ORDER DENYING PLAINTIFFS' THIRD MOTION TO CERTIFY CLASS; DENYING AS MOOT APPLE'S MOTION TO EXCLUDE STEFAN BOEDEKER

Re: Dkt. Nos. 341, 362

LUCY H. KOH, District Judge.

Plaintiffs bring this putative class action against Defendant Apple, Inc. ("Apple") based on Apple's alleged failure to disclose an alleged defect with the iPhone 6 and iPhone 6 Plus. Before the Court are two motions: (1) Plaintiffs' third motion for class certification, ECF No. 341; and (2) Apple's motion to exclude Stefan Boedeker and his amended expert report. ECF No. 362. Having considered the submissions of the parties, the relevant law, and the record in this case, the Court DENIES Plaintiffs' third motion for class certification and DENIES as moot Apple's motion to exclude.

I. BACKGROUND

A. Factual Background

Apple is the designer, manufacturer, marketer, and seller of the iPhone smartphone. ECF No. 172 ¶ 25 (Fourth Amended Class Action Complaint, or "FACC"). The iPhone utilizes a touchscreen for users to interact with the device, and use of the touchscreen is required to send text messages, capture video, browse the internet, and access applications, among other functions. Id. ¶¶ 26, 28. Apple released the iPhone 6 and iPhone 6 Plus on September 19, 2014. Id. ¶ 25. The iPhone 6 and 6 Plus both have a larger touchscreen than Apple's prior iPhone models. Id. ¶ 30. Purchasers of the iPhone 6 and 6 Plus had 14 days after purchase to return their iPhones for a full refund. ECF No. 54-2.

According to Plaintiffs, the iPhone 6 and 6 Plus "suffer from a material manufacturing defect that causes the touchscreen to become unresponsive to users' touch inputs" (hereinafter, the "touchscreen defect"). Id. ¶ 43. Plaintiffs allege that the touchscreen defect is caused by a defect in the iPhone's external casing. Id. ¶ 45. Specifically, "the touchscreen function fails because the phones' external aluminum casing, whose primary purpose is to protect the sensitive internal components from strain, is insufficient to prevent the phones from bending during normal use." ECF No. 174 at 4. This bending causes two main forms of damage to the iPhone's circuits: "trace cracks . . . and/or solder ball cracks, both of which negatively affect the flow of electricity." Id. at 6. Specifically, the damage interrupts the flow of electricity within circuits, thereby preventing the iPhone from recognizing when a user is touching the screen. Id. The problem is at first intermittent, but becomes permanent as time passes and the trace and solder ball cracks worsen. Eventually, the touchscreen defect "causes the touchscreen to completely fail to respond to user inputs." Id. at 7.

Plaintiffs allege that Apple knew about the touchscreen defect before releasing the iPhone 6 and 6 Plus on September 19, 2014. A consumer posted on Apple's website about "iPhone 6 touchscreen problems" on September 18, 2014, the day before the iPhone 6 and 6 Plus were released to the public. FACC ¶ 54; see also id. ¶¶ 66-69 (other consumer complaints from shortly after the release date). Moreover, Apple's internal testing "determined that the iPhone 6 was 3.3 times more likely to bend than the iPhone 5s (the model immediately prior to the subject iPhones) and that the iPhone 6 Plus was 7.2 times more likely to bend than the iPhone 5s." ECF No. 174 at 8. Underscoring the point, one of the major concerns Apple identified prior to launching the iPhones was that they were "likely to bend more easily when compared to previous generations," something that Apple described as "expected behavior." ECF No. 173-18 at 15.

Plaintiffs allege that within days of the iPhones' release on September 19, 2014, "there were widespread consumer complaints about the iPhones bending." ECF No. 174 at 8. Plaintiffs state that Apple then publicly denied that there was a bending problem, an incident the media termed "BendGate." FACC ¶ 54; ECF No. 174 at 8. Specifically, Apple stated:

FACC ¶ 79.

After internal investigation, Apple determined underfill was necessary to resolve the problems caused by the touchscreen defect. As Plaintiffs explain, "[u]nderfill is a bead of epoxy encapsulant that is placed on a circuit chip to reinforce its attachment to the board substrate and to stiffen the surrounding assembly. . . . Underfill is used to prevent the manifestation of chip defects induced by bending because it reinforces the connections and prevents them from bending away from the substrate." ECF No. 174 at 11 (internal citations omitted). Apple had used underfill on the preceding iPhone generation but did not start using it on the Meson (U2402) chip in the iPhone 6 and iPhone 6 Plus until May 2016. Id.

On November 18, 2016, Apple announced a customer service program related to the touchscreen defect called the "Multi-Touch Repair Program." FACC ¶ 119. Prior to the Multi-Touch Repair Program, Apple charged approximately $349 for a refurbished iPhone when a consumer complained of the touchscreen defect outside of Apple's warranty. Id. Through the Repair Program, Apple has offered to repair consumers' devices for $149 if the iPhone is otherwise working, and the screen is not broken. Id. Apple also offers to reimburse consumers for amounts previously paid over $149. Id. ¶ 120.

Plaintiffs allege that Apple did not disclose the existence of the touchscreen defect despite having exposed consumers to materials in which Apple could have disclosed the defect. Each new iPhone 6 and 6 Plus came in an identical box designed by Apple that contained various disclosures and advertisements about the iPhones' capabilities. See ECF No. 173-22 at 4-7. Inside the iPhone box, Apple included two documents that made additional representations about the iPhone. ECF No. 174 at 10. Apple also requires users to navigate an iPhone setup process before the iPhone can be used, and periodically releases updates to the iPhones' software. ECF No. 174-4; ECF No. 174 at 10, 22. However, Apple did not disclose the defect in any of these materials.

Each named Plaintiff experienced the touchscreen defect after purchasing their iPhone. Below is a chart summarizing the relevant details of the three named Plaintiffs whose claims are at issue in the instant motion for class certification. ECF No. 341 at 1.

B. Procedural History

On August 27, 2016, Plaintiffs Thomas Davison, Jun Bai, and Todd Cleary filed a putative class action complaint against Apple that alleged causes of action under (1) California's Consumer Legal Remedies Act, Cal. Civ. Code § 1750; (2) Unfair Competition Law, Cal Bus. & Prof. Code § 17200; (3) False Advertisement Law ("FAL"), Cal. Bus. & Prof. Code § 17500; (4) common law fraud; (5) negligent misrepresentation; (6) unjust enrichment; (7) breach of implied warranty; (8) violation of the Magnuson-Moss Warranty Act, 15 U.S.C. § 2301; and (9) violation of the Song-Beverly Consumer Warranty Act, Cal. Civ. Code § 17290. ECF No. 1.

On October 7, 2016, Plaintiffs filed a First Amended Class Action Complaint that added several named Plaintiffs and added causes of action under the consumer fraud statutes of Illinois, New Jersey, Florida, Connecticut, Texas, Colorado, Michigan, New York, and Washington. See ECF No. 20. On December 2, 2016, Plaintiffs filed a Second Amended Class Action Complaint ("SACC"), which added a Utah Plaintiff and a cause of action under Utah's consumer fraud statute. Plaintiffs sought to represent a Nationwide Class of "[a]ll persons or entities in the United States that purchased an Apple iPhone 6 or 6 Plus." Alternatively, Plaintiffs sought to represent state sub-classes. Id.

Given the breadth of the Plaintiffs' action, the Court ordered the parties at the November 30, 2016 initial case management conference to each select 5 causes of action—for a total of 10 causes of action—to litigate through summary judgment. See ECF No. 44. On December 5, 2016, the parties selected (1) New Jersey Consumer Fraud Act ("NJCFA"), N.J. Stat. Ann. § 56:8-1; (2) Florida Deceptive and Unfair Trade Practices Act ("FDUTPA"), Fla. Stat. § 501.201; (3) Washington Consumer Protection Act ("WCPA"), Wash. Rev. Code § 19.86.010; (4) Illinois Consumer Fraud and Deceptive Trade Practices Act ("ICFDTPA"), Ill. Comp. Stat ¶ 505; (5) Texas Deceptive Trade Practices Act ("TDTPA"), Tex. Bus. & Com. Code ¶ 17.41; (6) Colorado Consumer Protection Act ("CCPA"), Colo. Rev. Stat. § 6-1-105; (7) common law fraud; (8) breach of express warranty; (9) breach of implied warranty; and (10) Magnuson-Moss Act. ECF No. 44 at 1-2. The parties did not select any California statutory claims. See id.

1. Apple's Motions to Dismiss

On January 6, 2017, Apple filed a motion to dismiss the SACC. ECF No. 54. On March 14, 2017, the Court dismissed all 10 of the selected claims with leave to amend. See ECF No. 84; Davidson v. Apple, Inc., 2017 WL 976048 (N.D. Cal. Mar. 14, 2017) ("Davidson I").

On April 4, 2017, Plaintiffs filed the Third Amended Class Action Complaint ("TACC"). ECF No. 86. On April 18, 2017, Apple moved to dismiss the TACC. See ECF No. 87.

On July 25, 2017, the Court granted in part and denied in part Apple's motion to dismiss the TACC. See ECF No. 103; Davidson v. Apple, Inc., 2017 WL 3149305 (N.D. Cal. July 25, 2017) ("Davidson II"). To start, the Court found that some Plaintiffs lacked standing to seek injunctive relief enjoining Apple's allegedly fraudulent misrepresentations and omissions about the iPhones because they did not intend to buy a new phone or participate in Apple's Multi-Touch Repair Program. Id. at *7-8. Conversely, other Plaintiffs did have standing to seek injunctive relief because they intended to participate in the Multi-Touch Repair Program, or were at least willing to consider doing so. Id. at *8-9.

The Court then turned to Plaintiffs' fraud claims. First, the Court dismissed Plaintiffs' fraud claims based on affirmative misrepresentations "because Plaintiffs have failed to identify an actionable misrepresentation in the September 25, 2014 statement—and because this statement is the only statement that forms the basis of Plaintiffs' affirmative misrepresentation claims." Id. at *13. Second, the Court declined to dismiss Plaintiffs' fraud claims based on an omission theory because "Plaintiffs have sufficiently alleged the information about the iPhone to which Plaintiffs were exposed either prior to their purchase or immediately after their purchase and within the time window in which they could have returned their iPhone for a full refund." Id. at *14. The Court also found that Plaintiffs had adequately alleged that Apple knew of the touchscreen defect at the time of the Plaintiffs' purchases. See id. at *14-15.

The Court next dismissed Plaintiffs' claims under the NJCFA and Pennsylvania common law fraud with prejudice. Plaintiffs' NJCFA claim failed because the only New Jersey Plaintiff experienced the touchscreen defect after the expiration of Apple's one-year limited warranty period, and New Jersey law provides that "[a] defendant cannot be found to have violated the CFA when it provided a part—alleged to be substandard—that outperforms the warranty provided." Perkins v. DaimlerChrysler Corp., 890 A.2d 997, 1004 (N.J. App. Div. 2006). The Court dismissed Plaintiffs' claim for Pennsylvania common law fraud as barred by the economic loss doctrine, which bars a plaintiff "from recovering in tort economic losses to which their entitlement flows only from a contract." Werwinski v. Ford Motor Co., 286 F.3d 661, 670 (3d Cir. 2002).

Finally, the Court dismissed Plaintiffs' breach of express and implied warranty claims under Illinois law with prejudice. After finding that the limited warranty was not unconscionable, the Court dismissed the breach of express warranty claim because the limited warranty excluded design defects, and Plaintiffs alleged only a design defect. Davidson II, 2017 WL 3149305, at *24. Similarly, the Court dismissed the breach of implied warranty claim because the limited warranty was not unconscionable and expressly disclaimed an implied warranty. Id. at *26. The Court also dismissed Plaintiffs' Magnuson-Moss Act claim because the parties did not dispute that the claim rose or fell with Plaintiffs' express and implied warranty claims under state law. Id.

Thus, five causes of action (all premised on a fraudulent omissions theory) survived Apple's motion to dismiss the TACC: (1) a Colorado Consumer Protection Act ("CCPA") claim; (2) a Florida Deceptive and Unfair Trade Practices Act ("FDUTPA") claim; (3) an Illinois Consumer Fraud and Deceptive Trade Practices Act ("ICFDTPA") claim; (4) a Texas Deceptive Trade Practices Act ("TDTPA") claim; and (5) a Washington Consumer Protection Act ("WCPA") claim.

On December 21, 2017, the Court granted the parties' stipulation to file a Fourth Amended Class Action Complaint ("FACC"). ECF No. 169. The FACC was materially identical to the TACC save for the substitution of Plaintiff Eric Siegal, an Illinois resident, for Adam Benelhachem, the previous Illinois Plaintiff. See FACC. On January 3, 2018, Plaintiffs filed the FACC. ECF No. 172. On January 17, 2018, Apple filed its Answer to the FACC. ECF No. 177.

2. Court Denies Plaintiffs' First Motion for Class Certification Because Plaintiffs' First Damages Expert Report Failed to Satisfy Comcast

On January 5, 2018, Plaintiffs filed their first motion for class certification. ECF No. 174. Plaintiffs sought to certify the following proposed class under Rule 23(b)(3) or Rule 23(c)(4):

Id. at 3.1 Plaintiffs argued in the alternative that the Court should certify subclasses to account for differences between the states' laws. Id. at 23 n.13; TACC ¶ 129 ("In the alternative, Plaintiffs seek to represent the following state sub-classes."). On February 10, 2018, Apple filed its opposition, ECF No. 183, and on March 2, 2018, Plaintiffs filed their reply. ECF No. 199.

On May 8, 2018, the Court denied Plaintiffs' motion for class certification. ECF No. 225; Davidson v. Apple, 2018 WL 2325426 (N.D. Cal. May 8, 2018) ("Davidson III"). The Court rejected Apple's arguments that Plaintiffs lack standing, id. at *7-10, and explained that Plaintiffs' Colorado claim was categorically barred because the applicable Colorado law bars class actions for damages. Id. at *10-11.

As relevant to this motion, the Court turned to whether Plaintiffs' remaining state law claims satisfied Rule 23(a) and Rule 23(b)(3). The Court determined that Plaintiffs' remaining claims under Florida, Illinois, Washington, and Texas law satisfied Rule 23(a). Id. at *11-13. The Court concluded that predominance was met for Illinois, Florida, and Washington law because those statutes permit inferring reliance on a classwide basis, but denied certification of Plaintiffs' Texas claim because Texas law does not permit inferring reliance and individualized inquiries would predominate on Plaintiffs' Texas claim. Id. at *15-16. The Court also rejected Apple's arguments that Plaintiffs could not prove exposure and manifestation on a classwide basis. Id. at *17-20.

The Court then determined that Plaintiffs' damages model failed to satisfy the requirements set forth in Comcast v. Behrend, 569 U.S. 27 (2013). The Court explained that in Comcast, the United States Supreme Court held that "a plaintiff bears the burden of providing a damages model showing that `damages are susceptible of measurement across the entire class for purposes of Rule 23(b)(3).'" Davidson III, 2018 WL 2325426, at *21 (quoting Comcast, 569 U.S. at 35). "The damages model `must measure only those damages attributable to' the plaintiff's theory of liability." Id. (quoting Comcast, 569 U.S. at 35).

The Court then turned to Plaintiffs' damages model. Plaintiffs' theory of liability was that Apple's failure to disclose the touchscreen defect caused consumers "to pay more for those products than they otherwise would have." ECF No. 174 at 26. Plaintiffs' damages expert, Dr. Stefan Boedeker, proposed a "Choice-Based Conjoint Analysis" damages model to analyze the value consumers attach to specific product attributes. Davidson III, 2018 WL 2325426, at *21. In a survey, Boedeker asked respondents to choose between phones differing in storage capacity, screen size, talk time, price, and defectiveness (no defect, defect costing $100 to repair or replace, or defect costing $200 to repair or replace). Id. Then, Boedeker "isolated the (negative) value associated with defectiveness," and determined that the economic loss was $323 or 51.7% of the phone's price for the $100 defect and $432 or 69.2% of the phone's price for the $200 defect. Id.

Apple argued, and the Court agreed, that Boedeker's damages model failed to satisfy Comcast "because it assumes that the touchscreen defect will manifest in all iPhones." Id. at *22. By contrast, Plaintiffs' theory of liability was that Apple "failed to disclose the existence of a touchscreen defect that manifests in approximate 5.6 percent of the iPhone 6 Plus (after two years of use) and at a somewhat lower rate for the iPhone 6." Id. (citing ECF No. 174, Plaintiffs' motion for class certification). Thus, the Court held that to correspond to Plaintiffs' theory of liability, Boedeker's damages model should measure "how much consumers overpaid for iPhones assuming a roughly 5.6 percent or less chance that consumers would experience the touchscreen defect," rather than what Boedeker actually measured: "how much consumers overpaid for a touchscreen defect that is certain to manifest in all iPhones." Id. (emphasis in original). The options in Boedeker's survey failed to convey that the touchscreen defect might not manifest. Id. at *23. This deficiency alone meant that Boedeker's damages model failed to satisfy Comcast.

The Court stated that Boedeker's damages model failed to satisfy Comcast in other respects. For one, Boedeker's damages model assumed that the touchscreen defect rendered iPhones "inoperable," but "none of the named Plaintiffs have experienced complete iPhone inoperability as a result of the touchscreen defect." Id. at *23. In addition, Boedeker's survey asked consumers "about a generic defect instead of one specifically affecting a phone's touchscreen," and thus "necessarily assumed that respondents would value all defects equally." Id. The Court held that this model was inconsistent with Plaintiffs' theory of liability "because it unmoors Plaintiffs' damages from the specific touchscreen defect alleged to have harmed them." Id. Thus, the Court concluded that Plaintiffs failed to satisfy the Rule 23(b)(3) predominance requirement because Plaintiffs' damages model was inadequate under Comcast. Id. at *24.

In addition, the Court denied Plaintiffs' request for injunctive relief as waived, id. at *25, and denied Plaintiffs' motion to certify a Rule 23(c)(4) issues class because Plaintiffs failed to show why an issues class "would materially advance the litigation as a whole." Id. at *25-26.

3. The Court Denies Plaintiffs' Second Motion for Class Certification Because Plaintiffs' Motion Relied on a Hypothetical Expert Report

On May 16, 2018, in the parties' joint case management statement, Plaintiffs stated that they intended to either file a motion for leave to file a motion for reconsideration of the Court's May 8, 2018 denial of class certification or an appeal under Federal Rule of Civil Procedure 23(f), or both. ECF No. 232 at 1. Plaintiffs also asked the Court to stay the case. Id.

On May 22, 2018, Plaintiffs declined to file a motion for reconsideration, and instead filed in the Ninth Circuit a petition for permission to appeal the Court's denial of class certification under Federal Rule of Civil Procedure 23(f). ECF No. 235. On May 23, 2018, the Court held a case management conference, and stayed the case pending resolution of Plaintiffs' 23(f) petition. ECF No. 236.

On August 9, 2018, Plaintiffs moved to voluntarily dismiss their Rule 23(f) petition, ECF No. 248, and on August 24, 2018, the Ninth Circuit granted Plaintiffs' motion and dismissed Plaintiffs' Rule 23(f) petition. ECF No. 252.

On September 12, 2018, Plaintiffs asked to lift the stay of the case and to file a second motion for class certification by September 27, 2018. ECF No. 255 at 2-3. Plaintiffs stated that Plaintiffs had withdrawn their Rule 23(f) petition on August 9 in light of the Court's statements at the May 23, 2018 case management conference that the Court would probably certify a risk class if Plaintiffs presented a compliant damages model:

Id. at 2 (quoting ECF No. 238 at 8:10-16). Plaintiffs requested that the second motion for class certification be heard as early as November 1, 2018. Id. at 3. The Court set the next case management conference for October 31, 2018. ECF No. 258.

On October 24, 2018, Plaintiffs proposed to file their second motion for class certification by November 8, 2018 and that the second motion for class certification be heard as early as December 6, 2018. ECF No. 264 at 1.

On October 31, 2018, the Court held a further case management conference, lifted the stay in the case, and adopted Plaintiffs' proposed November 8, 2018 deadline for Plaintiffs to file their second motion for class certification. ECF No. 268.

At the October 31, 2018 case management conference, Plaintiffs' counsel stated that after receiving the Court's May 8, 2018 order denying Plaintiff's first motion for class certification, "we immediately worked with our expert following the receipt of that order and following the status conference. And the, the methodology that Your Honor talked about in her order is easily done, has been done as a matter of fact." ECF No. 276 at 6:4-8.

On November 8, 2018, Plaintiffs filed their second motion for class certification, which included a Boedeker supplemental declaration that discussed only a hypothetical survey. ECF No. 270. On December 6, 2018, Apple filed its opposition to Plaintiffs' second motion for class certification, ECF No. 283, and on December 20, 2018, Plaintiffs filed their reply. ECF No. 288.

On December 6, 2018, Apple moved to exclude Boedeker's supplemental declaration. ECF No. 284. On December 20, 2018, Plaintiffs filed their opposition to Apple's motion to exclude Boedeker's supplemental declaration, ECF No. 289, and on December 27, 2018, Apple filed its reply. ECF No. 295.

After Plaintiffs' second motion for class certification was fully briefed on December 20, 2018, and after Apple's motion to exclude Boedeker's supplemental declaration was fully briefed on December 27, 2018, Plaintiffs produced Boedeker's second completed damages survey and second expert report ("Second Boedeker Report") on January 18, 2019. ECF No. 314 at 3 (citing ECF No. 309).

On February 7, 2019, a week after filing its summary judgment reply brief on February 1, 2019, Apple filed an administrative motion for leave to file a supplemental brief in opposition to Plaintiffs' second motion for class certification and in support of Apple's motion to exclude Boedeker's supplemental declaration. ECF No. 309. Apple requested leave to brief how the Second Boedeker Report affected Plaintiffs' pending second motion for class certification. Id. at 2. Apple also stated that the survey discussed in the Second Boedeker Report told survey respondents the wrong risk manifestation rates: "Mr. Boedeker told respondents that the iPhone 6 suffers from a defect 5.6% of the time, which is in fact the alleged rate of touchscreen issues for the iPhone 6 Plus, and vice versa." Id. at 2-3. On February 11, 2019, Plaintiffs opposed Apple's administrative motion, but did not address the erroneous risk rates in Boedeker's second survey. ECF No. 310.

On February 12, 2019, the Court denied without prejudice Plaintiffs' second motion for class certification. ECF No. 314. The Court stated that in Plaintiffs' second motion for class certification, Plaintiffs represented that "Plaintiffs' expert, Stefan Boedeker, has stated how he will update his damages model to track this theory of liability and thereby satisfy Comcast." Id. at 2 (emphasis in original) (citing ECF No. 270 at 5). The Court continued:

Id.

Thus, the entire briefing on Plaintiffs' second motion for class certification concerned a hypothetical survey even though a completed survey now existed. Id. Accordingly, the Court denied without prejudice Plaintiffs' second motion for class certification "[b]ecause analysis of whether Boedeker's new survey satisfies Comcast is essential to a class certification determination; because Plaintiffs did not complete this survey before filing their renewed motion for class certification despite having 6 months to do so; and because briefing on class certification should address Boedeker's new survey." Id.

4. Apple's Motion for Summary Judgment

On December 21, 2018, while the parties were briefing class certification, Apple filed a motion for summary judgment. ECF No. 291. On January 17, 2019, Plaintiffs filed their opposition to Apple's motion for summary judgment, ECF No. 302, and on February 1, 2019, Apple filed its reply. ECF No. 305.

On February 21, 2019, the Court granted in part and denied in part Apple's motion for summary judgment. ECF No. 324. The Court (1) denied the motion as to Plaintiffs' Colorado claim, id. at 11-14; (2) denied the motion as to Plaintiffs' Florida claim, id. at 14-15; (3) denied the motion as to Plaintiffs' Washington claim, id. at 15-17; (4) granted the motion as to Plaintiffs' Illinois claim; id. at 17-20; (5) granted the motion as to Plaintiffs' Texas claim, id. at 20-22; and (6) granted the motion as to Plaintiff Bauer's injunctive relief claim. Id. at 22-24.

On February 28, 2019, Apple filed a motion for leave to file a motion for reconsideration of the Court's denial of Apple's motion for summary judgment on Plaintiffs' Florida claim. ECF No. 325. On March 1, 2019, the Court filed an order reaffirming the Court's denial of Apple's motion for summary judgment on Plaintiffs' Florida claim, ECF No. 333, and an order denying Apple's motion for leave to file a motion for reconsideration. ECF No. 334.

5. The Court Denies Plaintiffs' Attempt After the Deadline for Exchanging Expert Reports to Serve a Third Damages Expert Report

On March 1, 2019, four days before Plaintiffs' March 5, 2019 deadline to file their third motion for class certification, Plaintiffs filed an administrative motion to (1) serve a corrected expert report of Stefan Boedeker; and (2) modify the briefing schedule on Plaintiffs' anticipated third motion for class certification. ECF No. 335.

In the March 1, 2019 motion, Plaintiffs stated that "[o]n February 28, 2019, Plaintiffs' counsel discovered that the consumer survey" discussed in the Second Boedeker Report contained errors "resulting in inaccurate data being used to calculate economic loss." Id. at 2. Specifically, the survey transposed the defect risk rates for the iPhone 6 and the iPhone 6 Plus. Id.

Plaintiffs' representation that they only discovered the error on February 28, 2019 was not credible. Plaintiffs' March 1, 2019 motion concerned the same survey errors that Apple had identified almost a month earlier in a February 7, 2019 motion filed on the public docket. See ECF No. 309 at 2 ("The survey also tells respondents the wrong risk rate: Mr. Boedeker told respondents that the iPhone 6 suffers from a defect 5.6% of the time, which is the fact the alleged rate of touchscreen issues for the iPhone 6 Plus."). On February 11, 2019, Plaintiffs filed a response to Apple's February 7, 2019 motion. ECF No. 310. Moreover, on February 15, 2019, Apple served two rebuttal expert reports identifying the same risk rate errors and other flaws in the second survey. ECF No. 342.

Given the errors in their second survey and in the Second Boedeker Report premised on the survey, Plaintiffs requested leave to redo the second survey and serve a corrected, third Boedeker expert report. ECF No. 335 at 2. Plaintiffs also asked to modify the briefing schedule on Plaintiffs' third motion for class certification "because Mr. Boedeker's new survey results and corresponding corrected report will not be finalized until after the current deadline (March 5) for Plaintiffs to file their motion for class certification." Id. at 3.

On March 4, 2019, Apple filed an opposition to Plaintiffs' administrative motion to redo Boedeker's second survey and second expert report and modify the briefing schedule on Plaintiffs' third motion for class certification. ECF No. 336.

On March 6, 2019, the Court denied Plaintiffs' administrative motion to redo Boedeker's second survey and second expert report and modify the briefing schedule. ECF No. 342. The Court recounted the long history of Plaintiffs' attempts to certify a class, and stated that "[t]his Court generally does not give plaintiffs another opportunity to a file a motion for class certification after this Court has denied a class certification motion." Id. at 2. Rather, "[t]he Court made a singular exception" in this case to allow Plaintiffs to attempt to remedy the flaws in their first damages model via a second motion for class certification, but then had to order briefing of a third motion for class certification to account for Plaintiffs' belated completion of Boedeker's second survey. Id. at 4.

Further, the Court observed that Plaintiffs knew about the manifestation risk rate errors in the Second Boedeker Report and underlying survey since at least February 7, 2019, when Apple filed a motion on the public docket stating that Boedeker's survey used the wrong risk rates. Id. at 4-5 (citing ECF No. 309). Plaintiffs responded to Apple's motion on February 11, 2019. ECF No. 310. Moreover, on February 15, 2019, Apple served on Plaintiffs two rebuttal expert reports that identified the same and other flaws in Boedeker's second survey. ECF No. 342 at 5 (citing ECF No. 336). Thus, Plaintiffs forced Apple to prepare multiple sets of rebuttal expert declarations, expert reports, and briefs to respond to Boedeker's completed first and second surveys and to respond to Boedeker's hypothetical survey discussed in Plaintiffs' second motion for class certification. Id. Accordingly, the Court denied Plaintiffs' administrative motion to redo Boedeker's second survey and report and modify the briefing schedule. Id. at 5-6.

6. Plaintiffs Serve a Third Damages Expert Report in Violation of the Court's Order

On March 26, 2019, Apple filed an administrative motion to enforce the Court's March 6, 2019 order denying Plaintiffs' administrative motion to redo Boedeker's survey and serve a third Boedeker report. ECF No. 357. Apple stated that late on March 25, 2019—four days before Apple's opposition to Plaintiffs' third motion for class certification was due on March 29, 2019— Plaintiffs served Apple with a Corrected Amended Expert Report of Stefan Boedeker, which was a putative Third Boedeker Report "premised on a new survey." Id. at 1. Apple sought to prevent Plaintiffs "from using Mr. Boedeker's latest report and survey for any purpose." Id. Apple stated that the putative Third Boedeker Report "comes to entirely different conclusions of economic loss, and changes the survey instructions and attribute descriptions that [survey] respondents saw." Id.

On March 27, 2019, the Court granted Apple's administrative motion to enforce the Court's March 6, 2019 order. ECF No. 359. The Court stated that Plaintiffs' service of the putative Third Boedeker Report "flies in the face" of the Court's March 6, 2019 order, which "explicitly prohibited" Plaintiffs from redoing the second survey and completing a third Boedeker report. Id. at 2. Further, allowing Plaintiffs to serve a third survey and third expert report would result in severe prejudice to Apple, which had already served two expert reports addressing Plaintiffs' second survey. Id. Finally, Plaintiffs' last-minute service of the putative Third Boedeker Report was untimely because Plaintiffs had missed by more than a month the February 15, 2019 deadline for the exchange of expert reports. Id.

7. Briefing on the Instant Motions

On March 5, 2019, Plaintiffs filed the instant third motion for class certification. ECF No.

341 ("Class Cert. Mot."). Plaintiffs seek to certify the following class:

Id. at 1.2 Plaintiffs no longer seek to certify classes under Illinois, Texas, or Colorado law.

On March 29, 2019, Apple filed a motion to exclude Stefan Boedeker and the opinions in the Second Boedeker Report. ECF No. 362 ("Boedeker Mot.").

That same day, Apple also filed its opposition to Plaintiffs' third motion for class certification. ECF No. 363 ("Class Cert. Opp.").

On April 11, 2019, Plaintiffs filed their reply in support of their third motion for class certification. ECF No. 366 ("Class Cert. Reply").

On April 12, 2019, Plaintiffs filed an opposition to Apple's motion to exclude Stefan Boedeker and the Second Boedeker Report. ECF No. 278 ("Boedeker Opp.").

On April 18, 2019, Apple filed objections to evidence submitted with Plaintiffs' reply in support of their third motion for class certification. ECF No. 370. On April 29, 2019, Plaintiffs filed a brief in response to Apple's objections. ECF No. 373.

On April 19, 2019, Apple filed its reply in support of its motion to exclude Stefan Boedeker and the Second Boedeker Report. ECF No. 371 ("Boedeker Reply").3

II. LEGAL STANDARD

A. Motion for Class Certification

Class actions are governed by Rule 23 of the Federal Rules of Civil Procedure, which 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).

If all four prerequisites of Rule 23(a) are satisfied, the Court must also find that the plaintiff "satisf[ies] through evidentiary proof" at least one of the three subsections of Rule 23(b). Comcast Corp. v. Behrend, 569 U.S. 27, 33 (2013). The Court can certify a Rule 23(b)(1) class when plaintiffs make a showing that there would be a risk of substantial prejudice or inconsistent adjudications if there were separate adjudications. Fed. R. Civ. P. 23(b)(1). The Court can certify a Rule 23(b)(2) class if "the party opposing the class has acted or refused to act on grounds that apply generally to the class, so that final injunctive relief or corresponding declaratory relief is appropriate respecting the class as a whole." Fed. R. Civ. P. 23(b)(2). Finally, the Court can certify 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 superior to other available methods for fairly and efficiently adjudicating the controversy." Fed. R. Civ. P. 23(b)(3).

"[A] court's class-certification analysis must be `rigorous' and may `entail some overlap with the merits of the plaintiff's underlying claim[.]" Amgen Inc. v. Conn. Ret. Plans & Tr. Funds, 568 U.S. 455, 465-66 (2013) (quoting Dukes, 564 U.S. at 351); see also Mazza, 666 F.3d at 588 ("`Before certifying a class, the trial court must conduct a `rigorous analysis' to determine whether the party seeking certification has met the prerequisites of Rule 23.'" (quoting Zinser, 253 F.3d at 1186)). This "rigorous" analysis applies to both Rule 23(a) and Rule 23(b). Comcast, 569 U.S. at 34 (stating that Congress included "addition[al] . . . procedural safeguards for (b)(3) class members beyond those provided for (b)(1) or (b)(2) class members (e.g., an opportunity to opt out)" and that a court has a "duty to take a `close look' at whether common questions predominate over individual ones").

Nevertheless, "Rule 23 grants courts no license to engage in free-ranging merits inquiries at the certification stage." Amgen, 568 U.S. at 466. "Merits questions may be considered to the extent—but only to the extent—that they are relevant to determining whether the Rule 23 prerequisites for class certification are satisfied." Id. If a court concludes that the moving party has met its burden of proof, then the court has broad discretion to certify the class. Zinser, 253 F.3d at 1186.

B. Admissibility of Expert Testimony Under Federal Rule of Evidence 702 and Daubert

Federal Rule of Evidence 702 allows admission of "scientific, technical, or other specialized knowledge" by a qualified expert if it will "help the trier of fact to understand the evidence or to determine a fact in issue." Fed. R. Evid. 702. Expert testimony is admissible pursuant to Rule 702 if it is both relevant and reliable. Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 589 (1993).

When considering expert testimony offered pursuant to Rule 702, the trial court acts as a "gatekeeper" by "making a preliminary determination that the expert's testimony is reliable." Elsayed Mukhtar v. Cal. State Univ., Hayward, 299 F.3d 1053, 1063 (9th Cir. 2002); see Kumho Tire Co. v. Carmichael, 526 U.S. 137, 147-48 (1999); Gen. Elec. Co. v. Joiner, 522 U.S. 136, 142 (1997); Daubert, 509 U.S. at 589-90. In Daubert, the United States Supreme Court identified "four factors that may bear on the analysis": (1) whether a theory or technique can be and has been tested; (2) whether the theory or technique has been subjected to peer review and publication; (3) the known or potential rate of error; and (4) whether the theory is generally accepted in the scientific community. Murray v. S. Route Maritime SA, 870 F.3d 915, 922 (9th Cir. 2017) (citing Daubert, 509 U.S. at 593-94). However, the Ninth Circuit has explained that "the reliability analysis remains a malleable one tied to the facts of each case" and that the "Daubert factors are exemplary, not constraining." Id. The Ninth Circuit has also stated that "[i]t is important to remember that the factors are not `equally applicable (or applicable at all) in every case,'" and that "[a]pplicability `depend[s] on the nature of the issue, the expert's particular expertise, and the subject of his testimony.'" Id. (first quoting Daubert v. Merrell Dow Pharm., Inc., 43 F.3d 1311, 1317 (9th Cir. 1995), then quoting Kumho Tire, 526 U.S. at 150) (third alteration in original).

"Because of the fluid and contextual nature of the inquiry, district courts are vested with `broad latitude' to `decid[e] how to test an expert's reliability' and `whether or not [an] expert's relevant testimony is reliable.'" Id. (quoting Kumho Tire, 526 U.S. at 152-53) (alterations and emphasis in original). Thus, the Court "may permissibly choose not to examine factors that are not `reasonable measures of reliability in a particular case.'" Id. (quoting Kumho Tire, 526 U.S. at 153); see also id. at 924 ("District courts have broad range to structure the reliability inquiry and may choose not to comment on factors that would not inform the analysis.").

Moreover, the inquiry into admissibility of expert opinion is a "flexible one," where "[s]haky but admissible evidence is to be attacked by cross examination, contrary evidence, and attention to the burden of proof, not exclusion." Primiano v. Cook, 598 F.3d 558, 564 (9th Cir. 2010) (citing Daubert, 509 U.S. at 594, 596). "Under Daubert, the district judge is `a gatekeeper, not a fact finder.' When an expert meets the threshold established by Rule 702 as explained in Daubert, the expert may testify and the jury decides how much weight to give that testimony." Id. (quoting United States v. Sandoval-Mendoza, 472 F.3d 645, 654 (9th Cir. 2006)).

III. DISCUSSION

The two motions before the Court both concern Stefan Boedeker's January 18, 2019 expert report ("Second Boedeker Report") and the consumer survey discussed therein. The Court first addresses Plaintiffs' third motion for class certification.

A. Third Motion for Class Certification

The Court denied Plaintiffs' first motion for class certification because Plaintiffs' damages model failed to satisfy Comcast, although the Court determined that Plaintiffs met the other requirements of Rule 23(b)(3). Davidson III, 2018 WL 2325426, at *20-21. The sole issue at stake in Plaintiffs' third motion for class certification is whether Plaintiffs' revised damages model satisfies Comcast.

Although individual damages calculations alone do not make class certification inappropriate under Rule 23(b)(3), see Leyva v. Medline Indus., Inc., 716 F.3d 510, 514 (9th Cir. 2013) ("[T]he amount of damages is invariably an individual question and does not defeat class action treatment."), the United States Supreme Court has held that a plaintiff bears the burden of providing a damages model showing that "damages are susceptible of measurement across the entire class for purposes of Rule 23(b)(3)." Comcast, 569 U.S. at 35. The damages model "must measure only those damages attributable to" the plaintiff's theory of liability. Id.

1. The Deficient Damages Model in the First Boedeker Report

As set forth at length in the Procedural History, the Court denied Plaintiffs' first motion for class certification because Plaintiffs' damages model failed to measure only those damages attributable to Plaintiffs' theory of liability. Davidson III, 2018 WL 2325426, at *21-23. The Court explained that Plaintiffs' theory of liability is that "Apple failed to disclose the [touchscreen defect] in the iPhone 6 and 6 Plus, causing consumers to pay more for those products than they otherwise would have." Id. at *21 (quoting Plaintiffs' first motion for class certification, ECF No. 174). To prove damages, Plaintiffs submitted the Expert Report of Stefan Boedeker ("First Boedeker Report"), which relied on a conjoint analysis survey that Boedeker had conducted. Id.

However, Plaintiffs' damages model had three fatal flaws: Boedeker's survey (1) "assume[d] that the touchscreen defect will manifest in all iPhones," even though Plaintiffs contended that the touchscreen defects "manifests in approximately 5.6 percent of the iPhone 6 Plus (after two years of use) and at a somewhat lower rate for the iPhone 6," id. at *22; (2) "assume[d] that the touchscreen defect renders affected iPhones inoperable," when none of the named Plaintiffs experienced complete iPhone inoperability, id. at *23; and (3) "asked respondents about a generic defect instead of one specifically affecting a phone's touchscreen," and thus "necessarily assumed that respondents would value all defects equally." Id. The Court stated that each of the three flaws provided an independent basis to deny certification. Id.

The Court gave Plaintiffs an opportunity to fix the identified issues with Plaintiffs' damages model. ECF No. 268. Thus, Plaintiffs filed a second motion for class certification, which relied on a damages model premised on a hypothetical Boedeker survey. ECF No. 270. After that second motion for class certification had been fully briefed on December 20, 2018, and after Apple's motion to exclude the hypothetical survey and Boedeker's supplemental declaration was fully briefed on December 27, 2018, Plaintiffs served on Apple on January 18, 2019 the Second Boedeker Report, which discusses a second survey that Boedeker conducted, rather than the hypothetical survey discussed in Plaintiffs' second motion for class certification. ECF No. 314. Accordingly, the Court denied without prejudice Plaintiffs' second motion for class certification and instructed Plaintiffs to file a third motion for class certification premised on the Second Boedeker Report and Boedeker's second survey. Id. Plaintiffs then filed the instant motion.

2. The Second Boedeker Report Fails to Satisfy Comcast

Like the First Boedeker Report, the Second Boedeker Report analyzes the results of a choice-based conjoint analysis study that Boedeker conducted. ECF No. 341-3 ("Second Boedeker Report"), ¶¶ 3-4. As Boedeker explains, "[t]he premise of conjoint analysis is that the market value for a given product is derived by features or descriptions of features in the product." Id. at ¶ 12; see also Davidson III, 2018 WL 2325426, at *21 (explaining that in conjoint analysis, survey respondents "choose between different sets of product attributes, the responses are aggregated, and statistical methods are then used to determine the value (often termed `part-worth') that consumers attach to each specific attribute").

Plaintiffs contend that Boedeker's second survey and the Second Boedeker Report remedy the three deficiencies that the Court identified in Davidson III. However, Plaintiffs have again failed to provide a damages model that accounts for how much consumers overpaid based on the defect, and have thus failed to cure the first deficiency identified in Davidson III. Plaintiffs' theory of liability is as follows: "Plaintiffs contend Apple failed to disclose the existence of a touchscreen defect that manifests in approximately 5.6 percent of the iPhone 6 Plus (after two years of use) and at a somewhat lower rate for the iPhone 6." Davidson III, 2018 WL 2325426, at *22. Accordingly, the Court held in denying Plaintiffs' first motion for class certification that Plaintiffs' damages model should reflect "how much consumers overpaid for iPhones" assuming those manifestation rates. Id.

However, Plaintiffs have failed to provide a damages model that accords with Plaintiffs' theory of liability because Plaintiffs' damages model overstates the cost to purchasers of addressing the touchscreen defect. This leads to elevated economic loss calculations. The FACC acknowledges that Apple provides iPhone purchasers a one-year warranty. FACC ¶ 106. As a result, the second survey was supposed to inform respondents that a purchaser could replace her iPhone for free if the defect touchscreen defect manifested within the warranty period, as follows: "You are further informed that Apple provides a one-year warranty. If the touchscreen defect manifests during the warranty period, Apple will provide you with a refurbished phone at no charge. If the touchscreen defects manifests after the warranty expires, Apple will provide you with a refurbished phone for $149." Second Boedeker Report ¶ 48 (describing proposed survey).

However, the actual second survey omitted that material information about the free warranty replacement, and instead informed respondents that the defect would always cost $149 to repair: "You are further informed that Apple offers to repair the phone for $149 when the touchscreen defect manifests on the phone you are about to buy." Appx. B at 14-15. In conflict with Plaintiffs' theory of liability, the second survey informed respondents that even if the defect manifested within the warranty period, a purchaser's only remedy was to pay $149 for repairs.

As Boedeker admitted at his deposition, the second survey thus failed to measure how purchasers would value a defect that could be remedied for free within the one-year warranty period, and inflated the economic loss calculations. Boedeker Depo. 422:1-3 ("So what the screen — what participants saw, they only test with the repair cost knowledge because the refurbished replacement is not mentioned."). Logically, the economic harm of a defect that can be remedied for free within one year is lower than a defect that can only be remedied by paying $149 for repairs. Boedeker acknowledged in his own supplemental declaration that not disclosing the one-year warranty affected his damages calculations: "How much this additional disclosure . . . affects class wide damages is an empirical question." ECF No. 366-1, Supplemental Declaration of Stefan Boedeker, ¶ 25. It is an empirical question that Plaintiffs' damages model ignored. Accordingly, Plaintiffs' damages model fails to satisfy Comcast because the second survey did not account for Apple's one-year warranty period and its effect on class damages.

Plaintiffs cite a line of Ninth Circuit cases holding that "individualized calculations do not defeat predominance," and appear to argue that these cases excuse the deficiencies in Plaintiffs' damages model. See Pulaski & Middleman, LLC v. Google, Inc., 802 F.3d 989, 988 (9th Cir. 2015) (citing Yokoyama v. Midland Nat'l Life Ins. Co., 594 F.3d 1087, 1092 (9th Cir. 2010)). However, those cases concern whether predominance is defeated when damages vary among class members. Pulaski, 802 F.3d at 987. In Pulaski, the Ninth Circuit reaffirmed that Comcast stands "for the proposition that `plaintiffs must be able to show that their damages stemmed from the defendant's actions that created the legal liability.'" Id. at 987-88 (quoting Leyva, 716 F.3d at 514). Here, Plaintiffs' damages model has a Comcast problem rather than an individualized calculation problem: Plaintiffs created a classwide damages model that overstates the cost to purchasers of addressing the touchscreen defect. Thus, Plaintiffs' damages model is not "consistent with [their] liability case, particularly with respect to the . . . effect of the violation." Comcast, 569 U.S. at 35.

In addition, Plaintiffs' damages model also fails to satisfy Comcast because Plaintiffs' damages survey used the wrong manifestation rates for the touchscreen defect. Although the touchscreen defect manifestation rates are 0.7% for the iPhone 6 and 5.6% for the iPhone 6 Plus, Boedeker's second survey never disclosed these percentages. Instead, Boedeker's second survey used raw numbers and switched the numbers for the two phones. Specifically, Boedeker's second survey asked about a defect that manifested in "approximately 2.86 million out of the 51,139,595 iPhone 6 models that were manufactured" and a defect that manifested in "approximately 125,000 out of the 17,527,727 iPhone 6 Plus models that were manufactured." ECF No. 363-2, Appx. B at 14-15 (screenshots of the second survey). Thus, Boedeker's second survey used the wrong manifestation rates.

Plaintiffs concede that Boedeker's second survey switched the manifestation rates for the two phones. Class Cert Mot. at 7. At his deposition, Boedeker admitted that the second survey did not ask respondents about the correct manifestation rates: "So what this resulted in was then to test a larger than empirically validated failure rate for the iPhone 6 and a lower than empirically validated failure rate for iPhone 6 Plus." ECF No. 361-7, Deposition of Stefan Boedeker ("Boedeker Depo."), 401:17-21. Specifically, the survey assumed a manifestation rate for the iPhone 6 that is approximately eight times larger than the actual manifestation rate for that phone model, and a manifestation rate for the iPhone 6 Plus that is approximately eight times smaller than the actual manifestation rate for that phone model. Apple identified these errors on February 7, 2019 on the public docket, but Plaintiffs did not acknowledge them in Plaintiffs' February 11, 2019 response to Apple's February 7, 2019 motion. Moreover, Plaintiffs did not request an opportunity to conduct a third survey until almost a month later, on March 1, 2019, after Plaintiffs had the benefit of Apple's two February 15, 2019 rebuttal expert reports identifying additional flaws in Plaintiffs' damages model. ECF No. 342.

The Second Boedeker Report, which was served on January 18, 2019, also does not acknowledge the manifestation rate errors in the second survey. In fact, as Boedeker admitted at his March 2019 deposition, Boedeker never reviewed the screenshots of the second survey as actually conducted before Boedeker signed the Second Boedeker Report. Id. at 409:14-17 ("I have not looked at this document with screenshots until after the — way after the report when I was made aware of this — I called the transposition error of the units and the rates."). Thus, the Second Boedeker Report assumes that respondents were asked about the correct manifestation rates, in accordance with Plaintiffs' theory of liability—but that is not what respondents to the second survey were asked. See Second Boedeker Report at ¶¶ 48-49 (describing second survey's purported questions).

Accordingly, the Second Boedeker Report reaches economic loss conclusions premised on a consumer survey conducted using manifestation rates not aligned with Plaintiffs' theory of liability. Specifically, the Second Boedeker Report projects that the median economic loss was $412.90 for an iPhone 6 and $382.50 for an iPhone 6 Plus. Id. ¶ 84. Plaintiffs concede that because of the errors in the second survey, these figures are inaccurate and useless. Class Cert Mot. at 1 ("[T]he exact economic loss calculation requires revision."). This is a commonsense conclusion. If survey respondents assumed that the risk of manifestation in the iPhone 6 was eight times higher than Plaintiffs' theory of liability presumes, survey respondents then overvalued the economic harm caused by the touchscreen defect, and Boedeker's economic loss calculations are inaccurate. See Class Cert. Mot. at 8 (acknowledging that the second survey and Second Boedeker Report lead to an "over-valuation of the economic loss associated with the iPhone 6"). Accordingly, Plaintiffs' damages model does not satisfy Comcast's requirement that a damages model "measure only those damages attributable to" the plaintiff's theory of liability. Comcast, 569 U.S. at 35. Instead, for the iPhone 6, Plaintiffs' damages model measures the damages attributable to a defect that is eight times more likely to manifest than is "empirically validated," in the words of Plaintiffs' own expert. Boedeker Depo. 401:17-21.

Nonetheless, Plaintiffs contend that the second survey's overvaluation of the iPhone 6 defect and undervaluation of the iPhone 6 Plus defect are immaterial because Plaintiffs' damages model generally "accounts for the allegation of a risk of manifestation." Class Cert. Mot. at 7. Plaintiffs effectively argue that a damages model testing the effect of a 5.6% manifestation rate somehow measures the damages attributable to a defect with a 0.7% manifestation rate. However, varying manifestation rates lead to vast differences in economic loss calculations. The Second Boedeker Report recognizes this, as the Second Boedeker Report (using erroneous manifestation rates) calculates that purchasers of the iPhone 6 experienced on median $30 more in economic loss than purchasers of the iPhone 6 Plus, even though the iPhone 6 has a 0.7% manifestation rate compared to the iPhone 6 Plus's 5.6% manifestation rate, and the iPhone 6 costs $100 less than the iPhone 6 Plus. Second Boedeker Report ¶ 84.

Even though Plaintiffs concede that Plaintiffs need to redo the second survey to generate economic loss figures that accord with Plaintiffs' theory of liability, Plaintiffs also contend that the errors in the second survey are immaterial to class certification because Plaintiffs need only propose a damages model at this stage. To be sure, this Court and others have granted motions for class certification based on proposed damages models that measure damages in accordance with the plaintiffs' theory of liability. See, e.g., In re Arris Cable Modem Consumer Litig., 327 F.R.D. 334, 367 (N.D. Cal. 2018) (certifying class based on proposed damages model including detailed proposed survey); In re Twitter Inc. Sec. Litig., 326 F.R.D. 619, 630 (N.D. Cal. 2018) (concluding that "proposed damages model" with a sound methodology satisfied Comcast).

However, those cases are readily distinct from the instant case, where Plaintiffs relied on actual conducted surveys in their first and third motions for class certification and relied on a hypothetical survey in their second motion for class certification.

Like this case, Brazil v. Dole Packaged Foods, LLC, indicates how vast differences can exist between a hypothetical damages model and an actual damages model. 2014 WL 5794873 (N.D. Cal. Nov. 6, 2014). In Brazil, this Court granted the plaintiff's motion for class certification based on the plaintiff's proposed damages model, but then decertified the class because the plaintiff's actual damages model did not deliver what was promised. The Ninth Circuit affirmed this Court's decertification of the class. Brazil v. Dole Packaged Foods, LLC, 660 F. App'x 531, 534-35 (9th Cir. 2016).

Moreover, it makes little sense for the Court to certify a class based on Plaintiffs' or Boedeker's representations about what the second survey was supposed to ask respondents when the second survey that Boedeker actually conducted fails to satisfy Comcast. Cases in which courts have certified classes based on proposed damages models have no applicability to the instant case. Plaintiffs have filed three motions for class certification, two completed damages surveys, and one hypothetical damages survey, but have been unable to produce a compliant damages model—even with the benefit of a Court order and Apple's many briefs and expert reports identifying flaws in Plaintiffs' damages model.

After the Court denied Plaintiffs' first motion for class certification and identified the Comcast flaws in Plaintiffs' damages model on May 8, 2018, the Court gave Plaintiffs an opportunity to file a second motion for class certification on November 8, 2018 as Plaintiffs requested. ECF No. 268. At the October 31, 2018 case management conference, Plaintiffs' counsel stated that after receiving the Court's May 8, 2018 order denying Plaintiff's first motion for class certification, "we immediately worked with our expert following the receipt of that order and following the status conference. And the, the methodology that Your Honor talked about in her order is easily done, has been done as a matter of fact." ECF No. 276 at 6:4-8.

Yet Plaintiffs' November 8, 2018 second motion for class certification relied on a hypothetical Boedeker survey. ECF No. 314 at 2. Then, after Plaintiffs' second motion for class certification was fully briefed on December 20, 2018 and Apple's motion to exclude Boedeker's supplemental declaration and hypothetical survey were fully briefed on December 27, 2018, Plaintiffs served Apple with a second completed survey and the Second Boedeker Report on January 18, 2019. Id. at 3. Accordingly, the Court denied without prejudice Plaintiffs' second motion for class certification, and instructed Plaintiffs to file a third motion for class certification based on the completed second survey and the Second Boedeker Report. Id.

Then, four days before Plaintiffs' third motion for class certification was due, Plaintiffs asked the Court for permission to complete a third survey and to serve a putative Third Boedeker Report. ECF No. 342 at 4. Plaintiffs disingenuously claimed that they discovered the manifestation rate errors in the second survey on February 28, 2019, and thus needed to conduct a third survey. Id. However, Plaintiffs had known about that manifestation rate errors since at least February 7, 2019, when Apple identified the errors in a motion on the public docket. Id. at 4-5. On February 11, 2019, Plaintiffs responded to Apple's February 7, 2019 motion. ECF No. 310. Moreover, Apple's two rebuttal expert reports served on February 15, 2019 identified the same errors and other flaws with the second survey.

Thus, Plaintiffs' late March 1, 2019 request to complete a third survey and third Boedeker report came after Apple's two February 15, 2019 expert reports identified additional flaws in Boedeker's second survey. Id. at 5. At that point, Apple had already had to brief Plaintiffs' first and second motions for class certification and rebut three different surveys: Boedeker's first completed survey, Boedeker's hypothetical survey, and Boedeker's second completed survey. Id. Accordingly, on March 6, 2019, the Court denied Plaintiffs' late request to create yet another survey and damages expert report for the following reasons:

Id.

Plaintiffs violated this Court's March 6, 2019 order and served a Third Boedeker Report and third completed survey on Apple on March 25, 2019, only four days before the March 29, 2019 due date for Apple's opposition to Plaintiffs' third motion for class certification. ECF No. 359 at 1. Apple represents that the Third Boedeker Report was "substantially different than the last version, and it does far more than simply correct `transposed numbers.'" ECF No. 357 at 1. Accordingly, the Court reaffirmed its order denying Plaintiffs' request to serve a Third Boedeker Report. ECF No. 359; see Quevedo v. Trans-Pacific Shipping, Inc., 143 F.3d 1255, 1258 (9th Cir. 1998) (holding that the district court did not abuse its discretion in excluding late expert report produced after the deadline ordered by the district court).

Plaintiffs have repeatedly failed to produce a damages model that satisfies Comcast, despite three motions for class certification, two completed Boedeker surveys, and one hypothetical Boedeker survey, and despite the benefit of the many filings identifying flaws in Plaintiffs' damages models, including the Court's May 8, 2018 order, Apple's three oppositions to class certification, and Apple's rebuttals to two completed Boedeker surveys and one hypothetical Boedeker survey. Allowing Plaintiffs a fourth attempt to produce a Comcast-compliant damages model would be severely prejudicial to Apple. Comcast does not require the Court or Apple to engage in an endless cycle of flawed attempts to produce a satisfactory damages model. Thus, because Plaintiffs have failed to present a damages model that measures "only those damages attributable" to Plaintiffs' theory of liability, Comcast, 569 U.S. at 35, class certification is not warranted.

The Court also observes that when Plaintiffs asked the Court for leave to complete a third survey and Third Boedeker Report on March 1, 2019, Plaintiffs stated that the manifestation risk rate error was the only "clerical error" in Boedeker's second survey and Second Boedeker Report. ECF No. 335 at 2. Then, on March 29, 2019, Apple filed its opposition to the third motion for class certification, in which Apple pointed out the second survey's failure to account for Apple's one-year warranty, among other flaws. Class Cert. Opp. at 4. Only after Apple's March 29, 2019 opposition did Plaintiffs belatedly claim on April 12, 2019 that the warranty omission too was a "clerical error." Boedeker MTE Opp. at 3-4. Plaintiffs continually claim that each successive flaw with their damages model is a "clerical error."

Finally, it is not clear whether Plaintiffs' second survey resolves the flaws this Court identified with Plaintiffs' first damages model. Plaintiffs' first damages model erroneously assumed both that the touchscreen defect will manifest in all iPhones and that the touchscreen defect renders affected iPhones completely inoperable. Davidson III, 2018 WL 2325426, at *22-23. Based on that erroneous model, Plaintiffs estimated economic losses of $323 and $432 per phone. Curiously, Boedeker's second survey generated economic loss estimates of $383 and $413 per phone, which are remarkably similar to Plaintiffs' original estimates—even though Plaintiffs' revised damages model is supposed to reflect that the touchscreen defect will manifest in as few as 0.7% of phones and does not render those phones completely inoperable. The Court also notes that Plaintiffs' damages estimates vary dramatically from Plaintiffs' own expert's estimate of the cost to repair the phone, which is $150. Davidson III, 2018 WL 2325426, at *21 n.11.

As Apple argues, the second survey could mislead respondents and suggest that the touchscreen defect manifests in many more than 0.7% or 5.6% of iPhones. The second survey never discloses these specific risk manifestation risk rates. Instead, the second survey states that the touchscreen defect manifests in "2.86 million out of the 51,139,595 iPhone 6 models that were manufactured" and "125,000 out of the 17,527,727 iPhone 6 Plus models that were manufactured," and uses the term "probability" to refer to the likelihood of manifestation. See Class Cert. Opp. at 8-13. Not surprisingly, Carol Scott, an Apple expert, concluded that the survey design confused respondents. Cheung Decl., Ex. A, Expert Report of Carol Scott ("Scott Report"), ¶¶ 49-52.

Screenshots from Plaintiffs' second survey, one of which is displayed below, show how respondents could be confused. Scott Report, Appx. B at 24. Of the five product choices shown to respondents, all manifest the touchscreen defect, as indicated by the "Yes" selection in the Touchscreen Defect field—even though Plaintiffs' theory of liability is that as few as 0.7% or at most 5.6% of iPhones manifest the defect. Moreover, in 4 out of 5 options, the manifestation of the touchscreen defect renders the phones inoperable 75% to 95% of the time. Id.

In fact, the ten survey screenshots include 50 different product choices, and 43 of those product choices display "Yes" for the touchscreen defect. Then, 34 of those product choices with "Yes" selected include inoperability percentages of 60% or higher. Id. at 16-25. Accordingly, the survey could lead respondents to assume that the defect manifests in a much higher percentage of phones. As Apple's expert explains, respondents "may have inferred that a `Yes' value for `Touchscreen Defect' means that this particular iPhone definitely is one that will (not may) exhibit a touchscreen issue at least some of the time, and that the `probability' they would experience the touchscreen issue was that listed as `Manifestation of Touchscreen Defect Leads to Inoperability.'" Scott Report ¶ 16. Although the survey has no difficulty including inoperability percentages, the survey never discloses that the risk of manifestation of the defect is 0.7% in iPhone 6 phones and 5.6% in iPhone 6 Plus phones. Therefore, it is not entirely clear to the Court that Plaintiffs' second survey has cured the defects in the first survey.

Nevertheless, under Ninth Circuit law, Apple's arguments that the format of the survey and its interpretation render it misleading go to the weight of the Second Boedeker Report rather than its inadmissibility. See Fortune Dynamic, Inc. v. Victoria's Secret Stores Brand Mgmt., Inc., 618 F.3d 1025, 1036 (9th Cir. 2010) (holding that challenges to a survey relating to "the format of the questions or the manner in which it was taken [] bear on the weight of the evidence, not its admissibility") (internal quotation marks and citations omitted). Moreover, even if the Court considers the Second Boedeker Report, Plaintiffs' damages model again fails to satisfy Comcast. The Court therefore DENIES as moot Apple's motion to exclude Boedeker and the opinions in the Second Boedeker Report under Federal Rule of Evidence 702 and Daubert.

Finally, the Court need not consider Apple's various objections to the Supplemental Declaration of Stefan Boedeker, which Plaintiffs submitted with their reply brief in support of their third motion for class certification. See ECF No. 370 (Apple's objections). Even if considered, the statements in the Supplemental Boedeker Declaration do not rectify the deficiencies in Boedeker's second survey or the Second Boedeker Report, which fail to provide a Comcast-compliant damages model.

IV. CONCLUSION

For the foregoing reasons, the Court DENIES Plaintiffs' third motion for class certification and DENIES as moot Apple's motion to exclude Stefan Boedeker and the Second Boedeker Report.

IT IS SO ORDERED.