Case 1:18-cv-02340-RJL Document 120 Filed 06/21/19 Page 1 of 22 UNITED STATES DISTRICT COURT FOR THE DISTRICT OF COLUMBIA UNITED STATES OF AMERICA, et al. Plaintiffs, v. CVS HEALTH CORPORATION and AETNA INC. Defendants. ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) Case No. 1:18-cv-02340-RJL AMICUS CURIAE AMERICAN MEDICAL ASSOCIATION’S POST-HEARING BRIEF IN OPPOSITION TO THE UNITED STATES’ MOTION FOR ENTRY OF THE PROPOSED FINAL JUDGMENT Case 1:18-cv-02340-RJL Document 120 Filed 06/21/19 Page 2 of 22 The evidentiary hearing in this case showed why the merger of CVS and Aetna, and the divestiture of Aetna’s prescription drug plan (PDP) business to WellCare, will increase concentration in an already concentrated PDP market. Basic economic theory and empirical analysis both associate increased concentration in this market with higher premiums, which will harm the vulnerable elderly and the American taxpayer. The hearing also showed that the loss of Aetna will increase consolidation in the market for pharmacy benefit managers (PBMs), of which the top three participants already control more than 70%. Because the merger of CVS and Aetna will harm competition, even with the divestiture of Aetna’s PDP business, the proposed final judgment should be denied. ARGUMENT I. The Merger and Divestiture Will Harm Competition and Raise Premiums in the PDP Market. A. Unrebutted Evidence Showed That the Transaction Will Increase Market Concentration and Likely Will Raise Premiums. The PDP market, which serves a vulnerable population, has been consolidating for years.1 In every region of the country, the PDP market is considered “moderately concentrated” under the Department of Justice’s Horizontal Merger Guidelines, except for two regions that are “highly concentrated.”2 CVS’s merger with Aetna will eliminate Aetna as a competitor in the PDP market.3 This loss is especially significant because CVS and Aetna have been close competitors.4 Even if WellCare can retain every one of the Aetna subscribers it inherits, this transaction will increase concentration in several regions to an extent that “potentially raise[s] 1 6/4/19 Tr. at 40:6–14; [Dkt. # 62 at 4]. 6/4/19 Tr. at 43:20–24; [Dkt. # 62-1 at 8]. 3 6/4/19 Tr. at 49:7–11; [Dkt. # 62 at 5]. 4 6/4/19 Tr. at 18:5–9; [Dkt. # 62 at 9]. 2 Case 1:18-cv-02340-RJL Document 120 Filed 06/21/19 Page 3 of 22 significant competitive concerns” under the Guidelines.5 In fact, in some regions it is mathematically impossible for WellCare to mitigate such an increase in concentration, no matter how well it competes.6 New entrants are also unlikely to mitigate this increase in concentration, as the PDP market has high barriers to entry.7 Economic theory predicts that higher concentration in the PDP market will lead to higher premiums, and empirical studies of the market unanimously conclude the same, associating mergers with higher premiums.8 There are no merger-specific efficiencies in the PDP market that would outweigh this loss of competition.9 All the evidence described in the paragraph above has something in common: Professor Sood presented it at the evidentiary hearing, and it was unrebutted by CVS’s own witnesses. Some evidence, such as the calculations of market concentration, was confirmed to be correct by CVS’s expert and the DOJ. 6/5/19 Tr. at 273:16–18; [Dkt. # 56 at 22 (DOJ conceding that after the divestiture, “some regions fall into the category of ‘potentially’ raising concerns under the Horizontal Merger Guidelines”)]. Other evidence, like the close competition between CVS and Aetna, the difficulty of entry into the PDP market, and the lack of merger-specific efficiencies, is based on the DOJ’s own filings in this case.10 No matter what the evidence may show about 5 6/4/19 Tr. at 58:14–19; [Dkt. # 62 at 6]. 6/4/19 Tr. at 62:23–64:4. 7 6/4/19 Tr. at 66:13–67:24; [Dkt. # 62 at 6–7]. 8 6/4/19 Tr. at 45:8–48:5. 9 6/4/19 Tr. at 69:18–70:16; [Dkt. # 62 at 23–24]. 10 [Dkt. # 1 ¶ 31 (“The proposed acquisition would substantially lessen competition and harm consumers by eliminating significant head-to-head competition between CVS and Aetna. Indeed, throughout the country, CVS and Aetna have been close competitors.”); Dkt. # 3 at 6 (“Because of these barriers to entry, entry or expansion into the sale of individual PDPs is unlikely to be timely or sufficient to remedy the anticompetitive effects from this merger.”); Dkt. # 1 ¶ 38 (“The proposed merger is also unlikely to generate verifiable, merger-specific efficiencies sufficient to outweigh the anticompetitive effects that are likely to occur in the sale of individual PDPs in the relevant Part D regions.”)] 6 2 Case 1:18-cv-02340-RJL Document 120 Filed 06/21/19 Page 4 of 22 other potential effects of this merger, the likely effect on the PDP market is clear: competition will be harmed, and premiums will increase. The harm to competition in the PDP market that will result from the merger and divestiture alone is enough to deny entry of the proposed final judgment. See United States v. Anthem, Inc., 236 F. Supp. 3d 171, 259 (D.D.C. 2017) (enjoining the merger of Anthem and Cigna based on its likely effect in the market for the sale of health insurance to large group employers in the Richmond, Virginia area), aff’d, 855 F.3d 345 (D.C. Cir. 2017). The DOJ itself has acknowledged that “[t]he standard for an acceptable divestiture is that it restore competition to premerger levels.”11 The unrebutted evidence has shown that the divestiture here will not restore competition to premerger levels, at least in some regions.12 And as the Court put it at the hearing, “if [prices are] likely to go up, [the transaction is] not going to be in the public interest.” 6/4/19 Tr. at 41:12–13. Both economic theory and empirical analysis predict that this transaction is likely to raise premiums for PDPs. On these grounds alone, the Court should find that the merger and divestiture are not in the public interest, and deny the DOJ’s motion for entry of the proposed final judgment. B. CVS’s and the DOJ’s Cursory Rebuttals to Certain Parts of Professor Sood’s Testimony Do Not Even Begin to Explain How the Divestiture of Aetna’s PDP Business Will Restore Competition. Professor Sood’s unrebutted testimony alone is enough to establish that the CVS–Aetna merger is not in the public interest. He also identified several other aspects of the PDP market 11 United States Department of Justice, “United States v. CVS and Aetna, Questions and Answers for the General Public,” https://www.justice.gov/opa/press-release/file/1099806/download. 12 While the effect of the Anthem–Cigna merger in just one metropolitan area was enough reason to enjoin the merger, the effect of the CVS–Aetna merger is far broader: even if WellCare can retain all of its former Aetna enrollees, the transaction will increase market concentration beyond the Guidelines’ threshold of concern in seven regions encompassing nine states. [Dkt. # 62 at 6]; 6/4/19 Tr. at 58:14–59:17. 3 Case 1:18-cv-02340-RJL Document 120 Filed 06/21/19 Page 5 of 22 and the transaction that confirm its likely anticompetitive effect. And while this testimony was rebutted, the rebuttals were particularly weak. First, WellCare will lose access to the Aetna brand at the end of this year, and CVS will get it back in 2021. Professor Sood explained that WellCare is unlikely to retain all of its Aetna enrollees because it does not have the same brand recognition or size and scale that Aetna does. 6/4/19 Tr. at 64:25–66:10. Even CVS’s own expert, relying on CVS’s own data, concluded that retention could be expected to be less than 100%. 6/5/19 Tr. at 254:1–9. CVS’s witness, Terri Swanson, downplayed the importance of brand in the PDP market, but did not try to claim that WellCare’s brand is as strong as Aetna’s. Instead, she pointed out that in one year, certain WellCare plans gained more new enrollment than more expensive Aetna plans. 6/5/19 Tr. at 346:17–347:24. But she also testified that Aetna’s and WellCare’s premiums have been generally comparable over time. Id. at 348:4–11. The results of the competition between Aetna plans and comparably priced WellCare plans speak for themselves: Source: April snapshots of PDP enrollment data published by CMS from 2009 to 2018, 4 Case 1:18-cv-02340-RJL Document 120 Filed 06/21/19 Page 6 of 22 https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-andReports/MCRAdvPartDEnrolData/Monthly-Enrollment-by-Contract-Plan-State-County.html Note: PDP Market Share Index = (National PDP Market Share in Year X / National PDP Market Share in Year 2009) * 100. Faced with a choice between WellCare-branded plans or comparably priced Aetna-branded plans, consumers have overwhelmingly chosen the Aetna brand.13 With respect to WellCare’s relative size and scale, it is indisputable that Aetna is much larger than WellCare, both in its PDP business and overall. Nevertheless, Ms. Swanson claimed that WellCare has been able to compete successfully despite having a fraction of the number of Aetna’s enrollees, and having no commercial business (i.e., business other than government programs). 6/5/19 Tr. at 346:1–16. Again, WellCare’s inability to grow its enrollment significantly over time, compared to Aetna’s robust growth, belies this claim. Second, Professor Sood noted that on a per-enrollee basis, Aetna’s PDP business was sold to WellCare for about $107 million, or about $50 per enrollee. 6/4/19 Tr. at 56:18–57:3. By comparison, the rest of Aetna’s business was sold to CVS for $69 billion, or more than $3,100 per enrollee. The relatively minuscule price for Aetna’s PDP business suggests that WellCare doubted its ability to retain all its former Aetna enrollees. CVS’s expert, Dr. Lawrence Wu, responded (based on internal CVS data that has never been presented to the Court or the amici) that $50 per enrollee is “within the range of prior sale prices,” and that CVS acquired the PDP enrollees of the insurer CareFirst for $40 each. 6/5/19 Tr. at 256:3–22. But Dr. Wu never stated what this range was, or where in the range the price of Aetna’s PDP business fell. The AMA has been unable to verify Dr. Wu’s figure, but even if it is correct, it appears to be an outlier. Two CVS acquisitions of PDP contracts with publicly available prices are those of Torchmark and 13 In their paper cited below at note 16, Professor Claudio Lucarelli and others created a model of pricing in the PDP market. Although it was not discussed in the published paper, the model estimated that, holding other factors equal, CVS’s SilverScript brand commanded a premium $3.19 higher than WellCare’s brand, and Aetna’s brand was worth $2.76 more than WellCare’s. 5 Case 1:18-cv-02340-RJL Document 120 Filed 06/21/19 Page 7 of 22 Universal American. CVS appears to have paid $179 per enrollee to Torchmark, whose business was declining and had been sanctioned by CMS for various violations, and $658 per enrollee to Universal American, or many times the $50 per enrollee that WellCare paid here.14 Dr. Wu described the Torchmark acquisition as “the best way I can describe my conclusion” about likely retention rates; he said that CVS modeled that it would retain 95% of Torchmark’s enrollees. 6/5/19 Tr. at 253:16–254:7. If that is true, and CVS paid $179 per enrollee, then it stands to reason that at $50, WellCare expected a significant loss of its Aetna enrollees. Third, while Dr. Wu identified what he saw as shortcomings in the academic literature, which unanimously concludes that reducing the number of competitors in the PDP market leads to increased premiums, he did not raise any serious questions about the literature’s conclusions. With respect to the paper by Princeton researcher Dr. Anna Chorniy and others,15 Dr. Wu stated that “the paper does estimate the cost efficiencies, which more than offset what they label as the price increase.” 6/5/19 Tr. at 276:5–7. This is wrong. For mergers between insurers who competed against each other (as CVS, Aetna, and WellCare have done in every market nationwide), the paper found that premiums increase by 7.3% on average, and “rebate increases [i.e., lower drug prices from the manufacturers] dwarf in comparison to merging insurers’ gain in market power and no sizable cost efficiencies are realized and/or passed to the enrollees through 14 CVS paid $21.8 million in cash for the 122,000 enrollees of Torchmark subsidiary United American. SilverScript Quarterly Statement at 10.1, https://insurance.arkansas.gov/uploads/pages/silverscript_2017_1st_qtr.pdf; Kaiser Family Foundation, “Medicare Part D: A First Look at Prescription Drug Plans in 2017” at 3, http://files.kff.org/attachment/Issue-Brief-Medicare-Part-D-A-First-Look-at-Prescription-DrugPlans-in-2017. CVS paid $1.25 billion for the 1.9 million enrollees of Universal American. “CVS Caremark to Purchase Universal American's Medicare Part D Business,” https://cvshealth.com/newsroom/press-releases/cvs-caremark-purchase-universal-americansmedicare-part-d-business. 15 Chorniy et al., Mergers in Medicare Part D: Assessing Market Power, Cost Efficiencies, and Bargaining Power (April 2018), available at http://www.princeton.edu/~achorniy/research/papers/MergersPartDChorniyMillerTang.pdf. 6 Case 1:18-cv-02340-RJL Document 120 Filed 06/21/19 Page 8 of 22 premiums.” Chorniy at 6. In other words, the merged insurers’ size may have allowed them to drive a better bargain with drug manufacturers, but they kept the savings for themselves and increased their premiums. With respect to another paper by Wharton Professor Claudio Lucarelli and others,16 Dr. Wu took issue with the paper’s use of data from 2006, the first year that PDPs were available, because “there may be serious miscalculations by consumers and sellers.” 6/5/19 Tr. at 275:1–10. But Dr. Wu did not propose, much less explain, that these miscalculations would be biased in such a way that the paper’s conclusions would be incorrect. Moreover, the paper was revised in 2011 (without changing the relevant conclusions) and was accepted into a peer-reviewed journal, the International Economic Review. And as to a working paper by the non-partisan Congressional Budget Office (CBO),17 Dr. Wu cited an appendix in which an alternative specification of the model would reduce the statistical significance of the relationship between the number of PDP sponsors and PDP premiums. 6/5/19 Tr. at 275:16–22. But the authors ultimately concluded that the most relevant specification of the model did show a statistically significant effect, as they summarized: “Consistent with economic theory, we find that an increasing number of plan sponsors active in the Part D market is associated with lower bids submitted by plan sponsors.” CBO at 19–20.18 In summary, three different papers, applying 16 Lucarelli et al., The welfare impact of reducing choice in Medicare Part D: A comparison of two regulation strategies. International Economic Review 53, no. 4 (2012): 1155-1177. Manuscript available at https://www.nber.org/papers/w14296.pdf. 17 Stocking, et al., Examining the Number of Competitors and the Cost of Medicare Part D. CBO Working Paper Series. 2014-04, http://cbo.gov/sites/default/files/cbofiles/attachments/45553PartD.pdf. 18 Dr. Wu also complained that all three papers were dated because they used data no more recent than 2012. 6/5/19 Tr. at 275:7–10, 275:14–15, 276:11–12. There has been no change since 2012 to the competitive structure of the PDP market, in which plan sponsors submit bids and the government subsidizes about 75% of the premiums. Dr. Wu did not identify any such changes, or any analyses concluding that reducing the number of competitors results in anything other than higher premiums. 7 Case 1:18-cv-02340-RJL Document 120 Filed 06/21/19 Page 9 of 22 three different methods, all concluded that reducing competition in the PDP market leads to increased premiums, consistent with economic theory. Fourth, in its motion to present rebuttal testimony, the DOJ claims that “Dr. Sood’s analysis contains basic errors in its HHI calculations and wrongly assumes that CVS would win any business that WellCare lost.” [Dkt. # 116 at 7.] The DOJ did not disclose what these basic errors are, but it is has already agreed with Professor Sood’s main conclusion stating in its response to public comments that “"some regions fall into the category of ‘potentially’ raising concerns under the Horizontal Merger Guidelines,” even if WellCare keeps all of its Aetna enrollees. [Dkt. # 56 at 22.] Arguments about whether WellCare’s lost enrollees will go to CVS or other insurers is essentially a question of how many more regions will suffer from reduced competition, a question that does not change the fundamental fact that the merger and divestiture will significantly increase concentration in several regions.19 Finally, the DOJ has criticized Professor Sood for pointing out that the divestiture would give WellCare more than 35% of the market for low-income-subsidy beneficiaries in Arkansas and Hawaii—a fact the DOJ itself cited as “likely result[ing] in further loss of competition.” [Dkt. # 116 at 8; Dkt. # 3 at 5.] The problem, according to the DOJ, is that Professor Sood’s calculation used only the PDP market, while the DOJ’s threshold referred to combined enrollment in PDPs and Medicare Advantage plans. [Dkt. # 116 at 8.] It is unclear why the DOJ is insisting on combining two markets that it has argued should be kept separate for economic analysis. [Dkt. # 1 ¶ 25 (“Medicare Advantage is not a close substitute for beneficiaries enrolled in individual PDPs”)]. But if that is what the DOJ wants, the fact is that in the combined 19 Professor Sood also showed that in two regions, Arkansas and Mississippi, it is impossible for WellCare to bring the increase in concentration below the DOJ’s level of concern no matter how much market share it can take from any competitor, not just CVS. 6/4/19 Tr. at 63:23–64:4. 8 Case 1:18-cv-02340-RJL Document 120 Filed 06/21/19 Page 10 of 22 PDP/Medicare Advantage market, WellCare will have more than 35% in the Mississippi region, meaning the transaction is still anticompetitive.20 C. CVS Has the Incentive and Ability to Disadvantage WellCare Through Input Foreclosure. WellCare relies on CVS to provide two important inputs to its PDP business: PBM services and pharmacies. Without competitive pricing for those inputs, WellCare’s PDP offerings will be inferior to those of CVS, which controls (by some measures) the largest PBM and the largest pharmacy chain in the United States. As WellCare’s competitor, CVS has twin incentives not to offer WellCare as good a deal as it gives itself: higher prices for inputs to WellCare result in higher profits to CVS, and if WellCare competes vigorously, it could erode CVS’s PDP business. If CVS acts in line with its incentives, WellCare will not be able to retain all of its former Aetna subscribers, and market concentration will increase even more.21 In CVS’s testimony on input foreclosure, Dr. Wu began by building a strawman: that input foreclosure meant that CVS would not offer its services to WellCare at all: a “blackout.” 6/5/19 Tr. at 259:17–260:2. As Dr. Sood explained, the reality is less black and white: CVS merely has to increase its prices, or offer less service, at a level that WellCare will not be able to detect, or that WellCare will not be able to improve on by seeking out other suppliers. 6/4/19 Tr. at 26:3–30:13. CVS’s witnesses claimed that it is easy for an insurer to detect that it is getting noncompetitive pricing, citing the practice of “market checks,” in which the insurer can see what the market is offering and demand a lower price from its PBM. 6/5/19 Tr. at 326:8–22, 358:25– 20 Richard Scheffler, Ph.D., “The Impact of Aetna’s Proposed Medicare Part D Stand-Alone Prescription Drug Plan Divestiture to WellCare” at 10, http://petris.org/impact-aetnas/. 21 That said, it is not essential to find vertical foreclosure in order to conclude that the merger and divestiture are not in the public interest. Even if WellCare maintains or grows its business, the transaction will significantly increase concentration in the PDP market in multiple regions. See supra p. 8. 9 Case 1:18-cv-02340-RJL Document 120 Filed 06/21/19 Page 11 of 22 359:5. This is not nearly as easy as it sounds. For example, Anthem, one of the largest insurers in the country and a sophisticated consumer of PBM services, is embroiled in a years-long litigation with its PBM, Express Scripts, about a market check. In 2014, Anthem undertook a market check, determined that it was paying well above the market rate, and asked Express Scripts for $13 billion in pricing concessions. Anthem, Inc. v. Express Scripts, Inc., 2017 WL 1134765, at *2 (S.D.N.Y. Mar. 23, 2017). Express Scripts refused to give any concessions, and Anthem sued. Id. The case is now well into its fourth year. Similarly, insurers who contract with CVS can have a difficult time finding out exactly what they are paying for. In 2018, the Auditor of the State of Ohio released a report about CVS and another PBM, which administered the prescription drug benefit for Ohio’s managed Medicaid program.22 The report determined that in just one year, CVS and the other PBM took about $225 million in “spread,” which is the difference between the amount the PBM pays for a drug and what it charges its client. Report at 2. The Report pointed out that the Medicaid plans could not have discovered this information: “[W]hile we [the Auditor’s office] were able to obtain information on payments made by CVS Caremark to pharmacies, the Plans that paid CVS Caremark were not permitted by this PBM to see this information.” Id. at 9. Even CVS and Aetna have disagreed about reimbursement for PDPs. Aetna’s former chief Medicare actuary has filed a False Claims Act suit alleging that CVS Caremark has fraudulently caused the government to make more than a billion dollars in excess payments since 2010. United States ex rel. Behnke v. CVS Caremark Corp., No. 14-cv-824 (E.D. Pa.). While the AMA has no position on whether the allegations are true, the fact that CVS and Aetna’s chief Medicare actuary could disagree completely on something so important speaks volumes about the opacity of pricing in the PBM industry. 22 Ohio’s Medicaid Managed Care Pharmacy Services, available at https://ohioauditor.gov/ auditsearch/Reports/2018/Medicaid_Pharmacy_Services_2018_Franklin.pdf (“Report”). 10 Case 1:18-cv-02340-RJL Document 120 Filed 06/21/19 Page 12 of 22 Even if a PDP sponsor like WellCare could cut through the opacity of PBM pricing, its alternatives to CVS are limited because just three PBMs control roughly 70% of the market. 6/5/19 Tr. at 216:10–20; [Dkt. # 56-1 at 219]. The next largest PBMs are much smaller and cannot offer the same scale when negotiating with drug manufacturers. Moreover, the “Big Three” are all integrated with PDP sponsors and therefore are WellCare’s competitors, with the same incentives not to give WellCare pricing comparable to what they give their own plans.23 Even if the risk were high that CVS could lose WellCare as a PBM customer by charging more or offering fewer services, the risk might be worth taking for a large PDP sponsor like CVS, which could gain enrollees that WellCare loses. Dr. Sood’s research suggests that insurance, pharmacy, and PBM services have profit margins of 3.0%, 4.0%, and 2.3%, respectively.24 Therefore, an insurance, pharmacy, and PBM customer is worth four times as much as a customer who only uses PBM services. If CVS could gain one out of every four WellCare customers, it may be able to make up its losses.25 Finally, CVS’s witnesses did little to address Dr. Sood’s claims that CVS can use its pharmacies to foreclose WellCare. In 2015, CVS stated in its Form 10-K, “We currently operate in 98 of the top 100 United States drugstore markets and hold the number one or number two market share in 93 of these markets.” [Dkt. # 56-1 at 117.] As a result, a pharmacy network without CVS will be seen as less attractive than one with it, and CVS can charge a premium for access to its pharmacy network. 6/4/19 Tr. at 39:2–21. To supply a competitive product, 23 The other two large PBMs are Express Scripts (owned by Cigna) and OptumRx (owned by UnitedHealthcare). 24 Sood et al., The Flow of Money Through the Pharmaceutical Distribution System, at 5, available at https://healthpolicy.usc.edu/wp-content/uploads/2017/06/USC_Flow-ofMoneyWhitePaper_Final_Spreads.pdf. 25 Dr. Wu implied that this analysis is implausible because CVS’s revenue from its PBM contract with WellCare is more than double the revenue of CVS’s own PDP business. 6/5/19 Tr. at 259:1–14. But he did not say anything about the relative profitability of these two businesses. 11 Case 1:18-cv-02340-RJL Document 120 Filed 06/21/19 Page 13 of 22 WellCare would have to pay that premium whether it used CVS’s PBM, another PBM, or tried to develop a pharmacy network on its own. This puts WellCare at a competitive disadvantage to CVS, which does not have to pay this extra expense. Between its pharmacy network and its PBM, CVS has the ability and incentive to put WellCare at a competitive disadvantage. D. The Same Problem of Input Foreclosure Extends More Broadly to All Health Insurance, not Just PDPs. As a provider of PBM and pharmacy services to health insurers other than PDPs, CVS will gain a new incentive to engage in input foreclosure by acquiring Aetna. Professor Sood has estimated that an insurance customer is fourteen times as profitable as a PBM customer. [Dkt. # 62 at 18.] If CVS thinks it can gain even a small fraction of health insurance customers for its Aetna business, it would be in CVS’s interest to raise its prices for PBM and pharmacy services, degrade its service, or both. Health insurance customers would be harmed as a result. The AMA has discussed this issue in its amicus brief, as well as in its public comments to the DOJ. [Dkt. # 62 at 17–19; Dkt. # 56-1 at 113–19.] II. The Merger of CVS and Aetna Is Plainly Anticompetitive in the PBM Market. CVS has denied that its merger with Aetna is a horizontal one in the PBM market. 6/5/19 Tr. at 243:18–248:25, 348:12–19. The evidence is overwhelmingly to the contrary. A. Aetna Has a Significant Market Share as a Supplier of PBM Services. While Aetna purchases certain PBM services from CVS under a 2010 agreement that expires in 2022, the companies’ joint press release announcing the supply agreement says that “Aetna will maintain and manage its core pharmacy benefits business.”26 The press release 26 https://news.aetna.com/news-releases/aetna-awards-long-term-contract-to-cvs-caremark-toprovide-pbm-services/. 12 Case 1:18-cv-02340-RJL Document 120 Filed 06/21/19 Page 14 of 22 quotes Aetna’s CEO as claiming, “we retain our PBM and our ability to integrate medical care with clinical and pharmacy programs and actionable data.” In an attempt to walk back this admission, Dr. Wu testified that he had seen amended versions of the CVS/Aetna PBM contract and suggested that the press release was no longer accurate. 6/5/19 Tr. at 228:9–229:7. The actual relationship between CVS and Aetna before the merger, Dr. Wu testified, was that “what Aetna does is mostly functions related to customerfacing activities.” Id. at 239:6–7. Dr. Wu’s testimony contradicts Aetna’s final annual report, which states, “We offer pharmacy benefit management services …. We also perform various pharmacy benefit management services for Aetna pharmacy customers consisting of: product development, commercial formulary management, pharmacy rebate contracting and administration, sales and account management and precertification programs.”27 Dr. Adam Fein, a leading authority on the PBM industry, recently concluded that “Aetna controls medical and pharmacy policy, formulary design, pharmacy/medical benefit integration, rebate contracting and many other core PBM functions.”28 Likewise, in the California Department of Insurance hearing on this merger, the Department found that “Aetna currently undertakes numerous important PBM services in-house, including rebate contracting with drug manufacturers for most of its commercial clients, as well as formulary development.” [Dkt. # 62-2 at 4.] Aetna has been very successful as a supplier of PBM services. Thus, two years into its supply agreement with CVS, then-FTC Commissioner Julie Brill found that Aetna was the “Big Three’s nearest competitor.”29 Before the merger of CVS and Aetna, PBM expert Dr. Fein 27 Aetna Form 10-K (2017); see also AMA Letter to Makan Delrahim, [Dkt. # 56-1 at 191–92]. See Drug Channels, Dec. 5, 2017, available at https://www.drugchannels.net/2017/12/the-cvsaetna-deal-five-industry-and.html. 29 Dissenting statement of Commissioner Julie Brill concerning the proposed acquisition of Medco Health Solutions by Express Scripts (April 2, 2012) at 3, 28 13 Case 1:18-cv-02340-RJL Document 120 Filed 06/21/19 Page 15 of 22 reported Aetna’s share of the PBM market to be four percent, a figure which may not sound large, but would rank Aetna among the largest PBMs outside of the “Big Three.”30 B. The Merger Is Anticompetitive in a PBM Market That Is Noncompetitive and Poorly Performing. The merger is occurring in a PBM market that Professor Sood has found to be “significantly noncompetitive” under criteria found in the Areeda & Hovenkamp Antitrust Law treatise.31 Moreover, the “Big Three” collectively account for at least 70% of the market. Professor Sood has calculated the PBM market HHI to be roughly 1900 (“Moderately Concentrated” under the Guidelines). The merger would increase HHI by roughly 200 points, twice the increase that potentially raises significant competitive concerns in moderately concentrated markets under the Guidelines. [Dkt. #56-1 at 245.] Moreover, the PBM market is poorly performing. A February 2018 report from the President’s Council of Economic Advisers (CEA Report) states that the existing market structure allows PBMs “to exercise undue market power.”32 Similar concerns were recently expressed by U.S. Food and Drug Administration Commissioner Scott Gottlieb, M.D.: The top three PBMs control more than two-thirds of the market: the top three wholesalers more than 80%; and the top five pharmacies more than 50%. Market concentration may prevent optimal competition. And so the saving may not always be passed along to employers or consumers. https://www.ftc.gov/sites/default/files/documents/public_statements/dissenting-statementcommissioner-julie-brill/120402medcobrillstatement.pdf. 30 “PBM Market Share, by Totally Equivalent Prescription Claims Managed, 2017,” https://www.drugchannels.net/2018/05/cigna-express-scripts-vertical.html. 31 [Dkt. # 56-1 at 242 (applying criteria found in Herbert Hovenkamp, Antitrust Law, ¶ 1121 (2016)).] The criteria are that the same four firms have accounted for at least 75% of the market for at least five years preceding the merger and that the HHI of the market is at least 1800. 32 Council of Economic Advisers, Reforming Bio Pharmaceutical Pricing at Home and Abroad (February, 2018) at 10, https://www.whitehouse.gov/wp-content/uploads/2017/11/CEA-RxWhite-Paper-Final2.pdf. See also “The Flow of Money through the Pharmaceutical Distribution System,” supra note 24. 14 Case 1:18-cv-02340-RJL Document 120 Filed 06/21/19 Page 16 of 22 Too often, we see situations where consolidated firms—the PBMs, the distributors, and the drugstores—team up with payors. They use their individual market power to effectively split some of the monopoly rents with large manufacturers and other intermediaries rather than passing on the savings garnered from competition to patients and employers.33 As recommended by the CEA Report, “policies to decrease concentration in the PBM market … can increase competition and further reduce the price of drugs paid by consumers.” CEA Report at 10. Allowing a CVS/Aetna merger would be at war with those policies. C. The PBM Market Has High Barriers to Entry. The lost competition from this merger is likely to be permanent because barriers to entry prevent new entrants from restoring competitive pricing. One barrier is the scale required to negotiate favorable discounts from pharmaceutical manufacturers. PBM entrants need to attract customers with competitive discounts from pharmaceutical firms. However, the magnitude of discounts that a PBM can negotiate with these firms depends on the number of covered lives represented by the PBM, with the size of the discount rising with the size of the PBM. Hence, the three largest incumbent PBMs comprising at least 70% of the market have a durable price advantage. [Dkt. #62-8 at 2.] Given that there are numerous health insurance markets where Aetna is the first or second largest health insurer, [Dkt. # 56-1 at 144–48], the CVS acquisition of Aetna significantly reduces the size of the PBM market available to new entrants, and thus reduces the incentives for entry into the market. [Dkt. # 56-1 at 122.] Several of the largest PBM’s in the United States such as Express Scripts, OptumRx and Humana Pharmacy Solutions are also owned by health plans. So new standalone PBM entry is unlikely given that several health plans already have their own PBM’s. Id. 33 Scott Gottlieb, M.D., “Capturing the Benefits of Competition for Patients” (March 7, 2018), https://www.fda.gov/news-events/speeches-fda-officials/capturing-benefits-competition-patients03072018. 15 Case 1:18-cv-02340-RJL Document 120 Filed 06/21/19 Page 17 of 22 The significant barriers to entry in the PBM market were acknowledged by David Snow, the CEO of one of the largest PBMs, when Walmart, which ought to be a formidable competitor, tried to enter the PBM market in 2008: “Many people shake in their boots when they hear the name Wal-Mart in any industry,” Medco CEO David Snow told the Newark Star-Ledger. “This is a very complicated business with serious barriers to entry. I just don’t think they’re going to pull it off. You just don’t snap your fingers and say you’re going to be a pharmacy-benefits manager.”34 In addition, the PBM needs to form a national pharmacy network with the ability to contract and process claims from pharmacies within the network. According to Professor Sood, forming such a network “is no small feat for a new entrant.” [Dkt. # 56-1 at 165.] At the hearing in this case, Dr. Wu held up OptumRx as a counterexample, noting that it grew from a small PBM to one of the “Big Three.” 6/5/19 Tr. at 225:5–20. But this is no surprise; OptumRx is integrated with UnitedHealthcare, the largest commercial health insurer in the United States. OptumRx was able to take advantage of these members to grow its business, just as CVS will be able to leverage its access to Aetna members to steer them to CVS’s own PBM, further consolidating the industry. D. Potential Disruptive Competition Will Be Lost. The merger would not just result in the significant lost PBM market competition discussed above. Professor Sood has shown that it would also eliminate Aetna as an important potential competitor in the PBM market and under criteria found in the Areeda & Hovenkamp antitrust law treatise, should be deemed detrimental to competition on that basis also.35 First, the PBM market is noncompetitive, with the same four firms controlling at least 75% of the market for at least five years preceding the merger. Second, CVS is a dominant firm in the PBM market 34 Managed Care, “Could a Wal-Mart PBM Succeed?” https://www.managedcaremag.com/archives/2008/4/could-wal-mart-pbm-succeed. 35 See V Philip E Areeda & Herbert Hovenkamp, Antitrust Law, ¶ 1121 (2016); see also [Dkt. # 56-1 at 213–29 (AMA correspondence with Peter J. Mucchetti), id. at 240–48 (Sood presentation to DOJ)]. 16 Case 1:18-cv-02340-RJL Document 120 Filed 06/21/19 Page 18 of 22 and the merger will increase the market share of CVS and eliminate Aetna, increasing HHI by 200 points. Third, Aetna has the size, skill and economic incentives to compete. Finally, there are few other potential entrants due to significant barriers to entry. In the event of a merger, the PBM market would lose Aetna, a national company with an established brand, significant customer base (Aetna health insurance), expertise, capital, and years of experience as a major player in the PBM market. There would be no PBMs that could defeat the coordinated conduct of the three largest PBMs. E. PBMs Will Have Enhanced Incentives and Ability to Engage in Coordinated Input Foreclosure. Merger law does not require evidence of actual effects. In Hospital Corp. of America v. FTC, Judge Posner concluded that “all that is necessary is that the merger create an appreciable danger of collusive practices in the future. A predictive judgment, necessarily probabilistic and judgmental rather than demonstrable is called for.” 807 F.2d. 1381, 1389 (7th Cir. 1986) (emphasis added). As a result of the merger of CVS and Aetna, all suppliers of PBM services with the scale to drive deep discounts with pharmaceutical companies would be vertically integrated into health insurers. Facing the same incentives of not bidding aggressively for contracts that would strengthen their health insurer rivals, there is an appreciable danger that the “Big Three” would engage in market-wide express or tacit collusion or oligopoly behavior of not competing aggressively for PBM customers competing with the “Big Three’s” health insurer arms.36 Such coordinated conduct of “raising rivals’ costs” would likely be profitable. [Dkt. #56-1 at 114–19, 248.] The CEA Report observes that drug pricing suffers from a lack of transparency characterized by price obfuscation. CEA Report at 10. PBM customers have scant 36 Thomas L. Greaney, The New Health Care Merger Wave: Does the “Vertical, Good” Maxim Apply?, 46 J.L. Med. & Ethics 918 (2018). 17 Case 1:18-cv-02340-RJL Document 120 Filed 06/21/19 Page 19 of 22 information about the rebates supposedly negotiated on their behalf because contracts between PBMs and drug manufacturers are claimed as trade secrets.37 Not even large payers like Blue Cross or Walmart know the net prices of the drugs they are buying.38 One expert has concluded that most of the increase in drug pricing can be attributed to rebates pocketed by PBMs.39 In short, the strategy of raising rivals’ costs is likely to work. F. CVS Can Engage in Customer Foreclosure in the PBM Market. There are worrisome reports, including a Wall Street Journal article, that CVS’s acquisition of Aetna is a defensive play to block market entry.40 Aetna has approximately 22.1 million members and says that it has “an estimated 39 million people relying on Aetna to help them make decisions about their healthcare.”41 Thus, the merger would significantly narrow the market for the services of competing PBMs. The merger would also result in a PBM market that is 90% vertically integrated into health insurance markets. [Dkt. # 56-1 at 247.] Lacking customers for PBM services, any new PBM market entrant would need to engage in two-tiered 37 Robin Feldman, “Perverse Incentives: Why Everyone Prefers High Drug Prices-Except For Those Who Pay the Bills,” forthcoming Harv. J. on Legis., available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3162432) (PBM payment structure “minimizes the competitive pressure to reduce prices” and creates incentives “operating to drive prices higher”). 38 Id. In a letter to the Court, CVS claimed that it passes on 98% of its rebates to its clients other than Medicare plans. [Dkt. # 118-1 at 4.] This is not as impressive as it sounds, as PBMs have reacted to criticisms of their rebate policies by recharacterizing their compensation from drug manufacturers as “administrative fees,” even when those fees are not for bona fide administrative services. National Prescription Coverage Coalition, “Express Scripts Lawsuit Should Raise Everyone’s Eyebrows,” https://nationalprescriptioncoveragecoalition.com/express-scriptslawsuit-should-raise-everyones-eyebrows/. 39 Robert Goldberg, Center for Medicine in the Public Interest, “Drug Costs Driven by Rebates,” http://bionj.org/wp-content/uploads/2015/11/drug-costs-driven-by-rebates.pdf. 40 “A Force behind the Aetna Bid: Amazon,” the Wall Street Journal (Oct. 27, 2017); see also “CVS's Megadeal to Change US Healthcare Faces Stiff Challenges,” Bloomberg News (Dec. 22, 2017); “Aetna-CVS Deal a Defensive Play as Amazon Threat Looms,” Bloomberg First Word (Dec 15, 2017). 41 “Aetna Facts,” https://www.aetna.com/about-us/aetna-facts-and-subsidiaries/aetna-facts.html. 18 Case 1:18-cv-02340-RJL Document 120 Filed 06/21/19 Page 20 of 22 entry by also entering the highly concentrated health insurance markets. Unfortunately, these markets too have high entry barriers.42 CVS’s witness, Dr. Wu, seemed to suggest that because CVS has been providing PBM services to Aetna under a long-term exclusive contract, the acquisition of Aetna would not result in any competitive change in the market, including its ability to engage in foreclosure. 6/5/19 Tr. at 218:11–222:5. This testimony obscures the important difference between exclusive dealing and vertical mergers, which requires antitrust condemnation of vertical mergers at lower foreclosure percentages than is appropriate in exclusive dealing cases. See Herbert Hovenkamp, Federal Antitrust Policy § 9.4 at 346 (1994) (“Federal Antitrust Policy”). The vertical merger is more permanent than exclusive dealing contracts. Id. Even long-term service contracts maintain competition in the marketplace. This is called “competition for the contract,” which, in the words of Judge Posner, is a “form of competition the antitrust laws protect.”43 A merger forever eliminates the considerable ongoing and consumer-benefitting competition for the exclusive contract. Id. For example, then-AMA President Barbara McAneny, M.D., testifying on behalf of the AMA and herself as a practicing oncologist before the California Department of Insurance, observed that when quality-of-care issues arise between her and a PBM concerning one of her cancer patients, she takes the problem to the insurer. At contract renewal time, Aetna is free to weigh her consumer quality demands against financial concerns and perhaps opt for a different PBM.44 However, once Aetna has a permanent ownership interest in CVS, Aetna will be less responsive to her consumer demands. As Professor Hovenkamp observes: “When the integration 42 See United States v. Aetna, 240 F. Supp. 3d 1 (D.D.C. 2017); United States v. Anthem, 855 F.3d 345 (D.C. Cir. 2017). 43 Methodist Health Servs. Corp. v. OSF Healthcare Sys., 859 F.3d 408, 411 (7th Cir. 2017); see also Harold Demsetz, “Why Regulate Utilities?” 11 J.L. & Econ. 55 (1968). 44 Video of the hearing is available at https://youtu.be/mvpTLGuyuPg. Dr. McAneny’s testimony begins at 4:22:03. 19 Case 1:18-cv-02340-RJL Document 120 Filed 06/21/19 Page 21 of 22 occurs by merger … the downstream business becomes part of the colluding firm itself.” Federal Antitrust Policy at 346. CONCLUSION The motion to enter the proposed final judgment should be denied for two independent reasons: it will likely reduce competition and increase premiums in the PDP market, and it will reduce competition in the PBM market as well, to the detriment of consumers of PBM services. Dated: June 21, 2019 Respectfully submitted, /s/ Henry C. Quillen Henry C. Quillen (Bar ID 986686) WHATLEY KALLAS LLP 159 Middle St., Suite 2C Portsmouth, NH 03801 Telephone: (603) 294-1591 Fax: (800) 922-4851 hquillen@whatleykallas.com Henry S. Allen, Jr. (admitted pro hac vice) Senior Attorney, Advocacy Resource Center AMERICAN MEDICAL ASSOCIATION 330 N. Wabash Chicago, IL 60611 Telephone: (312) 464-4271 Fax: (312) 224-6919 henry.allen@ama-assn.org Counsel for the American Medical Association 20 Case 1:18-cv-02340-RJL Document 120 Filed 06/21/19 Page 22 of 22 CERTIFICATE OF SERVICE I, Henry C. Quillen, hereby certify that on June 21, 2019, I caused a copy of the foregoing document to be served upon Plaintiffs State of California, State of Florida, State of Hawaii, State of Washington, and Defendants CVS Health Corporation and Aetna, Inc., via the Court’s CM/ECF system, and to be served upon Plaintiff State of Mississippi by mailing the document electronically to its duly authorized legal representative: Counsel for the State of Mississippi: Crystal Utley Secoy Consumer Protection Division Mississippi Attorney General’s Office P.O. Box 22974 Jackson, Mississippi 39225 Phone: (601) 359-4213 cutle@ago.state.ms.us /s/ Henry C .Quillen Henry C. Quillen (Bar ID 986686)