Call it a comeback, now that they’ve been here for years

FTC successes in hospital merger enforcement during the Obama administration

In the 1990s, the Federal Trade Commission and the US Department of Justice lost seven hospital merger cases in a row before taking a nearly 10-year hiatus from challenging such tie-ups.1 In this time, the FTC turned losing into a teachable moment, using the courtroom defeats as a fresh data set with which to analyse the effects of hospital mergers and build support for its arguments in favour of blocking future hospital consolidation. The agency came out of its brief retirement in 2008 when it successfully challenged a hospital deal between Evanston and Highland Park. Since that win, the FTC has been unbeatable, with courts granting practically every injunction the FTC has requested against proposed hospital mergers.2 This has made hospital mergers practically impossible, where deals are continually blocked by the FTC – a stark difference from the pre-­hiatus period when the agencies could not buy a win. And there are no signs that this will change, as the FTC has found success even in the face of (rare) losses at the district court level and new healthcare regulations, including the Patient Protection and Affordable Care Act (ACA).

Why the slump?

During the mid-to-late 1990s, the agencies and state enforcers lost all seven of the hospital mergers they challenged.3 Courts rejected the agencies’ arguments with respect to relevant geographic market definition, entry, potential anti­competitive effects or that the deals would lack any pro-competitive benefits.4 An extensive regulatory framework governing hospitals at federal, state and local levels, combined with broad geographic markets based on the courts’ acceptance of a test called the Elzinga-Hogarty patient flow analysis, made what used to be near-certain courtroom victories for the FTC considerably more difficult.5

The agencies’ failed hospital merger challenges: 1994-1999

Merger

Challenging agency

Reason for loss

Hospital Board of Directors of Lee County, d/b/a Lee Memorial Hospital/Cape Coral Medical Center Inc (1994)

FTC

State action doctrine immunises the proposed acquisition from federal antitrust laws.

Freeman Hospital/Tri-State Osteopathic Hospital Association, d/b/a Oak Hill Hospital Inc (1995)

FTC

The court rejected the FTC’s relevant geographic market.

Mercy Health Services/Finley
Tri-States Health Group
(1995)

DOJ

The court ruled the geographic market was larger than alleged by the government

Butterworth Health Corporation/Blodgett Memorial Medical Center (1997)

FTC

FTC failed to show that the market power would be exercised post-merger due to the efficiencies presented and the non-profit status.

Long Island Jewish Medical Center/North Shore Health System Inc (1997)

DOJ

The government’s relevant product and geographic markets were rejected.

Tenet Health Care Corporation/Poplar Bluff Physicians Group Inc, d/b/a Doctors Regional Medical Center (1999)

FTC

The FTC failed to establish a relevant geographic market.

Sutter Health/Summit Medical Center (1999)

California AG

The state’s geographic market was rejected as too narrow and expanded ultimately, including competition from numerous other hospitals.

During this losing streak, courts denied the FTC’s preliminary injunctions for two primary reasons: they rejected the FTC’s alleged geographic markets as too narrow, instead relying upon broad geographic markets created by the now-outdated test; and they believed that the live-saving mission and non-profit status of the merging hospitals meant that they lacked the motivation to use their market power to increase prices. What’s more, courts believed that the heavy regulation of the healthcare industry meant that competition was less important in healthcare markets.6 This framework had not previously proven an obstacle for the agencies when asking for an injunction; the FTC had prevailed in all but one of the cases it challenged in the 1980s.7 But both agencies quickly found themselves on the wrong end of a losing battle.

Geographic markets

During the agencies’ string of losses, the definition of the relevant geographic market was heavily disputed as courts began accepting hospitals’ claims that they, in fact, served wide swathes of territory and competed with other hospitals miles away. The precedent was set in Rockford Memorial, when the court accepted the use of patient flow analysis in the Elzinga-Hogarty test to measure patient inflows and outflows to determine the geographic market.8 Using Elzinga-Hogarty, the geographic market keeps expanding until the inflows and outflows are small enough that they are assumed to indicate a lack of alternatives. With the larger geographic areas, more competitors were included in the relevant geographic market – reducing the likelihood of anticompetitive effects. This flow analysis became fundamental in disputing, and ultimately expanding, the FTC’s relevant geographic markets, leading to the agency’s courtroom defeats.

In Freeman, the FTC argued for a relevant market that was a 54-mile ­diameter around the two merging hospitals. The hospitals, relying on Elzinga-Hogarty, argued that the relevant geographic market included the 13 surrounding counties – an area measuring nearly 100 miles in diameter.9 The hospitals’ market included a dozen more hospitals than the FTC’s analysis. The expansion of the relevant geographic market and the subsequent inclusion of the additional hospitals led the court to find the market to be highly competitive.

Similarly in Long Island, the courts denied the FTC’s request for a preliminary injunction because its geographic market was too narrow.10 The FTC did not specify a precise relevant geographic market, but noted that generally the relevant geographic market was Nassau and Queens counties – specifically excluding nearby Suffolk and Manhattan counties.11 The courts’ expansion of the relevant geographic market to include New York City for tertiary care services led it to conclude: “it would be virtually impossible for the government to obtain injunctive relief if Manhattan were included in the geographic market.”12 In expanding the market, the court relied on patient origin data to note that the merging hospitals received patients from Queens, Nassau and Suffolk counties. Additionally, patients in those counties travelled to Suffolk, Nassau, Queens and Manhattan for medical treatment.

Non-profit status

There also seemed to be a halo above the non-profit status of the hospitals proposing the mergers. The courts favoured the hospitals in comparison to the insurance providers, who were often the leading voices complaining about the anticompetitive effects of the proposed hospital mergers. This treatment came from an emphasis on the merging hospitals non-profit status, their supposedly inherent altruistic motivations and their life-saving contributions to the communities served. Instead, it was believed that insurance companies, not the hospitals, were primarily responsible for driving up healthcare costs and consolidation would give the providers more leverage in their negotiations.

The agencies initially struggled with courts highlighting the non-profit status of the merging hospitals as evidence that the deals wouldn’t harm consumers. For example, the court in Carilion found that the “defendants’ non-profit status also militate[d] in favor of finding their combination reasonable.”13 In Butterworth, the court decided that the proposed merger would not harm competition because the post-transaction board of directors of the combined entity would have a community-oriented composition, making it unlikely that they would harm consumers through price increases.14 And the chairman had convincingly testified to the hospital’s motivation to lower healthcare costs and improve the quality of care provided to the community.15 This testimony was “consistent with the hospitals’ altruistic motivations”,16 and as such was an important factor “support[ing] the defendants’ contention that community service, not profit maximisation, is the hospitals’ mission.”17 The life-saving motives were essentially deemed more significant than those of other hospitals, and were thus granted more weight than the traditional good motives pursued by other non-profit entities outside the hospital context.18

Similarly, witnesses employed by the hospitals were given more weight than the agencies’ expert economists, especially when explaining the efficiencies created by the transaction. Hospitals were held up on pedestals, their altruistic missions seemingly providing their stated transaction efficiencies with more credence than traditionally granted to merging parties. The testimony of hospital executives was taken on face value to rebut the government’s prima facie case. For example, in Long Island, the court noted that the main reasons for the merger were “to continue and advance the high quality of treatment of the hospitals’ patients, to seek advancement of medical technology, to foster the education and training of its physicians, to further pursue medial research at both hospitals and to avoid duplication, thereby reducing costs.”19

However, throughout this run of success for merging hospitals, courts caveated this deference to hospitals, stating they gave “only limited and non-determinative effect to the not-for-profit status”.20 Courts continually noted that “there is nothing inherent in the structure of the corporate board or the non-profit status of the hospitals which would operated [sic] to stop any anticompetitive behavior.”21 Yet the only preliminary injunction granted during this time period was overturned because the district court placed “inordinate emphasis on price competition” and should have considered the hospitals’ claim that the transaction would lead to better patient care.22 It was clear that, with respect to competition in the healthcare market, the courts favoured the hospitals over the complainants – insurance companies that, as for-profit entities, did not merit the same type of goodwill that hospitals and their boards were afforded.

Back on track

All streaks come to an end, and the agencies’ losing streak was no different. Except the end of the streak didn’t come with an FTC win – it simply stopped challenging hospital mergers altogether. But instead of simply letting hospital deal review go by the wayside, the FTC used the time to re-evaluate its hospital merger strategy. After a nearly 10-year hiatus from challenging hospital mergers in the 2000s, the FTC undertook to develop a different approach using the lessons learned from its losses. The FTC retooled its arguments, focusing on improved methods for defining geographic markets, better illustrating anticompetitive effects and moving courts away from their characterisation of merging hospitals as do-gooders without profit maximisation motives. This strategy worked. The FTC successfully challenged Evanston/Highland Park in 2008 and has won all eight hospital mergers it has challenged since. And the two challenges that were not litigated led to the hospitals abandoning those transactions.23

The belief that competition was not important because non-profit hospitals were inherently good and would not raise prices even if the merged hospital would be able to, had become so widespread that merging parties argued that non-profit status should protect almost any hospital deal, even “a ‘three-to-two’ combination”.24 In fact, in Rockford/OSF, the defendants did not even attempt to challenge the FTC’s market share and market concentration evidence, believing that, under recent case law, doing so was unnecessary to convince a court to deny injunctive relief.25

But, after its string of losses and a break from challenging hospital mergers, the FTC had natural experiments that it could use to rebut this proposition by determining what actually happened to hospital prices post-merger. The joint FTC and DOJ report produced during the FTC’s hiatus noted that the United States spends significantly more on healthcare than other countries – approximately 14% of gross domestic product in 2002.26 The agencies observed a dramatic cost increase in the early 2000s that was attributed primarily to price increases in inpatient hospital care and pharmaceuticals.27 Inpatient hospital care prices had increased, although inpatient hospital visits were decreasing. After looking back at the hospital mergers the courts cleared in the 1990s, the agencies had evidence to show that, when hospitals and other non-profits achieve market power, their pricing behaviour does not systematically differ from that of for-profits. The agencies’ analysis attributed hospitals’ price increases to their “increasing ability to negotiate higher prices from private payers”.28 Backed by evidence of the same price increases that the FTC had predicted would happen if the challenged mergers were allowed, the agencies could move out of the hypothetical and calculate the actual effects of hospital consolidations, even among non-profit hospitals.

Armed with this new evidence, the FTC was able to convince courts that non-profit hospitals are not deserving of special treatment. In ProMedica, the buyer claimed that it only sought a cost-coverage ratio of a small per cent from commercial health plans. However, the district court found that ProMedica used its bargaining leverage to obtain the most favourable reimbursement rates possible from commercial health plans, a claim that was conceded by ProMedica’s own witnesses.29 ProMedica was no different than a profit-maximising business in any other industry, and its non-profit status did not prevent it from exercising its market power.

Likewise, in OSF Healthcare, the merging hospitals argued that non-profit hospitals, unlike for-profit entities, were not maximising profits in highly concentrated markets.30 The district court, however, disagreed and found that the evidence in the case showed that non-profit hospitals seek to maximise the reimbursement rates they receive. The court also noted its disagreement with the conclusion in Butterworth that non-profit hospitals eschewed maximum profit, citing expert testimony that non-profit hospitals will negotiate higher rates whenever they can.31

Narrowing geographic markets

Armed with the evidence to convince courts that non-profit status should not insulate hospital mergers from antitrust challenges, the FTC still needed to overcome its problem convincing courts of its proposed geographic markets. In its past losses, “judicial acceptance of implausibly large geographic markets [resulted in] judicial approval of mergers that would not be permitted in any other industry . . . .”32 In order to win on geographic market definition, the FTC needed to move courts away from the Elzinga-Hogarty test.

The FTC succeeded in showing that the test was invalid and unreliable.33 The key problem with the test is that it assumes that patients would simply use a hospital outside of the core geographic area if the merger resulted in price increases at the merging hospitals. In fact, there are many reasons, in addition to cost, why patients travel (or choose not to do so) for medical services, such as reputation of the facilities, recommendations by other physicians, and convenience and proximity to patients’ communities. Without more, it was premature to assume patients would travel solely due to a price increase. Instead, the FTC claimed that the hypothetical monopolist test of the Horizontal Merger Guidelines, which is used in merger analysis in all other industries, is appropriate for hospital mergers. With evidence highlighting the flaws in the Elzinga-Hogarty ­analysis – and the observed price increases showing that non-profit hospitals are profit-­maximising competitors – courts began to apply the same hypothetical monopolist test used to show anticompetitive effects in mergers in other industries. The FTC was then free to analyse the actual competitive effects of hospital mergers under the typical merger framework. With this tool in hand, the FTC could overcome the obstacles that led to its losses and convince courts that anticompetitive effects should drive merger analysis, just like other industries.

The FTC was able to provide courts with factual support for their narrowly defined geographic markets, disproving the approach based on patient flow data that hospitals had relied on to expand geographic markets in prior agency losses. In Evanston, the FTC noted that defining a relevant market based on analysis of post-merger price increases reflects the fundamental relationship between market definition and competitive effect analysis, and was not circular as the respondent had alleged.34 Consistent with the Horizontal Merger Guidelines, the proper market was determined to be the “smallest possible group of competing products (or geographic area) over which a hypothetical monopolist that sells those products could profitably impose a [small but significant price increase].”35

More recently, in Penn State Hershey, the Third Circuit disagreed with the district court’s decision to employ the Elzinga-Hogarty test instead of the hypothetical monopolist test, which, the district court believed, resulted in too narrow a geographic market. The circuit court noted that the patient flow data the district court used was “particularly unhelpful in hospital merger cases” as it “did not consider that Hershey is a leading, academic medical center that provides highly complex medical services.”36 Ultimately, the circuit noted that “relying solely on patient flow data is not consistent with the hypothetical monopolist test.” It’s now clear that the hypothetical monopolist test is the correct test to determine the relevant geographic market for hospital merger cases. In fact, in reversing the district court’s denial of a preliminary injunction, the Third Circuit went as far as to refer to Elzinga-Hogarty as a “discredited economic theory”.37

Losses – or not?

Not only have the agencies won numerous challenges in district courts, they have been able to turn their rare “losses” at the district court level into wins on appeal.

In both Advocate and Penn State Hershey, the FTC lost at the district court level because the courts found the government defined the geographic markets too narrowly, but won at the circuit court level. In Advocate, the district court denied the FTC’s motion for an injunction because the hypothetical monopolist test used by the FTC’s economic expert to define the relevant geographic market was too circular to show a likelihood of success on the merits. But the Seventh Circuit reversed, as the evidence from the hypothetical monopolist test supported the distinction between academic medical centres and other hospitals. The court noted that the relevant geographic market does not include every competitor, but only the area of effective competition – the competitors that will discipline attempts by the merging entities to raise prices. Likewise, as described above, Penn State Hershey was reversed because the district court erred in “both its formulation and its application of the proper legal test” for determining the relevant geographic market.38

After losses at the district and circuit courts, the Supreme Court – in a unanimous decision – sided with the FTC’s claim that Phoebe Putney’s merger withPalmyra Medical Center was not protected from the antitrust laws due to state action immunity. Although the FTC ultimately did not pursue a case on the merits because divestitures were prevented by Georgia’s certificate of need laws, the FTC was able to extract a consent agreement from Phoebe Putney.

A is (still) for antitrust

The FTC’s re-emergence in challenging hospital mergers coincided with the debate and ultimate passage of the ACA in March 2010. With the introduction of the ACA, it was thought that it would provide a new rational for hospital mergers as part of a broader movement to reduce healthcare costs in the United States. However, the ACA has not influenced the FTC’s decisions to litigate hospital mergers that it believed would be anticompetitive, nor have courts relied upon the ACA in denying the FTC’s challenges. In fact, the FTC has enjoyed continued success in blocking hospital mergers, even after implementation of the ACA.

For example, in Penn State Hershey, the district court noted that the FTC needed to adapt to “an evolving landscape of healthcare that includes, among other changes, the institution of the ACA, fluctuation in Medicare and Medicaid reimbursement” in denying its request for an injunction.39 As the district court saw it, “the same federal government under which the FTC operates has created a climate that virtually compels institutions to seek alliances such as the hospitals intend here.” But the Third Circuit found the district court’s reliance upon the ACA produced “analysis economically unsound and not reflective of the commercial reality of the healthcare market.”40 The implementation of the ACA and its emphasis on reducing inefficiencies by providing integrated services to patients has not been a green light for hospital consolidation.

Conclusion

If one were to ask antitrust practitioners in the mid-1990s to predict the industry where the agencies would have the most courtroom success over the next 20 years, nobody would have imagined it would be hospital mergers. But by taking some time off to reflect on their approach, the FTC turned what was once a great weakness into arguably its greatest area of strength. District court losses have turned into wins. Regulations such as the ACA, that some believed would allow for merging hospitals to argue that efficiencies are more important than competitive effects, have not supplanted traditional antitrust analysis. And it is unlikely this will change anytime soon.

Notes

  1. See FTC v. Freeman Hosp., 69 F.3d 260, 272 (8th Cir. 1995) (holding that the FTC failed to establish the relevant market); U.S. v. Mercy Health Serv., 902 F.Supp. 968, 975 (N.D. Iowa 1995), vacated as moot, 107 F.3d 632,634 (8th Cir. 1997) (finding that the FTC failed to establish the relevant market; the deal was subsequently abandoned by the hospitals); FTC v. Butterworth Health Corp., 946 F. Supp 1285 (W.D. Mich. 1996), aff’d. per curiam, No. 96-2440 (6th Cir. July 8, 1997).
  2. See FTC v. OSF Healthcare System, 852 F. Supp.2d 1069, 1080 (N.D Ill. 2012); Hosp. Corp. of Am. v. FTC, 807 F.2d 1381, 1393 (7th Cir. 1986) (affirming the district court’s decision to grant a preliminary injunction).
  3. Federal Trade Commission and the Department of Justice, Improving Health Care: A Dose of Competition 14 (2004).
  4. Improving Health Care at p. 15.
  5. Improving Health Care at p. 4.
  6. See Butterworth, 946 F. Supp. at 1302 (noting that non-profit hospitals operate differently in highly-concentrated markets than do profit-maximising firms); Improving Health Care at p. 4 (stating that an extensive regulatory framework developed over decades, at both the federal and state levels of government, affects where and how competition takes place in healthcare markets).
  7. See United States v. Carilion Health Service, 707 F. Supp. 840 (W.D. Va. 1989), aff’d, 892 F. 2d 1042 (4th Cir. 1989); see also Julie Brill, Address at 2014 Hal White Antitrust Conference (June 9, 2014).
  8. U.S. v. Rockford Mem’l Corp., 898 F.2d 1278 (7th Cir. 1990) (affirming the district court’s decision to grant a preliminary injunction)
  9. Freeman, 911 F. Supp at 1218-19.
  10. See United States v. Long Island Jewish Medical Center, 983 F. Supp. 121, 140-141 (E.D.N.Y. 1997).
  11. Id. at 141.
  12. Id. at 142.
  13. See United States v. Carilion Health Service, 707 F. Supp. 840, 847 (W.D. Va. 1989).
  14. See Butterworth, 946 F. Supp at 1297.
  15. Id.
  16. Long Island, 983 F. Supp at 146.
  17. Id.
  18. Compare Carilion Health System, 707 F. Supp at 847 with NCAA v. Bd. of Regents, 468 U.S. 85, 101 n. 23, 104 S. Ct 2948, 2960, 82 L.Ed.2d 70 (1984) (“good motives will not validate an otherwise anti-competitive practice”).
  19. Id. at 142.
  20. Id. at 126.
  21. Long Island, 983 F. Supp. at 126 (citing Mercy Health Servs., 902 F. Supp. at 988).
  22. FTC v. Tenet Healthcare Corp., 186 F.3d 1045, 1054-10 (8th Cir. 1999).
  23. In the Matter of Reading Health System and Surgical Institute of Reading, FTC Docket No. 9353 (Dec. 7, 2012); Richard Feinstein, Statement on Today’s Announcement by Capella Healthcare that it will abandon its Plan to Acquire Mercy Hot Spring (June 27, 2013).
  24. OSF Healthcare System, 852 F. Supp.2d at 1080.
  25. Id.
  26. Improving Health Care at p. 2.
  27. Id. at p. 3.
  28. Id. at p. 8.
  29. FTC v. Promedica Health System, No. 3:11-CV-47, 2011 WL 1219281, at * 22 (N.D. Ohio Mar. 29, 2011).
  30. OSF Healthcare System, 852 F. Supp. at 1081.
  31. Id. at 1081.
  32. Improving Health Care at 6.
  33. Improving Health Care at 26.
  34. In the Matter of Evanston Northwestern Healthcare Corp, 2007 WL 2286195 at *50 (F.T.C. Aug. 6, 2007).
  35. Id. 2007 WL 2286195 at *51.
  36. FTC v. Penn State Hershey Medical Center, et al., 838 F.3d 332, 341 (3d. Cir. 2016).
  37. Id. at 327.
  38. Id. at 341.
  39. Penn State Hershey, 185 F. Supp. 3d 552, 564 (M.D. Pa. 2016).
  40. Penn State Hershey, 838 F.3d at 344.

Get unlimited access to all Global Competition Review content