Federal Trade Commission v. Meta Platforms, Inc.

UNITED STATES DISTRICT COURT

FOR THE DISTRICT OF COLUMBIA

FEDERAL TRADE COMMISSION,

Plaintiff,

v. Civil Action No. 20-3590 (JEB)

META PLATFORMS, INC.,

Defendant.

MEMORANDUM OPINION

How times have changed. Not so long ago, a company called Facebook so dominated the

provision of personal social-networking services that a movie charting its rise was entitled

simply “The Social Network.” Fifteen years later — and four years into the Federal Trade

Commission’s antitrust suit against it — that company (now called Meta) argues that the market

with which it was once nearly synonymous does not even exist. Gazing across an online

landscape teeming with companies catering to every conceivable interest, Meta sees a bloody

battle for users’ time and attention in which its products face withering competition. After over

two years of discovery, it now moves for summary judgment on the FTC’s claims that it violated

Section 2 of the Sherman Act through its acquisitions of Instagram and WhatsApp. The FTC

opposes that Motion and has itself cross-filed for partial summary judgment on certain of Meta’s

affirmative defenses. Both parties have buttressed their filings with a voluminous evidentiary

record, mere summaries of which span thousands of pages.

1

Although the field of the parties’ combat has shrunk somewhat over time, the FTC’s story

has remained “essentially unchanged.” FTC v. Facebook, Inc. (Facebook II), 581 F. Supp. 3d 34,

40 (D.D.C. 2022). Armed now with experts and evidence, Plaintiff continues to argue that a

relevant antitrust market exists for personal social-networking (PSN) services and that, since at

least 2012, Meta has enjoyed monopoly power in that market. The Commission further contends

that Defendant unlawfully maintained that monopoly by acquiring two actual or nascent

competitors, Instagram and WhatsApp, that posed a threat to its dominance at the time. Meta

retorts that PSN services do not constitute a relevant antitrust market over which the company

exercises any degree of monopoly power and that no other direct effects of monopoly power

have been shown. Even if it did possess such power, Meta denies that its Instagram and

WhatsApp acquisitions were anticompetitive, especially considering the significant investments

it has since made to improve and grow both applications.

In the end, while the parties’ legal jousting is both impressive and comprehensive, it

leaves no clear victor. This case must go to trial. Under the forgiving summary-judgment

standard, the FTC has put forward evidence sufficient for a reasonable factfinder to rule in its

favor. Prevailing here, however, does not obscure the fact that the Commission faces hard

questions about whether its claims can hold up in the crucible of trial. Indeed, its positions at

times strain this country’s creaking antitrust precedents to their limits. All the same, except for

Count II, which the Court previously found lacking but could not dismiss, and one of Meta’s

defenses, for which Defendant has not proffered sufficient evidence to survive summary

judgment, the Court will deny Meta’s Motion for Summary Judgment and the FTC’s Cross-

Motion for Partial Summary Judgment.

2

Table of Contents

I. Background ............................................................................................................................. 4

A. Factual Background ............................................................................................................ 4

1. Facebook ...................................................................................................................... 4

2. The Acquisitions ........................................................................................................... 6

a. Instagram................................................................................................................... 6

b. WhatsApp ................................................................................................................. 8

3. Other Applications ...................................................................................................... 10

B. Procedural History ............................................................................................................ 12

II. Legal Standard ...................................................................................................................... 14

III. Analysis............................................................................................................................. 16

A. Monopoly Power............................................................................................................... 17

1. Market Definition ....................................................................................................... 18

a. Brown Shoe Factors ................................................................................................ 22

b. Expert Testimony .................................................................................................... 39

2. Market Share .............................................................................................................. 43

3. Barriers to Entry......................................................................................................... 47

B. Anticompetitive Conduct .................................................................................................. 52

1. Legal Framework ....................................................................................................... 53

a. Anticompetitive Effects .......................................................................................... 53

b. Nascent Threats ....................................................................................................... 62

2. Acquisitions ................................................................................................................ 65

a. Instagram................................................................................................................. 65

b. WhatsApp ............................................................................................................... 72

3. Procompetitive Justifications ..................................................................................... 76

a. Legal Framework .................................................................................................... 77

b. Application.............................................................................................................. 83

IV. Conclusion ........................................................................................................................ 92

3

I. Background

As many of the relevant facts remain contested, the Court will address them later when

examining the parties’ contentions. Here, it provides an overview of the undisputed facts and the

history of the litigation to this point.

A. Factual Background

1. Facebook

Although those sufficiently interested to be reading this Opinion are likely familiar with

Meta’s history, it is worth a brief retelling. Mark Zuckerberg founded Facebook in 2004 as “The

Facebook,” a “social-networking service initially limited to college students.” FTC v. Facebook,

Inc. (Facebook I), 560 F. Supp. 3d 1, 6 (D.D.C. 2021). The parties continue to vigorously dispute

the precise definition and boundaries of PSN services, but, at a minimum, they are online

platforms that enable “users to virtually connect with others in their network and to digitally

share their views and experiences by posting about them in a shared, virtual social space.” Id. at

5–6. While Facebook may have been early to the game, it was not the first: other PSN

companies include now-defunct Friendster and Myspace, both of which launched in 2002. Id. at

5. By 2009, however, Facebook had “surpassed Myspace to become the most widely used online

social networking service in the United States.” ECF No. 363-2 (FTC’s Statement of Material

Facts), ¶ 1560.

As originally conceived, Facebook consisted of a few simple features: a “Profile,” where

users could share information about themselves, coupled with the ability to add other Profiles as

“Friends” — “reciprocal relationship[s]” that grant users access to information and updates about

each other. See ECF No. 325-2 (Meta’s Statement of Material Facts), ¶¶ 18–22. Within a few

years of its founding, Facebook enabled users to upload photos and videos to their Profiles,

4

followed shortly thereafter by the creation of the “Feed” (formerly the “News Feed”), which

presented a mix of content from people’s Friends. Id., ¶¶ 23–25. Other features (also known as

“surfaces”) have since rolled in: “Pages” (2007), which permits users to follow businesses,

organizations, and public figures to get updates; “Groups” (2010), which allows users to form a

subcommunity for any reason or none at all; “Live” (2015), which is Facebook’s mobile video

livestreaming service; “Marketplace” (2016), where users can buy and sell items with people in

their community; “Stories” (2017), which permits users to share “full-screen, short-form,

ephemeral forms of content” with their Friends and followers; “Gaming” (2019), where users can

play games, watch gaming videos, and connect with gaming Groups; “Dating” (2019), which

facilitates romantic connections; and “Reels” (2021), which displays short-form videos. Id.,

¶¶ 26–44 (internal quotation marks omitted). In addition, while Facebook users have been able

to message one another almost since its launch, id., ¶ 94, in 2011 Facebook launched

“Messenger” as a stand-alone application. Id., ¶ 95. Currently, Messenger enables users to,

among other things, send individual messages, start group chats, conduct voice and video calls,

and post Stories. Id., ¶¶ 96–112.

Facebook managed fantastic growth during its first two decades of existence. By the end

of 2009, it boasted 112 million monthly active users in the United States and Canada, a number

that expanded to 158.9 million U.S. monthly active users by early 2012 and such

users by the first half of 2022. See FTC SMF, ¶ 1561; Meta SMF, ¶ 725. And it charged those

users nothing to use the service. See Meta SMF, ¶ 5. That remains true today across all four

applications that constitute the “family” of Meta’s products: Facebook, Messenger, Instagram,

and WhatsApp. Id., ¶¶ 4–5. Instead, Meta generates revenue primarily through advertising on

Instagram and Facebook. Id., ¶¶ 132–33. Through its “Ads Manager,” Meta lets advertisers

5

customize their campaigns to target specific segments of users on various surfaces, with options

to increase or decrease spending based on feedback about how the advertisements are performing

with the target audience. Id., ¶¶ 134, 136. In 2021 alone, Meta earned $114.9 billion in

advertising revenue. See FTC SMF, ¶ 1439(c).

Since its founding, Zuckerberg has remained at the helm. He is its CEO and the

Chairman of its Board of Directors, and he controls a majority of the company’s voting shares.

Id., ¶ 1562. He has testified that he cannot “think of any examples” of when the Board ever

overruled a decision that he had made. See ECF No. 356-1 (Pl. Docs. Vol. 58) at ECF p. 839

(FTC Investigational Hearing of Mark Zuckerberg at 24:11–24).

2. The Acquisitions

Facebook did not merely invent new features and collect new users, however. As is by

now clear, it also grew by acquisition. Indeed, the company changed its name to Meta in 2021 to

unite its various “apps and technologies under one new company brand.” Introducing Meta: A

Social Technology Company, Meta (Oct. 28, 2021), https://perma.cc/5MPH-7WU5. The Court

will refer to “Meta” when discussing the company formerly known as “Facebook,” and to

“Facebook” when discussing the specific PSN platform discussed above, which is also called

“Facebook Blue”—“what its millions of users think of when they think of ‘Facebook.’”

Facebook I, 560 F. Supp. 3d at 6. Two acquisitions are at the center of this lawsuit: those of

Instagram and WhatsApp.

a. Instagram

Founded in 2010 by Kevin Systrom and Mike Krieger, see Meta SMF, ¶ 55, Instagram

was a photo-editing and -sharing service with similar features to Facebook. For instance, users

could likewise set up a “Profile” with some of their personal information, connect with other

6

users via a one-way “Follow,” and browse content posted by accounts they follow on

Instagram’s “Feed.” Id., ¶¶ 67–69, 73. Initially, users were limited to sharing photos that they

had either taken with an in-app camera or uploaded from their own photo library. Id., ¶¶ 70, 672.

Currently, however, users can share both photos and videos as posts to the Feed or as “Stories,”

id., ¶¶ 71–72, 82–83, post short-form videos or “Reels,” id., ¶¶ 89–90, message other users

directly via Instagram “Direct,” id., ¶¶ 79–80, broadcast a video live to their followers through

Instagram “Live,” id., ¶ 85, shop online via Instagram “Shop,” id., ¶ 87, and post text-based

content to the public on Instagram “Threads.” Id., ¶ 92.

Prior to its acquisition by Meta in 2012, Instagram had just over a dozen employees and

approximately 3.9 million U.S. monthly active users. Id., ¶¶ 657–58; see also FTC SMF, ¶ 1608

(35 million total users). Its chief differentiator was its focus on mobile rather than desktop use.

The technological “shift” from desktop to mobile created a “brief moment and opening . . . for [a

company] like Instagram to create a mobile-optimized experience.” Pl. Docs. Vol. 58 at ECF

p. 551 (FTC Investigational Hearing of Kevin Systrom at 82:9–17). That mobile orientation

drove Instagram’s decision to focus on photo sharing, taking advantage of the increasingly high-

quality cameras found on consumers’ smartphones. See id. at 44:20–24 (“[W]ith the camera in

your pocket in the form of a cell phone[,] . . . the primary way we would communicate would be

with photos, and possibly videos some day . . . .”); FTC SMF, ¶ 1567 (describing launch of

iPhone 4, which featured significantly improved camera). The company leveraged its mobile-

first orientation by permitting users to “rebroadcast” their posts onto the other platforms, thus

. See Meta SMF, ¶¶ 660–61.

In February and March of 2012, Zuckerberg began communicating to Systrom his

interest in acquiring Instagram. Id., ¶¶ 696–97. The parties dispute Meta’s motivations for the

7

acquisition. Meta points to evidence that it sought to “grow Instagram as an independent brand

and product,” id., ¶ 698 (quoting email from Zuckerberg to Systrom), while the FTC insists that

Meta “acquired Instagram to eliminate a competitive threat,” ECF No. 328-2 (FTC Resp. to Meta

SMF) at 564, particularly in light of the challenges Meta faced in transitioning the desktop

version of Facebook to the mobile environment. See FTC SMF, ¶¶ 1569–74, 1576–86. On

April 8, 2012, Zuckerberg and Systrom agreed that Meta would acquire Instagram for $1 billion.

See Meta SMF, ¶ 701.

Just over a week later, Meta submitted to the FTC a Hart-Scott-Rodino (HSR) Act

Premerger Notification Form for the transaction. Id., ¶ 702; see 15 U.S.C. § 18a. The FTC’s

review of the transaction took over four months and involved a “rare” request for additional

information and documents. See Facebook I, 560 F. Supp. 3d at 7; Meta SMF, ¶ 704.

Eventually, however, the FTC cleared the acquisition to proceed. See Meta SMF, ¶ 707. In the

first half of 2022, Instagram had more than U.S. monthly active users. Id., ¶ 724.

b. WhatsApp

In September 2009, Brian Acton and Jan Koum launched WhatsApp as an “over-the-top”

(OTT) mobile-messaging service. See FTC SMF, ¶¶ 1732, 1765. OTT services use the internet,

rather than traditional telephone infrastructure, to enable users to communicate with one another.

Id., ¶ 1714. As a cross-platform application, WhatsApp could be downloaded and used on

different types of devices. Id., ¶ 1733. Android users of WhatsApp paid a yearly subscription

fee of $0.99, while iPhone users could use the application for free after paying — at least for a

time — a one-time download fee of $1.00. See Meta SMF, ¶ 763; see also FTC Resp. to Meta

SMF, ¶ 763 (pointing out that WhatsApp removed download fee in July 2013).

8

Users quickly flocked to the service: by March 2011, the company had 17 million

registered users, 11 million monthly active users, and 7.5 million daily active users, and it was

registering 100,000 new users each day. See FTC SMF, ¶ 1735(a). By April 2013, those

numbers had grown to 320 million registered users, 160 million daily active users, and 800,000

new registrations per day. Id., ¶ 1735(h)–(i). Most of those users lived outside the United

States. See ECF No. 370-3 (Meta Reply to FTC SMF), ¶ 1735 (citing only 6.9 million U.S.

monthly active users by April 2013); see also ECF No. 361-1 (Pl. Docs. Vol. 57) at ECF p. 614

(FTC Investigative Hearing of Brian Acton at 147:13–15) (calling U.S. market a “hard market”

because “people were using adjacent messaging products”); Meta SMF, ¶ 776 (9% penetration of

smartphone users in United States, compared to 99% in Spain, 94% in Mexico, 90% in Brazil,

65% in India, and 49% in United Kingdom). Nonetheless, by some estimations, in September

2013 WhatsApp was second only to Facebook Messenger in OTT cross-platform mobile-

messaging services in the United States. See FTC Resp. to Meta SMF, ¶¶ 774, 776.

Meta soon began making acquisition overtures. See FTC SMF, ¶ 1831. As before, the

parties dispute its purpose in pursuing WhatsApp. Defendant points to various rationales,

including that the acquisition would make it more difficult to exclude Meta’s applications from

mobile-operating-system platforms and would also help Facebook to grow through exposure to

potential new users. See Meta SMF, ¶ 813. The FTC observes that Facebook Messenger

struggled to compete with WhatsApp, see FTC SMF, ¶¶ 1727–28, and that Meta feared

WhatsApp’s pivot into PSN services. Id., ¶¶ 1799. In any event, on February 19, 2014, Meta

announced that it would pay approximately $19 billion to purchase WhatsApp. See Meta SMF,

¶ 814 ($16 billion purchase price, plus $3 billion paid to WhatsApp employees, including

founders).

9

On March 13, Meta submitted to the FTC and DOJ an HSR Premerger Notification Form

for the acquisition. See id., ¶ 815. Just under a month later, it was publicly reported that the

FTC had closed its investigation, permitting the merger to proceed. Id., ¶ 818. Meta believes it

highly relevant that the FTC previously approved both purchases but now reverses field to

challenge them as anticompetitive.

3. Other Applications

Although Meta, Facebook, Instagram, and WhatsApp are the main characters in this

drama, other online platforms and applications play supporting roles. Two such platforms that

the FTC contends compete directly with Facebook and Instagram in the relevant market are

Snapchat and MeWe. Snapchat first became available to download in 2011 and permits users to

send and receive disappearing photos and videos to their friends, called “Snaps.” Meta SMF,

¶¶ 483–84. As of 2013, users can also post collections of Snaps as “Stories,” which are

transmitted in a broadcast format to their friends. Id., ¶ 485. Snaps and Stories are shared only

between bidirectional friends as a default. Id., ¶ 489. Users can, however, set their privacy

settings to allow any Snapchat user to view their stories, or they can choose to have a public

profile. See FTC Resp. to Meta SMF, ¶ 485. In addition to sending photos and videos, friends

can also “Chat” with one another via direct messages, see Meta SMF, ¶ 492, as well as browse

short-form videos posted primarily by non-friends on “Spotlight.” Id., ¶ 496. Snapchat

generates revenue through advertising. Id., ¶ 144(a).

MeWe, which launched for mobile use in October 2014, is a PSN service that seeks to

provide a “privacy-focused” service that does not rely on advertising or tracking user activities.

Id., ¶ 522. Instead, users choose between a free, basic version of the site or a paid subscription

that unlocks additional features, such as data storage. Id. Users create a profile with personal

10

information and can post text and images on their timelines, share other users’ content, send

messages, join and create groups, and participate in group chats. Id., ¶¶ 523–24.

Outside of the FTC’s proposed market are other platforms with which Meta claims to

compete. Launched in 2005, YouTube — acquired by Google (now Alphabet, Inc.) in 2006 —

permits users to watch and share videos either through a mobile application or on its webpage.

See Meta SMF, ¶¶ 156–57, 160, 162.

. See ECF No. 346-1 (Pl. Docs. Vol. 65) at ECF p. 2,758 (Deposition of Thomas

Kim at 347:7–8, 14–17). Another service, X — which launched in 2006 as “Twitter” and

adopted its new name in July 2023 — permits users to upload posts containing text, images,

videos, or links to their followers or to the public at large on a “Feed,” where they can be “liked,”

reposted, or replied to. See Meta SMF, ¶¶ 272–73, 283, 288–91. Users can choose to allow only

their followers to view their posts, and they can use either their real name or a pseudonym. Id., ¶

298; see also FTC Resp. to Meta SMF, ¶ 288 (posts are public “by default”). Finally, and

relatively new to the scene, TikTok was launched in the United States in 2018 primarily as a

platform for users to share short-form video or photo content, also presented in various “feeds.”

See Meta SMF, ¶¶ 204–05, 211–213, 218. Account holders can create content and interact with

others’ content by liking the videos or commenting on them, id., ¶¶ 230–31, become “Friends”

with another user if they mutually follow one another, id., ¶ 223, and send in-app messages to

other users via the “Inbox.” Id., ¶ 237. Accounts can be either public or private; anyone can

follow and watch the content of users with public accounts. Id., ¶ 236. Like virtually all other

digital platforms, YouTube, X, and TikTok largely offer their services to users for free and

generate revenue through advertising. Id., ¶¶ 144(c)–(d), 159, 274, 278, 288.

11

The Court also takes judicial notice of the Protecting Americans from Foreign Adversary

Controlled Applications Act, Pub. L. No. 118-50, 138 Stat. 895, 955–60 (2024), which requires

TikTok to be divested from its Chinese parent company, ByteDance, or else face an effective ban

within the United States. See Earle v. District of Columbia, 707 F.3d 299, 308 n.10 (D.C. Cir.

2012) (noting that courts “may take judicial notice of statutes”). The constitutionality of that

statute is currently before this Circuit. See TikTok, Inc. v. Garland, No. 24-1113 (D.C. Cir.

2024) (oral argument heard Sept. 16, 2024).

B. Procedural History

The FTC filed this action on December 9, 2020, alleging one count of monopoly

maintenance under Section 2 of the Sherman Act. See ECF No. 3 (Redacted Compl.), ¶¶ 169–

74. In particular, Plaintiff alleged that Meta had abused its dominant position in the market for

PSN services by (a) acquiring Instagram and WhatsApp, which it viewed as potential threats to

its monopoly, and (b) adopting and enforcing several anticompetitive policies that blocked rivals

from interconnecting their product with Facebook (the “interoperability allegations”). Id., ¶¶ 23–

26, 71–73. The Commission brought suit under Section 13(b) of the FTC Act, which authorizes

it to seek an injunction against an entity that “is violating” or “is about to violate” the antitrust

laws, including Section 2. See 15 U.S.C. § 53(b). It asked for an injunction to prevent such

conduct in the future as well as an order mandating, among other relief, divestiture of Instagram

and WhatsApp. See Redacted Compl. at 51.

As the Court has noted in previous Opinions, although the suit was initially assigned to

Judge Christopher R. Cooper of this district, he subsequently re-assigned it to this Court, which

was handling an earlier-filed and related State case against Meta. See Facebook I, 560 F. Supp.

3d at 11; see also Minute Order of Jan. 12, 2021. Facebook moved to dismiss both cases. The

12

Court granted dismissal of the States’ entire case, finding that the acquisition-based allegations

were barred by laches and that the interoperability allegations both failed to state a claim and

came too late to permit equitable relief. See New York v. Facebook, Inc., 549 F. Supp. 3d 6, 24,

34, 49 (D.D.C. 2021), aff’d sub nom. New York v. Meta Platforms, Inc., 66 F.4th 288 (D.C. Cir.

2023). As for the FTC’s suit, the Court held that the Commission had not adequately alleged

that Meta possessed monopoly power in the relevant market, resting instead on a “naked

allegation” that the company had a dominant position. Facebook I, 560 F. Supp. 3d at 4.

Acknowledging that such defect “could conceivably be overcome” by an Amended Complaint,

however, the Court dismissed only the Complaint, not the entire case, “leaving the agency the

chance to replead if it believes it can successfully remedy the infirmities described” in the

Court’s Opinion. Id. at 4, 11. The Court noted, however, that it remained unconvinced that the

interoperability allegations could be a basis for injunctive relief. Id. at 20, 28, 30.

Eager for a rematch, the FTC accepted the Court’s invitation. In August 2021, the

agency filed an Amended Complaint once more alleging unlawful monopoly maintenance under

Section 2 of the Sherman Act and bringing suit under Section 13(b) of the FTC Act. See ECF

No. 76 (Amended Complaint filed under seal); ECF No. 82 (Redacted Am. Compl.), ¶¶ 230–43.

This new-and-improved Complaint, however, now contained two counts: Count I encompassed

only the acquisition-based allegations, while Count II incorporated those allegations while

appending the interoperability allegations. See Redacted Am. Compl., ¶¶ 232, 238. As before,

the FTC sought an injunction against any such conduct in the future and divestiture of the

Instagram and WhatsApp acquisitions. Id. at 79.

As the reader has by now no doubt surmised, Plaintiff’s second effort fared better than its

first. This time around, the Court found that the Commission had plausibly alleged the elements

13

of monopoly power — that is, it had plausibly alleged the existence of a relevant antitrust market

of PSN services, that Meta possessed a dominant market share of that market, and that

significant barriers to entry made its share durable. See Facebook II, 581 F. Supp. 3d at 43–52.

As for its allegations of anticompetitive conduct, the FTC batted .500. While the Court found its

Count I (acquisition-related) allegations sufficient to survive the motion to dismiss, it remained

unimpressed with its Count II allegations regarding interoperability. Id. at 53–59. Because

Count II also encompassed the agency’s acquisition allegations, however — and a court

evaluating a Rule 12(b)(6) motion to dismiss cannot “parse” the viable from the unviable in the

same claim — the Court did not dismiss the count. Id. at 59–60. Rather, it indicated that the

interoperability arguments could be “sliced out at summary judgment,” id. at 60, and it barred

discovery on such policies. Id. at 61.

After over two years of discovery on the remaining allegations, Meta has now moved for

summary judgment. See ECF No. 325-1 (Meta MSJ). The FTC opposes that Motion and has

filed a Cross-Motion for Partial Summary Judgment on Meta’s asserted procompetitive

justifications for the acquisitions. See ECF No. 363-1 (Updated FTC Opp. and Cross MSJ).

Those Motions are now ripe.

II. Legal Standard

“When faced with cross-motions for summary judgment, the court must review each

motion separately on its own merits ‘to determine whether either of the parties deserves

judgment as a matter of law.’” Fam. Tr. of Mass., Inc. v. United States, 892 F. Supp. 2d 149,

154 (D.D.C. 2012) (quoting Rossignol v. Voorhaar, 316 F.3d 516, 523 (4th Cir. 2003)). If the

court determines that one party is not entitled to summary judgment, it “changes tack on the

cross motion and gives the unsuccessful movant ‘all of the favorable factual inferences that it has

14

just given to the movant’s opponent.’” Nucap Indus., Inc. v. Robert Bosch LLC, 273 F. Supp. 3d

986, 998 (N.D. Ill. 2017) (quoting Hotel 71 Mezz Lender LLC v. Nat’l Ret. Fund, 778 F.3d 593,

603 (7th Cir. 2015)). It is nonetheless still possible for a court to deny summary judgment to

both sides.

Summary judgment is appropriate “only if one of the moving parties is entitled to

judgment as a matter of law upon material facts that are not genuinely disputed.” Airlie

Foundation v. IRS, 283 F. Supp. 2d 58, 61 (D.D.C. 2003); see also Fed. R. Civ. P. 56(a);

Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 247–48 (1986); CEI Wash. Bureau, Inc. v. DOJ,

469 F.3d 126, 129 (D.C. Cir. 2006). A fact is “material” if it is capable of affecting the

substantive outcome of the litigation. Liberty Lobby, 477 U.S. at 248; Holcomb v. Powell, 433

F.3d 889, 895 (D.C. Cir. 2006). A dispute is “genuine” if the evidence is such that a reasonable

factfinder could return a verdict for the non-moving party. Liberty Lobby, 477 U.S. at 248; see

Scott v. Harris, 550 U.S. 372, 380 (2007); Holcomb, 433 F.3d at 895. “A party asserting that a

fact cannot be or is genuinely disputed must support the assertion” by “citing to particular parts

of materials in the record” or “showing that the materials cited do not establish the absence or

presence of a genuine dispute, or that an adverse party cannot produce admissible evidence to

support the fact.” Fed. R. Civ. P. 56(c)(1).

In considering a motion for summary judgment, “[t]he evidence of the non-movant is to

be believed, and all justifiable inferences are to be drawn in [its] favor.” Liberty Lobby, 477

U.S. at 255; see also Mastro v. Potomac Elec. Power Co., 447 F.3d 843, 850 (D.C. Cir. 2006);

Aka v. Wash. Hosp. Ctr., 156 F.3d 1284, 1288 (D.C. Cir. 1998) (en banc). The court must

“eschew making credibility determinations or weighing the evidence.” Czekalski v. Peters, 475

F.3d 360, 363 (D.C. Cir. 2007). To defeat summary judgment, however, an opposition must

15

consist of more than mere unsupported allegations or denials and must be supported by

affidavits, declarations, or other competent evidence, setting forth specific facts showing that

there is a genuine issue for trial. See Fed. R. Civ. P. 56(c), (e); Celotex Corp. v. Catrett, 477

U.S. 317, 324 (1986). The non-movant is required to provide evidence that would permit a

reasonable factfinder to find in its favor. Laningham v. Navy, 813 F.2d 1236, 1241 (D.C. Cir.

1987). If the non-movant’s evidence is “merely colorable” or “not significantly probative,

summary judgment may be granted.” Liberty Lobby, 477 U.S. at 249–50 (citations omitted).

III. Analysis

A violation of Section 2 of the Sherman Act via monopoly maintenance “has two

elements: ‘(1) the possession of monopoly power in the relevant market and (2) the

willful . . . maintenance of that power as distinguished from growth or development as a

consequence of superior product, business acumen, or historic accident.’” United States v.

Microsoft Corp., 253 F.3d 34, 50 (D.C. Cir. 2001) (en banc) (quoting United States v. Grinnell

Corp., 384 U.S. 563, 570–71 (1966)). As before, Meta seeks summary judgment against the

FTC on both of those prerequisites. Specifically, it argues that the Commission has not created a

genuine dispute of material fact over whether Meta possesses monopoly power in a properly

defined market, see Meta MSJ at 10–31, or has engaged in any exclusionary conduct by

acquiring Instagram and WhatsApp. See id. at 31–47. The agency’s burden on the latter should

be particularly heavy, Defendant argues, given its initial preclearance of both mergers during the

HSR process. See id. at 47–50. Plaintiff opposes Meta on all fronts and fires its own volley in

return, arguing that not only should the Court find that the agency survives Defendant’s Motion,

but it should also hold that Meta cannot raise any of its procompetitive justifications as

16

affirmative defenses to the charge of anticompetitive conduct. See FTC Opp. at 64–80. The

Court will address each of these issues in turn.

A. Monopoly Power

Start with the critical threshold inquiry: whether the FTC has put forth enough evidence

to convince a reasonable factfinder that Meta has a monopoly in the relevant market.

Traditionally, monopoly power has been understood as “the power to control prices” while

“exclud[ing] competition.” Microsoft, 253 F.3d at 51 (quoting United States v. E.I. du Pont de

Nemours & Co., 351 U.S. 377, 391 (1956)); see IIB Philip E. Areeda & Herbert Hovenkamp,

Antitrust Law, ¶ 501, at 117 (5th ed. 2021). More specifically, a firm is considered a monopolist

if it can maintain prices above the competitive level. See Microsoft, 253 F.3d at 51. When

plaintiffs can therefore provide “direct proof” of supracompetitive prices that remain protected

from erosion through competition, “the existence of monopoly power is clear.” Id. Such proof,

however, is “only rarely available.” Id. More typically, courts infer the existence of a monopoly

through three pieces of “circumstantial evidence”: the existence of (1) a relevant antitrust market,

in which the defendant holds (2) a dominant market share, protected by (3) barriers to the entry

of rivals. Id.; see also S. Pac. Commc’ns Co. v. Am. Tel. & Tel. Co., 740 F.2d 980, 1000–01

(D.C. Cir. 1984). The first two prerequisites measure a firm’s market power, while the latter

recognizes that market power becomes monopoly power “only if it is durable.” Lenox

MacLaren Surgical Corp. v. Medtronic, Inc., 762 F.3d 1114, 1124 (10th Cir. 2014).

The FTC argues that it has provided evidence sufficient to support a finding of monopoly

power through either the direct or indirect methods. See FTC Opp. at 3. As before, however, the

agency “devotes far more attention to the indirect-proof argument.” Facebook II, 581 F. Supp.

3d at 44. Because the Court concludes that the FTC has proffered sufficient evidence to get to

17

trial on that approach, it need not address whether the Commission’s direct evidence also

suffices. See id. (taking similar approach at motion-to-dismiss stage). The Court thus separately

examines the three prerequisites for monopoly power under the indirect approach.

1. Market Definition

While in some antitrust cases, the definition of the relevant market is often the simplest of

the issues (e.g., bicycles or kitchen appliances), that is not the case here. Instead, defining the

denominator of the market-power question — viz., the boundaries of the market Meta is claimed

to dominate — is no walk in the park. See Facebook I, 560 F. Supp. 3d at 14. The definition of

a relevant antitrust market is “basically a fact question dependent upon the special characteristics

of the industry involved.” Twin City Sportservice, Inc. v Charles O. Finley & Co., 676 F.2d

1291, 1299 (9th Cir. 1982). It includes “two components: the product market and the geographic

market.” Sky Angel U.S., LLC v. Nat’l Cable Satellite Corp., 947 F. Supp. 2d 88, 102 (D.D.C.

2013). A relevant geographic market is the geographic arena “where sellers compete for a group

of buyers.” IIB Areeda & Hovenkamp, ¶ 550a, at 351; see FTC v. Arch Coal, Inc., 329 F. Supp.

2d 109, 119 (D.D.C. 2004). Here, the parties agree that the relevant geographic market is the

United States. See FTC Opp. at 8 n.2.

The heart of the dispute — indeed, the question on which this case may ultimately turn at

trial — is whether the FTC has adequately defined a relevant product market. Such a market “is

a term of art in antitrust analysis,” United States v. H & R Block, Inc., 833 F. Supp. 2d 36, 50

(D.D.C. 2011), and includes “all products reasonably interchangeable by consumers for the same

purposes.” Microsoft, 253 F.3d at 52 (internal quotation marks and citation omitted). Whether

products are “reasonably interchangeable” with one another depends on two factors: functional

interchangeability — i.e., “whether buyers view similar products as substitutes,” FTC v. Sysco

18

Corp., 113 F. Supp. 3d 1, 25 (D.D.C. 2015) — and cross-elasticity of demand — i.e., whether a

slight increase in the price of one product causes a considerable number of customers to switch

to a different product. See Arch Coal, 329 F. Supp. 2d at 120. Although cross-elasticity of

demand typically turns on prices, it also depends on other intangible factors, such as whether

“the switch can be accomplished without the consumer incurring undue expense or

inconvenience.” Sysco, 113 F. Supp. 3d at 26. “In other words, courts look at whether two

products can be used for the same purpose, and, if so, whether and to what extent purchasers are

willing to substitute one for the other.” H & R Block, 833 F. Supp. 2d at 51 (internal quotation

marks and citation omitted).

This question of the relevant product market is pivotal, at least when proving monopoly

power via the indirect method. The FTC continues to argue that the market here comprises PSN

services in the United States, which it defines as online services that possess the “core

functionality” of “maintaining relationships and sharing with friends and family in a shared

space.” FTC SMF, ¶¶ 1007–08; compare Redacted Am. Compl., ¶¶ 165–80. Meta rejoins that

no such product market exists. See Meta MSJ at 10–21; ECF No. 370-1 (Meta Reply) at 4–28.

To make out its case at this juncture, the Commission must put forth evidence of such a market

sufficient to support a verdict against Meta at trial.

That evidence typically comes in one — or both — of two forms. The first, largely

qualitative, examines “such practical indicia as industry or public recognition of the submarket as

a separate economic entity, the product’s peculiar characteristics and uses, unique production

facilities, distinct customers, distinct prices, sensitivity to price changes, and specialized

vendors.” Brown Shoe Co. v. United States, 370 U.S. 294, 325 (1962). Because these Brown

Shoe factors are simply “evidentiary proxies for direct proof” of substitutability and cross-

19

elasticity of demand, see Rothery Storage & Van Co. v. Atlas Van Lines, Inc., 792 F.2d 210,

218–19 (D.C. Cir. 1986), “submarkets can exist even if only some of these factors are present.”

FTC v. Staples, Inc., 970 F. Supp. 1066, 1075 (D.D.C. 1997). The second, usually quantitative

approach involves examining expert testimony in the field of economics. See Sysco, 113 F.

Supp. 3d at 33. Such testimony very often includes a “hypothetical monopolist test” (HMT),

which “asks whether a hypothetical profit-maximizing firm . . . that was the only present and

future seller of a group of products . . . likely would undertake at least a small but significant

non-transitory increase in price (‘SSNIP’) or other worsening of terms (‘SSNIPT’) for at least

one product in the group.” DOJ & FTC Merger Guidelines § 4.3.A (2023) (“2023 Merger

Guidelines”); see FTC v. Whole Foods Mkt., Inc., 548 F.3d 1028, 1038 (D.C. Cir. 2008) (Brown,

J.) (writing for court, although other judge in majority concurred only in judgment) (describing

this test); H & R Block, 833 F. Supp. 2d at 51–52 (also describing test). As both types of

evidence are in play, the Court will examine each in turn.

Before doing so, however, it pauses to emphasize a few principles that will guide its

evaluation of the record. First, “the mere fact that a firm may be termed a competitor in the

overall marketplace does not necessarily require that it be included in the relevant product

market for antitrust purposes.” Staples, 970 F. Supp. at 1075; see Whole Foods, 548 F.3d at

1040 (Brown, J.) (similar). Rather, “[w]ithin a general product market, ‘well-defined submarkets

may exist which, in themselves, constitute product markets for antitrust purposes.” Hicks v.

PGA Tour, Inc., 897 F.3d 1109, 1121 (9th Cir. 2018) (quoting Brown Shoe, 370 U.S. at 325).

That means, as relevant here, that competition between Meta and other social-media firms for

users’ time and attention does not necessarily make them part of the same relevant product

market. See Sysco, 113 F. Supp. 3d at 26.

20

Second, and related, “market definition is guided by the ‘narrowest market’ principle.”

Id. In other words, “[a] relevant market cannot meaningfully encompass [an] infinite range [of

products]. The circle must be drawn narrowly to exclude any other product to which, within

reasonable variations in price, only a limited number of buyers will turn.” Times–Picayune

Publ’g Co. v. United States, 345 U.S. 594, 612 n.31 (1953). The relevant product market is “the

most narrowly-defined product or group of products sold by the merging firms” that is supported

by the pertinent evidence. Arch Coal, 329 F. Supp. 2d at 120 (emphasis added).

Finally, the determination of the relevant product market must ultimately “take into

account the economic and commercial realities” of the social-media landscape. FTC v. Cardinal

Health, Inc., 12 F. Supp. 2d 34, 46 (D.D.C. 1998); see also Grinnell, 384 U.S. at 572 (product

market should “reflect[] commercial realities”). Most importantly here, the proposed PSN-

services market — and, for that matter, most of the broader market of social-media services — is

characterized by the absence of prices on the user side. See Facebook II, 581 F. Supp. 3d at 55.

Recall that Meta and many of its competitors do not charge for use, preferring instead to capture

advertising revenues. That makes this case an awkward fit with many of this nation’s antitrust

precedents, which have confronted anticompetitive practices mostly in markets subject to more

traditional price and output effects. While that unusual aspect of this suit does not relieve

Plaintiff of its burden to put forth evidence sufficient to support a relevant product market at

trial, it does call for greater flexibility in evaluating that evidence, particularly at the summary-

judgment stage. See, e.g., United States v. Google LLC, --- F. Supp. 3d ----, Nos. 20-3010 & 20-

3715, 2024 WL 3647498, at *68–69 (D.D.C. Aug. 5, 2024) (acknowledging similar market

features and adjusting analysis accordingly).

21

With those principles in mind, the Court will now assess the evidence in the light most

favorable to the FTC. The take-home message is this: although Plaintiff may have a tall task

awaiting it at trial, it has provided enough evidence of a relevant product market comprising PSN

services to preclude judgment against it as a matter of law. The Court starts with the Brown

Shoe factors — which, though “old school antitrust law,” nonetheless “bind the court.” Id. at

*67 (cleaned up).

a. Brown Shoe Factors

Consistent with how other courts have deployed the Brown Shoe factors, see, e.g., FTC v.

IVQVIA Holdings Inc., 710 F. Supp. 3d 329, 355 (S.D.N.Y. 2024), the Court will analyze them

with respect to the reasonably interchangeable substitutes proposed by Meta — i.e., the full

universe of social-media platforms and applications. See Meta MSJ at 10–11. The factors,

again, are: (i) the product’s peculiar characteristics and uses, (ii) industry or public recognition of

the submarket as a separate economic entity, (iii) unique production facilities, (iv) distinct

customers, distinct prices, and sensitivity to price changes, and (v) specialized vendors. See

Brown Shoe, 370 U.S. at 325. As neither of the parties addresses the last, see FTC Opp. at 16

n.3, the Court will also skip it.

i. Peculiar Characteristics and Uses

The Court begins with the “peculiar characteristics and uses” supposedly unique to

products in Plaintiff’s proposed market. As for characteristics, the FTC focuses on three: (1) a

“social graph” focused on friends-and-family sharing, coupled with (2) tools to foster building

network connections with those friends and family, in (3) a shared social space. See id. at 13–

14.

22

Starting with the first, the Commission has provided evidence that Facebook, Instagram,

Snapchat, MeWe, and other now-defunct products in its proposed market differentiate

themselves by constructing a social graph focused on friends and family sharing. A social graph,

according to the agency, “is the set of connections that users make among each other within a

social network.” FTC SMF, ¶ 1088. So stated, it is a feature that is apparently common to many

social-media products, as Meta is quick to point out. See Meta MSJ at 19; Meta Reply at 24–25;

see also, e.g., FTC SMF, ¶ 1089; Meta SMF, ¶¶ 209–10 (TikTok), 280 (X), 401 (LinkedIn). But

the Commission’s evidence — which includes depositions of executives at many of these firms

and statements in internal documents — suggests that PSN services like Facebook and Instagram

distinguish themselves by social graphs specifically “built around friends and family

connections.” FTC Opp. at 14; see FTC SMF, ¶ 1090. That is, while many online platforms

create networks of various connections, only some of them create a network focused on personal

relationships, as opposed to professional connections (LinkedIn), see FTC SMF, ¶ 1139, or

various other interests (YouTube, TikTok, X). See id., ¶¶ 1182, 1208, 1228. Meta’s products

are self-avowedly of the former type. See, e.g., id., ¶¶ 1090(a)(i), (b)(ii) (internal document

describing both Facebook and Instagram as products offering “one-to-many sharing bounded by

a friend graph”); cf. id., ¶ 1139(c) (different internal document describing LinkedIn’s social

graph as “orthogonal” to Facebook’s because it focused on “coworkers” instead of “friends and

family”). Indeed, an internal Meta presentation referred to

Id., ¶ 1090(a)(iii). A strange term to use if the feature is,

as Meta claims, no different from what is on offer at every other social-media company. See

Meta MSJ at 19.

23

The salience of these differences is widely recognized, including by other companies.

Pinterest (a social-media company that permits users to “pin” various “interests” to their

“boards,” see Meta SMF, ¶¶ 330–40), has submitted in public filings that “social networks can

be distinguished based on use cases and target user base,” with “‘[p]ersonal’ versus

‘professional’” being “one axis of differentiation.” FTC SMF, ¶ 1108(i). Adam Presser — the

head of operations for TikTok, one of the largest competitors that Meta seeks to include in any

relevant product market — testified that such a social graph was a “core differentiated feature”

of services like Instagram and Facebook, in distinction to TikTok. Id., ¶ 1090(b)(iii); see also

id., ¶ 1089(b) (TikTok executive explaining that TikTok’s graph is a “content graph” that “relies

on our interests” as opposed to “relationships between friends and family”); id., ¶ 1208 (other

TikTok executives testifying similarly).

. Id., ¶ 1090(a)(v).

So, too, executives at YouTube repeatedly testified that their service does not use, and did not

plan to leverage, a social graph built on friends-and-family connections. Id., ¶ 1182.

The FTC has also submitted evidence that PSN services deploy certain other design tools

or norms designed to leverage their friends-and-family-focused social graphs. For instance, they

suggest new “friends” or “follows” based on friends in common and address-book information.

See id., ¶¶ 1092–96. They also emphasize the use of real identities, which facilitates the core

purpose of enabling users to connect with their actual friends and family. See id., ¶¶ 1092(b),

1094(b). Other services that Defendant claims as competitors, conversely, do not have such a

norm or, indeed, have the opposite norm: one of anonymity or pseudonymity. See, e.g.,

¶¶ 1108(k) (Reddit wants users to be able to share personal information “while maintaining their

24

privacy” and does not require users to share their real identity); (

); 1231 (pseudonymity common on X). Such evidence

tends to support the FTC’s contention that PSN services distinguish themselves through a suite

of characteristics — some, admittedly, rather subtle — that are peculiar to platforms seeking to

enable friends-and-family sharing.

Consistent with the above, Plaintiff has also presented evidence that friends-and-family

sharing is a “peculiar . . . use[]” of PSN services. Brown Shoe, 370 U.S. at 325. The FTC’s

expert Clifford Lampe — a professor at the University of Michigan School of Information, an

acknowledged expert in human-computer interactions and previously a close advisor to Meta

— submitted a report summarizing academic research on computer-mediated communications

systems. See ECF No. 331-6 (Pl. Docs. Vol. 74) at ECF pp. 155–56 (Report of Clifford Lampe),

¶¶ 1–7. He concluded that “only a subset of computer-mediated technologies” — which he

termed “Personal Social Network Sites” — “combine design with user motivations and norms

centered in efficient personal relationship maintenance across an individual’s social network,

including their strong and weak ties.” Id., ¶¶ 9, 13. That conclusion followed from his analysis

of “three main factors”: the “technical features” of a social-networking site, the “motivation” of

the people using the site, and the “norms of use for that platform that people have evolved over

time.” Id., ¶ 79. PSN services, according to Lampe, are “platforms whose design, motivations,

and norms enable masspersonal, low cost maintenance of broad personal relationship networks

that are comprised of real life connections.” Id.; see also id., ¶¶ 81–99 (outlining specific design

features, such as emphasis on real IDs). On the basis of that complex interaction among design,

motivations, and norms, Lampe concluded that Facebook, Instagram, and Snap were examples of

25

PSN services, but X, TikTok, YouTube, and LinkedIn were not. Id., ¶¶ 14–23; see also FTC

SMF, ¶¶ 895–901 (summarizing findings).

Meta itself appears to recognize the power of such norms in differentiating its products as

suited for friends-and-family sharing. Take just a few examples. During his investigational

hearing before the FTC, Zuckerberg repeatedly acknowledged that “the use cases that [Meta has]

focused on the most over time are around helping you connect with . . . your friends and family,”

as opposed to “primarily being around interests.” Zuckerberg Hr. at 184:12–15. He also

testified that “the culture of the apps and which features get emphasized can lead to . . . different

use cases,” and that “Facebook has always kind of emphasized this semi-private community

space,” somewhere between the “fully public” YouTube and the “one-on-one, private network.”

Id. at 181:23–25, 183:14–18. Indeed, the former head of the Facebook application stated in an

email that, although both Facebook and YouTube offer video products, they “serve nearly

disjoint use cases.” FTC SMF, ¶ 1010(h). The company’s internal documents have also noted

that some applications’ norm of anonymity make them a bad fit for friends-and-family sharing,

see id., ¶ 1082(c), while other emails acknowledge that Instagram and TikTok are “hired” by

users for “very different jobs,” as the latter product “is not hired for sharing with friends and

family.” Id., ¶ 1010(c).

Other actors also acknowledge that friends-and-family sharing is a peculiar use of PSN

services. See, e.g., id., ¶ 1084(h) (

).

Pl. Docs. Vol. 65 at ECF p. 876 (Deposition of Keith Coleman, Vol. 2, at 434:9–

21) (also describing that as Facebook’s “core job”). Such use cases, he explained, delimit a

26

company’s competition: X therefore “didn’t talk a lot about competing for attention because you

can’t do much with that . . . . I mean, there’s so many things that grab your attention. . . . [W]hen

I think of like what competition is useful to me in developing the product, . . . it’s competition or

alternatives that serve the same purpose.” Pl. Docs. Vol. 65 at ECF pp. 469–70 (Deposition of

Keith Coleman, Vol. 1, at 88:15–89:5) (emphasis added). An internal presentation

similarly noted that the “professional purpose of [LinkedIn] creates norms and expectation[s]

about the responsibilities and conduct of both LinkedIn and its members.” FTC SMF,

¶ 1082(d)(ii); see also id., ¶ 1084(i) (

).

. See id., ¶ 1154(b)–(c).

Id., ¶ 1154(e).

Finally, Plaintiff has also submitted data that, while far from a slam dunk, nonetheless

support this use-case differentiation. For instance, Meta records a metric that it calls

“meaningful social interactions” (MSI) to “capture the concept of active engagement with friend

and family (and group) content” on Facebook. Id., ¶ 1081(a). From 2017–2020, MSI activity

data saw “significant peaks . . . in time periods that tightly coincide with seven holidays

“ordinarily associated with spending time with friends and family.” Id., ¶ 1081(b).

. Id., ¶ 1081(d).

27

These data suggest that, consistent with the FTC’s thesis, these platforms serve separate core

uses and see differential demand based on changing patterns of user needs.

ii. Industry or Public Recognition

The Commission has also put forward evidence of industry and public recognition of a

distinct market of PSN services typified by the characteristics and uses outlined above. Such

evidence “matters” for distinguishing product markets “because we assume that economic actors

usually have accurate perceptions of economic realities.” Rothery Storage, 792 F.2d at 218 n.4.

In particular here, statements both from industry executives and in ordinary-course

documents tend to show that market participants understand their products to serve a distinct

demand for friends-and-family sharing. See Sysco, 113 F. Supp. 3d at 27–28, 30 (relying on

executives’ statements to help determine relevant product market); H & R Block, 833 F. Supp.

2d at 52 (“When determining the relevant product market, courts often pay close attention to the

defendants’ ordinary course of business documents.”). For instance, Plaintiff points to repeated

statements by Meta executives — including Zuckerberg — that fostering connections with

friends and family remains the primary use or focus for Facebook, see FTC SMF, ¶ 1320, and

that they understand Instagram to target the same core use, see id., ¶ 1321, even as both services

expanded to other use cases. Zuckerberg even went so far as to say that, despite the

“explo[sion]” of public content on Facebook in recent years, Meta would still seek to “put

friends and family at the core of the experience.” Id., ¶ 1320(c). Likewise, Sheryl Sandberg,

Meta’s former COO, testified before the FTC that staying connected with friends and family is

“a core part” of Facebook and “the majority of what people do” on the platform. See Pl. Docs.

Vol. 58 at ECF p. 110 (FTC Investigational Hearing of Sheryl Sandberg at 44:10–14). And

Adam Mosseri, the head of Instagram within Meta, agreed that “connect[ing] with . . . friends

28

and family” is a “core use case” of Instagram, that it is one of the “distinct jobs” that the

application is “hired for,” and that Instagram is “uniquely good” at integrating it with another of

the platform’s primary uses (to “entertain”). See ECF No. 355-1 (Pl. Docs. Vol. 61) at ECF

pp. 456–57 (Deposition of Adam Mosseri at 129:23–131:12).

Such statements align with the companies’ internal assessments. Internal presentations

and emails at Meta refer to Facebook’s and Instagram’s

. See FTC SMF, ¶¶ 1057, 1061; see also ECF No. 331-8 (Pl. Docs. Vol. 76) at ECF

pp. 9–11 (Expert Rebuttal Report of Clifford Lampe, ¶¶ 24–25) (collecting statements). For

example, early brand-positioning documents placed Facebook “closest” to MySpace (which

Meta viewed as “offer[ing] users a community where they can share . . . with a growing network

of mutual friends”) and further from Google and YouTube. See FTC SMF, ¶ 1070. A more

recent Meta analysis similarly concluded:

Id., ¶ 1323(b).

Internal presentations at Snapchat similarly convey its understanding that its value

proposition is “being the place to connect to close groups of friends,” while its “Stories” product

was used to build “relationship value” (as opposed to “entertainment value”). Id., ¶ 1065; see

also Lampe Rebuttal Report, ¶ 26 (collecting statements about other platforms). Likewise,

TikTok’s former Director of UX Research testified that during his time at TikTok he

“constantly” saw in various surveys and forms of user research that “Instagram and

Facebook . . . would be apps that people would primarily use to connect with people that they

already know, in some way or another, versus TikTok [which] was a little bit more random than

29

that.” ECF No. 343-1 (Pl. Docs. Vol. 66) at ECF p. 1,360–61 (Deposition of Eric Morrison at

192:14–193:9).

Evidence also shows that public expectations match these internal assessments.

Extensive user surveys conducted by Meta itself march to the same tune: users see Facebook,

Instagram, and Snapchat as the best applications for keeping up with family and friends and

sharing content with those groups. See FTC SMF, ¶ 1075. For instance, user research

conducted by Meta in 2017 on Facebook, X, Instagram, Snapchat, WhatsApp, and YouTube

found that Facebook “was seen as the best app to keep up with family/friends who are far away,”

and Instagram “was seen as the best app for seeing interesting photos/videos shared by friends.”

Id., ¶ 1075(a). A 2018 survey conducted by Meta also concluded that “Facebook’s core value

proposition is robustly anchored in ‘keeping up with friends and family,’” id., ¶ 1075(d), while a

different survey conducted the same year found that

Id., ¶ 1075(c). Meta’s survey results from 2019 through 2022

are of a piece. See, e.g., Pl. Docs. Vol. 74 at ECF pp. 483–90 (Expert Report of Michal

Malkiewicz, ¶¶ 121–37) (summarizing results from numerous Meta surveys between 2019 to

2022). TikTok’s research and surveys likewise showed that staying connected to friends and

family was the reason individuals used Facebook, Instagram, and Snapchat. See FTC SMF,

¶ 1075(f)–(g).

Plaintiff has, in addition, submitted the expert report of Michal Malkiewicz, who

conducted a consumer web survey of 3,228 respondents showing that users selected “to keep up

with [their] friends’ and family’s lives in one place” more than any other reason for using

Facebook, Instagram, Snapchat, and MeWe, yet did not commonly select that reason for other

services like X, LinkedIn, TikTok, and YouTube. See Malkiewicz Report, ¶¶ 11, 33. For

30

instance, 60.1% of respondents indicated that they used Facebook for that purpose. Id., ¶ 66.

While the numbers are not quite as impressive for Instagram, Snapchat, and MeWe — where

36.7%, 34.1%, and 22.9%, respectively, said they turned to those platforms to keep up with

friends and family — that reason was still selected more than any other. Id., ¶¶ 66, 78.

Meanwhile, only 4.5% of respondents picked friends-and-family sharing as their primary reason

for using TikTok, 2.8% for YouTube, 4.6% for LinkedIn, and 5.5% for X. Id., ¶ 67. Such

numbers shore up Plaintiff’s contention that a publicly recognized focus on friends-and-family

sharing differentiates products in its proposed market from others that Meta seeks to include.

iii. Unique Production Facilities

It is somewhat unclear what, if any, “unique production facilities” are relevant to

Plaintiff’s case. As our Circuit has explained, courts look to this factor as a proxy for “the ability

of other producers to shift resources” to make the product in the proposed market, should prices

rise or output be restricted. Rothery Storage, 792 F.2d at 218 n.4. There is little evidence,

however, that TikTok, YouTube, or other such platforms lack the technical know-how to offer

PSN services. Rather, the FTC argues that such services require “the presence of a network of

users producing friends and family content to share with their connections,” FTC SMF, ¶ 1119,

and that recreating that network from scratch is difficult without significant time and expense.

See id., ¶¶ 1120–21. So conceived, this factor overlaps substantially with Plaintiff’s argument

for high entry barriers, so the Court will address it more fully then.

iv. Distinct Customers and Prices, Sensitivity to Price

Changes

Finally, the Commission has presented evidence that Meta can price discriminate within

its PSN-services market. See Whole Foods, 548 F.3d at 1038–39 (explaining that monopolies

can extract monopoly profits from “captive” consumers through price discrimination); IIB

31

Areeda & Hovenkamp, ¶ 517a, at 155, 157 (same). Here, the available evidence of such

discrimination is not straightforward, as users access Meta’s products free of charge. Even with

zero-price products, however, a monopolist can increase the “quality-adjusted price” by

degrading the quality of the product while keeping prices constant (at zero). See FTC SMF,

¶ 1435; Google, 2024 WL 3647498, at *126. The FTC argues that consumers of Meta’s

products generally “pay” by watching advertisements and providing their personal

information. See id., ¶ 1437. An increase in the quality-adjusted price of Meta’s products thus

might be shown through an increased “ad load” — or “the percentage of advertisements served

to users as a fraction of total items shown” — a decrease in privacy, or other indications of

declining quality. Id., ¶¶ 1440–42; see also id., ¶¶ 1478–83 (likening ad load to “tax” on user

experience and drag on user engagement). Meta could raise the quality-adjusted price,

moreover, only for “core” users whose demand is relatively inelastic, which signals that they

have few other substitutes. See Whole Foods, 548 F.3d at 1037–38.

The Commission contends that users who come to Meta’s products for friends-and-

family sharing (“core users”) do, in fact, face higher quality-adjusted prices. It points to two key

pieces of evidence. First, multiple internal Meta surveys showed that

. See id., ¶ 1503 (collating results from various user surveys). For instance, a 2020

internal analysis of

. Id., ¶ 1503(b). Another, more recent survey

found that

32

. Id., ¶ 1503(h). And things appear to have . See, e.g., id.,

¶¶ 1500 ( ), 1502

(Meta executives fretting over underinvestment in friends-and-family sharing over time). Such

evidence suggests Meta has raised the quality-adjusted price for users who come to it for friends-

and-family sharing.

Second, and more directly, there is evidence both that Meta raises the ad load for users

whose demand is more inelastic, and that such users overlap with those who tend to use Meta’s

products for friends-and-family sharing. To start, Defendant openly admits that it adjusts ad load

based on a user’s perceived sensitivity to advertisements. See id., ¶ 1529. That alone suggests

that at least some users may have an inelastic demand for Meta’s products, in that increases in

the “price” (ad load) do not translate into a measurable drop-off in demand. Cf. IIB Areeda &

Hovenkamp, ¶ 517a, at 155 (noting that “‘persistent’ and systematic price discrimination” over

“two (or more) groups of customers with different demands” can indicate market power in a

relevant product market). Plaintiff points out, moreover, that those adjustments have

recognizable patterns: Meta shows fewer ads

. See FTC SMF, ¶¶ 1525 (evidence of

ad-load variations); 1528 (evidence of ad-load variation based on

). Why? Meta’s own analyses indicate that

users of Instagram and Facebook are more likely to use the applications for

friends-and-family sharing than . See id., ¶ 1530;

ECF No. 331-7 (Pl. Docs. Vol. 75) at ECF pp. 38–39 (Expert Rebuttal Report of Scott

Hemphill), ¶ 108. Increasing the quality-adjusted price for these users therefore does not

33

comparably reduce their demand for the service. Collectively, such evidence tends to support a

finding that Meta identifies users who come to its products for friends-and-family sharing and

exercises monopoly power over such users by raising the quality-adjusted price of accessing its

services.

v. Meta’s Counterarguments

Meta, of course, does not accept this evidence lying down. Its arguments, however, tend

to show — rather than refute — that genuine disputes of material fact continue to exist over the

relevant product market. They thus do not demonstrate that Meta is entitled to judgment as a

matter of law.

Defendant’s most consistent attack reduces to the same core claim: it competes against

many companies, not just those the FTC identifies as part of the PSN-services market. See, e.g.,

Meta MSJ at 10–14; Meta Reply at 4–6. This claim really consists of two different and repeated

assertions: first, that Meta competes against numerous other online offerings in a market for

users’ “time and attention,” see, e.g., Meta SMF, ¶¶ 187, 189, 196–97; Meta MSJ at 18–19, and

second, that the FTC has not shown that there is a narrower product market. See, e.g., Meta

Reply at 1–2, 4–7.

The first of these assertions is true but beside the point. The FTC readily admits that

Meta has incorporated some features (like Reels) to “bring [it] into more direct competition with

non-PSN apps such as (e.g.) YouTube and TikTok.” FTC SMF, ¶ 1325. And, no doubt, copious

evidence demonstrates both that the company views itself as competing against many other

companies for users’ time and attention, and that users seek out Meta’s products for far more

than just friends-and-family sharing. See Meta Reply at 6. As the Commission points out,

however, it is beyond cavil that companies “can operate in multiple and concentric relevant

34

markets.” FTC Opp. at 15; see, e.g., Sysco, 113 F. Supp. 3d at 26 (“[T]he mere fact that a firm

may be termed a competitor in the overall marketplace does not necessarily require that it be

included in the relevant product market for antitrust purposes.”) (quoting Staples, 970 F. Supp. at

1075); FTC v. IQVIA Holdings, Inc., 710 F. Supp. 3d 329, 368 (S.D.N.Y. 2024) (same); United

States v. Bertelsmann Se & Co. KGaA, 646 F. Supp. 3d 1, 28 (D.D.C. 2022) (same). The

narrowest-market principle, moreover, requires laser-like focus here on “the most narrowly

defined product or group of products” that is supported by the evidence. Arch Coal, 329 F.

Supp. 2d at 120. Meta’s extensive evidence of head-to-head competition outside the PSN-

services market therefore does nothing to show that its proposed “time and attention” market is

the only relevant antitrust market. Indeed, if the Court were to credit Defendant’s preferred

market, it sees no obvious limiting principle: when it comes to so broad a market as “time and

attention,” Meta competes not just with YouTube, TikTok, and X, but also with watching a

movie at a friend’s house, reading a book at the library, and playing online poker. Antitrust law

does not require consideration of such an “infinite range” of possible substitutes. Times-

Picayune, 345 U.S. at 612 n.31.

Meta’s second assertion comes nearer to the mark. Although Plaintiff need not disprove

the existence of some broader market for time and attention, it still must prove the existence of a

relevant product market composed of PSN services. See FTC Opp. at 15. To put the issue

another way: the FTC must show that PSN services are not merely different products, but that

they compose a separate product market, which means that they cannot readily be substituted for

other products outside the proposed market. See Microsoft, 253 F.3d at 52. On this point, Meta

argues that the Commission has boxed itself into a corner by claiming that virtually all time spent

on PSN apps counts for the purposes of computing Meta’s market share, while the same time

35

spent on other applications doing similar activities does not. See Meta Reply at 4–5. According

to Meta, that position “sweeps in” all sorts of activities that occur on Meta’s platforms that have

ready substitutes elsewhere, such as passively watching a video or shopping for furniture. Id. at

1. The Court will have more to say about this issue when discussing Meta’s market share, but it

addresses its relevance to the product market here.

The FTC first defends its market definition by disputing Meta’s premise. Because a

sports clip watched on Facebook is “integrated” within a social graph focused on friends-and-

family sharing, the agency argues, it is a fundamentally different experience from watching that

same clip on YouTube or TikTok. See FTC Opp. at 18. Some evidence bears out Plaintiff’s

claim. See FTC SMF, ¶¶ 1309–14. For instance, Meta’s executives appear to view unconnected

content as a complement to Meta’s core offering, creating a feedback loop in which content

discovery drives engagement with friends and family, which in turn drives more content

discovery. See, e.g., id., ¶ 1310. Intuitively, however, the argument faces an uphill climb,

especially when much of the same content is available on multiple applications outside the

FTC’s proposed market.

Perhaps sensing the difficulties of that path, the Commission tries another sortie: even

accepting that a Reel on Instagram, like a rose by another name, would smell as sweet as a

TikTok video, Plaintiff argues that Meta’s monopoly power stems from its dominance over a

“core” service that it offers alongside many other services — namely, friends-and-family

sharing. See FTC Opp. at 18–19. For this point, the Commission relies on Whole Foods, where

the D.C. Circuit held that Whole Foods competed in a relevant product market for premium,

natural, and organic supermarkets (PNOS), even though it also competed in a broader market of

conventional supermarkets. See 548 F.3d at 1038–40 (Brown, J.); see also id. at 1043–49 (Tatel,

36

J., concurring in the judgment). Recognizing that “cross-shopping [with conventional

supermarkets] is entirely consistent with the existence of a core group of PNOS customers,”

Judge Brown wrote for the court that a “core group of particularly dedicated, distinct customers,

paying distinct prices, may constitute a recognizable submarket, whether they are dedicated

because they need a complete cluster of products, because their particular circumstances dictate

that a product is the only realistic choice, or because they find a particular product uniquely

attractive.” Id. at 1039–40 (cleaned up); cf. United States v. Conn. Nat’l Bank, 418 U.S. 656,

662, 664 (1974) (despite “fierce” competition between savings banks and commercial banks for

“identical or essentially fungible services,” former can be excluded from relevant product market

because they do not offer same “cluster of services”). Because Plaintiff has proffered evidence

that only PSN applications offer “a core use and functionality that serves the demand for”

friends-and-family sharing, the Court agrees that it has met its burden to show that other

applications are not reasonable substitutes, notwithstanding the presence of “non-PSN” activity

on PSN applications. See FTC Opp. at 19. And, given the record evidence the Court has just

summarized, a reasonable factfinder could determine that Meta has such a “core use” of friends-

and-family sharing and exploits a customer base seeking out that use.

Meta’s response to this maneuver is to argue that Whole Foods can be invoked only

where there is evidence of price discrimination against that core group of customers, which

Plaintiff has failed to proffer. See Meta Reply at 7–15. The Court is not convinced. While it is

certainly true that evidence of distinct prices is highly probative for delineating product markets

— and that evidence of price discrimination against inelastic consumers can be direct evidence

of monopoly power, see FTC SMF, ¶ 1531(a) — it is not legally required. See Whole Foods,

548 F.3d at 359 (Tatel, J., concurring in the judgment) (“Brown Shoe lists ‘distinct prices’ as

37

only one of a non-exhaustive list of seven ‘practical indicia’ that may be examined to determine

whether a separate market exists.”). Even were this market a typical one, Meta’s position is

essentially that of then-Judge Kavanaugh’s dissenting opinion in Whole Foods. See id. at 367

(Kavanaugh, J., dissenting) (lack of price discrimination “all-but-dispositive”); id. at 371 (calling

Brown Shoe a “1960s-era relic”). Whatever the merits of that view, it is not the law of this

Circuit, nor have courts uniformly required such evidence. See, e.g., Google, 2024 WL

3647498, at *68–74 (finding relevant product market without looking to prices); United States v.

Bazaarvoice, 2014 WL 203966, at *22–26 (N.D. Cal. 2014) (same). Exalting price

discrimination as a legal requirement is, moreover, particularly inappropriate in a zero-price

market. See Google, 2024 WL 3647498 at *68 (declining to “consider factors related to pricing”

because “general search is a free product”).

Besides, the FTC has submitted evidence of price discrimination, as detailed above, in the

form of increased ad load and a degraded user experience for those who access Meta’s products

for friends-and-family sharing. To be sure, “the pricing evidence here is unquestionably less

compelling than the pricing evidence in some other cases,” Whole Foods, 548 F.3d at 360 (Tatel,

J., concurring in the judgment), and Meta raises serious doubts about its ultimate persuasiveness,

see Meta Reply at 11–15 (attacking Plaintiff’s price-discrimination evidence) — particularly in

light of other evidence that this so-called “core” use case “represents a small and declining share

of time spent on Facebook and Instagram.” Id. at 6. But those are factual disputes. They are not

reasons to conclude, at this stage, that Meta is entitled to judgment as a matter of law.

Finally, Meta raises various objections to the other Brown Shoe factors, mostly by

minimizing differences between products or offering competing evidence. See, e.g., Meta Reply

at 23–28; Meta Reply to FTC SMF, ¶¶ 1207–1208 (features of social graph are not “relevant to

38

substitution”). Whether the distinctions the FTC observes are akin to those between, for

instance, broadline and specialty food distributors, cf. Sysco, 113 F. Supp. at 27–37 (finding a

relevant product market), or are ultimately immaterial to the question of substitutability is, again,

a factual dispute that must be resolved at trial. The FTC has done enough to make it that far.

b. Expert Testimony

As the above is sufficient to preclude summary judgment on the existence of a relevant

product market, the Court will only briefly address the additional expert testimony to explain

why it does not alter that conclusion.

To start, as the FTC’s lead expert admits, a traditional hypothetical monopolist test is not

readily available in this case. See ECF No. 331-5 (Pl. Docs. Vol. 73) at ECF p. 76 (Expert

Report of Scott Hemphill, ¶ 179). Recall that the HMT typically requires asking whether a

“hypothetical monopolist” with control over all products in the proposed market could

“profitably raise prices on those products.” Sysco, 113 F. Supp. 3d at 33. The purpose of

imposing that “SSNIPT” is to identify potentially substitutable goods: if consumers can

substitute goods outside the proposed market, the price increase would not be profitable, and thus

those other goods must be included in the market. See id. As already noted, however, a zero-

price market does not readily lend itself to price-based analyses. Any HMT would thus require

evaluation largely of non-price data, such as “quality, service, capacity investment, choice of

product variety or features, or innovative effort.” 2023 Merger Guidelines, § 4.3.B. And while a

qualitative HMT is hardly “oxymoronic,” as Meta claims, see Meta Reply at 17, it lacks the

empirical clarity of the traditional test.

The central claim in this case, moreover, is that Meta already exercises monopoly power

over the relevant product market. It therefore makes little sense to ask if a “hypothetical”

39

monopolist would impose a SSNIPT. Rather, the key question is whether Defendant can already

do so. See IIB Areeda & Hovenkamp, ¶ 539a2, at 343 (“In contrast [to the typical merger case],

in a monopolization the given assumption is that a firm already has monopoly power . . . . The

relevant question is . . . typically whether it already has the power to charge supracompetitive

prices.”) (emphasis added). Indeed, to measure substitution in response to price increases

imposed on an already monopolized market engages in a fallacy so common it has its own name:

the Cellophane fallacy. See United States v. E.I. du Pont de Nemours & Co., 351 U.S. 377

(1956) (origin of fallacy); FTC Opp. at 9–10 (explaining it). In short, when a monopoly already

exists, any worsening of terms is apt to induce customers to substitute to products that are not, in

a competitive market, true substitutes. See Epic Games, Inc. v. Apple, Inc., 67 F.4th 946, 975

n.7 (9th Cir. 2023). Including those products in the market based on that substitution therefore

risks delineating an erroneously broad product market. See IIB Areeda & Hovenkamp ¶ 516h, at

153–54.

The upshot of all this is that conducting an HMT in this case is likely to overlap

substantially with the type of analysis that Brown Shoe already requires — viz., a largely

qualitative examination of whether any adequate substitutes exist outside the proposed product

market that have prevented Meta from imposing a “worsening of terms” (SSNIPT) on customers

of its products. The FTC’s halfhearted attempt to repackage Hemphill’s analysis as a

thoroughgoing HMT implicitly acknowledges this reality. See FTC Opp. at 21–23. It is

therefore incorrect to claim, as Meta does, that a lack of any traditional quantitative evidence of

substitution is “fatal” to Plaintiff’s case. See Meta Reply at 18. On the contrary, “[t]here is no

legal requirement that a plaintiff supply quantitative proof to define a relevant market,” Google,

40

2024 WL3647498, at *68; see also McWane, Inc. v. FTC, 783 F.3d 814, 829–30 (11th Cir. 2015)

(same), especially when alleging the monopolization of a zero-price market.

Undeterred, however, Meta offers four studies from its two lead experts purporting to be

the only quantitative evidence in the record, and it argues that their conclusions require granting

summary judgment in its favor. Not only is a motion for summary judgment not the appropriate

place to weigh competing expert testimony, see Robinson v. Pezzat, 818 F.3d 1, 3–4 (D.C. Cir.

2016), but the Court also highlights a few reasons why, viewing the evidence in the light most

favorable to the FTC, these studies do not compel a contrary result.

First, two of the studies are of questionable relevance. One measured user behavior in

India in response to a complete ban of all Chinese-owned applications, including TikTok and

WeChat, and found an uptick in usage of Facebook and Instagram. See ECF No. 331-9 (Pl.

Docs. Vol. 77) at ECF p. 483 (Expert Report of John A. List, ¶¶ 19, 21). Setting aside that the

study examines data from an entirely separate geographic market and that the ban applied to

applications like WeChat that Plaintiff believes are substitutes for Meta’s products abroad, Meta

also confuses the cross-elasticity of demand of its own products with that of TikTok, in effect

attempting an inverse HMT. See Hemphill Rebuttal Report, ¶¶ 448–49. As one of Meta’s lead

experts explains in his own academic writing, however, cross-elasticities of demand between

products are not necessarily symmetrical. See id., ¶ 449 (citing Dennis W. Carlton & Jeffrey M.

Perloff, Modern Industrial Organization 648 (2004)). Data showing substitution from TikTok to

Meta do not, therefore, support the inference that similar substitution occurs from Meta to

TikTok. The second study, meanwhile, shows that a reduction in the share of time spent on

Facebook or Instagram correlates with an increase in the share of time spent on TikTok,

YouTube, and other non-PSN applications. See List Report, ¶¶ 147–63. But the study does not

41

even purport to show that users were substituting one service for another; rather, the changes in

proportional time spent could reflect any number of non-market factors, including increased

usage of one application while holding constant usage of Meta’s apps. See Hemphill Rebuttal

Report, ¶ 471.

Second, as the FTC points out, three of the studies involved exorbitant “price” increases

imposed on the existing market — e.g., a 6-hour outage of all Meta’s apps in October 2021, see

Pl. Docs. Vol. 77 at ECF p. 815 (Expert Report of Dennis Carlton, ¶¶ 25–26); a significant

(+20%) payment for participants to reduce time spent on Meta products, see List Report, ¶¶ 52,

74, 76; FTC SMF, ¶ 1345(a); and a ban on a purported competitor in India. See List Report,

¶ 19. An HMT typically requires, however, a price increase of no more than 5%, see 2023

Merger Guidelines, § 4.3.B, because severe price increases can induce substitution outside the

relevant product market. Cf. Insight Equity v. Transitions Optical, Inc., 252 F. Supp. 3d 382,

390 (D. Del. 2017) (“[A]t a high enough price, even poor substitutes look good to the

consumer.”). By imposing steep (and, in some cases, transitory) price increases on a market that

Plaintiff argues is already monopolized, these studies both fail to implement any recognized

“quantitative” test and also risk falling prey to the aforementioned Cellophane fallacy. See FTC

Opp. 24–26.

Finally, even putting aside any methodological reservations that a reasonable factfinder

might have, the results themselves do not necessarily dance to Meta’s tune. To take just one

example, Meta makes much of the fact that in both the “outage” study and the “pricing”

experiment, users diverted more of their non-Meta time to applications outside the proposed PSN

market than they did to Snapchat, which Plaintiff includes in the market. See, e.g., Meta Reply

at 17–18. To hear Meta’s side of the evidentiary story, that means that if Snapchat is in the

42

market, all the other apps must be as well. Id. As the FTC notes, however, in both cases a large

plurality of users switched to “off-device time” — suggesting that, even when Meta’s apps

become “pricey,” many users have no real substitutes for them. See FTC SMF, ¶ 1369. Indeed,

in the pricing experiment, nearly two-thirds of users switched from Facebook to nothing at all

(39%), playing games (14%), or using their browser (10%), and a nearly equal proportion

switched from Instagram to nothing at all (29%), using their browser (18%), or Facebook (13%)

— hardly airtight evidence of widespread substitution away from Meta’s products to non-PSN

applications like TikTok and YouTube. See List Report at 47–48, figs. II-5 & II-6. So, too, in

the “switching” study, a reduction in the share of time spent on Facebook was correlated with

greater increases in shares of time spent on , , and

than on , , or even — puzzlingly —

. See List Report at 67, fig. II-13. To interpret this evidence the way Meta

urges would require accepting the improbable contention that are a closer substitute for

Facebook than Instagram is merely because more people “switched” to it.

Ultimately, the weighing of this evidence is a task for trial, where Meta’s experts may yet

win the day. But viewing the evidence in the light most favorable to the FTC — as it must at this

juncture — the Court finds that Meta’s experts have not done enough to entitle it to judgment as

a matter of law on the relevant product market.

2. Market Share

After ferociously contesting Plaintiff’s market definition, Meta offers little more than a

whimper against the Commission’s market-share calculations. To show monopoly power, the

FTC must put forward evidence that Meta holds a dominant share of the market for PSN

services. See Microsoft, 253 F.3d at 51. Here, Defendant’s market share was calculated in

43

reliance on three sources: (1) first-party data provided in discovery by market participants; (2)

data provided in discovery by Comscore and data.ai, which are data-analytics companies Meta

relies on, and Onava, which is Meta’s former in-house data source for third-party app-usage data;

and (3) data the FTC licensed from Comscore outside the discovery process. See FTC SMF,

¶ 1398. Those sources, in turn, permitted the calculation of several metrics relevant to market

share, including daily active users (DAUs), monthly active users (MAUs), and time spent on app.

See id., ¶ 1400; Facebook II, 581 F. Supp. 3d at 47 (explaining why these metrics “are

appropriate indicators” of market share).

Whichever way the data is sliced — and even accounting for a purported double-counting

issue that one of Meta’s experts identified, see Carlton Report, ¶ 96 & pp. 90–91, tbls. 14–16 —

the story is the same: Meta’s share of the relevant market has exceeded (at times greatly) the

level typically associated with monopoly power. See FTC SMF, ¶¶ 1400–01; Facebook II, 581

F. Supp. 3d at 47–48 (discussing threshold as 60% market share). According to the FTC,

between 2011 and 2022, Meta had a market share ranging between 62% and 100% in the PSN-

services market. See FTC SMF, ¶ 1397. The most recent numbers, which are the most relevant

for this claim, are : in 2022, market participants’ data show that Meta gobbled

up of time spent on PSN services, of MAUs, and of DAUs. See FTC SMF,

¶ 1401. Those numbers, even if approximations, more than meet Plaintiff’s burden to provide

evidence of Meta’s dominance in the relevant product market.

Defendant does not dispute these numbers, instead resuscitating its market-definition

complaint: the FTC’s calculations are “meaningless” because the market excludes applications

like X, YouTube, and TikTok. See Meta MSJ at 22–23; Meta Reply to FTC SMF, ¶ 1396. That

response essentially concedes that the market-share question will rise or fall in tandem with the

44

definition of the relevant product market. Indeed, Meta’s lead economic expert agreed that,

assuming that the Commission’s market definition was correct (which, of course, he disputed),

its estimates of market share were likely accurate. See ECF No. 342-1 (Pl. Docs. Vol. 67) at

ECF p. 1,005 (Deposition of Dennis Carlton at 258:6–11).

That said, Plaintiff’s have-it-both-ways market definition rears its head here, as the Court

previously indicated. The above statistics would seem to capture all time spent on Meta’s

products — i.e., not just time spent on its asserted core use case of friends-and-family sharing,

but also any time spent watching highlight reels of the Red Sox game on Instagram or buying an

antique table from a stranger on Facebook Marketplace. Perhaps anticipating this issue, the

FTC’s lead expert also calculated Meta’s market share both “based on the parts of PSN apps that

are most directly related to broadcast sharing with friends and family” (i.e., Facebook Feed,

Facebook Stories, Instagram Feed, Instagram Stories, and Snapchat Stories) and by prorating the

original market share based on various surveys capturing why users turn to PSN and non-PSN

applications. See FTC SMF, ¶¶ 1402–05. The result was , even when

excluding time spent watching videos on Facebook Feed as well as “certain Meta users who

spent little time” on Feed or Stories. Id., ¶¶ 1404–05. More specifically, under the “broadcast

sharing” approach, Meta had a market share of of time spent (excluding time spent

watching videos on Facebook Feed), of MAUs, and of DAUs. Id., ¶ 1404. Under the

proration approach, Meta’s minimum prorated market share for DAUs, MAUs, or time spent

based on any of the survey data is 69.3%. Id., ¶ 1405.

Meta raises issues with these friends-and-family-focused market-share measurements,

such as that the broadcast approach incorrectly assumes that most of what happens on Feed and

Stories is related to friends-and-family sharing while the proration approach relies on surveys

45

that did not measure actual usage data. See Meta Reply to FTC SMF, ¶¶ 1402–05. Resolving

those disputes, however, is inappropriate at the summary-judgment stage, especially without any

evidence that better methods of market-share measurement exist. See United States v. Anthem,

Inc., 236 F. Supp. 3d 171, 207 (D.D.C. 2017) (“[P]laintiffs need not present market

shares . . . with the precision of a NASA scientist. The closest available approximation often

will do.”) (internal quotation marks omitted). The FTC has put forward enough evidence that a

reasonable factfinder could conclude, based on either the all-things-considered measure or the

more targeted one, that Meta dominates the relevant product market. At this juncture, that meets

Plaintiff’s burden.

The bigger question for the Commission, however, is which route it will take at trial.

Significant unresolved questions remain over the ultimate viability of the agency’s apparent

position that all time spent on PSN applications relates to PSN services (even passive

consumption of a video or buying a table from a stranger), while none of the time spent on

applications like YouTube, TikTok, or X does so (even sharing a TikTok with a friend in-app).

This counterintuitive assertion poses more difficulties, admittedly, for metrics like “time spent on

the application” than it does for counting MAUs or DAUs; if the millions of people who open

their Facebook or Instagram applications every day do so in part because those applications meet

a specific need for friends-and-family sharing, it is plausible that all those users should “count”

as part of Meta’s MAUs or DAUs, even if some (or even most) of their time spent on Meta’s

products does not strictly relate to such sharing. And it is true that, as the Court noted in its

previous Opinion, market shares must be calculated with the best available metrics — even if

they are approximations — which can be considered “holistically.” Facebook II, 581 F. Supp.

3d at 48–49. The FTC, however, has not fully clarified the relationship of its metrics to its

46

ultimate theory of the market, preferring instead to cover all its bases. While that approach

suffices to get to trial, more precision will be expected at the main event.

3. Barriers to Entry

What remains is the final ingredient of proving monopoly power by indirect evidence:

entry barriers. See Microsoft, 253 F.3d at 51. As the Court explained above, market power

matters only if it is “durable,” Lenox MacLaren Surgical Corp, 762 F.3d at 1124, which can be

achieved if the market has high barriers to the entry of new firms. “For antitrust purposes, a

barrier to entry is best defined as any factor that permits firms already in the market to earn

returns above the competitive level while deterring outsiders from entering.” IIB Areeda &

Hovenkamp, ¶ 420a, at 77; see also Microsoft, 253 F.3d at 51 (“‘Entry barriers’ are

factors . . . that prevent new rivals from timely responding to an increase in price above the

competitive level.”). Here, the FTC argues that three such barriers exist: network effects,

switching costs, and capital costs. See FTC SMF, ¶ 1409. The Court will discuss Plaintiff’s

evidence for each in turn, before addressing Meta’s counterarguments.

“For purposes of analyzing barriers to entry, a network effect is a feature that makes a

network more valuable as the number of users increases.” IIB Areeda & Hovenkamp, ¶ 421h, at

95. In the world of PSN services, the theory is relatively simple: because users come to products

like Meta’s to connect with a network of their friends, acquaintances, and family members, they

have no real reason to join unless enough of their social world is already on the platform. See

Facebook II, 581 F. Supp. 3d at 50. That means incumbents with large and established networks

have a significant advantage over those who would seek to build out their own network.

Unsurprisingly, Plaintiff has mustered ample evidence that industry participants —

including Meta itself — understand their advantages in precisely such terms. A document

47

prepared for potential investors in 2012, for instance, described Facebook’s network as a “real

competitive moat,” one that was, at the time, holding at bay a now-defunct rival: Google+. See

FTC SMF, ¶ 1414(b)–(c). Zuckerberg’s own statements, moreover, are saturated with the

recognition that network effects protected Facebook from competition. See, e.g., id., ¶ 1416

(noting in 2008 email that “the answer to your question of why it is better to buy than compete is

network effects”). Other executives within Meta agree, see, e.g., id., ¶ 1416(c)–(d), as do those

at Meta’s competitors both within and outside of the PSN-services market. See, e.g., id., ¶ 1417

(collecting statements from Google, Snap, X, Pinterest, MeWe, and others). Even one of Meta’s

lead expert witnesses has previously written in one of his books that, “as a network scales, the

benefits to its members get bigger and bigger and bigger — often to the point of what is known

as ‘lock-in,’ which means members are almost shackled to using these platforms.” Id., ¶ 1414(e)

(quoting John A. List, The Voltage Effect 104 (2022)). Such widespread acknowledgment of

this entry barrier, including by industry participants, is highly probative evidence of its

existence. See Rothery Storage, 792 F.2d at 218 n.4.

The flip side of network effects are switching costs: the difficulty of leaving one PSN for

another after users “invest in their social graphs and accumulate a history of photos, posts, and

other content that they contribute over time.” FTC SMF, ¶ 1419. Or, to use words from one

2012 Meta document: “[Y]our Timeline on Facebook is more than a conversation. It’s the

complete story of your life on a single page . . . .” Id., ¶ 1419(a). Here, too, record evidence

reveals extensive recognition of such costs amongst industry participants, including Meta itself.

One former Meta executive, for example, wrote to Zuckerberg that the incorporation of photos

“is perhaps one of the most important ways we can make switching costs very high for users”

because it “will be very tough for a user to switch if they can’t take those photos and associated

48

data/comments with them.” Id., ¶ 1419(b). Another former executive wrote that investors in

Meta liked that social-media products in general were “sticky” because “after you have invested

hours and hours in your friend graph or interest graph or follower graph, you are less likely to

leave for a new or different service that offers similar functionality.” Id., ¶ 1419(c). Indeed,

Meta has apparently even coined its own term for these switching costs: the “ratchet effect.” Id.,

¶ 1420(c) (“[O]nce you get a user on your app[,] it’s hard to lose them.”). According to Meta’s

own analysis of the ratchet effect, it means that investing early in a market “can confer a

permanent advantage.” Id., ¶ 1420(d).

The evaluations of those outside Meta are no different. To take just one example, a 2020

Snapchat document recognized that “[t]he burden of reestablishing friend connections on a new

service means high switching cost[s] for new users and locks them into existing services.

Facebook’s social graph creates strong lock-in, as users become dependent on the service to

reach their family and friends.” Id., ¶ 1419(f); see also id., ¶ 1421 (collecting similar statements

from third parties). The Court can put it little better than Meta itself: in this market, the

“competitive moats are deep.” Id., ¶ 1420 (quoting 2012 document prepared for Meta’s IPO

roadshow).

Finally, the FTC has put forth evidence that, in part because of these entry barriers exist,

a third one appears: “high capital costs.” Id., ¶ 1422. New PSN entrants require capital to build

new applications and then to operate and scale their products without adequate offsetting

revenue. See id., ¶¶ 1422–23. But depositions and declarations in the record show a decreasing

willingness to invest in such companies. In the words of one witness,

49

ECF No. 357-1 (Pl. Docs. Vol. 59) at ECF

pp. 930–31 . Another witness,

, stated:

ECF No. 331-4 (Declaration of

), ¶ 17. And the founder of MeWe, a Meta competitor in the PSN-services market,

explained that “Facebook’s dominance made it virtually impossible for MeWe to raise money

from venture capitalists in Silicon Valley.” ECF No. 331-2 (Declaration of Mark Weinstein),

¶ 9.

In the face of this evidence, Meta offers only feeble resistance. Some of its efforts are

purely rhetorical. It swats away Plaintiff’s evidence of network effects and switching costs as

mere “buzzwords” without meaningfully contesting that such dynamics typify the PSN-services

market or the broader social-media market generally. See Meta Reply at 30–31; Meta Reply to

FTC SMF, ¶¶ 1403–21. It also asserts, without apparent support, that there is “no evidence that

money is no longer available for investment in promising companies,” Meta Reply at 30, leaving

largely undisputed Plaintiff’s evidence, from actual market participants, that capital costs remain

high and funding scarce. More substantively, Meta points to historical market entries as

evidence that the barriers are low. It claims, for instance, that the entry of Snapchat and the

growth of TikTok disprove the Commission’s thesis, and it points out that companies outside the

PSN-services market can pivot to compete with Meta within the market by developing new

features. See Meta MSJ at 23–24; Meta Reply at 31. But Snapchat entered the market almost a

decade ago, and TikTok is not in the FTC’s proposed product market. The existence of some

successes, moreover, does not refute the existence of high entry barriers. See Microsoft, 253

50

F.3d at 55 (“That some developers write applications for other operating systems is not at all

inconsistent with the finding that the applications barrier to entry discourages many from writing

for these less popular platforms.”). Barriers to entry “deter[]” the entry of rivals, see IIB Areeda

& Hovenkamp, ¶ 420a, at 77 — they do not banish competition altogether. And Plaintiff rightly

points out that many companies have attempted to pivot into the PSN-services market, only to

find the barriers too high. See FTC Opp. at 32.

On balance, then, the Court finds that the FTC has produced enough evidence of entry

barriers that a reasonable factfinder could find in its favor at trial.

***

Establishing the existence of a monopoly is no easy sail, and the FTC has chosen to

navigate into gusty winds. In addition to the complexities of a platform market that offers its

products to end users at no cost while charging for advertising, time and technological change

pose serious challenges to the market analysis here. For instance, one of the platforms Meta

claims as its chief rival may be banned entirely in the United States by the time this case goes to

trial. See TikTok, Inc., No. 24-1113. The Facebook and Instagram of 2024, moreover, are

different — in some cases dramatically so — from what they were when these acquisitions took

place, offering users content, commerce, and connection far beyond what was available in 2012

or 2014. The challenge for the Commission will be to show that the “personal social

networking” service that it highlights as Meta’s core use case is distinct enough from the many

other social-media products vying for consumers’ online attention that it still constitutes its own

relevant antitrust market. While the jury (so to speak) is still out on that question, Plaintiff has

earned the chance to make its case.

51

B. Anticompetitive Conduct

Merely possessing monopoly power, however, does not violate Section 2. See Microsoft,

253 F.3d at 51, 58. It is only when a monopolist engages in anticompetitive conduct that the

Sherman Act is properly invoked. See Verizon Commc’ns Inc v. Law Offs. of Curtis V. Trinko,

540 U.S. 398, 407 (2004) (“To safeguard the incentive to innovate, the possession of monopoly

power will not be found unlawful unless it is accompanied by an element of anticompetitive

conduct.”). Plaintiff’s Complaint lists two counts. Count I alleges that Meta engaged in such

conduct by acquiring Instagram and WhatsApp. Count II alleges that those acquisitions, together

with Meta’s interoperability policies, constitute a separate course of anticompetitive conduct. In

its prior Opinion, the Court held that the Count II interoperability allegations failed to state a

claim for injunctive relief because they concerned long-past conduct. See Facebook II, 581 F.3d

at 58–59. Because those allegations were packaged with the acquisition-related ones, however,

and the Court lacked the authority to dismiss only part of a claim, it held that the interoperability

portion of Count II could be “sliced out at summary judgment.” Id. at 60. It now does so. That

leaves on the table only the acquisitions.

In evaluating these, the Court is guided by the Circuit’s holding in Microsoft. It there

held that “[w]hether any particular act of a monopolist is exclusionary, rather than merely a form

of vigorous competition, can be difficult to discern: the means of illicit exclusion, like the means

of legitimate competition, are myriad. The challenge . . . lies in stating a general rule for

distinguishing between exclusionary acts, which reduce social welfare, and competitive acts,

which increase it.” 253 F.3d at 58. To aid in that task, the Microsoft court articulated several

principles that “emerge” from “a century of case law on monopolization under § 2.” Id. “First,

to be condemned as exclusionary, a monopolist’s act must have an ‘anticompetitive effect.’ That

52

is, it must harm the competitive process and thereby harm consumers.” Id. “Second, the

plaintiff, on whom the burden of proof of course rests, must demonstrate that the monopolist’s

conduct indeed has the requisite anticompetitive effect.” Id. at 58–59 (citations omitted).

“Third, if a plaintiff successfully establishes a prima facie case under § 2 by demonstrating

anticompetitive effect, then the monopolist may proffer a ‘procompetitive justification’ for its

conduct.” Id. at 59. And “[f]ourth, if the monopolist’s procompetitive justification stands

unrebutted, then the plaintiff must demonstrate that the anticompetitive harm of the conduct

outweighs the procompetitive benefit.” Id.

While agreeing on the applicability of the first two principles, the parties fiercely dispute

what counts here as an “anticompetitive effect.” After examining this complex issue and setting

forth the proper legal framework, the Court then analyzes the two acquisitions. It next considers

the procompetitive justifications that principles three and four raise.

1. Legal Framework

a. Anticompetitive Effects

The parties’ disagreement over the FTC’s burden to show an anticompetitive effect

merits an extended discussion. Meta argues that Plaintiff can make its prima facie case only by

offering evidence of reduced consumer welfare — primarily in the form of higher prices, lower

output, “or (maybe) reduced overall quality by some objective measure” — compared to the

but-for world in which the acquisitions never took place. See Meta Reply at 37–47; Meta MSJ at

32–36. Because prices are zero, output has expanded, and Meta’s products have improved,

Defendant asserts, the FTC cannot prove anticompetitive effects. See Meta Reply at 47–55. The

Commission, meanwhile, contends that a monopolist’s acquisition of an actual or nascent rival is

itself an anticompetitive action, regardless of any price or output evidence. In other words, the

53

exclusion or acquisition of “nascent threats” suffices to make out a prima facie Section 2

case. See FTC Opp. at 55–60. Since both Instagram and WhatsApp qualified as such rivals to

Meta when they were acquired, Plaintiff argues, its burden has been met. See id. at 40–55.

While the law is far from pellucid on this issue, the Court is persuaded that the FTC has

the better of the argument. In particular, Meta conflates the need to demonstrate

“anticompetitive effects” with the need to demonstrate an impact on consumer welfare via

measurable (and, in this case, counterfactual) price or output changes. To be sure, the Supreme

Court has held that “reduced output, increased prices, or decreased quality in the relevant

market” should be considered “[d]irect evidence of anticompetitive effects.” Ohio v. American

Express Co., 585 U.S. 529, 542 (2018). But price- or output-related evidence of competitive

harm is downstream from the anticompetitive effect itself. As Microsoft teaches, an

anticompetitive effect is anything that “has a substantial effect in protecting [a firm’s] market

power, and does so through a means other than competition on the merits.” 253 F.3d at 62. Or

consider the Supreme Court’s articulation of proscribed Section 2 conduct: “[T]he willful

acquisition or maintenance of [monopoly] power as distinguished from growth or development

as a consequence of a superior product, business acumen, or historic accident.” Grinnell, 384

U.S. at 570–71.

Both formulations teach the same lesson: while conduct with anticompetitive effects

indeed harms “the competitive process and thereby harms consumers,” Microsoft, 253 F.3d at 58

(emphases altered), the Section 2 requirement to demonstrate anticompetitive effects is not a

requirement to separately show consumer harm. Rather, plaintiffs must show that the action in

question maintained a monopoly other than through “competition on the merits” and thus

presumably harmed consumers. Indeed, while Meta repeatedly mischaracterizes Plaintiff’s

54

position as an attempt to evade its burden to demonstrate anticompetitive effects, see Meta Reply

at 38–42 (“The FTC misstates the law — it must prove anticompetitive effects.”), the FTC never

disavows that burden; it argues only that it can show anticompetitive effects without pointing to

specific price or output changes. See FTC Opp. at 55.

The caselaw bears out this important distinction. Microsoft is the key case, and each

party seeks refuge in it. See FTC Opp. at 55–56; Meta Reply at 40–42. There, the D.C. Circuit

sitting en banc confronted the charge that Microsoft had unlawfully maintained a monopoly in

the market for “Intelcompatible PC operating systems” by implementing “anticompetitive terms

in its licensing and software developer agreements.” Microsoft, 253 F.3d at 45, 47. In

identifying the anticompetitive effects of Microsoft’s actions, the court repeatedly (and

unanimously) focused on whether the actions foreclosed competition on the merits, not whether

they led to price or output changes. Microsoft had, for example, implemented license restrictions

that prevented original equipment manufacturers from encouraging the use of Netscape

Navigator — a rival browser to Microsoft’s Internet Explorer — thus protecting Microsoft’s

monopoly in the operating-systems market from the threat that such “middleware” might present

if it reached a critical mass of users. Id. at 60–61. Evaluating this conduct, the court wrote:

“Because this prohibition has a substantial effect in protecting Microsoft’s market power, and

does so through a means other than competition on the merits, it is anticompetitive.” Id. at 62.

Microsoft also took steps to “technologically bind[]” Internet Explorer to the Windows operating

system, thereby deterring installation or use of other browsers. Id. at 64–65. Again, the court

was clear about the harm: Microsoft’s conduct was “anticompetitive” because it had “the effect

of significantly reducing usage of rivals’ products and hence protecting its own operating system

monopoly” “through something other than competition on the merits.” Id. at 65. So, too, for

55

Microsoft’s exclusive-dealing arrangements with most internet-access providers, which

foreclosed to its rivals “a substantial percentage of the available opportunities for browser

distribution.” Id. at 70. Because those deals had a “significant effect in preserving” Microsoft’s

monopoly, they “demonstrated harm to competition.” Id. at 71.

Critically, Microsoft’s anticompetitive conduct was aimed at what the court called

“nascent threats” — middleware products that did not yet compete with Microsoft in the

operating-system market but posed the threat of doing so. Id. at 53–54, 79. The court went on to

articulate a “rather edentulous test” for determining that Microsoft’s exclusionary conduct

contributed to its continued monopoly: causation could be inferred “from the fact that a

defendant has engaged in anticompetitive conduct that reasonably appears capable of making a

significant contribution to maintaining monopoly power.” Id. at 79 (cleaned up). Thus, the

court wrote, the analysis came down to two key questions: first, “whether as a general matter the

exclusion of nascent threats is the type of conduct that is reasonably capable of contributing

significantly to a defendant’s continued monopoly power,” and second, “whether [the

middleware products] reasonably constituted nascent threats at the time Microsoft engaged in the

anticompetitive conduct at issue.” Id. As to the first question, the court had no doubt that such

conduct was proscribed. Id. On the second, it found the record supported that the middleware

products “showed potential as . . . threats.” Id.

Meta strains to read into Microsoft a focus on price or output effects, see Meta Reply at

40–42, but it comes up empty. Again and again, Microsoft holds that anticompetitive conduct by

a monopolist is simply conduct that maintains or expands its monopoly other than through

competition on the merits. The “exclusion” of competitors, the court held, qualifies as such

conduct. This Court sees no reason why a monopolist’s acquisition of its competitors should be

56

any different. See III Philip E. Areeda & Herbert Hovenkamp, Antitrust Law, ¶ 701b, at 228

(5th ed. 2022) (“The monopolist’s acquisition of a rival must be treated . . . clearly as

anticompetitive to the extent that it eliminates competition that might have otherwise dissipated

the monopolist’s power. As such, it is a classic strategy of monopoly maintenance.”). To hold

otherwise would imply that, had Microsoft simply acquired Netscape Navigator rather than

sought to exclude it from the market, Section 2 would have remained dormant. Meta provides no

authority — nor is the Court aware of any — that would support such a conclusion.

Indeed, it is hard to imagine an action that better fits the definition of conduct with

anticompetitive effects than a monopolist’s buying out its rivals. It has been long established, for

instance, that a horizontal merger between two competitors that “produces a firm controlling an

undue percentage share of the relevant market” is presumptively anticompetitive. United States

v. Phila. Nat’l Bank, 374 U.S. 321, 363 (1963); see also Polyplore Int’l, Inc. v. FTC, 686 F.3d

1208, 1215–16 (11th Cir. 2012) (merger between two firms to create a duopoly is subject to

Philadelphia National presumption of anticompetitive effects); III Areeda & Hovenkamp, ¶ 701a,

at 226 (“Historically and today, merging viable competitors to create a monopoly is a clear § 2

offense . . . .”). Extending that logic, horizontal acquisitions must also presumptively qualify as

anticompetitive when one of the merging firms is already a monopolist, as is the case in any

monopoly-maintenance suit. See III Areeda & Hovenkamp, ¶ 701a, at 226 (“[A] monopolist’s

acquisition of the productive assets or stock of an actual or likely potential competitor is properly

classified as anticompetitive, for it tends to augment or reinforce the monopoly by means other

than competition on the merits.”).

This view also finds support in binding precedent. In Grinnell, for instance, the Supreme

Court held that a dominant conglomerate violated Section 2 by engaging in multiple

57

anticompetitive practices, which included restrictive agreements, various pricing practices, and

acquisitions of competitors. See 384 U.S. at 576. While Meta is correct that the Court

considered evidence of increased prices as one type of anticompetitive conduct, the opinion

described each of the challenged practices — including the acquisitions— as “unlawful and

exclusionary” in their own right. Id. (noting that restrictive agreements “were one”

anticompetitive device, pricing practices “were another,” and acquisitions “were still another”);

see BRFHH Shreveport, LLC v. Willis Knighton Medical Center, 176 F. Supp. 3d 606, 619–20

(W.D. La. 2016) (endorsing this reading); see also Behrend v. Comcast Corp., 2012 WL

1231794, at *20 (E.D. Pa. Apr. 12, 2012) (permitting plaintiffs to “rely upon the [acquisitions]

themselves and evidence of [defendant’s] resultant increase in market share as proof” of Section

2 offense).

Under Microsoft, moreover, the presumption of illegality described above would appear

to apply to acquisitions of nascent as well as actual competitors. “Indeed, one of the most

effective ways for a dominant firm to maintain its monopoly position is to acquire incipient

rivals as they appear on the horizon.” Herbert Hovenkamp, Federal Antitrust Policy, § 7.2, at

314 (4th ed. 2011). “[W]hen the dominant firm in the market” merges with a nascent rival, the

merger “tends to maintain a monopoly by cutting off an avenue of future competition before it

has had a chance to develop. As a result, condemnation under § 2 is appropriate.” IV Philip E.

Areeda & Herbert Hovenkamp, Antitrust Law, ¶ 912b, at 92 (4th ed. 2016).

Meta’s rejoinder to these principles is not persuasive. It leans heavily on Rambus Inc v.

FTC, 522 F.3d 456 (D.C. Cir. 2008), and Brooke Grp. Ltd. v. Brown & Williamson Tobacco

Corp., 509 U.S. 209 (1993), to contend that the FTC must show price or output effects, but

neither case provides such support. In Rambus, a developer had misrepresented the scope of its

58

patent interests to a standard-setting body while that body was formulating the standards for

technologies covered by those interests, enabling the developer to monopolize certain markets.

See 522 F.3d at 460–62. Had the defendant fully disclosed its interests, the organization might

have either excluded its technology from the standards or demanded assurances of reasonable

license fees — but the FTC explicitly did not find that either outcome was more likely. See id. at

462–63. In light of that uncertainty, our Circuit determined, it was possible that the only “effect”

of defendant’s conduct was higher prices than would have otherwise obtained absent the

deception. See id. at 464. The court thus held that anticompetitive effects had not been shown

because “an otherwise lawful monopolist’s use of deception simply to obtain higher prices

normally has no particular tendency to exclude rivals and thus to diminish competition.” Id. at

464–65. That conclusion bolsters, rather than undermines, Plaintiff’s contention that the

exclusion of rivals (e.g., via acquisition) is itself anticompetitive conduct. Indeed, Rambus

confirms that the analysis must focus the “effect on competitive structure” rather than on

“rais[ed] prices” alone. Id. at 466.

Brooke Group (like Rambus) did not even concern a monopolist’s acquisition of a

competitor. There, the Supreme Court evaluated an alleged predatory-pricing scheme in the

domestic cigarette industry. See 509 U.S. at 212. It held only that, because the “linchpin” of the

allegation was that the defendant could raise prices along with other oligopolists — but there was

no evidence of raised prices or reduced output — anticompetitive effects had not been

shown. Id. at 232–33. Nothing in the opinion supports Meta’s contention that evidence of

reduced output or increased prices is required in the entirely separate context of a monopolist’s

acquisition of actual or nascent rivals.

59

Meta’s other citations serve only to underscore its misunderstanding of this crucial

context. Asserting that “acquisitions are not per se unlawful,” Meta Reply at 42, Defendant cites

two cases in which the acquiring firm was not a monopoly — either because it was a minor

player before the acquisitions and likely did not achieve monopoly power via the mergers,

see Eastman v. Quest Diagnostics, 724 F. App’x 556, 559 (9th Cir. 2018) (affirming dismissal

where largest acquisition required divestiture of assets, thus creating “a significant [new]

competitor,” and other acquisitions were “relatively insubstantial”), or because there were no

entry barriers, and the firm’s dominant market share post-acquisition thus quickly eroded. See

United States v. Syufy Enters., 903 F.2d 659, 664–65 (9th Cir. 1990). It is no doubt true that

“the elimination of rivalry by the joinder of rivals . . . is not, per se, an unlawful restraint of

trade,” Rothery Storage, 792 F.2d at 223, but that is not this case.

At issue here is emphatically not the merger of rivals in a competitive industry, but the

acquisition of a nascent competitor by the dominant firm in a market protected by entry barriers.

Cf. McWane, 783 F.3d at 836 (“[C]ompetitors and competition are linked, particularly in the

right market settings . . . .”); Spirit Airlines, Inc v. Nw. Airlines, Inc., 431 F.3d 917, 951 (6th Cir.

2005) (“[I]n a concentrated market with very high barriers to entry, competition will not exist

without competitors.”). When a monopolist acquires a rival, even a nascent one, a rebuttable

presumption therefore arises that the merger is anticompetitive. See IV Areeda & Hovenkamp,

¶ 912a, at 92 (acquisition by dominant firm of nascent rival “eliminates an important route by

which competition could have increased in the immediate future” and “thus bears a very strong

presumption of illegality that should rarely be defeated”).

To be clear, price and output effects remain highly probative evidence of anticompetitive

effects, especially where the same actions could plausibly be categorized as either vigorous

60

competition or illicit exclusion. And Plaintiff is free to present evidence of consumer harm, as it

has done here, that it believes is confirmation that the acquisitions were anticompetitive. See

FTC Opp. at 59–60. But the fact that anticompetitive conduct typically goes hand in glove with

price increases or reductions in output does not mean that, as a matter of law, such obvious

effects on consumer welfare must be demonstrated for a plaintiff to make out a prima facie

Section 2 case in circumstances such as those presented here. In so holding, the Court remains

especially mindful that typical price and output effects may not obtain in this market. See, e.g.,

Google, 2024 WL 3647498, at *80 (“[R]educed output is an ill-fitting indicia of monopoly

power in a market like search” because the “cost of responding to one additional query is near

zero. In such a market, a dominant firm has no incentive to restrict output to earn monopoly

profits.”); Microsoft, 253 F.3d at 79 (finding that middleware could become competitive threats

while relying on admission that “[m]arginal costs are essentially zero” and “[t]here are no

constraints on output” in relevant market).

Following Microsoft, then, the Court holds that, as a general matter, the acquisition of

actual competitors or nascent threats by a monopoly is “the type of conduct that is reasonably

capable of contributing significantly to a defendant’s continued monopoly power.” 253 F.3d at

79. This is so because such action tends to maintain monopoly by means other than competition

on the merits. The key question in this case thus becomes whether Instagram and WhatsApp

“reasonably constituted nascent threats” (or actual competitors) “at the time” Meta acquired

them. Id. If so, then the Commission has made out a prima facie case of monopolization, see

Google, 2024 WL 3647498, at *114 (“The loss of nascent competitors is a clear anticompetitive

effect.”), and the burden shifts to Meta to show that the acquisitions nonetheless have sufficient

procompetitive justifications. Defendant’s arguments concerning low prices, increased output,

61

and increased innovation are thus properly viewed, if anything, as procompetitive rebuttals to

Plaintiff’s prima facie case, rather than as evidence that the acquisitions themselves had no

anticompetitive effects. The Court stresses, moreover, that Microsoft’s toothless causation

standard applies only to the question of liability. If Plaintiff achieves a favorable verdict at trial,

it must meet a stricter standard to merit a structural remedy like divestiture. See Microsoft, 253

F.3d at 80.

b. Nascent Threats

So formulated, this standard places pressure on what it means to “reasonably constitute[]

a nascent threat.” Id. at 79. Neither party has thoroughly briefed this question. For its part,

Meta at times makes use of precedent from a parallel but different context: that of the potential-

competition doctrine, which is typically employed in cases brought under Section 7 of the

Clayton Act when an acquisition eliminates the possibility that one firm would have entered a

new market and created procompetitive effects. See, e.g., Meta Reply at 53 (citing United States

v. Marine Bancorporation, Inc., 418 U.S. 602, 639 (1974)). Although the viability of the

doctrine is a subject of considerable debate, most courts that accept it appear to hold that it

requires at least something like a “reasonable probability” that the competitor would have

entered the target market and matured into a competitive force. See FTC v. Meta Platforms, Inc.,

654 F. Supp. 3d 892, 926–27 (N.D. Cal. 2023) (discussing cases). Our Circuit, however, has not

endorsed or applied the potential-competition doctrine even in the Section 7 context, and its strict

standard is in significant tension with Microsoft itself, which never adverted to it.

The doctrine has, moreover, mostly arisen to evaluate the competitive harm that arises

when the acquiring firm chooses to merge with an existing market participant rather than to

compete in a concentrated market itself. See, e.g., Marine Bancorporation, 418 U.S. at 605;

62

United States v. Phillips Petrol. Co., 367 F. Supp. 1226, 1228–29 (C.D. Cal. 1973), aff’d sub

nom. Tidewater Oil Co. v. United States, 418 U.S. 906 (1974); Meta, 654 F. Supp. 3d at 908; cf.

FTC v. Steris Corp., 133 F. Supp. 3d 962, 966 (N.D. Ohio 2015) (applying theory to acquisition

of potential entrant rather than established participant). That is a materially different

circumstance from a monopolist’s acquisition of a nascent threat that may develop into a future

competitor. Cf. III Areeda & Hovenkamp, ¶ 701d, at 233 (while potential-competition doctrine

is “rightfully criticized for involving courts in undue speculation about likely consequences,”

“anticompetitive concerns are much less speculative” when “one of the merging firms is a

monopolist and the other is a potential entrant into the same market in which the monopolist has

power”). Here, moreover, Meta has not explicitly pushed the adoption of such a standard, and

the Court is not inclined to adopt it sua sponte as appropriate for a Section 2 case.

That leaves the Court with little other guidance. Microsoft observed only that, in the

context of that case, the excluded competitors “showed potential as . . . threats,” relying in part

on the estimation of Microsoft itself. See 253 F.3d at 79 (noting concession by counsel for

Microsoft). That formulation appears simply to reword the court’s “reasonably constituted

nascent threats” standard. Professors Areeda and Hovenkamp have advocated applying the

potential-competition doctrine to such circumstances while considerably lowering the bar and

would thus label as presumptively anticompetitive a monopolist’s “acquisition of any firm that

has the economic capabilities for entry and is a more-than-fanciful possible entrant.” III Areeda

& Hovenkamp, ¶ 701d, at 235. Neither party to this case has provided an alternative.

Bound by Microsoft, the Court will implement its laconic standard as best it can. At a

minimum, if the acquiring firm itself — a rational economic actor in possession of relevant

information about the market’s competitive dynamics, see III Areeda & Hovenkamp, ¶ 701d, at

63

233 (“The single-firm monopolist is in a position to make unilateral decisions about the risk and

impact of new entry . . . .”) — reasonably viewed the acquisition as a nascent competitor, that is

likely to be highly probative evidence that the acquired firm constituted such a threat. This

approach aligns with the accepted rule that evidence of anticompetitive intent “is relevant only to

the extent it helps us to understand the likely effect of the monopolist’s conduct.” Microsoft, 253

F.3d at 59; see Chi. Bd. of Trade v. United States, 246 U.S. 231, 238 (1918) (“[K]nowledge of

intent may help the court to interpret facts and to predict consequences.”). A monopolist’s

fearful acquisition of a nascent competitor tends to show that the “likely effect” of the

acquisition is to eliminate a threat to its dominance, and ipso facto that the acquired firm

“reasonably constituted” such a threat.

That said, objective evidence of the acquired firm’s capability or willingness to develop

into such a threat — where available — will also be probative. Threats can be overstated,

especially by firms protective of their dominance. Evidence that the acquired firm could not

reasonably have matured into a real competitor must cut heavily against a monopolist’s

subjective expectation that it would. Conversely, evidence that the acquired firm had the

capability and willingness to enter the monopolist’s market may suffice to show it was a nascent

threat, even in the absence of strong evidence that the monopolist so viewed its acquisition

target. Cf. Mercantile Tex. Corp. v. Bd. of Govs. of Fed. Rsrv. Sys., 638 F.2d 1255, 1270 (5th

Cir. Unit A 1981) (absence of subjective intent not fatal to potential-competition claim if

objective evidence supports reasonable probability of procompetitive entry). In the sections that

follow, the Court will thus focus on Meta’s own assessment of Instagram and WhatsApp,

coupled with any other objective evidence of the target firms’ capabilities or willingness to

64

compete in Defendant’s market, to form a realistic understanding of the competitive threat they

posed.

Before doing so, however, one last piece of housekeeping is in order. Meta has re-upped

its argument that a presumption of legality should be extended to its acquisitions because the

FTC previously cleared them to go forward during the Hart-Scott-Rodino process. See Meta

MSJ at 47–50; Meta Reply at 55–58. The Court previously rejected this argument, see Facebook

II, 581 F. Supp. 3d at 57, and it sees no reason to revisit that issue. As it previously explained,

the closing of the HSR process is explicitly “not to be construed as a determination that a

violation may not have occurred,” and “it would be improper to draw a merits conclusion about

the legality of the mergers on the basis of” such decisions. Id. (internal quotation marks

omitted). While unusual, the Commission’s decision to revisit mergers it previously declined to

block is entitled to full consideration on the merits.

2. Acquisitions

The Court will now assess whether Plaintiff’s evidence could convince a reasonable

factfinder that Instagram and WhatsApp either were actual competitors with Meta or “reasonably

constituted nascent threats” to Meta’s monopoly at the time of their acquisitions. Microsoft, 253

F.3d at 79. It finds that the FTC has met its burden on this front, as it has proffered evidence that

Instagram arguably qualified as an actual competitor and that WhatsApp constituted a nascent

threat.

a. Instagram

The FTC’s story of the Instagram acquisition begins with what it calls the “shift to

mobile”: a period beginning in 2010 with the launch of the iPhone 4 where the competitive

dynamics of social media shifted. See FTC Opp. at 40; FTC SMF, ¶¶ 1565–68. Designed for

65

desktop computers, Facebook had “enormous disadvantages” in adapting to the new environment

compared to “mobile first” start-ups. See FTC SMF, ¶¶ 1569–70. According to the FTC,

moreover, Facebook’s transition to mobile was not going well: users complained of the mobile

application’s poor quality, lack of speed, and propensity to crash, even as, by 2012, most of its

users were accessing Facebook on their mobile devices. See id., ¶¶ 1576–79. The application’s

photo features were a particular pain point. In 2012, for instance, Zuckerberg wrote that “when

you look at Instagram and Path and then go back to our Newsfeed, it looks like ours was built in

the stone age.” Id., ¶ 1580 (cleaned up). A few months later, he admitted that Facebook was

“behind” Instagram in developing a compelling mobile experience. Id., ¶ 1584.

Meanwhile, Instagram was pitching itself to investors as a “mobile-first network focused

on photos.” Systrom Hr. at 100:4–9. Launched in 2010, it “was designed specifically to take

advantage of the shift to mobile.” FTC SMF, ¶ 1603. In contrast to Facebook’s mobile

application, Instagram’s was “praised as exemplifying the speed, reliability, and simplicity

necessary to succeed in the mobile environment.” Id., ¶ 1603(b) (quoting Meta white paper); see

id., ¶¶ 1606(a), 1617 (noting positive user reviews and high engagement in 2011 and 2012). It

accordingly grew quickly: from 25,000 registered users on its first day to 35 million users by the

time Meta agreed to acquire the company. Id., ¶ 1608. By 2012, it was growing faster than the

Facebook application’s mobile photo-sharing capability. Id., ¶ 1612. Evidence in the record

suggests, moreover, that Instagram was eyeing expansions beyond photo sharing, including

video features and direct messaging. Id., ¶ 1618.

This growth posed competitive problems for Meta. At the time, Instagram’s founders,

Kevin Systrom and Mike Krieger, viewed their start-up as either already or imminently in

competition with Defendant. Systrom has testified that Instagram was “clearly competitive”

66

with Meta, see Systrom Hr. at 140:10–11, and that it would be “inevitable” that Zuckerberg

would “think that we were in conflict” if Instagram remained independent. See ECF No. 349-1

(Pl. Docs. Vol. 64) at ECF p. 475 (Deposition of Kevin Systrom at 37:6–8). Krieger similarly

testified that Instagram’s “continued growth posed a challenge” to Meta because, “for people

who used Instagram heavily,” it was “a primary way in which they kept up with their friends and

family potentially, and that is also a use case that Facebook has.” Pl. Docs. Vol. 57 at ECF

p. 951 (FTC Investigational Hearing of Michael Krieger at 95:14–20).

Meta, meanwhile, fretted over the challenge posed by Instagram. Zuckerberg noted in

2011 that Instagram was “kick[ing] ass,” and he worried that “they could easily add pieces of

their service that copy what we’re doing now,” which could pose “a real issue for [Meta].” FTC

SMF, ¶ 1632(a). “[F]rom a defense perspective,” a Meta engineer also told Zuckerberg,

Instagram was a “compelling” acquisition “because of the potential for someone like Apple to

use them as a foothold.” Id., ¶ 1632(c). By February 2012, Zuckerberg described Meta as “very

behind” Instagram, and the prospect of having to catch up was “really scary” — so Meta “might

want to consider paying a lot of money” to acquire it. Id., ¶ 1646(b). Four months later, his

worries only intensified. He noted that Instagram could “hurt [Meta] meaningfully” by shifting

“a core use case[]” away from Meta, and that “it’s kind of going to be tough to dislodge”

Instagram. Id., ¶¶ 1649–50.

These worries were not helped by the fact that Facebook’s own photo-sharing forays

were not seeing comparable success. A few months after Instagram launched, a Meta executive

informed Zuckerberg that the Facebook photos team was “now focused almost exclusively on a

new mobile photo app as we gawk at Instagram’s simple photo-sharing app taking off.” Id.,

¶ 1671. Meta began to develop a standalone product — initially known as “Snap,” then

67

“Facebook Camera” — to compete with Instagram, given initial difficulties with integrating

photo-sharing on the Facebook Blue application. Id., ¶¶ 1674–75; see Zuckerberg Hr. at 253:25–

254:1. But progress was slow, and Instagram was growing quickly. See FTC SMF, ¶ 1679(b)

(Meta executive asking colleagues, “[W]hy is mobile photos app development moving so

slowly? We still are getting our ass kicked by Instagram.”). In April 2012, Zuckerberg urged his

team “to crank and get [Facebook Camera] out,” or otherwise Instagram “may really be

unreachable.” Id., ¶ 1674(f).

Small surprise, then, that the case for acquiring Instagram focused heavily on neutralizing

a competitive threat. By February 2012, Zuckerberg was suggesting to colleagues that Meta

“should consider buying Instagram, even if it costs ~$500m” because “they seem to have two

things that we don’t: a really good camera and a photo-centric sharing network.” Id., ¶ 1690. On

camera features, he worried that it would take Meta “a huge amount of work” and “too long to

catch up [to Instagram], if we even will,” and the company would thus “be very behind in both

functionality and brand on how one of the core use cases of Facebook will evolve in the mobile

world, which is really scary and why we might want to consider paying a lot of money for this.”

Id. Discussing the potential acquisition with his CFO, Zuckerberg wrote that Instagram and

another potential acquisition target — Path — were “nascent but the networks are established,

the brands are already meaningful[,] and if they grow to a large scale they could be very

disruptive to us.” Id., ¶ 1693. He then wrote that his purpose in acquiring Instagram and

potentially other startups would be a combination of “neutraliz[ing] a potential competitor” and

“integrat[ing] their products with ours in order to improve” Meta’s service. Id., ¶ 1694.

He then explained his reasoning at length:

68

The basic plan would be to buy these companies and leave their

products running while over time incorporating the social dynamics

they’ve invented into our core products.

One thing that may make [neutralizing a potential competitor] more

reasonable here is that there are network effects around social

products and a finite number of different social mechanics to invent.

Once someone wins at a specific mechanic, it’s difficult for others

to supplant them without doing something different. . . .

[Integrating their products to improve Meta] is also a factor but in

reality we already know these companies’ social dynamics and will

integrate them over the next 12–24 months anyway. The integration

plan involves building their mechanics into our products rather than

directly integrating their products if that makes sense. . . .

[O]ne way of looking at this is that what we’re really buying is time.

Even if some new competitors spring[] up, buying Instagram, Path,

Foursquare, etc[.] now will give us a year or more to integrate their

dynamics before anyone can get close to their scale again. Within

that time, if we incorporate the social mechanics they were using,

those new products won’t get much traction since we’ll already have

their mechanics deployed at scale.

Id.

In light of these justifications, Zuckerberg and his team were willing to pay a pretty

penny to buy Instagram. By April 2012, he was expressing the view that Instagram could be

worth the “really expensive — probably ~$1 billion” cost because it was “pretty threatening” and

could “hurt [Meta] meaningfully.” Id., ¶ 1701. Sure enough, Meta paid $1 billion to acquire the

company. Id., ¶ 1710. That amount was double Instagram’s recent Series B valuation, id., and,

according to Instagram’s founders, an “unprecedented” amount of money. See Krieger Hr. at

158:9–13; Systrom Hr. at 199:17–25. After the acquisition, Zuckerberg nonetheless explained to

Meta employees that the buyout “was a no brainer” because Instagram had grown into a “mega-

network” that was “on the path to win” had Meta not acquired it. Id., ¶ 1635(e). Another Meta

executive explained that the “high level rationale” for the acquisition was that Instagram “had

69

escape velocity in terms of user traction (growth, engagement) and had/would become a ‘core’

social, mobile service in a strategically important area for us (photo sharing) and in a different

but important way (follower graph).” Id., ¶ 1708(b).

Evidence put forward by the FTC also suggests that Instagram was uniquely positioned to

pose this competitive threat to Meta. Other competitors at the time — namely, Path, PicPlz, and

Treehouse — failed to reach a comparable level of success. Path,

, grew slowly, reaching only 2 million users (to Instagram’s 30

million) by April 2012, even though it was . Id., ¶ 1663. It limped

along until 2018, when it was eventually shuttered. Id. Internal documents suggest that Meta

accordingly took a more dismissive attitude toward it. Id., ¶ 1664. And PicPlz and Treehouse

faded even faster. Id., ¶¶ 1665–66.

This evidence is sufficient to convince a reasonable factfinder that Instagram was already

a competitive threat to Meta at the time of its acquisition; that few other firms posed a

comparable threat; and that Meta’s acquisition of Instagram while it enjoyed monopoly power

was thus presumptively anticompetitive. Cf. Microsoft, 253 F.3d at 79. Were there doubt,

however, Plaintiff has also provided evidence of specific harms to consumers. Most relevantly,

Instagram’s ad load has grown enormously over time: by 2022, ad load on Instagram Feed was

what it was in 2015, while the increase on Instagram stories was the 2015

level. See FTC SMF, ¶ 1924. That includes a over just one year, from 2017–18,

over the objections of Systrom himself. Id., ¶¶ 1930, 1932. In addition, Meta has been able to

adjust ad loads on both Instagram and Facebook to make up for revenue shortfalls on one of the

products. See id., ¶¶ 1926–27, 1955. Because displeased users would simply switch between

70

the products, “Meta knew it would recapture ad revenues” lost on one platform from the

switching. Id., ¶ 1955.

The FTC has collected other evidence, moreover, of consumer harm, including Meta’s

explicit policy of not prioritizing Instagram’s growth so as not to cannibalize too much from

Facebook, id., ¶¶ 1942, 1957; declining user sentiment on privacy,

, id., ¶¶ 1992–94; and Meta’s

, even as some other platforms have done so. See id., ¶¶ 2011–16. Finally, Plaintiff

points out that Defendant decided to discontinue Facebook Camera — its competitive response

to Instagram — as a standalone application shortly after the acquisition. See FTC Opp. at 42;

FTC SMF, ¶ 1852 (Zuckerberg pushing to “scale back or cancel” Facebook Camera “since we’re

acquiring Instagram”).

Meta argues that the above evidence overstates Instagram’s competitive threat and

underappreciates the procompetitive benefits that flowed from the merger. At the time of the

acquisition, Defendant points out, “Instagram was a startup with just 13 employees that had

limited features, no plan to make money, and a third-party infrastructure that suffered repeated

service outages and spam attacks.” Meta MSJ at 39–40. It is thus mere “conjecture” that

Instagram would have even survived, much less thrived, had Meta not bought it. See id. at 41.

After the acquisition, according to Meta, it poured billions into developing and growing

Instagram, adding “dozens of new features” and providing “hundreds of billions of dollars of

consumer surplus” since 2012. Id. at 36–37. Indeed, Meta argues, Instagram could not have

achieved such success and growth without Meta’s expertise. See Meta Reply at 50.

Defendant’s arguments, however, mostly demonstrate the existence of factual disputes

over whether Instagram was a competitive threat (or very likely to become one) at the time of the

71

acquisition. Plaintiff has put forward ample evidence that it was, and that evidence carries its

burden at this stage in the proceedings. Defendant is also wrong to suggest that Plaintiff must

prove that Instagram would have avoided the “graveyard . . . of promising startups” in order to

make its case. See Meta MSJ at 41. On the contrary, Microsoft held that, where it can be shown

that a monopolist squashed an actual or nascent competitor, “the defendant is made to suffer the

uncertain consequences of its own undesirable conduct.” 253 F.3d at 79 (internal quotation

marks omitted). Meta’s evidence of consumer surplus and the procompetitive benefits of the

merger are, moreover, best understood as rebuttals to the Commission’s prima facie case — i.e.,

as evidence that the elimination of a monopolist’s competitor did not actually result in reduced

consumer welfare, as expected. The Court will thus consider that evidence when assessing

Plaintiff’s Cross-Motion for Partial Summary Judgment.

b. WhatsApp

Unlike Instagram, WhatsApp was not Meta’s actual competitor when acquired. Instead,

the FTC’s theory is that WhatsApp was a nascent competitor: a messaging application with the

potential to expand into the PSN market. See FTC Opp. at 42–44. Although Plaintiff’s evidence

on this front is less extensive than it is for Instagram, it also creates a triable issue of fact over

whether the WhatsApp acquisition was anticompetitive.

The switch to mobile again provides the starting point for this tale. As part of that

transformation, OTT mobile-messaging applications — which relied on the internet to send

messages — became newly available. See FTC SMF, ¶¶ 1714–17. One such application was

Meta’s own Facebook Messenger; another was WhatsApp. As previously recounted, WhatsApp

grew quickly, especially overseas. See supra Section I.A.2.b; see also FTC SMF, ¶¶ 1734–36

(documenting growth). Meta executives assessed at the time that WhatsApp was outperforming

72

Facebook Messenger along multiple dimensions, including its ability to notify users that they had

received a message. See FTC SMF, ¶ 1723; see also id., ¶ 1724 (Messenger under-performing

relative to LINE, KakaoTalk, and WeChat). As a result, more users relied on WhatsApp than on

Messenger to send messages. Id., ¶ 1726.

The real danger, however, was not WhatsApp’s dominance of OTT mobile messaging. It

was the possibility that it might pivot into providing PSN services. As Plaintiff notes, other OTT

mobile-messaging services had made such a pivot in Asia by 2012, mostly to grand success. See

id., ¶¶ 1729–31. Meta executives, noting WhatsApp’s growth as a messaging service, see id.,

¶¶ 1736–37, began tracking such pivots. See, e.g., id., ¶¶ 1799–1800. Leadership was openly

nervous: one internal presentation called mobile messengers “one of the most significant

competitive threats [Meta] face[s]” because they were “increasingly moving into core social-

networking product areas,” posing “a direct threat to [Meta’s] primary products.” Id.,

¶ 1802(c)(i). In February 2013, Zuckerberg told Meta’s board of directors that he wanted to “use

M&A to build a competitive moat around [Meta] on mobile and ads.” Id., ¶ 1802(g)–(h). He

explained that “the biggest competitive vector for [Meta] is for some company to build out a

messaging app for communicating with small groups of people, and then transform[] that into a

broader social network.” Id., ¶ 1802(g). A month later, he wrote of “rumors of WhatsApp”

doing just that. Id., ¶ 1807(f). A different Meta executive said in an email that this possibility

kept him “awake every night.” Id., ¶ 1802(j).

Record evidence suggests that these fears had some basis in reality. WhatsApp may have

been capable of the pivot: the FTC’s expert witness opined that it showed “extraordinary promise

as indicated by industry metrics for scale, growth, and engagement.” Pl. Docs. Vol. 76 at ECF

p. 720 (Expert Rebuttal Report of Jihoon Rim, ¶ 32). Sequoia Capital, a prominent venture firm

73

that took a seat on WhatsApp’s board as part of its growth-round investment, see FTC SMF,

¶ 1763, provided comprehensive assistance to WhatsApp, from hiring, to acquisitions, to money

management and more. See id., ¶¶ 1767–75. Such evidence supports the thesis that WhatsApp

could have pulled off an expansion into new markets. In addition to the capability to pivot,

moreover, WhatsApp had a reason to do so: it needed a viable monetization approach. While it

was operating on a paid subscription model for some users (and a single download fee for

others), . See id., ¶¶ 1778–82; see also id., ¶ 1785 (

). Advertising typically provides such a path, but it

was a poor match for a messaging service because it detracted from the user experience

significantly. See, e.g., id., ¶¶ 1792–93.

. See id., ¶ 1794(a)

.

Meta also appeared to worry that another company could snap up WhatsApp and develop

it into a PSN service. By April 2012, Zuckerberg was under the impression that WhatsApp had

turned down a sizable offer from Google and stated that he was willing to pay $1 billion “if

[Meta] could get them.” Id., ¶ 1811(a). Meta leadership speculated that Google would be

interested in trying again, especially after the failure of its own PSN service, Google+. See id.,

¶ 1811(c)–(e). That outcome, all agreed, would be a major issue for Meta. See id., ¶ 1811(f)–

(g). And, as it turns out, Google had been trying to buy WhatsApp — since at least 2010, to the

tune of up to , id., ¶ 1812 — potentially to develop it (according to one Google

presentation) “into a mobile social network.” Id., ¶ 1813. Meta was also aware that other

74

suitors, such as WeChat’s owner Tencent, were pursuing a WhatsApp acquisition. Id., ¶¶ 1817–

20.

So, Meta acted. Spurred on by renewed interested from Google, Meta leadership began

more explicit overtures to WhatsApp’s Jan Koum in early February 2014. Id., ¶ 1832. By

February 17, Meta’s board of directors had met to approve the acquisition, at a total cost of $19

billion. Id., ¶ 1835. Two days later, the acquisition was announced at that price. Id., ¶ 1837.

Plaintiff’s evidence suggests that most valuations of WhatsApp prior to that date, either explicit

or implied, reached at most $12.5 billion. See id., ¶ 1838. It also suggests that Meta did not

conduct such a valuation itself. See id., ¶¶ 1839, 1842. Indeed, using a post-hoc valuation

conducted by accounting firm KPMG at Meta’s behest in 2015, the FTC asserts that Meta

incurred “a $10.9 billion or 135% transaction premium for its acquisition of WhatsApp.” Id.,

¶¶ 1843–44.

Defendant amusingly contends that the FTC’s narrative “makes Jack-and-the-Beanstalk

sound like a documentary.” Meta MSJ at 41. It points in rebuttal to several pieces of evidence

that WhatsApp did not constitute the threat to Meta that the Commission claims, including

statements from WhatsApp’s founders that it had no intention of pursuing growth in the U.S.

market, pivoting into providing PSN services, or pursuing monetization via advertising, to which

it had a “deep aversion.” Id. at 43–44; see also Meta SMF, ¶¶ 786–95 (outlining relevant

evidence). Defendant also observes that several other messaging products could have been well

positioned to make the pivot, putting pressure on the FTC’s theory that the acquisition of

WhatsApp materially reduced competitive pressure on Meta. See Meta MSJ at 45. As for the

story about Google’s interest in acquiring WhatsApp to build it into a PSN service? A “chain of

speculation,” id. at 46, built on “spectral possibilities.” Meta Reply at 53.

75

Whatever the persuasiveness of Meta’s arguments on this front, however — and they are

surely stronger than its maneuvers regarding the Instagram acquisition — they merely raise the

existence of deep factual disputes requiring the weighing of various pieces of evidence. At this

stage, a reasonable factfinder could conclude, based on the evidence Plaintiff has presented, that

WhatsApp constituted a nascent threat to Meta at the time of its acquisition, and that the merger

was therefore anticompetitive. The evidence certainly provides reason to think that Meta feared

WhatsApp’s potential as a nascent PSN service and that WhatsApp was capable of making such

a pivot, as many of its rivals had done outside the United States. That evidence is sufficient to

preclude summary judgment.

***

In Microsoft, our Circuit prohibited monopolists from “squash[ing] nascent, albeit

unproven, competitors at will — particularly in industries marked by rapid technological

advance and frequent paradigm shifts.” 253 F.3d at 79. Today, the Court holds that a trier of

fact could reasonably conclude that Meta did just that when it acquired Instagram and WhatsApp

and eliminated them as actual or nascent competitive threats. Abundant evidentiary disputes of

course remain, just as they do for whether Meta possesses monopoly power in the first place.

But those disputes, whatever their ultimate resolution, hold open the courthouse doors for now.

3. Procompetitive Justifications

What remains to be decided is the FTC’s Cross-Motion for Partial Summary Judgment on

certain of Defendant’s affirmative defenses. Having survived Meta’s barrage against its own

case, Plaintiff fires back that there is no further dispute of material fact as to whether Meta’s

procompetitive justifications for the acquisitions outweigh their anticompetitive effects. It thus

asks the Court to bar Defendant from raising these defenses at trial. See FTC Opp. at 64–80;

76

ECF No. 372-1 (FTC Reply) at 1–22. Here, however, the (brown) shoe is on the other foot. In

evaluating much of the same record evidence as above, the Court changes tack and gives Meta

all of the favorable factual inferences that it has just given to the FTC. It first discusses the

applicable legal principles for evaluating such justifications. It then applies that framework to

Meta’s asserted justifications and finds that each but the last survives to trial.

a. Legal Framework

Unfortunately for the reader, who has already traversed hill and dale, more tricky terrain

must be navigated before we are home. The central issue is this: for many decades, mergers in

this nation have been challenged mostly under Section 7 of the Clayton Act, which permits

enforcement agencies to block any merger where “the effect of such acquisition may be

substantially to lessen competition, or to tend to create a monopol

This text is long and has been trimmed here. Open the source document for the complete record.