Klein et al v. Facebook, Inc.

Filing 214

ORDER by Judge Lucy H. Koh Granting in Part and Denying in Part (97) Motion to Dismiss in case 5:20-cv-08570-LHK. (lhklc3, COURT STAFF) (Filed on 1/14/2022)

1
2
3
4
5
6
7
8

UNITED STATES DISTRICT COURT

9

NORTHERN DISTRICT OF CALIFORNIA

10
SAN JOSE DIVISION

United States District Court
Northern District of California

11
12

MAXIMILIAN KLEIN, et al.
Plaintiffs,

13
14
15
16

v.
FACEBOOK, INC.,

Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND
DENYING IN PART MOTION TO
DISMISS WITH LEAVE TO AMEND
Re: Dkt. No. 97

Defendant.

17
18
19
20
21
22
23
24
25
26
27

Plaintiffs Maximilian Klein; Sarah Grabert; and Rachel Banks Kupcho (collectively,
“Consumers”) and Affilious, Inc.; Jessyca Frederick; Mark Young; Joshua Jeon; 406 Property
Services, PLLC; Mark Berney; Jessica Layser; Katherine Looper; and Zahara Mossman
(collectively, “Advertisers”) individually and on behalf of all others similarly situated, sue
Defendant Facebook, Inc. (“Facebook”).
Before the Court is Facebook’s motion to dismiss the Consolidated Consumer Class
Action Complaint and the Consolidated Advertiser Class Action Complaint. ECF No. 97. Having
considered the parties’ submissions, the parties’ arguments at the hearing, the relevant law, and the
record in this case, the Court GRANTS IN PART and DENIES IN PART Facebook’s motion to
dismiss with leave to amend.

28
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

TABLE OF CONTENTS

1
I.

Procedural History ..................................................................................................................... 1

II.

2

Legal Standard ........................................................................................................................... 3

3
4
5
6
7
8

A.

Motion to Dismiss Under Federal Rule of Civil Procedure 12(b)(6) ................................. 3

B.

Leave to Amend .................................................................................................................. 4

III. Discussion ................................................................................................................................. 5
A.

9

Consumers Adequately Allege that Facebook Has Monopoly Power in the Social
Network and Social Media Markets ................................................................................ 9

10

United States District Court
Northern District of California

11

Advertisers Adequately Allege that Facebook Has Monopoly Power in the Social
Advertising Market ....................................................................................................... 33

12
13

B.

The Court Denies Facebook’s Motion to Dismiss Consumers’ Data Privacy Claims ..... 38
Consumers Allege that Facebook Obtained and Maintained Monopoly Power by
Repeatedly Deceiving Users About Facebook’s Data Privacy Practices ...................... 40

14
15

Consumers Allege with Sufficient Particularity that Facebook Made Numerous
“Clearly False” Representations About Its Collection and Monetization of Data ........ 53

16

Consumers’ Data Privacy Claims Are Timely.............................................................. 55

17

Consumers Adequately Allege that Facebook’s False Representations About Its Data
Privacy Practices Were Not Readily Susceptible of Neutralization ............................. 58

18

Consumers Adequately Allege that Facebook’s False Representations Were “Clearly
Material” ....................................................................................................................... 61

19
20

Consumers Adequately Allege Causal Antitrust Injury................................................ 65

21

Consumers’ Request for Injunctive Relief as a Remedy for Consumers’ Data Privacy
Claims Is Not Barred by the Doctrine of Laches .......................................................... 69

22
23

Consumers and Advertisers Adequately Allege that Facebook Has Monopoly Power in
Cognizable Product Markets............................................................................................... 7

C.

24

The Court Grants Facebook’s Motion to Dismiss Consumers’ and Advertisers’ “Copy,
Acquire, Kill” Claims With Leave to Amend .................................................................. 70

25

Consumers and Advertisers Allege that Facebook’s “Copy, Acquire, Kill” Strategy
Allowed Facebook to Maintain Monopoly Power ........................................................ 72

26

Consumers’ and Advertisers’ “Copy, Acquire, Kill” Claims Are Untimely ................ 88

27
28

D.

The Court Denies Facebook’s Motion to Dismiss Advertisers’ GNBA Claims ............ 100
i

Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

The Google Network and Bidding Agreement ........................................................... 100

1

Advertisers Adequately Allege that the GNBA Caused Them Injury ........................ 102

2
3

E.

The Court Grants Facebook’s Motion to Dismiss Consumers’ Unjust Enrichment Claim
With Leave to Amend ..................................................................................................... 106

4
IV. Conclusion ............................................................................................................................. 107
5
6
7
8
9
10

United States District Court
Northern District of California

11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28

ii
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

I. PROCEDURAL HISTORY

2

On December 3, 2020, Plaintiffs Klein and Grabert filed an initial complaint against

3

Defendant Facebook. ECF No. 1. Subsequently, 11 other antitrust cases were filed by consumers

4

and advertisers against Facebook. On February 9, 2021, the Court: (1) granted motions to relate

5

Sherman v. Facebook, Kupcho v. Facebook, Dames v. Facebook, Steinberg v. Facebook, Layser v.

6

Facebook, and Rosenman v. Facebook to the instant case; (2) concluded that Affilious v. Facebook

7

was related to the instant case; and (3) consolidated these cases with the instant case. ECF No. 47.

8

On February 11, 2021, Plaintiff Rosenman voluntarily dismissed her case. Facebook v.

9

Rosenman, No. 21-CV-00336-LHK, ECF No. 17.
On February 25, 2021, the Court granted motions to relate Kovacevich v. Facebook and

11

United States District Court
Northern District of California

10

Garvin v. Facebook to the instant case and consolidated these cases with the instant case. ECF

12

No. 50. On March 16, 2021, the Court granted a motion to relate Wasvary v. Facebook to the

13

instant case and consolidated it with the instant case. ECF No. 68. On April 9, 2021, the Court

14

granted a motion to relate Ryan v. Facebook to the instant case and consolidated that case with the

15

instant case. ECF No. 85.

16

After voluntarily dismissing her federal case on February 11, 2021, Plaintiff Rosenman

17

refiled her case in state court, and Facebook removed the refiled case to federal court. See

18

Rosenman v. Facebook, No. 21-CV-2108, ECF No. 1. On April 9, 2021, the Court related the

19

refiled Rosenman Case to the instant case. ECF No. 85. On April 26, 2021, Rosenman filed a

20

motion to remand, which the Court denied on August 27, 2021. Rosenman v. Facebook, No. 21-

21

CV-02108-LHK, ECF Nos. 17, 26.

22

On March 18, 2021, the Court held a hearing on motions for appointment as interim class

23

counsel. ECF No. 77. That same day, the Court appointed Stephen A. Swedlow of Quinn

24

Emanuel Urquhart & Sullivan, LLP and Shana A. Scarlett of Hagens Berman Sobol Shapiro LLP

25

as Interim Class Counsel for the Consumer class (“Consumers”) and appointed Warren Postman

26

of Keller Lenkner and Brian D. Clark of Lockridge Grindal Nauen P.L.L.P. to serve on Plaintiffs’

27

Executive Committee for Consumers. ECF No. 73.

28

1
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

On March 18, 2021, the Court appointed Yavar Bataee of Bathaee Dunne LLP and Kristen

1
2

M. Anderson of Scott + Scott LLP as Interim Class Counsel for the Advertiser class

3

(“Advertisers”) and appointed Tina Wolfson of Ahdoot & Wolfson, PC and Keith J. Verrier of

4

Levin Sedran & Berman LLP to serve on Plaintiffs’ Executive Committee for Advertisers. Id.
On April 22, 2021, Consumers filed a Consolidated Consumer Class Action Complaint.

5
6

ECF No. 87 (“CC”). Consumers are individuals who use Facebook’s services, including

7

Facebook, Facebook Messenger, Instagram, and WhatsApp. Id. ¶¶ 19, 23, 26. Consumers allege

8

that, “[a]bsent Facebook’s anticompetitive scheme, fair competition would have required

9

Facebook to provide consumers greater value in return for consumers’ data on a market-wide
basis.” Id. ¶ 10. Consumers seek to represent a class of “[a]ll persons in the United States who

11

United States District Court
Northern District of California

10

maintained a Facebook profile at any point from 2007 up to the date of the filing of this action.”

12

Id. ¶ 248. Consumers assert five claims: (1) monopolization of the Social Network Market in

13

violation of § 2 of the Sherman Act; (2) attempted monopolization of the Social Network Market

14

in violation of § 2 of the Sherman Act; (3) monopolization of the Social Media Market in violation

15

of § 2 of the Sherman Act; (4) attempted monopolization of the Social Media Market in violation

16

of § 2 of the Sherman Act; and (5) unjust enrichment under California common law. Id. ¶¶ 260–

17

317.

18

On April 22, 2021, Advertisers filed a Consolidated Advertiser Class Action Complaint.

19

ECF No. 86 (“AC”). Advertisers are individuals, entities, and corporations who purchased

20

advertising from Facebook. Id. ¶¶ 24–33. Advertisers allege that they paid prices for advertising

21

that were “higher than they would have been absent Facebook’s anticompetitive conduct and

22

unlawfully acquired and/or maintained monopoly.” Id. ¶ 33. Affilious, Inc.; Jessyca Frederick;

23

Joshua Jeon; and 406 Property Services, PLLC seek to represent a class of “[a]ll persons, entities,

24

and/or corporations in the United States who purchased advertising from Facebook between

25

October 1, 2012, and April 3, 2018, but not after April 3, 2018.” AC ¶ 529. Mark Berney, Mark

26

Young, Jessica Layser, Katherine Looper, and Zahara Mossman seek to represent a class of “[a]ll

27

persons, entities, and/or corporations in the United States who purchased advertising from

28

2
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

Facebook between April 4, 2018, and the present.” Id. ¶ 532. Advertisers assert three claims: (1)

2

monopolization of the Social Advertising Market in violation of § 2 of the Sherman Act; (2)

3

attempted monopolization of the Social Advertising Market in violation of § 2 of the Sherman

4

Act; and (3) restraint of trade in violation of § 1 of the Sherman Act. Id. ¶¶ 547–69.

5

On May 7, 2021, Facebook filed a motion to disqualify Keller Lenkner. ECF No. 93. On

6

July 13, 2021, the Court granted Facebook’s motion to disqualify Keller Lenkner in advance of the

7

July 15, 2021 hearing on Facebook’s motion to dismiss in light of the importance of the issues

8

raised by the motion to disqualify. ECF No. 123. The Court noted that the Court was focusing on

9

preparation for the motion to dismiss hearing and would issue a written decision on the motion to
disqualify shortly. Id. On July 20, 2021, the Court issued a written decision on the motion to

11

United States District Court
Northern District of California

10

disqualify. ECF No. 127.

12

On May 20, 2021, Facebook filed the instant motion to dismiss. ECF No. 97 (“Mot.”). On

13

June 17, 2021, Plaintiffs filed an opposition. ECF No. 109 (“Opp.”). On June 28, 2021, the

14

United States District Court for the District of Columbia issued decisions on Facebook’s motions

15

to dismiss in FTC v. Facebook, Inc., No. 20-CV-3590-JEB, and State of New York v. Facebook,

16

Inc., No. 20-CV-3589-JEB. On July 1, 2021, the Court permitted Plaintiffs to file a supplemental

17

brief regarding those decisions. ECF No. 115. On July 5, 2021, Plaintiffs filed a supplemental

18

brief. ECF No. 116 (“Pls. Supp. Br.”). On July 7, 2021, Facebook filed a reply. ECF No. 117

19

(“Reply”). The Court held a hearing on the instant motion on July 15, 2021.

20

Plaintiff Layser voluntarily dismissed her case on August 5, 2021. ECF No. 129. Plaintiff

21

Mossman voluntarily dismissed her case on November 4, 2021. ECF No. 182. Thus, a total of 11

22

cases remain.

23

II. LEGAL STANDARD

24
25

A. Motion to Dismiss Under Federal Rule of Civil Procedure 12(b)(6)
Rule 8(a) of the Federal Rules of Civil Procedure requires a complaint to include “a short

26

and plain statement of the claim showing that the pleader is entitled to relief.” Fed. R. Civ. P.

27

8(a)(2). A complaint that fails to meet this standard may be dismissed pursuant to Federal Rule of

28

3
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

Civil Procedure 12(b)(6). Rule 8(a) requires a plaintiff to plead “enough facts to state a claim to

2

relief that is plausible on its face.” Bell Atlantic Corp. v. Twombly, 550 U.S. 544, 570 (2007). “A

3

claim has facial plausibility when the plaintiff pleads factual content that allows the court to draw

4

the reasonable inference that the defendant is liable for the misconduct alleged.” Ashcroft v. Iqbal,

5

556 U.S. 662, 678 (2009). “The plausibility standard is not akin to a probability requirement, but

6

it asks for more than a sheer possibility that a defendant has acted unlawfully.” Id. (internal

7

quotations omitted).

8
9

For purposes of ruling on a Rule 12(b)(6) motion, the Court “accept[s] factual allegations
in the complaint as true and construe[s] the pleadings in the light most favorable to the nonmoving
party.” Manzarek v. St. Paul Fire & Marine Ins. Co., 519 F.3d 1025, 1031 (9th Cir. 2008).

11

United States District Court
Northern District of California

10

Additionally, the Court may consider materials referenced in the complaint under the

12

incorporation by reference doctrine, even if a plaintiff failed to attach those materials to the

13

complaint. Knievel v. ESPN, 393 F.3d 1068, 1076 (9th Cir. 2005).

14

The Court, however, need not accept as true allegations contradicted by judicially

15

noticeable facts, see Shwarz v. United States, 234 F.3d 428, 435 (9th Cir. 2000), and it “may look

16

beyond the plaintiff’s complaint to matters of public record” without converting the Rule 12(b)(6)

17

motion into a motion for summary judgment, Shaw v. Hahn, 56 F.3d 1128, 1129 n.1 (9th Cir.

18

1995). Nor must the Court “assume the truth of legal conclusions merely because they are cast in

19

the form of factual allegations.” Fayer v. Vaughn, 649 F.3d 1061, 1064 (9th Cir. 2011) (per

20

curiam) (quoting W. Mining Council v. Watt, 643 F.2d 618, 624 (9th Cir. 1981)). Mere

21

“conclusory allegations of law and unwarranted inferences are insufficient to defeat a motion to

22

dismiss.” Adams v. Johnson, 355 F.3d 1179, 1183 (9th Cir. 2004).

23
24

B. Leave to Amend
If the Court determines that a complaint should be dismissed, it must then decide whether

25

to grant leave to amend. Under Rule 15(a) of the Federal Rules of Civil Procedure, leave to

26

amend “shall be freely given when justice so requires,” bearing in mind “the underlying purpose

27

of Rule 15 to facilitate decision[s] on the merits, rather than on the pleadings or technicalities,”

28

4
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

Lopez v. Smith, 203 F.3d 1122, 1127 (9th Cir. 2000) (en banc) (alteration and internal quotations

2

omitted). When dismissing a complaint for failure to state a claim, “a district court should grant

3

leave to amend even if no request to amend the pleading was made, unless it determines that the

4

pleading could not possibly be cured by the allegation of other facts.” Id. at 1130 (internal

5

quotations omitted). Accordingly, leave to amend generally shall be denied only if allowing

6

amendment would unduly prejudice the opposing party, cause undue delay, or be futile, or if the

7

moving party has acted in bad faith. Leadsinger, Inc. v. BMG Music Publ’g, 512 F.3d 522, 532

8

(9th Cir. 2008).

9

III. DISCUSSION

10

Section 2 of the Sherman Act makes it unlawful for any person to “monopolize, or attempt

United States District Court
Northern District of California

11

to monopolize, or combine or conspire with any other person or persons, to monopolize any part

12

of the trade or commerce among the several States, or with foreign nations.” 15 U.S.C. § 2. To

13

state a claim for monopolization under Section 2 of the Sherman Act, a plaintiff must allege: “(1)

14

[p]ossession of monopoly power in the relevant market”; (2) “willful acquisition or maintenance

15

of that power” through exclusionary conduct; and (3) “causal antitrust injury.” SmileCare Dental

16

Grp. v. Delta Dental Plan of Cal., Inc., 88 F.3d 780, 783 (9th Cir. 1996) (internal quotation

17

omitted).

18

In turn, Section 1 of the Sherman Act prohibits “[e]very contract, combination in the form

19

of trust or otherwise, or conspiracy, in restraint of trade or commerce.” 15 U.S.C. § 1. “To

20

establish liability under § 1, a plaintiff must prove (1) the existence of an agreement, and (2) that

21

the agreement was in unreasonable restraint of trade.” Aerotec Int’l, Inc. v. Honeywell Int’l, Inc.,

22

836 F.3d 1171, 1178 (9th Cir. 2016).

23

Consumers allege that Facebook violated Section 2 of the Sherman Act through a “two-

24

pronged anticompetitive strategy.” See CC ¶ 219. First, Consumers allege that Facebook acquired

25

and maintained monopoly power in the Social Network and Social Media Markets by making

26

false representations to users about Facebook’s data privacy practices. For ease of reference, the

27

Court refers to Consumers’ claims based on this theory of liability as Consumers’ “data privacy

28

5
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

claims.” Second, Consumers allege that Facebook’s “Copy, Acquire, Kill” strategy allowed

2

Facebook to maintain monopoly power in the Social Network and Social Media Markets. The

3

Court refers to Consumers’ claims based on this theory of liability as Consumers’ “‘Copy,

4

Acquire, Kill’ claims.”

5
6
7

Consumers also allege that Facebook’s conduct constitutes unjust enrichment under
California common law. The Court refers to this as Consumers’ “Unjust Enrichment claim.”
Like Consumers, Advertisers challenge Facebook’s “Copy, Acquire, Kill” strategy under

8

Section 2 of the Sherman Act. Specifically, Advertisers allege that the “Copy, Acquire, Kill”

9

strategy allowed Facebook to maintain monopoly power in the Social Advertising Market. The
Court refers to Advertisers’ claims based on this theory of liability as Advertisers’ “‘Copy,

11

United States District Court
Northern District of California

10

Acquire, Kill’ claims.”

12

Additionally, Advertisers allege that Facebook maintained monopoly power in the Social

13

Advertising Market by entering a contract with Google called the Google Network Bidding

14

Agreement (“GNBA”). Advertisers also allege that the GNBA was an unreasonable restraint of

15

trade. Thus, Advertisers challenge the GNBA under Section 1 of the Sherman Act and Section 2

16

of the Sherman Act. The Court refers to these claims as Advertisers’ “GNBA claims.”

17

Facebook moves to dismiss all of these claims. Specifically, Facebook argues that: (1)

18

neither Consumers nor Advertisers have alleged cognizable product markets; (2) Consumers have

19

not plausibly alleged monopoly power; (3) Consumers’ data privacy claims are untimely; (4)

20

Consumers’ and Advertisers’ “Copy, Acquire, Kill” claims are untimely; (5) Consumers have not

21

plausibly alleged that Facebook’s deceptive data privacy practices were anticompetitive; (6)

22

neither Consumers nor Advertisers have plausibly alleged that Facebook’s “Copy, Acquire, Kill”

23

strategy was anticompetitive; (7) neither Consumers nor Advertisers have adequately alleged

24

causal antitrust injury; and (8) Consumers have not adequately stated a claim for unjust

25

enrichment. Facebook also argues that Consumers’ requests for injunctive relief are barred by the

26

doctrine of laches.

27
28

The Court rules as follows: (1) Consumers and Advertisers have adequately alleged that
6
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

Facebook has monopoly power in cognizable product markets; (2) because Consumers have

2

adequately alleged that their data privacy claims are timely, that Facebook’s false representations

3

about its data privacy constitute exclusionary conduct, and that Consumers have suffered a causal

4

antitrust injury, the Court DENIES Facebook’s motion to dismiss Consumers’ data privacy claims;

5

(3) because Consumers’ and Advertisers’ “Copy, Acquire, Kill” claims are untimely, the Court

6

GRANTS Facebook’s motion to dismiss Consumers’ and Advertisers’ “Copy, Acquire, Kill”

7

claims with leave to amend; (4) because Advertisers have adequately alleged that they were

8

injured by the GNBA, the Court DENIES Facebook’s motion to dismiss Advertisers’ GNBA

9

claims; and (5) because Consumers have failed to state a claim for unjust enrichment, the Court
GRANTS Facebook’s motion to dismiss Consumers’ Unjust Enrichment claim with leave to

11

United States District Court
Northern District of California

10

amend. Below, the Court discusses each of these conclusions in turn.

12
13
14
15
16
17
18

A. Consumers and Advertisers Adequately Allege that Facebook Has Monopoly Power
in Cognizable Product Markets
The first element of a claim for monopolization under the Sherman Act is the “[p]ossession
of monopoly power in [a] relevant market.” SmileCare, 88 F.3d at 783. Accordingly, Consumers
and Advertisers must establish “both that a ‘relevant market’ exists and that [Facebook] has power
within that market.” Newcal Indus., Inc. v. Ikon Off. Sol., 513 F.3d 1038, 1044 (9th Cir. 2008).
To “plead a relevant market” for purposes of a Sherman Act claim, a plaintiff must allege
“both a geographic market and a product market.” Hicks v. PGA Tour, Inc., 897 F.3d 1109, 1120

19
(9th Cir. 2018). All parties agree that the relevant geographic market is the United States.
20
21
22

A product market “must encompass the product at issue as well as all economic substitutes
for the product.” Newcal Indus., 513 F.3d at 1045. “Economic substitutes have a ‘reasonable
interchangeability of use’ or sufficient ‘cross-elasticity of demand’ with the relevant product.”

23
Hicks, 897 F.3d at 1120 (internal citation omitted). Whether products are reasonably
24

interchangeable depends on the products’ “price, use[,] and qualities.” United States v. E. I. du

25
Pont de Nemours & Co., 351 U.S. 377, 404 (1956); see also United States v. Microsoft Corp., 253
26
F.3d 34, 51–52 (D.C. Cir. 2001) (explaining that products are in the same market if they are
27
28

7
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

“reasonably interchangeable by consumers for the same purposes”). In turn, there is a cross-

2

elasticity of demand between two products where “an increase in the price of one product leads to

3

an increase in demand for another.” Olin Corp. v. FTC, 986 F.2d 1295, 1298 (9th Cir. 1993).

4

Additionally, “[w]ithin a general product market, ‘well-defined submarkets may exist

5

which, in themselves, constitute product markets for antitrust purposes.’” Hicks, 897 F.3d at 1121

6

(quoting Brown Shoe Co. v. United States, 370 U.S. 294, 325 (1962)). A plaintiff may allege a

7

distinct submarket for a product by alleging “practical indicia” of the submarket. Brown Shoe,

8

370 U.S. at 325. Examples of “practical indicia” include: “industry or public recognition of the

9

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

11

United States District Court
Northern District of California

10

specialized vendors.” Id.

12

The Ninth Circuit has held that “what constitutes a relevant market is a factual

13

determination for the jury.” Image Tech. Servs. v. Eastman Kodak, 125 F.3d 1195, 1203 (9th Cir.

14

1997). Furthermore, the Ninth Circuit has held that the “definition of the relevant market is

15

basically a fact question dependent upon the special characteristics of the industry involved.”

16

Twin City Sportservice, Inc. v. Charles O’Finley & Co., 676 F.2d 1291, 1299 (9th Cir. 1982).

17

Although a defendant may raise market definition in a motion to dismiss, “the question of whether

18

the market should include other products is better resolved at the summary judgment stage.”

19

Datel Holdings Ltd. v. Microsoft Corp., 712 F. Supp. 2d 974, 997 (N.D. Cal. 2010). Accordingly,

20

“[o]n a motion to dismiss, the court need not engage in extensive analyses of reasonable

21

interchangeability and cross elasticity of demand.” In re Webkinz Antitrust Litig., 695 F. Supp. 2d

22

987, 995 (N.D. Cal. 2010). In other words, a plaintiff is “not required to identify every alleged

23

competitor in its pleadings.” FTC v. Facebook, Inc., No. 20-CV-3590-JEB, 2021 WL 2643627, at

24

*13 (D.D.C. June 28, 2021).

25

In turn, a plaintiff can establish market power either with “direct evidence” or with

26

“circumstantial evidence.” Rebel Oil Co. v. Atl. Richfield Co., 51 F.3d 1421, 1434 (9th Cir. 1995).

27

Direct evidence includes “evidence of restricted output and supracompetitive prices.” Id. (citing

28

8
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

FTC v. Indiana Fed’n of Dentists, 476 U.S. 447, 460–61 (1986)). However, “such direct proof is

2

only rarely available.” Microsoft, 253 F.3d at 51. Accordingly, a plaintiff typically establishes

3

market power with circumstantial evidence. Id.

4

The strongest circumstantial evidence of market power is evidence that the “defendant

5

owns a dominant share” of the relevant market and that the market has “significant barriers to

6

entry.” Rebel Oil, 51 F.3d at 1434. A showing that the defendant has a market share of greater

7

than 65% typically is sufficient to “establish a prima facie case of market power.” Eastman

8

Kodak, 125 F.3d at 1206. By contrast, “numerous cases hold that a market share of less than 50

9

percent is presumptively insufficient to establish market power.” Rebel Oil, 51 F.3d at 1438. In
turn, “[e]ntry barriers are ‘additional long-run costs that were not incurred by incumbent firms but

11

United States District Court
Northern District of California

10

must be incurred by new entrants,’ or ‘factors in the market that deter entry while permitting

12

incumbent firms to earn monopoly returns.’” Id. at 1439 (citing Los Angeles Land Co. v.

13

Brunswick Corp., 6 F.3d 1422, 1427–28 (9th Cir. 1993)).

14

Consumers Adequately Allege that Facebook Has Monopoly Power in the Social
Network and Social Media Markets

15
Consumers allege that Facebook has monopoly power in the Social Network and Social
16
Media Markets. Facebook contends that, as defined by Consumers, the Social Network and Social
17
Media Markets are not cognizable product markets. Facebook also contends that, even if those
18
markets are cognizable, Consumers have not plausibly alleged that Facebook has monopoly
19
power. For the reasons below, the Court rejects these arguments.
20
21
22

Consumers Adequately Allege that the Social Network and Social Media
Markets are Cognizable Product Markets
Consumers Adequately Allege the Social Network Market

23

According to Consumers, a social network service is a distinct type of social media

24

service, which is an online service that enables users “to distribute various forms of media—such

25

as text messages, photos, videos, and music—to other users of the same application.” CC ¶ 72.

26

Specifically, a social network service enables “users to find, communicate, and interact with

27

friends, family, personal acquaintances, and other people with whom the users have shared

28

9
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

interests or connections.” Id. ¶ 56. Facebook, for example, allows users to create profiles, to

2

designate other users as “friends,” to share media and information with other users, to form groups

3

with other users based on common interests, to plan events with other users, and to play games

4

with other users. Id. ¶ 51.

5

Consumers allege that social network services are not “reasonably interchangeable” with

6

other online services and that there is a distinct Social Network Market, which is a “part or sub-

7

part of the Social Media Market.” Id. ¶ 56. Under Consumers’ definition, Facebook is the only

8

significant social network service and other social network services, such as Diaspora, Ello, Vero,

9

Clubhouse, and Reddit “constitute ‘only a very small drop in the ocean compared to Facebook.’”
Id. ¶ 68. However, in the past, Myspace, Friendster, Orkut, Bebo, Flip.com, and Google+

11

United States District Court
Northern District of California

10

competed with Facebook in the Social Network Market. Id. ¶¶ 36–37.

12

Facebook contends that Consumers’ allegations are inadequate because Consumers have

13

not provided a basis for determining which products are in the Social Network Market and

14

because Consumers unreasonably exclude “countless” services that allow users to “kill time.”

15

Mot. at 16–17. For the reasons below, the Court rejects Facebook’s contentions.

16

To plead a product market based on “reasonable interchangeability,” a plaintiff must allege

17

details about a product’s “‘price, use and qualities’” and explain why products without those

18

characteristics are not reasonably interchangeable. See Webkinz, 695 F. Supp. 2d at 994 (quoting

19

United States v. Oracle Corp., 331 F. Supp. 2d 1098, 1131 (N.D. Cal. 2004)); see also FTC v.

20

Facebook, 2021 WL 2643627, at *11 (explaining that a plaintiff “may permissibly plead that

21

certain ‘factors’ of both the service at issue and its potential substitutes—e.g., their ‘price, use[,]

22

and qualities’—render them not ‘reasonably interchangeable’ in the eyes of users”). In other

23

words, the plaintiff must allege that other types of products are not “reasonably interchangeable by

24

consumers for the same purposes.” Microsoft, 253 F.3d at 51–52. For example, allegations that a

25

product has “a distinct core functionality,” “distinct perceptions by consumers,” “distinct customer

26

targeting by manufacturers,” “distinct analyses by industry analysts,” and “distinct pricing and

27

pricing patterns” are sufficient to establish that there is a distinct market for that product. See

28

10
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

Datel Holdings, 712 F. Supp. 2d at 997. Additionally, statements by companies explaining which

2

products they see as competitors for their own products are highly probative. See Epic Games,

3

Inc. v. Apple Inc., No. 20-CV-05640-YGR, 2021 WL 4128925, at *51 (N.D. Cal. Sept. 10, 2021).

4

Furthermore, “[b]ecause every market that encompasses less than all products is, in a

5

sense, a submarket,” all the factors relevant for determining whether there is a distinct submarket

6

also “are relevant . . . in determining the primary market.” Olin, 986 F.2d at 1299 (citing United

7

States v. Continental Can Co., 378 U.S. 441, 449, 453–55 (1964)). Thus, a plaintiff may support

8

its product market definition by pleading facts which show “industry or public recognition of the

9

[market] as a separate economic entity, the product’s peculiar characteristics and uses, unique
production facilities, distinct customers, distinct prices, sensitivity to price changes, and

11

United States District Court
Northern District of California

10

specialized vendors.” Brown Shoe, 370 U.S. at 325.

12

Consumers allege that social network services have three “peculiar characteristics.” Id.

13

First, every social network service uses a “social graph,” which is a system for tracking

14

connections between users. CC ¶ 57. A social graph records both direct connections, like whether

15

two users have designated each other as “friends,” and indirect connections, like whether two

16

users list the same interests. Id. Social network services often use social graphs to “encourage

17

their users to expand their networks by suggesting new people the users may connect to.” Id.

18

Second, social network services “provide substantive features to users which facilitate a wide

19

array of interactions among the wide array of people that make up a user’s social graph.” Id. ¶ 58.

20

For example, social networks allow users to create groups based on common interests, hobbies,

21

and backgrounds; organize events; and play interactive games. Id. Third, social networks

22

“provide users with the convenience of a ‘one-stop shop’” by “combin[ing] multiple substantive

23

features and functionalities into one product.” Id. ¶ 59. As Facebook CEO Mark Zuckerberg has

24

put it, this “multi-functionality” means that social network services are the “digital equivalent of

25

the town square.” Id.

26
27
28

Together, these characteristics give social network services a “distinct core functionality”:
fostering genuine social experiences between users. Datel Holdings, 712 F. Supp. 2d at 997.
11
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

Facebook itself has stressed to investors that this distinct core functionality sets Facebook’s social

2

network service apart from other online services. For example, in a 2012 presentation to investors,

3

Facebook described its service as “social by design, with people and relationships at the center of

4

every experience.” ECF No. 97-6 at 1. According to Facebook, this core functionality “creates

5

social experiences that are powerful and universal” and has “achieved a level of engagement

6

unlike anything else on the web.” Id. (emphasis added).

7

Indeed, statements by Facebook executives provide strong evidence that Facebook has

8

always viewed other social network services as its primary competitors. The most striking

9

example is a statement made after Google’s launch of the Google+ social network. Prior to the
2010 launch of Google+, Google offered at least three significant online services: YouTube,

11

United States District Court
Northern District of California

10

Gmail, and Google Search. CC ¶¶ 73, 131. However, it was not until Google announced Google+

12

that Facebook began to view Google as a direct competitive threat. With Google+, Google

13

“attempt[ed] to build out a ‘social graph’ that would leverage a common user identity across

14

Google products, including YouTube and Gmail.” Id. ¶ 131. Specifically, the “planned features

15

for Google+ included a continuous scroll product called the ‘stream’; a companion feature called

16

‘sparks,’ which related the ‘stream’ to users’ individual interests; and a sharing app called

17

‘Circles’ to share information with one’s friends, family, contacts, and the public at large.” Id.

18

¶ 132. Facebook immediately realized that Google+ posed a competitive threat unlike any other

19

online service. Indeed, although Facebook had been planning to remove a privacy feature in 2011,

20

a Facebook executive vetoed that plan and stated: “IF ever there was a time to AVOID

21

controversy, it would be when the world is comparing our offerings to G+.” Id. ¶ 134. The

22

executive added that Facebook should postpone any privacy changes “until the direct competitive

23

comparisons begin to die down.” Id. These statements indicate Facebook’s belief that Google+, a

24

social network service, was a competitive threat in a way that YouTube, Gmail, and Google

25

Search were not.

26
27
28

Other internal communications show that Facebook employees have frequently
distinguished between social network services and other types of online services. For example, in
12
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

August 2012, Facebook VP of Business and Marketing Partnerships David Fischer stated that

2

there are at least six distinct categories of online applications: “Social network apps,” which

3

included Google+ and Twitter; “Photo sharing apps”; “Messaging apps”; “Local apps”; “Social

4

search apps”; and “Platforms.” AC ¶ 136. Similarly, in November 2012, Facebook VP of Global

5

Operations Justin Osofsky stated that Facebook would be making an effort to “define competitive

6

networks.” Id. ¶ 143. Indeed, when Facebook began planning to remove public access to

7

Facebook’s Platform for developers, a Facebook engineer characterized the plan as a

8

“protectionist grab to make sure no one else can make a competing social network.” Id. ¶ 180.
These statements strongly support Consumers’ allegation that there is a distinct Social

10

Network Market. As discussed, a company’s statements indicating that it does not view other

11

United States District Court
Northern District of California

9

products as competitive with its own product support a conclusion that the products are not in the

12

same market. Epic Games, 2021 WL 4128925, at *51. In Epic, for example, the court highlighted

13

statements by Microsoft, the manufacturer of the Xbox console gaming system, which indicated

14

that Microsoft did “not consider cellular or tablet devices such as the iPhone or iPad as

15

competitors to the Xbox.” Id. These statements supported a conclusion that there were separate

16

markets for “console gaming” and “mobile gaming.” Id. at *51–52. Similarly, all of Facebook’s

17

statements about the unique properties of social network services support a conclusion that there is

18

a distinct Social Network Market.

19

Moreover, Consumers explain in detail why social network services and generic social

20

media services are not “reasonably interchangeable by consumers for the same purposes.”

21

Microsoft, 253 F.3d at 51–52. In contrast to social network services, most social media services

22

do not focus on facilitating social connections and fostering relationships between users. CC ¶ 65.

23

Instead, most social media services focus on the “distribution and consumption of content.” Id.

24

Consumers provide two examples to illustrate this distinction: YouTube and TikTok.

25

YouTube is a social media service which “facilitates the sharing of videos of varying temporal

26

length.” Id. ¶ 36. Unlike a social network service, in which users share content with friends and

27

acquaintances, most videos posted on YouTube can be viewed by “users with a wide range of

28

13
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

relationships to the person posting, including by strangers.” Id. ¶ 65. For example, whereas a

2

“user might use YouTube to access and watch a complete stranger’s video—such as a cooking

3

recipe—the same user would use Facebook, not YouTube, to share that video with the user’s

4

friends, family, and real-world connections.” Id. (emphasis in original). As the U.S. House of

5

Representative has explained, there is a “fundamental difference” between sharing personal

6

content, like a video of a “child’s first steps,” with friends on a social network service and

7

“broadcasting it publicly on YouTube.” Investigation of Competition in Digital Markets, Majority

8

Staff Report and Recommendations, Subcommittee on Antitrust, Commercial, and Administrative

9

Law of the Committee on the Judiciary at 91 (Oct. 6, 2020) (“House Report”),
https://bit.ly/32JPE1r. Thus, social network services like Facebook and social media services like

11

United States District Court
Northern District of California

10

YouTube have “distinct core functionalit[ies],” Datel Holdings, 712 F. Supp. 2d at 997, and are

12

not “reasonably interchangeable by consumers for the same purposes.” Microsoft, 253 F.3d at 51–

13

52.

14

In turn, TikTok is a social media service which “allows for the sharing of short-form

15

videos of limited temporal length.” CC ¶ 36. Unlike a social network service, users of TikTok

16

primarily post videos to be viewed by strangers. Indeed, TikTok’s stated goal is to be a “global

17

platform for users to express their ideas by sharing videos with a broader community.” House

18

Report at 91. Accordingly, for all the reasons that apply in the context of YouTube, social

19

network services like Facebook and social media services like TikTok have “distinct core

20

functionalit[ies],” Datel Holdings, 712 F. Supp. 2d at 997, and are not “reasonably

21

interchangeable by consumers for the same purposes.” Microsoft, 253 F.3d at 51–52.

22

For slightly different reasons, Consumers allege that social network services are distinct

23

from mobile messaging service like Apple’s “iMessage.” CC ¶ 60. Although iMessage is

24

designed to allow users to share media with friends and acquaintances, iMessage cannot be used to

25

communicate with anybody outside a pre-defined list of contacts and does not offer substantive

26

features besides “delivering messages.” Id. Thus, unlike social network services, iMessage does

27

not “facilitat[e] a broader online social experience.” Id. Moreover, iMessage cannot be used by

28

14
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

people without Apple devices, whereas social network services can “be used across devices.” See

2

CC ¶ 60 n.23; House Report at 136 n.752. Thus, social network services like Facebook and

3

mobile messaging services like iMessage have “distinct core functionalit[ies],” Datel Holdings,

4

712 F. Supp. 2d at 997, and are not “reasonably interchangeable by consumers for the same

5

purposes.” Microsoft, 253 F.3d at 51–52.

6

Additionally, Consumers explain in detail why social network services are distinct from

7

professional networking services like LinkedIn. Although professional networking services use a

8

social graph to facilitate connections between their users, these services have a relatively narrow

9

purpose: help their users find jobs and learn about their industries. CC ¶ 61. This narrow purpose
means that many users “have both a LinkedIn and a Facebook account, not one or the other,” id.

11

United States District Court
Northern District of California

10

¶ 62, and that most consumers do not view LinkedIn as an “economic substitute” for a social

12

network service. Newcal Indus., 513 F.3d at 1045. Indeed, LinkedIn itself has stressed that

13

people use social network services and professional networking services for fundamentally

14

different reasons. As explained by LinkedIn in a presentation to potential advertisers, users of

15

“personal networks” typically share “info on friends” and “info on personal interests,” whereas

16

users of “professional networks” typically share “career info.” Id. ¶ 64. Thus, social network

17

services like Facebook and professional networking services like LinkedIn have “distinct core

18

functionalit[ies],” Datel Holdings, 712 F. Supp. 2d at 997, and are not “reasonably

19

interchangeable by consumers for the same purposes.” Microsoft, 253 F.3d at 51–52.

20

Consumers also explain why social network services are not reasonably interchangeable

21

with online services outside the sphere of social media. For example, online entertainment

22

services like Hulu limit users to viewing media provided by the services themselves and do not

23

allow users to communicate with each other. See CC ¶ 75. Thus, such services do not facilitate

24

the distribution of media between “users of the same application,” id. ¶ 72, let alone

25

“communication and sharing content among groups of friends,” House Report at 91. Moreover,

26

whereas social network services make money by showing users advertisements, online

27

entertainment services like Hulu typically “collect from users a per-use monetary fee or a

28

15
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

regularly-occurring (monthly or annual) fee.” CC ¶ 76. Accordingly, social network services and

2

online entertainment services have both “distinct core functionalit[ies]” and “distinct pricing.”

3

Datel Holdings, 712 F. Supp. 2d at 997. Social network services are even more distinct from

4

search engines like Google, Yahoo, and Bing, which do not host their own media and do not allow

5

for communications between users. CC ¶ 60. Accordingly, social network services, online

6

entertainment services, and search engines are not “reasonably interchangeable by consumers for

7

the same purposes.” Microsoft, 253 F.3d at 51–52.

8
9

Finally, Consumers point out that there has been significant “industry or public recognition
of the” Social Network Market “as a separate economic entity.” Brown Shoe, 370 U.S. at 325.
For example, the U.S. House of Representatives found during its recent investigation of

11

United States District Court
Northern District of California

10

competition in digital markets that social network services have the distinct purpose of

12

“facilitating communication and sharing content among groups of friends who choose each other

13

and enjoy content in large part because of those relationships.” House Report at 91. Based on this

14

finding, the House concluded that “the Social Network Market and the Social Media Market” are

15

related, but distinct, economic markets. See CC ¶ 55; House Report at 90. Additionally,

16

Germany’s Federal Cartel Office and the United Kingdom’s Competition and Markets Authority

17

have made the same distinction. Id. This “public recognition” of the Social Network Market

18

reinforces the Court’s conclusion that Consumers have adequately alleged the Social Network

19

Market. Brown Shoe, 370 U.S. at 325.

20

FTC v. Facebook, Inc., No. 20-CV-3590-JEB, 2021 WL 2643627 (D.D.C. June 28, 2021),

21

confirms that Consumers have adequately alleged the Social Network Market. In FTC v.

22

Facebook, the Federal Trade Commission (“FTC”) alleged that Facebook possessed “monopoly

23

power in the market for Personal Social Networking Services.” 2021 WL 2643627, at *7. The

24

FTC’s definition of “Personal Social Networking Services” or “PSN services” largely resembled

25

Consumers’ definition of “social network services.” Id. at *10. Specifically, the FTC defined

26

“PSN Services” as “online services that enable and are used by people to maintain personal

27

relationships and share experiences with friends, family, and other personal connections in a

28

16
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

shared social space.” Id. Additionally, like Consumers, the FTC alleged that PSN services have

2

“[t]hree key elements”: (1) they “are built on a social graph”; (2) they “include features that many

3

users regularly employ to interact with personal connections”; and (3) they “include features that

4

allow users to find and connect with other users.” Id.

5

The FTC’s complaint also “explain[ed] why four different kinds of arguably comparable

6

online services are not ‘reasonably interchangeable’ with PSN services.” Id. First, the FTC

7

alleged that PSN services are distinct from professional networking services like LinkedIn, which

8

are “designed for and used primarily by professionals for sharing professional content.” Id.

9

Unlike PSN services, the FTC explained, professional networking services are not used to
“maintain personal relationships and share experiences with friends, family, and other personal

11

United States District Court
Northern District of California

10

connections.” Id. Second, the FTC alleged that PSN services are distinct from “interest-based”

12

networking services that have a specific focus, like physical exercise. Id. Third, the FTC alleged

13

that PSN services are distinct from “services that allow for consuming and sharing video or audio

14

content, such as YouTube, Spotify, Netflix, or Hulu.” Id. “[U]sers of such services,” the FTC

15

explained, “mostly consume such content passively or share content created by others (rather than

16

content they have created), and such sharing, where it occurs, is not to the user’s network of

17

personal connections but rather to a general and wide audience of unknown users.” Id. Finally,

18

the FTC alleged that PSN services are distinct from “mobile messaging services,” which “lack a

19

‘shared social space’ for interaction” and “do not employ a social graph to facilitate users’ finding

20

and ‘friending’ other users.” Id.

21

Taking these allegations as a whole, the court concluded that the FTC had adequately

22

alleged a market for PSN services. Id. at *11–12. Although Consumers do not address “interest-

23

based” network services, Consumers specifically address every other type of online service

24

addressed by the FTC and provide a detailed explanation for why each service is not reasonably

25

interchangeable with a social network service. Accordingly, the FTC v. Facebook court’s

26

conclusion that the FTC adequately alleged a market for PSN services confirms that Consumers

27

have adequately alleged the Social Network Market.

28

17
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

Facebook offers several meritless arguments. First, Facebook argues that Consumers “fail

2

to make any factual allegations about ‘cross-elasticity of demand.’” Mot. at 17. However, the

3

United States Supreme Court has explained that a plaintiff may establish the “outer boundaries of

4

a product market” with reference to “the reasonable interchangeability of use or the cross-

5

elasticity of demand between the product itself and substitutes for it.” Brown Shoe, 370 U.S. at

6

325 (emphasis added). Thus, there is “no authority . . . supporting Facebook’s argument that

7

Plaintiff must plead specific facts regarding the price or non-price terms under which . . . users

8

would switch (if ever) to alternatives.” FTC v. Facebook, 2021 WL 2643627, at *11 (emphasis in

9

original). Indeed, all that is required at the pleading stage is a qualitative description of the
product market which shows that “certain ‘factors’ of both the service at issue and its potential

11

United States District Court
Northern District of California

10

substitutes—e.g., their ‘price, use[,] and qualities’—render them not ‘reasonably interchangeable’

12

in the eyes of users.” Id.; see also Datel Holdings, 712 F. Supp. 2d at 997 (denying motion to

13

dismiss where complaint described the “distinct” qualities of the product and explained why

14

products without those qualities were not “reasonably interchangeable”). Accordingly,

15

Consumers’ failure to address the cross-elasticity of demand is irrelevant.

16

Facebook also contends that Consumers “fail[] to provide a plausible basis for identifying

17

which firms provide products that consumers consider acceptable substitutes, and which do not.”

18

Mot. at 16. According to Facebook, the Social Network Market is implausible because it excludes

19

the “countless gaming, news, messaging, and other apps” that allow users to “kill time.” Id.

20

However, even leaving FTC v. Facebook aside, Facebook ignores the many cases in which

21

courts have found that entertainment markets may contain multiple distinct submarkets. Indeed, in

22

Kickflip, Inc. v. Facebook, Inc., 999 F. Supp. 2d 677, 687 (D. Del. 2013), the court held that the

23

plaintiff adequately alleged a market for “social-game networks.” According to the plaintiff, that

24

market included “Facebook, MySpace, Google+, and other social networks that offer social games

25

to users” and did not include “other platforms that offer games through websites, mobile devices,

26

or stores selling games.” Id. The plaintiff explained that, in contrast to these other gaming

27

platforms, social network games “allow interactions between players who are not directly

28

18
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

connected to a console, are less elaborate and expensive, derive revenue primarily from

2

advertising or in-game purchases, leverage an existing social network, and have a large user base.”

3

Id. The court held that these allegations were “sufficient to survive Facebook’s motion to

4

dismiss.” Id. at 687–88. Not only does Kickflip contradict Facebook’s argument that the relevant

5

market must include all online services that allow individuals to “kill time,” it directly supports

6

Consumers’ argument that social network services are distinct from other types of online services.

7

Similarly, several courts have found that the market for video games contains multiple
distinct submarkets. For example, in Epic Games v. Apple, the court rejected Apple’s argument

9

that the relevant product market was the “market for all digital video games.” 2021 WL 4128925,

10

at *1. Specifically, after a bench trial, the court found that although there was “some competition

11

United States District Court
Northern District of California

8

amongst the players in the general video game market,” this general competition was not

12

“sufficient for purposes of defining a relevant product market.” Id. at *55. Instead, the evidence

13

showed that the “wider video game market . . . . include[d] at least four distinct submarkets”—(1)

14

“mobile gaming”; (2) “PC gaming”; (3) “console gaming”; and (4) “cloud-based game streaming.”

15

See id. at *47–55. Similarly, in Datel Holdings, the court found plausible the plaintiff’s allegation

16

that Microsoft’s Xbox 360 and Sony’s Playstation 3 were not in the same product market as

17

Nintendo’s Wii. 712 F. Supp. 2d at 982, 997.

18

Thus, Facebook’s suggestion that all “gaming, news, messaging, and other apps” must be

19

included in the same market is implausible. Although it is not Facebook’s burden to establish the

20

relevant product market, Facebook’s failure to suggest a plausible alternative market supports

21

Consumers’ alleged market. Moreover, Facebook’s suggestion that social network services are

22

reasonably interchangeable with all online services that can be used to “kill time” directly

23

contradicts Facebook’s statement that its social network service has “achieved a level of

24

engagement unlike anything else on the web.” ECF No. 97-6 at 1 (emphasis added).

25

Finally, Facebook summarily asserts that Consumers have failed to explain why Twitter,

26

Snapchat, LinkedIn, and TikTok are not substitutes for social network services. Mot. at 16. To

27

the contrary, Consumers explain precisely why social network services, LinkedIn, and TikTok are

28

19
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

not “reasonably interchangeable by consumers for the same purposes.” See Microsoft, 253 F.3d at

2

51–52. Given Consumers’ detailed allegations, Facebook’s conclusory argument has no merit.

3

Although the Court agrees that Consumers have not explained why Twitter and Snapchat

4

are not social network services, Facebook does not explain why this omission is fatal to

5

Consumers’ complaint. As discussed, the Ninth Circuit has held that “what constitutes a relevant

6

market is a factual determination for the jury.” Eastman Kodak, 125 F.3d at 1203. Indeed, “the

7

question of whether the market should include other products is better resolved at the summary

8

judgment stage,” Datel Holdings, 712 F. Supp. 2d at 997, and a plaintiff is “not required to

9

identify every alleged competitor in its pleadings,” FTC v. Facebook, 2021 WL 2643627, at *13.
Thus, because Consumers “consider and reject multiple conceivably interchangeable substitutes in

11

United States District Court
Northern District of California

10

great factual detail,” Consumers have no obligation to address every possible substitute.

12

RealPage, Inc. v. Yardi Systems, Inc., 852 F. Supp. 2d 1215, 1225 (C.D. Cal. 2012).

13

Thus, Consumers have adequately alleged the Social Network Market.

14
15

Consumers Adequately Allege the Social Media Market
Consumers allege that the “Social Network Market is a distinct part or sub-part of the

16

Social Media Market.” CC ¶ 56. Below, the Court finds that Consumers adequately allege the

17

Social Media Market.

18

As discussed, social media services are online services that enable users “to distribute

19

various forms of media—such as text messages, photos, videos, and music—to other users of the

20

same application.” Id. ¶ 72. Additionally, instead of “charg[ing] users a monetary price for

21

access,” social media services “subject the users to ads . . . or require the users to give up some

22

form of limited data.” Id. ¶ 76.

23

Consumers allege that social media services are not “reasonably interchangeable” with

24

other online services and that there is a distinct Social Media Market. Id. ¶¶ 72–81. In addition to

25

Facebook, the largest companies in this market are Twitter, TikTok, LinkedIn, Snapchat, and

26

YouTube. Id. ¶¶ 36–37. Facebook contends that Consumers’ allegations are inadequate because

27

Consumers have failed to identify characteristics that differentiate social media services from

28

20
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

other online services and have failed to provide details about the cross-elasticity of demand. Mot.

2

at 17–18. For the reasons below, the Court rejects Facebook’s contentions.
To plead a product market based on “reasonable interchangeability,” a plaintiff must allege

3
4

details about a product’s “‘price, use and qualities’” and explain why products without those

5

characteristics are not reasonably interchangeable. See Webkinz, 695 F. Supp. 2d at 994 (quoting

6

Oracle, 331 F. Supp. 2d at 1131). In other words, the plaintiff must allege that other types of

7

products are not “reasonably interchangeable by consumers for the same purposes.” Microsoft,

8

253 F.3d at 51–52. For example, allegations that a product has “a distinct core functionality,”

9

“distinct perceptions by consumers,” “distinct customer targeting by manufacturers,” “distinct
analyses by industry analysts,” and “distinct pricing and pricing patterns” are sufficient to

11

United States District Court
Northern District of California

10

establish that there is a distinct market for that product. See Datel Holdings, 712 F. Supp. 2d at

12

997. Additionally, statements by companies explaining which products they see as competitors

13

for their own products are highly probative, see Epic, 2021 WL 4128925, at *51, as is “industry or

14

public recognition of the” product market “as a separate economic entity,” Brown Shoe, 370 U.S.

15

at 325.

16

As an initial matter, Facebook executives have recognized that Facebook primarily

17

competes with other social media services. For example, Consumers highlight a 2012 presentation

18

by Facebook COO Sheryl Sandberg which directly compared Facebook to other “social media”

19

companies and stated that Facebook was “95% of all social media.” CC ¶ 77. This statement

20

strongly supports a conclusion that there is a distinct Social Media Market. See Epic, 2021 WL

21

4128925, at *51.

22

Moreover, Consumers explain in detail why social media services and other online services

23

which facilitate the consumption of media are not “reasonably interchangeable by consumers for

24

the same purposes.” Microsoft, 253 F.3d at 51–52. First, Consumers explain that social media

25

services are not reasonably interchangeable with online entertainment services like Hulu.

26

Specifically, online entertainment services like Hulu limit users to viewing media provided by the

27

services themselves and do not allow users to communicate with each other. See CC ¶ 75. Thus,

28

21
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

such services do not facilitate the distribution of media between “users of the same application,”

2

id. ¶ 72, let alone “communication and sharing content among groups of friends,” House Report at

3

91. Moreover, whereas social media services make money by showing users advertisements,

4

online entertainment services like Hulu typically “collect from users a per-use monetary fee or a

5

regularly-occurring (monthly or annual) fee.” CC ¶ 76. Accordingly, because social media

6

services and online entertainment services have both “distinct core functionalit[ies]” and “distinct

7

pricing,” they are not “reasonably interchangeable.” See Datel Holdings, 712 F. Supp. 2d at 997.

8
9

Second, Consumers explain that social media services are distinct from search engines like
Google, Yahoo, and Bing, which do not host their own media and do not allow for
communications between users. CC ¶ 73. Thus, social media services and search engines are not

11

United States District Court
Northern District of California

10

“reasonably interchangeable by consumers for the same purposes.” Microsoft, 253 F.3d at 51–52.

12

Finally, Consumers point out that there has been “industry or public recognition” of the

13

Social Media Market. See Olin, 986 F.2d at 1299. For example, the U.S. House of

14

Representatives found during its recent investigation of competition in digital markets that social

15

media services have the distinct purpose of “facilitat[ing] the distribution and consumption of

16

content.” House Report at 91. Based on this finding, the House concluded that the “Social Media

17

Market” is distinct from markets for other online services. House Report at 89. This “public

18

recognition” of the Social Media Market reinforces the Court’s conclusion that Consumers have

19

adequately alleged the Social Media Market. See Brown Shoe, 370 U.S. at 325.

20

Epic Games v. Apple confirms that Consumers have adequately alleged a distinct Social

21

Media Market. The Epic court found after a bench trial that, for at least two reasons, the “wider

22

video gaming market” contained “distinct submarkets” for “mobile gaming,” “PC gaming,”

23

“console gaming,” and “cloud-based game streaming.” Epic, 2021 WL 4128925, at *47–55.

24

First, the court found that each type of gaming system appealed to a specific demographic group

25

interested in playing a specific type of game. For example, whereas mobile gaming appealed to

26

“women gamers of all ages . . . and gen-x male gamers” with a “focus and interest on casual

27

games,” console gaming appealed to “millennial male gamers . . . with an interest in playing action

28

22
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

games.” Id. at *49, *52. Second, the court found that each type of gaming system had a distinct

2

pricing model. For example, whereas the “overwhelming majority of gaming revenue in mobile

3

gaming derives from free-to-play games,” “the gaming revenue generated by console games . . .

4

derived overwhelmingly from pay-to-play or buy-to-play games.” Id. at *49, *52.

5

The distinctions between gaming systems identified in Epic mirror the distinctions between

6

social media services and online entertainment services identified by Consumers in the instant

7

case. First, like the different gaming systems described in Epic, social media services and online

8

entertainment services offer fundamentally different types of content and thus appeal to different

9

types of users. See Epic, 2021 WL 4128925, at *49, 52; CC ¶ 72, 74–75. Specifically, whereas
social media services provide content created by users, online entertainment services provide

11

United States District Court
Northern District of California

10

content created by the service itself. CC ¶ 75. Second, like the different gaming systems

12

described in Epic, social media services and online entertainment services have distinct pricing

13

models. See Epic, 2021 WL 4128925, at *49, 52; CC ¶ 76. Whereas social media services earn

14

revenue by allowing users to access the services for free and then “subject[ing] the users to ads,”

15

online entertainment services “collect from users a per-use monetary fee or a regularly-occurring

16

(monthly or annual) fee.” CC ¶ 76. Accordingly, like the different gaming systems in Epic, social

17

media services and online entertainment services are not reasonably interchangeable.

18

Facebook’s arguments to the contrary are unconvincing. Facebook argues that Consumers

19

“fail to define the contours of the ‘social media’ market by reference to cross-elasticity of

20

demand.” Mot. at 18. However, as discussed, the United States Supreme Court has explained that

21

a plaintiff may establish the “outer boundaries of a product market” with reference to “the

22

reasonable interchangeability of use or the cross-elasticity of demand between the product itself

23

and substitutes for it.” Brown Shoe, 370 U.S. at 325 (emphasis added). Indeed, there is “no

24

authority . . . supporting Facebook’s argument that Plaintiff must plead specific facts regarding the

25

price or non-price terms under which . . . users would switch (if ever) to alternatives.” FTC v.

26

Facebook, 2021 WL 2643627, at *11 (emphasis in original). Accordingly, Consumers’ failure to

27

address the cross-elasticity of demand is irrelevant at this stage.

28

23
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

Additionally, Facebook contends that Consumers “fail to allege any plausible basis for

2

excluding competitors that provide other forms of online entertainment since they, like Facebook,

3

compete for user attention.” Mot. at 18. To the contrary, Consumers specifically allege that social

4

media services are distinct from online entertainment services because social media services and

5

online entertainment services have “distinct core functionalit[ies]” and “distinct pricing and

6

pricing patterns.” Datel Holdings, 712 F. Supp. 2d at 997. Facebook’s conclusory argument

7

regarding online entertainment fails to address these specific allegations.

8

Moreover, as discussed, Facebook’s argument that all online services which provide
“entertainment” are in the same market is implausible. For example, multiple courts have found

10

that, although all video game systems provide users with “entertainment,” the “wider video game

11

United States District Court
Northern District of California

9

market” contains multiple distinct submarkets. See Epic, 2021 WL 4128925, at *55; Datel

12

Holdings, 712 F. Supp. 2d at 996–97 (holding that the plaintiff plausibly alleged that Microsoft’s

13

Xbox 360 is not in the same market as Nintendo’s Wii); Kickflip, 999 F. Supp. 2d at 687 (holding

14

that the plaintiff plausibly alleged a market for “social-game networks”).

15

Finally, Facebook contends that Consumers unreasonably include within the market

16

applications “that offer in-app distribution (such as WhatsApp and iMessage) while excluding

17

competitors that allow for the distribution of media outside their own applications,” such as SMS

18

messaging and email. Mot. at 17–18. Although the Court agrees with Facebook that Consumers

19

have not made clear why SMS messaging and email are not “reasonably interchangeable” with

20

social media services, Facebook fails to explain why this is fatal to Consumers’ complaint. As

21

discussed, the Ninth Circuit has held that “what constitutes a relevant market is a factual

22

determination for the jury.” Eastman Kodak, 125 F.3d at 1203. Indeed, “the question of whether

23

the market should include other products is better resolved at the summary judgment stage,” Datel

24

Holdings, 712 F. Supp. 2d at 997, and a plaintiff is “not required to identify every alleged

25

competitor in its pleadings.” FTC v. Facebook, Inc., 2021 WL 2643627, at *13. Thus, because

26

Consumers “consider and reject multiple conceivably interchangeable substitutes in great factual

27

detail,” Consumers have no obligation to address every possible substitute. RealPage, 852 F.

28

24
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1
2
3

Supp. 2d at 1225.
Thus, Consumers have adequately alleged the Social Media Market.
Consumers Adequately Allege that Facebook Has Market Power in the
Social Media and Social Network Markets

4
As discussed, the Ninth Circuit has held that a plaintiff establish monopoly power at the
5
6

pleading stage by alleging that: (1) the defendant’s share of the relevant market exceeds 65%,
Eastman Kodak, 125 F.3d at 1206, and (2) the relevant market has “significant barriers to entry,”

7
Rebel Oil, 51 F.3d at 1434. Facebook argues that, even if Consumers have adequately alleged the
8
Social Network and Social Media Markets, Consumers have not adequately alleged market power
9
10

United States District Court
Northern District of California

11

in either market. Specifically, Facebook argues that, “[i]n both market,” Consumers “claim
without support that Facebook’s ‘share’ exceeds 65%.” Mot. at 18–19. Facebook also argues that
Consumers’ allegations about barriers to entry are insufficient. Id. at 19. The Court addresses

12
these arguments in turn.
13
14
15
16
17

Consumers Adequately Allege that Facebook’s Shares of the Social
Media Market and the Social Network Market Are Greater than
65%
Consumers allege that Facebook’s share of the Social Media Market is between 80% and
95% and that Facebook’s share of the Social Network Market is close to 100%. CC ¶¶ 262, 286.
Facebook contends that Consumers’ market share allegations are implausible because Consumers

18
do not explain how they calculated the market share and because the alleged markets include other
19
large companies. Mot. at 18–19. For the reasons below, the Court rejects these contentions.
20
Consumers identify two metrics for measuring market share in the Social Media Market:
21
(1) time spent on social media applications; and (2) advertising revenue. Under either metric,
22
23
24

Consumers allege, Facebook’s share of the Social Media Market exceeds 65%.
First, Consumers allege that because “more than 80% of the time that consumers in the
United States spend using social media is spent on Facebook and Instagram,” Facebook’s share of

25
the Social Media Market exceeds 80%. CC ¶ 286. Consumers present several pieces of evidence
26

which directly support this allegation. In a 2012 presentation to a “large telecommunications

27
28

25
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

firm,” Facebook COO Sheryl Sandberg asserted that “Facebook is now 95% of all social media in

2

the US” in terms of “monthly minutes of use.” See CC ¶ 77; House Report at 138. The same

3

year, a report by the data analytics firm Comscore, on which Facebook previously relied for

4

market research, found that Facebook was the “third largest web property in the world . . . and

5

accounted for approximately 3 in every 4 minutes spent on social networking sites and 1 in every

6

7 minutes spent online around the world.” See House Report at 144; CC ¶ 159. Finally, a 2018

7

Facebook report showed that, even excluding Instagram, individuals spent more time on Facebook

8

than on Snapchat and Twitter combined. See CC ¶ 78; House Report at 137–38.

9

Second, Consumers allege that, because social media services “generate virtually all of
their value to shareholders in the form of advertising revenue,” advertising revenue is an effective

11

United States District Court
Northern District of California

10

metric for measuring market share. CC ¶ 79. Using that metric, “Facebook (including Instagram)

12

has over 85% of the U.S. market, with the second-place competitor, Twitter, claiming less than

13

3.5% market share.” Id. (emphasis in original). Consumers do not describe Facebook’s

14

competitors’ revenue in detail, but Consumers highlight that in 2018, “Facebook earned over $55

15

billion in revenue, almost completely from selling targeted advertising,” id. ¶ 147, and that in

16

2019, “Facebook collected $70.7 billion in revenue almost entirely from allowing companies to

17

serve ads to its users,” id. ¶ 311.

18

Although Consumers do not calculate Facebook’s precise share of the Social Network

19

Market, Consumers allege that Facebook is the only significant social network service and that its

20

share of the Social Network Market is close to 100%. Id. ¶ 68. Specifically, Consumers allege

21

that Facebook is the “undisputed king of the social network market” and that other social network

22

services, such as Diaspora, Ello, Vero, Clubhouse, and Reddit “constitute ‘only a very small drop

23

in the ocean compared to Facebook.’” Id. (internal citation omitted). Additionally, Consumers

24

allege that, because the Social Network Market is a “distinct part or sub-part of the Social Media

25

Market,” Facebook’s share of the Social Network Market necessarily is “higher than its share in

26

the Social Media Market.” See id. ¶¶ 56, 71, 262. The Court agrees with Consumers that, because

27

the Social Network Market is a submarket of the Social Media Market, it follows that Facebook’s

28

26
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

share of the Social Network Market must be greater than Facebook’s share of the Social Media

2

Market.

3

Consumers provide additional evidence to bolster their market share calculations. For
example, Consumers point out that “Facebook’s products include three of the seven most popular

5

mobile apps, measured by monthly active persons, reach, and percentage of daily and monthly

6

active persons.” Id. ¶ 78. Indeed, as of December 2019, the mobile application for Facebook’s

7

social network service “had the third highest reach of all mobile apps” in the United States,

8

“reaching 74% of smartphone users,” and the mobile application for Facebook Messenger “had the

9

fourth highest reach” of all mobile apps, “reaching 54.1% of U.S. smartphone users.” House

10

Report at 137. By contrast, Snapchat’s mobile application reached only 31.4% of smartphone

11

United States District Court
Northern District of California

4

users in the United States. Id. Similarly, Facebook’s 2012 presentation to investors showed that

12

Facebook had “97% penetration [with] 25-34 year olds” in the United States. ECF No. 97-6 at 2.

13

Under Ninth Circuit law, Consumers’ allegations are more than sufficient. Indeed, because

14

the Ninth Circuit has held that a plaintiff need not plead market share with specificity, courts

15

frequently have held that a rough estimate of the defendant’s market share is sufficient at the

16

pleading stage. See Newcal Indus., 513 F.3d at 1045 (“There is no requirement that these

17

elements of the antitrust claim be pled with specificity.”). For example, in United Energy

18

Trading, LLC v. Pacific Gas & Electric Co., 200 F. Supp. 3d 1012 (N.D. Cal. 2016), the plaintiff

19

brought a Sherman Act claim against Pacific Gas & Electric Company (“PG&E”) and offered only

20

a single allegation regarding market share: “PG&E currently holds between 70-90% of the natural

21

gas commodity market.” Second Amended Complaint ¶ 2, United Energy Trading, LLC v. Pacific

22

Gas & Electric Co., No. 15-CV-02383-RS (N.D. Cal. May 13, 2016), ECF No. 89. In an order

23

denying PG&E’s motion to dismiss, the court held that this allegation was sufficient. United

24

Energy Trading, 200 F. Supp. 3d at 1020.

25

Similarly, in Teradata Corporation v. SAP SE, No. 18-CV-03670-WHO, 2018 WL

26

6528009 (N.D. Cal. Dec. 12, 2018), the plaintiff brought a Sherman Act claim against SAP and

27

alleged only that SAP “possesses a market share that ranges, on information and belief, from 60%

28

27
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

to 90%.” First Amended Complaint ¶ 67, Teradata Corporation v. SAP SE, No. 18-CV-03670-

2

WHO (N.D. Cal. Jul. 11, 2018), ECF No. 24. In an order denying SAP’s motion to dismiss, the

3

court stated that this allegation “suffice[d] to plead SAP’s market power.” Teradata, 2018 WL

4

6528009, at *15.

5

Finally, in Datel Holdings, the plaintiff brought a Sherman Act claim against Microsoft

6

and alleged only that “Microsoft’s share of the Multiplayer Online Dedicated Gaming Systems

7

market was approximately 66% as of July 2009.” Complaint ¶ 44, Datel Holdings Ltd. v.

8

Microsoft Corp., No. 09-CV-05535-EDL (N.D. Cal. Nov. 20, 2009), ECF No. 1. In an order

9

denying Microsoft’s motion to dismiss, the court held that this allegation was sufficient. Datel

10

United States District Court
Northern District of California

11

Holdings, 712 F. Supp. 2d at 997–98.
Consumers’ allegations are far more detailed than the allegations deemed sufficient in

12

United Energy Trading, Teradata, and Datel Holdings. The plaintiffs in United Energy Trading,

13

Teradata, and Datel Holdings did not provide any explanation for how they calculated market

14

share, let alone any specific facts that could support such calculations. See United Energy

15

Trading, 200 F. Supp. 3d at 1020; Teradata, 2018 WL 6528009, at *15; Datel Holdings, 712 F.

16

Supp. 2d at 997–98. By contrast, Consumers provide two alternate methods for calculating

17

Facebook’s share of the Social Media Market and cite detailed facts, including data produced by

18

Facebook itself, which support Consumers’ calculations. Additionally, Consumers have provided

19

a plausible explanation for why Facebook’s share of the Social Network Market must be even

20

greater. Thus, Consumers’ allegations regarding Facebook’s shares of the Social Media and

21

Social Network Markets are sufficient.

22

FTC v. Facebook confirms that Consumers’ allegations are sufficient. In FTC v.

23

Facebook, the FTC alleged that Facebook’s share of the market for “PSN Services” exceeded

24

65%. No. 20-CV-3590-JEB, 2022 WL 103308, at *6 (D.D.C. Jan. 11, 2022). The FTC offered

25

three metrics to support that calculation, including “Facebook’s share of users’ time spent on such

26

services.” Id. Like Consumers, the FTC cited “Facebook’s internal presentations” which assessed

27

Facebook’s performance using “time spent per month.” Id. at *7. According to the court, these

28

28
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

allegations plausibly established that “Facebook has maintained a dominant market share during

2

the relevant time period.” Id. at *7. Given Consumers’ similar allegations, the Court reaches the

3

same conclusion in the instant case.

4

Facebook’s arguments to the contrary are unconvincing. According to Facebook,

5

Consumers “claim without support that Facebook’s ‘share’ exceeds 65% . . . but they make no

6

allegations about what this number consists of.” Mot. at 18. Facebook adds that “[m]erely

7

reciting a percentage without alleging to what it refers—let alone how it was or could be

8

calculated—is conclusory and is not creditable.” Id. As an initial matter, there is no authority

9

which suggests that antitrust plaintiffs must explain in their complaints how they calculated
market share. Indeed, as discussed, courts have repeatedly held that a rough estimation of the

11

United States District Court
Northern District of California

10

defendant’s market share, with no explanation of how that estimation was made, is sufficient at the

12

pleading stage. See United Energy Trading, 200 F. Supp. 3d at 1020; Teradata, 2018 WL

13

6528009, at *15; Datel Holdings, 712 F. Supp. 2d at 997–98. Regardless, Facebook’s argument

14

fails on its own terms because Consumers provide two independent methods for measuring

15

Facebook’s market share and explain why each method results in a market share that exceeds

16

80%. CC ¶¶ 71, 77–80, 262, 286.

17

Moreover, although Facebook relies heavily on Korea Kumho Petrochemical v. Flexsys

18

America LP, No. C07-01057 MJJ, 2008 WL 686834 (N.D. Cal. Mar. 11, 2008), and

19

Rheumatology Diagnostics Lab., Inc. v. Aetna, Inc., No. 12-CV-05847-JST, 2013 WL 3242245

20

(N.D. Cal. June 25, 2013), those cases do not support Facebook’s argument. Indeed, in Korea

21

Kumho, 2008 WL 686834, at *9, the court emphasized that a plaintiff “need not necessarily

22

quantify . . . market share with precision.” However, because the plaintiff’s sole allegation related

23

to market power was that the defendant had achieved “dominance” in the relevant market, the

24

court dismissed the plaintiff’s claim. Id.

25

By contrast here, Consumers have alleged that Facebook’s share of the Social Media

26

Market and share of the Social Network Market fall within relatively narrow ranges and have

27

provided detailed explanations for how they calculated those ranges.

28

29
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

In turn, in Rheumatology Diagnostics, 2013 WL 3242245, at *1, *14, although the plaintiff

2

alleged that the defendant had a “seventy percent market share of the outpatient testing market in

3

Northern California,” this allegation provided no information about the defendant’s share of the

4

“relevant markets identified in the Complaint”: the “Routine Clinical Laboratory Testing market”;

5

the “Anatomic Pathology Laboratory Testing market”; the “Specialty Rheumatologic Laboratory

6

Testing market”; the “Advanced Lipid Testing market”; and the “Specialty Breast Pathology

7

Testing market.” Accordingly, the court concluded that the plaintiff had failed to allege any facts

8

regarding defendant’s share of those “relevant markets.” Id. at *14. That conclusion has no

9

bearing on the instant case.

10

Finally, the Court rejects Facebook’s argument that Consumers’ market share allegations

United States District Court
Northern District of California

11

are implausible because Consumers fail to account for “large rivals like Apple’s iMessage,

12

Twitter, YouTube, Snapchat, and TikTok.” Mot. at 18. As discussed, Consumers allege that, in

13

2018, individuals spent more time on Facebook than on Snapchat and Twitter combined. See CC

14

¶ 78; House Report at 137–38. Similarly, Facebook has provided evidence that, as of December

15

2019, the mobile application for Facebook’s social networking service had more than double the

16

number of users as Snapchat. House Report at 137. Additionally, Consumers allege that Twitter

17

is the second largest social media service and that, measuring market share using advertising

18

revenue, Twitter has “less than 3.5% market share.” CC ¶ 80 (emphasis in original).

19
20
21
22
23
24

Accordingly, Consumers have adequately alleged that Facebook’s shares of the Social
Media Market and Social Network Market exceed 65%.
Consumers Plausibly Allege that the Social Network and Social
Media Markets Have Significant Barriers to Entry
As discussed, “[e]ntry barriers are ‘additional long-run costs that were not incurred by
incumbent firms but must be incurred by new entrants,’ or ‘factors in the market that deter entry
while permitting incumbent firms to earn monopoly returns.’” Rebel Oil, 51 F.3d at 1439 (citing

25
Los Angeles Land, 6 F.3d at 1427–28).
26
Consumers allege in detail that the Social Network and Social Media Markets have at least
27
28

30
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

two classic barriers to entry: (1) network effects and (2) switching costs. In response, Facebook

2

offers one conclusory sentence: “naked assertions about barriers to entry do not suffice.” Mot. at

3

19. However, as Mr. Zuckerberg has explained, “there are network effects around social products

4

and a finite number of different social mechanics to invent.” See CC ¶ 96. Thus, “[o]nce someone

5

wins at a specific mechanic, it’s difficult for others to supplant them without doing something

6

different.” See House Report at 143. Given this and other contemporaneous statements by

7

Facebook about the power of network effects, Facebook’s conclusory argument in the instant

8

motion has no merit. Regardless, the Court discusses Consumers’ allegations below.
First, social network services and social media services have strong network effects, which

10

means that the value of a service increases as the number of users increases. See CC ¶ 83; see also

11

United States District Court
Northern District of California

9

United States v. Microsoft Corp., 84 F. Supp. 2d 9, 20 (D.D.C. 1999) (“A positive network effect

12

is a phenomenon by which the attractiveness of a product increases with the number of people

13

using it.”). A direct network effect is present when a “service becomes more valuable to users as

14

additional others use the . . . service.” CC ¶ 83. Because the goal of a social media service is to

15

provide users with content created by other users of the service, social media services exhibit

16

strong direct network effects. Id. ¶ 84. In turn, because the goal and design of a social network

17

service is to enable users to form social connections with other users of the service and to facilitate

18

users’ sharing of information with other users of the service, social network services exhibit even

19

stronger direct network effects. Id. Indeed, in an October 2018 presentation, an economist

20

working at Facebook “recognized that the network effects of Facebook and its products are ‘very

21

strong.’” Id. ¶ 96.

22

An indirect network effect is present when a service becomes more valuable to third parties

23

as the service’s user base grows. Id. ¶ 83. Because social media services and social network

24

services make money by showing advertisements to users, both types of services exhibit strong

25

indirect network effects. Id. ¶ 86.

26

These network effects create high barriers to entry in the Social Network and Social Media

27

Markets because, even if a new service offers higher quality features than incumbent services, the

28

31
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

new service is inherently less valuable because it has a smaller user base than incumbent services.

2

Id. ¶ 87.

3

As noted, Facebook CEO Mark Zuckerberg has displayed a keen understanding of the
barriers to entry created by network effects. In 2012, when Facebook was in the process of

5

acquiring Instagram, Mr. Zuckerberg explained to Facebook CFO David Ebersman that “there are

6

network effects around social products and a finite number of different social mechanics to

7

invent.” See Id. ¶ 96; House Report at 152. “Once someone wins at a specific mechanic,” Mr.

8

Zuckerberg stated, “it’s difficult for others to supplant them without doing something different.”

9

See House Report at 143. Other Facebook employees have described Facebook’s network effects

10

as a “flywheel” and have stated that “[n]etwork effects make it very difficult to compete with us.”

11

United States District Court
Northern District of California

4

CC ¶ 96 (internal citation omitted). Moreover, as the U.S. House of Representatives found during

12

its recent investigation of competition in digital markets, “attracting a critical mass of users is

13

essential to delivering a viable social network, as there is no reason for users to start using a social

14

network if there is no one there with whom they can connect.” House Report at 89.

15

Second, social network services and social media services are characterized by “high

16

switching costs.” CC ¶ 87. The United States Supreme Court has explained that “switching

17

costs” are present when consumers have become “locked in” to a product and thus “will tolerate

18

some level of . . . price increases before changing . . . brands.” See Eastman Kodak Co. v. Image

19

Tech. Servs., Inc., 504 U.S. 451, 476 (1992).

20

Social network services and social media services have high switching costs because,

21

“[o]nce a significant number of users adopt” a social network service or social media service,

22

“they will be reluctant to switch to even a superior competitive alternative because the newer

23

offering will not deliver as much value unless many other users make the switch simultaneously.”

24

CC ¶ 87.

25

As a Facebook executive explained in 2014, “the idea is that after you have invested hours

26

and hours in your friend graph or interest graph or follower graph, you are less likely to leave for a

27

new or different service that offers similar functionality.” House Report at 145. Indeed,

28

32
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

Facebook’s “near monopoly” means that, for most users, “the vast majority of the people with

2

whom they want to exchange feeds are likely on Facebook already.” House Report at 144 n.817.

3

“[T]he switching cost for any one user is therefore enormous.” Id.

4

These switching costs create high barriers to entry because, “[t]o induce a user to adopt a

5

new” service, a new service “must incur not only the expense of building a superior product, but

6

also the added compensation to defray a user’s cost of switching.” CC ¶ 88. Accordingly, new

7

social network and social media services must incur “additional long-run costs that were not

8

incurred by incumbent firms.” Rebel Oil, 51 F.3d at 1439.

9

Because of the powerful network effects and high switching costs present in social
networking services, the Social Network Market has uniquely high barriers. Thus, “there are

11

United States District Court
Northern District of California

10

strong tipping points in the social networking market that create competition for the market, rather

12

than competition within the market.” CC ¶ 85 (quoting House Report at 41) (emphasis in

13

original). Facebook itself has recognized that, “with respect to social networks, ‘either everyone

14

uses them, or no-one uses them.’” Id. ¶ 87 (internal citation omitted).

15

FTC v. Facebook confirms that Consumers’ allegations are sufficient. In FTC v.

16

Facebook, the FTC alleged that “Facebook’s dominant position in the U.S. personal social

17

networking market is durable due to significant entry barriers, including direct network effects and

18

high switching costs.” 2022 WL 103308, at *9. Like Consumers, the FTC pointed to Facebook’s

19

statements acknowledging the strong networks effects and high switching costs that are present in

20

social products. Id. After explaining that “network effects and switching costs are commonly

21

recognized types of barriers to entry,” the court concluded that the FTC’s allegations were

22

sufficient. Id. Given Consumers’ detailed allegations, the same is true in the instant case.

23
24
25

Accordingly, Consumers have adequately alleged that Facebook has monopoly power both
in the Social Media Market and in the Social Network Market.
Advertisers Adequately Allege that Facebook Has Monopoly Power in the Social
Advertising Market

26
Advertisers allege that Facebook has monopoly power in the Social Advertising Market.
27
28

33
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

Although Facebook argues that Advertisers have not adequately alleged the Social Advertising

2

Market, Facebook does not dispute that, if Advertisers have adequately alleged the Social

3

Advertising Market, Facebook has monopoly power. See Mot. at 13–18. Thus, the Court need not

4

address whether Advertisers have adequately alleged monopoly power.

5

According to Advertisers, a social advertisement is a distinct type of online advertisement

6

offered by social media services like Facebook. AC ¶¶ 413–14. Specifically, a social

7

advertisement gives companies the option to target a group of users on a social media service

8

“based on a range of targeting criteria—location, age group, gender, hobbies, interests.” Id. ¶ 417.

9

Advertisers allege that, because social advertisements are not “reasonably interchangeable”
with other types of online advertising, there is a distinct Social Advertising Market, which is a

11

United States District Court
Northern District of California

10

“submarket of online advertising.” Id. ¶ 413. Under Advertisers’ definition, Facebook, Twitter,

12

and LinkedIn are the “three largest firms competing in the Social Advertising Market.” Id. ¶ 450.

13

Facebook contends that these allegations are inadequate because markets limited to a single form

14

of advertising are implausible and Advertisers have failed to define the “contours” of the Social

15

Advertising Market. Mot. at 14–15. For the reasons below, the Court rejects Facebook’s

16

contentions.

17

As discussed, a plaintiff may plead a product market based on “reasonable

18

interchangeability” by alleging details about a product’s “‘price, use and qualities’” and explaining

19

why products without those characteristics are not reasonably interchangeable. See Webkinz, 695

20

F. Supp. 2d at 994 (quoting Oracle, 331 F. Supp. 2d at 1131). In other words, the plaintiff must

21

allege that other types of products are not “reasonably interchangeable by consumers for the same

22

purposes.” Microsoft, 253 F.3d at 51–52. For example, allegations that a product has “a distinct

23

core functionality,” “distinct perceptions by consumers,” “distinct customer targeting by

24

manufacturers,” “distinct analyses by industry analysts,” and “distinct pricing and pricing

25

patterns” are sufficient to establish that there is a distinct market for that product. See Datel

26

Holdings, 712 F. Supp. 2d at 997. Additionally, statements by companies explaining which

27

products they see as competitors for their own products are highly probative. See Epic, 2021 WL

28

34
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

4128925, at *51. Furthermore, a plaintiff may plead a distinct submarket by alleging facts which

2

show “industry or public recognition of the submarket as a separate economic entity, the product’s

3

peculiar characteristics and uses, unique production facilities, distinct customers, distinct prices,

4

sensitivity to price changes, and specialized vendors.” Brown Shoe, 370 U.S. at 325.

5

Facebook executives have recognized that social advertising is distinct from other

6

prominent forms of advertising. For example, in October 2012, Facebook COO Sheryl Sandberg

7

stated to investors: “We’re not TV, we’re not search. We are social advertising.” AC ¶ 422.

8

Similarly, in a May 2013 earnings call, Ms. Sandberg stated: “As I said before, the thing about

9

brand advertisers is that they got very used to TV, then they got very used to search, and we are a
third thing.” Id. These statements strongly support a conclusion that there is a distinct Social

11

United States District Court
Northern District of California

10

Advertising Market. See Epic, 2021 WL 4128925, at *51.

12

Moreover, there has been widespread “industry . . . recognition of the” Social Advertising

13

Market as distinct from other markets for online advertising. Brown Shoe, 370 U.S. at 325. For

14

example, the advertising company Outbrain has stated that “social ads” and “search ads” have

15

fundamentally different purposes. AC ¶¶ 416–18. Outbrain stated that, whereas “[s]ocial ads are

16

best for targeting audience segments who may be interested in your products or services,” “search

17

ads” “are great for targeting customers when they are already looking for you.” Id. Additionally,

18

in October 2019, “industry publication Marketing Lands” published an article which “predicted

19

growth in the social media advertising segment as distinguished from search and television

20

advertising.” Id. ¶ 421. This industry recognition supports a conclusion that the Social

21

Advertising Market is a “separate economic entity.” Brown Shoe, 370 U.S. at 325.

22

Additionally, Advertisers explain in detail why social advertisements and other types of

23

online advertisements are not “reasonably interchangeable by consumers for the same purposes.”

24

Microsoft, 253 F.3d at 51–52. First, social advertisements are not reasonably interchangeable with

25

search advertisements. Search advertisements are shown to people who enter certain terms into

26

search engines. AC ¶ 418. Accordingly, these advertisements allow companies to target

27

customers who are “already looking for [the company] (i.e., they search [the] company name or

28

35
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

product”) or who “are searching for a specific product, service, or piece of information.” Id.

2

¶¶ 418, 437. By contrast, social advertisements help a company find customers who are not

3

already looking for the company’s products. Id. Additionally, to buy search advertisements,

4

companies must bid against each other for certain search terms. Id. ¶¶ 438, 442. By contrast,

5

because “social advertisers like Facebook can tailor audiences,” companies who want to purchase

6

social advertisements do not have to bid against each other. Id. ¶ 440. Thus, whereas prices for

7

search advertisements are “sensitive to changes in demand” for particular search terms, the price

8

of a social advertisement “is proportional to the generality of the targeting.” Id. ¶ 439, 442. As a

9

result, “[s]mall businesses that do not generally have the budget to bid on coveted search results”
can negotiate with companies that offer social advertisements to show advertisements to

11

United States District Court
Northern District of California

10

“granularly defined audiences.” Id. ¶ 437. Accordingly, because social advertisements and search

12

advertisements have distinct “uses,” “distinct prices,” and “distinct customers,” Brown Shoe, 370

13

U.S. at 325, they are not “reasonably interchangeable by consumers for the same purposes.”

14

Microsoft, 253 F.3d at 51–52.

15

For similar reasons, social advertisements are not “reasonably interchangeable” with

16

banner advertisements, which are advertisements that appear as rectangular panels at the top of

17

websites. See AC ¶¶ 431–32. Because companies place banner advertisements on pre-determined

18

websites, banner advertisements are most useful for companies who already know what kinds of

19

websites their customers are visiting. Thus, banner advertisements do not enable companies to

20

find new customers in the way that social advertisements do. Id. Additionally, as with search

21

advertisements, companies must bid against each other to place banner advertisements on the most

22

coveted websites, which means that the prices of banner advertisements are “sensitive to changes

23

in demand.” Id. ¶ 439. Thus, because social advertisements and banner advertisements have

24

distinct “uses” and “distinct prices,” Brown Shoe, 370 U.S. at 325, they are not “reasonably

25

interchangeable by consumers for the same purposes.” Microsoft, 253 F.3d at 51–52.

26
27
28

Epic Games v. Apple confirms that Advertisers’ allegations regarding the Social
Advertising Market are sufficient. As discussed, the Epic court found after a bench trial that, for
36
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

at least two reasons, the “wider gaming market” had to be divided into “distinct submarkets” for

2

“mobile gaming,” “PC gaming,” “console gaming,” and “cloud-based game streaming.” 2021 WL

3

4128925, at *47–55. First, the court found that each type of gaming system appealed to a specific

4

demographic group interested in playing a specific type of game. For example, whereas mobile

5

gaming appealed to “women gamers of all ages . . . and gen-x male gamers” with a “focus and

6

interest on casual games,” console gaming appealed to “millennial male gamers . . . with an

7

interest in playing action games.” Id. at *49, *52. Second, the court found that each type of

8

gaming system had a distinct pricing model. For example, whereas the “overwhelming majority

9

of gaming revenue in mobile gaming derives from free-to-play games,” “the gaming revenue
generated by console games . . . derived overwhelmingly from pay-to-play or buy-to-play games.”

11

United States District Court
Northern District of California

10

Id. at *49, *52.

12

The distinctions between gaming systems identified in Epic mirror the distinctions between

13

social advertisements and other types of online advertisements. First, like the different gaming

14

systems described in Epic, social advertisements and other types of online advertisements appeal

15

to different types of customers. See id. Specifically, whereas social advertisements appeal to

16

companies who want to find new customers, other types of online advertisements appeal to

17

companies with established brand or products. AC ¶ 431. Additionally, social advertisements

18

have a particular appeal to small businesses. Id. ¶ 437. Second, like the different gaming systems

19

described in Epic, social advertisements and other types of online advertisements have distinct

20

pricing models. See Epic, 2021 WL 4128925, at *49, *52. Whereas prices for social

21

advertisements determined on a company-by-company basis, prices for search-based

22

advertisements and banner advertisements are “sensitive to changes in demand” for particular

23

search terms and particular websites. AC ¶¶ 439, 442. Thus, like the different gaming systems in

24

Epic, social advertisements and other types of online advertisements are not reasonably

25

interchangeable.

26
27
28

Facebook’s arguments to the contrary are unconvincing. Although Facebook is correct
that Hicks v. PGA Tour, Inc., 897 F.3d 1109 (9th Cir. 2018), requires courts to be skeptical of
37
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

narrowly defined advertising markets, Hicks does not apply where a plaintiff has alleged that two

2

types of advertising have fundamentally different purposes. In Hicks, the plaintiffs alleged that

3

there was a distinct submarket for “‘in-play’ or ‘in-action’ advertising during professional golf

4

tournaments (i.e., between commercial breaks).” 897 F.3d at 1121. The Ninth Circuit concluded

5

that this market was implausible because “[c]ompanies aiming to target professional golf fans”

6

had many other options for advertising and that these forms of advertising were used by the same

7

companies for the same fundamental objective: targeting golf fans. Id. at 1121–22. By contrast,

8

Advertisers allege that social advertising has a fundamentally different purpose from other forms

9

of online advertising: enabling advertisers to reach specific purchasers with specific attributes.
AC ¶¶ 424–30. Advertisers also have explained in detail that this form of advertising appeals to

11

United States District Court
Northern District of California

10

different types of companies. Thus, unlike the plaintiffs in Hicks, Advertisers have established

12

that social advertising has a distinct “use” and “distinct customers.” Brown Shoe, 370 U.S. at 325.

13

Additionally, Facebook contends that “Advertisers do not include factual allegations

14

sufficient to define the contours of the market” and “offer[] no plausible basis for excluding other

15

forms of advertising.” Mot. at 15. Given Facebook’s previous statements recognizing the distinct

16

Social Advertising Market, Facebook’s argument is not credible. As Advertisers point out,

17

Facebook has previously stated to investors: “We’re not TV, we’re not search. We are social

18

advertising.” AC ¶ 422. Moreover, the Ninth Circuit has held that “what constitutes a relevant

19

market is a factual determination for the jury.” Eastman Kodak, 125 F.3d at 1203. Indeed, “the

20

question of whether the market should include other products is better resolved at the summary

21

judgment stage.” Datel Holdings, 712 F. Supp. 2d at 997. Thus, because Advertisers “consider

22

and reject multiple conceivably interchangeable substitutes in great factual detail,” Advertisers

23

have sufficiently defined the contours of the market. RealPage, 852 F. Supp. 2d at 1225.

24

Thus, Advertisers have adequately alleged the Social Advertising Market.

25
26
27
28

B. The Court Denies Facebook’s Motion to Dismiss Consumers’ Data Privacy Claims
Consumers’ first theory of Sherman Act liability is that Facebook acquired and maintained
monopoly power by making false representations to users about Facebook’s data privacy
38
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

practices. Consumers claim that these misrepresentations were “instrumental to Facebook gaining

2

and maintaining market share at the expense of its rivals.” CC ¶¶ 108, 217.

3

According to a leading antitrust treatise, a “monopolist’s misrepresentations encouraging

4

the purchase of its product can fit [the] general test for an exclusionary practice when the impact

5

on rivals is significant.” See Phillip E. Areeda & Herbert Hovenkamp, Antitrust Law ¶ 782b (4th

6

and 5th Eds., 2021 Cum. Supp.) (“Areeda”). In American Professional Testing Service, Inc. v.

7

Harcourt Brace Jovanovich Legal & Professional Publications, Inc., 108 F.3d 1147 (9th Cir.

8

1997), the Ninth Circuit relied on an older version of this treatise and held that, under certain

9

circumstances, a company’s “false and misleading advertising” may “constitute[] exclusionary
conduct” for purposes of the Sherman Act. See id. at 1152 (citing Phillip E. Areeda & Donald F.

11

United States District Court
Northern District of California

10

Turner, Antitrust Law ¶ 737b at 280–81 (3d ed. 1978)). Specifically, to state a Sherman Act claim

12

based on “false and misleading advertising,” a plaintiff must show that the company made

13

representations about its own products or its rivals’ products that “were [1] clearly false, [2]

14

clearly material, [3] clearly likely to induce reasonable reliance, [4] made to buyers without

15

knowledge of the subject matter, [5] continued for prolonged periods, and [6] not readily

16

susceptible of neutralization or other offset by rivals.” Id.1 Although American Professional

17

Testing Service dealt with a claim that the defendant had disparaged its rivals’ products, courts

18

have applied the same test to a claim that the defendant made false statements about its own

19

product. See, e.g., National Ass’n of Pharmaceutical Mfrs., Inc. v. Ayerst Laboratories, 850 F.2d

20

904, 916 (2d Cir. 1988) (applying test to claim that pharmaceutical company misrepresented that

21

its generic product was superior to a name brand product); Genus Lifesciences Inc. v. Lannett Co.,

22

Inc., 378 F. Supp. 3d 823, 841 (N.D. Cal. 2019) (applying test to claim that pharmaceutical

23

company misrepresented that its product was approved by the FDA).

24
25
26
27
28

Courts of Appeals across the country have adopted the same six part test. See National Ass’n of
Pharmaceutical Mfrs., Inc. v. Ayerst Laboratories, 850 F.2d 904, 916 (2d Cir. 1988); Am. Council
of Certified Podiatric Physicians and Surgeons v. Am. Board of Podiatric Surgery, Inc., 323 F.3d
366, 371 (6th Cir. 2003); Covad Communications Co. v. Bell Atlantic Corp., 398 F.3d 666, 674–
75 (D.C. Cir. 2005); and Lenox MacLaren Surgical Corp. v. Medtronic, Inc., 762 F.3d 1114, 1127
(10th Cir. 2014).
39
1

Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

Because fraud is an “essential element” of a Sherman Act claim based on

2

misrepresentations to buyers, a plaintiff that brings such a claim must satisfy the heightened

3

pleading standard set forth by Federal Rule of Civil Procedure 9(b). Vess v. Ciba-Geigy Corp.

4

USA, 317 F.3d 1097, 1103 (9th Cir. 2003). Federal Rule of Civil Procedure 9(b) provides that,

5

“[i]n alleging fraud or mistake, a party must state with particularity the circumstances constituting

6

fraud or mistake.” If any “particular averments of fraud are insufficiently pled under Rule 9(b),”

7

the Court must “‘disregard’ those averments, or ‘strip’ them from the claim.” Vess, 317 F.3d at

8

1105. Then, the Court will “examine the allegations that remain.” Id.

9

Facebook argues that Consumers’ data privacy claims are barred by the statute of
limitations. See Mot. at 10–11. Additionally, Facebook argues that Consumers’ data privacy

11

United States District Court
Northern District of California

10

claims must be dismissed because (1) Consumers have failed to allege fraud with particularity; (2)

12

Consumers have failed to show that Facebook’s alleged misrepresentations were “clearly false”;

13

(3) Consumers have failed to show that Facebook’s alleged misrepresentations were “not readily

14

susceptible of neutralization”; (4) Consumers have failed to show that Facebook’s

15

misrepresentations were “clearly material”; and (5) Consumers have failed to allege causal

16

antitrust injury. See Mot. at 10–11, 19–22. Facebook also argues that Consumers’ claim for

17

injunctive relief is barred by the doctrine of laches. Mot. at 8–9. For the reasons below, the Court

18

rejects these arguments.

19

The Court begins by describing Consumers’ allegation that Facebook obtained and

20

maintained monopoly power by deceiving Consumers about Facebook’s data privacy practices.

21

Because the Court must “disregard” all non-particularized allegations of fraud and “examine the

22

allegations that remain,” the Court then addresses whether Consumers have alleged with sufficient

23

particularity that Facebook made clearly false representations about Facebook’s collection and

24

monetization of user data. Vess, 317 F.3d at 1105. After this threshold inquiry, Court addresses

25

the remainder of Facebook’s arguments in turn.

26

Consumers Allege that Facebook Obtained and Maintained Monopoly Power by
Repeatedly Deceiving Users About Facebook’s Data Privacy Practices

27
28

40
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

According to Consumers, Facebook knew that users placed a premium on privacy and that

2

the best way to attract users was to promise to keep user data private. At the same time, Facebook

3

could not make money without selling user data to third parties. Thus, Consumers allege,

4

Facebook systematically deceived users about the extent to which Facebook was collecting and

5

selling user data. This deception allowed Facebook to beat out companies that were truthful about

6

their user data practices or did not collect and sell user data.

7

The Court begins by describing Consumers’ allegations about the relationship between
privacy and the value of Facebook’s platform. The Court then describes Consumers’ allegations

9

that Facebook deceived users with false statements about Facebook’s privacy practices. Finally,

10

the Court describes Consumers’ allegations that Facebook’s deceptive privacy practices allowed

11

United States District Court
Northern District of California

8

Facebook to obtain and to maintain monopoly power.

12
13

Facebook Knew That Privacy Would Attract Users and Repel Advertisers
According to Consumers, users of social network and social media services always have

14

valued data privacy highly. In a 2004 consumer survey, a majority of such users indicated that

15

privacy was a “really important issue that [they] care about often.” CC ¶ 108. In another study,

16

Facebook users indicated that they cared more about privacy policies than about terrorism. Id.

17

Accordingly, “[p]rivacy practices were a crucial form of competition in the early days of the

18

Social Network and Social Media Markets.” Id. ¶ 102.

19

Indeed, Facebook’s initial success in the Social Network and Social Media Markets arose

20

directly from competitors’ failure to keep users’ data private. Initially, a company called Myspace

21

dominated the Social Network Market. Id. ¶ 102. Founded in 2003, Myspace offered an “open”

22

social network that allowed individuals to sign up with pseudonyms. Id. ¶¶ 102–03. By 2006,

23

Myspace was the most visited website in the United States. Id.

24

However, in 2007, negative headlines began drawing attention to Myspace’s “lax privacy

25

practices.” Id. ¶ 104. Specifically, users and commentators “attributed sexual assaults, suicides,

26

and murders to Myspace,” whose open network “cloaked wrong-doers with relative anonymity.”

27

Id. These controversies provided an opportunity for Myspace’s nascent competitors, which

28

41
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

included Friendster, Orkut, Bebo, Flip.com, and Facebook. Id. ¶ 105.

2

Facebook immediately recognized that it could take market share from Myspace by

3

offering a social network that required users to register with their real names and allowed users to

4

share information only with users designated as “friends.” See id. ¶¶ 44, 106. The original

5

version of Facebook “was closed to all but those users who could validate their own real-world

6

identities, such as by verifying that their identities were legitimate via an e-mail address issued by

7

an organization, such as a university or firm.” Id. ¶ 48.

8
9

These features provided users with “a perceived degree of safety and comfort” that
Myspace did not offer, and users left Myspace for Facebook. Id. (internal citation omitted).
Between 2006 and 2009, Facebook went from having one-tenth the users of Myspace to having

11

United States District Court
Northern District of California

10

more users than Myspace. Id. ¶¶ 107, 122. In 2011, Mr. Zuckerberg explained that Facebook

12

“became the world’s biggest community online” because Facebook made users feel safe with

13

regard to the privacy of their data:

14
15
16
17
18
19

I founded Facebook on the idea that people want to share and connect with people
in their lives, but to do this everyone needs complete control over who they share
with at all times.
This idea has been the core of Facebook since day one. When I built the first
version of Facebook, almost nobody I knew wanted a public page on the internet.
That seemed scary. But as long as they could make their page private, they felt safe
sharing with their friends online. Control was key. With Facebook, for the first
time, people had the tools they needed to do this. That’s how Facebook became the
world’s biggest community online. We made it easy for people to feel comfortable
sharing things about their real lives.

20

See Id. ¶ 109; Mark Zuckerberg, Our Commitment to the Facebook Community, Facebook

21

Newsroom (Nov. 29, 2011), https://bit.ly/3zDEs2A.

22

Additionally, Facebook’s purported privacy features allowed Facebook to obtain valuable

23

data. In a 2008 internal report titled “Facebook Secret Sauce,” Facebook observed that “‘[u]sers

24

will share more information if given more control over who they are sharing with and how they

25

share.’” CC ¶ 109 (internal citations omitted). Because the registration and “friending” process

26

“fostered an incredible amount of trust,” users were willing to “post their cell phone numbers and

27

their birthdays, offer personal photos, and otherwise share information they’d never do outside

28

42
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

their circle of friends.” Id. ¶ 48. Such data was more valuable to advertisers and other third

2

parties than the data generated by a service like Myspace. For example, “[k]nowing the internet

3

habits of ‘YankeesFan123’ is less valuable than knowing the browsing habits of a specific

4

individual whose love of the Yankees can be linked with data about his shopping habits, income,

5

family, friends, travel, dining, dating, and myriad other data points.” Id. ¶ 50. As Mr. Zuckerberg

6

put it, Facebook is “one of the only services on the web where people are sharing pretty personal

7

and intimate information.” Id. ¶ 130.

8
9

However, Facebook knew that users would not give Facebook their data if Facebook did
not “promise privacy protection.” Id. ¶ 46. To a certain degree, privacy is inherent in Facebook’s
“closed-network approach.” Id. ¶ 106. Because Facebook is a “walled garden,” advertisers and

11

United States District Court
Northern District of California

10

other third parties cannot access user data without Facebook’s permission. See id. Nevertheless,

12

Facebook made additional promises to gain additional trust. For example, in a 2004 Privacy

13

Policy, “Facebook promised users that it would disclose its ‘information and privacy practices’

14

and that it would ‘not use cookies to collect private information from any users.’” Id.

15

Indeed, Facebook knew that the success of its business depended on two competing forces.

16

The first force was users’ desire for increased control over their data. The second force was

17

advertisers’ and other third parties’ desire to purchase as much data as Facebook could gather.

18

Whereas increasing privacy protections would reduce Facebook’s revenue from advertisers and

19

third parties, reducing privacy protections would potentially cause users to stop sharing data or to

20

leave Facebook altogether. For example, when Google launched a rival platform in 2010,

21

Facebook decided to “postpone any changes to its privacy policies ‘until the direct competitive

22

comparisons beg[a]n to die down.’” Id. ¶ 134.

23

Facebook “spent the next fifteen years deceiving consumers about the data privacy

24

protections that it provided to users in exchange for access to their data.” Id. ¶ 111. By selling

25

increased amounts of data to third parties while representing to users that it was keeping data

26

private, Facebook increased its user base and its profits. In the next section, the Court describes

27

Consumers’ specific allegations about Facebook’s data practices and the misrepresentations

28

43
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

Facebook made to hide those practices from users.

2

Facebook’s Deceptive Collection and Sale of User Data to Third Parties

3

Consumers allege specific instances in which Facebook misled users about the scope of

4

Facebook’s data collection and monetization practices. For example, despite repeated promises

5

that it would protect users’ privacy and give users the tools to keep their data safe, Facebook

6

tracked users across the internet with various deceptive tools and provided users’ private

7

information to third party application developers without permission. Below, the Court describes

8

each of the instances.

9

The News Feed Protest
In 2006, an incident involving Facebook’s “News Feed” feature reaffirmed that Facebook

11

United States District Court
Northern District of California

10

users cared strongly about controlling their data. The “News Feed” provided each Facebook user

12

with a dynamic list of information and media that the user’s friends recently have posted and “was

13

intended as a central destination so users did not have to browse through friends’ profiles for

14

updates.” CC ¶ 113. However, when the News Feed initially launched, users had little control

15

over what information was pulled from their profiles into their friends’ News Feeds. Id. ¶ 115.

16

Indeed, many users “argu[ed] that the feature was too intrusive, showing every little moment such

17

as Lucy friending Ben or that John and Sarah broke up.” See Alyssa Newcomb, Can You Even

18

Remember How You Coped Before Facebook’s News Feed?, NBC News (Sept. 2, 2016),

19

https://nbcnews.to/3nmr7Xp.

20

After one million of Facebook’s then eight million users joined a “Facebook News Feed

21

protest group,” Facebook CEO Mark Zuckerberg publicly stated: “This was a big mistake on our

22

part, and I’m sorry for it. But apologizing isn’t enough. I wanted to make sure we did something

23

about it, and quickly. So we have been coding nonstop for two days to get you better privacy

24

controls.” CC ¶¶ 114–15. Facebook then instituted settings that “purportedly allowed users

25

greater ability to keep their activities private.” Id. ¶ 115.

26

Facebook Deceptively Collected Data Despite Beacon’s “No, Thanks”
Option

27
28

44
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

Despite its promise that it would give users tools to control their privacy, Facebook

2

deceived users in 2007 about Facebook’s practice of using a tool called “Beacon” to track users’

3

purchases on third party websites. See CC ¶ 119. Specifically, when Facebook users made

4

purchases on participating third party websites, Beacon would send the users “‘pop-up’

5

notification[s] . . . requesting permission to share details regarding the user[s] [purchase] with

6

Facebook.” Id. The “‘pop-up’ notifications” provided users with a “No, Thanks” option, which

7

gave users the impression that they could decline to share their information with Facebook. Id.

8

However, Stefan Berteau, a senior research engineer at California’s Threat Research
Group, examined the software code used to create Beacon and discovered that “Beacon allowed

10

Facebook to track the activity of even those users that clicked the ‘No, Thanks’ prompt.” See id.

11

United States District Court
Northern District of California

9

¶ 120; ECF No. 97-4 at 57–58. Accordingly, the “No, Thanks” prompt deceived users into

12

thinking that Facebook was providing them with control over their purchasing data when, in

13

reality, Facebook was collecting and selling that data.

14

When rumors about the full extent of Beacon’s functionality began to spread, Facebook

15

executives attempted to silence the rumors. In a 2007 interview, Brad Stone of the New York

16

Times asked Chamath Palihapitiya, Facebook’s Vice President of Marketing and Operations,

17

whether Beacon provided Facebook with information about a purchase even if a user “decline[d]

18

to publish the purchase” by clicking “No, Thanks.” See CC ¶ 120 n.92; ECF No. 97-4 at 57 n.87.

19

Mr. Palihapitiya responded: “Absolutely not.” ECF No. 97-4 at 57 n.87. Mr. Palihapitiya also

20

stated that Facebook was trying to “dispel a lot of misinformation that [was] being propagated

21

unnecessarily.” Id.

22

After an FTC investigation revealed the truth about Beacon, Facebook modified Beacon so

23

that users had a genuine choice to opt out. CC ¶ 121. In response to significant public outrage

24

and litigation, Facebook CEO Mark Zuckerberg called Beacon a mistake and stated: “I’m not

25

proud of the way we’ve handled this situation and I know we can do better.” Id.

26

Facebook Provided Users’ Private Information to Third Party
Applications Without Users’ Permission

27
28

45
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

Around the same time that Facebook launched Beacon, and unbeknownst to users,

2

Facebook began giving “app developers access to user content and information, including content

3

marked private.” CC ¶ 117.

4

Indeed, “Facebook explained to third-party app developers in May 2007 that Facebook’s

5

core value proposition and business model was ‘providing access to a new kind of data—social

6

data, which enables you to build applications that are relevant to users.’” Id. ¶ 118 (internal

7

citation omitted). Although Facebook had not expressly promised to keep app developers from

8

accessing user data, Facebook’s terms of service stated only that user data would be used for

9

targeted advertising. See id. ¶¶ 5–6.
In 2009, Facebook started to provide users with a “Central Privacy Page.” Id. ¶ 123. The

11

United States District Court
Northern District of California

10

Central Privacy Page invited users to “control who can see your profile and personal information”

12

and gave users the option of sharing their information with “Only Friends.” See id.; Facebook

13

Settles FTC Charges That It Deceived Consumers By Failing To Keep Privacy Promises, Federal

14

Trade Commission (Nov. 29, 2011) (“2011 FTC Settlement”), https://bit.ly/3318YHU. The

15

Central Privacy Page did not mention the possibility that application developers could receive

16

users’ “personal information.” Additionally, although Facebook’s “Privacy Settings” stated that

17

applications would receive some of users’ information, Facebook assured users that applications

18

would only receive information necessary “for [the applications] to work.” Complaint for

19

violation of the Federal Trade Commission Act in the matter of Facebook Inc., No. 0923184 at 10

20

(F.T.C. Dec. 5, 2011) (“2011 FTC Complaint”), https://bit.ly/33eqwAa. Taken together, these

21

statements gave users the impression that Facebook was not providing third party application

22

developers with users’ personal data for profit. Id.

23

Then, in May 2010, Facebook made three statements that confirmed this impression. First

24

on, May 24, 2010, Facebook CEO Mark Zuckerberg wrote an op-ed in the Washington Post

25

stating that Facebook “do[es] not share your personal information with people or services you

26

don’t want,” “do[es] not give advertisers access to your personal information,” and “do[es] not and

27

never will sell any of your information to anyone.” See CC ¶ 238(m) n.212; Anita Balakrishnan,

28

46
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

et al., Mark Zuckerberg Has Been Talking About Privacy for 15 Years – Here’s Almost Everything

2

He’s Said, CNBC (Apr. 9, 2018), https://cnb.cx/3n7iPTn.

3

Second, on May 26, 2010, Facebook issued an official press release which stated that

4

Facebook would “give the more than 400 million people who use Facebook the power to control

5

exactly who can see the information and content they share”; that “People have control over how

6

their information is shared”; that “Facebook does not share personal information with people or

7

services users don’t want”; that “Facebook does not give advertisers access to people’s personal

8

information”; and that “Facebook does not sell any of people’s information to anyone.” CC

9

¶ 238(c).

10

Third, on May 27, 2010, Facebook CEO Mark Zuckerberg stated in an interview with

United States District Court
Northern District of California

11

National Public Radio that “[t]here’s this false rumor that’s been going around which says that

12

we’re sharing private information with applications and it’s just not true.” Id. ¶ 238(d).

13

Despite these statements, Facebook subsequently increased the amount of user data that

14

developers could purchase. Specifically, Facebook allowed “third parties to access the data of a

15

Facebook user’s friends,” even if those friends had not signed up for the third parties’ applications.

16

Id. ¶ 145.

17

Facebook Deceptively Collected Users’ Data With “Like” Buttons on
Third Party Websites

18
Around the same time that Facebook increased the amount of user data it was providing to
19
20

third party applications, Facebook started using the “Like” button tool to collect user data without
permission. In early 2010, Facebook began allowing participating third parties to add “Like”

21
buttons as plugins on their external websites. See CC ¶ 125. These buttons allowed users to
22
23
24
25
26

express public approval for a “particular piece of content on a third party’s website.” Id. As of
2018, “Like” buttons “appear[ed] on nearly 3 million websites.” ECF No. 97-4 at 41 n.6.
Although users knew that clicking a “Like” button transmitted information to Facebook,
“Facebook’s ‘Frequently Asked Questions’ page indicated that ‘No data is shared about you when
you see a social plug-in on an external website.’” See CC ¶ 126. However, in 2011, a Dutch

27
28

47
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

computer scientist named Arnold Roosendaal discovered that Facebook used “Like” buttons on

2

third party websites “to monitor users and obtain their data anytime users visited those third-party

3

websites, even when users did not click the ‘Like’ button.” See id. (emphasis in original); ECF

4

No. 97-4 at 65.

5

As with Beacon, Facebook attempted to keep the true function of the “Like” button secret.

6

For example, in a May 2011 interview with the Wall Street Journal, Facebook CTO Bret Taylor

7

stated, in reference to the “Like” buttons, “[w]e don’t use them for tracking and they’re not

8

intended for tracking.” ECF No. 97-4 at 66.

9
10

The 2011 Federal Trade Commission Settlement
In 2011, the FTC filed a complaint against Facebook alleging that Facebook had “deceived

United States District Court
Northern District of California

11

customers by telling them they could keep their information on Facebook private, and then

12

repeatedly allowing it to be shared and made public.” CC ¶ 127. In particular, the FTC

13

highlighted Facebook’s practice of allowing third party app developers to access the data of users’

14

friends. See id. ¶ 145; 2011 FTC Complaint, supra.

15

Shortly thereafter, Facebook and the FTC agreed to a settlement which “barred Facebook

16

from making any further deceptive privacy claims.” CC ¶ 127. Facebook also agreed “to obtain

17

consumers’ affirmative express consent before enacting changes that override their privacy

18

preferences,” “to prevent anyone from accessing a user’s material more than 30 days after the user

19

has deleted his or her account,” and “to establish and maintain a comprehensive privacy program

20

designed . . . to protect the privacy and confidentiality of consumers’ information.” Id. ¶ 128.

21

The settlement also directly addressed Facebook’s practice of giving user data to third

22

party app developers. For example, the settlement required Facebook to “give consumers ‘clear

23

and prominent’ notice and obtain their consent before sharing their information beyond those

24

entities clearly enumerated in consumers’ privacy settings.” Id. ¶ 153.

25

On November 29, 2011, Mr. Zuckerberg issued a press release which stated that “everyone

26

needs complete control over who they share with at all times.” Id. ¶ 238(e). Mr. Zuckerberg

27

added that “[t]his idea has been the core of Facebook since day one” and that the reason people

28

48
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

originally signed up for Facebook was that, “for the first time,” they “could make their page

2

private.” Mark Zuckerberg, Our Commitment to the Facebook Community, Facebook Newsroom

3

(Nov. 29, 2011), https://bit.ly/3zDEs2A. Indeed, Mr. Zuckerberg explained, the reason “Facebook

4

became the world’s biggest community online” was that Facebook “made it easy for people to feel

5

comfortable sharing things about their real lives.” Id.

6

Mr. Zuckerberg further stated that, pursuant to the 2011 FTC Settlement, Facebook would
be giving users “tools to control who can see your information and then making sure only those

8

people you intend can see it.” CC ¶ 238(f). Specifically, Mr. Zuckerberg stated that Facebook

9

would be providing users with a “new apps dashboard to control what your apps can access” and a

10

“new app permission dialog that gives you clear control over what an app can do anytime you add

11

United States District Court
Northern District of California

7

one.” Mark Zuckerberg, Our Commitment to the Facebook Community, Facebook Newsroom

12

(Nov. 29, 2011), https://bit.ly/3zDEs2A.

13

Mr. Zuckerberg reinforced this message in a February 2012 letter to investors which stated

14

that “giving people control over what they share is a fundamental principle” and, in accordance

15

with that principle, Facebook was dedicated to “build[ing] tools to help people . . . share what they

16

want.” CC ¶ 238(g).

17

After the 2011 FTC Settlement, Facebook Deceptively Tracked Users
Across the Internet With “View Tags”

18
However, Facebook quickly resumed its deceptive practices. For example, in 2012,
19
20
21

Facebook “began using ‘View Tags,’ which allow advertisers to track Facebook users across the
Internet using cookies.” CC ¶ 137. View Tags allowed companies to determine whether users
who viewed the companies’ advertisements on Facebook purchased the companies’ products. Id.

22
(citing Rebecca Greenfield, 2012: The Year Facebook Finally Tried to Make Some Money, The
23
Atlantic (Dec. 14, 2012), https://bit.ly/33aUFjF).
24
25
26

This practice contradicted Facebook’s 2007 Privacy Policy, which stated that although
Facebook would use cookies to collect users’ “browser type and IP address,” Facebook would use
this information only to “confirm that users are logged in.” ECF No. 97-4 at 49 n.43, 58. The

27
28

49
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

Privacy Policy also stated that “[t]hese cookies terminate once the user closes the browser.” Id.

2

Furthermore, around the time that Facebook started using “View Tags,” Facebook made

3

several statements to assure users that Facebook was not using cookies for commercial purposes.

4

For example, in a May 2011 interview with the Wall Street Journal, Facebook CTO Bret Taylor

5

stated that, although “Facebook places cookies on the computers of people that visit

6

facebook.com,” the sole purpose of these cookies was to “protect users’ Facebook accounts from

7

cyber-attacks.” ECF No. 97-4 at 66. Additionally, in September 2011, a Facebook engineer

8

named Arturo Bejar wrote in a blog post that Facebook used cookies only “for safety and

9

protection.” Id. at 67. Mr. Bejar further stated: “we don’t use our cookies to . . . target ads or sell
your information to third parties.” Id. The same month, a Facebook spokesperson stated that

11

United States District Court
Northern District of California

10

Facebook does not use cookies to “track users across the web.” Id.

12
13
14

After the 2011 FTC Settlement, Facebook Resumed its Practice of
Providing Users’ Private Information to Third Party Applications
Without Users’ Permission
Most critically, Facebook resumed its practice of giving third party app developers the data

15

of users who had not signed up for those developers’ applications. See CC ¶ 145. Indeed, the

16

“volume of data third parties were acquiring from Facebook led one Facebook employee to state:

17

‘I must admit, I was surprised to find out that we are giving out a lot here for no obvious reason.’”

18

Id. This practice continued until at least 2018. Id. ¶ 146.

19

Before this practice was uncovered, Facebook made several deceptive statements that

20

obscured the practice from users. For example, in a February 2017 filing with the SEC, Facebook

21

stated that, although “some of our developers or other partners, such as those that help us measure

22

the effectiveness of ads, may receive or store information provided by us or by our users through

23

mobile web applications,” Facebook provides only “limited information to such third parties based

24

on the scope of services provided to us.” See id. ¶ 238(l); Form 10-K Filing for Facebook, Inc.,

25

Securities and Exchange Commission (Feb. 2, 2017), https://bit.ly/3q7CtAl.

26
27
28

The Cambridge Analytica Scandal
A 2018 scandal involving the British political consulting firm Cambridge Analytica
50
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

revealed Facebook’s practice of providing users’ private information to third party app developers.

2

Specifically, the “Cambridge Analytica scandal” revealed that, although the Cambridge Analytica

3

Facebook application had only 270,000 users, Cambridge Analytica “was able to access the

4

personal data of up to 87 million Facebook users.” See CC ¶ 143 (emphasis added).
Mr. Zuckerberg called the Cambridge Analytica incident a “mistake,” pledged to take

5
6

action against “rogue apps,” and stated that “[w]e have a responsibility to protect your data, and if

7

we can’t then we don’t deserve to serve you.” 2 See CC ¶ 238(o); Facebook’s Zuckerberg Speaks

8

Out Over Cambridge Analytica ‘Breach’, BBC News (Mar. 22, 2018), https://bbc.in/3GmpZKT.

9

However, despite Mr. Zuckerberg’s statement that Cambridge Analytica was a “mistake,”
Facebook subsequently announced that at least 10,000 applications had been able to access similar

11

United States District Court
Northern District of California

10

data for the entire period since the FTC settlement. CC ¶ 148.

12

The 2019 Federal Trade Commission Penalty
Shortly after the Cambridge Analytica Scandal, the Department of Justice (“DOJ”) and the

13
14

FTC filed a complaint alleging that Facebook had breached the 2011 FTC settlement agreement by

15

“deceiv[ing] users about their ability to control the privacy of their personal information.” CC

16

¶¶ 145, 149.
In July 2019, “the FTC and the DOJ announced that Facebook would pay a $5 billion

17
18

penalty.” Id. ¶ 149. Although this penalty exceeded the FTC’s largest ever penalty by an order of

19

magnitude, the FTC explained that the size was justified because Facebook had made $55.8 billion

20

in 2018 through targeted advertising and because Facebook’s promises to users “that they can

21

control the privacy of their information” were central to collecting information for that targeted

22

advertising. Id. ¶ 150.
Facebook’s Deceptive Privacy Practices Helped Facebook Obtain and
Maintain Monopoly Power

23
24
25
26
27
28

In 2017, “Facebook disclosed to congressional investigators that it had sold to a Russian
company, Internet Research Agency (IRA), ads which psychographically targeted American
voters, and that up to approximately 126 million could have been targeted with such ads.” ECF
No. 97-4 at 42 n.10.
51
2

Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

Consumers allege that Facebook’s deceptive practices allowed Facebook to maximize its

2

user base and its revenue. Between 2007 and 2010, Facebook doubled its revenue every year. CC

3

¶ 124. By 2010, Facebook made over $1 billion in annual revenue and was the “largest social

4

network in the world.” Id.

5

In 2012, when Facebook executed an initial public offering, Facebook “had 845 million

6

monthly active users” worldwide. Id. ¶ 138. Additionally, data analytics firm Comscore reported

7

that Facebook was the “third largest web property in the world . . . and accounted for

8

approximately 3 in every 4 minutes spent on social networking sites and 1 in every 7 minutes

9

spent online around the world.” House Report at 144 (internal citation omitted). Around the same
time, Facebook COO Sheryl Sandberg stated in a presentation to a “large telecommunications

11

United States District Court
Northern District of California

10

firm” that Facebook was “95% of all social media” in the United States as measured by “monthly

12

minutes of use.” See CC ¶ 286; House Report at 138.

13

In 2018, “Facebook earned over $55 billion in revenue” and had “over 217 million users in

14

the United States.” CC ¶ 147. In 2019, Facebook earned over $70 billion in revenue, id. ¶ 41,

15

which gave it an 85% share of the Social Media Market as measured by revenue, id. ¶ 80.

16

Additionally, Facebook’s deception of users allowed Facebook to prevent sophisticated

17

rivals, like Google, from entering the market. In 2010, Google launched a new social network

18

called Google+. Id. ¶ 131. With Google+, Google attempted “to build out a ‘social graph’ that

19

would leverage a common user identity across Google products, including YouTube and Gmail.”

20

Id. Google poured a “massive” amount of resources into Google+ and “conscripted almost all of

21

the company’s products to help build Google+.” Id. ¶ 133. At its peak, “Google+ involved 1,000

22

employees from divisions across the country.” Id.

23

However, around the time Google+ entered the social network market, “prominent

24

investors . . . noted that the social networking market had ‘extreme network effects,’ making it

25

‘increasingly hard to see a materially successful new entrant, even with all of Google’s

26

resources.’” House Report at 144. Accordingly, Google+ needed a way to distinguish itself from

27

Facebook.

28

52
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

Indeed, Facebook realized that it could not allow users to find out about Facebook’s

2

privacy practices while Google+ was a viable alternative. For example, in 2011, a Facebook

3

executive stated that it would be unwise to remove privacy protections because “IF ever there was

4

a time to AVOID controversy, it would be when the world is comparing our offerings to G+.” CC

5

¶ 134. The same executive stated that Facebook could remove those protections after “the

6

directive competitive comparisons begin to die down.” Id.

7

Without any way to distinguish itself from Facebook and to overcome the extreme network

8

effects, Google+ was not able to compete with Facebook. Id. ¶ 135. Thus, in 2018, Google+ left

9

the market. Id. ¶ 140.

10

United States District Court
Northern District of California

11
12

Consumers Allege with Sufficient Particularity that Facebook Made Numerous
“Clearly False” Representations About Its Collection and Monetization of Data
Facebook contends that none of the representations identified by Consumers are “‘clearly
false’” and that, in any event, Consumers have failed to allege those representations with sufficient

13
particularity. Mot. at 20–21 (quoting Am. Pro. Testing Serv., 108 F.3d at 1152). For the reasons
14
15
16
17
18
19
20
21
22
23
24

below, the Court rejects Facebook’s arguments.
Although the Ninth Circuit has not specifically addressed what constitutes a “clearly false”
statement in the context of the Sherman Act, the Ninth Circuit’s cases dealing with securities fraud
claims provide useful guidance. The Securities Exchange Commission (“SEC”) has interpreted
Section 10(b) of the Securities Exchange Act to prohibit “any untrue statement of a material fact
. . . in connection with the purchase or sale of any security.” 17 C.F.R. § 240.10b-5(b). Thus,
“[t]o plead a claim under section 10(b),” a plaintiff must allege “a material misrepresentation or
omission.” Oregon Pub. Emps. Ret. Fund v. Apollo Grp. Inc., 774 F.3d 598, 603 (9th Cir. 2014).
To qualify as a “material misrepresentation,” a statement must be “capable of objective
verification.” Id. at 606. By contrast, “subjective assessments” and “vague statements of
optimism like ‘good,’ ‘well-regarded,’ or other feel good monikers” are “non-actionable puffing.”

25
See In re Cutera Sec. Litig., 610 F.3d 1103, 1111 (9th Cir. 2010).
26

The same principles apply to a Sherman Act claim based on “false and misleading

27
28

53
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

advertising.” Just as the purpose of a securities fraud claim is to hold a company liable for making

2

misrepresentations that caused investors to purchase the company’s securities, the purpose of a

3

“false and misleading advertising” claim under the Sherman Act is to hold a “monopolist” liable

4

for making “misrepresentations [that] encourag[ed] the purchase of its product.” Areeda ¶ 782b.

5

Indeed, the Ninth Circuit’s statement that misrepresentations are anticompetitive only if they are

6

“clearly false” and “clearly material” mirror the basic requirements of a securities fraud claim.

7

Am. Pro. Testing Serv., 108 F.3d at 1152.

8

Additionally, as discussed, Federal Rule of Civil Procedure 9(b) requires that “the
circumstances constituting the alleged fraud be specific enough to give defendants notice of the

10

particular misconduct . . . so that they can defend against the charge and not just deny that they

11

United States District Court
Northern District of California

9

have done anything wrong.” Vess, 317 F.3d at 1106. Thus, a plaintiff must allege the “who, what,

12

when, where, and how” of each alleged misrepresentation. Id.

13

Accordingly, the Court must assess whether Consumers have alleged misrepresentations

14

that were “capable of objective verification” and whether Consumers have alleged the “who, what,

15

when, where, and how” of each misrepresentation. As an initial matter, the Court concludes that

16

several of the representations that Consumers highlight are “non-actionable puffing.” See In re

17

Cutera, 610 F.3d at 1111. For example, Consumers highlight Facebook’s statement that

18

“[k]eeping the global community safe is an important part of our mission – and an important part

19

of how we’ll measure our progress going forward.” CC ¶ 238(n). The Court agrees with

20

Facebook that such representations are “vague statements of optimism” and are not “capable of

21

objective verification.” See In re Cutera Sec. Litig., 610 F.3d at 1111; Apollo Grp., 774 F.3d at

22

606. Accordingly, the Court must disregard these representations.

23

However, the Court finds that Consumers have alleged with sufficient particularity that

24

Facebook made numerous “clearly false” representations about Facebook’s data privacy practices.

25

Specifically, Consumers have adequately alleged at least the following “clearly false”

26

representations that are “capable of objective verification,” Apollo Grp., 774 F.3d at 606: (1)

27

Facebook falsely represented that it was not sharing users’ private information with third parties,

28

54
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

beginning with statements made after the News Feed protest and culminating in statements made

2

during the Cambridge Analytica scandal; (2) Facebook falsely represented that users could prevent

3

the “Beacon” tool from collecting their data; (3) Facebook falsely represented that users could

4

prevent the “Like” button from collecting their data; and (4) Facebook falsely represented that it

5

would not use cookies to collect users’ data for commercial purposes, such as tracking users across

6

the internet with “View Tags.” Additionally, because Consumers identify the “who, what, when,

7

where, and how” of each of these alleged representations, each alleged representation satisfies

8

Federal Rule of Civil Procedure 9(b). Vess, 317 F.3d at 1106.

9
10

United States District Court
Northern District of California

11

Consumers’ Data Privacy Claims Are Timely
Facebook contends that Consumers’ data privacy claims are barred by the statute of
limitations. See Mot. at 6–7. For the reasons below, the Court rejects this contention.

12

A claim may be dismissed under Rule 12(b)(6) on the ground that it is barred by the

13

applicable statute of limitations only when “the running of the statute is apparent on the face of the

14

complaint.” Huynh v. Chase Manhattan Bank, 465 F.3d 992, 997 (9th Cir. 2006). Thus, “a

15

complaint cannot be dismissed unless it appears beyond doubt that the plaintiff can prove no set of

16

facts that would establish the timeliness of the claim.” Supermail Cargo, Inc. v. United States, 68

17

F.3d 1204, 1206 (9th Cir. 1995).

18

The limitations period for private damages claims brought under Section 2 of the Sherman

19

Act is four years. 15 U.S.C. § 15b. However, the “period of limitations for antitrust litigation

20

runs from the most recent injury caused by the defendants’ activities rather than from the

21

violation’s inception.” Xechem, Inc. v. Bristol-Myers Squibb Co., 372 F.3d 899, 902 (7th Cir.

22

2004) (Easterbrook, J.) (citing Zenith Radio Corp. v. Hazeltine Research, Inc., 401 U.S. 321

23

(1971)). Under this “continuing violation” doctrine, “each overt act that is part of the [antitrust]

24

violation and that injures the plaintiff . . . starts the statutory period running again.” Klehr v. A.O.

25

Smith Corp., 521 U.S. 179, 189 (1997) (internal quotations omitted). However, to qualify as an

26

“overt act,” the act must satisfy “two criteria: 1) It must be a new and independent act that is not

27

merely a reaffirmation of a previous act; and 2) it must inflict new and accumulating injury on the

28

55
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

plaintiff.” Samsung Elecs. Co., Ltd. v. Panasonic Corp., 747 F.3d 1199, 1202 (9th Cir. 2014)

2

(internal quotations omitted).

3

Accordingly, because Consumers filed their complaint on December 3, 2020, Consumers

4

must allege “overt act[s] that [are] part of the [antitrust] violation” that occurred after December 3,

5

2016. See Klehr, 521 U.S. at 189.

6

Because Consumers have adequately alleged that Facebook made at least two false

7

representations after December 3, 2016, Consumers’ claim is timely. First, on February 2, 2017,

8

Facebook stated in an SEC filing that Facebook provides only “limited information to [third party

9

application developers] based on the scope of services provided to us.” See CC ¶ 238(l); Form 10K Filing for Facebook, Inc., Securities and Exchange Commission (Feb. 2, 2017),

11

United States District Court
Northern District of California

10

https://bit.ly/3q7CtAl. Second, in March 2018, Mr. Zuckerberg called the Cambridge Analytica

12

incident a “mistake,” pledged to take action against “rogue apps,” and stated that “[w]e have a

13

responsibility to protect your data, and if we can’t then we don’t deserve to serve you.” See CC

14

¶ 238(o); Facebook’s Zuckerberg Speaks Out Over Cambridge Analytica ‘Breach’, BBC News

15

(Mar. 22, 2018), https://bbc.in/3GmpZKT.

16

As discussed, Consumers have alleged with sufficient particularity that these

17

representations were “clearly false.” Facebook’s 2017 statement to the FTC would have given

18

reasonable users the impression that Facebook was not providing third party applications with

19

private information. To the contrary, the Cambridge Analytica scandal revealed that Facebook

20

had provided users’ private information to numerous third party applications, including

21

applications for which users were not registered. See CC ¶ 143. For example, although

22

Cambridge Analytica had only 270,000 users, Cambridge Analytica “was able to access the

23

personal data of up to 87 million Facebook users.” See id. (emphasis added). Similarly, Mr.

24

Zuckerberg’s 2018 statement about Cambridge Analytica would have given reasonable users the

25

impression that Cambridge Analytica was a “rogue app” and that Facebook had not been

26

systematically providing users’ private information to third party application developers.

27

However, Facebook subsequently announced that at least 10,000 applications had been able to

28

56
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

access similar data for the entire period since the FTC settlement. Id. ¶ 148. Accordingly,

2

because each of these representations is an “overt act that is part of the [antitrust] violation” and

3

occurred after December 3, 2016, Consumers’ claim is timely. Klehr, 521 U.S. at 189.

4

Facebook’s arguments to the contrary are unconvincing. Facebook argues that any false

5

representation made after 2016 “is, as alleged, a ‘reaffirmation of a previous’ strategy, not a ‘new

6

and independent act’ capable of restarting the statute of limitations.” Reply at 4–5 (quoting Bay

7

Area Surgical Mgmt. LLC v. Aetna Life Ins. Co., 166 F. Supp. 3d 988, 999 (N.D. Cal. 2015)).

8

This argument selectively and misleadingly quotes Bay Area Surgical Management, which stated

9

in relevant part that an “overt act restarts the statute of limitations if it . . . is ‘a new and
independent act that is not merely a reaffirmation of a previous act.’” 166 F. Supp. 3d at 999

11

United States District Court
Northern District of California

10

(quoting Pace Indus. v. Three Phoenix Co., 813 F.2d 234, 237 (9th Cir. 1987)) (emphasis added).

12

Thus, Facebook provides no authority for its argument that an act is not “new and independent”

13

simply because the defendant has previously committed the same type of act as part of a unified

14

anticompetitive strategy.

15

Facebook’s argument ignores the Ninth Circuit’s clear guidance that, if a defendant

16

commits the same anticompetitive act multiple times, each new act restarts the statute of

17

limitations for all the acts. See Oliver v. SD-3C LLC, 751 F.3d 1081, 1086 (9th Cir. 2014)

18

(“[E]ach time a defendant sells its price-fixed product, the sale constitutes a new overt act causing

19

injury to the purchaser and the statute of limitations runs from the date of the act.”); Hennegan v.

20

Pacifico Creative Serv., Inc., 787 F.2d 1299, 1301 (9th Cir.1986) (holding that a new overt act

21

occurred each time tour operators shepherded tourists away from the plaintiffs’ shop in exchange

22

for payment). Accordingly, the two representations discussed above were overt acts that restarted

23

the limitations period.

24

Additionally, Facebook summarily asserts that “even if the post-December 2016 acts could

25

be sufficient, [Consumers] cannot allege—as they must—that the acts inflicted ‘new and

26

accumulating injury’ on them.” Reply at 5. However, this argument ignores Consumers’

27

allegation that each of Facebook’s false representations allowed Facebook to maintain a “critical

28

57
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

mass of users” by convincing users that Facebook was protecting their data. CC ¶ 217. As the

2

Seventh Circuit has explained, “improperly prolonging a monopoly is as much an offense against

3

the Sherman Act as is wrongfully acquiring market power in the first place.” Xechem, 372 F.3d at

4

902. Moreover, as explained in the following sections, Consumers have adequately alleged each

5

of the elements required to show that Facebook’s false representations had a significant effect on

6

competition. Accordingly, Facebook’s conclusory argument is insufficient to show that

7

Consumers “can prove no set of facts that would establish” that each of Facebook’s false

8

representations contributed to Consumers’ injury. Supermail Cargo, 68 F.3d at 1206.
Thus, Consumers’ data privacy claims are timely.

9

Consumers Adequately Allege that Facebook’s False Representations About Its
Data Privacy Practices Were Not Readily Susceptible of Neutralization

10
11

United States District Court
Northern District of California

Facebook contends that, even if Consumers have adequately alleged that Facebook made
12

false representations, Consumers have failed to allege that these representations were “‘not readily

13
susceptible of neutralization or other offset by rivals.’” Mot. at 21–22 (quoting Am. Pro. Testing
14
Serv., 108 F.3d at 1152). For the reasons below, the Court rejects this argument.
15
16

Although the Ninth Circuit has not explained how to evaluate whether a company’s false
representations about its own products are reasonably susceptible of neutralization,3 three cases

17
from the D.C. Circuit provide useful guidance. First, in Caribbean Broadcasting Sys., Ltd. v.
18
Cable & Wireless PLC, 148 F.3d 1080, 1087 (D.C. Cir. 1998), a radio station in the Caribbean
19
brought a Sherman Act claim alleging that a rival radio station had falsely represented to
20
21
22

advertisers that the rival station’s services “could reach the ‘entire Caribbean.’” In reality, the
station “reached only a fraction of that area.” Id. The D.C. Circuit held that these “fraudulent
misrepresentations to advertisers” were “well within” the definition of “anticompetitive conduct.”

23
Id.
24
25
26
27
28

The Ninth Circuit’s decision in American Professional Testing Service, 108 F.3d at 1152, dealt
with a claim that the defendant had disparaged a rival by posting fliers which falsely stated that the
rival “was in financial trouble” and “had been accused of fraud by the SEC.” Because these
statements were about the rival, and the rival knew the statements were false, the rival was in a
good position to “neutralize” the statements. See id.
58
3

Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

Second, in United States v. Microsoft Corp., 253 F.3d 34, 76 (D.C. Cir. 2001), the

2

Department of Justice brought a Sherman Act claim alleging that Microsoft had falsely

3

represented to developers that Microsoft’s software tools could be used to design applications that

4

would be compatible with other platforms. In reality, “Microsoft’s tools included ‘certain

5

keywords and compiler directives that could only be executed properly by Microsoft’s

6

[software].’” Id. (internal citation omitted). The D.C. Circuit held that these false representations

7

fell within the scope of the Sherman Act. Id.

8
9

Third, in Covad Communications Co. v. Bell Atlantic Corp., 398 F.3d 666, 670 (D.C. Cir.
2005), a provider of Digital Subscriber Line (“DSL”) services brought a Sherman Act claim
alleging that Bell Atlantic had falsely represented to customers in certain areas that Bell Atlantic

11

United States District Court
Northern District of California

10

could provide DSL services. Id. at 674. In reality, Bell Atlantic’s DSL services were not yet

12

available in those areas. Id. However, the D.C. Circuit held that Bell Atlantic’s representations

13

were readily susceptible of neutralization because the “falsity of” Bell Atlantic’s representation

14

was “necessarily dispelled whenever a consumer trie[d] to obtain the [DSL] service.” Id. at 675.

15

The takeaway from these three cases is that a company’s false representation is not readily

16

susceptible of neutralization when the representation is about a highly technical aspect of the

17

company’s product. In Caribbean Broadcasting, 148 F.3d at 1087, the defendant made false

18

representations about the technical capabilities of its radio broadcasting system. Similarly, in

19

Microsoft, 253 F.3d at 76, Microsoft made false representations about the technical features of its

20

programming tools. Thus, it would have been difficult for anybody without technical expertise to

21

determine that the representations considered in Caribbean Broadcasting and Microsoft were

22

false. By contrast, in Covad, 398 F.3d at 670, Bell Atlantic falsely represented that its services

23

were available to customers in certain geographic areas. Because any customer who tried to

24

obtain the defendant’s services could discover that this representation was false, technical

25

expertise was not required. Indeed, in Covad, the D.C. Circuit expressly contrasted Bell Atlantic’s

26

representations with the representations considered in Microsoft. Id.

27
28

Under this standard, Consumers have adequately alleged that Facebook’s false
59
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

representations were “not readily susceptible of neutralization or other offset by rivals.” Am. Pro.

2

Testing Serv., 108 F.3d at 1152. Consumers explain that Facebook has created a “walled garden”

3

which gives Facebook complete control over user data. See CC ¶ 106. Thus, even sophisticated

4

third parties, such as developers and search engines, cannot access user data without Facebook’s

5

permission, let alone determine what Facebook is doing with user data. See id.

6

Moreover, Consumers have alleged that each of Facebook’s deceptive privacy practices—

7

providing private user data to third party application developers, collecting user data with the

8

Beacon tool, collecting user data with “Like” buttons, and tracking users across the internet with

9

cookies—could not have been revealed by anybody without significant technical expertise. For
example, the true nature of the “Beacon” tool was not revealed until “Stefan Berteau, a senior

11

United States District Court
Northern District of California

10

research engineer at California’s Threat Research Group” examined the software code used to

12

create the “Beacon” tool. ECF No. 97-4 at 57–58. Similarly, the true nature of the “Like” button

13

was not revealed until “Dutch researcher Arnold Roosendaal” published a white paper that

14

analyzed the software code used to create “Like” buttons. Id. at 65. In turn, Facebook’s provision

15

of user data to third party application developers over a seven year period was revealed only

16

because of the Cambridge Analytica scandal. CC ¶ 146. These examples show that, just like the

17

false representations considered by the D.C. Circuit in Caribbean Broadcasting, 148 F.3d at 1087,

18

and Microsoft, 253 F.3d at 76, users and rivals could not have discovered that Facebook’s

19

representations were false without significant technical expertise or access to Facebook’s software

20

code. Thus, Facebook’s false representations were “not readily susceptible of neutralization or

21

other offset.” Am. Pro. Testing Serv., 108 F.3d at 1152.

22

Facebook’s conclusory argument to the contrary is unconvincing. Facebook merely states

23

that the “alleged misrepresentations are on their face reasonably susceptible to neutralization by

24

rivals in obvious ways.” Mot. at 21. “For example,” Facebook contends, “those firms could have

25

improved their own policies, or called attention to Facebook’s supposed misstatements.” Id.

26

However, Facebook makes no effort to explain how rival firms could have known that Facebook’s

27

statements were false when Facebook made them.

28

60
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

Additionally, although Facebook relies on Genus Lifesciences, 378 F. Supp. 3d 823, that

2

case does not support Facebook’s argument. In Genus Lifesciences, the plaintiff, who owned the

3

only FDA approved version of a certain drug, brought a Sherman Act claim alleging that the

4

defendant had falsely represented that the defendant’s version of the same drug also was FDA

5

approved. Id. at 841–42. The court held that this representation was readily susceptible of

6

neutralization because FDA approval is a matter of public knowledge and the plaintiff was fully

7

aware that the defendant’s drug was not approved. Id. By contrast, in the instant case, there was

8

no publicly available information that Facebook’s rival could have consulted to determine whether

9

Facebook’s representations about its data privacy practices were true.

10

Accordingly, Consumers have adequately alleged that Facebook’s false representations

United States District Court
Northern District of California

11

about its data privacy practices were “not readily susceptible of neutralization or other offset by

12

rivals.” Am. Pro. Testing Serv., 108 F.3d at 1152.

13
14
15
16

Consumers Adequately Allege that Facebook’s False Representations Were
“Clearly Material”
Facebook argues that Consumers’ data privacy claims must be dismissed because
Consumers have not adequately alleged that “Facebook’s representations impeded competition.”
Mot. at 21. According to Facebook, Consumers have failed to explain how “Facebook’s alleged

17
misrepresentations prevented other well-resourced firms—like Google or Snapchat—from
18
19
20

competing effectively.” Id. at 22. Additionally, Facebook argues that Consumers’ “underlying
theory is completely implausible” because there are other “competing theories for Facebook’s
success,” “including Facebook’s ‘realness,’ which is alleged to be Facebook’s ‘distinguishing

21
feature.’” Id.
22
23

As an initial matter, Facebook’s arguments ignore the Ninth Circuit’s specific test for
determining whether a “monopolist’s misrepresentations encouraging the purchase of its product”

24
impede competition. See Areeda ¶ 782b. Indeed, as discussed, the Ninth Circuit has stated that a
25
26

monopolist’s misrepresentations about its products constitute “exclusionary conduct” if the
misrepresentations “were [1] clearly false, [2] clearly material, [3] clearly likely to induce

27
28

61
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

reasonable reliance, [4] made to buyers without knowledge of the subject matter, [5] continued for

2

prolonged periods, and [6] not readily susceptible of neutralization or other offset by rivals.” Am.

3

Pro. Testing Serv., 108 F.3d at 1152. Thus, if Consumers have adequately alleged these six

4

elements, Consumers have adequately alleged that Facebook’s misrepresentations impeded

5

competition. See Genus Lifesciences, 378 F. Supp. 3d at 841 (explaining that a Sherman Act

6

claim based on false statements to buyers survives a motion to dismiss if the plaintiff “allege[s]

7

cumulative facts that would prove the statements” meet the six elements).

8

Facebook’s arguments are conclusory and inconsistent with the six elements articulated by

9

the Ninth Circuit in American Professional Testing Service, 108 F.3d at 1152. However, the Court
interprets Facebook’s argument about “competing theories for Facebook’s success” to be an

11

United States District Court
Northern District of California

10

argument that Facebook’s false representations were not “clearly material” to users’ decision to

12

sign up for Facebook. For the reasons below, the Court rejects this argument as well.

13

Although the Ninth Circuit has not addressed the meaning of the term “clearly material” in

14

the context of the Sherman Act, cases dealing with securities fraud claims and cases dealing with

15

false advertising claims under the Lanham Act provide useful guidance. In the context of a

16

securities fraud claim, an “omitted fact is material if there is a substantial likelihood that a

17

reasonable shareholder would consider it important in deciding how to vote.” TSC Industries, Inc.

18

v. Northway, Inc., 426 U.S. 438, 449 (1976). Similarly, in the context of a false advertising claim,

19

a deceptive statement is material if “it is likely to influence the purchasing decision.” Southland

20

Sod Farms v. Stover Seed Co., 108 F.3d 1134, 1139 (9th Cir. 1997). Accordingly, to show that a

21

defendant’s false representation about its product was “clearly material” for purposes of the

22

Sherman Act, the plaintiff must show that customers would consider the representation important

23

in deciding whether to use the defendant’s product or that the representation was likely to

24

influence customers to use the defendant’s product.

25

Under this standard, Consumers adequately allege that Facebook’s false representations

26

about its data privacy practices were “clearly material.” Consumers explain in detail that

27

“[p]rivacy practices were a crucial form of competition in the early days of the Social Network

28

62
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

and Social Media Markets.” CC ¶ 102. In a 2004 consumer survey, a majority of users indicated

2

that privacy was a “really important issue that [they] care about often.” Id. ¶ 108. In another

3

study, users indicated that they cared more about privacy than about terrorism. Id.

4

Indeed, Facebook recognized on multiple occasions that users would not use Facebook unless

5

Facebook “promise[d] privacy protections.” Id. ¶ 46. For example, in Facebook’s 2008 internal

6

report titled “Facebook Secret Sauce,” Facebook observed that “‘[u]sers will share more

7

information if given more control over who they are sharing with and how they share.’” Id. ¶ 109

8

(internal citations omitted). Additionally, when Google+ launched in 2010, Facebook decided to

9

“postpone any changes to its privacy policies ‘until the direct competitive comparisons beg[a]n to

10

die down.’” Id. ¶ 134.

United States District Court
Northern District of California

11

Indeed, in the aftermath of the 2011 FTC settlement, Mr. Zuckerberg stated that “everyone

12

needs complete control over who they share with at all times” and that this “idea has been the core

13

of Facebook since day one.” Mark Zuckerberg, Our Commitment to the Facebook Community,

14

Facebook Newsroom (Nov. 29, 2011), https://bit.ly/3zDEs2A. Mr. Zuckerberg added that the

15

reason people originally signed up for Facebook was that, “for the first time,” they “could make

16

their page private.” CC ¶ 109. Indeed, Mr. Zuckerberg explained, the reason “Facebook became

17

the world’s biggest community online” was that Facebook “made it easy for people to feel

18

comfortable sharing things about their real lives.” Id. As Mr. Zuckerberg put it, Facebook’s

19

privacy protections made Facebook “one of the only services on the web where people are sharing

20

pretty personal and intimate information.” Id. ¶ 130.

21

Given the value that Facebook’s users placed on privacy and given Facebook’s express

22

recognition that privacy was the reason “Facebook became the world’s biggest online

23

community,” it is more than plausible that users would have considered Facebook’s

24

representations about its data privacy practices important in determining whether to use Facebook.

25

Id. Accordingly, Facebook’s false representations about its data privacy practice were clearly

26

material. See TSC Industries, 426 U.S. at 449; Southland Sod Farms, 108 F.3d at 1139.

27
28

Genus Lifesciences supports this conclusion. In Genus Lifesciences, the plaintiff, who
63
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

owned the only FDA approved version of a certain drug, brought a Sherman Act claim alleging

2

that the defendant had falsely represented that the defendant’s version of the same drug also was

3

FDA approved. 378 F. Supp. 3d at 841–42. To support this claim, the plaintiff alleged that the

4

“FDA approval status of a prescription drug is material to customers because approved drugs

5

provide customers assurance as to the quality of the product not afforded to unapproved

6

prescription drugs.” Complaint ¶ 101, Genus Lifesciences Inc. v. Lannett Co., Inc., No. 18-CV-

7

7603 (N.D. Cal. Dec. 18, 2018), ECF No. 1. Citing this allegation, the court concluded that the

8

plaintiff had adequately alleged that the defendant’s statements were “clearly material.” Genus

9

Lifesciences, 378 F. Supp. 3d at 842. Accordingly, because Consumers provide far more detail
about the importance of data privacy to users, Genus Lifesciences supports Consumers’ argument

11

United States District Court
Northern District of California

10

that Facebook’s representations about its data privacy practices were “clearly material.”

12

Facebook’s arguments to the contrary are unconvincing. Citing Rambus Inc. v. FTC, 522

13

F.3d 456, 464 (D.C. Cir. 2008), Facebook argues that Consumers’ “theory is not cognizable if they

14

fail to plausibly allege that deception was the but-for cause of Facebook’s supposed monopoly.”

15

Mot. at 22. However, Rambus says nothing of the sort. Instead, Rambus explains that deceptive

16

statements are anticompetitive if they “impaired rivals in a manner tending to bring about or

17

protect a defendant’s monopoly power.” 522 F.3d at 464 (emphasis added). Thus, Facebook cites

18

no authority which supports its assertion that conduct is anticompetitive only if it is the “but-for

19

cause” of a defendant’s monopoly power.

20

Additionally, Facebook contends that there are “competing theories for Facebook’s

21

success, all more plausible than deception.” Mot. at 22. Specifically, Facebook identifies

22

Facebook’s “realness” and “its unique emphasis on connecting friends, family, and other ‘verified

23

relationships.’” Id. (citing CC ¶ 48). However, there is no requirement that deceptive statements

24

be the only reason for Facebook’s success in order for those deceptive statements to be “clearly

25

material.” Instead, as discussed, deceptive statements are clearly material if they influenced users’

26

decision to use Facebook or would have been considered important by users. See TSC Industries,

27

426 U.S. at 449; Southland Sod Farms, 108 F.3d at 1139.

28

64
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

Moreover, Facebook ignores Consumers’ allegations that Facebook’s representations about

2

its data privacy practices were essential to creating Facebook’s “realness.” Although Facebook

3

encouraged users to share their real-world identities and personal information, Facebook knew that

4

it could only induce users to provide their identities and information if it “promise[d] privacy

5

protection.” CC ¶ 46. Indeed, the original version of Facebook “was closed to all but those users

6

who could validate their own real-world identities, such as by verifying that their identities were

7

legitimate via an e-mail address issued by an organization, such as a university or firm.” Id. ¶ 48.

8

Facebook’s “Secret Sauce” report expressly recognized that “‘[u]sers will share more information

9

if given more control over who they are sharing with and how they share.’” Id. ¶ 109 (internal
citations omitted). As Mr. Zuckerberg put it, Facebook’s privacy protections made Facebook “one

11

United States District Court
Northern District of California

10

of the only services on the web where people are sharing pretty personal and intimate

12

information.” Id. ¶ 130. Thus, Facebook’s perceived “realness” would not have been possible

13

without promises of data privacy.

14

For all these reasons, Consumers adequately allege that Facebook’s false representations

15

about its data privacy practices were “clearly material.” Accordingly, Consumers have adequately

16

alleged that Facebook’s false representations about its data privacy practices constitute

17

exclusionary conduct under the meaning of the Sherman Act.

18
19

Consumers Adequately Allege Causal Antitrust Injury
Consumers allege that Facebook’s monopolization of the Social Network and Social Media

20

Markets has harmed users because, without competition, Facebook can extract additional

21

“personal information and attention” from users. CC ¶¶ 220–22. Facebook contends that

22

Consumers have not adequately alleged causal antitrust injury because “lost ‘information and

23

attention’ is not a cognizable injury,” because users’ alleged injury is speculative, and because

24

users’ alleged injury was “not caused by lost competition.” Mot. at 28–31. For the reasons below,

25

the Court rejects Facebook’s arguments.

26

To plead causal antitrust injury, a plaintiff must allege that the defendant’s unlawful

27

conduct caused an injury “that flows” from the unlawful conduct and that is “the type the antitrust

28

65
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

laws were intended to prevent.” Somers v. Apple, Inc., 729 F.3d 953, 963 (9th Cir. 2013). A

2

cognizable antitrust injury includes harm to a plaintiff’s “business or property.” 15 U.S.C. § 15.

3

Business is “that which occupies the time, attention, and labor of [a plaintiff] for the purpose of

4

. . . pecuniary reward.” Fine v. Barry & Enright Prods., 731 F.2d 1394, 1397 (9th Cir. 1984). In

5

turn, “property” is a “broad and inclusive” term that includes “anything of material value owned or

6

possessed.” Reiter v. Sonotone Corp., 442 U.S. 330, 338 (1979).

7

Because Consumers explain in detail that “information and attention” has significant
“material value,” Consumers adequately allege harm to their “business or property.” Reiter, 442

9

U.S. at 338. Consumers allege that their “information and attention” has “material value” because

10

Facebook sells users’ “information and attention to third parties, including advertisers.” CC ¶ 222.

11

United States District Court
Northern District of California

8

Specifically, “Facebook monetizes user information through targeted advertising, which generated

12

most of the company’s $55.7 billion in revenues in 2018.” Id. ¶ 150. In other words, users

13

provide significant value to Facebook by giving Facebook their information—which allows

14

Facebook to create targeted advertisements—and by spending time on Facebook—which allows

15

Facebook to show users those targeted advertisements. If users gave Facebook less information or

16

spent less time on Facebook, Facebook would make less money. Indeed, as Consumers point out,

17

Facebook “describes its massive advertising earnings in terms of average revenue per user

18

(‘ARPU’) in its public filings.” Id. ¶ 7. In 2019, for example, “Facebook reported that its ARPU

19

was over $41.00 per user in the United States and Canada.” Id. Thus, there is no doubt that

20

Consumers’ “information and attention” has “material value.” Reiter, 442 U.S. at 338; see also

21

United Food & Commercial Workers Local 1776 & Participating Emp’rs Health & Welfare Fund

22

v. Teikoku Pharma USA, Inc., 74 F. Supp. 3d 1052, 1070 (N.D. Cal. 2014) (explaining that “courts

23

need not restrict the definition of ‘payments’ . . . to cash”).

24

This Court’s recent decision in Brown v. Google, No. 20-CV-3664-LHK, 2021 WL

25

6064009 (N.D. Cal. Dec. 22, 2021), confirms that Consumers’ allegations are sufficient. In

26

Brown, the plaintiffs brought a claim under California’s Unfair Competition Law (“UCL”)

27

alleging that Google had unlawfully collected their browsing data. Id. at *14. Citing the UCL’s

28

66
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

requirement that a plaintiff must establish “lost money or property as a result of the unfair

2

competition,” Google moved to dismiss the UCL claim and argued that the plaintiffs had not

3

“plausibly alleged that they ‘lost money or property’ as a result of Google’s conduct.” Id. (citing

4

Cal. Bus. & Prof. Code § 17204). The Court rejected this argument because the plaintiffs had

5

“allege[d] that the ‘cash value’ of the data which Google collected ‘can be quantified’ and that

6

there is an active market for such data.” Id. at *15 (internal citation omitted). Specifically, the

7

plaintiffs cited a “recent study that found that internet users are willing to pay up to $52.00/year to

8

keep their browsing histories private” and identified several companies that were willing to pay

9

users for browsing data. Id.

10

Similarly, in the instant case, Consumers allege that users’ “information and attention” has

United States District Court
Northern District of California

11

a cash value, which can be approximated by Facebook’s reported ARPU. CC ¶ 7. Additionally,

12

like the plaintiffs in Brown, Consumers identify examples of companies that have been willing to

13

pay users for information and attention. See id. ¶¶ 223–24. Indeed, Facebook itself paid certain

14

users “up to $20.00 per month in return for access to those users’ emails, private messages in

15

social media apps, photos and videos, web browsing and search activity, and even location

16

information.” Id. Thus, Consumers have adequately alleged that, by providing Facebook with

17

their information and attention, they “lost money or property.” Brown, 2021 WL 6064009, at *16.

18

Indeed, outside the antitrust context, numerous courts have recognized that plaintiffs who

19

lose personal information have suffered an economic injury. See Calhoun v. Google LLC, 526 F.

20

Supp. 3d 605 (N.D. Cal. 2021) (“[T]he Ninth Circuit and a number of district courts, including this

21

Court, have concluded that plaintiffs who suffered a loss of their personal information suffered

22

economic injury and had standing.”); In re Marriott Int’l, Inc., Customer Data Sec. Breach Litig.,

23

440 F. Supp. 3d 447, 461 (D. Md. 2020) (“[T]he growing trend across courts that have considered

24

this issue is to recognize the lost property value of this information.”); In re Yahoo! Inc. Customer

25

Data Sec. Breach Litig., No. 16-MD-02752-LHK, 2017 WL 3727318, at *13 (N.D. Cal. Aug. 30,

26

2017) (holding that plaintiffs had adequately alleged injury in fact based on the loss of value of

27

their personal information); In re Anthem, Inc. Data Breach Litig., No. 15-MD-02617-LHK, 2016

28

67
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

WL 3029783, at *14 (N.D. Cal. May 17, 2016) (concluding that the plaintiffs had plausibly

2

alleged injury from the loss of value of their personal information). Thus, there is ample support

3

for Consumers’ argument that lost information and attention has material value.

4

Additionally, Consumers adequately allege that their injury “flows” from Facebook’s
monopolization of the Social Network and Social Media Markets. Somers, 729 F.3d at 963.

6

Consumer allege that, had Facebook not eliminated competition in these markets, Consumers

7

would have been able to “select a social network or social media application which offers

8

consumers services that more closely align the consumers’ preferences, such as with respect to the

9

content displayed, quantity and quality of advertising, and options regarding data collection and

10

usage practices.” CC ¶ 226. To support this allegation, Consumers point out that, in competitive

11

United States District Court
Northern District of California

5

markets, some companies pay users for their data. For example, “[w]hen consumers agree to use

12

Microsoft’s ‘Bing’ search engine and allow Microsoft to collect their data, Microsoft . . .

13

compensates consumers with items of monetary value.” Id. ¶ 223.

14

Furthermore, Consumers allege that, had Facebook not eliminated competition,

15

“Consumers could have benefitted from Facebook’s social network and social media offerings

16

without having to surrender as much personal data to Facebook and other third parties that use

17

Facebook for app development or targeted advertising.” Id. ¶ 226. Indeed, during the period of

18

time when Google+ posed a competitive threat to Facebook, a Facebook executive stated that it

19

would be unwise to remove privacy protections because “IF ever there was a time to AVOID

20

controversy, it would be when the world is comparing our offerings to G+.” Id. ¶ 134.

21

Taken together, these allegations establish that Consumers have suffered an injury because

22

Facebook “detrimentally changed the market make-up and limited consumers’ choice to one

23

source of output.” Glen Holly Ent., Inc. v. Tektronix Inc., 343 F.3d 1000, 1010–11 (9th Cir.

24

2003). Accordingly, Consumers have adequately alleged that their injury “flows” from

25

Facebook’s anticompetitive conduct. Id.

26
27
28

For all these reasons, the Court concludes that Consumers have adequately alleged causal
antitrust injury. Thus, the Court DENIES Facebook’s motion to dismiss Consumers’ data privacy
68
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1
2
3

claims.
Consumers’ Request for Injunctive Relief as a Remedy for Consumers’ Data
Privacy Claims Is Not Barred by the Doctrine of Laches
Consumers seek two forms of injunctive relief related to their data privacy claims: “(i) an

4
order prohibiting Facebook from continuing to engage in the wrongful acts described herein; and
5
(ii) requiring Facebook to engage third-party auditors to conduct audits and evaluations of
6
7
8
9
10

United States District Court
Northern District of California

11
12
13

Facebook’s data privacy practices, commercial surveillance, and acquisition conduct, and ordering
them to promptly correct any problems or issues detected by these auditors.” CC ¶ 318(b).
Facebook argues that the doctrine of laches bars this injunctive relief because Facebook’s 2011
“settlement with the FTC regarding the privacy representations at issue was . . . public.” Mot. at
8. For the reasons below, the Court rejects Facebook’s argument.
“Claims for injunctive relief . . . are subject to the equitable defense of laches.” Oliver,
751 F.3d at 1085. “Laches is an equitable time limitation on a party’s right to bring suit, resting
on the maxim that one who seeks the help of a court of equity must not sleep on his rights.”

14
Jarrow Formulas, Inc. v. Nutrition Now, Inc., 304 F.3d 829, 835 (9th Cir. 2002) (internal citations
15
16
17

and quotations omitted). To invoke a laches defense, the defendant must show that: (1) “the
plaintiff knew (or should have known) of the allegedly [unlawful] conduct” yet delayed “the
initiation of the lawsuit,” and (2) the delay was unreasonable. Danjaq LLC v. Sony Corp., 263

18
F.3d 942, 952, 954 (9th Cir. 2001).
19

In antitrust cases, “the four-year statute of limitations in 15 U.S.C. § 15b ‘furnishes a

20
guideline for computation of the laches period.’” Samsung Elecs., 747 F.3d at 1205 (quoting Int.
21

Tel & Tel. Corp. v. Gen. Tel. & Elec. Corp., 518 F.2d 913, 926 (9th Cir. 1975)). “Therefore, in

22
applying laches, [the Ninth Circuit] look[s] to the same legal rules that animate the four-year
23
24
25

statute of limitations.” Oliver, 751 F.3d at 1086.
Thus, because Consumers’ data privacy claims are timely, “the strong presumption is that
laches is inapplicable.” See Jarrow Formulas, Inc., 304 F.3d at 835. Facebook’s one sentence

26
argument is insufficient to rebut that presumption. Moreover, Facebook fails to explain why
27
28

69
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

Consumers’ knowledge of the 2011 FTC settlement would prevent Consumers from seeking

2

injunctive relief based on the numerous false representations which Facebook made after the 2011

3

FTC settlement. See pp. 49–51, supra.

4
5
6

Thus, the Court rejects Facebook’s argument that Consumers may not seek injunctive
relief related to their data privacy claims.
C. The Court Grants Facebook’s Motion to Dismiss Consumers’ and Advertisers’
“Copy, Acquire, Kill” Claims With Leave to Amend

7
According to Consumers and Advertisers, Facebook maintained monopoly power through
8

a multipart strategy called “Copy, Acquire, Kill.” Specifically, Facebook identified particularly

9
popular mobile social media applications by using Onavo, a spyware application that was
10
disguised as a security and encryption application, but that deceptively harvested information

United States District Court
Northern District of California

11

about its users’ mobile applications. Using this deceptively obtained information, Facebook

12
executed a three part strategy to eliminate potential competitors. First, Facebook copied several
13

potential competitors’ products. Second, Facebook acquired numerous potential competitors,

14
including several of its most successful potential competitors. Third, Facebook enticed thousands
15

of potential competitors to build their products using Facebook’s “Platform” then removed access

16
to the Platform.
17

Consumers and Advertisers both allege that Facebook’s “Copy, Acquire, Kill” strategy

18
violated Section 2 of the Sherman Act. Additionally, both are clear that they are not challenging
19
20
21
22
23
24
25
26

individual aspects of the “Copy, Acquire, Kill” strategy. Consumers state that they are not
challenging “single acquisitions in isolation,” Opp. at 24, and that their theory of liability “do[es]
not depend on any ‘duty to deal’ between Facebook and ‘third-party app developers,’” Opp. at 27
(emphasis in original). Instead, Consumers challenge “Facebook’s overall anticompetitive
scheme.” Opp. at 25. Similarly, Advertisers state that they are “not assert[ing] claims that impose
upon Facebook a duty to deal with competitors.” Opp. at 27. Instead, Advertisers “allege[] a
multi-part scheme that includes Platform-based conduct.” Id.
Thus, Consumers and Advertisers rely on what some courts have “referred to as a ‘course

27
28

70
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

of conduct’ or ‘monopoly broth’ theory” of antitrust liability. New York v. Facebook, Inc., No.

2

20-CV-3589-JEB, 2021 WL 2643724, at *26 (D.D.C. June 28, 2021) (quoting In re Intuniv

3

Antitrust Litig., 496 F. Supp. 3d 639, 680 (D. Mass. 2020)). That theory gets its name from City of

4

Mishawaka v. American Electric Power Co., Inc., 616 F.2d 976, 986 (7th Cir. 1980), in which the

5

Seventh Circuit affirmed the district court’s decision to analyze the aggregate effect of a series of

6

anticompetitive acts committed by the defendant. The Seventh Circuit rejected the defendant’s

7

argument that the district court had erred by failing to analyze each act separately and stated that

8

“[i]t is the mix of the various ingredients of utility behavior in a monopoly broth that produces the

9

unsavory flavor.” Id.

10

In City of Anaheim v. Southern California Edison Co., 955 F.2d 1373, 1376, 1378 (1992),

United States District Court
Northern District of California

11

the Ninth Circuit endorsed this “monopoly broth” theory and stated that “it would not be proper to

12

focus on specific individual acts of an accused monopolist while refusing to consider their overall

13

combined effect.”

14

Similarly, in LePage’s Inc. v. 3M, 324 F.3d 141 (3d Cir. 2003), the Third Circuit held that

15

it was necessary to analyze the aggregate anticompetitive effect of the defendant’s bundled

16

discount and exclusive dealing contracts, rather than the anticompetitive effect of each individual

17

contract.

18

Under a “monopoly broth” theory of liability, a plaintiff “can state a Section 2 claim by

19

alleging a series of practices that are anticompetitive, even if some of the activities would be

20

lawful if viewed in isolation.” Free FreeHand Corp. v. Adobe Systems Inc., 852 F. Supp. 2d 1171,

21

1180 (N.D. Cal. 2012) (citing City of Anaheim, 955 F.2d at 1376); see also Intuniv, 496 F. Supp.

22

3d at 680 (explaining that “a plaintiff can allege a series of actions that when taken together make

23

out antitrust liability even though some of the individual actions, when viewed independently, are

24

not all actionable”); Tele Atlas N.V. v. NAVTEQ Corp., No. C-05-01673 RS, 2008 WL 4911230, at

25

*2 (N.D. Cal. Nov. 13, 2008) (“To appreciate the effect of otherwise lawful acts, the jury must

26

consider the acts’ aggregate effect.”).

27
28

Facebook argues that Advertisers’ and Consumers’ “Copy, Acquire, Kill” claims are
71
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

barred by the statute of limitations. See Mot. at 6–7, 8–9. Additionally, Facebook argues that

2

both “Copy, Acquire, Kill” claims must be dismissed because (1) copying competitors’ products is

3

not anticompetitive; (2) Facebook’s acquisitions were not exclusionary; and (3) Facebook did not

4

have a duty to give potential competitors access to Facebook’s “Platform.” Id. at 22–27.

5

Facebook also argues that Consumers’ request for injunctive relief is barred by the doctrine of

6

laches. Id. at 8–9. For the reasons below, the Court concludes that Consumers’ and Advertisers’

7

“Copy, Acquire, Kill” claims are barred by the statute of limitations. Accordingly, the Court need

8

not reach the remainder of Facebook’s arguments.

9

The Court first describes Consumers’ and Advertisers’ allegations regarding Facebook’s
“Copy, Acquire, Kill” strategy. Second, the Court explains why Consumers’ and Advertisers’

11

United States District Court
Northern District of California

10

“Copy, Acquire, Kill” claims are untimely.

12
13

Consumers and Advertisers Allege that Facebook’s “Copy, Acquire, Kill”
Strategy Allowed Facebook to Maintain Monopoly Power
Below, the Court describes Facebook’s “Copy, Acquire, Kill” strategy in detail. First, the

14
Court explains why Facebook viewed mobile social media applications as the biggest competitive
15
threats. Next, the Court describes how Facebook deceptively used Onavo to determine which
16
mobile social media applications were the most popular. In the following sections, the Court
17

describes each part of Facebook’s strategy. Although Consumers and Advertisers allege the same

18
basic facts, Advertisers include significantly more detail about some aspects of Facebook’s
19

strategy. Accordingly, the Court notes where Consumers’ and Advertisers’ allegations differ.

20
21

Facebook Realized that Mobile Social Media Applications Posed the
Biggest Competitive Threat to Facebook

22

Advertisers explain that, besides MySpace and Google+, the biggest threat to Facebook’s

23

monopoly power came from mobile social media applications. In 2007, shortly before Facebook

24

overtook Myspace, Apple Inc. (“Apple”) released its first iPhone. AC ¶ 84. On July 10, 2008,

25

Apple opened the Apple App Store (“App Store”), which provided a marketplace where

26

developers could sell applications that would run on iPhones. Id. ¶ 85. Accordingly, mobile

27

applications proliferated rapidly. Id. ¶ 87. By September 2008, the App Store had 100 million

28

72
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1
2

downloads, and by 2009, it had 1 billion. Id. ¶ 86.
This development was a problem for Facebook. Unlike desktop alternatives to Facebook,

3

mobile social media applications “presented their own specialized value propositions.” Id. ¶ 90.

4

Streamlined, single-purpose applications were easier to use on mobile devices than Facebook’s

5

platform, which was not easily transferrable to a mobile version. Id. ¶¶ 94–95. Indeed, the first

6

version of Facebook’s mobile application was buggy, prone to crashes, and painfully slow. Id.

7

¶ 95. Thus, because it was clear “that within five years, the number of users who accessed the

8

Internet from mobile devices would surpass the number who accessed it from PCs,” Facebook

9

knew that mobile applications would continue to peel away Facebook’s users. Id. ¶ 92. In its
2012 annual report filed with the Securities Exchange Commission (“SEC”), Facebook disclosed

11

United States District Court
Northern District of California

10

that “[g]rowth in the use of Facebook through our mobile products for a substitute for use on

12

personal computers may negatively affect our revenue and financial results.” Id. ¶ 98.

13

Although Consumers do not provide the same level of detail, Consumers allege that

14

Facebook was concerned with the competitive threat posed by mobile applications and was intent

15

on finding a way to measure this threat. According to Consumers, Facebook initially “used its

16

own internal data and data from Comscore, a data analytics and measurement firm, to track the

17

growth of competitive threats.” CC ¶ 159. However, in April 2012, Facebook Director of Growth

18

Javier Olivan emailed Facebook CEO Mark Zuckerberg about improving Facebook’s “competitive

19

research.” Id. ¶ 160. Mr. Olivan stated: “I keep seeing the same suspects (Instagram,

20

pinterest, . . . ) both on our competitive radar . . . I think having the exact data about their users

21

engagement . . . would help us make more bold decisions on whether they are friends or foes.

22

Back to your thread about ‘copying’ vs. ‘innovating’ we could also use this info to inspire our next

23

moves.” Id. Mr. Zuckerberg responded: “Yeah, let’s do it.” Id. ¶ 161.

24
25
26

Facebook Deceptively Used Onavo Spyware to Obtain Information About
Which Mobile Social Media Applications Were Popular
After Facebook realized that mobile social media applications threatened Facebook’s
market power, Facebook engaged Onavo, an “Israeli mobile web analytics company,” to

27
28

73
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

determine the precise scope of this threat. CC ¶ 162; AC ¶ 247. Onavo had a suite of mobile

2

applications, including a virtual private network (“VPN”) application called “Onavo Protect.” CC

3

¶¶ 162, 166; AC ¶ 252. As Consumers explain, a VPN typically “provides security and encryption

4

to a user” by “scrambling” the data sent from a user’s phone to internet servers. CC ¶ 162.

5

Indeed, Onavo billed Onavo Protect, which was downloaded over 33 million times, as a tool that

6

mobile phone owners could use to keep their “data safe” and “[a]dd an extra layer of protection to

7

all [their] mobile data traffic.” Id. ¶ 166; AC ¶ 252.

8
9

Regardless of whether Onavo Protect functioned as a VPN, Onavo Protect was valuable to
Facebook for another reason: Onavo Protect operated as a spyware program by “monitor[ing] all
web and mobile application traffic on a user’s mobile device.” CC ¶ 166; AC ¶ 252. Indeed,

11

United States District Court
Northern District of California

10

Onavo Protect collected users’ data “whether their screens were on or off, whether they used WiFi

12

or cellular data, and even when the VPN was turned off.” CC ¶ 210; AC ¶ 261.

13

As relevant here, Onavo Protect collected information about which social media

14

applications users were accessing most frequently. Consumers allege that Onavo collected the

15

following information: “every app the user has accessed”; “the number of seconds the user spent

16

in the app per day”; “the percent of time the user spent in a specific app out of their total mobile

17

usage time”; “the actions taken by the user in each app”; and the user’s demographic information.

18

CC ¶ 165. In turn, Advertisers allege that Onavo kept track of “which apps and features people

19

were using in real time, how frequently they used the apps, and for how long.” AC ¶ 253.

20

Accordingly, Onavo permitted Facebook to assess “the two most important metrics for competing

21

mobile applications”: an application’s reach, i.e., the size of its user base, and an application’s

22

engagement, i.e., how frequently users use the application. Id. ¶ 249.

23

Facebook used data collected by Onavo to determine which mobile social media

24

applications were competitive threats. Advertisers allege that Facebook began paying Onavo for

25

this data in 2011. Id. ¶ 256. “By April 2013,” Facebook was “using Onavo to track Snapchat,

26

Pinterest, WhatsApp, Tumblr, Foursquare, Google, Path, Vine, Kik, Voxer, MessageMe, Viber,

27

GroupMe, Skype, Line, and Tango.” Id. ¶ 166. Although Consumers do not identify precisely

28

74
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

when Facebook engaged Onavo, Consumers allege that, by 2013, Facebook had been acquiring

2

data through Onavo “for years.” CC ¶ 164.

3

In October 2013, Facebook acquired Onavo and placed Onavo under the supervision of

4

Facebook’s “Growth team.” CC ¶ 164; AC ¶ 250. At that point, Onavo announced on its blog

5

that it “remain[ed] committed to the privacy of people who use our application, and that

6

commitment will not change.” AC ¶ 251.

7

From thereon, Onavo automatically provided all data it gathered directly to Facebook. See

8

CC ¶ 166; AC ¶ 253. However, users of Onavo Protect were unaware of Onavo’s association with

9

Facebook because “nowhere on the Onavo Protect website, the Apple App Store, the Google Play
Store, or in the advertising for Onavo Protect did Facebook or Onavo disclose to consumers . . .

11

United States District Court
Northern District of California

10

that Onavo Protect would not protect and keep secret users’ personal activity data, or that

12

Facebook or Onavo would use personal activity data collected from users for the commercial

13

benefit of Facebook or Onavo.” CC ¶ 232.
Facebook and Onavo also released a product called “Onavo Bolt,” “which locked apps

14
15

behind a passcode or fingerprint while it covertly surveilled users—and sent Facebook the

16

results.” AC ¶ 266. Onavo Bolt was “installed approximately 10 million times.” Id.

17

In 2018, Apple removed Onavo Protect from the App Store “because it violated Apple’s

18

rules prohibiting applications from using data in ways far beyond what is required to run the app

19

and provide advertising.” CC ¶ 209; AC ¶ 260. “In other words, because Onavo Protect was

20

leveraging far more data than any VPN could conceivably need, it was clear that the true purpose

21

of the app was to spy on Onavo users, and Apple would not allow it.” CC ¶ 209; AC ¶ 260.

22

Facebook attempted to circumvent this removal by repackaging Onavo Protect as an

23

application called “Facebook Research.” CC ¶ 211; AC ¶ 262. However, when Apple uncovered

24

this ruse in January 2019, Facebook immediately withdrew Facebook Research from the App

25

Store. CC ¶ 212; AC ¶ 263. In a subsequent investigation of Facebook Research, Apple

26

concluded that Facebook clearly had violated Apple’s rules for the App Store. CC ¶ 212; AC

27

¶ 264.

28

75
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1
2
3

Similarly, when the “surveillance functionality” of Onavo Bolt was discovered, Facebook
immediately “shut that app down.” AC ¶ 266.
By that time, however, Facebook already had used data acquired from Onavo to eliminate

4

potential competitive threats. As explained in the following sections, Facebook copied the

5

features of potential competitors identified through Onavo; acquired or attempted to acquire

6

potential competitors identified through Onavo; and weakened potential competitors identified

7

through Onavo by denying those competitors access to Facebook’s Platform.

8

Facebook Copied the Features of Applications Identified Through Onavo
and Acquired Several Applications Identified Through Onavo

9
Consumers and Advertisers both allege that, after identifying potential competitors through
10
Onavo, Facebook copied or acquired several of its most successful potential competitors. Overall,
11

United States District Court
Northern District of California

Facebook acquired at least 63 companies since its founding in 2004 to approximately April 22,
12
2021. CC ¶ 157. In 2014, a senior executive at Facebook described this acquisition strategy as a
13
14
15

“land grab.” Id. ¶ 98. Mr. Zuckerberg later boasted that Facebook “can likely always just buy any
competitive startups.” Id. Accordingly, Facebook grew from a single social network service to a
technology conglomerate. However, because all parties use the term “Facebook” when referring

16
both to the social network service and the conglomerate, the Court follows that convention.
17
Consumers and Advertisers highlight that Facebook copied or acquired six social media
18
19
20

applications. Both Consumers and Advertisers highlight Facebook’s acquisitions of Instagram
and WhatsApp and note that Facebook copied several of Snapchat’s core features. Consumers
also allege that Facebook copied Instagram’s core feature before acquiring Instagram, that

21
Facebook copied the core feature of Houseparty, and that Facebook acquired Giphy and tbh.
22

In the following sections, the Court discusses, in order, Facebook’s conduct with respect to

23
Instagram, Whatsapp, Snapchat, Houseparty, Giphy, and tbh.
24
Instagram
25
Instagram launched in October 2010 as a streamlined photo sharing system that
26
encouraged users to share a single photo at a time. See CC ¶¶ 186–87; AC ¶¶ 268, 273. As
27
28

76
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

explained by Advertisers, this feature was appealing to smartphone users who took photos often

2

and shared them individually. AC ¶ 278. By contrast, Facebook’s then-current photo sharing

3

system was designed for bulk uploads and thus appealed to digital camera users. Id. Accordingly,

4

because the increased popularity of smartphones meant that users were more likely to upload

5

individual photos, Instagram posed a unique threat to Facebook. Id.

6

By December 2010, Instagram had been downloaded 1 million times, and by March 2012,

7

Instagram had 27 million users. CC ¶¶ 187–88; AC ¶¶ 274–76. Although Instagram competed

8

only in the Social Media Market, Facebook believed that Instagram had the potential to move into

9

the Social Network Market because it could attract a large user base with its simple photo-sharing

10

United States District Court
Northern District of California

11

feature and then add other features after building up that base. CC ¶ 189; AC ¶ 279.
Thus, in 2011, Facebook started tracking Instagram through Onavo. CC ¶ 189; AC ¶ 268.

12

According to Consumers, Facebook initially attempted to lessen the appeal of Instagram by

13

copying Instagram’s core feature. In June 2011, Facebook launched Facebook Camera, “a

14

standalone app allowing users to shoot, filter, and share photos from their mobile device.” CC ¶

15

191. One Facebook employee called Facebook Camera “an Instagram clone.” Id. Other members

16

of the Facebook Camera team later admitted that their job was to “kill” Instagram. House Report

17

at 153 n.882.

18

After Facebook Camera failed to catch on, Facebook decided to take more drastic

19

measures. In February 2012, Facebook CEO Mark Zuckerberg met with Facebook CFO David

20

Ebersman and proposed that Facebook acquire Instagram. House Report at 151–52. Mr.

21

Ebersman asked Mr. Zuckerberg whether the purpose of the acquisition “would be to: (1)

22

neutralize a potential competitor; (2) acquire talent; or (3) integrate Instagram’s product with

23

Facebook’s to improve its service.” Id. at 152.

24

Mr. Zuckerberg responded that the purpose would be “a combination of (1) and (3).” Id.

25

Displaying his knowledge of the relevant barriers to entry, Mr. Zuckerberg stated that “there are

26

network effects around social products and a finite number of different social mechanics to

27

invent.” Id. Accordingly, Mr. Zuckerberg explained, “[o]nce someone wins at a specific

28

77
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

mechanic, it’s difficult for others to supplant them without doing something different.” Id. Mr.

2

Zuckerberg added that “one way of looking at this is that what we’re really buying is time”

3

because, “[e]ven if some new competitors spring[] . . . up, buying Instagram, Path, Foursquare, etc

4

. . . now will give us a year or more to integrate their dynamics before anyone can get close to their

5

scale again.” Id.
In March 2012, Mr. Zuckerberg told Facebook CTO Mike Schroepfer that “acquiring

7

Instagram would provide the company with ‘[i]nsurance for Facebook’s main product.” AC

8

¶ 284. Then, in an April 5, 2012, meeting with Facebook employees, Mr. Zuckerberg stated that,

9

because of “Instagram’s rapid growth,” Facebook would need to “dig [itself] out of a hole.” Id.

10

¶ 285. He added that the “bad news is that [Instagram is] growing really quickly, they have a lot

11

United States District Court
Northern District of California

6

of momentum, and it’s going to be tough to dislodge them.” Id. ¶ 285; CC ¶ 190.

12

In April 2012, Facebook offered to acquire Instagram for $1 billion. CC ¶ 192–94; AC

13

¶ 286. Consumers allege that, after Facebook made its initial offer, Mr. Zuckerberg threatened

14

Instagram CEO Kevin Systrom that “[h]ow we engage now will determine how much we’re

15

partners vs. competitors down the line.” CC ¶ 193. Following this conversation, Mr. Systrom

16

spoke with a former Facebook executive, who stated that Mr. Zuckerberg would try to “destroy”

17

Instagram if Instagram rejected Facebook’s offer. Id.

18

Shortly thereafter, Instagram accepted Facebook’s offer. CC ¶ 194; AC ¶ 286. Advertisers

19

allege that, on March 6, 2019, Facebook announced that it would integrate Instagram and

20

Facebook so that users of Instagram could connect with users of Facebook. AC ¶ 524.

21
22

WhatsApp
WhatsApp is a messaging service that uses an internet connection, instead of a cellular

23

network connection, to send text messages and photographs between cell phones. CC ¶ 201; AC

24

¶ 305. Unlike other messaging applications, WhatsApp “monetized its service through

25

subscriptions for a nominal fee after the first year of use.” House Report at 156. Indeed, in a blog

26

post on the WhatsApp website, WhatsApp’s founders stated that they strongly opposed advertising

27

because “when advertising is involved you the user are the product.” Id. (emphasis in original).

28

78
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1
2
3
4
5

They further explained:
At every company that sells ads, a significant portion of their engineering team
spends their day tuning data mining, writing better code to collect all your personal
data, upgrading the servers that hold all the data and making sure that it’s all being
logged and collated and sliced and packaged and shipped out.”
Id. Indeed, in June 2012, WhatsApp stated in its privacy policy that it did “not collect names,
emails, location data, or the contents of messages sent through WhatsApp.” Id.

6
By 2011, WhatsApp was one of the top twenty paid iPhone applications and by February
7
2013, WhatsApp had 200 million users. AC ¶¶ 307–09. After WhatsApp achieved initial success,
8

Facebook started tracking WhatsApp through Onavo and discovered that WhatsApp “held third

9
place in terms of user reach among mobile messenger apps for iPhone in the U.S. as of April
10

United States District Court
Northern District of California

11
12
13
14

2013.” Id. ¶ 310; CC ¶ 202. Data collected through Onavo also showed that users of WhatsApp
sent more than double the messages of users of Facebook’s similar Messenger service. AC ¶ 311;
CC ¶ 202. As another example of WhatsApp’s popularity, Onavo data showed that “WhatsApp
reached 99.9% of the smartphone population in Spain.” AC ¶ 316. According to Advertisers,
Facebook understood this data to mean that WhatsApp had the “unique potential” to overcome the

15
barriers to entry faced by most aspiring social network companies. Id. ¶¶ 320–23. Thus, as with
16
Instagram, Facebook suspected that WhatsApp would pivot to being a social networking service
17
after solidifying its core user base. See id.
18
In 2014, Facebook acquired WhatsApp. Although WhatsApp had lost $232 million
19
20
21
22
23
24
25
26

during the first six months of 2014, Facebook’s final purchase price was $22 billion. CC ¶ 203;
AC ¶ 321. Accordingly, Consumers and Advertisers allege, Facebook’s acquisition of WhatsApp
did not make “economic sense” unless it was aimed at preventing WhatsApp from becoming a
competitor to Facebook. CC ¶ 203; AC ¶ 321. Indeed, Facebook had paid “thousands of times
WhatsApp’s revenues” in order “to acquire a money-losing company that created software
functionality Facebook itself already had as part of its own products.” AC ¶ 321.
After the acquisition was finalized, Mr. Zuckerberg stated: “[w]e are absolutely not going
to change plans around WhatsApp and the way it uses user data.” House Report at 157. Shortly

27
28

79
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

thereafter, the FTC advised Facebook “that WhatsApp ‘must continue to honor’ its privacy data

2

security commitments to its users.” Id.
When Facebook acquired WhatsApp in 2014, Facebook represented to European

3
4

regulators that Facebook did not have the capability “to establish reliable automated matching

5

between Facebook’s users’ accounts and WhatsApp users’ accounts.” AC ¶ 518. Advertisers

6

equate this representation to a statement that Facebook would not be able to integrate WhatsApp.

7

See id.

8
9

However, in an August 2016 update to WhatsApp’s “terms of service and privacy policy,”
Facebook announced that it had the capability to “link[] WhatsApp users’ phone numbers with
Facebook users’ identities.” Id. Then, in December 2016, the European Commission found that

11

United States District Court
Northern District of California

10

Facebook had misled European regulators in 2014 because the “possibility of automatically

12

matching Facebook and WhatsApp users’ identities already existed in 2014.” Id. On March 6,

13

2019, Facebook officially announced that it would begin the process of integrating Facebook and

14

WhatsApp so that users of WhatsApp could connect with users of Facebook. Id. ¶ 524.

15
16

Snapchat
Facebook also used Onavo to track Snapchat. CC ¶ 197; AC ¶ 166. Consumers explain

17

that Snapchat allows users “to send each other communications—including text, photos, and

18

videos—which appear only for a fixed period of time and then disappear.” CC ¶ 196. By 2013,

19

Snapchat was among the five most popular social media applications. Id. ¶ 202; AC ¶ 312.

20

Facebook attempted to acquire Snapchat in early 2013. Consumers allege that, before

21

Facebook made an offer to Snapchat, Facebook tried to weaken Snapchat’s position by launching

22

an identical product called “Poke.” CC ¶ 198. Then, when Facebook CEO Mark Zuckerberg met

23

with Snapchat founder Evan Spiegel, Mr. Zuckerberg intimated that Facebook would “crush”

24

Snapchat. Id. Spiegel, however, rejected Facebook’s $3 billion offer. Id. ¶ 199; AC ¶ 298.

25

Having failed to acquire Snapchat, Facebook continued trying to weaken Snapchat. In

26

“late 2016,” Facebook copied Snapchat’s “stories” feature, which allows a user to post a series of

27

images and videos that remain online for twenty-four hours. CC ¶ 200; AC ¶ 298. According to

28

80
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

Consumers, Facebook’s version of “stories” was more popular than Snapchat’s by April 2017.

2

CC ¶ 200.

3
4
5
6

Houseparty, Giphy, and tbh
Finally, Consumers allege that Facebook used Onavo to track at least three other social
media applications and copied or acquired all three applications.
First, Facebook copied Houseparty, a video chat application that referred to itself as “the

7

internet’s living room.” CC ¶ 204. Specifically, after Houseparty turned down Facebook’s

8

acquisition offer, “Facebook announced that its Messenger app would become a ‘virtual living

9

room.’” Id. ¶ 205. “Houseparty’s active user base fell by half between 2017 and 2018.” Id.

10

United States District Court
Northern District of California

11
12
13
14

Second, in October 2017, Facebook acquired tbh, an “anonymous social media app.” Id.
¶ 207. However, Facebook shut down tbh after less than a year. Id.
Third, in May 2020, Facebook acquired Giphy, an application that allows users to share
animated images called “GIFs.” Id. ¶ 206. Facebook paid $400 million for Giphy. Id.
Facebook Enticed Potential Competitors to Build Their Products Using
Facebook’s Platform Then Removed Access to the Platform

15
Consumers and Advertisers also allege that Facebook harmed potential competitors by
16

convincing them to build their products using Facebook’s Platform and then removing access to

17
the Platform. Additionally, Advertisers allege that Facebook lied to the public about the reasons
18
for removing access to the Platform. Below, the Court describes this conduct in detail.
19
20
21

Facebook Initially Allowed Third Party Developers to Use Facebook’s
Platform to Build Social Media Applications
In response to the competitive threat posed by mobile social media applications, Facebook

22

decided to increase the number of applications available through Facebook. However, instead of

23

building these applications itself, Facebook provided third party developers with access to the

24

Facebook Platform, a set of tools that allowed developers to build applications that worked in

25

conjunction with Facebook. CC ¶ 170; AC ¶ 99.

26
27
28

At the center of the Platform were Facebook’s custom Application Programming Interfaces
(“APIs”). CC ¶ 170; AC ¶ 101. In the computer programming field, an API is a prewritten piece
81
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

of software code that allows a programmer to execute a complex task without having to write the

2

code from scratch every time the programmer wants to perform the task. See Google LLC v.

3

Oracle Am., Inc., 141 S. Ct. 1183, 1191 (2021). Facebook’s APIs allowed programmers to

4

communicate with the Facebook network and obtain Facebook data. CC ¶ 170; AC ¶ 101.

5

Consumers and Advertisers highlight two Facebook APIs that were particularly important

6

for developing applications: the “Friends” API and the “Newsfeed” API. See CC ¶ 175; AC

7

¶ 126. The Friends API allowed a programmer to obtain information from Facebook about a

8

particular Facebook user’s friends. CC ¶ 174; AC ¶ 126. In turn, the “Newsfeed” API, which

9

Advertisers also refer to as the “Timeline API,” allowed a programmer to obtain information from

10

United States District Court
Northern District of California

11

Facebook about a user’s news feed posts. CC ¶ 174; AC ¶ 128.
Advertisers provide a detailed explanation of how developers used these APIs. First, in

12

accordance with Facebook’s original plan, many developers used these APIs to build applications

13

for Facebook itself. For example, developers built “social games that allowed [Facebook] users to

14

play with and against each other.” AC ¶ 114. These applications directly increased the value of

15

Facebook by stimulating user engagement. Id.; CC ¶ 174. Second, Facebook created a way for

16

developers building applications independent of Facebook to show advertisements to Facebook

17

users. AC ¶¶ 118–19. Such developers could use the APIs to identify Facebook users who might

18

like the developers’ applications and show those users advertisements for the applications. Id.

19

Every time a user downloaded an application because of an advertisement, the developer would

20

pay Facebook a fee. Id. ¶ 120.

21

According to Advertisers, Facebook recognized that giving third parties access to the APIs

22

had risks. Indeed, in a 2012 report to investors, Facebook stated that “Platform partners may use

23

information shared by our users through the Facebook Platform in order to develop products or

24

features that compete with us . . . As a result, our competitors may acquire and engage users at the

25

expense of the growth or engagement of our user base, which may negatively affect our business

26

and financial results.” Id. ¶ 105.

27
28

Initially, Facebook determined that the benefits outweighed the risks. The 2012 report
82
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

explained that “[e]ngagement with our Platform developers’ apps and websites can create value

2

for Facebook in multiple ways: our Platform supports our advertising business because apps on

3

Facebook create engagement that enables us to show ads; our Platform developers may purchase

4

advertising on Facebook to drive traffic to their apps and websites; Platform developers use our

5

Payment infrastructure to facilitate transactions with users on personal computers; Platform apps

6

share content with Facebook that makes our products more engaging; and engagement with

7

Platform apps and websites contributes to our understanding of people’s interest and preferences,

8

improving our ability to personalize content.” Id. ¶ 107.

9

However, as described in the following sections, Facebook eventually decided that the
short-term benefits of allowing third party developers to access the Platform were not worth the

11

United States District Court
Northern District of California

10

risk that one of these developers might leverage Facebook’s data to build a service that could

12

compete directly with Facebook.

13

In 2012, Facebook Covertly Decided to Threaten to Block Certain
Selected Competitors

14
Towards the end of 2011 and continuing into 2012, Facebook CEO Mark Zuckerberg held
15
discussions with other top Facebook executives about how to prevent third party developers from
16
using Facebook APIs to build services that could compete with Facebook. CC ¶ 173; AC ¶ 126.
17
Using data obtained through Onavo, Mr. Zuckerberg and other Facebook executives identified
18
Line, WeChat, and Instagram as applications that relied on Facebook APIs and posed significant
19
threats to Facebook. CC ¶ 173; AC ¶ 123.
20
Advertisers allege that, during these 2011 and 2012 meetings, Mr. Zuckerberg and the
21

other Facebook executives “proposed modifying the APIs” to make them less useful. AC ¶ 127.

22
Specifically, the group proposed modifying the Friends API so that a developer could not access
23
24

“information about a user’s friends (and the friends of their friends) unless that developer’s
application was already installed by a user’s friends to begin with.” Id. The group also proposed

25
removing the Newsfeed API entirely. Id. ¶ 128.
26

Additionally, in mid-2012, Facebook began charging “major competitors” a premium for

27
28

83
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

access to Facebook APIs. CC ¶ 176; AC ¶ 133. Facebook required “major competitors” to sign

2

“reciprocity” agreements which obligated these competitors to provide Facebook with their own

3

data. CC ¶ 176; AC ¶ 133. Most notably, Facebook required “reciprocity” from “Twitter,

4

Instagram, Pinterest, and Foursquare.” CC ¶ 176; AC ¶ 134.
Advertisers describe this tactic in further detail. Advertisers allege that, in August 2012,

5
6

Facebook created an expanded list of competitors and proposed either demanding reciprocity from

7

or banning these competitors. AC ¶ 136. Advertisers highlight an email from Facebook VP of

8

Business and Marketing Partnerships David Fischer which observed that a “large part of the

9

market for [Facebook’s] network will come from current and potential competitors.” Id. Mr.

10

Fischer divided “current and potential competitors” into the following categories:

United States District Court
Northern District of California

11

•

Social network apps (Google+, Twitter, Path, etc.)

12

•

Photo sharing apps (Picasa, Flickr, LiveShare, Shutterfly, etc.)

13

•

Messaging apps (WhatsApp, Viber, Imo, KakaoTalk, etc.)

14

•

Local apps (Google+ local, Google Offers, Yelp, yp, etc.)

15

•

Social search apps (HeyStaks, Wajam, etc.)

16

•

Platforms (Google Play, Amazon, etc.)

17
18

Id.
In the Fall of 2012, Mr. Zuckerberg formalized a plan to “limit the ability for competitive

19

networks to use [Facebook’s] platform without a formal deal in place.” Id. ¶ 141 (internal citation

20

omitted). As detailed in an October 2012 email from Facebook VP for Engineering Michael

21

Vernal, the plan was to modify the key Friends and News Feed APIs to make them less useful, to

22

provide competitors with API access only if the competitors agreed to “formal deals,” and to

23

require data reciprocity from all developers. Id. ¶¶ 141–42. Mr. Vernal later characterized the

24

amount of data that Facebook would require from developers as “crazy.” Id. ¶¶ 147, 151. In

25

November 2012, Facebook VP of Global Operations Justin Osofsky provided the following

26

summary of the plan:

27
28

Policy changes: define competitive networks + require they have a deal with us,
84
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1
2
3
4
5
6
7
8
9
10

United States District Court
Northern District of California

11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28

regardless of size. Maintain size-based thresholds for all other developers to force
business deals. Require data reciprocity for user extended info to ensure we have
richest identity.
Id. ¶ 143.
On November 19, 2012, Mr. Zuckerberg “broadly announced” to Facebook employees
“his decision to block competitors or require full data reciprocity for continued access.” Id. ¶ 145.
However, instead of announcing the decision to the public, Mr. Zuckerberg “decided to enforce
the decision selectively and covertly.” Id. ¶ 153. Indeed, when Mr. Osofsky “pleaded with” Mr.
Vernal “to make an announcement that would send a clear signal to developers,” Mr. Vernal
“responded that” Mr. Zuckerberg “had already rejected that approach.” Id. ¶ 157.
In January 2013, Mr. Zuckerberg initiated the plan and “ordered that WeChat, Kakao, and
Line be restricted from using the Friends and News Feed APIs and even from advertising on
Facebook[].” Id. ¶ 154. Mr. Fischer protested that “blocking competitors even from the
advertising platform was irrational and unworkable” and stated that Facebook “should be secure
enough in the quality of [its] products to enable them to compete effectively in the open
marketplace.” Id. ¶ 155. Mr. Zuckerberg was not swayed. Indeed, later in January, Mr.
Zuckerberg “ordered that Facebook competitor Vine be ‘shut down’ from Facebook’s API and
Platform, including from advertising.” Id. ¶ 158. Over the next few months, Facebook demanded
reciprocity agreements from Yahoo, Amazon, and Refresh.io. Id. ¶¶ 162, 164, 165.
The engineers responsible for Facebook’s platform expressed serious concerns about Mr.
Zuckerberg’s plan. For example, in March 2013, Facebook Director of Engineering for Platform
Douglas Purdy stated that the “post facto” nature of this conduct was unfair to the developers who
had relied on Facebook. Id. ¶ 163. Specifically, Mr. Purdy stated: “The way we are structured
today, you build an app on FB and then launch and then we may just shut you down, harming
users and the developer.” Id. Additionally, in August 2013, an engineer named Bryan Klimt
stated in an email to Facebook Head of Developer Products Ilya Sukhar that the Friends API “is so
core to the developer experience . . . that removing it would be ridiculous on its face.” Id. ¶ 177.
According to Mr. Klimt, there were no “solid technical reasons” to remove API access and the
85
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

only purpose was a “protectionist grab to make sure no one else can make a competing social

2

network.” Id. ¶ 180. Mr. Klimt’s email further asserted: “We’re removing the core API in our

3

developer platform. Out of concerns that someone will steal our social network product.” Id.

4
5
6
7
8

In 2015, Facebook Began to Grant API Access on a Discriminatory
Basis Despite Contrary Public Statements
Consumers and Advertisers both allege that, in 2015, Facebook “cut off all public access to
the Friends and News Feed APIs.” CC ¶ 180; AC ¶ 225. However, Facebook continued to
provide access to certain developers on the condition that those developers sign “Whitelist and
Datasharing Agreements.” CC ¶ 181; AC ¶ 225. Although these agreements allowed chosen

9
developers to access non-public APIs, the agreements obligated developers to make advertising
10
purchases from Facebook. CC ¶ 181; AC ¶¶ 217–18. According to Advertisers, Facebook made
11

United States District Court
Northern District of California

agreements in 2015 with the following developers: Tinder, Hinge, Netflix, Nissan, Lyft,
12
Microsoft, Hootsuite, and Walgreens. AC ¶¶ 221, 223, 225. Although Advertisers allege that
13
14
15

Facebook made similar agreements with “dozens” of other developers, Advertisers state that it is
impossible to know “the precise number and identity” of those developers. Id. ¶ 226.
Advertisers provide further details about the process that led to Facebook’s decision to

16
remove public access to the APIs. In Summer 2013, Facebook determined that over 40,000
17
applications were using Facebook APIs. Id. ¶ 169. Using data obtained through Onavo, Facebook
18
19
20
21
22
23

divided these applications into three categories: (1) applications that “may cause negative press” if
their access to APIs were shut down; (2) applications that “provide strategic value”; and (3)
applications that were “competitive” or “not useful.” Id. ¶ 170. As a direct result of this analysis,
Facebook immediately decided to restrict all “lifestyle apps” from using the Friends API because
Facebook was “ultimately competitive with all of them.” Id. ¶ 171. Additionally, Facebook
developed “a small list of strategic competitors that” Mr. Zuckerberg “personally reviewed” to

24
determine whether they would receive API access. Id. ¶ 189.
25
Facebook continued to keep these changes as secret as possible. Specifically, Mr.
26

Zuckerberg “decided to announce the API removal under the cover of a major change to the

27
28

86
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

Facebook Platform . . . which would be announced at the next Facebook F8 developers

2

conference.” Id. ¶ 192. Accordingly, “Facebook’s engineers were . . . instructed in September

3

2013 to bury the changes to the API.” Id. Indeed, “[a]lthough Facebook knew that the APIs were

4

going to be removed by the next F8 conference, it continued to tell developers to rely on them.”

5

Id. ¶ 194.

6

In January 2014, Facebook decided that it would be easier to remove API access from

7

“[t]he bulk of the 41,191 that relied on the Friends or News Feed APIs” and decide on an

8

individual basis which developers should be whitelisted. Id. ¶¶ 203–05. For example, “Facebook

9

decided that it would whitelist Tinder and other anointed dating apps and shut down the rest.” Id.

10

However, even after this decision was made, “Facebook continued to evangelize the APIs

United States District Court
Northern District of California

11

to developers.” Id. ¶ 206. Indeed, in January 2014, an engineer named George Lee reported to

12

Mr. Purdy and Mr. Vernal, Facebook employees were “still telling people to use” the APIs to

13

build products. Id. One Facebook employee remarked that the decision to remove the APIs

14

“seems a little unfair especially when our stance on some of these policies is that they’re about

15

ensuring trust[] and a great experience.” Id. ¶ 208. Facebook Head of Developer Products Ilya

16

Sukhar noted that developers would “get totally fucked,” id. at 187, by Facebook’s removal of the

17

APIs and began referring to the removal as the “switcharoo plan,” id. ¶ 207. In April 2014, Mr.

18

Vernal remarked to Mr. Sukhar that if the reasons for removing the APIs became public, there

19

would be a “high likelihood of breaking into jail.” Id. ¶ 210.

20

On April 30, 2014, Facebook announced the removal of the APIs at the bottom of a long

21

website post about changes to Facebook’s Login system. Id. ¶ 211. After describing in detail the

22

changes to Facebook’s Login system, Facebook stated: “In addition to the above, we are removing

23

several rarely used API endpoints; visit our changelog for details.” Id. However, the APIs were

24

not rarely used; five out of the top ten Facebook applications in December 2012 relied heavily on

25

the APIs. Id. ¶ 212.

26
27
28

The same day, Mr. Zuckerberg took the stage at the F8 Conference and emphasized that
Facebook’s Platform was a “stable mobile platform”:
87
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1
2
3
4
5
6
7

Id. ¶ 213. Mr. Zuckerberg did not mention that Facebook was removing API access. Id.
A year later, in April 2015, Facebook “cut off all public access to the Friends and News

8
9

Feed APIs” and began providing API access only to certain whitelisted developers. Id. ¶ 225; CC
¶ 180. According to Advertisers, this practice persisted until April 4, 2018, when Facebook

11

United States District Court
Northern District of California

10

removed access to APIs entirely. AC ¶ 241.
In November 2019, NBC News published internal Facebook documents which revealed to

12
13

the public Facebook’s motivations for removing API access. Id. ¶ 516.
Consumers’ and Advertisers’ “Copy, Acquire, Kill” Claims Are Untimely

14

Because Consumers and Advertisers filed their complaint on December 3, 2020, they must

15
16

allege that at least one “overt act” that was part of Facebook’s “Copy, Acquire, Kill” strategy

17

occurred after December 3, 2016. See Klehr, 521 U.S. at 189; 15 U.S.C. § 15b. They must also

18

allege that this overt act “meets two criteria: 1) It must be a new and independent act that is not

19

merely a reaffirmation of a previous act; and 2) it must inflict new and accumulating injury on the

20

plaintiff.” Samsung Elecs., 747 F.3d at 1202.
Consumers contend that three overt acts that were part of Facebook’s “Copy, Acquire,

21
22

Kill” strategy occurred after December 3, 2016. 4 First, Consumers contend that Facebook

23
24
25
26
27
28

Consumers’ complaint could be read as alleging that Facebook’s copying of Houseparty’s video
messaging feature occurred after December 3, 2016. Specifically, Consumers allege that, after
Facebook copied Houseparty, “Houseparty’s active user base fell by half between 2017 and
2018.” CC ¶ 205. However, Facebook contends that Consumers’ allegations with respect to
Houseparty are insufficient, Mot. at 13, and Consumers provide no response, see Opp. at 8.
Moreover, Consumers incorporate into their complaint an article which suggests that Facebook
made plans to copy Houseparty prior to December 3, 2016. See CC ¶ 164 n.137; Betsy Morris &
Deepa Seetharaman, The New Copycats: How Facebook Squashes Competition from Startups, The
88
4

Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

“deceptively obtain[ed] data” using Onavo after December 3, 2016. Opp. at 8 (citing CC ¶¶ 209–

2

13). Additionally, Consumers contend that Facebook “acquir[ed] additional competitors” after

3

December 3, 2016. Id. (citing CC ¶¶ 206–07). This contention refers to Facebook’s October 2017

4

acquisition of tbh and May 2020 acquisition of Giphy.
In turn, Advertisers contend that four overt acts that were part of Facebook’s “Copy,

6

Acquire, Kill” strategy occurred after December 3, 2016. First, similar to Consumers, Advertisers

7

contend that Facebook “spied on users’ mobile applications with Onavo spyware” after December

8

3, 2016. Opp. at 10 (citing AC ¶¶ 247–67). Second, Advertisers contend that Facebook

9

“anticompetitively cloned Snapchat features . . . culminating in the cloned release of Instagram’s

10

‘stories’ feature at the end of 2016.” Id. (citing AC ¶¶ 298–99). Third, Advertisers contend that

11

United States District Court
Northern District of California

5

Facebook discriminatorily denied access to the “Platform” “at least until April 4, 2018.” Id.

12

(citing AC ¶¶ 234–44). Fourth, Advertisers contend that Facebook “continued back-end

13

integration of Instagram and WhatsApp into 2020.” Id. (AC ¶¶ 517–26). In the alternative,

14

Advertisers contend that the limitations period should be tolled under the doctrine of fraudulent

15

concealment. Id. at 13–14.

16

Below, the Court addresses: (1) Consumers’ and Advertisers’ argument about Facebook’s

17

use of Onavo; (2) Consumers’ argument about Facebook’s acquisitions of tbh and Giphy; (3)

18

Advertisers’ argument about Facebook’s copying of Snapchat’s “stories” feature; (4) Advertisers’

19

argument about Facebook’s “Platform”; (5) Advertisers’ argument about Instagram and

20

WhatsApp; and (6) Advertisers’ fraudulent concealment argument.

21
22
23

Consumers and Advertisers Fail to Allege Any Facts That Allow the Court
to Evaluate Whether Facebook’s Use of Onavo After December 3, 2016
Inflicted a “New and Accumulating Injury”
Consumers and Advertisers both allege that Facebook’s use of Onavo Protect was an

24

integral part of the “Copy, Acquire, Kill” strategy and that Facebook collected data with Onavo

25

Protect after December 3, 2016. CC ¶¶ 209–13; AC ¶¶ 247–67. Thus, Consumers and

26
27
28

Wall Street Journal (Aug. 9, 2017), https://on.wsj.com/3qbF0tE. Thus, the Court need not address
Consumers’ allegations regarding Houseparty.
89
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

Advertisers contend, their “Copy, Acquire, Kill” claims are timely. Opp. at 8, 10. In response,

2

Facebook argues that Consumers and Advertisers fail to allege that Facebook’s “deceptive access

3

to data through Onavo is itself anticompetitive.” See Reply at 4.

4

With an important caveat, the Court agrees with Facebook. As an initial matter,

5

Facebook’s use of Onavo Protect was deceptive, and Facebook made deceptive statements about

6

Onavo Protect up until 2019. For example, Facebook and Onavo, which Facebook acquired in

7

2013, represented to users of Onavo Protect that Onavo Protect would keep their “data safe” and

8

“[a]dd an extra layer of protection to all [their] mobile data traffic.” CC ¶ 166; AC ¶ 252. To the

9

contrary, Onavo Protect “monitored all web and mobile traffic application on a user’s mobile
device” and sent this data to Facebook. CC ¶ 166; AC ¶ 252. Onavo Protect collected users’ data

11

United States District Court
Northern District of California

10

“whether screens were on or off, whether they used WiFi or cellular data, and even when the VPN

12

was turned off.” CC ¶ 210; AC ¶ 261. Facebook and Onavo also released a product called Onavo

13

Bolt, “which locked apps behind a passcode or fingerprint while it covertly surveilled users—and

14

sent Facebook the results.” AC ¶ 266.

15

In 2018, Apple removed Onavo Protect from the App Store for violating Apple’s “rules

16

prohibiting applications from using data in ways far beyond what is required to run the app.” CC

17

¶ 209; AC ¶ 260. Facebook attempted to circumvent this removal by deceptively repackaging

18

Onavo Protect as an application called “Facebook Research,” but Apple uncovered Facebook’s

19

deception in January 2019, and Facebook immediately removed Facebook Research from the App

20

Store. CC ¶ 212; AC ¶ 263. Similarly, when the “surveillance functionality” of Onavo Bolt was

21

discovered, Facebook immediately “shut that app down.” AC ¶ 266.

22

However, neither Consumers nor Advertisers have alleged any facts to establish that

23

Facebook used Onavo Protect to inflict “new and accumulating injury” after December 3, 2016.

24

Samsung Elecs., 747 F.3d at 1202. Specifically, Consumers and Advertisers have not alleged how

25

Facebook used data obtained through Onavo after December 3, 2016. Accordingly, they have

26

failed to show how Facebook’s use of Onavo was an overt act in furtherance of any theory of

27

antitrust liability. Indeed, neither Consumers nor Advertisers identify any companies which

28

90
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

Facebook tracked using Onavo after December 3, 2016. Nor do they specify whether any users of

2

Facebook’s social networking services also were users of Onavo Protect. To the contrary,

3

Consumers’ and Advertisers’ allegations indicate that the purpose of Onavo Protect was to track

4

users of other social media applications in order to determine which applications besides Facebook

5

were the most popular. CC ¶ 164; AC ¶ 166.

6

Given the absence of allegations regarding how Facebook used Onavo data after December

7

3, 2016, the Court cannot evaluate whether Facebook’s collection of data through Onavo inflicted

8

a “new and accumulating injury.” Samsung Elecs., 747 F.3d at 1202. Thus, the Court cannot

9

conclude that Facebook’s collection of data through Onavo after December 3, 2016 was an “overt

10

United States District Court
Northern District of California

11
12
13

act” that renders Consumers’ and Advertisers’ “Copy, Acquire, Kill” claims timely. Id.
Consumers Fail to Allege Any Facts Which Would Allow the Court to
Evaluate Whether Facebook’s Acquisitions of tbh and Giphy Inflicted
“New and Accumulating Injury”
Consumers allege that Facebook’s 2017 acquisition of tbh and 2020 acquisition of Giphy

14

were part of Facebook’s “Copy, Acquire, Kill” strategy. CC ¶¶ 206–07. Thus, Consumers

15

contend, Consumers’ “Copy, Acquire, Kill” claims are timely. Opp. at 8. Facebook argues that,

16

when Facebook filed the instant motion, Facebook’s acquisition of Giphy “ha[d] not yet been

17

completed” and thus could not have harmed Consumers. Mot. at 13 n.8. Additionally, Facebook

18

argues that Consumers have offered “no[] allegations to explain how the tbh acquisition lessened

19

competition.” Mot. at 13.

20

The Court agrees that Consumers fail to allege any facts which could allow the Court to

21

determine whether Facebook’s acquisitions of tbh and Giphy had anticompetitive effects.

22

Typically, plaintiffs establish that an acquisition is anticompetitive by offering evidence about the

23

“concentration of firms in [the relevant] market” before and after the acquisition. See FTC v. H.J.

24

Heinz Co., 246 F.3d 708, 715 (D.C. Cir. 2001). Alternatively, plaintiffs can offer direct evidence

25

about the acquired firm to show that removing that firm as a competitor would have a particular

26

anticompetitive effect. See, e.g., United States v. Aetna Inc., 240 F. Supp. 3d 1, 43 (D.D.C. 2017)

27

(explaining that an acquisition that “would result in the elimination of a particularly aggressive

28

91
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

competitor in a highly concentrated market” could be anticompetitive).

2

Here, Consumers have failed to offer either type of evidence. Consumers’ complaint

3

contains a single sentence about tbh: “Facebook acquired ‘tbh’—an anonymous social media

4

app—in October 2017.” CC ¶ 207. In turn, although Consumers at least describe the nature of

5

Giphy’s business, Consumers provide no information about Giphy which could allow the Court to

6

determine whether Facebook’s acquisition of Giphy had anticompetitive effects.5 See id. ¶ 206.

7

Consumers do not, for example, provide any details about Giphy’s revenue, user base, or plans for

8

growth.
Given the absence of allegations regarding tbh’s and Giphy’s businesses, the Court cannot

9

evaluate whether Facebook’s 2017 acquisition of tbh and 2020 acquisition of Giphy inflicted a

11

United States District Court
Northern District of California

10

“new and accumulating injury” on Consumers. Samsung Elecs., 747 F.3d at 1202. Accordingly,

12

the Court cannot conclude that those acquisitions were “overt acts” that render Consumers’ “Copy,

13

Acquire, Kill” claims timely. Id.
Advertisers Do Not Adequately Allege that Facebook’s Copying of
Snapchat Was an Overt Act That Occurred After December 3, 2016

14
15

Advertisers allege that “by late 2016,” Facebook copied Snapchat’s “stories” feature,

16
which allows a user to post a series of images and videos that remain online for twenty-four hours.
17
18

AC ¶ 298. Because this act was part of Facebook’s “Copy, Acquire, Kill” strategy, Advertisers
argue, Advertisers’ “Copy, Acquire, Kill” claims are timely. Opp. at 10.

19
The Court rejects this argument. Because Advertisers filed their complaint on December
20
21

3, 2016, Advertisers’ “Copy, Acquire, Kill” claims are timely only if Advertisers allege overt acts
that were part of the “Copy, Acquire, Kill” strategy and occurred after December 3, 2016. Klehr,

22
521 U.S. at 189.
23

Advertisers’ allegations, which state only that Facebook “had launched” a product that

24
25
26
27
28

5

As noted, Facebook asserts that, as of May 20, 2021, Facebook had not completed its acquisition
of Giphy. Mot. at 13 n.8. Moreover, in November 2021, the UK’s Competition and Markets
Authority blocked Facebook’s acquisition of Giphy. Morgan Meaker, Meta’s Failed Giphy Deal
Could End Big Tech’s Spending Spree, Wired (Dec. 3, 2021), https://bit.ly/3tkoRUx. Any
amended complaint should provide an update on the status of Facebook’s acquisition of Giphy.
92
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

copied Snapchat “by late 2016,” are insufficient to establish that Facebook copied Snapchat’s

2

stories feature after December 3, 2016 for two reasons. AC ¶ 298. First, Consumers’ allegation

3

that Facebook copied Snapchat’s stories feature “by late 2016” is too vague to support a

4

conclusion that Facebook copied Snapchat’s stories feature after December 3, 2016. Second,

5

Consumers’ use of the word “launched” does not provide sufficient details about the actions

6

Facebook took to copy Snapchat’s stories feature.

7

Without more specific details, the Court cannot conclude that Facebook’s copying of

8

Snapchat’s stories feature occurred after December 3, 2016. Thus, that act does not render

9

Advertisers’ “Copy, Acquire, Kill” claims timely. Klehr, 521 U.S. at 189.

10

United States District Court
Northern District of California

11
12
13
14
15

Advertisers Do Not Adequately Allege that Facebook Committed Overt
Acts Related to Facebook’s APIs After December 3, 2016
Advertisers allege that, in 2015, Facebook “cut off all public access to the Friends and
News Feed APIs” and discriminatorily allowed access only to developers who signed “Whitelist
and Datasharing Agreements.” AC ¶ 225. Accordingly, Advertisers contend, their “Copy,
Acquire, Kill” claims are timely because Facebook continued this discriminatory practice “at least
until April 4, 2018.” Opp. at 10. For the reason below, the Court rejects this contention.

16
As an initial matter, Advertisers do not allege that Facebook granted or refused to grant
17

any “Whitelist and Datasharing Agreements” after December 3, 2016. AC ¶ 241. Advertisers

18
allege only that:
19

•

In January 2015, Facebook provided Whitelist and Data Sharing Agreements to the

20
dating apps Tinder and Hinge.
21

•

In February 2015, Facebook refused to provide Airbiquity a Whitelist and Data Sharing

22
Agreement.
23

•

In February 2015, Facebook provided Whitelist and Datasharing Agreements to

24
Netflix, Nissan, and Lyft.
25

•

In April 2015, Facebook received requests for Whitelist and Data Sharing Agreements

26
with Microsoft, Hootsuite, and Walgreens.
27
28

93
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

Id. ¶¶ 221–25. Although Advertisers also allege that “dozens of app developers entered into such

2

Agreements with Facebook,” Advertisers concede that the “precise number and identity” of these

3

other developers “cannot be known.” Id. ¶ 226. Moreover, because Advertisers’ allegations are

4

“bereft of any dates or details with regards” to Facebook’s agreements with these other

5

developers, they do not support Advertisers’ argument that Facebook entered Whitelist and

6

Datasharing Agreements after December 3, 2016. Ryan v. Microsoft Corp., 147 F. Supp. 3d 868,

7

883 (N.D. Cal. 2015).
Because Advertisers have failed to allege that Facebook entered new Whitelist and Data

8

Sharing Agreements after December 3, 2016, Advertisers have failed to allege that Facebook

10

committed an “overt act” related to its discriminatory API restrictions during the limitations

11

United States District Court
Northern District of California

9

period. Id. at 885. Thus, those restrictions do not render Advertisers’ “Copy, Acquire, Kill”

12

claims timely. Klehr, 521 U.S. at 189.
Advertisers Do Not Adequately Allege that Facebook’s Integration of
Instagram and WhatsApp Inflicted a “New and Accumulating Injury”

13
14

Advertisers allege that, in 2019, Facebook officially announced that it planned to integrate
15
Facebook, Instagram, and WhatsApp. AC ¶ 524. In essence, integrating Facebook, Instagram,
16
and WhatsApp means that a user of one of those services will be able to communicate with users
17
of the other services without making accounts for those services. See Mike Isaac, Zuckerberg
18
Plans to Integrate WhatsApp, Instagram and Facebook Messenger, The New York Times (Jan.
19
20

25, 2019), https://nyti.ms/3zRQo0Q.6 Accordingly, Advertisers contend, that their “Copy,
Acquire, Kill” claims are timely because Facebook’s plan to integrate Facebook, Instagram, and

21
WhatsApp occurred after December 3, 2016 and is an overt act that is part of the Copy, Acquire,
22
Kill strategy. Opp. at 10. For the reasons below, the Court rejects this argument.
23

Although Advertisers’ argument is sparse, Advertisers appear to invoke the “new use”

24
exception to the general rule that, when one firm acquires another firm, subsequent actions by the
25
26
27
28

The Court takes judicial notice of the article in order to aid its understanding of Advertisers’
argument. See United States v. Black, 482 F.3d 1035, 1041 (9th Cir. 2007)
94
6

Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

“merged firm” do not inflict new and accumulating injuries that allow a plaintiff to challenge the

2

original acquisition. See Midwestern Machinery Co., Inc. v. Northwest Airlines, Inc., 392 F.3d

3

265, 271–74 (8th Cir. 2004) (“Even if the merger itself was unlawful, the continued existence of

4

the merged entity is not a continuing violation: It is simply the natural unabated inertial

5

consequence of the merger.”). “Under the ‘new use’ exception, ‘[i]f assets are used in a different

6

manner from the way that they were used when the initial acquisition occurred, and that new use

7

injures the plaintiff, he or she has four years from the time that the injury occurs to sue.’” See

8

Free FreeHand, 852 F. Supp. 2d at 1188 (quoting Midwestern Machinery, 392 F.3d at 273).

9

However, an important limitation is that the “new use” exception does not apply to an alleged
“anti-competitive threat” that “was clear at the time of the merger.” Midwestern Mach., 392 F.3d

11

United States District Court
Northern District of California

10

at 273. “Otherwise, every business decision could qualify as a continuing violation to restart the

12

statute of limitations as long as the firm continued to desire to be merged.” Id. at 271.

13

As an initial matter, Advertisers fail to explain why, when a firm acquires another firm, it

14

is not “clear at the time” of the acquisition that the two firms are going to integrate. Midwestern

15

Mach., 392 F.3d at 273. Indeed, a firm’s acquisition of another firm is potentially anticompetitive

16

precisely because the acquisition could result in a single firm that controls an undue share of the

17

market. See H.J. Heinz Co., 246 F.3d at 719. Without additional facts or law, the Court cannot

18

conclude that a firm’s decision to integrate a purchased asset into its existing business generally is

19

a “new use” that allows a plaintiff to challenge the original purchase.

20

Advertisers also fail to explain why, in this specific instance, Facebook’s 2019

21

announcement about WhatsApp and Instagram satisfies the “new use” doctrine. Specifically,

22

Advertisers highlight Facebook’s 2014 representation that it would not be able to “establish

23

reliable automatic matching between Facebook’s users’ accounts and WhatsApp users’ accounts”

24

and equates this representation to a statement that Facebook would not be able to integrate

25

WhatsApp. See AC ¶ 518. Accordingly, Advertisers allege, it was not clear at the time of the

26

acquisition that Facebook planned to integrate WhatsApp. Id. However, even if Advertisers are

27

correct, Facebook publicly announced in August 2016 that it would be able to “link[] WhatsApp

28

95
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

users’ phone numbers with Facebook users’ identities.” Id. ¶ 518. Thus, under Advertisers’ own

2

logic, Facebook’s plan to integrate WhatsApp was clear in August 2016.

3
4
5

In turn, Advertisers allege no facts which suggest that Facebook’s plan to integrate
Instagram was not clear at the time Facebook acquired Instagram.
Accordingly, the Court cannot conclude that Facebook’s 2019 announcement of its plan to

6

integrate Instagram and WhatsApp qualifies as “overt act” that inflicted “new and accumulating

7

injury” after December 3, 2016. Samsung Elecs., 747 F.3d at 1202.

8
9

Advertisers Do Not Adequately Allege Tolling Under the Fraudulent
Concealment Doctrine
Under the doctrine of fraudulent concealment, a “statute of limitations may be tolled if the

10
defendant fraudulently concealed the existence of a cause of action in such a way that the plaintiff,
11

United States District Court
Northern District of California

acting as a reasonable person, did not know of its existence.” Hexcel Corp. v. Ineos Polymers,
12
13
14
15

Inc., 681 F.3d 1055, 1060 (9th Cir. 2012). The purpose of the doctrine “is to prevent a defendant
from ‘concealing a fraud . . . until such time as the party committing the fraud could plead the
statute of limitations to protect it.’” In re Animation Workers Antitrust Litig., 123 F. Supp. 3d
1175, 1194 (N.D. Cal. 2015) (quoting Bailey v. Glover, 88 U.S. (21 Wall.) 342, 349 (1874)). “To

16
plead fraudulent concealment, the plaintiff must allege that: (1) the defendant took affirmative acts
17
18
19

to mislead the plaintiff; (2) the plaintiff did not have ‘actual or constructive knowledge of the facts
giving rise to its claim’; and (3) the plaintiff acted diligently in trying to uncover the facts giving
rise to its claim.” Id. (quoting Hexcel, 681 F.3d at 1060). Additionally, the plaintiff must plead

20
these elements with the particularity required by Rule 9(b). Id.
21
Advertisers argue that Facebook fraudulently concealed two sets of anticompetitive acts.
22

First, Advertisers argue that Facebook “made false and misleading statements (and omissions)

23
about the Instagram and WhatsApp mergers . . . to prevent discovery of the anticompetitive
24

purposes of those mergers.” Opp. at 14 (citing AC ¶¶ 517–26). Second, Advertisers argue that

25
Facebook fraudulently concealed the anticompetitive nature of its API restrictions by stating false
26
reasons for those restrictions. Opp. at 13–14 (citing AC ¶¶ 211–15, 491–92, 500–16). The Court
27
28

96
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

addresses these arguments in turn.

2

Advertisers Fail to Allege that Facebook’s 2014 Representation About
WhatsApp Tolls Advertisers’ “Copy, Acquire, Kill” Claims

3
Advertisers allege that Facebook fraudulently concealed the anticompetitive purpose of the
4
WhatsApp acquisition by representing to European regulators in 2014 that Facebook did not have
5

the means “to establish reliable automated matching between Facebook’s users’ accounts and

6
WhatsApp users’ accounts.” AC ¶ 518. Advertisers equate this statement to a representation that
7
Facebook would not be able to integrate WhatsApp. Id. Thus, Advertisers contend, Facebook’s
8
2014 representation tolled the statute of limitations until 2019, when Facebook officially
9
announced its plan to integrate WhatsApp. Opp. at 10.
10
However, as discussed, Facebook announced in August 2016 that it did have ability to
11

United States District Court
Northern District of California

match Facebook’s users’ accounts and WhatsApp users’ accounts. See pp. 95–96, supra. Thus,
12
13
14
15
16
17
18

even assuming that Facebook’s 2014 statement to European regulators misled Advertisers about
the anticompetitive purpose of the WhatsApp acquisition, Advertisers had “actual or constructive
notice” of this purpose prior to December 3, 2016. Hexcel, 681 F.3d at 1060.
Moreover, Advertisers have failed to allege that they “acted diligently in trying to uncover
the facts giving rise to [their] claim.” Id. “Where a plaintiff’s suspicions have been or should
have been excited, there can be no fraudulent concealment where he [or she] ‘could have then
confirmed his [or her] earlier suspicion by a diligent pursuit’ of further information.” Conmar

19
Corp. v. Mitsui & Co. (U.S.A.), Inc., 858 F.2d 499, 504 (9th Cir. 1988) (quoting Rutledge v.
20
Boston Woven Hose & Rubber Co., 576 F.2d 248, 250 (9th Cir. 1978)). Advertisers fail to explain
21

why Facebook’s August 2016 announcement would not have “excite[d] the inquiry of a

22
reasonable person.” Id. Thus, because Advertisers failed to allege any facts that suggest they
23

“acted diligently,” Advertisers may not invoke the fraudulent concealment doctrine. Hexcel, 681

24
F.3d at 1060.
25
26

Thus, Advertisers have not adequately alleged that Facebook’s 2014 statement to the
European Commission renders Advertisers’ “Copy, Acquire, Kill” claims timely under the

27
28

97
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

fraudulent concealment doctrine. Id.

2

Advertisers Allege No Facts That Show Diligence Regarding
Facebook’s 2015 Decision to Remove Public Access to APIs

3
Advertisers contend that Facebook made a series of misleading statements that concealed
4

the anticompetitive purpose of Facebook’s 2015 decision to remove public access to APIs. Opp.

5
at 13–14. According to Advertisers, these statements prevented Advertisers from timely bringing
6
7

the “causes of action asserted [in the instant case], which all require anticompetitive purpose
and/or effect as necessary elements.” Id. Facebook contends that Advertisers have failed to allege

8
that they were misled by these statements, that Advertisers had actual or constructive notice of the
9
conduct giving rise to their claims, and that Advertisers have failed to allege diligence. Mot. at 9–
10
10. For the reasons below, the Court agrees with Facebook that Advertisers have failed to allege
11

United States District Court
Northern District of California

diligence.
12
13
14
15
16
17
18
19
20

“Where a plaintiff’s suspicions have been or should have been excited, there can be no
fraudulent concealment where he [or she] ‘could have then confirmed his [or her] earlier suspicion
by a diligent pursuit’ of further information.” Conmar, 858 F.2d at 504 (quoting Rutledge, 576
F.2d at 250). “The requirement of diligence is only meaningful, however, when facts exist that
would excite the inquiry of a reasonable person.” Id.
As discussed, Advertisers’ allegations regarding APIs center on Facebook’s decision in
2015 to “cut off all public access to the Friends and News Feed APIs.” AC ¶ 225. According to
Advertisers, these APIs were so important to developers that Facebook’s 2015 decision to remove
public API access “simply br[oke] nearly all of the more than 40,000 third-party apps that relied

21
on the APIs.” Id. ¶ 214.
22
23
24
25
26

Advertisers do not allege that they were unaware of Facebook’s 2015 decision to remove
public access to APIs. Thus, even assuming that Facebook’s decision to remove public API
access did not itself give Advertisers “actual or constructive knowledge of the facts giving rise to
[their] claim,” Advertisers’ knowledge of Facebook’s decision obligated Advertisers to engage in
a “diligent pursuit” of “further information.” Hexcel, 681 F.3d at 1060. Indeed, Facebook’s

27
28

98
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

decision to remove public API access in 2015 is the core fact giving rise to Advertisers’ claims

2

and Advertisers fail to provide any explanation why this decision would not have “excite[d] the

3

inquiry of a reasonable person.” Conmar, 858 F.2d at 504.

4

Accordingly, because Advertisers have failed to allege any facts which suggest that they

5

“acted diligently in trying to uncover the facts” about Facebook’s decision to remove public access

6

to APIs in 2015, Advertisers may not invoke the fraudulent concealment doctrine. Hexcel, 681

7

F.3d at 1060. Advertisers’ complaint stands in stark contrast to the pleadings of plaintiffs who

8

have successfully invoked the fraudulent concealment doctrine. For example, in Animation

9

Workers, the plaintiffs brought a Sherman Act claim alleging that their former employers had
“engaged in a conspiracy to fix and suppress employee compensation.” 123 F. Supp. 3d at 1178–

11

United States District Court
Northern District of California

10

79. In support of their argument that their employers had fraudulently concealed the conspiracy,

12

the plaintiffs alleged that they had “repeatedly asked Defendants about how compensation was

13

determined and what steps Defendants were taking to retain and attract talented employees.”

14

Second Amended Complaint ¶ 183, In re Animation Workers Antitrust Litig., No. 14-CV-04062-

15

LHK (N.D. Cal. May 15, 2015), ECF No. 121. The Court found that this allegation, combined

16

with allegations describing specific instances in which the plaintiffs asked their employers about

17

compensation, were sufficient to show diligence. See Animation Workers, 123 F. Supp. 3d at

18

1204–05. Although Advertisers do not necessarily need to provide this level of detail, Advertisers

19

must allege some facts which suggest that they “acted diligently in trying to uncover the facts”

20

about Facebook’s removal of public access to its APIs. Hexcel, 681 F.3d at 1060.

21

Because Advertisers have failed to allege any facts about diligence, the Court need not

22

decide what level of allegations would suffice. However, Advertisers’ failure to allege diligence is

23

fatal to their fraudulent concealment argument. Id.

24

Thus, the Court GRANTS Facebook’s motion to dismiss Consumers’ and Advertisers’

25

“Copy, Acquire, Kill” claims. However, because the Court finds that additional allegations may

26

cure the deficiencies outlined above, amendment would not be futile. Leadsinger, 512 F.3d at

27

532. Additionally, neither Consumers nor Advertisers have acted in bad faith, and allowing leave

28

99
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

to amend would not unduly prejudice Facebook or cause undue delay. Id. Accordingly, the Court

2

grants Consumers and Advertisers leave to amend their “Copy, Acquire, Kill” claims.

3

D. The Court Denies Facebook’s Motion to Dismiss Advertisers’ GNBA Claims

4

Facebook concedes that Advertisers’ GNBA claims are timely. Additionally, Facebook

5

does not dispute Advertisers’ allegations that the GNBA was an unreasonable restraint of trade

6

under Section 1 of the Sherman Act and exclusionary conduct under Section 2 of the Sherman

7

Act. Instead, Facebook argues that Advertisers have not adequately alleged that the GNBA caused

8

them a cognizable antitrust injury. See Mot. at 31–34. For the reasons below, the Court rejects

9

this argument.

10

United States District Court
Northern District of California

11
12

The Court begins by discussing Advertisers’ allegations regarding the GNBA. The Court
then explains why Advertisers have adequately alleged causal antitrust injury.
The Google Network and Bidding Agreement

13

In addition to alleging that Facebook maintained monopoly power by strategically

14

eliminating potential competitors, Advertisers allege that Facebook preserved its monopoly power

15

by agreeing with Google not to enter each other’s market. This agreement, which Facebook and

16

Google signed in 2018, was called the Google Network Bidding Agreement (“GNBA”).

17

Prior to signing the GNBA, Facebook had taken steps to improve its general advertising

18

capabilities. In 2013, Facebook acquired an advertising program called Atlas. AC ¶ 347. In

19

addition to serving advertisements, Atlas tracked whether a user who clicked on an advertisement

20

eventually purchased a product as a result. Id. ¶ 339. Facebook used Atlas to create a robust

21

tracking system which allowed Facebook to determine whether a Facebook user purchased a

22

product at an offline retail store after seeing an advertisement for that product on Facebook. Id.

23

¶¶ 341, 352. After purchasing and implementing Atlas, Facebook acquired a real-time bidding

24

service called LiveRail. Id. ¶ 355.

25

In 2014, Facebook announced a new advertising system called the Facebook Audience

26

Network (“FAN”). Id. ¶ 326. FAN allowed advertisers to use Facebook’s granular targeting

27

strategies to advertise not just on Facebook but also in mobile applications. Id. ¶ 327. Initially,

28

100
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

FAN only showed advertisements to users who logged into mobile applications with their

2

Facebook accounts. Id. ¶ 334. However, in May 2016, Facebook extended FAN to track

3

Facebook users even when they did not login with Facebook accounts and to show targeted

4

advertisements to those users. Id. ¶ 335. Indeed, on April 16, 2018, Facebook revealed in

5

documents provided to Congress that Facebook tracks its users even when they are not logged into

6

Facebook based on: (1) social plugins, such as the Like and Share buttons; (2) Facebook Login,

7

which allows users to use their Facebook account to log into another website or app; (3) Facebook

8

Analytics; and (4) Facebook ads and measurement tools. Id. ¶ 366.

9

These steps positioned Facebook to compete with Google, which provides advertising
services on websites across the internet. Id. ¶ 369. Google’s Ad Manager (“GAM”) service helps

11

United States District Court
Northern District of California

10

websites who wish to sell advertising space to find companies who want to place advertisements

12

on those websites. Id. ¶ 371. Google uses GAM both to find direct purchasers for advertising

13

space and to sell advertising space to advertising exchanges, “where marketers bid for the

14

[advertising space] in real time.” Id. Google also operates its own advertising exchange, which is

15

called Google Ad Exchange. Id. ¶ 375. Google charges companies fees when they purchase

16

advertising space through Google Ad Exchange. Id.

17

However, while Facebook was getting better at providing advertisements outside its core

18

applications, Google was getting better at determining the identities of people who visited the

19

websites with which Google worked to sell advertising. From the beginning, Google could obtain

20

basic information about people who visited websites that contracted with GAM or Google Ad

21

Exchange. Id. ¶ 378. For example, Google could obtain such a person’s “IP address, device

22

identification information, or browser information.” Id. ¶ 378. Then, in January 2014, Google

23

acquired an artificial intelligence company called DeepMind. Id. ¶ 384. This technology allowed

24

Google to “make granular identity determinations” from the basic data that Google had always

25

collected. Id. ¶ 378. For example, Google used this information to “study a [person’s] decisions

26

and preferences” and make recommendations on its Google Play Store. Id. ¶¶ 385–86.

27
28

In 2017, Facebook made an overt threat to Google’s business. By that time, websites that
101
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

wanted to sell advertising space had “beg[u]n to adopt a practice called ‘header bidding,’” which

2

allowed websites to send a “standardized” advertising space “to several exchanges” at once. Id.

3

¶¶ 391–92. This practice “threatened to cut Google out of the picture” because it obviated the

4

need to go through Google to place advertising space on advertising exchanges. Id. ¶ 393. “In

5

March 2017, Facebook publicly announced it would support header bidding.” Id. ¶ 395.

6

Following Facebook’s announcement, Google reached out to Facebook to broker a deal.

7

Id. ¶ 398. Then, in September 2018, Facebook and Google reached the GNBA. Id. ¶ 400. Under

8

this agreement, Facebook agreed to drop its support for header bidding. Id. ¶ 401. In return,

9

Google agreed that it would: (1) provide Facebook with powerful tools to identify, target, and
monetize Facebook’s own users on the web and across third party mobile applications and (2) give

11

United States District Court
Northern District of California

10

Facebook the right to show ads to 90% of Facebook’s users and twice the amount of time to bid on

12

advertising to Facebook’s users. Id. ¶¶ 18, 405. Because this arrangement would enable

13

Facebook to identify Facebook users across the internet, Facebook would have more data on those

14

users and could better target advertisements to them. Id. ¶ 406.

15

“Put simply,” Advertisers allege, “Google and Facebook agreed to divide and segment

16

markets, allowing Facebook to continue charging a significant price premium for its targeted

17

advertising sold in the Social Advertising Market.” Id. ¶ 411.

18
19

Advertisers Adequately Allege that the GNBA Caused Them Injury
To state a claim under either Section 1 or Section 2 of the Sherman Act, a plaintiff must

20

allege causal antitrust injury. See Brantley v. NBC Universal, Inc., 675 F.3d 1192, 1197 (9th Cir.

21

2012) (“In order to state a Section 1 claim . . . plaintiffs must plead . . . . that they were harmed by

22

the defendant’s anti-competitive contract, combination, or conspiracy, and that this harm flowed

23

from an ‘anti-competitive aspect of the practice under scrutiny.’”) (internal citation omitted);

24

SmileCare Dental, 88 F.3d at 783 (“In order to state a claim for monopolization under Section 2 of

25

the Sherman Act, a plaintiff must prove . . . causal antitrust injury.”) (internal quotation omitted).

26

This requirement applies even if the plaintiff has alleged that the defendant’s conduct is per se

27

illegal. Atl. Richfield Co. v. USA Petroleum Co., 495 U.S. 328, 344 (1990) (“[P]roof of a per se

28

102
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

violation and of antitrust injury are distinct matters that must be shown independently.”).

2

To plead causal antitrust injury, a plaintiff must allege that the defendant’s unlawful

3

conduct caused an injury “that flows” from the unlawful conduct and that is “the type the antitrust

4

laws were intended to prevent.” Somers, 729 F.3d at 963. Additionally, “the party alleging the

5

injury must be either a consumer of the alleged violator’s goods or services or a competitor of the

6

alleged violator in the restrained market.” Id. (quoting Glen Holly, 343 F.3d at 1008).

7

Because Advertisers allege that the GNBA allowed Facebook to maintain

8

supracompetitive prices, Advertisers have adequately alleged a cognizable antitrust injury.

9

Specifically, Advertisers allege that the GNBA prevented Google from “leverag[ing] its ability to
identify and target Facebook users.” AC ¶ 485. Thus, the GNBA “bolstered and reinforced

11

United States District Court
Northern District of California

10

Facebook’s dominant position in the Social Advertising Market” and allowed Facebook “to

12

maintain and raise prices with little or no competitive check.” Id.

13

Indeed, Advertisers cite specific sources which show that Facebook’s advertising prices

14

“increased 122 percent” the year the GNBA was signed and “grew 90 percent” the year after the

15

GNBA was signed. Id. ¶ 443. As this Court has explained, “charg[ing] Plaintiffs

16

supracompetitive prices” is a “type of injur[y] that commonly satisf[ies] the antitrust standing

17

requirement.” See Free FreeHand, 852 F. Supp. 2d at 1185; see also Glen Holly Ent., Inc. v.

18

Tektronix, Inc., 352 F.3d 367, 374 (9th Cir. 2003); Pool Water Prods. v. Olin Corp., 258 F.3d

19

1024, 1034 (9th Cir. 2001) (“[T]he antitrust laws are only concerned with acts that harm

20

‘allocative efficiency and raise[ ] the price of goods above their competitive level or diminish[ ]

21

their quality.’” (quoting Rebel Oil Co. v. ARCO, 51 F.3d 1421, 1433 (9th Cir. 1995))); Abbyy USA

22

Software House, Inc. v. Nuance Communications Inc., No. 08-CV-01035, 2008 WL 4830740, at

23

*4 (N.D.Cal. Nov. 6, 2008) (noting that direct purchasers of software would have antitrust

24

standing stemming from injury of paying supracompetitive prices).

25

Additionally, Advertisers have adequately explained how Advertisers’ injury “flow” from

26

the GNBA. Somers, 729 F.3d at 963. Advertisers allege that, by 2018, Google had developed a

27

way to determine the true identities of people who visited Google’s partners’ websites. See AC

28

103
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

¶¶ 377–86. Thus, Google was close to offering “a new, highly targeted form of advertising” that

2

could compete with Facebook’s social advertising business by providing companies with the

3

opportunity to target people or groups with specific traits and interests. Id. ¶¶ 379–80. However,

4

under the GNBA, Google agreed that, on Google’s partners’ websites, “Facebook would receive

5

the right to show ads to 90% of the users it recognized as its own” and that, for the remaining

6

10%, Facebook would receive extra time to place bids. Id. ¶ 405. According to Advertisers, the

7

primary effect of this agreement was to make it nearly impossible for companies to show online

8

advertisements to Facebook users without negotiating with Facebook, even when those people are

9

not using Facebook or Facebook-affiliated applications. Id. ¶ 406. Thus, “Facebook’s users
remained uniquely Facebook’s to advertise to” and, if Advertisers wanted to reach those users,

11

United States District Court
Northern District of California

10

Advertisers “had to pay Facebook (at a premium).” Id. ¶ 409. By providing this description,

12

Advertisers have adequately “sketch[ed] the outline of [the injury to competition] with allegations

13

of supporting factual detail.’” Brantley, 675 F.3d at 1198 (quoting Les Shockley Racing, Inc. v.

14

Nat’l Hot Rod Ass’n, 884 F.2d 504, 508 (9th Cir. 1989)) (alteration in original).

15

Facebook’s arguments to the contrary are unconvincing. Although Facebook cites Intel

16

Corporation v. Fortress Investment Group LLC, 511 F. Supp. 3d 1006 (N.D. Cal. 2021), to argue

17

that Advertisers are required to “identify the competitive price,” Mot. at 32, Intel imposes no such

18

requirement. In Intel, the plaintiff brought a claim alleging that “companies who licensed [certain]

19

patents from [the defendant] paid supracompetitive prices.” 511 F. Supp. 3d at 1027. However,

20

because the terms of the relevant patent licenses were confidential, the plaintiff “provided no

21

information” about the prices paid under the license agreements. Id. Accordingly, the court held

22

that the plaintiff’s allegations were too speculative to support an antitrust claim. Id.

23

In contrast to the Intel plaintiff, Advertisers have cited specific sources showing that

24

Facebook’s advertising prices “increased 122 percent” the year the GNBA was signed and “grew

25

90 percent” the year after the GNBA was signed. AC ¶ 433. Numerous courts have found similar

26

allegations sufficient. For example, in Free FreeHand, the plaintiff supported its claim of

27

supracompetitive prices by alleging that the defendant had raised its prices by 25% the year after

28

104
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

committing the alleged antitrust violation. See First Amended Complaint ¶ 68, Free FreeHand

2

Corp. v. Adobe Systems Inc., No. 11-CV-02174-LHK (N.D. Cal. Jul. 20, 2011), ECF No. 19. This

3

Court found that allegation sufficient to establish supracompetitive prices. See Free FreeHand,

4

852 F. Supp. 2d at 1185. Similarly, in In re Magnesium Oxide Antitrust Litigation, No. 10-CV-

5

5943-DRD, 2011 WL 5008090, at *6 (D.N.J. Oct. 20, 2011), the court found sufficient the

6

plaintiffs’ allegation “that they were ‘injured by having paid more for [the product] than they

7

otherwise would have paid.’” Thus, not only is there no authority for Facebook’s argument that

8

Advertisers must allege the “competitive price,” there is ample authority showing that Advertisers’

9

allegations regarding prices are sufficient.

10

Facebook also relies on Associated General Contractors of California, Inc. v. California

United States District Court
Northern District of California

11

State Council of Carpenters, 459 U.S. 519 (1983), but that case is inapposite. In Associated

12

General Contractors, a construction workers union brought antitrust claims on behalf of

13

contractors and subcontractors who were members of the union. Id. at 527–28. The United States

14

Supreme Court held that the union did not have standing to bring claims on behalf of its members.

15

Id. at 541. Although the union alleged that it “suffered unspecified injuries in its ‘business

16

activities,’” the Court explained that it was “obvious that any such injuries were only an indirect

17

result of whatever harm may have been suffered by ‘certain’ construction contractors and

18

subcontractors.” Id. at 541–42. Thus, Associated General Contractors merely provides that an

19

entity that has no direct connection with an antitrust violator may not bring claims on behalf of

20

entities who have suffered direct harm. By contrast, in the instant case, Advertisers allege that

21

they are “consumer[s] of [Facebook’s] goods or services” and have suffered direct harm because

22

of the Facebook’s anticompetitive acts. See Somers, 729 F.3d at 963. Facebook has not identified

23

a case in which similarly situated plaintiffs have not been allowed to bring antitrust claims.

24

The remainder of Facebook’s arguments involve complicated factual disputes. For

25

example, Facebook argues that Advertisers’ theory is implausible because there is no evidence that

26

Google had the capability to peel customers away from Facebook. See Mot. at 33. However, this

27

kind of factual dispute cannot be resolved at the pleading stage, and because Advertisers have

28

105
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

“‘raise[d] a reasonable expectation that discovery will reveal evidence of’ an injury to

2

competition,” Advertisers’ allegations regarding injury from the GNBA are sufficient. Brantley,

3

675 F.3d at 1198 (citing Twombly, 550 U.S. at 556).
Accordingly, because Advertisers have adequately alleged causal antitrust injury, the Court

4
5
6

DENIES Facebook’s motion to dismiss Advertisers’ GNBA claims.
E. The Court Grants Facebook’s Motion to Dismiss Consumers’ Unjust Enrichment
Claim With Leave to Amend

7
In addition to alleging claims under the Sherman Act, Consumers allege a claim against
8
Facebook for common law unjust enrichment. CC ¶¶ 308–17. Specifically, Consumers allege that
9

Facebook obtained user data through “misrepresentations and deception” and that Facebook has

10
made billions of dollars from selling this data. Id. ¶¶ 313–15.

United States District Court
Northern District of California

11

However, as “this Court has repeatedly held, ‘California does not recognize a separate

12
cause of action for unjust enrichment.’” Abuelhawa v. Santa Clara Univ., No. 20-CV-0404513
LHK, 2021 WL 1176689, at *9 (N.D. Cal. Mar. 29, 2021) (quoting Brodsky v. Apple Inc., 445 F.
14
Supp. 3d 110, 132–33 (N.D. Cal. 2020) (collecting California and federal cases)). California law
15

is clear: “Unjust enrichment is not a cause of action.” De Havilland v. FX Networks, LLC, 21 Cal.

16
App. 5th 845, 870, 230 Cal. Rptr. 3d 625 (Cal. Ct. App. 2018) (quoting Hill v. Roll Int’l Corp.,
17
18

195 Cal. App. 4th 1295, 1307, 128 Cal. Rptr. 3d 109 (Cal. Ct. App. 2011)). Thus, “courts have
consistently dismissed stand-alone claims for unjust enrichment.” Brodsky, 445 F. Supp. 3d at

19
132.
20

Accordingly, the Court GRANTS Facebook’s motion to dismiss Consumers’ Unjust

21
Enrichment claim. Although the Court questions whether Consumers can cure the above
22
deficiencies, the Court cannot conclude that amendment would be futile. Leadsinger, 512 F.3d at
23
532. Additionally, the Court cannot conclude that Consumers have acted in bad faith, or that
24
allowing leave to amend would unduly prejudice Facebook or cause undue delay. Id. Therefore,
25
the Court grants leave to amend.
26
27
28

106
Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND

1

IV. CONCLUSION
For the foregoing reasons, the Court GRANTS IN PART and DENIES IN PART

2
3

Facebook’s motion to dismiss.
Specifically, the Court GRANTS with leave to amend Facebook’s motion to dismiss the

4
5

following claims:
•

8
9

•

United States District Court
Northern District of California

16
17
18
19

Advertisers’ GNBA claims.

Plaintiffs shall file any amended consolidated complaints within 45 days of this Order.

12

15

Consumers’ data privacy claims; and

•

11

14

Consumers’ Unjust Enrichment claim.

The Court DENIES Facebook’s motion to dismiss the following claims:

10

13

Advertisers’ “Copy, Acquire, Kill” claims; and

•

7

Consumers’ “Copy, Acquire, Kill” claims;

•

6

Failure to do so, or failure to cure deficiencies identified herein or identified in the instant motion
to dismiss, will result in dismissal of the deficient claims with prejudice. Plaintiffs may not add
new causes of action or add new parties without stipulation or leave of the Court. Plaintiffs are
directed to file a redlined complaint comparing the complaint to any amended complaint as an
attachment to Plaintiffs’ amended complaint.
IT IS SO ORDERED.
Dated: January 14, 2022
______________________________________
LUCY H. KOH
United States Circuit Judge*

20
21
22
23
24
25
26
27
28

*

Sitting by designation on the United States District Court for the Northern District of California.
107

Case No. 20-CV-08570-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION TO DISMISS WITH LEAVE TO AMEND