United States of America v. Cvs Health Corporation

UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF COLUMBIA
UNITED STATES OF AMERICA, et al., )
)
Plaintiffs, )
)
v. ) Civil Case No. 18-2340 (RJL)
)
CVS HEALTH CORPORATION, et al., )
)
Defendants. )
~
MEMORANDUM OPINION
(September~, 2019) [Dkt. # 57]
The United States of America ("the Government") filed this lawsuit to challenge
CVS Health Corporation's ("CVS's") acquisition of Aetna Inc. ("Aetna") as a violation
of Section 7 of the Clayton Act, 15 U.S.C. § 18. It now moves to resolve the case
through entry of a negotiated consent judgment. See U.S. Mot. & Memo. in Supp. of
Entering Prop. Final J. ("Mot. for Prop. Final J.") [Dkt. # 57]. When the Government
seeks to settle a civil antitrust suit through a consent judgment, a court must
independently "determine that ... entry of [the proposed] judgment is in the public
interest" before granting the Government's request. 15 U.S.C. § 16(e)(l) ("the Tunney
Act" 1). As such, this Court must determine whether the proposed consent judgment here
is in the public interest.
That determination in this particular case, however, is no small matter. Industry
players, consumer groups, and state regulatory bodies have all raised concerns about
1
The Antitrust Procedures and Penalties Act, 15 U .S.C. § 16(6)-(h), is also known as the Tunney'Act.
CVS's acquisition of Aetna. The merger combines two healthcare giants. Its effects, for
better or worse, will be felt by millions of consumers. As I explained to the parties near
the outset of this case, with so much at stake, the congressionally mandated public
interest inquiry must be thorough. Indeed, if the Tunney Act is to mean anything, it
surely must mean that no court should rubberstamp a consent decree approving the
merger of "one of the largest companies in the United States" and "the nation's third-
largest health-insurance company," Compl. i1i115-16 [Dkt. # 1], simply because the
Government requests it!
My determination of whether the Government's proposed final judgment is in the
public interest will, of course, be based on the existing record, which has been
meaningfully supplemented by the briefs and testimony presented by the parties and
amici curiae ("the amici"). 2 Indeed, the amici raised substantial issues that deserved
serious consideration. Unfortunately for the amici, · however, the r.ecord did not
persuasively undermine the parties' contention that the proposed final judgment is in the
public interest. Accordingly, for the following reasons, I have concluded that the
Government's Motion to Enter the Proposed Final Judgment must be GRANTED.
BACKGROUND
On October 10, 2018, the Government, along · with the States of California,
Florida, Hawaii, Mississippi, and Washington, sued to enjoin CVS' s sixty-nine-billion-
dollar acquisition of Aetna. See Compl. i1i11, 41. According to the Government's
2
Four amici-the AIDS Healthcare Foundation, the American Medical Association ("the AMA"),
Consumer Action, and U.S. PIRG-submitted briefs and participated in the hearings in this matter. Two
additional amici-Pharmacists United for Truth and Transparency and the Pharmacists Society of the
State of New York-filed briefs but declined to participate in the hearings.
2
complaint, "CVS ... is one of the largest companies in the United States." Id. ,-r 15.
Indeed, it is currently listed as number eight in the Fortune 500 list, see Fortune.com,
Fortune 500, CVS Health, https://fortune.com/fortune500/2019/cvs-health, and "operates
the nation's largest retail pharmacy chain; owns a large pharmacy benefit manager called
Caremark; and is the nation's second-largest provider of individual [Medical Part D
prescription drug plans ("PDPs")], with over 4.8 million members," Compl. ,-r 15. By
acquiring Aetna, CVS purchased "the nation's third-largest health-insurance company
and fourth-largest individual PDP insurer." Id. ,-r 16. Both companies earn billions of
dollars in annual revenue. See id. ,-r,-r 15-16. The Government alleged in its complaint
that their merger would "lessen competition substantially in the sale of individual PDPs"
in sixteen of the geographic regions 3 established by the Centers for Medicare & Medicaid
Services ("CMS"), the agency that administers Medicare Part D. See id. ,-r,-r 1, 2, 39;
United States ex rel. Fox Rx, Inc. v. Omnicare, Inc., 38 F. Supp. 3d 398, 402 (S.D.N.Y.
2014) (explaining that CMS "administers the Government's Medicare and Medicaid
programs").
As soon as the complaint was filed, however, the Government submitted a notice
attaching a proposed consent judgment that would settle. the case. See U.S. Explanation
of Consent Decree Procedures at 1 [Dkt. # 2]. To comply with the proposed judgment,
Aetna would have to divest its individual PDP business to an independently owned
competitor, WellCare Health Plans, Inc. ("WellCare"). The Government describes its
3
The complaint alleges anticompetitive effects in "Arkansas, California, Florida, Georgia, Hawaii,
Kansas, Louisiana, Mississippi, Missouri, New Mexico, North Carolina, Ohio, Oklahoma, South
Carolina, Wisconsin, and the multistate region of Iowa, Minnesota, Montana, Nebraska, North Dakota,
South Dakota, and Wyoming." Compl. ~ 29.
3
proposed remedy as having five primary components:
First, CVS must divest both of Aetna's individual PDP contracts with the
Centers for Medicare and Medicaid Services. . . . Second, the proposed
Final Judgment required CVS and Aetna to transfer all data relating to
Aetna's individual PDP business to WellCare, including information
regarding the amount that Aetna pays to retail pharmacies in exchange for
filling prescriptions for Aetna members and any contracts with brokers that
currently sell Aetna's individual PDPs. Third, during the 60-day period
following the sale to WellCare, the proposed Final Judgment gave
WellCare the opportunity to interview and hire Aetna's current employees
with expertise related to the individual PDP business. Fourth, CVS must, at
WellCare's option, enter into an administrative services agreement to
provide WellCare with all of the services required to manage the divestiture
assets through the 2019 plan year, which ends on December 31, 2019,
including contracting with pharmacy networks, administering the plans'
formularies, and providing back-office support and claims administration
functions. Finally, CVS and Aetna must allow WellCare to use the Aetna
brand for the divestiture assets through the 2019 plan year.
Mot. for Prop. Final J. at 2-3.
Because this is a civil antitrust suit brought by the Government, the proposed
consent judgment is subject to the Tunney Act. See 15 U.S.C. § l 6(b ). That statute
requires the Government to take several procedural steps before moving for entry of its
proposed judgment. 4 See id. § l 6(b )-( d). The Government must publish its proposed
final judgment and a competitive impact statement 5 in the Federal Register at least sixty
4
The Tunney Act also imposes, subject to a statutory exception, the procedural requirement that
defendants "file with the district court a description of any and all written or oral communications" made
by them or on their behalf "with any officer or employee of the United States concerning or relevant to"
the proposed judgment. 15 U .S.C. § 16(g).
5
The Tunney Act requires that a competitive impact statement recite:
(1) the nature and purpose of the proceeding;
(2) a description· of the practices or events giving rise to the alleged violation of the
antitrust laws;
(3) an explanation of the proposal for a consent judgment, including an explanation of
any unusual circumstances giving rise to such proposal or any provision contained
therein, relief to be obtained thereby, and the anticipated effects on competition of such
relief;
4
days before the effective date of the proposed judgment. See id. § 16(b ). During the
sixty-day period, the Government must receive and consider written comments about its
proposed judgment. See id. § 16(b ), ( d). And at the close of the sixty-day period, it must
publish a response to those comments in the Federal Register and file the same response
with the Court. See id. The Government is also required to publish the proposed final
judgment and competitive impact statement in a newspaper of general circulation in the
district where the case is pending and to furnish the competitive impact statement to
members of the public upon request. See id. § 16(b )-( c }. When the Government moved
for entry of final judgment in this case, it certified that it had completed all required
procedural steps. See Cert. of Compliance with Provisions of the Antitrust Procedures
and Penalties Act [Dkt. # 57-2].
In addition to these procedural steps, the Tunney Act "requires that before a
proposed consent judgment" is "approved by the Court~ the Court must determine that
'the entry of such judgment is in the public interest."' United States v. Airline Tariff Pub.
Co., 836 F. Supp. 9, 11 (D.D.C. 1993) (quoting 15 U.S.C. § 16(e)). To establish that its
proposed judgment meets this standard, the Government incorporated its response to the
comments received during the sixty-day notice and comment period into its motion for
entry of final judgment. See Mot. for Prop. Final J. at 4-5. To say the least, that
response left much to be desired. It is rife with conclusory assertions that merely reiterate
( 4) the remedies available to potential private plaintiffs damaged by the alleged violation
in the event that such proposal for the consent judgment is entered in such proceeding;
(5) a description of the procedures available for modification of such proposal; and
(6) a description and evaluation of alternatives to such proposal actually considered by
the United States.
15 U.S.C. § 16(b).
5
the Government's confidence in its proposed remedy, but shed little light on the reasons
for that confidence. Indeed, the Government's perfunctory response to the public
comments was particularly disappointing in light of the volume and quality of the
comments to which it was responding !6
For example, the AMA's comments criticized the Government's proposed
divestiture remedy because the buyer-WellCare-relies on CVS for pharmacy benefit
management7 ("PBM") and retail pharmacy services. See Resp. to Comments at 26-27;
see also id., Ex. TC-003 at 9-12. The AMA contended that CVS has the ability to deny
or restrict WellCare's access to those PBM and pharmacy services and, in so doing,
threaten the success of the Government's proposed remedy. See Resp. to Comments at
26-27. In response, the Government merely asserted that "such foreclosure-whether
directed at WellCare or any other insurer-is unlikely to occur." Id. at 27. This
6
The Government received 173 comments, totaling over 1,800 pages, on its proposed judgment. See U.S.
Resp. to Pub. Comments on Prop. Final J. ("Resp. to Comments") at 1 [Dkt. # 56]; id., Exs. TC-001 to
TC-085. The amici paiiicipating in this case each submitted comments raising plausible objections. The
AMA's submission alone ran 224 pages, including multiple attachments authored by expetis. See id., Ex.
TC-003. AIDS Healthcare Foundation submitted a letter from its Chief of Operations. See id., Ex. TC-
001. Consumer Action and U.S. PIRG submitted a letter on their own behalf and joined a second letter
submitted with Consumer Reports and the Universal Health Care Foundation of Connecticut. See id.,
TC-023 to TC-024. The Pharmacists Society of the State of New York and Pharmacists United for Truth
and Transparency submitted the amicus brief they filed in this action as a comment on the proposed
judgment. See id., Ex. TC-060. In addition to amici's comments, the Government received submissions
from American Pharmacy Cooperative, Inc.; the Association of American Physicians and Surgeons, Inc.;
the Massachusetts Independent Pharmacists Association; the Medical Society of the State of New York;
the National Community Pharmacists Association; and the West Virginia Independent Pharmacy
Association, among many others.
7
Pharmacy benefit managers, or PBMs, "implement[] and administer[]" drug and pharmacy benefits on
behalf of clients, such as health plans, employers, and unions. LtL to the Court from Gary A. Loeber of
CVS ("Loeber Ltr.") at 1 (June 19, 2019) [Dkt. # 118-1]. This involves "contract[ing] with ... retail
pharmacies," '" adjudicat[ing]' ... claims," and "negotiat[ing] discounts and rebates with pharmaceutical
manufacturers." Id. at 1-2. To provide these services, PBMs build nationwide networks of retail
\
pharmacies and design drug formularies, which list the drugs that will be covered by the PBM client. See
Hr'g Tr. at 233:11-19 (June 5, 2019 A.M.) [Dkt. # 111]; Hr'g Tr. at 299:12-300:15, 322:21-323:14 (June
5, 2019 P.M.) [Dkt. # 112].
6
conclusion was apparently based on the Government's review of "evidence [that]
showed ... CVS is unlikely to be able to profitably raise its PBM or retail pharmacy
costs post-merger." Id. at 26. But the Government did not describe the evidence it
reviewed. Nor did it explain how that evidence supports its conclusion that CVS will not
likely be able to profitably raise its prices. Without such a description and explanation,
the Government's response to the AMA's criticism is little more than a bald assertion that
it is right and the AMA is wrong. 8
Rather than risk an uninformed public interest determination that relied too
heavily on responses like these from the Government, I decided to hold hearings on the
Motion to Enter the Proposed Final Judgment. The hearings were designed to assist the
Court in evaluating the public record. The parties and the amici were given the
opportunity to propose up to three witnesses who could be called to testify. The Court
alone would decide which of those witnesses it believed would be most helpful to its
analysis and how much time would be allotted to each· witness. In the end, the amici
were allowed to call a combined total of three witnesses who were permitted to testify for
a total of four hours. CVS and the Government were allowed the same combined total of
witnesses and the same combined total number of hours of testimony. To reinforce my
repeated emphasis that the hearings were not a trial, cross-examination was not permitted.
Only the Court was allowed to ask follow-up questions during the direct examination of
each witness.
At the hearings, which lasted two days, the AMA examined Dr. N eeraj Sood, a
8
See supra note 6.
7
college professor who is an expert on health policy. Consumer Action and U.S. PIRG
jointly examined Dr. Diana Moss, an economist who is president of the American
Antitrust Institute. And the AIDS Healthcare Foundation elicited factual testimony from
the Foundation's Chief Medical Officer, Dr. Michael Wohlfeiler.
After amici's testimony, the parties to the case were permitted a rebuttal
presentation. I heard testimony from Dr. Alan Lotvin, CVS' s Executive Vice President
and Chief Transformation Officer, and from Dr. Lawrence Wu, an expert in economics
offered by CVS. Thereafter, the Government and CVS jointly designated Terri Swanson,
Vice President for Medicare Part D products at Aetna, to testify about the PDP assets
Aetna sold to WellCare. 9
The parties and amici then submitted supplemental briefing that addressed the
evidence presented at the hearings. Finally, oral argument on the Government's Motion
for Entry of Final Judgment was held on July 19, 2019.
STANDARD OF REVIEW
The Tunney Act provides that, when making a public interest. determination, "the
court shall consider":
(A) the competitive impact of such judgment, including termination of
alleged violations, provisions for enforcement and modification, duration of
9
The Government cross-designated Ms. Swanson as a hearing witness, see U.S. Witness List at 5 [Dkt.
# 84], but, for reasons known only to the Government, declined to question her after CVS's examination
was complete, even though I expressly reserved half of the witness's time for questioning by the
Department of Justice, see Minute Order (May 31, 2019); Hr'g Tr. at 336:10-15, 363:2-5 (June 5, 2019
P.M.). Curiously, the Government, in subsequent pleadings, misrepresented that it was never permitted to
call its own witnesses at the hearings. Such phantasmagorical claims are, of course, demeaning to the
credibility of the Antitrust Division's litigation position. As I said when I corrected the record at our most
recent hearing, "I hope and trust the Department of Justice will refrain from these misrepresentations
going forward. Doing so reflects poorly on the important work that they are doing on a regular basis for
the good of the American people." Hr'g Tr. at 10: 14-17 (July 19, 2019) [Dkt. # 133].
8
relief sought, anticipated effects of alternative remedies actually
considered, whether its terms are ambiguous, and any other competitive
considerations bearing upon the adequacy of such judgment that the court
deems necessary to a determination of whether the consent judgment is in
the public interest; and ·
(B) the impact of entry of such judgment upon competition in the relevant
market or markets, upon the public generally and individuals alleging
specific injury from the violations set forth in the complaint ....
15 U.S.C. § 16(e)(l).
The public interest mqmry is "not ... 'a de nova determination of facts and
issues."' United States v. Newpage Holdings Inc., No. 14-2216, 2015 WL 9982691, at *5
(D.D.C. Dec. 11, 2015) (quoting United States v. Western Elec. Co., 993 F.2d 1572, 1577
(D.C. Cir. 1993)). The Court need "only ... confirm that the ... settlement is within the
reaches of the public interest." United States v. Microsoft Corp., 56 F.3d 1448, 1460
(D.C. Cir. 1995) ( quotation marks, citation, and italics omitted).
But neither is the inquiry a mere formality or judicial rubberstamp. If, for
example, a proposed consent "decree is ambiguous, or the district judge can foresee
difficulties in implementation," the decree should not be entered until the problems are
fixed. Microsoft, 56 F.3d at 1462. "[I]f third parties contend that they would be
positively injured by the decree, a district judge might well hesitate before assuming that
the decree is appropriate." Id. And no "judge is ... obliged to accept [a consent decree]
that, on its face and even after government explanation, appears to make a mockery of
judicial power." Id.
Throughout this case, the Government has repeatedly asked this Court to dismiss
out of hand many of amici's objections to its proposed final judgment. Relying for the
9
most part on United States v. Microsoft Corporation, 56 F.3d 1448 (D.C. Cir. 1995), the
Government argues that consideration of harms that were not alleged in the complaint
would "aggravate ... 'constitutional difficulties that inhere'" in the Tunney Act. U.S.
Resp. to Order to Show Cause at 2 [Dkt. # 32] (quoting Microsoft, 56 F.3d at 1459). To
avoid this purported aggravation, the Government contends that the Court must ignore all
evidence regarding "harm outside of the individual PDP market," "theor[ies] of harm that
the United States did not allege," and "efficiencies" gained from the merger. U.S. Mot.
to Limit the Scope of the Tunney Act Hr' g at 4-6 [Dkt. # 82].
To say the least, these arguments severely understate the permissible scope of a
Tunney Act review. The "constitutional difficulties" our Circuit Court addressed in
Microsoft were implicated because a court "reach[ ed] beyond the complaint to evaluate
claims that the government did not make." 56 F.3d at 1459 (first italics added). Courts
cannot, of course, "force the government to make [a] claim." Id. at 1460. The
Government, alone, chooses which causes of action to allege in its complaint.
But Microsoft never says that allegations in the complaint are the only harms
courts may consider in a Tunney Act review .10 Indeed, such a holding would have
contradicted the Tunney Act itself. There, Congress ·directs courts to consider "the
impact of entry of [a] judgment" on both "individuals alleging specific injury from the
°
1
Curiously, this appears to be the Government's actual position. See Hr'g Tr. at 19:24-20:8 (July 19,
2019) ("THE COURT: So the government's position is even ifthe Comi were to somehow identify harms
that were not mentioned in the complaint and those harms would in some way undermine the public
interest, the Court is to turn a blind eye to those and focus on just the harms that were alleged in the
complaint? Is that essentially your position? [Counsel for the Government]: Yes, Your Honor. The
public-interest determination here is measured in light of the allegations made in the United States'
complaint.").
10
violations set forth in the complaint" and "the public generally." 15 U.S.C. § 16( e )(1 )(B)
(emphasis added); see also United States v. ENS Inc., 858 F.2d 456, 462-63 (9th Cir.
1988) ("[T]he [Tunney Act] does not authorize a district court to base its public interest
determination on antitrust concerns in markets other than those alleged in the
government's complaint. ... Nevertheless, the statute clearly indicates that the court may
consider the impact of the consent judgment on the public interest, even though that effect
may be on an unrelated sphere of economic activity." (emphasis added)). A Tunney Act
review that ignores harms that will flow from the entry of a proposed judgment to the
general public ignores the language Congress uses in the Act. See Advocate Health Care
Network v. Stapleton, 137 S. Ct. 1652, 1659 (2017) (Courts "'presum[e] that each word
Congress uses is there for a reason.").
The Microsoft Court recognized this very point. It explained that the Tunney Act
reflects Congress's understanding that "a consent decree might well do unexpected harm
to persons other than those 'alleging specific injury from the violations set forth in the
complaint,'" so "district court[ s] might ponder those sort of concerns in determining
whether to enter [a proposed] judgment." Microsoft, 56 F.3d at 1459 (quoting 15 U.S.C.
§ 16(e)(2) (1988), now codified at 15 U.S.C. § 16(e)(l)(B)). The Microsoft Court went
on to specifically identify three examples of when complaint allegations do not
circumscribe a Tunney Act inquiry. Courts, it said, may review whether a "decree is
I
ambiguous, or ... difficult[] [to] implement[]," whether "third parties ... w[ill] be
positively injured by the decree," and whether entry of a. decree would "make a mockery
of judicial power." Id. at 1462. Adequately reviewing any of these potential harms to the
11
public interest could require a court to look beyond the four corners of the. Government's
complaint. But none constitute the "evaluat[ion of a] claim[] that the government did not
make." Id. at 1459 (italics omitted). All, therefore, are fair game. See id. at 1462; ENS
Inc., 858 F.2d at 462-64.
The Government's suggestion here-that by narrowly drafting a complaint it can
effectively force the Court to shut its eyes to the real-world impact of a proposed
judgment-thus misconstrues Microsoft. 11 It also strikes-at the heart of the Tunney Act's
very purpose. Congress passed the law to "ensure[] that the economic power and
political influence of antitrust violators do not unduly influence the government into
entering into consent decrees that do not effectively remedy antitrust violations." Airline
Tariff Pub. Co., 836 F. Supp. at 11. The Government's position here could actually
facilitate such undue influence so long as unduly influenced attorneys strategically draft
complaints to shield their indifference to the public interest from judicial review. Neither
the statute, nor Microsoft, supports such a reading. Therefore, while the Government is
certainly entitled to great deference-if not a presumption of accuracy-when it contends
that a proposed final judgment is in the public interest, evidence by third parties that
11
Reading Microsoft too narrowly, as the Government does, also ignores Congress's response to that very
case. In 2004, Congress amended the Tunney Act and included in the amendments congressional
findings that reject a narrow reading of Microsoft. See Antitrust Criminal Penalty Enhancement and
Reform Act of 2004, Pub. L. 108-23 7, Title II, § 221 (a)( 1)(B), 118 Stat. 661, 668 (June 22, 2004) ("[I]t
would misconstrue the meaning and Congressional intent in enacting the Tunney Act to limit the
discretion of district comis to review antitrust consent judgments solely to detem1ining whether entry of
those consent judgments would make a 'mockery of the judicial function."'); 150 Cong. Rec. S3617-18
(Apr. 2, 2004) (statement of Sen. Kohl) ("The language quoted paraphrases the D.C. Circuit decisions in
Massachusetts School of Law v. US., 118 F .3d 776, 783 (D.C. Cir. 1997) and US. v. Microsoft, 56 F Jd
1448, 1462 (D.C. Cir. 1995). To the extent that these precedents are contrary to section 221 (a) of our bill
regarding the standard of review a court should apply in reviewing consent decrees 1.mder the Tunney Act,
these decisions are overruled by this legislation.").
12
persuasively demonstrates actual or likely harm to the public interest will overcome that
presumption and the proposed final judgment will be denied.
Thus, based on a correct reading of Microsoft and the Tunney Act, the potential
harms to the public interest raised by the amici in this case fall within the permissible
scope of this Court's review. As such, notwithstanding the Antitrust Division's
protestations to the contrary, a judicial evaluation of those alleged harms raises no
constitutional issue.
ANALYSIS
Amici's testimony and briefs alleged a variety of harms to the public interest that
the final judgment would cause if it were entered by the Court. While those concerns
shed a healthy light on this merger, the amici did not substantially undermine the parties'
public interest position by persuasively demonstrating that their concerns currently exfst
or are likely to develop. This was especially so in light of the rebuttal presentations by
CVS and the Government. A review of the amici's major concerns should be sufficient. 12
Principal among amici's concerns were the following three contentions:
(i) Aetna's divestiture to WellCare will not effectively remedy the harm .to the PDP
market alleged in the complaint; (ii) the proposed final judgment's failure to address
effects in markets adjacent to the PDP market-like the market for PBM services-will
undercut the effectiveness of the divestiture remedy and harm the public; and (iii) entry
of the proposed final judgment without modification will harm HIV and AIDS patients in
12
Amici had a number of other concerns, but they address harms that, by compariso,n, are less likely than
their principal concerns to undermine the parties' claim that the merger is in the public interest. As such,
they do not warrant futiher discussion.
13
need of affordable, quality healthcare. I will discuss each in turn.
First, amici contend that the divestiture to WellCare mandated by the proposed
judgment will not remedy the competitive harm to the PDP market alleged in the
Government's complaint. Amici argue that, according to the Herfindahl-Hirschman
Index ("HHI"), 13 the divestiture leaves the PDP market overly concentrated and
anticompetitive and that WellCare-which is smaller than Aetna, has a less recognizable
brand than Aetna, and simultaneously competes against and contracts with CVS-will
not be as strong a competitor in the PDP market as Aetna was.
In response, CVS and the Government provided a· more persuasive rebuttal. Terri
Swanson, the Vice President in charge of Aetna's Medicare Part D products prior to the
divestiture, was the hearing witness most familiar with the PDP market and the PDP
assets that WellCare purchased. See Hr'g Tr. at 336:24-337:16, 338:10-16, 339:9-22
(June 5, 2019 P.M.). Her testimony focused on how the PDP market is already highly
competitive. See id. at 342:22-344:5. She described CMS's Medicare Plan Finder tool,
which allows PDP customers to compare plans in granular detail. See id. at 343: 16-
344: 5. Using CMS' s tool, customers can see "the price of a specific drug at a specific
pharmacy for a specific plan" before purchasing a PDP plan. Id. CMS also makes it
"very easy for [customers] to switch" PDP plans online or over the phone. Id. at 344:6-
14. Ms. Swanson explained that, because PDP plans can be easily compared and
swapped, variables like brand recognition are far less important to maintaining PDP
13
"Market power or the lack of it is often measured by the HHI," which "is calculated by squaring the
individual market shares of all firms in the market and adding up the squares." FTC v. PPG Indus., 798
F.2d 1500, 1503 (D.C. Cir. 1986).
14
market share than simply offering the drug coverage customers want for a couple of
dollars less than competitors. See id. at 346: 17-348: 11.
She also emphasized that WellCare has been a successful competitor in these very
conditions. Prior to Aetna's divestiture, WellCare "added around half a million" new
members-growing from "about 1.1 million members" to "about 1.6 million
members"-by underselling Aetna "a dollar or two" on "similarly situated plans." Hr'g
Tr. at 340:8-342:3, 346:17-347:20 (June 5, 2019 P.M.). WellCare did so despite being
smaller than Aetna. See id. at 342:4-11. It also did so while both competing against
CVS in the PDP market and contracting with CVS for PBM services. See id. at 360: 17-
361 :7; Hr'g Tr. at 228:9-228:19; 251:10-19 (June 5, 2019 A.M.). WellCare thus
showed, before the divestiture, that it could attract customers from larger competitors
under conditions similar to those it faces now. It stands to reason that, notwithstanding
amici's concerns, WellCare can and will continue to. compete post-merger, after its
market share has been bolstered by the purchase of Aetna's assets.
With respect to amici's other concern-PDP market concentration-CVS's
expert, Dr. Lawrence Wu, emphasized that amici' s HHI analysis shows that the PDP
market is "moderately," as opposed to "highly," concentrated under the Government's
guidelines. See Hr'g Tr. at 272:23-273:24 (June 5, 2019 A.M.). According to the
Government, the HHI scores at issue here merely indicate potential competitive concerns,
not a presumption of market power. 14 See Resp. to Comments at 21-22. And while
14
The AMA does argue that in one geographic market-Hawaii-the divestiture causes an increase in
HHI score that is sufficient to raise a presumption of enhanced market power. See Hr' g Tr. at 40:7-14
(July 19, 2019). But even "where the HHI calculation ... indicates a merger to be presumptively illegal,"
15
amici argue that studies have concluded that increased concentration in the PDP market
causes premiums to rise, see Hr'g Tr. at 46:9-19 (July 19, 2019), CVS and the
Government have pointed to evidence that shows the market has remained quite
competitive. For example, CMS reported in July 2018 that, "for the second year in a row,
the average basic premium for a Medicare Part D prescription d_rug plan ... is projected
to decline." CMS, Medicare Part D Premiums Continue to Decline in 2019 (July 31,
2018), https://www.cms.gov/newsroom/press-releases/medicare-part-d-premiums-contin
ue-decline-2019; see also Hr' g Tr. at 70:9-20 (July 19, 2019). Considering that the PDP
market was already moderately concentrated when this announcement was made by
CMS, see Hr'g Tr. at 43:16-44:3 (June 4, 2019 A.M.) [Dkt. # 109], amici's concerns are
not borne out by present circumstances.
In the final analysis, PDP customers do h 4ve numerous options when picking a
plan, and many of the plans are offered by major healthcare players, including United
Healthcare, Humana, Cigna, and Rite Aid. See Hr'g Tr. at 342:22-343:15 (June 5, 2019
P.M.). The moderate concentration in the PDP market has neither prevented WellCare
from competing in the market, nor prevented price competition from driving premium
prices down, in recent years. 15 Accordingly, amici's warnings of similar conditions after
the calculation "establish[ es] [only] a prima facie case of an anticompetitive merger," necessitating "a
more comprehensive and holistic assessment of whether the proposed merger is likely to create or
enhance market power or facilitate its exercise." FTC v. Arch Coal, Inc., 329 F. Supp. 2d 109, 124
(D.D.C. 2004). For the reasons discussed throughout this Memorandum Opinion, the evidence showed
that PDP markets remain competitive despite amici's HHI calculation. This is true even in Hawaii, where
the average PDP price fell in 2019. See Hr'g Tr. at 70:9-20 (July 19, 2019).
15
Amici also argued that the WellCare divestiture will be ineffective because WellCare is unlikely to
retain all of Aetna' s former PDP customers. But because the evidence showed that the individual PDP
market is competitive, with multiple plan options available to customers, concerns about WellCare's
customer retention rate do not establish the divestiture is contrary to the public interest. Former Aetna
16
Aetna's divestiture do not persuasively establish that the proposed judgment is ineffective
or will otherwise be contrary to the public interest.
Amici next argue that the proposed judgment's failure to address the merger's
effects in PDP-adjacent markets-such as the market for PBM services-threaten the
success of the Government's proposed remedy and potentially harm third parties and the
general public. How so?
CVS is a major player in the PBM market. It accounts for close to a quarter of
PBM market share, and with the two other top PBMs, CVS is part of "the big three" that
control about 70% of the market. See Adam J. Fein, The CVS-Aetna Deal: Five Industry
and Drug Channel Implications, Drug Channels (Dec. 5, 2017), https://www.
drugchannels.net/2017 / 12/the-cvs-aetna-deal-five-industry-and.html.
Amici argue that CVS can leverage this power in the PBM market to disadvantage
companies that compete against its newly expanded health insurance business. They
theorize that CVS could, for example, raise the price of its PBM services when selling the
services to health insurance competitors. 16 If the competitors cannot find cheaper PBM
services elsewhere, CVS' s rivals may then have to raise the price of their insurance
products or accept reduced profits, leaving CVS with the more attractive insurance
offerings or more profitable business. Amici argue that this would harm members of the
general public, who could end up overpaying for pharmacy benefits. And it would
customers who choose to leave WellCare will likely find themselves in a competitive market and will not
necessarily need to turn to CVS to purchase a new PDP plan.
16
Amici also theorize that CVS could reduce access to ce1iain formulary-listed drugs or PBM services
post-merger. See, e.g., Hr'g Tr. at 159:22-163:12 (June 4, 2019 P.M.) [Dkt. # 115]. Those theories raise
the same issues as amici's contention that CVS could raise the price of its PBM services, so they need not
be analyzed separately.
17
threaten the success of the Government's proposed divestiture remedy because WellCare,
which both competes against CVS in the PDP market and contracts with CVS for PBM
services, would be vulnerable to such a tactic.
But again, CVS presented more persuasive evidence that substantially undermines
amici's theory. Notwithstanding CVS's significant market share, the evidence showed
that CVS must compete vigorously to retain its PBM customers. Dr. Wu explained that
CVS's PBM competes on two fronts. First, rival PBMs try to underbid CVS. See Hr'g
Tr. at 222: 15-224: 1 (June 5, 2019 A.M.). Indeed, Dr. Wu concluded, based on a review
of bid data, that CVS "lose[ s] bids all the time." Id. In fact, it "lost business to more than
ten different PBM competitors" in 2017. Resp. to Comments, Ex. TC-003 at 133.
Second, CVS's PBM oftentimes competes against its own customers. See Hr'g Tr. at
263:3-264:9 (June 5, 2019 A.M.). After all, health insurance companies can move PBM
services in house when and if they consider CVS' s price for contract services too high.
See id.; Hr'g Tr. at 324:20-326:7 (June 5, 2019 P.M.).
Moreover, PBM customers have ways of ensuring that they receive the best deal
the market can offer. Clients, for example, "demand market checks, which means, even
if you have a 3-year contract, ... during the contract[,] ... that PBM client will have the
right to take the business back out for a market check." Hr' g Tr. at 326:8-22 (June 5,
2019 P .M. ). PBM clients also negotiate directly with pharmaceutical manufacturers to
confirm that their PBM is providing the lowest available price for the manufacturer's
drugs. See id. at 325:2-7.
This evidence combined strongly suggests that, if CVS were to raise its PBM
18
pnces, customers like WellCare could simply switch to a less expensive PBM or stop
contracting for those PBM services altogether. Were CVS to raise PBM prices in this
scenario, it would risk losing PBM market share without disadvantaging WellCare or
other competing insurers at all. To say the least, that would be an enormous risk for CVS
to take.
Indeed, Dr. Lotvin addressed that very problem when he testified that focusing
CVS's business only on its health insurance customers "would be economic suicide."
Hr'g Tr. at 328:15-329:3 (June 5, 2019 P.M.). CVS's PBM business covers about 90
million lives. See id. Its pharmacies 17 "probably see a third of the people in the country
over the course of [a] year." Id. The health insurance business CVS acquired from
Aetna, by contrast, covers about 22 million lives, and Aetna provided pharmacy benefits
to fewer than 10 million of them. 18 See id.; Loeber Ltr. at 2. Amici may well be correct
that CVS wants to grow its health insurance business. But that does not mean the
company would risk jeopardizing its 90-million-life PBM business or 100-million-life
17
Amici also raised concerns about CVS 's retail pharmacies. But the arguments about pharmacies do not
show harm to the public interest for the same reasons that amici's concerns about CVS's PBM business
do not. Again, the evidence showed that, if CVS were to raise prices for or reduce access to its retail
pharmacies, health insurers could take their business to CVS' s competitors. Dr. Lotvin testified that most
national pharmacy networks include around 50,000 to 60,000 pharmacies. See Hr'g Tr. at 335: 1-3 (June
5, 2019 P.M.). CVS owns less than 15% of the nation's approximately 70,000 retail pharmacies. See
Hr'g Tr. at 32:9-33:2 (June 4, 2019 A.M.). So a network of more than 50,000 pharmacies can be
constructed without including a single one owned by CVS. In fact, within the last three years, CVS's
pharmacies have been excluded from a major pharmacy network. See Shelby Livingston, CVS Pushed
Out of Tricare Pharmacy Network, Modern Healthcare (Oct. 3, 2016), https://www.
modernhealthcare.com/article/20161003/NEWS/l 61009986/cvs-pushed-out-of-tricare-pharmacy-network
("In a blow to CVS Health Corp., pharmacy benefit manager Express Scripts Holding Co. announced it
has ousted the provider from the pharmacy network it manages for Tricare, the U.S. Defense
Depatiment's health benefits program with 9.4 million beneficiaries."). Like with its PBM business, the
evidence showed that CVS is likely to lose customers if it tries to disadvantage competitors by raising the
price of, or reducing access to, its retail pharmacy services.
18
In addition, CVS's individual PDP business covers almost 5 million lives. See Compl. ~ 15.
19
pharmacy business to funnel customers toward its much, much smaller health insurance
business. As I pointed out during the hearings, that would surely be a case of "cutting off
your nose to spite your face"! Hr'g Tr. at 261:23-24 (June 5, 2019 A.M.).
As such, the record here did not persuasively establish amici' s contention that the
proposed final judgment's failure to address the PBM market will likely result in harm to
the public interest.
Finally, amici argue that the proposed final judgment will harm certain HIV and
· AIDS patients. Dr. Michael Wohlfeiler, of the AIDS Healthcare Foundation, testified
about the importance of comprehensive, HIV-and-AIDS-specific treatment programs in
addressing those conditions. See Hr' g Tr. at 107 :9-113: 17 (June 4, 2019 P .M.). Success
in treating HIV and AIDS can often turn on patients' adherence to complicated medicinal
regimens. See id. at 109:4-110:2. Those adherence rates increase when specialists
remain in frequent contact with patients. See id. at 110:13-112:10. And even simple
procedures, like flu shots, can be fraught with danger when a patient with HIV or AIDS
receives treatment from someone who lacks specialized training. See id. at 127:9-24.
Dr. Wohlfeiler' s testimony demonstrated that, if the proposed final judgment were to
cause patients to leave HIV-and-AIDS-specific treatment providers for providers that are
unequipped to treat those conditions, the judgment could cause harm.
For the reasons already discussed, however, the record did not establish that the
proposed final judgment will likely result in CVS gaining the ability to steer patients
away from their current healthcare providers. Indeed, that reasoning applies with even
more force to patients receiving healthcare from the AIDS Healthcare Foundation
20
because it embeds its own pharmacies in its clinics and contracts with Medlmpact, one of
CVS's competitors, for PBM services. See Hr'g Tr. at 103:8-15 (June 4, 2019 P.M.);
Hr'g Tr. at 224:9-10 (June 5, 2019 A.M.). Put simply, if the record did not establish that
CVS will be likely to steer customers away from WeliCare, which relies on CVS for
PBM services, it certainly did not establish that CVS will be likely to steer patients away
from the AIDS Healthcare Foundation, which uses a different PBM and maintains its
own pharmacies. As such, the potential harm to this segment of the public was not
persuasively established on the record either. 19
CONCLUSION
Although amici raised substantial concerns that warranted serious consideration,
CVS' s and the Government's witnesses, when combined with the existing record,
persuasively support why the markets at issue are not only very competitive today, but
are likely to remain so post-merger. Consequently, the harms to the public interest the
amici raised were not sufficiently established to undermine the Government's conclusion
to the contrary.
As such, for all of the above reasons, I have concluded that the proposed
settlement is well "within the reaches" of the public interest and the Government's
Motion to Enter the Proposed Final Judgment [Dkt. # 57] should therefore be
GRANTED. See Microsoft, 56 F.3d at 1460 (In the finalanalysis, "the court's function is
not to determine whether the resulting array of rights and liabilities is the one that will
19
The AIDS Healthcare Foundation requested several specific· modifications to the Government's
proposed final judgment. Since I have determined that the proposed judgment, as submitted, is in the
public interest, it will be entered without the modifications proposed by the AIDS Healthcare Foundation.
21
best serve society, but only to confirm that the resulting settlement is within the reaches
of the public interest." (quotation marks and citations omitted)). An Order consistent
with this decision and an executed version 20 of the Final Judgment that was submitted
with the Government's motion accompany this Memorandum Opinion.
20
The executed version of the Final Judgment was modified slightly in the last paragraph of the preamble
by deleting the words "before any testimony is taken," which, if left in, would obviously conflict with the
two days of hearings held by the Comi in June 2019.
22