits are likely to exceed consumer harm." (Tenn Rebuttal Report, ¶ 127, ECF No. 450-4.)
iii. Estimates of HPN enrollment are speculative and unsubstantiated
Dr. Eisenstadt's analysis might be helpful at least as a rough estimate of the range of potential savings if defendants were able to demonstrate with any certainty how many enrollees the HPN might attract. But they have not provided serious enrollment projections for the proposed ANHP HPN product. As support for their position that there would be high employer interest in the ANHP HPN, defendants rely principally on an employer survey conducted by their expert Dr. Kent Van Liere. (DFF ¶¶ 291-94.) Plaintiffs call this study "highly flawed," based in part on the opinion of their own marketing survey expert, Gary T. Ford. (PFF ¶ 115) (citing PX6023, Ford Report ¶ 78, ECF No. 450-7). Ford explains that the study suffers from a number of weaknesses. For one, it lacks a control group. It essentially just asks people whether their employers would be interested in the product, which is suggestive and not probative of where the respondents live or work or how many might be interested in an Advocate-only HPN (which Advocate could offer without merging with NorthShore). (Ford Report ¶¶ 29-44.) Further, Ford explains that the study is methodologically unsound in a number of ways and lacks any sufficient qualitative basis for its conclusions. (Id. ¶¶ 63-68.)
Defendants take issue with some of these criticisms, but in general, the Court finds Ford and plaintiffs more persuasive. But even if Ford overstated the significance of some of the flaws he identified, Dr. Van Liere's study does little more than "gauge employer interest," as defendants themselves put it (see DFF ¶ 295), in the ANHP HPN; it does not amount to a rigorous economic analysis or serious business-planning projection of the potential enrollment in the HPN. The Court agrees with defendants that they need not demonstrate with certainty the amount consumers will save, but they must provide firmer, more rigorous proof than they have offered. See FTC v. ProMedica Health Sys., Inc., No. 3:11 CV 47, 2011 WL 1219281, at *40 (N.D. Ohio Mar. 29, 2011) ("Projections of efficiencies may be viewed with skepticism, particularly if they are generated outside of the usual business planning process."); see also United States v. H & R Block, Inc., 833 F. Supp. 2d 36, 91 (D.D.C. 2011) ("The difficulty in substantiating efficiency claims in a verifiable way is one reason why courts generally have found inadequate proof of efficiencies to sustain a rebuttal of the government's case.") (internal quotation omitted)); cf. Sysco, 113 F. Supp. 3d at 82 (even efficiencies estimates that are the "product of meticulous analysis and planning" based on a "back-breaking amount of information" may be inadequate if the merging parties cannot demonstrate what amount of the claimed efficiencies are merger-specific). It may well be possible that the HPN will generate sufficient enrollment to offset any anticompetitive effects caused by the merger, and perhaps defendants will be able to show as much at the FTC administrative hearing, but the opinions of Dr. Eisenstadt and Dr. Van Liere do not establish the level of enrollment in the plan, or the amount consumers will save by enrolling in the plan, with reasonable confidence. Their estimates are essentially speculative and unable to withstand the "rigorous analysis" that the Clayton Act requires. Heinz, 246 F.3d at 721. Defendants therefore fail to shoulder their burden of demonstrating that consumer benefits due to enrollment in the HPN will offset anticompetitive effects of the merger.
The Court agrees with plaintiffs that, for the reasons discussed above, defendants have not carried their burden of proving that efficiencies will offset the anticompetitive effects. Plaintiffs have demonstrated a likelihood of success on the merits.
Plaintiffs also argue that the HPN is not a merger-specific benefit because NorthShore could contract to join the HPN without merging. The Court need not reach this argument because defendants' failure to prove that the merger's benefits outweigh the possible anticompetitive effects is a sufficient basis for concluding that they have not met their burden of establishing an efficiencies defense. --------
II. BALANCING THE EQUITIES
Because plaintiffs have demonstrated that they are likely to succeed in demonstrating that defendants' merger would cause harm to competition and damage consumers' interests, they have "created a presumption in favor of injunctive relief." Sysco Corp., 113 F. Supp. 3d at 86. Nevertheless, the Court must weigh the equities to determine whether an injunction is in the public interest. Id.
The public has strong interests in the effective enforcement of the antitrust laws and in preserving its ability to order effective relief if it succeeds after a trial on the merits. Id. These interests are plainly served by entering an injunction. Plaintiffs have shown a likelihood of success on the merits, and if the Court does not enter an injunction and allows the parties to merge, then it may be more difficult to order effective relief after a trial on the merits by "unscrambling" merged assets to "recreate pre-merger competition." See Heinz Co., 246 F.3d at 726. Defendants argue that consumers' interests in receiving the benefits of the transaction outweigh these public interests, but the Court has already found that defendants have not shown a likelihood that the consumer benefits of the merger outweigh its anticompetitive effects. Thus, any interest consumers may have in the consummation of the merger is too small to outweigh the public's interest in the enforcement of the antitrust laws. See Saint Alphonsus Med. Ctr. - Nampa, Inc. v. St. Luke's Health Sys., Ltd., No. 1:12-CV-00560-BLW, 2014 WL 407446, at *25 (D. Idaho Jan. 24, 2014) ("[T]he Clayton Act is in full force, and it must be enforced. The Act does not give the Court discretion to set it aside to conduct a health care experiment."), aff'd 778 F.3d 775 (9th Cir. 2015). The equities weigh in favor of the injunction.
CONCLUSION
Because plaintiffs have demonstrated a likelihood of success on the merits and that the equities weigh in their favor, their motion for preliminary injunction [152] is granted. Defendants are enjoined from consummating their proposed merger by affiliating or acquiring each other's assets or other interests, pending final disposition of the FTC's full administrative proceeding on the merits.
SO ORDERED.
ENTERED: March 16, 2017
/s/ _________
HON. JORGE L. ALONSO
United States District Judge