Grace v. Apple, Inc.

Filing 269

ORDER by Judge Lucy H. Koh granting in part and denying in part #174 Motion to Certify Class; denying #233 Motion to Strike. (lhklc3S, COURT STAFF) (Filed on 9/18/2018)

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

CHRISTINA GRACE, et al.,
Plaintiffs,

13

ORDER GRANTING IN PART AND
DENYING IN PART MOTION FOR
CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

v.

14
15

Case No. 17-CV-00551-LHK

APPLE, INC.,

Re: Dkt. Nos. 174, 233

Defendant.

16
17

Plaintiffs Christina Grace (“Grace”) and Ken Potter (“Potter”) (collectively, “Plaintiffs”)

18
19

bring this putative class action against Defendant Apple, Inc. (“Apple”) for trespass to chattels and

20

violation of California’s Unfair Competition Law (“UCL”), Cal. Bus. & Prof. Code § 17200.

21

Before the Court is Plaintiffs’ motion for class certification and Plaintiffs’ motion to strike the

22

report of Apple’s technical expert. Having considered the parties’ submissions, the relevant law,

23

and the record in this case, the Court GRANTS in part and DENIES in part Plaintiffs’ motion for

24

class certification, and DENIES Plaintiffs’ motion to strike.

25

I.

26
27
28

BACKGROUND
A. Factual Background
1

Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

1. Apple’s iPhone 4 and iPhone 4S and Apple’s FaceTime

2

Apple is the designer, seller, and manufacturer of the iPhone smartphone. ECF No. 36

3

(First Amended Complaint, or “FAC”) ¶ 29. Apple released the iPhone 4 device on June 7, 2010.

4

Id. ¶ 31. Apple released the iPhone 4S device on October 14, 2011. Id. ¶ 32.

5

All iPhones, including the iPhone 4 and iPhone 4S, operate through a mobile operating

6

system known as iOS. Id. ¶ 34. iOS runs the features and applications of the iPhone. Id. Prior to

7

September 13, 2013—the day that Apple released the “iOS 7” operating system, which is

8

discussed further below— iPhones operated on the “iOS 6” operating system or an earlier iOS

9

operating system. See, e.g., ¶ 9–10.
The iPhone 4 was the first iPhone device to offer a feature known as “FaceTime.” Id. ¶ 37.

11
United States District Court
Northern District of California

10

FaceTime is a video conferencing feature that allows iPhone users to communicate via video calls,

12

as opposed to traditional audio calls. Id. Apple’s press releases about the iPhone 4 described the

13

iPhone 4 as “the new iPhone 4 featuring FaceTime.” Id. In addition, FaceTime was featured

14

prominently in Apple’s advertising campaign for the iPhone 4. Id. ¶ 38. “[S]everal of Apple’s

15

television advertisements for the iPhone 4 focused exclusively on FaceTime” and FaceTime’s

16

ability to “bridge the gap between friends and loved ones no matter the geographic distance

17

between them.” Id. For example, an iPhone 4 advertisement depicted a deployed solider using

18

FaceTime to view his pregnant wife’s sonogram appointment, and a hearing-impaired couple

19

communicating over FaceTime in sign language. Id. ¶¶ 39–40.

20

Prior to November 2012, iPhone 4 and iPhone 4S devices used two methods to connect a

21

FaceTime call: (1) the peer-to-peer method, and (2) the relay method. Id. ¶ 52. “[T]he two

22

connection methods would occur in parallel, and the call would be connected through whichever

23

method achieved a connection first.” Id. ¶ 55. The call would connect through the peer-to-peer

24

method approximately 90 to 95% of the time. Id. Through the peer-to-peer method, the caller and

25

callee exchanged data directly between each other over the internet. Id. ¶ 53.

26
27
28

“[S]ometimes, however, it was not possible to connect a FaceTime call through the peer2
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

to-peer method,” and thus the FaceTime call would instead connect through the relay method. Id.

2

¶ 54. Through the relay method, the caller and the callee connected to a relay server that relayed

3

the data on behalf of the devices. Id. ¶ 5. Apple used servers from a company called Akamai

4

Technologies, Inc. (“Akamai”), to relay FaceTime calls through the relay method. Id. Akamai

5

charged Apple fees based on the amount of FaceTime calls that connected through Akamai’s

6

servers. Id. ¶¶ 5, 59. Accordingly, if a low number of FaceTime calls connected through the relay

7

method, Apple paid Akamai a lower fee than if a higher number of FaceTime calls connected

8

through the relay method. Id.

9
10

a. Patent Infringement Trial and Apple’s Transition to Only the Relay Method
On November 7, 2012, a jury found that Apple’s use of the peer-to-peer method to connect

United States District Court
Northern District of California

11

FaceTime calls infringed patents owned by VirnetX, Inc. (“VirnetX”). Id. ¶ 61. Indeed, due to the

12

willful nature of Apple’s infringement, the jury ordered Apple to pay an ongoing royalty to

13

VirnetX that was higher than the effective royalty for use of VirnetX’s patents. Id.

14

In order to avoid paying royalty fees to VirnetX for use of VirnetX’s patents, Apple

15

eliminated the peer-to-peer method for FaceTime calls. Id. ¶ 61. Thus, all FaceTime calls on

16

devices using iOS 6 or earlier operating systems connected via the relay method 100% of the time.

17

Id. However, Apple’s shift to the relay only method meant that Apple’s fees to use Akamai’s relay

18

servers increased substantially. Id. ¶ 62. According to the FAC, internal Apple emails show that

19

Apple’s switch to the relay only method for FaceTime calls caused Apple’s fees to Akamai to

20

increase substantially. Id.

21

b. Introduction of iOS 7

22

As a result of Apple’s mounting relay fees, Apple began to explore options for decreasing

23

relay usage for FaceTime calls. Id. ¶ 65. On September 13, 2013, Apple released a new operating

24

system, iOS 7. Id. ¶ 67. Unlike iOS 6 and earlier operating systems, iOS 7 allowed users to

25

connect a FaceTime call using a peer-to-peer method that had not yet been found to infringe

26

VirnetX’s patents. Id. ¶ 68. Thus, the introduction of iOS 7 provided Apple with a way to avoid

27
28

3
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1
2

the millions of dollars in increased relay fees that Apple was paying to Akamai. Id.
Problematically, however, the introduction of iOS 7 only reduced relay usage to the extent

3

that iPhone users who were operating iOS 6 or earlier operating systems voluntarily transitioned to

4

iOS 7. Id. ¶ 68. The upgrade to iOS 7 was free of cost. Nonetheless, iPhone 4 and iPhone 4S users

5

were reluctant to switch to iOS 7. Specifically, “iOS7 was a more robust and powerful operating

6

system that acted as a significant drain on the processing capability of any device on which it ran.”

7

Id. ¶ 71. Although new iPhone devices, such as the iPhone 5 and iPhone 5C, “were designed to

8

include a more powerful processor in order to function properly with iOS7,” the iPhone 4 and

9

iPhone 4S devices were not designed with processors that could handle the upgrade to iOS 7. Id.

10

United States District Court
Northern District of California

11

¶¶ 71–72.
According to the FAC, “transitioning to iOS7 on an iPhone 4 or iPhone 4S significantly

12

impaired” the device’s functionality. Id. ¶ 72. For example, iPhone 4 and iPhone 4S devices that

13

transitioned to iOS 7 experienced problems such as “non-responsiveness, keyboard sluggishness,

14

extremely slow app[lication] launching and device crashes.” Id. ¶ 72. Plaintiffs also allege that

15

“[f]or a sizeable portion of the iPhone 4 and 4S population, a defect in iOS7 meant that

16

transitioning to iOS7 would prevent [iPhone 4 and iPhone 4S users] from accessing Wi-Fi and/or

17

Bluetooth,” which are two critical features of the iPhone. Id. ¶¶ 80–86. Moreover, “Apple made it

18

impossible for users who had transitioned to iOS7 to revert back to an earlier version of iOS,”

19

even though Apple previously allowed users to install older operating systems, Id. ¶ 80. Thus, if

20

an iPhone 4 or iPhone 4S user transitioned to iOS 7 and experienced lost functionality, the user

21

could not transition back to iOS 6 or an earlier operating system. Id.

22

c. Apple’s Disabling of FaceTime for iOS 6 and Earlier Operating Systems

23

Because of the functionality problems discussed above, “a sizeable percentage of Apple’s

24

user base” continued to use iOS 6 or an earlier operating system after the introduction of iOS 7. Id.

25

¶ 88. Plaintiffs allege that this was problematic for Apple because, as set forth above, iPhone 4 and

26

iPhone 4S devices that used iOS 6 or earlier operating systems connected on FaceTime through

27
28

4
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

the relay method, rather than the new peer-to-peer method used on iOS 7. Id. ¶ 89. Thus, Apple

2

continued to pay Akamai substantial fees. Id.

3

According to the FAC, Apple decided to “break” FaceTime for iOS 6 and earlier operating
systems. Id. ¶ 90. Specifically, Apple “cause[d] a digital certificate necessary to the operation of

5

FaceTime on iOS 6 and earlier operating systems to prematurely expire” on April 16, 2014. Id. ¶

6

91. As a result, FaceTime stopped working on iPhone 4 and iPhone 4S devices that used iOS 6 or

7

earlier operating systems. Id. The Court will hereafter refer to this event as the “FaceTime Break.”

8

In order to regain usage of FaceTime, iPhone 4 and iPhone 4S users were required to

9

upgrade to iOS 7, which utilized the new peer-to-peer method to connect FaceTime calls. Id.

10

However, as set forth above, transitioning to iOS 7 on an iPhone 4 or an iPhone 4S caused the

11
United States District Court
Northern District of California

4

device to “buckle under the weight of the new software.” Id. ¶ 12.

12

Thus, Plaintiffs allege that after Apple disabled FaceTime for iOS 6 and earlier operating

13

systems on April 16, 2014, Apple in effect gave iPhone 4 and iPhone 4S users three options. The

14

iPhone 4 or 4S user could either: (1) continue to use iOS 6 or an earlier operating system, but

15

never regain the ability to use FaceTime; (2) upgrade to iOS 7, but suffer the lost functionality that

16

iOS 7 caused iPhone 4 and iPhone 4S devices; (3) purchase a new iPhone device, such as an

17

iPhone 5 or 5C, that had the processing capability to successfully operate using iOS 7, and thus

18

regain the ability to use FaceTime on iOS 7. See id.

19

The FAC alleges that, at the time that Apple decided to disable FaceTime for users of iOS

20

6 and earlier operating systems, Apple knew that “[a]ll users with [iOS] 6.0 and older c[ould not]

21

make FaceTime [calls] any longer” as a result of Apple’s conduct. Id. ¶ 98. Further, Apple “fully

22

recognized that FaceTime was a very important tool” for iPhone 4 and 4S owners “that allowed

23

loved ones separated by geographic distance to remain connected in a meaningful way.” Id. ¶ 95.

24

Nonetheless, Apple broke FaceTime in order to gain a financial advantage and reduce relay fees.

25

Id. Indeed, some Apple employees seem to have found the FaceTime Break quite funny—one

26

employee wrote “hilarity ensues” to describe the aftermath. Mot. at 9. At least some senior Apple

27
28

5
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

executives were aware of what was occurring and issued a directive to falsely blame the FaceTime

2

Break on a “bug”. Mot at 11; see FAC ¶ 108.

3

d. Named Plaintiffs’ Experiences

4

Plaintiff Christina Grace (“Grace”) is a California resident who purchased an iPhone 4 in

5

August 2010. ECF No. 199-2 at 83:15–84:3; FAC ¶ 23. Grace testified that she used FaceTime.

6

ECF No. 225-11 (excerpts of Grace’s deposition testimony). A review of Grace’s iPhone 4

7

revealed at least one FaceTime use and a number of entries that may or may not have been

8

FaceTime uses. ECF No. 223-4 at 1.

9

Plaintiff Ken Potter (“Potter”) is also a California resident. Id. ¶ 25. Potter purchased a 32
gigabyte (“GB”) iPhone 4 in September 2010, and a 16 GB iPhone 4 in March 2011. ECF No.

11
United States District Court
Northern District of California

10

199-3 at 50:14–51:23. In early 2011, Potter “jailbroke” his 32 GB iPhone 4. Id. at 222:5–22. Potter

12

did not jailbreak his 16 GB iPhone 4. Id. at 223:24–224:5. Potter upgraded his non-jailbroken

13

phone to iOS 7.1.1, on June 25, 2014, and regained the ability to use FaceTime on the device at

14

that time. Id. at 152:17–22.

15

B. Procedural History

16

On February 2, 2017, Grace filed a putative class action complaint in this Court against Apple.

17

ECF No. 1. Plaintiff sought to represent a class consisting of “All owners of Apple iPhone 4 or

18

Apple iPhone 4S devices in the United States who on April 16, 2014, had the iOS6 or earlier

19

operating system on their iPhone 4 or iPhone 4S.” Id. ¶ 107. The complaint alleged causes of

20

action against Apple for (1) trespass to chattels, and (2) violations of the UCL, each based on

21

Apple’s intentional disabling of FaceTime on iOS 6 and earlier operating systems. See id. ¶ 1.

22

On March 22, 2017, Apple moved to dismiss the complaint. ECF No. 33. According to

23

Apple, Grace lacked standing under Article III and the UCL because Grace had no right to

24

uninterrupted FaceTime service under the terms of Apple’s Software License Agreement. Id. at

25

13–17. Further, Apple argued that Grace lacked standing to bring claims based on the use of

26

iPhone 4S, which Grace did not own, and that Grace lacked standing to raise issues relating to the

27
28

6
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

upgrade to iOS 7, which Grace never alleged that she downloaded. Id. at 18. In addition, Apple

2

argued that Grace failed to state a claim for trespass to chattels or violations of the UCL. See, e.g.,

3

id.
Rather than oppose Apple’s motion to dismiss, Grace filed the FAC on April 5, 2017. ECF

4
5

No. 36. The FAC added Potter as a named Plaintiff. Potter alleged that he upgraded to iOS 7 on

6

one of his iPhone 4 devices, and that Potter experienced lost functionality as a result of his

7

upgrade. Id.
Grace’s amendment of the complaint on April 5, 2017 was an amendment as of right under

9

Federal Rule of Civil Procedure 15(a)(1)(B) because Grace amended the complaint within 21 days

10

of Apple’s motion to dismiss the original complaint. See Fed. R. Civ. P. 15(a)(1)(B). Accordingly,

11
United States District Court
Northern District of California

8

on April 6, 2017, the Court denied as moot Apple’s motion to dismiss the original complaint. See

12

ECF No. 37.

13

On April 19, 2017, Apple filed a motion to dismiss the FAC, which largely raised the same

14

arguments that Apple raised in its motion to dismiss the original complaint. ECF No. 38.1 On May

15

3, 2017, Plaintiffs filed an opposition to Apple’s motion to dismiss. ECF No. 47. On May 10,

16

2017, Apple filed a reply. ECF No. 50.
On July 28, 2017, the Court denied Apple’s motion to dismiss. Grace v. Apple Inc., 2017

17
18

WL 3232464 (N.D. Cal. July 28, 2017). The Court found that Plaintiffs possessed Article III

19

standing, that Plaintiffs had stated claims for violation of the UCL’s “unfair” prong and trespass to

20

chattels, and that Plaintiffs had adequately alleged entitlement to restitution and injunctive relief.

21

On May 4, 2018, Plaintiffs filed a motion for class certification. ECF No. 174 (“Mot.”).

22

Plaintiffs seek to certify the following nationwide class under Rule 23(b)(2) and Rule 23(b)(3).

23

This class is defined as the following:

24

All owners of Apple iPhone 4 or Apple iPhone 4S devices in the United States who

25
26
27
28

1

Apple also requested that the Court take judicial notice of Apple’s iOS 6 and iOS 7 License
Agreements. ECF No. 39. The Court granted this request. Grace v. Apple Inc., 2017 WL 3232464,
at *4 n.1 (N.D. Cal. July 28, 2017)
7
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

on April 16, 2014, had iOS 6 or earlier operating systems on their iPhone 4 or
iPhone 4S devices.

1
2

Mot. at vii. In the alternative, Plaintiffs seek to certify a California class under Rule 23(b)(2) and

3

Rule 23(b)(3). This class is defined as the following:

4

All owners of Apple iPhone 4 or Apple iPhone 4S devices in California who on
April 16, 2014, had iOS 6 or earlier operating systems on their iPhone 4 or iPhone
4S devices.

5
6
7

Id. On June 15, 2018, Apple filed its opposition. ECF No. 201 (“Opp.”). On July 13, 2018,
Plaintiffs filed their reply. ECF No. 225 (“Reply”).

8
9
10

On July 19, 2018, Plaintiffs filed a motion to strike the Declaration of Dr. Avi Rubin,
Apple’s technical expert. ECF No. 233. On August 2, 2018, Apple filed its opposition. ECF No.
242. On August 9, 2018, Plaintiffs filed their reply. ECF No. 246.

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

On July 20, 2018, Apple filed objections to evidence submitted in connection with
Plaintiffs’ reply. ECF No. 235.
II.

LEGAL STANDARD
Class actions are governed by Rule 23 of the Federal Rules of Civil Procedure. Rule 23

does not set forth a mere pleading standard. To obtain class certification, plaintiffs bear the burden
of showing that they have met each of the four requirements of Rule 23(a) and at least one
subsection of Rule 23(b). Zinser v. Accufix Research Inst., Inc., 253 F.3d 1180, 1186, amended by
273 F.3d 1266 (9th Cir. 2001). “A party seeking class certification must affirmatively demonstrate
… compliance with the Rule[.]” Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 350 (2011).
Rule 23(a) provides that a district court may certify a class only if: “(1) the class is so
numerous that joinder of all members is impracticable; (2) there are questions of law or fact
common to the class; (3) the claims or defenses of the representative parties are typical of the
claims or defenses of the class; and (4) the representative parties will fairly and adequately protect
the interests of the class.” Fed. R. Civ. P. 23(a). That is, the class must satisfy the requirements of
numerosity, commonality, typicality, and adequacy of representation to maintain a class action.
Mazza v. Am. Honda Motor Co., Inc., 666 F.3d 581, 588 (9th Cir. 2012).
8
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

If all four prerequisites of Rule 23(a) are satisfied, the Court must also find that the

2

plaintiff “satisf[ies] through evidentiary proof” at least one of the three subsections of Rule 23(b).

3

Comcast Corp. v. Behrend, 569 U.S. 27, 33 (2013). The Court can certify a Rule 23(b)(1) class

4

when plaintiffs make a showing that there would be a risk of substantial prejudice or inconsistent

5

adjudications if there were separate adjudications. Fed. R. Civ. P. 23(b)(1). The Court can certify a

6

Rule 23(b)(2) class if “the party opposing the class has acted or refused to act on grounds that

7

apply generally to the class, so that final injunctive relief or corresponding declaratory relief is

8

appropriate respecting the class as a whole.” Fed. R. Civ. P. 23(b)(2). Finally, the Court can certify

9

a Rule 23(b)(3) class if the Court finds that “questions of law or fact common to class members
predominate over any questions affecting only individual members, and that a class action is

11
United States District Court
Northern District of California

10

superior to other available methods for fairly and efficiently adjudicating the controversy.” Fed. R.

12

Civ. P. 23(b)(3).

13

“[A] court’s class-certification analysis must be ‘rigorous’ and may ‘entail some overlap

14

with the merits of the plaintiff’s underlying claim[.]’” Amgen Inc. v. Connecticut Ret. Plans & Tr.

15

Funds, 568 U.S. 455, 465–66 (2013) (quoting Dukes, 564 U.S. at 351); see also Mazza, 666 F.3d

16

at 588 (“‘Before certifying a class, the trial court must conduct a ‘rigorous analysis’ to determine

17

whether the party seeking certification has met the prerequisites of Rule 23.’” (quoting Zinser, 253

18

F.3d at 1186)). This “rigorous” analysis applies to both Rule 23(a) and Rule 23(b). Comcast, 569

19

U.S. at 34 (stating that Congress included “addition[al] ... procedural safeguards for (b)(3) class

20

members beyond those provided for (b)(1) or (b)(2) class members (e.g., an opportunity to opt

21

out)” and that a court has a “duty to take a ‘close look’ at whether common questions predominate

22

over individual ones”).

23

Nevertheless, “Rule 23 grants courts no license to engage in free-ranging merits inquiries

24

at the certification stage.” Amgen, 568 U.S. at 466. “Merits questions may be considered to the

25

extent—but only to the extent—that they are relevant to determining whether the Rule 23

26

prerequisites for class certification are satisfied.” Id. If a court concludes that the moving party has

27
28

9
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

met its burden of proof, then the court has broad discretion to certify the class. Zinser, 253 F.3d at

2

1186.

3

III.

4

DISCUSSION
Apple opposes Plaintiffs’ motion for class certification on four main grounds. First, Apple

5

argues that Plaintiffs cannot demonstrate Rule 23(a)(3) typicality because the proposed class

6

representatives’ iPhone usage is not typical of the class. Second, Apple advances a series of

7

challenges to Rule 23(b)(3) predominance. Specifically, Apple argues that Plaintiffs’ trespass to

8

chattels claim and Plaintiffs’ UCL claims require individualized inquiries, that Plaintiffs’ damages

9

theory does not satisfy Comcast, and that Plaintiffs’ proposed nationwide class is barred under
California’s choice of law test. Third, Apple briefly argues that Plaintiffs cannot demonstrate Rule

11
United States District Court
Northern District of California

10

23(b)(3) superiority. Fourth, Apple argues that no Rule 23(b)(2) injunctive class can be certified.

12

For their part, Plaintiffs contest Apple’s challenges to class certification arguments and have filed

13

a separate motion to strike the report of Dr. Avi Rubin, Apple’s technical expert. Apple has also

14

filed an objection to evidence Plaintiffs have submitted in connection with their reply.

15

The Court rejects Apple’s Rule 23(a)(3) typicality argument. The Court also rejects most

16

of Apple’s Rule 23(b)(3) arguments. In particular, the Court finds that individual inquiries are not

17

necessary for Plaintiffs’ trespass to chattels claim and Plaintiffs’ UCL claim, and the Court rejects

18

Apple’s Comcast challenge to Plaintiffs’ damages model. However, the Court agrees with Apple

19

that California’s choice of law test precludes certification of a nationwide class, and agrees further

20

that no Rule 23(b)(2) class is warranted. Finally, the Court closes by denying Plaintiffs’ motion to

21

strike the report of Apple’s expert and rejecting Apple’s evidentiary objections.

22
23

A. Rule 23(a) Analysis
“Parties seeking class certification must satisfy each of the four requirements of Rule

24

23(a)—numerosity, commonality, typicality, and adequacy—and at least one of the requirements

25

of Rule 23(b).” Briseno v. ConAgra Foods, Inc., 844 F.3d 1121, 1124 (9th Cir. 2017), cert. denied

26

sub nom. ConAgra Brands, Inc. v. Briseno, 138 S. Ct. 313 (Oct. 10, 2017). The Court takes each

27
28

10
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

of Rule 23(a)’s requirements in turn, and finds that Plaintiffs have satisfied all four.

2

1. Numerosity

3

Under Rule 23(a)(1), Plaintiffs must show that “the class is so numerous that joinder of all

4

members is impracticable.” Fed. R. Civ. P. 23(a)(1). “In determining whether numerosity is

5

satisfied, the Court may consider reasonable inferences drawn from the facts before it.” In re

6

Yahoo Mail Litig., 308 F.R.D. 577, 589–90 (N.D. Cal. 2015).

7

Apple concedes that there are at least 16 million class members spread across the country,

8

and does not dispute that numerosity would be satisfied if the Court certified a California-only

9

class. ECF No. 174-75 at 5:18–24. Consequently, because the proposed class “numbers well over
forty and joinder would be impracticable, the Court finds that the numerosity requirement is

11
United States District Court
Northern District of California

10

satisfied.” In re Yahoo Mail, 308 F.R.D. at 590; see, e.g., Rannis v. Recchia, 380 F. App’x 646,

12

651 (9th Cir. 2010) (“In general, courts find the numerosity requirement satisfied when a class

13

includes at least 40 members.”); Huynh v. Harasz, 2015 WL 7015567, at *5 (N.D. Cal. Nov. 12,

14

2015) (“As other district courts have noted, ‘the numerosity requirement is usually satisfied where

15

the class comprises 40 or more members.’” (quoting Twegbe v. Pharma Integrative Pharmacy,

16

Inc., 2013 WL 3802807, at *3 (N.D. Cal. July 17, 2013)).

17

2. Commonality

18

Next, Rule 23(a)(2) states that “[o]ne or more members of a class may sue or be sued as

19

representative parties on behalf of all members only if … there are questions of law or fact

20

common to the class.” Fed. R. Civ. P. 23(a)(2). To satisfy the commonality requirement, Plaintiffs

21

must show that the class members have suffered “the same injury,” meaning that class members’

22

claims must “depend upon a common contention” of such a nature that “determination of its truth

23

or falsity will resolve an issue that is central to the validity of each [claim] in one stroke.” Dukes,

24

564 U.S. at 350 (quotation marks and citation omitted). Plaintiffs must demonstrate not merely the

25

existence of a common question, but rather “the capacity of a classwide proceeding to generate

26

common answers apt to drive the resolution of the litigation.” Id. (quotation marks omitted)

27
28

11
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

(emphasis in original). Nevertheless, the “common contention need not be one that will be

2

answered, on the merits, in favor of the class.” Alcantar v. Hobart Serv., 800 F.3d 1047, 1053 (9th

3

Cir. 2015) (internal quotation marks omitted). Additionally, “for purposes of Rule 23(a)(2), even a

4

single common question will do.” Dukes, 564 U.S. at 359 (alteration and quotation marks

5

omitted); see also Mazza, 666 F.3d at 589 (“[C]ommonality only requires a single significant

6

question of law or fact.”).

7

Plaintiffs identify a number of issues common to the class: (1) whether Apple deliberately

8

caused the FaceTime Break; (2) whether Apple intentionally interfered with Plaintiffs’ devices; (3)

9

whether Apple concealed its conduct; and (4) whether iOS 7 impaired the functionality of iPhone
4 and iPhone 4S devices. Apple does not dispute that these are common questions capable of

11
United States District Court
Northern District of California

10

satisfying Rule 23(a)(2). In fact, Apple does not dispute Rule 23(a)(2) commonality at all. The

12

Court therefore finds that Plaintiffs have satisfied Rule 23(a)(2)’s commonality requirement.

13

3. Typicality

14

Rule 23(a)(3) requires plaintiffs to show that “the claims or defenses of the representative

15

parties are typical of the claims or defenses of the class.” Fed. R. Civ. P. 23(a)(3). Typicality is

16

satisfied “when each class member’s claim arises from the same course of events, and each class

17

member makes similar legal arguments to prove the defendants’ liability.” Rodriguez v. Hayes,

18

591 F.3d 1105, 1124 (9th Cir. 2010) (quoting Armstrong v. Davis, 275 F.3d 849, 868 (9th Cir.

19

2001)). This requirement is “permissive and requires only that the representative’s claims are

20

reasonably co-extensive with those of the absent class members; they need not be substantially

21

identical.” Hanlon v. Chrysler Corp., 150 F.3d 1011, 1020 (9th Cir. 1998). Claims that are

22

reasonably co-extensive with the claims of absent class members will satisfy the typicality

23

requirement, but the class must be limited to “those fairly encompassed by the named plaintiff’s

24

claims.” Dukes, 564 U.S. at 349. “[C]lass certification is inappropriate where a putative class

25

representative is subject to unique defenses which threaten to become the focus of the litigation.”

26

Hanon v. Dataproducts Corp., 976 F.2d 497, 508 (9th Cir. 1992) (citations omitted). “The purpose

27
28

12
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

of the typicality requirement is to assure that the interest of the named representative aligns with

2

the interests of the class.” Id.

3

Apple presents three arguments as to why Plaintiffs’ two proposed class representatives

4

cannot satisfy Rule 23(a)(3) typicality. First, Apple argues that Grace is not typical because she

5

never used FaceTime or did not use FaceTime enough. Second, Apple argues that Potter is not

6

typical because any interruption in FaceTime access was too brief to have injured him. Third,

7

Apple argues that Potter is not typical because he chose to “jailbreak” his iPhone. The Court

8

disagrees with Apple’s first two arguments and agrees partially with Apple’s third argument.

9

First, Apple’s opposition, which was based on the neutral evaluator’s initial findings,
argues that Grace is not typical because “[a] review of her phone demonstrates the absence of any

11
United States District Court
Northern District of California

10

FaceTime calls made on her iPhone 4.” Opp. at 7. However, Apple has since conceded that this

12

argument was ultimately determined to be empirically incorrect. ECF No. 223-4 at 1 (Apple

13

acknowledging this). Consequently, Apple’s new argument is that Grace is not typical because she

14

did not use FaceTime enough. The Court disagrees. Plaintiffs’ theory of the case is that Apple is

15

liable because the FaceTime Break reduced the monetary value of class members’ iPhones.

16

Determining precisely how often a class member like Grace used FaceTime is irrelevant to that

17

analysis. The Court therefore finds that Grace’s claims are typical of the class she seeks to

18

represent.

19

Second, Apple argues that Potter is not typical because he was only fleetingly deprived of

20

FaceTime access. That matters, Apple says, because a California trespass to chattels claim requires

21

that a plaintiff be deprived of personal property for a substantial time, a “mere momentary or

22

theoretical deprivation of use is not sufficient unless there is a dispossession.” Intel Corp. v.

23

Hamidi, 30 Cal. 4th 1342, 1357 (2003). This is unpersuasive. The FaceTime Break occurred on

24

April 16, 2014, and Apple’s expert states that Potter upgraded his 16 GB iPhone 4 to iOS 7 on

25

June 25, 2014. ECF No. 223-6, Second Declaration of Avi Rubin (“Second Rubin Decl.”) at 24.

26

Apple is therefore arguing that 71 days without FaceTime is “a mere momentary or theoretical

27
28

13
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

deprivation.” It is not, a point underscored by Apple’s failure to point to any case adopting its

2

view.

3

Apple’s argument is also irrelevant. California recognizes three ways to establish an injury

4

in the context of a trespass to a computer system or similar device. Specifically, injury exists

5

where “the purported trespass: (1) caused physical damage to the personal property, (2) impaired

6

the condition, quality, or value of the personal property, or (3) deprived plaintiff of the use of

7

personal property for a substantial time.” Fields v. Wise Media, LLC, 2013 WL 5340490, at *4

8

(N.D. Cal. Sept. 24, 2013). Apple’s argument about deprivation length is only relevant if Plaintiffs

9

base their injury on a deprivation theory of harm. However, Plaintiffs are instead proceeding under
an impaired value theory, which means that Apple’s argument is beside the point. The Court

11
United States District Court
Northern District of California

10

explores this issue further in the subsequent Rule 23(b)(3) predominance analysis.

12

Third, Apple argues that Potter is not typical because he chose to “jailbreak” his 32 GB

13

iPhone 4. Broadly speaking, to “jailbreak” an iPhone is to remove programmed limitations in

14

order to enlarge the iPhone’s functionality, run unauthorized applications, or otherwise change the

15

iPhone’s performance. United States v. Flores-Lopez, 670 F.3d 803, 808 (7th Cir. 2012)

16

(discussing this); United States v. Lee, 2012 WL 2031220, at *2 (D. Haw. June 6, 2012), aff’d, 585

17

F. App’x 632 (9th Cir. 2014) (same). Apple argues that Potter’s jailbreaking of his 32 GB iPhone

18

4 means he is subject to unique defenses (e.g. violation of the iOS Software License) and may not

19

even have suffered a cognizable injury.

20

The Court agrees that Potter’s claims based on his jailbroken 32 GB iPhone 4 are not

21

typical of the class. Only a small fraction of iPhone owners jailbreak their devices and one of the

22

main reasons to jailbreak an iPhone is to alter its performance and functionality relative to a

23

normal iPhone, i.e. to make it significantly different. Opp. at 8 (describing jailbreaking and its

24

effect on iPhones); Reply at 3 (stating less than 5 percent of iPhones are jailbroken). Accordingly,

25

claims based on Potter’s jailbroken 32 GB iPhone 4 would be subject to unique defenses, such as

26

that Potter violated the terms of Apple’s iOS Software License and that any performance issues

27
28

14
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

with Potter’s 32 GB iPhone 4 are attributable to the device being jailbroken rather than the

2

FaceTime Break. Opp. at 9.2
As a result, the Court will exercise its “broad discretion to modify class definitions” and

3
4

amend Plaintiffs’ proposed nationwide and California class definitions to exclude jailbroken

5

iPhones. Williams v. City of Antioch, 2010 WL 3632199, at *3 (N.D. Cal. Sept. 2, 2010) (quoting

6

Powers v. Hamilton County Public Defender Com'n, 501 F.3d 592, 619 (6th Cir. 2007));

7

Charlebois v. Angels Baseball, LP, 2011 WL 2610122, at *3 (C.D. Cal. June 30, 2011) (“[T]he

8

Court is free to limit the class definition or give Plaintiffs leave to do so.”); 7A Charles Alan

9

Wright & Arthur R. Miller, Federal Practice and Procedure § 1759 (3d ed.) (“[T]he court may
construe the complaint or redefine the class to bring it within the scope of Rule 23 or it may allow

11
United States District Court
Northern District of California

10

plaintiff to amend in order to limit the class.”).
Nevertheless, Potter still satisfies typicality. Potter owned a non-jailbroken 16 GB iPhone

12
13

4 in addition to his jailbroken 32 GB iPhone 4. ECF No. 199-3 at 50:14–51:23. Potter’s claims

14

based on the 16 GB iPhone 4 are not subject to any jailbreaking-related differences or defenses

15

because the 16 GB iPhone 4 was not jailbroken. Thus, even though Potter’s claims based on his 32

16

GB iPhone 4 do not satisfy Rule 23(a)(3) typicality, Potter’s 16 GB iPhone 4 does. Reply at 3

17

(noting Apple’s jailbreaking typicality challenges apply only to Potter’s 32 GB iPhone 4).
In sum, the Court finds that Grace and Potter, as to his 16 GB iPhone 4 only, satisfy Rule

18
19

23(a)(3) typicality. The Court also amends Plaintiffs’ proposed nationwide and California class

20

definitions to exclude jailbroken iPhones.

21

4. Adequacy

22

Rule 23(a)(4) requires “the representative parties [to] fairly and adequately protect the

23
24
25
26
27
28

2

The existence of jailbroken iPhones does not, however, bar certification for ascertainability
reasons because the Ninth Circuit has clarified that there is no freestanding ascertainability
requirement under Rule 23(a). Briseno, 844 F.3d at 1133 (“We therefore join the Sixth, Seventh,
and Eighth Circuits in declining to adopt an administrative feasibility requirement.”); see Bruton v.
Gerber Prod. Co., 703 F. App'x 468, 470 (9th Cir. 2017) (noting “ascertainability” is synonymous
with “administrative feasibility”).
15
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

interests of the class.” Fed. R. Civ. P. 23(a)(4). In the Ninth Circuit, to test the adequacy of a class

2

representative, a court must answer two questions: “(1) do the named plaintiffs and their counsel

3

have any conflicts of interest with other class members; and (2) will the named plaintiffs and their

4

counsel prosecute the action vigorously on behalf of the class?” Staton v. Boeing Co., 327 F.3d

5

938, 957 (9th Cir. 2003) (citing Hanlon, 150 F.3d at 1020).

6

Apple does not dispute the adequacy of Plaintiffs’ counsel or the prospective lead

7

plaintiffs. The Court finds that Plaintiffs’ counsel satisfies Rule 23(a)(4) adequacy. Specifically,

8

the Court finds that Plaintiffs’ counsel has experience in class action suits, has no conflicts of

9

interest in this case, has vigorously litigated this case, and will continue to do so. See ECF No.
174-77 (listing attorneys’ background and experience). The Court also finds that prospective lead

11
United States District Court
Northern District of California

10

plaintiffs Grace and Potter satisfy Rule 23(a)(4) adequacy. Specifically, the Court finds that Grace

12

and Potter do not have conflicts of interest with other class members, that they are willing and able

13

to serve as class representatives, and that they will vigorously pursue this suit. See ECF No. 174-

14

80, Declaration of Christina Grace; ECF No. 174-81, Declaration of Ken Potter.

15

In sum, the Court finds that Plaintiffs have satisfied the Rule 23(a) requirements of

16

numerosity, commonality, typicality, and adequacy. The Court now turns to whether Plaintiffs

17

have satisfied Rule 23(b)(3)’s predominance requirement.

18
19

B. Rule 23(b)(3) Predominance
Under Rule 23(b)(3), plaintiffs must show “that the questions of law or fact common to

20

class members predominate over any questions affecting only individual members.” Fed. R. Civ.

21

P. 23(b)(3). The Rule 23(b)(3) predominance requirement is “even more demanding” than Rule

22

23(a)’s commonality counterpart. Comcast, 569 U.S. at 34. Predominance “tests whether proposed

23

classes are sufficiently cohesive to warrant adjudication by representation.” Amchem, 521 U.S. at

24

623 (citation omitted). The Ninth Circuit has held that “there is clear justification for handling the

25

dispute on a representative rather than an individual basis” if “common questions present a

26

significant aspect of the case and they can be resolved for all members of the class in a single

27
28

16
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

adjudication.” Hanlon, 150 F.3d at 1022. As the U.S. Supreme Court recently explained, “[t]he

2

predominance inquiry asks whether the common, aggregation-enabling, issues in the case are more

3

prevalent or important than the non-common, aggregation-defeating, individual issues.” Tyson

4

Foods, Inc. v. Bouaphakeo, 136 S. Ct. 1036, 1045 (2016) (quoting 2 William Rubenstein,

5

Newberg on Class Actions, § 4:49 at 196–97 (5th ed. 2012)). “When ‘one or more of the central

6

issues in the action are common to the class and can be said to predominate, the action may be

7

considered proper under Rule 23(b)(3) even though other important matters will have to be tried

8

separately, such as damages or some affirmative defenses peculiar to some individual class

9

members.’” Id. (quoting 7AAC Charles Wright, Arthur Miller & Mary Kay Kane, Federal Practice

10

United States District Court
Northern District of California

11

and Procedure, § 1778 at 123–24 (3d ed. 2005)).
Plaintiffs argue that common questions predominate because this case stems from Apple’s

12

uniform and simultaneous creation of the FaceTime Break, which affected all class members at the

13

same time and in the same way. Newton v. Am. Debt Servs., Inc., 2015 WL 3614197, at *10 (N.D.

14

Cal. June 9, 2015) (“[W]here a plaintiff’s claim under the ‘unfair’ prong ‘hinges on the existence

15

of a uniform business practice or series of practices amenable to some degree of precise

16

definition,’ certification of a Rule 23(b)(3) class may be warranted.” (quoting Herskowitz v. Apple,

17

Inc., 301 F.R.D. 460, 476 (N.D. Cal. 2014)). Plaintiffs note further that common questions are

18

especially likely to predominate because this case centers almost exclusively on Apple’s conduct,

19

namely causing the FaceTime Break, as opposed to class members’ individual experiences. See In

20

re: Lenovo Adware Litig., 2016 WL 6277245, at *17–18 (N.D. Cal. Oct. 27, 2016) (finding

21

Lenovo’s decision to install VirtualDiscovery on class members’ laptops was subject to common

22

proof). In line with this view, Plaintiffs’ damages model bases class members’ damages on the

23

diminution in the resale market value of affected iPhone 4 and iPhone 4S devices, which

24

eliminates the need to base damages on class members’ personal experiences with the FaceTime

25

Break (e.g. length of time without FaceTime access).

26
27
28

Apple has four responses. First, Apple argues that Plaintiffs’ trespass to chattels claim
17
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

requires individual inquiries into whether class members have suffered legally cognizable harm.

2

Second, Apple argues that the same is true for Plaintiffs’ UCL claim. Third, Apple argues that

3

Plaintiffs’ classwide damages theory does not align with Plaintiffs’ theory of liability. Fourth,

4

Apple argues that California’s choice of law analysis bars applying California law to a nationwide

5

class, and that applying the laws of each individual state would defeat predominance. The Court

6

rejects Apple’s first three arguments, but agrees with Apple that California’s choice of law test

7

precludes certification of a nationwide class.

8

1. Trespass to Chattels

9

The California Supreme Court has explained that “while a harmless use or touching of
personal property may be a technical trespass (see Rest. 2d of Torts, § 217), an interference (not

11
United States District Court
Northern District of California

10

amounting to dispossession) is not actionable under modern California and broader American law

12

without a showing of harm.” Hamidi, 30 Cal. 4th at 1350–51. Specifically, in the context of

13

trespass to a computer system or other similar devices, injury is adequately alleged where the

14

plaintiff pleads “that the purported trespass: (1) caused physical damage to the personal property,

15

(2) impaired the condition, quality, or value of the personal property, or (3) deprived plaintiff of

16

the use of personal property for a substantial time.” Fields, 2013 WL 5340490, at *4.

17

Apple argues that individual inquiries will be necessary in order to determine whether the

18

FaceTime Break deprived class members of the use of FaceTime for a “substantial time.” Apple

19

says this is so because an actionable deprivation “must be for a time so substantial that it is

20

possible to estimate the loss caused thereby. A mere momentary or theoretical deprivation of use is

21

not sufficient unless there is a dispossession.” Hamidi, 30 Cal. 4th at 1357 (quoting Restatement

22

(Second) of Torts § 217 cmt. i). In other words, Apple argues that class members pursuing a

23

trespass to chattels claim based on a deprivation theory can only recover if class members were

24

deprived of FaceTime use for a substantial time as opposed to a momentary or theoretical

25

deprivation.

26
27
28

Apple’s argument misconstrues Plaintiffs’ theory of liability. Plaintiffs are not arguing that
18
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

Apple’s trespass deprived them “of the use of personal property for a substantial time” but rather

2

proceed under the alternative theory that the FaceTime Break “impaired the condition, quality, or

3

value” of their iPhones. Indeed, Plaintiffs’ briefing makes this point clear. See, e.g., Mot. at 1

4

(“The inability to use the FaceTime feature impacted the function and value of all Class members’

5

devices.”); id. at 22 (“Dr. Hastings has constructed an econometric damages model which shows

6

that the FaceTime Break caused all Class members’ iPhone 4 and 4S devices to lose value.”);

7

Reply at 4 (“Again, Plaintiffs’ trespass to chattels claim is premised on the impairment to the

8

value of all iPhone 4 and 4S devices affected by the FaceTime Break.”). So does the report of

9

Plaintiffs’ damages expert, Professor Justine Hastings. ECF No. 173-6, Expert Report of Justine S.
Hastings, Ph.D. (“Hastings Report”) at 21 (“Here, the harmful event was the [FaceTime] Break.

11
United States District Court
Northern District of California

10

The harm is the diminished value of the phone as measured by resale market prices.”). Apple’s

12

argument that individual inquiries will be necessary to determine whether Plaintiffs were deprived

13

of their property for a “substantial time” is therefore irrelevant—answering that question is

14

unnecessary in this case.

15

2. UCL

16

California’s Unfair Competition Law (“UCL”) prohibits unfair competition, including

17

“any unlawful, unfair or fraudulent business act or practice and unfair, deceptive, untrue or

18

misleading advertising.” Cal. Bus. & Prof. Code § 17200. The UCL creates a cause of action for

19

business practices that are (1) unlawful, (2) unfair, or (3) fraudulent. Each “prong” of the UCL

20

provides a separate and distinct theory of liability. Lozano v. AT & T Wireless Servs., Inc., 504

21

F.3d 718, 731 (9th Cir. 2007). Plaintiffs proceed under the unfair prong of the UCL. Compl. ¶¶

22

130–35. Although there are a number of ways to satisfy the unfairness prong, Plaintiffs here rely

23

on the balancing test. Grace, 2017 WL 3232464, at *14 (discussing different tests used to satisfy

24

unfairness prong and finding Plaintiffs proceeded under the balancing test). The “balancing test”

25

“asks whether the alleged business practice is ‘immoral, unethical, oppressive, unscrupulous or

26

substantially injurious to consumers and requires the court to weigh the utility of the defendant’s

27
28

19
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

conduct against the gravity of the harm to the alleged victim.’ ” In re Carrier IQ, Inc., 78 F. Supp.

2

3d 1051, 1115 (N.D. Cal. 2015) (quoting Drum v. San Fernando Valley Bar Ass’n, 182 Cal. App.

3

4th 247, 257 (2010)).

4

Apple argues that the balancing test precludes class certification because a class member’s

5

harm will depend on a number of individual inquiries such as when the class member upgraded to

6

iOS 7, and whether the class members suffered any performance-related issues as a result. Apple

7

therefore analogizes this case to the denial of class certification in Herskowitz, 301 F.R.D. 460.

8

The Herskowitz plaintiffs sought to certify a class of consumers who had been charged multiple

9

times for the same product on one of Apple’s online marketplaces. Id. at 467.3 As here, the
Herskowitz plaintiffs pursued a claim under the UCL’s unfair prong. Id. at 476. This Court denied

11
United States District Court
Northern District of California

10

certification in part because of a concern that different class members had experienced different

12

harms depending on, inter alia, whether or not they had requested a refund. Id. The Court

13

reasoned that while it might “be possible to determine harm uniformly for each of these (and

14

other) possible scenarios, determining which scenario applies to any given class member will

15

require individualized inquiries.” Id.
Apple’s Herskowitz argument again misapprehends Plaintiffs’ theory of liability. Apple

16
17

assumes that Plaintiffs seek to measure class members’ harm by the length of their time without

18

FaceTime or the degree of their iPhone performance problems after upgrading to iOS 7. If that

19

were the case, Herskowitz’s concern that individual inquiries would be necessary to determine the

20

type of harm that might apply. However, Plaintiffs have repeatedly stated that they are not

21

attempting to measure harm in this manner.4 Plaintiffs’ theory of the case is that the FaceTime

22
23
24
25
26
27
28

3

For example, one Herskowitz plaintiff bought a song on the iTunes Store, but the song did not
appear in her music library. Consequently, the plaintiff bought the song again, at which point the
song appeared in her music library. The plaintiff was charged for both purchases. Herskowitz, 301
F.R.D. at 466.
4
Apple also overlooks another distinction between Herskowitz and the instant case. In Herskowitz,
the Court found that “Plaintiffs have failed to provide any evidence to support the existence of a
uniform business practice or series of practices respecting refunds in cases of double-billing.”
Herskowitz, 301 F.R.D. at 476. By contrast, the evidence here suggests the FaceTime Break was
the result of a single action undertaken by Apple on a specific date which affected every class
20
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

Break made class members’ iPhones less valuable by forcing them to choose between (1) keeping

2

iOS 6 and losing FaceTime or (2) upgrading to iOS 7 and keeping FaceTime, but experiencing

3

substantial performance problems.5 Plaintiffs seek to measure how much this problem diminished

4

the monetary value of class members’ iPhones’ by comparing Best Buy’s resale prices for the

5

iPhone 4 and iPhone 4S before and after the FaceTime Break. This means Plaintiffs can measure

6

class members’ harm through the devices’ decreased market value instead of a case-by-case

7

determination of when class members upgraded to iOS 7 and what performance issues they

8

experienced thereafter. Best Buy’s data is especially useful in this regard because Best Buy

9

upgrades iPhones to the latest iOS prior to resale, meaning there is no concern that the amount
paid for resold iPhones could vary depending on whether a given iPhone was or was not upgraded

11
United States District Court
Northern District of California

10

to iOS 7 when it was resold. Opp. at 15 (noting Best Buy upgrades iPhones’ iOS prior to resale).

12

In sum, Apple’s Rule 23(b)(3) arguments based on Plaintiffs’ trespass to chattels and UCL

13

claims fail because they rest on a misapprehension of Plaintiffs’ theory of harm. The Court now

14

turns to Apple’s challenge to Plaintiffs’ damages model.

15

3. Damages Model

16

Although individual damages calculations alone do not make class certification

17

inappropriate under Rule 23(b)(3), see Leyva v. Medline Indus., Inc., 716 F.3d 510, 514 (9th Cir.

18

2013) (“[T]he amount of damages is invariably an individual question and does not defeat class

19

action treatment.”), the U.S. Supreme Court has held that the plaintiff bears the burden of

20

providing a damages model showing that “damages are susceptible of measurement across the

21

entire class for purposes of Rule 23(b)(3).” Comcast, 569 U.S. at 35. The damages model “must

22
23
24
25
26
27
28

member in the same way.
5
Both scenarios are grounded in fact. It is undisputed that FaceTime stopped working on iOS 6 on
April 14, 2014, and Plaintiffs point to credible evidence that iPhone 4 and iPhone 4S devices
performed poorly when running on iOS 7 because it overtaxes the devices’ older hardware.
Compl. ¶¶ 75–76 (quoting ARS Technica article that examined iPhone 4 and found that
“[e]verything is slower in iOS 7 and that delays due to the new operating system “can become a
significant percentage of the time you spend” ); Id. ¶ 72 (noting iPhone 4 and iPhone 4S have
approximately half the random access memory of the iPhone 5 and iPhone 5S).
21
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

measure only those damages attributable to” the plaintiff’s theory of liability. Id. If the plaintiff

2

does not offer a plausible damages model that matches her theory of liability, “the problem is not

3

just that the Court will have to look into individual situations to determine the appropriate measure

4

of damages; it is that Plaintiffs have not even told the Court what data it should look for.” In re

5

MyFord Touch Consumer Litig., 2016 WL 7734558, at *15 (N.D. Cal. Sept. 14, 2016).

6

Plaintiffs’ damages theory is presented by their expert, Professor Justine Hastings (“Dr.

7

Hastings”). Dr. Hastings states that “[b]ecause the Break stopped FaceTime from working on the

8

iPhone 4 and 4S absent a migration to iOS 7 (and resulting decrease in functionality with

9

migration), these models’ resale prices would have dropped due to a loss in value in the market for
a valued feature.” Hastings Report at 20. Class members’ damages are therefore equal to “the

11
United States District Court
Northern District of California

10

resale value of the phone compared to what the resale value would have been but-for the Break.”

12

Id. at 21. Dr. Hastings plans to isolate the effect of the FaceTime Break on resale value through

13

multivariate regression modeling. Id. at 22. In this approach, an economist creates a model that

14

“relates the price of the product in question to demand and supply factors which affect price based

15

on economic principles and facts about the market.” Id. The model includes a “dummy variable”

16

that “captures the average difference in price during the period the harmful act was in effect,

17

controlling for the influence on price of other demand and supply factors.” Id.

18

Dr. Hastings’ analysis relies on data from third-party smart phone resellers, in particular

19

data from Best Buy because it “contains millions of individual-level phone transactions with resale

20

prices for the phones and detailed characteristics of each phone, along with the date of the sale.”

21

Hastings Report at 24. Best Buy’s data also includes “iPhone 4 models through iPhone 6 models,

22

and Samsung Galaxy models S4 through S6.” Id. Dr. Hastings can therefore use these other

23

models as a “a comparison group for estimating the impact of the Break on iPhone 4 and iPhone

24

4S resale prices, controlling for other factors impacting demand and supply for various phone

25

models.” Id. Best Buy’s data for each phone includes relevant characteristics such as model,

26

condition, storage size, and color—which are among the primary factors consumers look to when

27
28

22
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

deciding how much they are willing to pay for a phone. Id. at 22, 24.
Dr. Hastings also plans to control for other variables that affect iPhone supply and

2

consumer demand, such as prevailing economic conditions and seasonal variation (e.g. higher

4

demand around the holidays). See id. at 22–27 (discussing this); ECF No. 224-6, Reply Report of

5

Justine S. Hastings, Ph.D. (“Hastings Reply Report”), at 13 (noting model can add further controls

6

if necessary), at 21 (“In fact, the multivariate regression in my report controls for 751 factors, all

7

of which are factors that market participants, like Best Buy, state are important demand and/or

8

supply factors impacting the price of phones.”). After accounting for outside variables, Dr.

9

Hastings’s basis for class members’ damages is the effect of the FaceTime Break on the “price of

10

the phone Best Buy received when it sold the phone to a third-party vendor for eventual resale on

11
United States District Court
Northern District of California

3

the used market to estimate damages.” Id. at 25. Dr. Hastings concludes that, based on Best Buy’s

12

data6, “the Break caused prices of used iPhone 4 and 4S devices to fall by an estimated 12.72%.”

13

Id. at 28.
It is important to be clear about what data Dr. Hastings employs. Consumers often trade-in

14
15

used iPhones to carriers like AT&T or retailers like Best Buy, receiving in-store credit or credit

16

towards the purchase of a new smartphone or in-store credit. For example, Best Buy will not offer

17

cash for used smartphones, but will provide consumers with Best Buy gift cards. ECF No. 224-16

18

at 36 (so noting); see ECF No. 200-3, Expert Report of Dana Trexler (“Trexler Report”), at 12

19

(“[I]n September 2013, Verizon and AT&T offered customers a minimum of $200 trade-in credit

20

for an iPhone 4 or iPhone 4S in ‘good and working condition.’”).7 Retailers like Best Buy then

21

resell the used iPhones in bulk to a range of third parties, such as enterprises, other retailers, and

22

operators. Trexler Report at 36. We have, then, two separate but linked markets. Significantly, Dr.

23

Hastings bases her findings off the latter retailer-to-third-party market, instead of the former

24
25
26
27
28

6

Dr. Hastings also cross-checked this finding against Amazon data, and found a comparable
impact of 17.58 percent. Hastings Report at 26–27.
7
This is as compared to a more traditional exchange of an iPhone for cash, as would be found on
sites like Craigslist or eBay. Trexler Report at 12 (so noting).
23
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

consumer-to-retailer market. Hastings Report at 25 (noting damages model is based on “the price

2

of the phone Best Buy received when it sold the phone to a third-party vendor” instead of the

3

“trade-in value for phones”).

4

Apple has nine responses to this, none of which are persuasive. First, Apple argues that Dr.

5

Hastings’s model is flawed because it measures the diminution in value for all iPhone 4 and

6

iPhone 4S devices, even though iPhones that had already upgraded to iOS 7 at the time of the

7

FaceTime Break were not affected. This is unpersuasive. Dr. Hastings’s model presumes that the

8

FaceTime Break had a measurable effect on how much the market valued the iPhone 4 and iPhone

9

4S, and that the reduction in market value can be expressed as a percentage. The fact that she has
examined the market as a whole is unproblematic because she is attempting to measure market

11
United States District Court
Northern District of California

10

demand in the aggregate rather than demand for un-upgraded iPhones. See In re Lidoderm

12

Antitrust Litig., 2017 WL 679367, at *12 (N.D. Cal. Feb. 21, 2017) (“[T]hat the experts dispute

13

what the appropriate inputs should be does not undermine the approach or the reliability of …

14

[Plaintiffs’ damages] model.”). This is especially true given that damages calculations under the

15

UCL are “particularly forgiving” such that California requires “only that some reasonable basis of

16

computation of damages be used … even if the result reached is an approximation.” Lambert v.

17

Nutraceutical Corp., 870 F.3d 1170, 1183 (9th Cir. 2017), cert. granted on other grounds, 138 S.

18

Ct. 2675 (2018); Pulaski & Middleman, LLC v. Google, Inc., 802 F.3d 979, 988 (9th Cir. 2015)

19

(“[T]he fact that the amount of [restitution] damage may not be susceptible of exact proof or may

20

be uncertain, contingent or difficult of ascertainment does not bar recovery.” (citation omitted)).

21

Second, Apple argues that Dr. Hastings bases her analysis on the wrong dataset. As noted

22

above, Best Buy provided Dr. Hastings with data indicating how much consumers received when

23

they traded in their used iPhones (the retail market) and with data indicating how much Best Buy

24

received when it resold those used iPhones to other entities (the wholesale market). Apple faults

25

Dr. Hastings for relying on the latter dataset, arguing that her model does not measure “the amount

26

class members could have received for their devices” but rather measures only how much Best

27
28

24
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

Buy received for reselling used iPhones in bulk to third parties. Opp. at 16. As proof, Apple points

2

to the finding of its damages expert, Dana Trexler, that iPhones’ trade-in value increased8 after the

3

FaceTime Break, which stands in stark contrast to Dr. Hastings’s finding that prices on the

4

wholesale market fell. Trexler Report at 44–52.
Plaintiffs respond that relying on trade-in value would be even worse. Plaintiffs note that

6

the headline prices for trading in used iPhones are distorted by rebates and promotions—a point

7

Apple later echoes—and that consumers usually receive credit towards a new smartphone or a

8

store gift card instead of cash. Hastings Reply Report at 18; Reply at 8 n.23; see Opp. at 17 (Apple

9

faulting Dr. Hastings’s model for failing to “control for promotions and news related to the iPhone

10

4 and 4S”). Plaintiffs therefore suggest that it is better to rely on wholesale data that can provide a

11
United States District Court
Northern District of California

5

reliable cash-per-iPhone figure than to rely on retail trade-in prices skewed by promotions and

12

rebates, and which pay consumers in hard-to-value gift cards and trade-in credits. See also

13

Saavedra v. Eli Lilly & Co., 2014 WL 7338930, at *5 (C.D. Cal. Dec. 18, 2014) (finding “the

14

numerous complicating factors in the prescription drug market sever the relationship between

15

price and value” such that “a consumer’s out-of-pocket cost for a drug is not a proxy for the drug's

16

value to the consumer”). Dr. Hastings even indicates that it would be impossible to construct a

17

damages model based on trade-in value because trade-in value “may include a combination of

18

market value of the phone, discounts, and incentives towards purchases of additional Best Buy

19

products” that “cannot be separated by the economist.” Hastings Report at 25; Hastings Reply

20

Report at 18 (stating trade-in price is not a suitable basis for damages modelling “because it often

21

contain[s] discounts and tie-ins to in-store purchases of new products bundled in with the trade-in

22

price, thus preventing the econometrician from being able to separately determine the diminished

23

value of the phone.”). More fundamentally, Plaintiffs dispute Apple’s assumption that the prices

24

for used iPhones on the retail and wholesale markets are divorced from one another. As Dr.

25
26
27
28

8

However, the increase in trade-in price was not uniform as Trexler freely admits that some
affected iPhones’ trade-in value decreased after the FaceTime Break. Trexler Report at 47.
25
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

Hastings points out, “all major platforms benchmark their phone offers against wholesale market

2

prices.” Hastings Reply Report at 19.

3

Plaintiffs have the better of this argument, though the case is close. Comcast holds that
damages models “must measure only those damages attributable to” the plaintiff’s theory of

5

liability. Comcast, 569 U.S. at 35. Plaintiffs’ theory of liability is that the FaceTime Break reduced

6

the market value of class members’ iPhones, i.e. how much class members could expect to receive

7

for their iPhones on the open market. Dr. Hastings concluded that the only feasible way to

8

determine this amount was to look to the prices Best Buy received from bulk sales to third parties

9

in the wholesale market and to use it as a proxy. The Court agrees with Apple that there are flaws

10

with this method, and it is possible that Plaintiffs’ model’s reliance on an indirect market measure

11
United States District Court
Northern District of California

4

has inflated class members’ damages. Nevertheless, the question before the Court is whether

12

Plaintiffs’ model “measure[s] only those damages attributable to” the plaintiff’s theory of liability.

13

Comcast, 569 U.S. at 35. Although “Plaintiffs’ model may be flawed, … [Apple] has not

14

established that the model is not tied to plaintiffs’ theory of liability.” In re: Lenovo, 2016 WL

15

6277245, at *21. Moreover, despite Apple’s assertion that the trade-in market is the appropriate

16

basis for a damages measurement, Apple does not explain how Plaintiffs could have accounted for

17

the distorting effects of promotions and rebates. The bottom line is that Apple’s arguments about

18

the appropriate dataset raise a factual dispute that goes to the weight of Dr. Hastings’s report, not

19

whether her damages model “measure[s] only those damages attributable to” Plaintiffs’ theory of

20

liability. Comcast, 569 U.S. at 35.

21

Third, Apple argues that Dr. Hastings has not accounted for the value of the free upgrade

22

to iOS 7. In particular, Apple argues that class members’ harm from losing access to FaceTime

23

was minimal because class members could quickly upgrade to iOS 7 at no cost and regain access

24

to FaceTime. This is unpersuasive. Plaintiffs allege that class members’ iPhones suffered a

25

significant decrease in performance upon upgrading to iOS 7. FAC ¶¶ 70–87. Thus, Apple’s claim

26

that upgrading to iOS 7 redressed all harm fails because it ignores this key aspect of Plaintiffs’

27
28

26
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

case. Apple’s technical expert, Dr. Avi Rubin, appears to suggest that there were not major

2

performance differences between iOS 6 and iOS 7, or that performance issues were attributable to

3

other sources. See ECF No. 223-6 at 13–15. Given that Plaintiffs have credible evidence pointing

4

the opposite direction, this is a factual dispute not suited for resolution at the class certification

5

stage. FAC ¶¶ 72–78. Again, Plaintiffs’ damages theory is that class members were harmed by the

6

decreased value of their iPhones that resulted from the FaceTime Break. This harm remains

7

constant whether or not class members upgraded. Indeed, the Court rejected a similar version of

8

this argument at the motion to dismiss phase. See Grace, 2017 WL 3232464, at *10 (“Apple’s

9

argument confuses Plaintiffs’ alleged injury. Plaintiffs’ alleged injury is that Apple permanently

10

deactivated FaceTime for iOS6, which has caused Plaintiffs’ iPhones to diminish in value.”).
Fourth, Apple argues that Dr. Hastings’s dataset is fatally undermined by the fact that Best

United States District Court
Northern District of California

11
12

Buy upgraded all used iPhone 4 and iPhone 4S phones to iOS 7 prior to resale during the time

13

period Dr. Hastings analyzed. This means all of the iPhones included in Dr. Hastings’s analysis

14

were sold with iOS 7. It follows, Apple contends, that none of the devices ever lost the ability to

15

use FaceTime because the FaceTime Break only affected iPhones running iOS 6. Apple argues

16

that this fatally undermines Dr. Hastings’s analysis because she cannot create a “benchmark for

17

the value of an iPhone 4 or 4S running FaceTime on iOS 6 before April 16, 2014.” Opp. at 16.

18

Apple’s theory merely reframes the argument that the Court rejected above. Plaintiffs’ theory is

19

that class members were harmed by the FaceTime Break and that this harm is reflected in the price

20

that the market was willing to pay for used iPhone 4 and iPhone 4S devices.9 Because Plaintiffs

21

have consciously chosen to measure class members’ harm through the market value of class

22

members’ iPhone 4 and iPhone 4S devices, what matters for benchmark purposes is how much the

23

devices were valued before and after the FaceTime Break instead of what version of iOS the

24

devices were running.

25
26
27
28

9

Any concern that this market price might not account for the potentially different value of
jailbroken iPhones is addressed by excluding jailbroken iPhones from the class definition. See
Section III.A.3, supra.
27
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

Fifth, Apple criticizes Dr. Hastings’s model for failing to “control for promotions and news

2

related to the iPhone 4 and 4S.” Opp. at 17. However, Apple’s leading examples are trade-in

3

promotions and Dr. Hastings deliberately relied on wholesale resale value to minimize the

4

distorting effect of trade-in promotions. Hastings Reply Report at 18 (so noting). In other words,

5

Apple faults Dr. Hastings for failing to adequately account for trade-in promotions after Dr.

6

Hastings structured her model to render them irrelevant. That is not a strong argument. Apple’s

7

argument is also inconsistent: Apple makes much of Trexler’s finding that trade-in values rose

8

after the FaceTime Break, but Trexler’s finding is based on averaging trade-in values without

9

controlling for other factors such as promotions. See id. at 23–24. Thus, Apple faults Dr. Hastings

10

for inadequate controls when its own expert applied even fewer controls to her analysis.

United States District Court
Northern District of California

11

Moreover, the Ninth Circuit has made clear that class certification does not require

12

damages models based on flawless price averages. In Lambert, a district court denied certification

13

because the plaintiffs’ damages model was based on suggested retail price instead of actual retail

14

price. The Ninth Circuit reversed, and reasoned that a “suggested retail price does not

15

automatically configure an average, but such a precise average is unnecessary for class

16

certification. At this stage, the question is only whether [the plaintiff] … has presented a workable

17

method.” Id. at 1183–84. Here, Dr. Hastings has proposed a damages model based on actual resale

18

data that controls for a number of external factors. That is enough. Indeed, this Court has

19

previously explained that assertions, like Apple’s, that models were flawed owing to their failure

20

to include this or that variable are “prototypical concern[s] that go[] to weight, not admissibility.”

21

In re High-Tech Employee Antitrust Litig., 2014 WL 1351040, at *20 (N.D. Cal. Apr. 4, 2014);

22

see Bazemore v. Friday, 478 U.S. 385, 400 (1986) (“Normally, failure to include variables will

23

affect the analysis’ probativeness, not its admissibility.”).

24

Sixth, Apple makes the related argument that Dr. Hastings’s report is fatally flawed

25

because it fails to control for devices that are “jailbroken, damaged, or ha[ve] undergone

26

unauthorized repair.” Opp. at 16. The Court disagrees. Apple’s argument applies to at most a small

27
28

28
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

percentage of the class. For instance, less than 5 percent of iPhone owners jailbreak their devices.

2

Hastings Reply Report at 14–15. Moreover, to the extent that certain class members are entitled to

3

less or no relief because their iPhones fall into these categories, this issue can easily be remedied

4

by the use of claims forms. See, e.g., In re Korean Ramen Antitrust Litig., 2017 WL 235052, at

5

*21 (N.D. Cal. Jan. 19, 2017) (“Post-judgment claims forms and other tools can be used to allow

6

defendants to test an absent class member’s purported entitlement to damages and to appropriately

7

apportion damages between class members.”); Nevarez v. Forty Niners Football Co., LLC, 2018

8

WL 3830137, at *9–10 (N.D. Cal. July 12, 2018) (finding that class members could certify

9

whether they encountered an actionable violation of a statute and were therefore entitled to

10

damages); see Lambert, 870 F.3d at 1183 (noting damages calculations under the UCL are

11
United States District Court
Northern District of California

1

“particularly forgiving” such that California requires “only that some reasonable basis of

12

computation of damages be used … even if the result reached is an approximation.”); id. at 1882

13

(“We have repeatedly emphasized that uncertain damages calculations should not defeat

14

certification.”); Pulaski, 802 F.3d at 988 (noting that California law “requires only that some

15

reasonable basis of computation of damages be used, and the damages may be computed even if

16

the result reached is an approximation.”). The Court notes also that, as with Apple’s last argument,

17

this type of missing variable argument goes to the weight of Dr. Hastings’s analysis rather than its

18

admissibility. High-Tech, 2014 WL 1351040, at *20; Bazemore, 478 U.S. at 400.

19

Seventh, Apple alleges that Dr. Hastings “arbitrarily” defined the damages period as

20

extending to December 31, 2016, over 2.5 years after the FaceTime Break. Opp. at 16. The Court

21

disagrees. Dr. Hastings selected a December 31, 2016 end date because the underlying dataset

22

extended to December 31, 2016. Trexler Report at 54. Apple may believe that the FaceTime

23

Break’s effects did not last that long, but a bald assertion that it was “arbitrary” for Dr. Hastings to

24

base her analysis on all the data before her is unpersuasive.

25
26
27
28

Eighth, Apple argues that Dr. Hastings erred in calculating a single average price impact
because the iPhone 4 and iPhone 4S are different devices with different attributes. To start, the
29
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

Court previously rejected a similar version of this theory at the motion to dismiss phase, when

2

Apple suggested that Grace and Potter lacked standing to bring claims on behalf of iPhone 4S

3

owners because Grace and Potter had purchased the iPhone 4. See Grace, 2017 WL 3232464, at

4

*8–9 (rejecting Apple’s argument because it was undisputed that both devices used iOS, had

5

FaceTime, and lost FaceTime as a result of the FaceTime Break if they had not upgraded). Yet

6

even if Apple is correct and separate price impacts should be calculated for the iPhone 4 and

7

iPhone 4S, Apple nowhere suggests that Dr. Hastings’s model is incapable of doing so. In fact,

8

Plaintiffs indicate that Dr. Hastings’s model could be combined with claims forms to determine

9

how much a class member was owed based on a range of iPhone characteristics. Reply at 10
(“[D]uring the claims administration process, the percentage diminution in value may be applied

11
United States District Court
Northern District of California

10

to each class member’s device to determine each class member’s damages, taking into account the

12

same handful of key characteristics that Apple itself uses to determine the price of a phone (model

13

and condition).”).

14

Ninth, Apple argues that Plaintiffs’ damages model does not align with Plaintiffs’ UCL

15

claim because the UCL only authorizes restitution. Restitution under the UCL “must be of a

16

measurable amount to restore to the plaintiff what has been acquired by violations of the statutes,

17

and that measurable amount must be supported by evidence.” Pulaski, 802 F.3d at 988 (citation

18

omitted). California law “requires only that some reasonable basis of computation of damages be

19

used, and the damages may be computed even if the result reached is an approximation.” Id.

20

(citation omitted); Lambert, 870 F.3d at 1182 (stating in case including UCL claim that “[w]e have

21

repeatedly emphasized that uncertain damages calculations should not defeat certification.”).

22

Plaintiffs’ diminution in value damages model meets this bar. On April 16, 2014, the market

23

learned that FaceTime had stopped working on iPhone 4 and iPhone 4S devices that had not

24

upgraded to iOS 7 and that the only way to restore FaceTime was to upgrade. Consequently,

25

comparing the market value of the iPhone 4 and iPhone 4S before and after that information was

26

revealed—i.e. before and after the FaceTime Break—seems an accurate way of determining how

27
28

30
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1
2

much class members lost as a result of the FaceTime Break.
The cases to which Apple points do not compel a contrary conclusion. Brazil v. Dole

3

Packaged Foods, LLC, 2014 WL 5794873 (N.D. Cal. Nov. 6, 2014); Philips v. Ford Motor Co.,

4

2016 WL 7428810 (N.D. Cal. Dec. 22, 2016), aff’d, 726 F. App’x 608 (9th Cir. 2018). In Brazil,

5

the plaintiff brought a class action that alleged the defendant had mislabeled its food products.

6

This Court granted certification because the plaintiff’s expert presented a damages model that

7

would compare the price of the food products before the misrepresentations and the price of the

8

food products after the misrepresentations. Brazil, 2014 WL 5794873, at *5. However, the Court

9

decertified when the model failed to materialize and the alternative model proved inadequate. Id.
at 6, 14. In brief then, Brazil approved a damages model that mirrors the before-and-after Plaintiffs

11
United States District Court
Northern District of California

10

envision here, and the Court decertified only when that model failed to materialize. That is not

12

strong support for finding that Dr. Hastings’s model fails to satisfy Comcast. Nor is Philips any

13

better. The Court there rejected the proposed damages model because the damages model assumed

14

that the true value of the allegedly defective part “was $0.” Philips, 2016 WL 7428810, at *21.

15

Plaintiffs’ damages model does not make that assumption here, so Philips is inapposite.

16

In sum, the Court rejects Apple’s challenges to Dr. Phillips’s damages model and finds that

17

Plaintiffs’ damages model complies with Comcast. The Court now turns to Apple’s choice of law

18

arguments.

19

4. Choice of Law

20

Apple argues that “common issues of law do not predominate because California’s

21

consumer protection statutes [and trespass to chattels law] may not be applied to a nationwide

22

class with members in … [50] jurisdictions.” Mazza, 666 F.3d at 589. The Court agrees. California

23

choice of law rules preclude applying California law to the proposed nationwide class. In turn,

24

“[b]ecause adjudication of the nationwide claims will require application of the laws of 50 states,

25

common questions of law would not predominate for the proposed nationwide class, as is required

26

by Rule 23(b)(3).” Brazil v. Dole Packaged Foods, LLC, 2014 WL 2466559, at *14 (N.D. Cal.

27
28

31
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

May 30, 2014). However, the Court notes at the outset that this problem does not prevent

2

certification of Plaintiffs’ alternatively proposed California class.

3

“A federal court sitting in diversity must look to the forum state’s choice of law rules to
determine the controlling substantive law.” Zinser, 253 F.3d at 1187. “Under California’s choice

5

of law rules, the class action proponent bears the initial burden to show that California has

6

‘significant contact or significant aggregation of contacts’ to the claims of each class member.”

7

Mazza, 666 F.3d at 589 (quoting Wash. Mut. Bank v. Superior Court, 24 Cal.4th 906, 921 (Cal.

8

2001)). This showing “is necessary to ensure that application of California law is constitutional.”

9

Id. at 589–90. “Once the class action proponent makes this showing, the burden shifts to the other

10

side to demonstrate ‘that foreign law, rather than California law, should apply to class claims.’” Id.

11
United States District Court
Northern District of California

4

at 590 (quoting Wash. Mut. Bank, 24 Cal.4th at 921).

12

Apple’s principal place of business is in California and it appears Apple’s actions that

13

caused the FaceTime Break were taken in California. Moreover, Apple does not dispute that

14

Plaintiffs have sufficiently alleged that California has a constitutionally sufficient aggregation of

15

contacts to the claims of each putative class member in this case. Accordingly, the Court is

16

satisfied that California has sufficient contacts with the class claims. See In re Yahoo Mail, 308

17

F.R.D. at 602 (finding application of California law was constitutionally permissible where

18

defendant’s corporate headquarters were in California, the defendant’s executive decision makers

19

were largely in California, and the processes at issue were developed and directed in California);

20

In re Qualcomm Antitrust Litig., 292 F. Supp. 3d 948, 978 (N.D. Cal. 2017) (reaching same

21

conclusion on similar facts).

22

The Court now turns to whether “foreign law, rather than California law, should apply.”

23

Mazza, 666 F.3d at 590 (citation omitted). California law may be applied on a classwide basis only

24

if “the interests of other states are not found to outweigh California’s interest in having its law

25

applied.” Id. (quoting Wash. Mut. Bank, 24 Cal.4th at 921). To determine whether the interests of

26

other states outweigh California’s interest, courts administer a three-step government interest test.

27
28

32
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

The court must first determine whether the law of the other states is materially different from

2

California law. Id. at 590. Second, if there are differences, the court determines whether the other

3

state has an interest in having its law applied to decide whether a true conflict exists. Id. at 591–

4

92. Third, if another state has an interest, the court determines which state’s interest would be

5

most impaired if its policy were subordinated to the law of another state. Id. at 593.

6
7

a. Material Differences in State Law
Plaintiffs bring a claim for violation of California’s Unfair Competition Law (“UCL”),

8

California Business and Professions Code § 17200 et seq., and a claim for trespass to chattels

9

under California law. The Court examines each in turn, and finds that both claims are materially

10

United States District Court
Northern District of California

11

different from the laws of other states.
Apple lists a number of ways in which California’s UCL is materially different from other

12

states’ consumer protection laws. Plaintiffs do not dispute Apple’s argument. Even so,

13

thoroughness compels pointing out at least some of the distinctions. First, pre-filing notice is not

14

required in California, but is a prerequisite in Alabama, Georgia, Indiana, Maine, Massachusetts,

15

Mississippi, Texas, West Virginia, and Wyoming. ECF No. 201-7 at 2–26 (Apple’s state law

16

appendix). Plaintiffs did not provide pre-filing notice, so this distinction is material because it is

17

potentially dispositive. In re Yahoo Mail, 308 F.R.D. at 602 (noting differences that would spell

18

difference between success and failure of a claim are material). Second, Alabama, Georgia,

19

Louisiana, Mississippi, Montana, South Carolina, and Tennessee prohibit consumer protection

20

class actions outright. See id. This too is a potentially dispositive distinction. Fourth, while the

21

UCL does not require scienter, but “many other states’ consumer protection statutes do require

22

scienter.” Mazza, 666 F.3d at 591 (listing examples). Fifth, “California also requires named class

23

plaintiffs to demonstrate reliance, while some other states’ consumer protection statutes do not.”

24

Id. (listing examples). Given this list of distinctions, the Court finds that there are material

25

differences between the UCL claim and other states’ consumer protection laws.

26
27
28

The Court now turns to Plaintiffs’ trespass to chattels claim. Apple argues that California’s
33
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

trespass to chattels law materially differs from other states’ trespass to chattels laws in three

2

respects. First, Apple argues that California’s scienter requirement differs from that of other states.

3

Second, Apple argues that applicable statutes of limitations also vary from state to state. Third,

4

Apple argues that some states do not require a meaningful harm showing. The Court rejects the

5

first two arguments, but agrees with the third.
Apple points to Kentucky, Arizona, Colorado, and New Mexico in support of its argument

7

that California’s scienter requirement for trespass to chattel claims differs materially from those of

8

other states. Opp. at 21. Yet according to the cases in Apple’s own Appendix, each state’s law—

9

like California’s—requires that the trespasser have acted intentionally. California law provides

10

that “trespass to chattels lies where an intentional interference with the possession of personal

11
United States District Court
Northern District of California

6

property has proximately caused injury.” Hamidi, 30 Cal. 4th at 1350–51 (internal quotation

12

marks and citation omitted). Arizona provides that “an intermeddling is not a dispossession unless

13

the actor intends to exercise a dominion and control over [property] inconsistent with a possession

14

in any other person other than himself.” Hagar v. Rodbell, 2012 WL 827068, at *6 (D. Ariz. Mar.

15

12, 2012). Kentucky requires an “intent to intermeddle with the particular property … .” Ingram

16

Trucking, Inc. v. Allen, 372 S.W. 3d 870, 873 (Ky. Ct. App. 2012). Colorado holds that “it is

17

necessary that the defendant have acted for the purpose of interfering with the chattel.” Mountain

18

States Tel. & Tel. Co. v. Horn Tower Const. Co., 363 P.2d 175, 178 (Colo. 1961). Finally, New

19

Mexico provides that “[t]he intention required to make one liable for trespass to a chattel is that he

20

must have acted for the purpose of interfering with the chattel or with knowledge that a

21

disturbance thereof is substantially certain to occur.” Texas-New Mexico Pipeline Co. v. Allstate

22

Const., Inc., 369 P.2d 401, 40 (N.M. 1962). In sum, because California, Kentucky, Arizona,

23

Colorado, and New Mexico all require intentional action Apple has failed to show that there is a

24

material difference between California’s scienter requirement and those of other states.

25
26
27
28

Apple next contends that the length of the applicable statute of limitations differs
materially from state to state. In particular, although California’s statute of limitations for this
34
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

claim (like many other states) is three years, Louisiana has a one-year statute of limitations, and

2

the statute of limitations is two years in Alaska, Arizona, Hawaii, Indiana, Kansas, Kentucky,

3

Montana, Oklahoma, Pennsylvania, Texas, and West Virginia. Cal. Civ. Proc. Code § 338; Alaska

4

Stat. § 09.10.070; Ariz. Rev. Stat. Ann. § 12-542; Haw. Rev. Stat. Ann. § 657-7; Ind. Code Ann. §

5

34-11-2-4; Kan. Stat. Ann. § 60-513; Ky. Rev. Stat. Ann. § 413.125; La. Civ. Code Ann. art.

6

3492; Mont. Code Ann. § 27-2-207; Okla. Stat. Ann. tit. 12, § 95; 42 Pa. Stat. and Cons. Stat.

7

Ann. § 5524; Tex. Civ. Prac. & Rem. Code Ann. § 16.003; W. Va. Code Ann. § 55-2-12.

8

It is not clear that these statute of limitations differences matter for this case. The
FaceTime Break occurred on April 16, 2014, and this suit was filed approximately 33 months later

10

on February 2, 2017. ECF No. 1. If Plaintiffs’ trespass to chattels claim accrued on April 16, 2014,

11
United States District Court
Northern District of California

9

it is likely barred in the 12 states with one or two year statutes of limitations. In that case, the

12

statute of limitations difference would be material because it “spell[s] the difference between the

13

success and failure of a claim.” In re Yahoo Mail, 308 F.R.D. at 602 (quoting Mazza, 666 F.3d at

14

591). However, Apple does not suggest that these 12 states diverge from the usual rule that claims

15

accrue when a reasonable person would have learned about the facts necessary to file suit. Indeed,

16

even Louisiana applies the “discovery rule” whereby the one year statute of limitations periods

17

begins to run “on the date the injured party discovers or should have discovered the facts upon

18

which his cause of action is based.” Body By Cook v. Ingersoll-Rand Co., 39 F. Supp. 3d 827, 836

19

(E.D. La. 2014). That matters here because Plaintiffs have consistently argued that Apple’s

20

conduct came to light only as the result of a 2016 patent trial. ECF No. 1 at 26 (“Apple refused to

21

disclose the truth behind why FaceTime had suddenly stopped working … Apple publicly

22

disclosed nothing about any of this (until it was reluctantly forced to do so at the VirnetX retrial in

23

2016).”); FAC at 28 (stating same). Given that the transcripts from that trial were not released

24

until May 9, 2016, Plaintiffs have a reasonable argument that class members’ claims did not

25

accrue until that date—in which case this action satisfies even Louisiana’s one year statute of

26

limitations. Reply at 13. Specifically, less than 9 months elapsed between the release of the

27
28

35
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1
2

transcripts on May 9, 2016, and Plaintiffs initiating this suit on February 2, 2017.
The foregoing is not meant to establish that Plaintiffs’ claims did or did not accrue on a

3

certain date. Rather, the point is that Apple’s failure to explain why the statute of limitations

4

distinction matters based on the facts of this case dooms its argument. The Court agrees that the

5

statute of limitations differences would be material if Plaintiffs’ claims accrued on April 16, 2014.

6

But Apple has not argued for an April 16, 2014 accrual date. In fact, Apple has not argued for an

7

accrual date at all. All Apple has done is make the point that different states have different statutes

8

of limitations. Opp. at 21. That is not enough. The mere “fact that two or more states are involved

9

does not itself indicate that there is a conflict of law problem. A problem only arises if differences
in state law are material, that is, if they make a difference in this litigation.” Mazza, 666 F.3d at

11
United States District Court
Northern District of California

10

590 (internal quotation marks and citation omitted). Because Apple has not attempted to explain

12

why differences in state statutes of limitations make a difference in this litigation, the Court cannot

13

find that the differences are material.

14

A similar problem confronted the district court in In re Packaged Seafood Products

15

Antitrust Litigation, 277 F. Supp. 3d 1167 (S.D. Cal. 2017). As here, the defendants argued that

16

statutes of limitations were materially different and therefore barred a nationwide class. Id. at

17

1180–81. The district court rejected this argument:

18

22

Defendants nowhere address whether the statutes of limitations might actually
preclude certain claims given the facts of this case, and even Defendants’ sole
authority (which is contradicted by several cases Plaintiffs offer) rested its relevant
analysis on such a showing. Furthermore, Plaintiffs assert that “all the consumer
claims at issue are timely under any of the relevant statutes of limitation” and
Defendants nowhere counter this statement. Accordingly, Defendants have not here
shown that the differing statutes of limitation are material within the meaning of
Mazza.

23

Id. at 1181 (internal citations omitted). The Court finds this reasoning persuasive, and adopts it. It

24

is not enough to simply note that different statutes of limitations exist. Apple must actually explain

25

why those differences might matter for this case, especially because it has been obvious since the

26

outset that Plaintiffs believe Apple’s conduct did not emerge until the 2016 patent litigation. See

19
20
21

27
28

36
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

Grodzitsky v. Am. Honda Motor Co. Inc., 2014 WL 718431, at *8–9 (C.D. Cal. Feb. 19, 2014)

2

(“California’s choice-of-law analysis must be conducted on a case-by-case basis because it

3

requires analyzing various states’ laws under the circumstances of the particular case and given the

4

particular legal issue in question.”); Mullins v. Premier Nutrition Corp., 2016 WL 1535057, at *10

5

(N.D. Cal. Apr. 15, 2016) (“At the class certification stage, other district courts have rejected blind

6

reliance on Mazza, and instead require defendants to provide case-specific analysis as to the

7

second and third prongs of the governmental interest test.”). In sum, the Court rejects Apple’s

8

argument that differences between statutes of limitations are material here.

9

Nevertheless, the Court agrees with Apple’s third argument that the level of interference
required for a trespass to chattels claim materially differs from state to state. To prevail on a claim

11
United States District Court
Northern District of California

10

for trespass to a computer system or similar device, California law requires the plaintiff to show

12

that the trespass “(1) caused physical damage to the personal property, (2) impaired the condition,

13

quality, or value of the personal property, or (3) deprived plaintiff of the use of personal property

14

for a substantial time.” Grace, 2017 WL 3232464, at *11 (quoting Fields, 2013 WL 5340490, at

15

*4). Plaintiffs are here proceeding under the second of these options and argue in particular that

16

Apple’s conduct harmed them by diminishing the value of their iPhones.

17

The problem for Plaintiffs is that Apple has provided authority—whose accuracy Plaintiffs

18

do not dispute—indicating that other states would permit liability even absent a showing of harm.

19

For instance, in Iowa “[i]f a party cannot prove actual injury, the law entitles him to recover

20

nominal damages.” Sheridan v. City of Des Moines, 2000 WL 35626707, at *4 (S.D. Iowa Aug.

21

10, 2000). Likewise, in Florida “[a]ny unlawful interference, however slight, with another’s

22

enjoyment of his or her personal property is a trespass, forcible dispossession being unnecessary

23

as a rule.” Am. Fed’n of Labor-Cong. of Indus. Organizations v. City of Miami, 2008 WL

24

11333331, at *9 (S.D. Fla. Sept. 4, 2008) (quoting 55 Fla. Jur. 2d Trespass § 3 (1984)) (emphasis

25

added). Thus, Florida would permit liability even absent physical damage, impairment, or

26

deprivation—the interference is enough. Similarly, North Carolina does not require “actual

27
28

37
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

damage as an essential element” of a trespass to chattels claim where “actual damage” is defined

2

as “some actual loss, hurt or harm resulting from the illegal invasion of a legal right.” Hawkins v.

3

Hawkins, 101 N.C. App. 529, 532–33 (1991). In other words, North Carolina permits liability for

4

trespass to chattels even if the plaintiff does not show they suffered “some actual loss, hurt or

5

harm resulting from the illegal invasion of a legal right.” Id. at 532. The bottom line, then, is that

6

Apple points to significant authority showing that California’s trespass to chattels law differs

7

materially from that of other states.

8
9

Plaintiffs have chosen not to contest or respond to Apple’s characterization of other states’
trespass to chattels laws. Instead, Plaintiffs argue that the differences Apple highlights are not
material and therefore do not preclude a nationwide class. The Court disagrees. Apple has

11
United States District Court
Northern District of California

10

consistently argued that it cannot be liable because the FaceTime Break did not cause class

12

members any actual harm. The distinction between a state like California that requires some

13

degree of harm and states like Florida, North Carolina, and Iowa that do not is therefore highly

14

material because applying those states’ laws in place of California’s could “spell the difference

15

between the success and failure of a claim.” In re Yahoo Mail, 308 F.R.D. at 602 (quoting Mazza,

16

666 F.3d at 591). Put otherwise, the differences are material because if Apple’s no harm theory is

17

correct, Apple cannot be liable in California but may still be liable in Florida, North Carolina, or

18

Iowa. Mazza itself underscores the point. The Ninth Circuit there found that the existence of

19

absence of a reliance requirement in different states’ consumer protection laws—in essence, a

20

requirement that a plaintiff show that a defendant had caused them harm—constituted a material

21

difference. Mazza, 666 F.3d at 591.

22

In sum, the Court finds that Plaintiffs’ California UCL and trespass to chattels claims differ

23

materially from other states’ analogous laws. The Court now turns to the second step in the

24

inquiry, whether other states have an interest in applying their laws.

25
26
27
28

b. Other States’ Interests
“[S]tates may permissibly differ on the extent to which they will tolerate a degree of
38
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

lessened protection for consumers to create a more favorable business climate for the companies

2

that the state seeks to attract to do business in the state.” Mazza, 666 F.3d at 592; see id. (“[T]he

3

district court erred by discounting or not recognizing each state’s valid interest in shielding out-of-

4

state businesses from what the state may consider to be excessive litigation.”). Consequently,

5

“each state has an interest in setting the appropriate level of liability for companies conducting

6

business within its territory.” Id.; In re Yahoo Mail, 308 F.R.D. at 603 (“Each state has an interest

7

in being able to delineate the appropriate standard of liability and the scope of recovery based on

8

its understanding of the balance between the interests of individuals and corporate entities

9

operating within its territory.”).

10

There is no doubt that Plaintiffs’ UCL claim implicates other states’ interests in balancing

United States District Court
Northern District of California

11

business climate and consumer protections. The Ninth Circuit in Mazza recognized that other

12

states’ interest in balancing business climate and consumer protections applied in the putative 44

13

state consumer class action, which likewise alleged inter alia a violation of the UCL. Mazza, 666

14

F.3d at 587, 592–93; Potter v. Chevron Prods. Co., 2018 WL 4053448, at *11 (N.D. Cal. Aug. 24,

15

2018) (another prospective class action involving a claim under the UCL that found states’

16

balancing interest was implicated); Frezza v. Google Inc., 2013 WL 1736788, at *7 (N.D. Cal.

17

Apr. 22, 2013) (same).

18

Similarly, the Ninth Circuit and the California Supreme Court have both recognized that

19

other jurisdictions have an interest in seeing their tort laws applied to injuries that occurred within

20

their borders. McGhee v. Arabian Am. Oil Co., 871 F.2d 1412, 1425 (9th Cir. 1989), as amended

21

on reh’g (Apr. 28, 1989) (“Saudi Arabia has some legitimate interest in seeing that Saudi law

22

determines the consequences of actions within its borders causing injury to people who reside

23

there.”); Offshore Rental Co. v. Cont’l Oil Co., 22 Cal. 3d 157, 160, 168 (1978). The California

24

Supreme Court’s Offshore Rental decision is especially relevant. Offshore Rental held that

25

Louisiana law should apply over California law where the plaintiff was a California resident, the

26

alleged tort (negligence) occurred in Louisiana, and defendant corporation was a citizen of neither

27
28

39
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

state. The opinion stressed Louisiana’s “vital interest in promoting freedom of investment and

2

enterprise within Louisiana’s borders” and found this interest prevailed over California’s interest

3

in compensating its residents. Offshore Rental, 22 Cal. 3d at 168; see McCann v. Foster Wheeler

4

LLC, 48 Cal. 4th 68, 91–94 (2010) (discussing Offshore Rental and affirming its continued

5

vitality).
In sum, other states have an interest in seeing their laws applied to Plaintiffs’ claims. The

6
7

Court now balances those states’ interests against California’s interest in seeing its law applied.
c. Balance of Interests

8

California law generally recognizes that “with respect to regulating or affecting conduct

10

within its borders, the place of the wrong has the predominant interest.” Mazza, 666 F.3d at 593

11
United States District Court
Northern District of California

9

(citation omitted). “California considers the ‘place of the wrong’ to be the state where the last

12

event necessary to make the actor liable occurred.” Id.
Plaintiffs argue that the place of the wrong is California because that is where Apple

13
14

undertook the actions which caused the FaceTime Break. Plaintiffs point to McCann, in which a

15

California plaintiff sued a Delaware10 corporation in tort for exposing the plaintiff to asbestos in

16

Oklahoma. McCann, 48 Cal. 4th at 74, 91. McCann applied California’s choice of law test and

17

concluded that Oklahoma law should apply over California’s. Specifically, McCann found that the

18

plaintiff’s claim had accrued where he was exposed to asbestos (i.e. Oklahoma) rather than where

19

the plaintiff fell ill and sued (i.e. California). Id. at 101–02.
It is unclear why Plaintiffs believe that McCann supports their argument. McCann did not

20
21

hold that the plaintiff’s claim accrued where the asbestos was manufactured, but rather held that

22

the claim accrued where the plaintiff himself was exposed to the asbestos. Likewise, class

23

members’ claims did not accrue in California where the FaceTime Break originated, but rather in

24

the state in which class members’ iPhones ceased being able to use FaceTime. Framed differently,

25
26
27
28

10

The corporation was incorporated in Delaware, but was headquartered in New York. McCann,
48 Cal. 4th at 91.
40
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

the last event necessary to make Apple liable was the loss of FaceTime access on class members’

2

iPhones, and that event took place wherever a given class member was on April 16, 2014.

3

Mazza underscores the point. There, a putative class with members in 44 states accused

4

Honda of misrepresenting the characteristics of its braking system in marketing materials. Mazza,

5

666 F.3d at 585, 587. Honda was a California corporation, and the marketing materials had been

6

created in California. Id. at 597 (Nelson, J., dissenting). Even so, Mazza found California law

7

could not apply to out of state class members because “the last events necessary for liability as to

8

the foreign class members—communication of the advertisements to the claimants and their

9

reliance thereon in purchasing vehicles—took place in the various foreign states, not in

10

United States District Court
Northern District of California

11

California.” Id. at 594.
The Court grants that Mazza was a false advertising case, and this suit is not. Even so,

12

Mazza likewise concerned a UCL claim and its key finding—that a claim accrues where the harm

13

is felt rather than where it originated—remains fully applicable here. The Court therefore finds

14

that the place of the wrong here is where class members were exposed to the FaceTime Break,

15

even if many of Apple’s actions preceding the exposure took place in California. Cf. Brazil, 2014

16

WL 2466559, at *14 (“The place of the wrong is the geographic location where the

17

misrepresentations were communicated to the consumer.”); Frezza, 2013 WL 1736788, at *5–7

18

(relying on Mazza to reach similar conclusion).

19

In sum, because the place of the wrong was the state in which each class member was

20

exposed to the FaceTime Break each class member’s claims should be governed by the law of the

21

state in which they were exposed to the FaceTime Break. Consequently, the Court will not certify

22

a nationwide class because doing so would require application of all 50 states’ laws. However, this

23

does not bar certification of Plaintiffs’ alternative California-only class.

24
25
26
27
28

C. Rule 23(b)(3) Superiority
Rule 23(b)(3) provides four factors that a court must consider in determining whether a
class action is superior to other methods of adjudication. These factors are:
41
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

4

(A) the class members’ interests in individually controlling the prosecution or
defense of separate actions;
(B) the extent and nature of any litigation concerning the controversy already begun
by or against class members;
(C) the desirability or undesirability of concentrating the litigation of the claims in
the particular forum; and
(D) the likely difficulties in managing a class action.

5

Fed. R. Civ. P. 23(b)(3). “[T]he purpose of the superiority requirement is to assure that the class is

6

the most efficient and effective means of resolving the controversy.” Wolin v. Jaguar Land Rover

7

N. Am., LLC, 617 F.3d 1168, 1175 (9th Cir. 2010) (quoting 7A Charles Wright, Arthur Miller &

8

Mary Kay Kane, Federal Practice and Procedure, § 1779 at 174 (3d ed. 2005)). “In cases in which

9

plaintiffs seek to recover relatively small sums and the disparity between litigation costs and the

10

recovery sought may render plaintiffs unable to proceed individually, ‘class actions may permit

11

the plaintiffs to pool claims which would be uneconomical to bring individually.’” Moore v. Ulta

12

Salon, Cosmetics & Fragrance, Inc., 311 F.R.D. 590, 622 (C.D. Cal. 2015) (quoting Local Joint

13

Executive Bd. Of Culinary/Bartender Tr. Fund v. Las Vegas Sands, Inc., 244 F.3d 1152, 1163 (9th

14

Cir. 2001)).

1
2

United States District Court
Northern District of California

3

15

Apple’s brief superiority argument is that the proposed class will be hard to manage owing

16

to the difficulty in determining who is actually a class member. Specifically, Apple argues that

17

determining membership will be difficult because there is no “systematic way to identify which

18

current iPhone 4 and 4S users owned the same devices on April 16, 2014 and which operating

19

system they were running at the time.” Opp. at 24. Apple’s argument assumes that class

20

membership is limited to individuals who still own the iPhone 4 or iPhone 4S that they owned at

21

the time of the FaceTime Break. This assumption is mistaken. As Plaintiffs point out, the proposed

22

class definitions base class membership on whether an individual owned an iPhone 4 or iPhone 4S

23

on April 16, 2014, i.e. when the FaceTime Break occurred. Thus, whether class members continue

24

to own the same devices over 4 years later, or have sold them, given them away, or otherwise

25

disposed of them is beside the point. What matters for class identification purposes is whether an

26

individual owned an iPhone 4 or iPhone 4S running iOS 6 on April 16, 2014. Apple has already

27
28

42
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

indicated that it can provide this information. ECF No. 224-19 at 2 (Apple responding to

2

interrogatory requesting identification of class members by “refer[ring] Plaintiffs to … [a] report

3

that shows the iOS version used by U.S. Apple iPhone 4 and/or iPhone 4S users as identified by

4

serial number for each of April 15, 16th, and 17th of 2014 and …[a] report … that shows U.S.-

5

only Apple iPhone 4 and/or iPhone 4S users with iOS versions older than iOS 7, as identified by

6

serial number, from April 16, 2014 onwards.”).

7

In sum, Plaintiffs have established Rule 23(b)(3) superiority. Plaintiffs have also

8

established Rule 23(b)(3) predominance for the amended California class, and have satisfied the

9

requirements of Rule 23(a). The Court will therefore certify a Rule 23(b)(3) amended damages

10

class. The Court now turns to whether certification of a Rule 23(b)(2) class is also appropriate.

United States District Court
Northern District of California

11
12

D. Rule 23(b)(2)
Rule 23(b)(2) allows class treatment when ‘the party opposing the class has acted or

13

refused to act on grounds that apply generally to the class, so that final injunctive relief or

14

corresponding declaratory relief is appropriate respecting the class as a whole.” Dukes, 564 U.S. at

15

360 (quoting Fed. R. Civ. P. 23(b)(2)). “[U]nlike Rule 23(b)(3), a plaintiff does not need to show

16

predominance of common issues or superiority of class adjudication to certify a Rule 23(b)(2)

17

class.” In re Yahoo Mail, 308 F.R.D. at 587. Rather, “[i]n contrast to Rule 23(b)(3) classes, the

18

focus [in a Rule 23(b)(2) class] is not on the claims of the individual class members, but rather

19

whether [the defendant] has engaged in a ‘common policy.” Id. at 599.

20

Plaintiffs’ 25 page motion for class certification devotes a single four-sentence paragraph

21

to argue for certification of a Rule 23(b)(2) class. Of that paragraph, the only statement

22

approaching a legal argument is that “[c]ertification under Rule 23(b)(2) is proper because Apple’s

23

wrongful conduct affected all Class members in the same way.” Mot. at 25. Underscoring the

24

point, Plaintiffs do not specify the injunctive relief they seek in their complaint or their motion for

25

class certification—both filings refer instead to a boilerplate request for “appropriate injunctive

26

relief” or “injunctive relief to remedy Apple’s continuing wrongful conduct.” Compl. at 35; Mot.

27
28

43
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

at 25. Similarly, over 14 months after filing this lawsuit, Plaintiffs refused on April 16, 2018, to

2

respond to Apple’s interrogatory which requested that Plaintiffs “[i]dentify precisely the injunctive

3

relief You seek in this case.” See ECF No. 199-5 at 1–2. Plaintiffs’ first attempt to define the

4

injunctive relief at issue is in the reply to their motion for class certification, which has deprived

5

Apple of any chance to respond to their request. Reply at 15 (suggesting the Court should require

6

Apple to “(i) disclos[e] when certificates are programmed to expire and what features will be

7

affected; and (ii) allow iPhone devices to install earlier iOS versions”). The Court therefore finds

8

that Plaintiffs’ argument as to the form of injunctive relief has been waived. Bazuaye v. INS, 79

9

F.3d 118, 120 (9th Cir. 1996) (“Issues raised for the first time in the reply brief are waived.”);
Zamani v. Carnes, 491 F.3d 990, 997 (9th Cir. 2007) (“The district court need not consider

11
United States District Court
Northern District of California

10

arguments raised for the first time in a reply brief.”); John-Charles v. California, 646 F.3d 1243,

12

1247 n.4 (9th Cir. 2011) (holding party “failed to develop any argument on this front, and thus has

13

waived it”); see Diacakis v. Comcast Corp., 2013 WL 1878921, at *9 (N.D. Cal. May 3, 2013)

14

(finding injunctive relief request in motion for class certification “not well taken” when

15

“request[ed] in a footnote which is devoid of any meaningful analysis”).

16
17
18

In sum, the Court denies certification of a Rule 23(b)(2) class. The Court now turns to
Plaintiffs’ motion to strike.
E. Motion to Strike

19

Plaintiffs have filed a motion to strike the report of Dr. Avi Rubin (“Dr. Rubin”), Apple’s

20

technical expert. ECF No. 200-4, First Declaration of Avi Rubin (“First Rubin Decl.”). Plaintiffs

21

object to four main aspects of Dr. Rubin’s report. First, Plaintiffs seek to strike Dr. Rubin’s

22

opinion that the FaceTime usage of Grace’s iPhone 4 is minimal. Second, Plaintiffs seek to strike

23

Dr. Rubin’s opinion that Potter is not typical of the class because the performance issues on his 32

24

GB iPhone 4 were due to Potter “jailbreaking” the 32 GB iPhone. Third, Plaintiffs seek to strike

25

Dr. Rubin’s opinion that Potter’s (un-jailbroken) 16 GB iPhone may not be typical because it was

26

subject to a factory reset. Fourth, Plaintiffs seek to strike Dr. Rubin’s testimony that Apple did not

27
28

44
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1
2

intentionally cause the FaceTime Break.
The Court rejects Plaintiffs’ attempt to strike Dr. Rubin’s analysis of Grace’s FaceTime

3

usage. Plaintiffs’ main argument in support of striking this analysis is that Dr. Rubin initially

4

stated that Grace did not use FaceTime, but later amended his report to state Grace’s iPhone 4

5

used FaceTime at least once. ECF No. 223-4 at 1. However, the neutral evaluator who examined

6

Grace’s iPhone 4 initially represented that it did not have any FaceTime calls logged, before later

7

providing an output file indicating at least one FaceTime use occurred. ECF No. 223-4 at 2. Dr.

8

Rubin’s amendment may therefore simply reflect this change. Plaintiffs can certainly cross-

9

examine Dr. Rubin on the change at trial, but the change speaks to the weight of Dr. Rubin’s
opinions rather than their admissibility. As this Court has noted, “[e]xpert testimony is admissible

11
United States District Court
Northern District of California

10

pursuant to Rule 702 if it is both relevant and reliable.” Hadley v. Kellogg Sales Co., 2018 WL

12

3954587, at *14 (N.D. Cal. Aug. 17, 2018) (citing Daubert v. Merrell Dow Pharm., Inc., 509 U.S.

13

579, 589 (1993)). Dr. Rubin’s opinion that Grace used FaceTime infrequently, if at all, is highly

14

relevant to the underlying question of damages. Second Rubin Decl. at 16–17. The minimal

15

FaceTime usage on Grace’s iPhone 4 suggests the feature was not an important part of the iPhone

16

4’s value, thereby undermining Plaintiffs’ theory that the FaceTime Break diminished affected

17

iPhones’ value by 12 percent. Dr. Rubin’s opinion is also reliable given that he is a Professor of

18

Computer Science at Johns Hopkins University with an extensive background in information

19

security issues and has explained the bases for his conclusions in his report. Id. at 1–2, 16–17.

20

The Court also rejects Plaintiffs’ attempt to strike Dr. Rubin’s opinion that Potter’s 32 GB

21

iPhone 4 cannot support typicality because performance problems may be attributable to

22

jailbreaking. As explained above, the Court agrees with Apple that Potter’s clams based on his

23

jailbroken 32 GB iPhone 4 are not typical of the class. Plaintiffs’ theory of the case is that the

24

FaceTime Break harmed class members were harmed by forcing them to choose between (1)

25

keeping iOS 6 and losing FaceTime, or (2) upgrading to iOS 7 and keeping FaceTime, but

26

experiencing performance problems. Dr. Rubin’s opinion is therefore relevant because it means a

27
28

45
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

jailbroken iPhone like Potter’s 32 GB iPhone 4 gives rise to the unique defense that performance

2

issues are not, in fact, attributable to the FaceTime Break. See Hanon, 976 F.2d at 508 (“[C]lass

3

certification is inappropriate where a putative class representative is subject to unique defenses

4

which threaten to become the focus of the litigation.”). Dr. Rubin’s opinion is also reliable because

5

examining the potentially deleterious effects of jailbreaking on iPhone performance is based on his

6

expertise in the information security area and he sets forth the bases for his opinion in his report.

7

Second Rubin Decl. at 18–24.

8
9

Finally, the Court declines to decide Plaintiffs’ remaining two contentions because they are
not relevant at the class certification phase. The factory reset of Potter’s 16 GB iPhone 4 has no
bearing on how much the FaceTime Break reduced the device’s market value, and that is the focal

11
United States District Court
Northern District of California

10

point of Plaintiffs’ case. Similarly, whether or not Apple intended to cause the FaceTime Break

12

may be relevant to the ultimate liability determination but does not speak to whether or not a class

13

can be certified at this juncture.

14
15
16

The Court therefore DENIES Plaintiffs’ motion to strike.
F. Evidentiary Objections
Apple has filed objections to evidence submitted by Plaintiffs in connection with

17

Plaintiffs’ reply in support of class certification. ECF No. 235. The Court DENIES Apple’s

18

objections because the Court would reach the same outcome with or without the material to which

19

Apple objects. The only objected-to evidence referenced in the instant order is Dr. Hastings’s

20

statement that “all major platforms benchmark their phone offers against wholesale market

21

prices.” Hastings Reply Report at 19. Even if the Court’s decision depended on this evidence—

22

and it does not—the Court would find that considering the evidence was appropriate because

23

Plaintiffs raise it in direct response to Defendants’ opposition brief. Opp. at 16 (“Dr. Hastings also

24

relies on a misleading data set, while ignoring more relevant data regarding the value of class

25

members’ phones. She does not evaluate the amount class members could have received for their

26

devices either before or after April 16, 2014.” (internal citation omitted)); Defs. of Wildlife v. U.S.

27
28

46
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE

1

Fish, 2016 WL 4382604, at *10 (N.D. Cal. Aug. 17, 2016) (court may consider reply evidence

2

“offered in direct response” to issues raised in opposition); see Terrell v. Contra Costa Cty., 232

3

F. App'x 626, 628–29 & n.2 (9th Cir. 2007) (evidence responding to recitation of facts in

4

opposition is not “new” and may be considered on reply).

5

IV.

CONCLUSION
For the foregoing reasons, the Court GRANTS in part and DENIES in part Plaintiffs’

6
7

motion for class certification. The Court CERTIFIES a class seeking restitution and damages for

8

violation of the UCL and trespass to chattels pursuant to Rule 23(b)(3):

9

All owners of non-jailbroken Apple iPhone 4 or Apple iPhone 4S devices in
California who on April 16, 2014, had iOS 6 or earlier operating systems on their
iPhone 4 or iPhone 4S devices.

10

United States District Court
Northern District of California

11

The Court APPOINTS Ken Potter, only as to his 16 GB iPhone 4, and Christina Grace as

12

representatives of the class. As Apple does not challenge the adequacy of proposed class counsel,

13

the Court also APPOINTS as class counsel Jill M. Manning of Steyer Lowenthal Boodrookas

14

Alvarez & Smith LLP, Daniel L. Warshaw of Pearson, Simon & Warshaw LLP, John Austin

15

Curry of Caldwell Cassady & Curry LLP, and David F.E. Tejtel of Friedman Oster & Tejtel

16

PLLC.

17

IT IS SO ORDERED.

18

Dated: September 18, 2018

19
20

______________________________________
LUCY H. KOH
United States District Judge

21
22
23
24
25
26
27
28

47
Case No. 17-CV-00551-LHK
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION; DENYING
MOTION TO STRIKE