Maldonado et al v. Apple, Inc et al

Filing 155

ORDER CERTIFYING CLASS AND DENYING DEFENDANTS' MOTION FOR SUMMARY JUDGMENT by Judge William H. Orrick. Case Management Statement due by 11/26/2019. Further Case Management Conference set for 12/3/2019 02:00 PM in San Francisco, Courtroom 02, 17th Floor. (jmdS, COURT STAFF) (Filed on 9/17/2019)

1
2
3
4

UNITED STATES DISTRICT COURT

5

NORTHERN DISTRICT OF CALIFORNIA

6
VICKY MALDONADO, et al.,

7

Plaintiffs,

8

ORDER CERTIFYING CLASS AND
DENYING DEFENDANTS’ MOTION
FOR SUMMARY JUDGMENT

v.

9

APPLE, INC, et al.,

10

Re: Dkt. Nos. 99, 100, 102, 103, 110, 111,

Defendants.

11
United States District Court
Northern District of California

Case No. 3:16-cv-04067-WHO

112, 121, 127, 128, 130, 132, 136, 138, 144,

12

147

13
14

According to plaintiffs Vicky Maldonado and Justin Carter, defendants Apple, Inc.,

15

AppleCare Service Company Inc., and Apple CSC Inc. (collectively, “Apple”) breach the

16

AppleCare and AppleCare+ agreements every time a consumer receives a remanufactured

17

replacement device because those devices are not “equivalent to new in performance and

18

reliability” as promised under the contract. Instead, the presence of non-new parts means

19

remanufactured devices can never be as reliable as new ones. Plaintiffs move for class

20

certification to pursue their claims against Apple, which opposes on predominance and other

21

grounds and further contends that it is entitled to summary judgment on Maldonado’s and Carter’s

22

claims. For the reasons set forth below, I will deny Apple’s motion for summary judgment and

23

grant plaintiffs’ motion for class certification.

24
25

BACKGROUND
I.

FACTUAL BACKGROUND

26

A.

27

AppleCare and AppleCare+ (“AC/AC+”) plans provide extended warranty and technical

28

AppleCare, AppleCare+, and the Remanufacturing Process

support for Apple consumers who wish for more than the standard one-year hardware warranty

1

request for an order relating it to the earlier filed case before me, English v. Apple, 14-cv-1619.

2

Dkt. Nos. 1, 21. On March 2, 2017, I granted in part and denied in part Apple’s motion to

3

dismiss.12 Order on MTD [Dkt. No. 64]. After a few continuations of the case schedule, plaintiffs

4

filed a motion for class certification on February 28, 2019.13 Motion for Class Certification

5

(“Cert. Mot.”) [Dkt. No. 102-4]. On March 29, 2019, pursuant to the parties’ stipulation, I

6

consolidated the hearings for the class certification motion and Apple’s forthcoming motion for

7

summary judgment. Dkt. No. 109. On April 8, Apple moved for summary judgment. Motion for

8

Summary Judgment (“MSJ”) [Dkt. No. 110-4]. On June 10, plaintiffs filed a conditional motion

9

for additional discovery under Federal Rule of Civil procedure 56(d). Dkt. No. 132. I heard

10

argument on all the motions on August 7, 2019. Dkt. No. 149.

United States District Court
Northern District of California

11
12

LEGAL STANDARD
I.

SUMMARY JUDGMENT
Summary judgment on a claim or defense is appropriate “if the movant shows that there is

13
14

no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of

15

law.” Fed. R. Civ. P. 56(a). In order to prevail, a party moving for summary judgment must show

16

the absence of a genuine issue of material fact with respect to an essential element of the non-

17

moving party’s claim, or to a defense on which the non-moving party will bear the burden of

18

persuasion at trial. See Celotex Corp. v. Catrett, 477 U.S. 317, 323 (1986). Once the movant has

19

made this showing, the burden then shifts to the party opposing summary judgment to identify

20

“specific facts showing there is a genuine issue for trial.” Id. The party opposing summary

21

judgment must present affirmative evidence from which a jury could return a verdict in that

22

party’s favor. Anderson v. Liberty Lobby, 477 U.S. 242, 257 (1986).
On summary judgment, the court draws all reasonable factual inferences in favor of the

23
24

non-movant. Id. at 255. In deciding the motion, “[c]redibility determinations, the weighing of the

25
26

12

27

13

28

I gave plaintiffs 20 days to amend their complaint, but they declined to do so.

Plaintiffs originally filed on February 25 but amended their motion on February 28. See Dkt.
Nos. 99, 100, 102, 103. The original motions at Dkt. Nos. 99 and 100 are TERMINATED AS
MOOT.
7

1

evidence, and the drawing of legitimate inferences from the facts are jury functions, not those of a

2

judge.” Id. However, conclusory and speculative testimony does not raise genuine issues of fact

3

and is insufficient to defeat summary judgment. See Thornhill Publ’g Co., Inc. v. GTE Corp., 594

4

F.2d 730, 738 (9th Cir. 1979).

5

II.

6

CLASS CERTIFICATION
“Before certifying a class, the trial court must conduct a rigorous analysis to determine

7

whether the party seeking certification has met the prerequisites of Rule 23.” Mazza v. Am. Honda

8

Motor Co., Inc., 666 F.3d 581, 588 (9th Cir. 2012) (internal quotation marks omitted). The party

9

seeking certification has the burden to show, by a preponderance of the evidence, that certain
prerequisites have been met. See Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 348-50 (2011);

11
United States District Court
Northern District of California

10

Conn. Ret. Plans & Trust Funds v. Amgen Inc., 660 F.3d 1170, 1175 (9th Cir. 2011).

12

Certification under Rule 23 is a two-step process. The party seeking certification must first

13

satisfy the four threshold requirements of Rule 23(a). Specifically, Rule 23(a) requires a showing

14

that: (1) the class is so numerous that joinder of all members is impracticable; (2) there are

15

questions of law or fact common to the class; (3) the claims or defenses of the representative

16

parties are typical of the claims or defenses of the class; and (4) the representative parties will

17

fairly and adequately protect the interests of the class. Fed. R. Civ. P. 23(a).

18

Next the party seeking certification must establish that one of the three grounds for

19

certification applies. See Fed. R. Civ. P. 23(b). Plaintiffs seek certification under Rule (b)(3),

20

which requires them to establish that “the questions of law or fact common to class members

21

predominate over any questions affecting only individual members, and that a class action is

22

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

23

R. Civ. P. 23(b)(3). They also seek certification under Rule 23(b)(2) for injunctive relief.

24

In the process of class-certification analysis, there “may entail some overlap with the

25

merits of the plaintiff’s underlying claim.” Amgen Inc. v. Connecticut Ret. Plans & Trust Funds,

26

568 U.S. 455, 465–66 (2013) (internal quotation marks omitted). However, “Rule 23 grants courts

27

no license to engage in free-ranging merits inquiries at the certification stage.” Id. at 466. “Merits

28

questions may be considered to the extent—but only to the extent—that they are relevant to
8

1

determining whether the Rule 23 prerequisites for class certification are satisfied.” Id.

2
3

DISCUSSION
I.

SUMMARY JUDGMENT
Apple moves for summary judgment on Maldonado’s and Carter’s breach of contract

4
5

claims and asserts that the remaining claims fall for the same reasons. According to Apple,

6

plaintiffs lack evidence to prove their essential elements of their breach of contract claims, and

7

Carter’s conduct should prevent him from pursuing his claims.
A. Plaintiffs’ Theory of Liability and Damages

9

The motion for summary judgment is based on three main arguments: (i) plaintiffs

10

contend that they were entitled to new devices, in clear contradiction of the contract language; (ii)

11
United States District Court
Northern District of California

8

plaintiffs lack evidence showing that their remanufactured devices were not equivalent to new;

12

(iii) plaintiffs cannot show that Apple’s alleged breach caused their damages because the

13

malfunctions they complained of do not relate to non-new part(s) in their devices. MSJ 8-16.

14

These arguments are based on a misunderstanding of plaintiffs’ theory of liability; none succeeds.
According to Apple, plaintiffs’ position boils down to an assertion that they were entitled

15
16

to new devices rather than the remanufactured devices they received. MSJ 8-9 (contending that in

17

their deposition testimony, plaintiffs “appear ultimately to take the position that ‘equivalent to new

18

in performance and reliability’ means ‘new’”). This theory contradicts the AC/AC+ language,

19

which unambiguously states that consumers may receive one of two types of replacement devices

20

under the contract: new or “equivalent to new.” MSJ 8-9.
I agree with Apple that given the language of the contract, equivalent-to-new devices

21
22

cannot be the same as new devices. But plaintiffs’ theory does not amount to a contention that

23

they were entitled to new devices. Their case rests on their ability to prove that remanufactured

24

devices are not “equivalent to new.” See Carter Depo. 108:11-18 (testifying that he understood

25

“equivalent to new” as meaning the phone would “operate exactly like [his] new phone did”).14 If

26
27
28

14

Apple uses the following definitions of these “accepted engineering terms”: “[T]o be equivalent
to new in ‘performance’ means that remanufactured devices meet the same engineering
specifications as new devices, and to be equivalent to new in ‘reliability’ means that the
remanufactured devices satisfy the same reliability test suites as new devices.” MSJ 9. But as
9

1

plaintiffs can prove this theory, consumers who received such devices did not receive the benefit

2

of their bargain.

3

Apple also challenges the evidence plaintiffs rely on to prove their theory: plaintiffs’

4

interpretation of the contract is “unrealistic and unsupportable” because the expert opinion of

5

Michael Pecht—that any device with a non-new component cannot be equivalent to new—“reads

6

‘equivalent to new’ out of the AC+ contract.” Reply MSJ 3; see infra Section II.B – Plaintiffs’

7

Classwide Theories (discussing Pecht’s opinions in more detail). I disagree. Pecht’s report sets

8

forth reasons why remanufactured devices do not meet that mark; it does not read “equivalent to

9

new” out of the contract. Apple’s performance must match its promise, and a reasonable fact

10

United States District Court
Northern District of California

11

finder could rely on this evidence to conclude that it does not.
Apple next contends that it is entitled to summary judgment because plaintiffs lack

12

evidence showing that their specific devices were not equivalent to new in performance and

13

reliability. MSJ 10-14. Instead, the evidence shows that remanufactured devices go through the

14

same manufacturing and testing process as new iPhones and iPads. Id. at 11-12. Apple presents

15

evidence of its remanufacturing and testing processes that could lead a fact finder to conclude that

16

the resulting remanufactured devices are equivalent to new. But a fact finder could also credit the

17

reports of Pecht and Robert Bardwell and conclude that remanufactured devices—including

18

Maldonado’s and Carter’s—are inferior. Plaintiffs’ theory of breach does not depend on the

19

nature of any individual device. They assert that load conditions prevent all devices with non-new

20

parts from being considered “equivalent to new.” See infra Section II.B.1 – Plaintiffs’ Classwide

21

Theories (discussing the Pecht and Bardwell reports in more detail).

22

Finally, Apple claims that plaintiffs cannot show that a non-new part caused the problems

23

they allegedly experienced, and thus there is no evidence to tie Apple’s alleged breach with

24

plaintiffs’ alleged damages. MSJ 14-16. For the reasons discussed above, plaintiffs’ success does

25

not depend on the functioning or malfunctioning of individual devices. Oppo. MSJ 17. Apple

26

promised plaintiffs equivalent-to-new devices under the AC+ contract. Plaintiffs assert that when

27
28

plaintiffs point out, given that AC/AC+ is a consumer contract, it is appropriate to construe the
terms in the way a consumer would understand them. See Oppo. MSJ 13-14
10

1

they submitted claims under the contract, instead of receiving the benefit of their bargain they

2

received inferior devices that were more likely to fail and have shorter lifespans. If a fact finder

3

credits this theory, then Apple breached—and caused plaintiffs’ damages—at the time of that

4

exchange.
Apple can challenge plaintiffs’ evidence at trial, but material disputes of fact preclude

5
6

summary judgment.

7

B.

8

Apple argues that Carter should not be permitted to pursue his claims because he engaged

9

Carter’s Conduct

in improper conduct that prejudiced it. MSJ 16-19. Someone opened and inspected Carter’s
replacement iPhones without Apple’s expert present, and Carter returned two of his replacement

11
United States District Court
Northern District of California

10

devices rather than preserving them for this litigation. As a result, Apple is unable to test the

12

phones or otherwise rely on that evidence to disprove Carter’s claims. Finally, Carter improperly

13

obtained his second and third replacements for purposes of this litigation, meaning he effectively

14

manufactured aspects of his claims.

15

I disagree with Apple that Carter’s conduct constitutes spoliation or that it will prejudice

16

Apple. As articulated above and discussed in more detail below, plaintiffs’ theory is not tied to

17

any specific remanufactured device, and detailed inspections are not necessary to defend against

18

Carter’s claims. Accordingly, Apple’s inability to test all of the replacement devices Carter

19

received will not cause it prejudice. Finally, Carter testified that he was experiencing problems

20

with his replacement devices separate and apart from his interactions with counsel. See Carter

21

Depo. 16:16-23, 150:19-151:6. Carter’s conduct does not merit summary judgment in favor of

22

Apple.

23

Apple’s motion for summary judgment on the remaining claims fails for the same reasons

24

as the breach of contract claim. See MSJ 19-22 (setting forth reasons why the remaining claims

25

fall with the breach of contract claims). For all of these reasons, Apple’s motion for summary

26

judgment is DENIED. Plaintiffs’ conditional motion for additional discovery under Federal Rule

27

of Civil Procedure 56(d) and the motions to seal related to the parties’ briefing on that motion are

28
11

1

DENIED AS MOOT. See Dkt. Nos. 138,15 144, 147.

2

II.

CLASS CERTIFICATION
Plaintiffs seek certification of the following class: “All individuals who purchased

3
4

AppleCare or AppleCare+, either directly or through the iPhone Upgrade Program, on or after

5

January 1, 2009, and received a remanufactured replacement Device.” Apple challenges

6

plaintiffs’ Rule 23 showing on several grounds: (i) the class is overbroad in terms of members and

7

time period; (ii) plaintiffs can establish neither commonality nor predominance; (iii) named

8

plaintiffs’ experiences are not typical; and (iv) named plaintiffs are not adequate class

9

representatives.
As an initial matter, I agree with Apple that plaintiffs have failed to justify a class period

11
United States District Court
Northern District of California

10

extending back to January 1, 2009 rather than to July 20, 2012, four years before they filed suit.

12

Oppo. Cert. Mot. 10; see Cal. Civ. Proc. Code § 337 (providing four years within which a contract

13

action can be brought). The parties discussed the class period in September and November 2017,

14

at which time plaintiffs told Apple they would support their tolling theory during class

15

certification briefing. Patel Decl. ¶ 2. Plaintiffs failed to support their proposed class period in

16

their motion or reply, despite the fact that Apple squarely addressed this issue in its opposition.16

17

See Oppo. Cert. Mot. 10. Plaintiffs have failed to meet their burden to show that it is appropriate

18

to extend the class period; accordingly, the class period will begin on July 20, 2012, four years

19

before this case was filed.

20

A. Rule 23(a)

21

1. Numerosity
Rule 23(a)(1) requires that the “the class [be] so numerous that joinder of all members is

22
23

impracticable.” Fed. R. Civ. P. 23(a)(1). The party seeking certification “do[es] not need to state

24

the exact number of potential class members, nor is a specific number of class members required

25
26
27
28

15

The motion at Dkt. No. 138 is denied only insofar as it related to the 56(d) motion rather than
the motion for summary judgment.
16

At the hearing plaintiffs asserted that the class period can extend to 2009 because claims do not
accrue until a party has reason to know of them. Hearing Transcript [Dkt. No.153] 14:7-23. This
untimely argument is not sufficient.
12

1

for numerosity.” In re Rubber Chemicals Antitrust Litig., 232 F.R.D. 346, 350 (N.D. Cal. 2005).

2

Courts generally find that numerosity is satisfied if the class includes forty or more members. See

3

Villalpando v. Exel Direct Inc., 303 F.R.D. 588, 605-06 (N.D. Cal. 2014); In re Facebook, Inc.,

4

PPC Adver. Litig., 282 F.R.D. 446, 452 (N.D. Cal. 2012).

5

Plaintiffs assert that their class is sufficiently numerous because Apple’s records show that

6

it sold over three million AC/AC+ plans where it provided at least one replacement device, many

7

of which were remanufactured. Cert. Mot. 16. Even with a class period beginning in 2012 rather

8

than 2009, this showing is sufficient to satisfy the numerosity requirement.

9
10

2. Commonality
Rule 23 requires that there be questions of law or fact common to the class. Fed. R. Civ. P.

United States District Court
Northern District of California

11

23(a)(2). Plaintiffs must show that the class members have suffered “the same injury,” meaning

12

their claims “depend upon a common contention” that is of such a nature that “determination of its

13

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

14

Dukes, 564 U.S. at 350 (internal quotation marks and citation omitted). Plaintiffs must

15

demonstrate not merely the existence of a common question, but rather “the capacity of classwide

16

proceedings to generate common answers apt to drive the resolution of the litigation.” Id. (internal

17

quotation marks and emphasis omitted). For purposes of Rule 23(a)(2), “even a single common

18

question will do.” Id. at 359 (internal quotation marks and modifications omitted).

19

Plaintiffs argue that the common question for all their claims is whether Apple’s

20

remanufactured devices are equivalent to new in performance and reliability. Cert. Mot. 17. For

21

the breach of contract claim, the common questions are: (i) “whether remanufactured devices’

22

higher rate of failure or shorter life span establishes those devices are not equivalent to new”; (ii)

23

whether Apple must employ comparison testing to assess whether remanufactured devices are

24

equivalent to new”; and (iii) “whether passing Apple’s uniform minimum test standards proves

25

that remanufactured devices are equivalent to new.” Id. For the warranty and UCL claims, the

26

common questions are: (i) “whether Apple’s promise to provide equivalent to new devices was a

27

misrepresentation”; (ii) “whether a reasonable consumer would have been deceived by Apple’s

28

misrepresentations”; and (iii) “whether Apple’s conduct constitutes an unfair or unlawful business
13

1
2

practice.” Id.
Apple argues that plaintiffs fail to show common questions because neither breach nor

3

causation nor injury can be adjudicated on a classwide basis. As discussed below for purposes of

4

the predominance inquiry, plaintiffs present evidence that, if credited, could establish

5

remanufactured devices are not equivalent to new in performance and reliability. Plaintiffs have

6

established common questions.

7
8

3. Typicality
“The test of typicality is whether other members have the same or similar injury, whether
the action is based on conduct which is not unique to the named plaintiffs, and whether other class

10

members have been injured by the same course of conduct.” Ellis v. Costco Wholesale Corp., 657

11
United States District Court
Northern District of California

9

F.3d 970, 984 (9th Cir. 2011) (internal quotation marks and citation omitted). Class certification

12

is not appropriate if unique defenses threaten to preoccupy the class representatives and thus cause

13

absent members to suffer. Hanon v. Dataproducts Corp., 976 F.2d 497, 508 (9th Cir. 1992). But,

14

“the defense of non-reliance is not a basis for denial of class certification.” Id. at 509.

15

Plaintiffs assert that they can satisfy the typicality requirement because the named

16

plaintiffs purchased AC/AC+ with the same promise from Apple and received remanufactured

17

devices that they assert were not equivalent to new. Cert. Mot. 18. Apple argues that named

18

plaintiffs’ claims are not typical of the class claims because they have not tied any problems they

19

had with their remanufactured devices to any non-new parts. Cert. Oppo. 24. Specifically, the

20

issues they experienced were likely related to software, usage, or geography rather than the non-

21

new parts. Id. Plaintiffs counter that their claims rest on their evidence that all remanufactured

22

devices are inferior to new devices. Reply ISO Class Certification (“Cert. Reply”) [Dkt. No. 121-

23

2] 14-15. Because the named plaintiffs received remanufactured devices with non-new parts, their

24

claims are typical. Id.

25

As I discuss below, plaintiffs present classwide evidence that remanufactured devices are

26

not equivalent to new in performance and reliability. Both Carter and Maldonado purchased AC+

27

plans, sought replacement devices, and received remanufactured devices. Accordingly, their

28

claims are typical of the class’s claims.
14

1

4. Adequacy
Finally, to establish adequacy under Rule 23(a)(4), named plaintiffs must show that they

2
3

“will fairly and adequately protect the interests of the class.” Fed. R. Civ. P. 23(a)(4). “To

4

determine whether named plaintiffs will adequately represent a class, courts must resolve two

5

questions: (1) do the named plaintiffs and their counsel have any conflicts of interest with other

6

class members and (2) will the named plaintiffs and their counsel prosecute the action vigorously

7

on behalf of the class?” Ellis, 657 F.3d at 985 (internal quotation marks omitted).
Apple argues that Carter is an inadequate class representative because he improperly

8

sought replacement devices for the purposes of this litigation, and counsel from Hagens Berman

10

Sobol Shapiro LLP is inadequate because they were involved in that conduct. Cert. Oppo. 24-25.

11
United States District Court
Northern District of California

9

Plaintiffs counter that Carter was already experiencing problems with a remanufactured device

12

before he contacted counsel about this case. Reply 15. Neither he nor counsel had reason to seek

13

replacements for the purposes of litigation because Carter had already received a remanufactured

14

device under AC+, which was enough to make him an appropriate plaintiff in this case.17 Id.

15

Carter’s conduct does not rise to the level of making him an inadequate class

16

representative. In addition, Hagens Berman has extensive experiencing litigating consumer

17

protection class actions. See Berman Decl. Ex. 15 [Dkt. No. 103-15] (firm resume). Plaintiffs

18

have made a sufficient showing of adequacy under Rule 23(a)(4).

19

B.

20

To proceed under Rule 23(b)(3) for damages, plaintiffs must show that it is superior to

Rule 23(b)(3)

21

proceed as a class action and “the questions of law or fact common to class members predominate

22

over any questions affecting only individual members.” Fed. R. Civ. P. 23(b)(3). The

23

predominance inquiry “tests whether proposed classes are sufficiently cohesive to warrant

24

adjudication by representation.” Amchem Prods, Inc. v. Windsor, 521 U.S. 591, 623 (1997). “The

25

focus is on the relationship between the common and individual issues.” Stearns v. Ticketmaster

26

Corp., 655 F.3d 1013, 1019 (9th Cir. 2011) (internal quotations and citation omitted).

27
28

17

Both parties agree that Carter’s third replacement, which Apple alleges he misleadingly
requested from Apple, is not part of this case. See Oppo. MSJ 20; Oppo. Cert. Mot. 13 n.13.
15

1

Predominance is established if “common questions present a significant aspect of the case and

2

they can be resolved for all members of the class in a single adjudication.” Mazza, 666 F.3d at

3

589. Commonality and predominance are related issues, and there is often substantial overlap

4

between the two tests, but the test for predominance is “far more demanding.” Amchein Prods.,

5

521 U.S. at 623-24.

6

Plaintiffs argue that common questions predominate over individual questions for each of

7

their causes of action because they can establish on a classwide basis that remanufactured devices

8

are not “equivalent to new in performance and reliability.” Cert. Mot. 20-24. With the expert

9

report of Lance Kaufman, plaintiffs also put forth a methodology to measure damages on a
classswide basis. Apple challenges plaintiffs’ classwide showings on several grounds, detailed

11
United States District Court
Northern District of California

10

below.

12

1.

Plaintiffs’ Classwide Theories

Plaintiffs rely on the expert report of Michael Pecht to assert that remanufactured devices

13
14

can never be equivalent to new because they have been subject to “load conditions.” Expert

15

Report of Michael Pecht (“Pecht Rpt.”), Pecht Decl. Ex. A [Dkt. No. 102-21]. Pecht has expertise

16

and experience in the fields of engineering and electronics testing and reliability. See id. at 3-4.

17

His report addresses whether remanufactured iPhones and iPads can be equivalent to new in terms

18

of performance and reliability and whether Apple’s testing procedures are sufficient to allow

19

Apple to make this representation to purchasers of AC/AC+. Id. at 5-6.
According to Pecht, “Electronic parts and products (device, equipment) are known to wear-

20
21

out with time, usage (operational) conditions, and environmental conditions.” Id. at 10. Such

22

degradation begins at the time the parts are made. Id. Both environmental conditions and

23

operation, which Pecht calls “load (stress) conditions,”18 can contribute to the degradation. Id.

24

These load conditions may not be visible, may not be detected during testing, and may not impact

25

the device’s performance right away, but they “use up (degrade) the life of a device[].” Id. at 10-

26
27
28

18

These conditions include “thermal ranges and changes, mechanical loads / stresses (including
handling and operation such as pushing buttons), humidity and moisture, vibration, shock, dust,
smoke and other contaminates, and even radiation.” Pecht Rpt. 10.
16

1

11. Pecht concludes that “[d]evices containing salvaged (used) components can never be as

2

reliable as devices containing new components” and that “Apple’s testing is insufficient for Apple

3

to represent that remanufactured devices are equivalent to new devices in reliability.” Id. at 13.

4

Plaintiffs’ theory is that the replacement devices are necessarily less reliable than new devices by

5

virtue of having non-new parts. If a jury were to credit this theory, as supported by Pecht’s

6

analysis, it could determine the question of equivalence for the entire class.

7

Plaintiffs also rely on the expert report of Robert Bardwell to assert that remanufactured

8

devices fail at a higher rate than new devices do. Expert Report of Robert Bardwell (“Bardwell

9

Rpt.”), Bardwell Decl. Ex. A [Dkt. No. 102-23]. Bardwell has experience and expertise in
statistical analysis and probability modeling. Id. at 21. He used the Mantel-Haenszel method to

11
United States District Court
Northern District of California

10

test whether manufactured devices have a higher failure rate than new replacement devices, as

12

measured by the rate at which the devices are returned. Id. at 7-9. He excluded the iPhone 5 from

13

his analysis because of “the abnormal number of failures in new phones.” Id. at 8. He estimates

14

that “remanufactured iPhones have over 2.3 times the odds of being returned than new

15

replacement iPhones” and “remanufactured iPads have over 1.7 times the odds of being returned

16

as new replacement iPads.” Id. at 9. Each of these figures is “statistically significant at an

17

extremely high level.” Id. He lists limitations on his analysis that suggest the failure rate is

18

actually higher than the estimates he reaches. Id. at 12.

19
20
21

2.

Apple’s Challenges
a.

Standing and Overbreadth

Apple argues the proposed class is overbroad because it includes individuals who never

22

experienced problems with their replacement devices. Cert. Oppo. 8-10. These individuals were

23

never injured and cannot establish Article III standing. Id. at 9. Plaintiffs counter that their class

24

is not overbroad because it is limited to “those who suffered a common injury—receipt of a

25

deficient remanufactured device under AC/AC+.” Cert. Reply 1.

26

There is a distinction between injury as a jurisdictional problem and injury as a Rule 23

27
28
17

1

problem.19 See Moore v. Apple Inc., 309 F.R.D. 532, 542 (N.D. Cal. 2015). In Moore, the Hon.

2

Lucy Koh declined to certify a class because it was overbroad under Mazza and Rule 23. Id. at

3

542. Plaintiffs sought to bring claims on behalf of individuals who had switched from using an

4

Apple iPhone to a non-Apple cell phone and subsequently did not receive all text messages sent to

5

them from Apple devices. Id. at 535-36, 538. Apple argued that the class included three groups

6

of individuals who were not injured: “(1) persons who experienced no disruption in their text

7

message services; (2) persons who failed to receive text messages because of technical issues

8

unrelated to the iMessages system; and (3) persons who failed to receive text messages because of

9

restrictions in their wireless contracts.” Id. at 542.
Judge Koh agreed with Apple that the class was overbroad because it “include[d]

10

United States District Court
Northern District of California

11

individuals who, by definition, could not have been injured by Defendant’s alleged wrongful

12

conduct.”20 Id. at 542 (emphasis in original). This was so because some members of the proposed

13

class had wireless contracts that did not in fact allow them to receive text messages. Id. Judge

14

Koh distinguished between this group—who, because they had no contractual right to receive text

15

messages, could not have been injured by non-receipt of text messages—and “proposed class

16

members who, by happenstance, may not have experienced disruption of text message services

17

due to Defendant’s alleged wrongful conduct.”21 Id.; see also Patel v. Trans Union, LLC, No. 14-

18

CV-00522-LB, 2016 WL 6143191, at *8 (N.D. Cal. Oct. 21, 2016) (distinguishing between

19

individuals who “by definition, [could not] be among those who may be entitled to recovery” and

20

“absent plaintiffs may ultimately fail to prove liability”); O’Shea v. Epson Am., Inc., No. CV 09-

21

8063 PSG CWX, 2011 WL 4352458, at *11 (C.D. Cal. Sept. 19, 2011), aff’d sub nom. Rogers v.

22

Epson Am., Inc., 648 F. App’x 717 (9th Cir. 2016) (noting that the class was defined to include

23
24
25
26
27
28

19

I already concluded that named plaintiffs have standing sufficient for jurisdiction. Order on
MTD 6-7.
20

Judge Koh concluded certification was inappropriate on a different, but related basis—that
individualized inquiries would be required to determine whether individuals were actually injured.
Moore, 309 F.R.D. at 544-46.
21

As discussed below, Judge Koh concluded that individualized inquiries would be required to
determine whether each class member actually experienced disruption in messaging and whether
iMessage had caused that disruption. Moore, 309 F.R.D. at 542-43; 545-46.
18

1

consumers who were never actually exposed to the allegedly deceptive representation).
The proposed class is defined in a way that avoids the overbreadth issues identified in

2
3

Moore, Patel, and Epson. Rather than including consumers who by definition could not have been

4

injured, this class includes only individuals who received replacement devices. Plaintiffs rightly

5

do not seek to include individuals who received new phones. Just as individuals who were never

6

entitled to receive messages could not be injured by not receiving messages, individuals who

7

received new phones could not have been injured even if a jury finds that Apple provides

8

remanufactured phones that are not equivalent to new. Because plaintiffs’ class avoids Article III

9

overbreadth, Apple’s arguments go to predominance under Rule 23. See Moore, 309 F.R.D. at

10

543.

United States District Court
Northern District of California

11

Apple’s predominance argument fails because it is based on a misapprehension of

12

plaintiffs’ theory of injury. Contrary to Apple’s assertions, plaintiffs’ injury occurred when they

13

filed a claim under AC/AC+ and received a device that was not “equivalent to new in performance

14

and reliability” because of load conditions or shorter lifespan. This injury occurred regardless of

15

whether an individual experienced problem with the device.22 See Nguyen v. Nissan N. Am., Inc.,

16

932 F.3d 811, 819 (9th Cir. 2019) (“Plaintiff’s theory is that the defect was inherent in each of the

17

Class Vehicles at the time of purchase, regardless of when and if the defect manifested.”). If a fact

18

finder credits plaintiffs’ theory, then all individuals who received a remanufactured replacement

19

device were injured. Accordingly, the class can include all consumers who purchased AC/AC+

20

during the class period and received a remanufactured device pursuant to the contract.

21

b.

Individualized Inquiries into Equivalence

22

Apple argues that plaintiffs cannot show predominance because individualized inquiries

23

will be necessary to determine which parts were not new and whether the non-new part actually

24

caused the problem a particular consumer experienced (as opposed to a new part or the device’s

25
26
27
28

22

In my Order on Apple’s motion to dismiss, I stated that plaintiffs would have to “point to some
‘problem’ with their devices to support their allegations that the devices were not ‘new or
equivalent to new in performance and reliability.’” Order on MTD 7. With plaintiffs’ benefit-ofthe-bargain theory crystallized, and in reliance on cases like Nguyen, I now conclude that all
individuals who received a remanufactured device allege an injury sufficient to confer standing.
19

1

software being the cause). Cert. Oppo. 13-14. Because all of the remanufactured iPhones and

2

iPads will “vary in the number and mix of non-new parts,” classwide proof cannot establish that

3

they are not “equivalent to new in performance and reliability” as a result of the non-new part(s).

4

Id.
In Moore, Judge Koh found that a few different individualized issues predominated over

5

common issues. Moore, 309 F.R.D. at 545. First, variations in service agreements regarding text

7

messages meant that individuals’ non-receipt of text messages could have been caused not by

8

Apple but by the individual having exceeded the number of messages paid for or having blocked a

9

number. Id. Plaintiffs had no way of answering these questions on a classwide basis. Id. at 546.

10

Second, individualized inquiries would be required to establish that an individual’s non-receipt of

11
United States District Court
Northern District of California

6

messages was in fact caused by iMessage rather than the many other possible causes.23 Id. at 546-

12

47. Judge Koh rejected plaintiffs’ argument that iMessage had a “systemic flaw” that prevented

13

16

delivery of messages:
[T]he question [was] not whether Plaintiff and members of the
proposed class [would] ultimately be able to prove that iMessage
could cause disruptions in text messaging services as a general matter.
Instead, the relevant question under Rule 23 [was] whether
determining if iMessage caused a class member’s injury require[d] an
individualized inquiry such that class treatment [was] inappropriate.

17

Id. at 547. The plaintiffs’ theory of causation inappropriately relied on the assumption “the

18

iMessage system actually interfered with a class member’s ability to receive text messages.” Id. at

19

548; see also Bruce v. Teleflora, LLC, No. 2:13-CV-03279, 2013 WL 6709939, at *6 (C.D. Cal.

20

Dec. 18, 2013) (noting that “one would have to assess each individual [flower] arrangement

21

delivered to each putative class member to determine whether she received an inferior-quality

22

arrangement”).

14
15

Apple argues that the same individualized inquiries into the following will be necessary in

23
24

this case because of the unique mix of new and non-new parts in each device. Each device will

25

have to be individually analyzed to assess which parts were not new and whether the non-new

26

part(s) in fact caused the problems the consumer experienced, rather than the myriad other

27
28

23

The court noted the frequency of text message delivery failure and the numerous reasons that
could cause those failures. Moore, 309 F.R.D. at 546.
20

1

possible causes. Plaintiffs counter that they have classwide proof of non-equivalence across the

2

remanufactured devices, regardless of the performance of a particular device. Through their

3

experts, they offer evidence that no remanufactured devices are equivalent to new because (i) load

4

conditions mean that used parts can never be equivalent to new parts and (ii) remanufactured

5

devices fail at significantly higher rates than new devices. Reply 4.
Apple’s arguments do not overcome plaintiffs’ predominance showing. Its reliance on the

6
7

Teleflora case, which addressed the quality of flower bouquets, is unpersuasive. Teleflora, 2013

8

WL 6709939, at *6-7. It hardly bears mentioning that there is a difference between assessing the

9

quality of unique, handmade floral arrangements and assessing the reliability of hardware.
Plaintiffs’ benefit-of-the-bargain theory of Apple’s liability is not dependent on the analysis of a

11
United States District Court
Northern District of California

10

particular device. See Nguyen, 932 F.3d at 819 (“Plaintiff’s theory is that the defect was inherent

12

in each of the Class Vehicles at the time of purchase, regardless of when and if the defect

13

manifested.”).24 Individualized inquiries—into whether a device experience problems or whether

14

those problems were tied to a non-new part—will not be necessary to prove plaintiffs’ case. See

15

Pecht Depo. 57:6-14 (noting that his opinions are “quite broad-based and fundamental reliability

16

engineering statements” that “hold true” regardless of an individual examination of a single

17

product). Apple’s remaining challenges to the Pecht report go to its merits, not whether his

18

opinions can serve as evidence for the entire class.
Apple next argues that its testing procedures are “the standard” in the industry and that

19
20

plaintiffs appear to take the “absurd position” that Apple should test replacements that are

21

provided to consumers in the extreme manner that it tests new devices. Cert. Oppo. 18-19. Such

22

testing on remanufactured devices would be impossible because the process often destroys the

23

device. Id. at 19. Both sides appear to agree that extreme reliability testing would make it

24

impossible to give remanufactured devices to consumers. But that reality does not necessarily

25

support a finding in Apple’s favor.25 Instead, a fact finder could rely on this fact to conclude not

26
27
28

24

The Ninth Circuit published this opinion after class certification briefing had concluded in this
case, but counsel discussed it during the hearing on these motions.
25

As plaintiffs point out, Pecht does not opine that Apple should perform reliability testing on all
21

1

that plaintiffs’ position is absurd but that despite Apple’s promises, it is not capable of ensuring

2

remanufactured devices are equivalent to new. Apple can raise these challenges to Pecht, but his

3

opinions and conclusions can nevertheless serve as classwide proof of plaintiffs’ claims.

4

Apple’s criticisms of the Bardwell report also go to the merits rather than to the question of
classwide proof. Apple argues that plaintiffs cannot rely on the return rate as the rate of failure

6

because a return does not necessarily indicate that the device failed or that the non-new part

7

caused the device’s problems. Cert. Oppo. 14-15. Plaintiffs counter that Apple itself “uses return

8

rates to assess reliability and calls them ‘failure rates,’ a common approach with consumer

9

electronics.” Cert. Reply 6; see Lanigan Depo. 108:6-109:8. Apple further criticizes Bardwell’s

10

exclusion of the iPhone 5 and from his analysis and argues that that its own expert’s report shows

11
United States District Court
Northern District of California

5

that there is no difference between the return rates of remanufactured and new replacements. Cert.

12

Oppo. 15-17. All of these challenges go to the merits of Bardwell’s conclusions, not to whether

13

the report, if credited by a fact finder, could serve as classwide proof of plaintiffs’ claims.

14

c.

Individualized Inquiries into Damages

15

As part of the predominance inquiry, plaintiffs must demonstrate that “damages are

16

capable of measurement on a classwide basis.” Comcast Corp. v. Behrend, 569 U.S. 27, 34

17

(2013). Plaintiffs must present a damages model consistent with their theory of liability—that is, a

18

damages model “purporting to serve as evidence of damages in this class action must only those

19

damages attributable to that theory.” Id. at 35. “Calculations need not be exact,” id., nor is it

20

necessary “to show that [the] method will work with certainty at this time,” Khasin, 2016 WL

21

1213767, at *3. “Restitution under the UCL and FAL ‘must be of a measurable amount to restore

22

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

23

must be supported by evidence.’” Pulaski & Middleman, LLC v. Google, Inc., 802 F.3d 979, 988

24

(9th Cir. 2015).

25
26

Plaintiffs advance two theories of classwide damage calculations through the expert report
of Dr. Lance Kaufman, an economist with experience calculating economic damages. Expert

27
28

remanufactured devices; instead he asserts that the testing Apple does perform is not sufficient to
establish equivalence as promised under the AC/AC+ contract. See Cert. Reply 11.
22

1

Report of Lance Kaufman (“Kaufman Rpt.”) [Dkt. No. 102-25]. He asserts that he can calculate

2

economic harm on a classwide basis using: (i) the difference in retail price between

3

remanufactured devices and new devices and (ii) the cost of AC/AC+ plans. Id. at 4. He asserts

4

that consumers value remanufactured devices less than new devices as evidenced by their

5

willingness to pay more for new devices. Id. at 5-7. “Damage related to historically received

6

remanufactured devices should be equal to the sum of the price difference at the time each

7

replacement device was received for all replacement devices received.” Id. at 7. In addition, he

8

asserts that class members should be able to choose contract rescission because the price

9

difference calculation may not fully mitigate the harm. Id. at 8.

10

Apple argues that the difference in retail price calculation is not tethered to plaintiffs’

United States District Court
Northern District of California

11

theory of liability; instead, any damages theory must be based on the price of the AC/AC+ plans

12

that Apple allegedly breached. Cert. Oppo. 21. Further, damages cannot be based on the retail

13

price of new devices because plaintiffs were not promised new devices. Id. at 22. Apple also

14

criticizes Kaufman’s model for being underdeveloped and failing to account for the discounted

15

prices most consumers pay for their phones thanks to a cellular service contract. Id.

16

Plaintiffs’ first damages model satisfies Comcast because it is tethered to their theory of

17

liability, namely that Apple’s breach deprived them of the benefit of their bargain. See Nguyen,

18

932 F.3d at 816 (finding a sufficient nexus between a benefit-of-the-bargain theory of liability and

19

a model based on the average cost of replacing the allegedly defective system in the car). In return

20

for their payment under the AC/AC+ plans, they were entitled to replacement devices that were

21

equivalent to new. Instead, they contend they received inferior remanufactured devices. See

22

Nguyen, 932 F.3d at 822 (concluding that plaintiffs’ damages model was tied to their theory of

23

liability, namely that “the allegedly defective clutch itself [was] the injury, regardless of whether

24

the faulty clutch caused performance issues”). One measure of the inferiority of the devices is the

25

difference in retail price between new devices and those the plaintiffs received. Although it is true

26

that “equivalent to new” means that consumers were not necessarily entitled to new devices under

27

AC/AC+, plaintiffs’ expert reasonably relies on the retail price of new devices to measure the

28

retail price of equivalent-to-new devices.
23

Apple urges that the model is inappropriate because individuals could receive a greater

1
2

amount in damages than they paid for the AC/AC+ coverage. That possibility does not preclude

3

plaintiffs’ damages model; all insurance schemes run the same risk. Apple offered consumers the

4

opportunity to purchase AC/AC+ at a certain price, likely determined by reference to the

5

(presumably) lower cost of producing remanufactured devices and based on an understanding that

6

not all purchasers would require replacement devices. For those purchasers who did require

7

replacement devices, they were entitled under the contract to one that met an equivalent-to-new

8

standard. If a fact finder determines that remanufactured devices do not meet that mark, the class

9

will be entitled to damages. Kaufman’s model appropriately measures the difference between the
value of what plaintiffs were promised—equivalent-to-new devices, as measured by the retail

11
United States District Court
Northern District of California

10

price of new devices—and what they received—remanufactured devices, as measured by their

12

retail value. Apple struck this bargain and was obligated to deliver on its promise.
For two reasons, plaintiffs cannot proceed on a rescission model. First, that model is not

13
14

tied to their theory of liability, which is that Apple breached when it provided purchasers of

15

AC/AC+ with remanufactured devices, which are not equivalent to new. Second, an inferior

16

replacement device does not render the AC/AC+ plans valueless because they provided other

17

benefits, including free technical support.
For the reasons set forth above, plaintiffs have met their Rule 23 burden to show that class

18
19

certification is appropriate.26 I will certify the following class: All individuals who purchased

20

AppleCare or AppleCare+, either directly or through the iPhone Upgrade Program, on or after July

21

20, 2012, and received a remanufactured replacement Device.

22

III.

Both parties filed motions to seal the briefing and exhibits associated with the pending

23
24

MOTIONS TO SEAL

motions. Plaintiffs made their requests based on Apple’s designation of the information as

25
26
27
28

26

Apple objects to the declarations of Bardwell and Kaufman submitted with plaintiffs’ reply in
support of class certification. Dkt. No. 126. Apple argues that the declarations go beyond the
proper scope of a reply because they are based on information that was available to both experts
when they authored their initial reports. I had no need to rely on this evidence to resolve the
motion for class certification. The objected-to portions are STRUCK.
24

1

confidential; they “take no position on whether these documents qualify for protection.” Motion

2

to Seal re: Cert. Mot. [Dkt. No. 102] 2.27
Both motions before me are “more than tangentially related to the merits of [the] case”;

3

accordingly, Apple’s sealing requests are subject to the compelling reasons standard. See Ctr. for

5

Auto Safety v. Chrysler Grp., LLC, 809 F.3d 1092, 1101 (9th Cir. 2016). In deciding summary

6

judgment in the companion case before me, 14-cv-1619 English v. Apple, I granted Apple’s

7

motion to seal “specific and narrowly tailored requests to seal information pertaining to its testing

8

processes and procedures, sales and services numbers, and databases.” Unredacted Order on

9

FRCP 56(d) and Summary Judgment [14-cv-1619 Dkt. No. 334] 30. Later I unsealed the

10

Unredacted Order because Apple had not provided compelling reasons why the redacted

11
United States District Court
Northern District of California

4

information should be sealed. Order Denying Defendants’ Sealing Requests [14-cv-1619 Dkt. No.

12

339]. The Order and motions before me in this case involve much more detailed and sensitive

13

information than at issue there; accordingly, more information is sealable. I will address each

14

category of requests in turn as set forth in the declarations of Pami Vyas.28 See Declaration of

15

Pami Vyas ISO Motions to Seal re: MSJ (“Vyas MSJ Decl. ”) [Dkt. No. 110-1]; Declaration of

16

Pami Vyas ISO Motions to Seal re: Class Certification (“Vyas Cert. Decl.”) [Dkt. No. 128-1];

17

Declaration of Pami Vyas ISO Motions to Seal MSJ Reply [Dkt. No. 138-1].
First, Apple seeks to seal information related to its remanufacturing and testing process,

18
19

including the specific parts that it uses in remanufactured devices and how those devices are

20

assembled and tested. Apple asserts that I have sealed such information in the past and that it is

21

“among the most competitively sensitive information that is at issue in this case.” Vyas MSJ

22

Decl. ¶ 3. Apple expends significant resources developing these processes and maintaining their

23

confidentiality. Id. ¶¶ 4-5. Second, Apple seeks to seal references to the information it tracks and

24

how it maintains that data. Apple has expended time and resources to develop these business

25
26
27
28

27

The plaintiffs amended their motion for class certification and accompanying motion to seal.
The original motions to seal at Dkt. No. 99 are TERMINATED AS MOOT.
28

Outside of the charts laying out specific redactions, the declarations are identical through
paragraph 10.
25

1

practices and maintain their confidentiality. Id. ¶¶ 8-9. Third, Apple seeks to seal the serial

2

numbers associated with Maldonado’s and Carter’s devices on the ground that third parties could

3

misuse them to plaintiffs’ detriment. Id. ¶ 12. Fourth, Apple seeks to seal information about its

4

sales and service numbers for AC/AC+ on the grounds that their disclosure would allow its

5

competitors to unfairly compete by using Apple’s numbers in their own forecasting, planning, and

6

marketing efforts. Vyas Cert. Decl. ¶¶ 12-13.
These categories are likely to include sealable information. That said, Apple’s requests are

7
8

overbroad. See Civ. L.R. 79-5(b) (“The request must be narrowly tailored to seek sealing only of

9

sealable material.”). I will not seal information that is necessary to understand plaintiffs’ theory of
liability and hence this Order. See Dkt. No. 128 (requesting to seal the following language from

11
United States District Court
Northern District of California

10

plaintiffs’ class certification motion: “Remanufactured devices with used parts are not equivalent

12

to new as they fail at a rate significantly higher than new devices.”29). Because I recognize the

13

potential sensitivity of the categories of information, and because in another context I allowed

14

Apple to seal more general information than I am contemplating allowing here, in this redacted

15

Order I have temporarily redacted information that does not appear sensitive and does appear

16

central to plaintiffs’ theory that will be tried in open court. But I intend to file an unredacted

17

version of the Order in ten days unless Apple submits a declaration meeting the compelling

18

reasons standard on any of that information.
With respect to the rest of the sealed information, the sheer number of lines for some

19
20

documents reveals the overbreadth; in some cases Apple seeks to seal entire pages of deposition

21

testimony. See generally Dkt. No. 128.30 Redactions to the Pecht, Bardwell, and Kaufman reports

22

must be narrowed. Information about the manufacturing process of remanufactured devices is

23
24
25
26
27
28

29

This assertion may be based on evidence that Apple considers confidential, but it does not
reveal anything about the data sets themselves.
30

To provide Apple an example of its overbreadth, it requests sealing of the majority of pages 24
and 25 of the Lanigan deposition. Those pages describe the difference (or lack thereof) between
the terms refurbished, certified refurbished, and remanufactured. This testimony does not describe
Apple’s manufacturing and testing procedures (and is not sealable on that basis) but rather
explains the relationship between terms that are potentially relevant to this outcome of this case.
As currently before me, there are no compelling reasons to seal such information.
26

1

certainly key to Apple’s defense; for this reason, redactions to depositions and declarations of

2

Lanigan, Fu, and Sen should be narrowly tailored.

3

With this guidance in mind, Apple shall file amended sealing requests within 30 days

4

of the date of this Order for all requests other than the information that is currently

5

redacted in this Order. Apple need not submit new versions of any documents, but it should

6

clearly reference the docket numbers as it has done in its prior filings. After I have reviewed the

7

narrower requests, I will order the parties to submit public versions of documents with approved

8

redactions as necessary.

9
10

CONCLUSION
As set forth above, plaintiffs’ motion for class certification is GRANTED for the following

United States District Court
Northern District of California

11

class: All individuals who purchased AppleCare or AppleCare+, either directly or through the

12

iPhone Upgrade Program, on or after July 20, 2012, and received a remanufactured replacement

13

Device. Maldonado and Carter are appointed as class representatives, and Hagens Berman is

14

appointed as class counsel. Apple’s motion for summary judgment is DENIED. Plaintiffs’

15

Conditional Motion under Rule 56(d) and the accompanying motions to seal are TERMINATED

16

AS MOOT. The remaining motions to seal are resolved in accordance with the discussion above.

17

A further Case Management Conference is set for December 3, 2019 at 2:00 p.m.

18

IT IS SO ORDERED.

19

Dated: September 17, 2019

20
21
William H. Orrick
United States District Judge

22
23
24
25
26
27
28
27