Summary
In Arris Cable, for example, Judge Koh contrasted the lack of injury in Birdsong and Lassen from claims that the Arris defendants' cable modems were defective because they “did not offer as fast and as reliable an Internet connection as advertised,” claims that constituted injury-in-fact.
Summary of this case from Drake v. Haier U.S. Appliance Sols.Opinion
[Copyrighted Material Omitted] [Copyrighted Material Omitted] [Copyrighted Material Omitted] [Copyrighted Material Omitted] [Copyrighted Material Omitted] [Copyrighted Material Omitted] [Copyrighted Material Omitted] [Copyrighted Material Omitted]
ORDER GRANTING IN PART AND DENYING IN PART MOTION FOR CLASS CERTIFICATION, DENYING MOTIONS TO EXCLUDE RICHARD NEWMAN AND STEVE GASKIN, AND DENYING AS MOOT MOTIONS TO EXCLUDE COLIN WEIR AND KRISTA HOLT Re: Dkt. Nos. 75, 94, 95, 96, 97, 98, 99, 124
LUCY H. KOH, United States District Judge
Plaintiffs bring this putative class action against Defendant Arris International plc ("Defendant"), based on Defendant’s alleged failure to disclose defects with the SB6190 cable modem. Before the Court are Plaintiffs’ motion for class certification, Plaintiffs’ motion to exclude the expert testimony of Krista Holt, and Defendant’s motions to exclude the expert testimony of Richard Newman, Steve Gaskin, and Colin Weir. Having considered the submissions of the parties, the relevant law, and the record in this case, the Court GRANTS IN PART AND DENIES IN PART Plaintiffs’ motion for class certification, DENIES the motions to exclude Newman and Gaskin, and DENIES AS MOOT the motions to exclude Holt and Weir.
I. BACKGROUND
A. Factual Background
"A cable modem is a device that enables a computer to transmit data over a coaxial cable line. The modem is located at the cable subscriber’s home and connects to the cable network to receive and transmit digital information between subscriber-owned devices (such as desktop PCs or routers) and the service provider’s head end or central office, providing Internet connectivity for data and/or voice services." Second Amended Consolidated Complaint ("SACC"), ECF No. 73 ¶ 40. Cable modems’ processing speeds are measured in the number of "bits per second" ("bps") they can process. Id. ¶ 44. " ‘Gbps’ is a measurement in billions of bits per second, ‘Mbps’ is a measurement in millions of bits per second, and ‘Kbps’ is a measurement in thousands of bits per second." Id. "Generally speaking, larger bps units denote higher data speed capability." Id. In addition, cable modems "are equipped with ‘upstream’ and ‘downstream’ channels for data." Id. "The more channels a cable modem has, the more bps it can handle, improving capability." Id.
A modem user’s experience is also affected by the amount of network latency. Id. ¶ 51. "[N]etwork latency refers to delays that occur in data communications over a network. Internet connections with low latency experience only small delay times, while those with high latency suffer from long delays." Id. ¶ 50. High latency can create "bottlenecks" that prevent data from "filing the network pipe" and limit the effective bandwidth of the connection. Id. ¶ 51. "Network latency is measured in milliseconds (‘ms’), where the number of milliseconds represents the amount of time each packet of data is delayed." Id. ¶ 52. According to Plaintiffs, "[f]or a cable modem, typical network latency between a computer and a cable modem ranges from approximately 5 ms to 40 ms. Latency above this range results in connection delays and prevents a cable modem from utilizing its maximum advertised bandwidth." Id. ¶ 53.
Defendant manufactures cable modem hardware for cable service providers and consumers. Id. ¶ 35. Plaintiffs seek to represent a class of consumers who purchased the SURFboard SB6190 cable modem (the "Modem") sold and manufactured by Defendant.
Plaintiffs are citizens of Alabama, Arizona, California, the District of Columbia, Georgia, Hawaii, Illinois, Kansas, Louisiana, Massachusetts, New Jersey, New York, Ohio, Oklahoma, Pennsylvania, Texas, Virginia, and Washington who bought the Modem between 2015 and 2017. Id. ¶¶ 1, 5-34. California Plaintiffs Greg Knowles, Jon Walton and Carlos Reyna bought Modems in 2016, and California Plaintiff Brian Alexander bought a Modem in 2017. Id. ¶¶ 10-13.
Plaintiffs allege that Defendant marketed the Modem as fast and reliable. Id. ¶ 45. Specifically, Plaintiffs allege that Defendant’s website describes the SURFboard line of cable modems as follows:
The Internet brings us together. It lets us experience the wealth of our global community with friends, family, and the entire world. If you want an unmatched Internet experience, look no further than ARRIS SURFboard modems. We’re the industry standard and the world’s standard— with over 175 million ARRIS modems sold.
ARRIS continuously evolves the SURFboard product line to deliver the fastest download speeds available.
When you choose an ARRIS SURFboard, you’re joining a 60-year legacy of innovation from the company that invented digital TV and brought wireless Internet into the home with the first cable modem gateway. The same company that the world’s leading service providers choose to connect millions of people around the world to the internet.
Id.
Plaintiffs also allege that on Defendant’s website and on Amazon.com Defendant emphasized the Modem’s speed and reliability. Id. ¶ 46. Defendant described the Modem as "the First Gigabit Cable Modem" able to achieve "download speeds up to 1.4 Gbps" and as a "DOCSIS 3.0 Cable modem" with "32 download and 8 upload channels." Id. ¶ 47. Plaintiffs allege that the SB6190 modem includes a chip manufactured by Intel Corporation called the "Puma 6" which suffers from network latency defects. Id. ¶¶ 54-55.
Cable modem capability is measured by a telecommunications standard named Data Over Cable Service Interface Specification ("DOCSIS"). The SB6190 modem uses the DOCSIS 3.0 standard.
Plaintiffs contend that in contrast to Defendant’s representations about the Modem’s speed and reliability, the Modem— specifically, the Modem’s Intel Puma 6 chip— has a defect that causes severe network latency. Id. ¶ 55. Plaintiffs allege that Defendant was "aware of at least some" of the Puma 6 latency defects while developing the Modem prior to public release. Id. ¶ 56. Plaintiffs allege that Intel had "experienced issues" with latency with the Puma 6 chip and had communicated these issues to Defendant. Id. ¶ 58.
In support of these contentions, Plaintiffs cite several occurrences since Defendant’s release of the Modem. Plaintiffs allege Defendant began to receive several complaints of high network latency quickly after commencing sales of the Modem to the retail market. Id. ¶¶ 59, 61. Several SB6190 Modem users posted online that they observed erratic latency, while one user reported latency issues to Defendant’s customer service. Id. ¶¶ 61, 63. On March 9, 2016 Defendant received pressure from Comcast Corporation to resolve the latency defect in the Modem. Id. ¶ 66. On April 29, 2016 Defendant released updated firmware for the SB6190, which was denoted as "9.1.93N" and was designed in part to fix the latency defects. Id. ¶ 68.
Plaintiffs allege that 9.1.93N only "partially mitigated" the latency issue and introduced a new latency defect: "DNS packet loss." Id. ¶¶ 69-70. On May 21, 2016 one Modem user on DSLReports.com, an online forum, reported experiencing "sporadic packet errors." Id. ¶ 71. On June 8, 2016, another Modem user on another online forum reported "around 5% packet loss that only stops when I reset my modem again." Id. ¶ 72. A Comcast engineer was alerted to the second post, and the Comcast engineer sent the post to Defendant’s engineers. Id. ¶ 73. On August 8, 2016, another Modem user posted on a blog that he was experiencing packet loss using the 9.1.93N firmware and experiencing "higher latency" and "more jitter" when downgrading back to 9.1.93K, the previous firmware. Id. ¶ 78.
On November 9, 2016, Plaintiff Christopher Stephens posted to DSLReports.com that he was experiencing network latency "through the Modem, to and from the Internet" not only "between the Modem and a user’s local computer." Id. ¶ 79. Plaintiffs allege that Defendant became aware of these complaints on November 11, 2016, and then attempted to "duplicate the latency defects on their own systems." Id. ¶ 81.
On November 16, 2016, Defendant and Intel held a meeting in which they discussed the latency issues along with the DNS packet loss. Id. ¶ 82. Plaintiffs allege Intel’s notes described the DNS packet loss as a side effect of the attempted solution to the "LAN-side ping latency problem." Id. ¶ 83.
On November 29, 2016, a journalist aggregated the reports into a single article on DSLReports.com discussing the latency issues with the Modem. Id. ¶ 85. The article stated that Defendant was "aware of the problem and currently working on a firmware update." Id. Defendant then publicly acknowledged the Modem’s latency problems by telling DSLreports.com that Defendant was "working closely with Intel" on the latency issues. Id. ¶ 86. A follow-up DSLReports.com article concluded from Defendant’s statement that the Intel Puma 6 chip was the "culprit in the jitter and latency problem." Id. ¶ 86.
Another article published on December 3, 2016 quoted Defendant promising to "quickly issue Intel’s firmware updates to resolve any latency." Id. ¶ 91.
On December 1, 2016, Plaintiffs allege that Defendant began its own internal testing of the ping latency defect. Id. ¶ 93. Defendant observed the SB6190 produced much higher and more erratic latency spikes compared to the Arris SB6183, which used a different chip by Broadcom. Id. ¶¶ 95, 101.
On January 10, 2017, Defendant released updated firmware, denoted as "9.1.93T," which Defendant claimed would address an issue with ping latency as well as corresponding DNS packet loss. Id. ¶¶ 104, 105. However, Plaintiffs allege that Defendant’s own testing showed the "DNS latency fixes were ineffective." Id. ¶ 109.
On February 16, 2017, Defendant and Intel had a meeting where Intel presented proposed fixes for the Puma 6 chip’s latency issues. Id. ¶ 111. On March 24, 2017, after additional testing and development by Defendant and Intel, Defendant released updated firmware designed to address the remaining latency defects, denoted as "9.1.93V." Id. ¶¶ 112-13. However, Plaintiffs allege that Defendant’s own testing of 9.1.93V firmware demonstrated the "ineffectiveness of its TCP and HTTP latency fixes," despite Defendant’s claim that the update would resolve that issue. Id. ¶ 115.
Intel continued to work on additional patches to fix the Modem’s continuing latency problems. Id. ¶ 118. On August 11, 2017, an Intel engineer shared a presentation with Defendant’s personnel describing ongoing improvements designed to address several latency issues. Id. ¶ 122. Intel performed benchmarking tests for its proposed fixes that showed vast improvements on SB6190’s latency performance. Id. ¶¶ 123-24. Plaintiffs allege that these improvements demonstrate that Intel was "able to improve the latency performance of the Puma 6 chip" as well as "improve the performance of the cable provider’s network. Id. ¶ 127. On December 28, 2017 Intel provided these tested firmware fixes for the Puma 6 chip to Defendant. Id. ¶ 128. Plaintiffs allege that Defendant "failed to release these Intel latency fixes to consumers." Id. ¶ 128.
Plaintiffs allege that since March 24, 2017, Defendant has failed to release new firmware for the Modem because Intel and Defendant had become distracted by a "widespread and ‘trivial to exploit’ denial-of-service (‘DoS’) security vulnerability" affecting all modems containing the Puma 6 chip. Id. ¶¶ 130-32.
The California Plaintiffs allege that they experienced network latency issues. Specifically, B. Alexander alleges that he "experienced high network latency, packet loss, unreliable Internet connectivity, or other related performance issues." Id. ¶ 153. Knowles experienced the same issues, including "significantly slow Internet performance" during online gaming. Id. ¶ 154. Reyna suffered the same issues as well, including Internet performance issues such as "stuttering and lag" during Reyna’s gaming activities. Id. ¶ 155. Walton experienced the same issues as well as "slow Internet speeds and freezing" during online gaming activities. Id. ¶ 156.
Plaintiffs allege that despite Defendant’s awareness of the multiple latency issues, Defendant "continues to fail to disclose the defect[s] in its marketing of the Modems and refuses to repair or replace the Modems." Id. ¶ 92. Plaintiffs also allege that Defendant represented the Modems as reliable "when, in fact, they have a defect that causes severe network latency." Id. ¶ 239(a). Plaintiffs contend that if they had "known that the Modems suffered from these defects, they would not have purchased the Modems," id. ¶ 179, or they "would have paid substantially less," id. ¶ 267.
Plaintiffs also allege Defendant has engaged in "false and misleading statements in its advertising, marketing and product packaging" of the Modem because Modem users bought a product that was "incapable of achieving" the speeds or channels the SB6190 was marketed to have. Id. ¶¶ 140-41, 146. Specifically, Plaintiffs contend that Defendant advertised the Modems as having 32 download channels and as compatible with Comcast, yet Defendant knew that Comcast provisioned the Modem to use only 24 download channels. Id. ¶¶ 140-42. Thus, the "Modem, when activated on Comcast, therefore cannot under any circumstances experience speeds equal to or greater than 1 Gbps— even if the Modem user subscribes to 1 Gbps service from Comcast." Id. ¶ 141.
Defendant claims that Defendant "made no representations regarding latency" in the Modem and "there is no standard, specification or limit for latency in DOCSIS 3.0 or otherwise in the industry." ECF No. 93 at 7.
B. Procedural History
On March 31, 2017, Plaintiff Carlos Reyna filed a putative class action complaint against Defendant. ECF No. 1. On May 11, 2017, Greg Knowles and 16 other plaintiffs brought a separate putative class action against Defendant. Case No. 5:17-CV-2740, ECF No. 1. On May 25, 2017, Reyna filed an administrative motion to consider whether his case and the Knowles case should be related. ECF No. 14. On June 5, 2017, the Court granted Reyna’s motion to relate the Knowles case. ECF No. 15. On June 30, 2017, the Court granted the parties’ stipulation to consolidate the cases. ECF No. 24. On July 5, 2017, the Court approved the parties’ stipulation to streamline the cases by adjudicating the California law claims before the claims from other states. ECF No. 40 at 5:5-5:17.
All references to the docket refer to Case No. 5:17-CV-1834 unless otherwise specified.
On July 21, 2017, Plaintiffs filed the consolidated amended complaint, which included several additional named plaintiffs. ECF No. 30. Plaintiffs alleged four causes of action under California law on behalf of the named California Plaintiffs and the California Subclass including (1) the California’s Song-Beverly Consumer Warranty Act, Cal Civ. Code § § 1790 et seq. ; (2) the California’s Consumer Legal Remedies Act ("CLRA"), Cal. Civ. Code § § 1750 et seq. ; (3) California’s False Advertising Law ("FAL"), Cal Bus. & Prof. Code § § 17500 et seq. ; and (4) California’s Unfair Competition Law ("UCL") Cal. Bus. & Prof. Code § § 17200 et seq. Plaintiffs individually and on behalf of the Nationwide Class also asserted a claim for unjust enrichment/quasi-contract. Id.
On August 21, 2017, Defendant filed a motion to dismiss and to strike parts of the CAC. ECF No. 33. On September 18, 2017, Plaintiffs filed an opposition. ECF No. 41. On October 5, 2017, Defendant filed a reply. ECF No. 42. On January 4, 2018, the Court granted Defendant’s motion to dismiss with leave to amend and denied the motion to strike. ECF No. 54.
On February 5, 2018, Plaintiffs filed the SACC. ECF No. 59-4. Defendant filed an answer and affirmative defenses to the SACC on February 20, 2018. ECF No. 68.
On May 3, 2018, Plaintiff filed the instant motion for class certification. ECF No. 75 ("Mot."). Plaintiffs seek certification for their UCL and FAL claims on behalf of the following proposed class: "All persons who purchased an Arris SURFboard SB6190 cable modem in California on or after October 1, 2015." Id. at 12. Plaintiffs seek certification of their claims under California’s Song-Beverly Consumer Warranty Act and the CLRA on behalf of the following consumer subclass: "All persons who purchased an Arris SURFboard SB6190 cable modem in California for personal, family, or household purposes on or after October 1, 2015." Id.
Excluded from the class and subclass are "governmental entities, Defendant, any entity in which Defendant has a controlling interest, and Defendant’s officers, directors, affiliates, legal representatives, employees, coconspirators, successors, subsidiaries, and assigns. Also excluded from the Class are any judges, justices, or judicial officers presiding over this matter and the members of their immediate families and judicial staff." Mot. at 12 n.13.
On May 31, 2018, Defendant filed an opposition to the motion for class certification. ECF No. 92 ("Opp’n"). Defendant also filed Daubert challenges to all three of Plaintiffs’ experts, ECF Nos. 94, 96, 98, and motions to strike those experts’ declarations, ECF Nos. 95, 97, 99. Plaintiffs filed oppositions to the Daubert motions on June 14, 2018. ECF Nos. 112, 113, 114. Defendant filed replies in support of their Daubert motions on June 21, 2018. ECF Nos. 118, 119, 120, 121.
On June 22, 2018, Plaintiffs filed a reply in support of their motion for class certification. ECF No. 122 ("Reply"). Plaintiffs also filed a Daubert challenge against Defendant’s expert Krista Holt, ECF No. 124, Defendant filed an opposition, ECF No. 130, and Plaintiffs filed a reply, ECF No. 135.
II. LEGAL STANDARD
Class actions are governed by Rule 23 of the Federal Rules of Civil Procedure. Rule 23 does not set forth a mere pleading standard. To obtain class certification, plaintiffs bear the burden of showing that they have met each of the four requirements of Rule 23(a) and at least one subsection of Rule 23(b). Zinser v. Accufix Research Inst., Inc., 253 F.3d 1180, 1186, amended by 273 F.3d 1266 (9th Cir. 2001). "A party seeking class certification must affirmatively demonstrate ... compliance with the Rule[.]" Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 350, 131 S.Ct. 2541, 180 L.Ed.2d 374 (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, 133 S.Ct. 1426, 185 L.Ed.2d 515 (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. Connecticut Ret. Plans & Tr. Funds, 568 U.S. 455, 465-66, 133 S.Ct. 1184, 185 L.Ed.2d 308 (2013) (quoting Dukes, 564 U.S. at 351, 131 S.Ct. 2541); 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, 133 S.Ct. 1426 (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, 133 S.Ct. 1184. "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.
III. DISCUSSION
Defendant opposes Plaintiffs’ motion for class certification on four main grounds. Specifically, Defendant argues that the proposed class definitions are overbroad and that no named California Plaintiffs have Article III standing. Opp’n at 10-13. Defendant also argues that the named California Plaintiffs are not typical of the putative class or adequate class representatives. Id. at 14-16. Finally, Defendant argues that common questions do not predominate over individual questions. Id. at 16-25. The Court addresses these arguments in turn.
A. The Class Definitions Are Not Overly Broad
Defendant first contends that Plaintiffs’ proposed classes are overbroad because they "sweep into the Class persons who suffered no injury in fact: purchasers who returned their modems to the retailer or ARRIS, received replacements from ARRIS, made successful warranty claims, purchased modems with updated firmware or downloaded firmware updates," or who resold their Modem or purchased a used Modem. Opp’n at 11.
As an initial matter, the Court agrees with Plaintiffs that "the categories of ‘non-injured’ plaintiffs that Arris identifies have actually been harmed in the same way that all class members have been harmed: by paying a price premium for a Modem with the latency defects when a reasonable consumer would expect to receive a Modem without the latency defects." Reply at 2. As the Ninth Circuit has explained, "[w]here plaintiffs are ‘deceived by misrepresentations into making a purchase, the economic harm is the same: the consumer has purchased a product that he or she paid more for than he or she otherwise might have been willing to pay if the product had been labeled accurately.’ " Pulaski & Middleman, LLC v. Google, Inc., 802 F.3d 979, 989 (9th Cir. 2015) (quoting Kwikset Corp. v. Super. Ct., 51 Cal.4th 310, 329, 120 Cal.Rptr.3d 741, 246 P.3d 877 (2011) ). In addition, Plaintiffs here have alleged that "because all Modems had the same latency problems throughout the class period, it is irrelevant whether class members received a replacement SB6190 (which would have contained the same latency issues) or whether class members purchased Modems with updated firmware (since all Modems are automatically updated to the latest firmware, and none of Arris’s firmware versions fixed the latency problems)." Reply at 2. Thus, purchasers who received replacements, bought modems with updated firmware, downloaded firmware updates, or resold their Modem would all have suffered the same economic harm of paying a price premium for a Modem with latency defects.
Next, Plaintiffs assert that "to the extent some class members returned or resold their Modems, Arris conflates the question of whether it is liable to these class members with whether they are owed damages. " Id. Plaintiffs are correct that "[e]ntitlement to restitution is a separate inquiry from the amount of restitution owed under California’s UCL and FAL." Pulaski & Middleman, 802 F.3d at 985. "To state a claim under the UCL or the FAL ‘based on false advertising or promotional practices, it is necessary only to show that members of the public are likely to be deceived.’ " Id. (quoting In re Tobacco II Cases, 46 Cal.4th 298, 312, 93 Cal.Rptr.3d 559, 207 P.3d 20 (2009) ). "This inquiry does not require ‘individualized proof of deception, reliance and injury.’ " Id. at 986 (quoting In re Tobacco II Cases, 46 Cal.4th at 320, 93 Cal.Rptr.3d 559, 207 P.3d 20). "[I]n effect, California has created what amounts to a conclusive presumption that when a defendant puts out tainted bait and a person sees it and bites, the defendant has caused an injury; restitution is the remedy." Id. (quoting Stearns, 655 F.3d at 1021 n.13) (alteration in original). Thus, "restitution is available on a classwide basis once the class representative makes the threshold showing of liability under the UCL and FAL." Id. To the extent that some Modem buyers may be owed $0 in restitution, that is a matter of damage calculations, and the Ninth Circuit has held that "damage calculations alone cannot defeat certification." Yokoyama v. Midland Nat’l Life Ins. Co., 594 F.3d 1087, 1094 (9th Cir. 2010); see Pulaski & Middleman, 802 F.3d at 988 (holding that "Yokoyama remains the law of this court, even after Comcast "); see also Vaquero v. Ashley Furniture Indus., Inc., 824 F.3d 1150, 1155 (9th Cir. 2016) (confirming that the U.S. Supreme Court’s decision in Tyson Foods, Inc. v. Bouaphakeo, __ U.S. __, 136 S.Ct. 1036, 194 L.Ed.2d 124 (2016), did not disturb the Ninth Circuit’s Yokoyama -based precedent).
Plaintiffs also argue that the inclusion of a small number of uninjured individuals does not render the class definition overbroad. Reply at 3. Indeed, the Ninth Circuit rejected an argument similar to Defendant’s in Torres v. Mercer Canyons, Inc., 835 F.3d 1125 (9th Cir. 2016). In Torres, the defendant argued that "a class cannot be certified if it contains both injured and non-injured parties." Id. at 1136. The Ninth Circuit held that this statement "is inaccurate, as even a well-defined class may inevitably contain some individuals who have suffered no harm as a result of a defendant’s unlawful conduct." Id. (citing NEWBERG ON CLASS ACTIONS § 2:3; Messner v. Northshore Univ. HealthSystem, 669 F.3d 802, 823 (7th Cir. 2012) ("[S]ome class members’ claims will fail on the merits if and when damages are decided, a fact generally irrelevant to the district court’s decision on class certification.") ). Moreover, the Ninth Circuit observed that "the possibility that an injurious course of conduct may sometimes fail to cause injury to certain class members" does not defeat predominance. Id. ; see also id. at 1137 ("We conclude that such fortuitous non-injury to a subset of class members does not necessarily defeat certification of the entire class, particularly as the district court is well situated to winnow out those non-injured members at the damages phase of the litigation, or to refine the class definition."). Accordingly, under Pulaski & Middleman and Torres, Defendant’s argument that some class members were not harmed and that the class is overbroad as a result fails.
Defendant also argues that the proposed class is overbroad as it relates to the Comcast compatibility theory of liability because the proposed class is "not limited to Comcast customers with gigabit service but would include customers of other cable operator systems as well as consumers, like Plaintiffs, who were Comcast customers without gigabit service." Opp’n at 11. Defendant argues that "[n]one of these consumers suffered or could suffer any injury relating to these representations regarding Comcast gigabit service even if Plaintiffs prevailed on their theory." Id. The Court disagrees. The fact that not all Modem purchasers were Comcast gigabit service subscribers does not mean that all class members were not harmed by overpaying for a Modem whose compatibility was falsely advertised. For example, several named Plaintiffs in this case testified that they relied on the statement about Comcast compatibility in anticipation of someday upgrading to gigabit service.
Thus, Defendant’s argument that the classes are overly broad fails. The Court next addresses Defendant’s argument that the Plaintiffs lack Article III standing.
B. The Named California Plaintiffs Have Article III Standing
Defendant next argues that "[n]ot even one of the California Plaintiffs has Article III standing." Opp’n at 11. "[A] suit brought by a plaintiff without Article III standing is not a ‘case or controversy,’ and an Article III federal court therefore lacks subject matter jurisdiction over the suit." Cetacean Cmty. v. Bush, 386 F.3d 1169, 1174 (9th Cir. 2004). "[A] plaintiff must demonstrate standing for each claim he seeks to press and for each form of relief that is sought." Davis v. Fed. Election Comm’n, 554 U.S. 724, 734, 128 S.Ct. 2759, 171 L.Ed.2d 737 (2008) (internal quotation marks omitted). In order to have standing under Article III, "a plaintiff must show (1) it has suffered an injury in fact that is [ ] concrete and particularized and [ ] actual or imminent, not conjectural or hypothetical; (2) the injury is fairly traceable to the challenged action of the defendant; and (3) it is likely, as opposed to merely speculative, that the injury will be redressed by a favorable decision." Friends of the Earth, Inc. v. Laidlaw Envtl. Servs. (TOC), Inc., 528 U.S. 167, 180-81, 120 S.Ct. 693, 145 L.Ed.2d 610 (2000) (internal quotation marks omitted); Spokeo, Inc. v. Robins, __ U.S. __, 136 S.Ct. 1540, 1547, 194 L.Ed.2d 635 (2016) (same). In class actions, "standing is satisfied if at least one named plaintiff meets the requirements." Bates v. United Parcel Serv., Inc., 511 F.3d 974, 985 (9th Cir. 2007); see also Stearns v. Ticketmaster Corp., 655 F.3d 1013 (9th Cir. 2011) ("[O]ur law keys on the representative party, not all of the class members, and has done so for many years."). In its January 4, 2018 order granting Defendant’s motion to dismiss, the Court determined that Plaintiffs had adequately established Article III standing because Plaintiffs "alleged that if they had known that the Modems were defective, they would not have purchased the Modems, or they would have paid substantially less for the Modems." ECF No. 54 at 11 (internal quotation marks and citations omitted). The Court reasoned that such an allegation— that Plaintiffs "spent money that, absent [Arris’s] actions, they would not have spent"— was "a quintessential injury-in-fact." Id. (quoting Maya v. Centex Corp., 658 F.3d 1060, 1069 (9th Cir. 2011) ).
Defendant now raises two new challenges to Plaintiffs’ Article III standing. First, Defendant argues that Plaintiffs lack standing to bring a Song-Beverly claim. Opp’n at 11-12. Second, Defendant argues that Plaintiffs lack standing to assert claims based on the alleged latency defects under the UCL, FAL, and CLRA because "Plaintiffs have failed to articulate a plausible legal theory of defect in the consumer fraud or warranty context." Id. at 13. The Court addresses these arguments in turn.
1. Named California Plaintiffs Have Standing to Bring Song-Beverly Claim
Defendant argues that "none of the California Plaintiffs have standing to bring a Song-Beverly claim because ARRIS expressly disclaimed any implied warranties in its Limited Warranty and required any retail purchaser to contact ARRIS at 1-877-466-8646 to take advantage of the Limited Warranty." Opp’n at 11-12. Thus, Defendant concludes that because the Plaintiffs did not contact Defendant with any warranty claims, Plaintiffs lack standing to bring their Song-Beverly claim. Id. at 12. Defendant also argues that Plaintiffs have not alleged all the required elements of an implied warranty claim because the defects that Plaintiffs allege are not so severe as to render the Modems unworkable. Id.
However, "[w]hether the plaintiffs can prove liability is one question; whether they are claiming a sufficient Article III injury is another." Patel v. Trans Union LLC, 2016 WL 6143191, at *7 (N.D. Cal. Oct. 21, 2016). Here, Defendant conflates Article III standing with the merits of Plaintiffs’ claim and the merits of Defendant’s ninth affirmative defense. See Kirola v. City & County of San Francisco, 860 F.3d 1164, 1175 (9th Cir. 2017) ("The district court seems to have improperly conflated Kirola’s standing with whether she would prevail on the merits."); ECF No. 68 at 27 (asserting as ninth affirmative defense that Plaintiffs’ claims are barred "by the failure to make a warranty claim to ARRIS within the applicable Warranty Period contained in the governing Software License, Warranty, Safety, and Regulatory Information"). Specifically, assuming that Defendant’s disclaimer argument is correct, which the Court need not decide here, Plaintiffs’ Song-Beverly claim would fail as a matter of law. Similarly, Plaintiffs’ Song-Beverly claim would also fail if Defendant is correct that the magnitude of the alleged harm is insufficient to state a claim for breach of the implied warranty of merchantability. These arguments go to the merits of Plaintiffs’ Song-Beverly claim and Defendant’s ninth affirmative defense.
By contrast, Defendant does not argue the elements of standing. Defendant does not contend that Plaintiffs have not suffered an injury in fact. Indeed, the Court previously determined that Plaintiffs’ alleged economic injury from overpaying for defective modems constituted an injury in fact. See ECF No. 54 at 11. In addition, with respect to the Song-Beverly claim in particular, Plaintiffs have alleged that the Modems were not fit for the ordinary purposes of fast and reliable Internet connectivity and that Plaintiffs were injured when they received a Modem that did not function as advertised due to the latency defects. SACC ¶¶ 227, 230-32. Nor does Defendant argue that Plaintiffs’ injury is not traceable to Defendant’s actions or that the injury is not redressable. It is thus clear that Defendant’s Song-Beverly Article III standing arguments go to the merits of the claim, not to Plaintiffs’ standing. See Brazil v. Dole Food Co., 935 F.Supp.2d 947, 962 (N.D. Cal. 2013) ("It may ultimately prove true, as defendants claim, that plaintiffs have no actionable claims. However, that is not the same as finding no standing." (quoting Astiana v. Ben & Jerry’s Homemade, Inc., No. 10-4387, 2011 WL 2111796, at *5 (N.D. Cal. May 26, 2011) ) ).
Moreover, Defendant cites no authority for its implied contention that Plaintiffs lack Article III standing if their claim is barred as a matter of law or if they fail to state a claim. To the contrary, the Ninth Circuit has found error where district courts conflate Article III standing and the merits. See Lazar v. Kroncke, 862 F.3d 1186, 1198 (9th Cir. 2017) (finding error where district court "conflated standing with the merits" of a Contracts Clause challenge); Kirola, 860 F.3d at 1175; see also, e.g., Cavka v. SoulCycle Inc., 2017 WL 2906034, at *3 (C.D. Cal. Jan. 30, 2017) (holding that SoulCycle’s argument that its liability waivers were enforceable went to the merits, not to standing); REO Bay Area L.P. v. Brambila, 2011 WL 2118707, at *2 (N.D. Cal. May 26, 2011) ("Defendants appear to argue that Plaintiff’s lawsuit lacks merit. This argument, however, erroneously conflates the establishment of standing with a determination of the merits of a claim."). Indeed, "[w]hether a plaintiff has a legally protected interest (and thus standing) does not depend on whether he can demonstrate that he will succeed on the merits. Otherwise, every unsuccessful plaintiff will have lacked standing in the first place." Claybrook v. Slater, 111 F.3d 904, 907 (D.C. Cir. 1997). Thus, because Plaintiffs have adequately alleged injury in fact, causation, and redressability, Plaintiffs have adequately established standing to pursue their Song-Beverly claim.
2. Named California Plaintiffs Have Standing to Bring UCL, FAL, and CLRA Claims
Defendant next argues that Plaintiffs lack Article III standing to assert their UCL, FAL, and CLRA claims. Opp’n at 13. First, Defendant argues that Plaintiffs’ claims are premised on alleged defects in the Modem, but "this is not a products liability action, and Plaintiffs have failed to articulate a plausible legal theory of defect in the consumer fraud or warranty context." Id. (internal quotation marks omitted). Defendant adds that Plaintiffs "do not allege they bargained for any type of latency feature, and it is well-established that a purchaser of a product who receives the benefit of his bargain has not suffered Article III injury-in-fact traceable to the defendant’s conduct." Id. (internal quotation marks omitted). Finally, Defendant argues that Plaintiffs must allege something more than diminished value to establish injury in fact because there is no evidence of decreased market demand for the Modem. Id. Defendant cites several district court cases in support of these arguments: Azoulai v. BMW of N. Am. LLC, No. 16-CV-589-BLF, 2017 WL 1354781 (N.D. Cal. Apr. 13, 2017); Lassen v. Nissan N. Am., Inc., 211 F.Supp.3d 1267, 1280-84 (C.D. Cal. 2016); and Barakezyan v. BMW of N. Am., LLC, No. CV 16-173-SJO, 2016 WL 2840803 (C.D. Cal. Apr. 7, 2016). Opp’n at 13. These district court cases, in turn, rely on Birdsong v. Apple, Inc., 590 F.3d 955, 961 (9th Cir. 2009).
However, Birdsong, Azoulai, Lassen, and Barakezyan are all distinguishable from the instant case. As an initial matter, Birdsong analyzed statutory standing under the UCL, not Article III standing. See Birdsong, 590 F.3d at 959-62. In addition, in Birdsong, Azoulai, and Lassen, the plaintiffs alleged that a product was defective for lack of an unbargained-for safety feature, yet the plaintiffs’ theory of injury was not based on physical harm, but rather on economic harm. Thus, at the center of the reasoning in Birdsong, Azoulai, and Lassen is a discomfort with no-injury products liability actions being tried as consumer fraud cases. See Birdsong, 590 F.3d at 961 ("Further, the alleged loss in value does not constitute a distinct and palpable injury that is actual or imminent because it rests on a hypothetical risk of hearing loss to other consumers who may or may not choose to use their iPods in a risky manner."); Azoulai, 2017 WL 1354781 at *4 ("Plaintiffs’ attempt to hybridize products liability and consumer fraud doctrines here fails...."); Lassen, 211 F.Supp.3d at 1281 (discussing cases where "courts have rejected for lack of standing or injury attempts to recast no-injury products-liability claims (which are not cognizable) as consumer fraud claims for contract-like economic damages (which may be cognizable") ). In the instant case, by contrast, Plaintiffs allege that the Modem was defective not because of any safety defect, but because it did not offer as fast and as reliable an Internet connection as advertised. Although Defendant is correct that Plaintiffs’ theory of the case relies on four alleged defects in the Modem, see Opp’n at 13, Plaintiffs do not use the term "defect" in the same sense that the term is used in a personal injury products liability case. Rather, Plaintiffs use the term "defect" to identify the causes of the Modem’s alleged subpar performance. The contrasts that Birdsong, Azoulai, and Lassen draw to personal injury tort actions are thus inapposite.
Specifically, in Birdsong, the plaintiffs alleged that Apple’s iPod was defective because it posed "an unreasonable risk of noise-induced hearing loss to its users" because the iPod lacked noise-reduction features and volume control software. Id. at 956-59. The Ninth Circuit held that a diminution in value stemming from a hypothetical risk of hearing loss, which in turn depended on how a consumer used the iPod, was not an actual or imminent harm. Id. at 961. The plaintiffs also argued that they were economically harmed because they were deprived of the benefit of their bargain, but the Ninth Circuit rejected this theory of injury in fact, too. The Ninth Circuit explained that "the plaintiffs admit that Apple provided a warning against listening to music at loud volumes," and that the alleged injury was "premised on the loss of a ‘safety’ benefit that was not part of the bargain to begin with." Id.
In Lassen, the plaintiffs alleged that their vehicles were defective because the vehicles were equipped with keyless fob ignition systems but lacked an auto-off feature, which meant that if a driver forgot to turn off the engine, the engine would continue to run even if unattended. 211 F.Supp.3d at 1271. The plaintiffs did not allege any personal injury stemming from the lack of an auto-off feature, but rather the plaintiffs contended that they would not have bought their vehicles or would have paid less for their vehicles if they had known about the lack of an auto-off feature. Id. at 1273, 1279-80. The district court recognized that "both overpayment and diminution in value are theoretically cognizable injuries-in-fact." Id. at 1280. However, because the plaintiffs’ claim was not based on any malfunction, but only a lack of additional safety features for which the plaintiffs had never bargained, the district court held that the plaintiffs had failed to establish injury in fact. Id. at 1283-84.
Similarly, in Azoulai, the plaintiffs contended that their BMW vehicles were defective because a feature that automatically closed the vehicle doors did not have sensors to detect fingers in the doors before activating. 2017 WL 1354781 at *1. Although the plaintiffs alleged that either they or their family members had been injured when their vehicle doors crushed their fingers, the plaintiffs asserted only economic injuries in their suit. Id. at *2, 4. Relying on Birdsong and Lassen, the district court found that the plaintiffs never bargained for the sensors that the plaintiffs alleged would make the vehicle doors safer. Id. at *6. Thus, in the absence of some malfunction, the plaintiffs had not adequately established an economic injury.
Finally, in Barakezyan, plaintiffs alleged that certain BMW vehicles contained brakes that made "extremely long, high pitched noise[s]." 2016 WL 2840803 at *1. However, the district court found that the plaintiff "has not alleged an economic injury because he has not alleged the Subject Vehicle failed to perform as advertised, is worth less because the brakes do not perform properly, that he purchased the Subject Vehicle and resold it at a lower than market price, or that he identified false representations about the noise the carbon ceramic brakes made by BMW." Id. at *4. In addition, although the plaintiff alleged that the value of his vehicle was diminished, he failed to allege any facts to support that assertion. Id.
Each of these four cases is therefore distinguishable from the instant case. Unlike Birdsong, Lassen, and Azoulai, Plaintiffs’ theory of economic injury is not based on the absence of an unbargained-for safety feature and is not a no-injury products liability case. Rather, Plaintiffs allege that Defendant advertised the Modem as delivering a fast, reliable internet connection, but Defendant in fact sold a Modem that offered a slower and less reliable connection than advertised due to the latency defects. Thus, although Plaintiffs did not explicitly bargain for any particular level of latency, Plaintiffs did bargain for speed and reliability. As a result, Plaintiffs contend that they bought Modems that they would not have otherwise bought, or paid more for Modems than they otherwise would have if they had understood the Modems’ true capabilities. This is precisely the type of economic injury that the Ninth Circuit, this Court, and other district courts routinely find to be a sufficient injury in fact to support Article III standing. See, e.g., Hinojos v. Kohl’s Corp., 718 F.3d 1098, 1101-02, 1104 n.3 (9th Cir. 2013) (finding that allegations that the plaintiff would not have purchased "on sale" merchandise, or would have paid less for it, in the absence of the defendant’s deceptive conduct was sufficient to establish Article III injury in fact); Mazza, 666 F.3d at 595 (finding that allegations that the plaintiffs would not have purchased Honda’s collision mitigation braking system, or would have paid less for it, if its limitations were fully disclosed was sufficient to establish Article III injury in fact); Maya, 658 F.3d at 1069 (calling allegation that the plaintiffs paid more for their homes than they otherwise would have "a quintessential injury-in-fact"); Zeiger v. WellPet LLC, 304 F.Supp.3d 837, 846 (N.D. Cal. 2018) (finding allegations that the plaintiffs would not have purchased the defendant’s product if it had been accurately labeled established an Article III injury in fact); Gustavson v. Wrigley Sales Co., 961 F.Supp.2d 1100, 1129-30 (N.D. Cal. 2013) (same). Moreover, unlike Barakezyan, Plaintiffs here allege that the Modems failed to perform as advertised and were worth less because they did not work properly. See Barakezyan, 2016 WL 2840803 at *4.
Accordingly, the Court concludes that the California Plaintiffs have sufficiently established Article III standing to assert each of their claims. The Court next addresses the requirements of Rule 23(a).
C. Rule 23(a) Analysis
"Parties seeking class certification must satisfy each of the four requirements of Rule 23(a)— numerosity, commonality, typicality, and adequacy— and at least one of the requirements of Rule 23(b)." Briseno v. ConAgra Foods, Inc., 844 F.3d 1121, 1124 (9th Cir. 2017), cert. denied sub nom. ConAgra Brands, Inc. v. Briseno, __ U.S. __, 138 S.Ct. 313, 199 L.Ed.2d 206 (2017).
1. Numerosity
Under Rule 23(a)(1), Plaintiffs must show that "the class is so numerous that joinder of all members is impracticable." Fed.R.Civ.P. 23(a)(1). "In determining whether numerosity is satisfied, the Court may consider reasonable inferences drawn from the facts before it." In re Yahoo Mail Litig., 308 F.R.D. 577, 589-90 (N.D. Cal. 2015). Here, Plaintiffs contend that Defendant likely sold at least 45,000 Modems in California during the applicable time period. Mot. at 13 (citing ECF No. 75-6 ("Schubert Decl.") ¶ 9); see also Exs. 54 (total Modem sales), 141 (California population data). Defendant does not dispute numerosity. See Opp’n at 11 n.4. Because the proposed classes "number[ ] well over forty and joinder would be impracticable, the Court finds that the numerosity requirement is satisfied." In re Yahoo Mail Litig., 308 F.R.D. at 590; see also, e.g., Rannis v. Recchia, 380 Fed.Appx. 646, 651 (9th Cir. 2010) ("In general, courts find the numerosity requirement satisfied when a class includes at least 40 members."); Huynh v. Harasz, 2015 WL 7015567, at *5 (N.D. Cal. Nov. 12, 2015) ("As other district courts have noted, ‘the numerosity requirement is usually satisfied where the class comprises 40 or more members.’ " (quoting Twegbe v. Pharma Integrative Pharmacy, Inc., 2013 WL 3802807, at *3 (N.D. Cal. July 17, 2013) ) ).
2. Commonality
Next, Rule 23(a)(2) states that "[o]ne or more members of a class may sue or be sued as representative parties on behalf of all members only if ... there are questions of law or fact common to the class." Fed.R.Civ.P. 23(a)(2). To satisfy the commonality requirement, Plaintiffs must show that the class members have suffered "the same injury," meaning that class members’ claims must "depend upon a common contention" of such a nature that "determination of its truth or falsity will resolve an issue that is central to the validity of each [claim] in one stroke." Dukes, 564 U.S. at 350, 131 S.Ct. 2541 (quotation marks and citation omitted). Plaintiffs must demonstrate not merely the existence of a common question, but rather "the capacity of a classwide proceeding to generate common answers apt to drive the resolution of the litigation." Id. (quotation marks omitted) (emphasis in original). Nevertheless, the "common contention need not be one that will be answered, on the merits, in favor of the class." Alcantar v. Hobart Serv., 800 F.3d 1047, 1053 (9th Cir. 2015) (internal quotation marks omitted). Additionally, "for purposes of Rule 23(a)(2), even a single common question will do." Dukes, 564 U.S. at 359, 131 S.Ct. 2541 (alteration and quotation marks omitted); see also Mazza, 666 F.3d at 589 ("[C]ommonality only requires a single significant question of law or fact.").
Here, Plaintiffs contend that all Modems contain the same four latency defects caused by the Puma 6 chip. Mot. at 13. Defendant disputes that the Modems contain any defects. See Opp’n at 2 (noting that there is no standard, specification or limit for latency in DOCSIS 3.0). As a result, whether the Modems contain any latency defects is a common issue of fact. Plaintiffs also highlight other common issues of fact, such as whether the latency defects would be perceivable to end users and whether Defendant knew of the defects. See Mot. at 13. Other common questions of law or common mixed questions of law and fact are readily apparent, such as whether Defendant had a duty to disclose the latency defects, whether the latency defects would be material to a reasonable consumer, whether Defendant permissibly disclaimed any implied warranties, and whether the Modems were fit for their typical use. These types of common questions satisfy "the relatively minimal showing required to establish commonality." Phillips v. Ford Motor Co., 2016 WL 7428810, at *7 (N.D. Cal. Dec. 22, 2016) (finding commonality requirement satisfied by "issues such as the existence of a defect and Ford’s knowledge of the defect"), aff’d, 726 Fed.Appx. 608 (9th Cir. 2018); see also Wolin v. Jaguar Land Rover N. Am., LLC, 617 F.3d 1168, 1172 (9th Cir. 2010) (holding commonality "easily satisfy[ied]" where common questions included allegations of the same defect with the same product, the defendant’s knowledge of the defect, the defendant’s obligations under its express and implied warranties, and whether the defendant’s conduct violated various laws). Indeed, Defendant does not dispute commonality. Thus, the Court finds that the commonality requirement is satisfied.
3. Typicality and Adequacy
Rule 23(a)(3) requires plaintiffs to show that "the claims or defenses of the representative parties are typical of the claims or defenses of the class." Fed.R.Civ.P. 23(a)(3). Typicality is satisfied "when each class member’s claim arises from the same course of events, and each class member makes similar legal arguments to prove the defendants’ liability." Rodriguez v. Hayes, 591 F.3d 1105, 1124 (9th Cir. 2010) (quoting Armstrong v. Davis, 275 F.3d 849, 868 (9th Cir. 2001) ). This requirement is "permissive and requires only that the representative’s claims are reasonably co-extensive with those of the absent class members; they need not be substantially identical." Hanlon v. Chrysler Corp., 150 F.3d 1011, 1020 (9th Cir. 1998). Claims that are reasonably co-extensive with the claims of absent class members will satisfy the typicality requirement, but the class must be limited to "those fairly encompassed by the named plaintiff’s claims." Dukes, 131 S.Ct. at 2550. "[C]lass certification is inappropriate where a putative class representative is subject to unique defenses which threaten to become the focus of the litigation." Hanon v. Dataproducts Corp., 976 F.2d 497, 508 (9th Cir. 1992) (citations omitted). "The purpose of the typicality requirement is to assure that the interest of the named representative aligns with the interests of the class." Id.
Rule 23(a)(4) requires "the representative parties [to] fairly and adequately protect the interests of the class." Fed.R.Civ.P. 23(a)(4). In the Ninth Circuit, to test the adequacy of a class representative, a court must answer two questions: "(1) do the named plaintiffs and their counsel have any conflicts of interest with other class members; and (2) will the named plaintiffs and their counsel prosecute the action vigorously on behalf of the class?" Staton v. Boeing Co., 327 F.3d 938, 957 (9th Cir. 2003) (citing Hanlon, 150 F.3d at 1020).
Plaintiffs contend that their claims are typical of the class "because they purchased the same Modems as the class (which contained the same latency defects as the class), experienced the same latency issues as those alleged by the class, and seek to recover damages under the same legal theories as the class." Mot. at 14. Plaintiffs also argue that there is no evidence of any conflict with the class and that the named plaintiffs will vigorously prosecute the action. Id. at 15. Defendant addresses typicality and adequacy together and argues that each named California Plaintiff is neither typical of the class nor an adequate representative. Opp’n at 14-16. The Court addresses each Plaintiff in turn.
a. Knowles
Defendant argues that Knowles is not a typical or adequate class representative because Knowles "is a convicted felon and current registered sex offender." Opp’n at 14. Because "[t]he only evidence Mr. Knowles has presented to support his claim that he experienced poor performance with his modem is his own testimony," Defendant asserts that Knowles is subject to the unique defense of impeachment based on his felony convictions. Id. Defendant’s argument is not persuasive.
A putative class representative’s "credibility may be a relevant consideration with respect to the adequacy analysis" and courts "must be concerned with the integrity of individuals it designates as representatives for a large class of plaintiffs." Marsh v. First Bank of Delaware, 2014 WL 554553, at *9 (N.D. Cal. Feb. 7, 2014) (citations omitted). However, "the ‘most important issue’ remains whether the class representative’s ‘interests are antagonistic to those of the class members.’ " Id. (quoting In re Computer Memories Sec. Litig., 111 F.R.D. 675, 682 (N.D. Cal. 1986) ). "Only when attacks on the credibility of the representative party are so sharp as to jeopardize the interests of absent class members should such attacks render a putative class representative inadequate." Harris v. Vector Mktg. Corp., 753 F.Supp.2d 996, 1015 (N.D. Cal. 2010). More precisely, a district court should find inadequacy "only where the representative’s credibility is questioned on issues directly relevant to the litigation or there are confirmed examples of dishonesty, such as a criminal conviction for fraud." Id.
Knowles does not approach that standard. Knowles’ convictions are from the late 1980s and 1993, did not involve fraud or dishonesty, have no relationship to this case, and his incarceration ended in 2006. See ECF No. 93-12 at 10:2-14 (stating that convictions were for a 1993 sex crime and two burglaries from the late 1980s); Reply at 7 n.6 (stating that Knowles was released from prison in 2006). The convictions, all of which are more than 20 years old, are therefore irrelevant, and are likely inadmissible as well. See Fed.R.Evid. 609(b); White v. E-Loan, Inc., 2006 WL 2411420, at *3 (N.D. Cal. Aug. 18, 2006) (finding class representative adequate despite past theft convictions). Indeed, when a past conviction does not relate to dishonesty— and even sometimes when it does— courts in this circuit typically find that the conviction does not render a putative class representative atypical or inadequate. See Meyer v. Portfolio Recovery Assocs., LLC, 707 F.3d 1036, 1042 (9th Cir. 2012) (holding that district court did not abuse its discretion in finding that class representative’s convictions involving deceptive conduct that were more than 10 years old did not render class representative inadequate); Nevarez v. Forty Niners Football Co., No. 16-CV-7013-LHK, ECF No. 186 at 27, 326 F.R.D. 562, 583, 2018 WL 3830137 (N.D. Cal. July 12, 2018) (finding that convictions for assaults with deadly weapons and DUI convictions that were more than 20 years old did not render class representative inadequate); In re Facebook Privacy Litig., 2016 WL 4585817, at *5-6 (N.D. Cal. Sept. 2, 2016) (finding that two year old felony embezzlement conviction did not render class representative inadequate or atypical); Stemple v. QC Holdings, Inc., 2014 WL 4409817, at *9 (S.D. Cal. Sept. 5, 2014) (finding class representative adequate despite 14 year old conviction for sex with a minor); Marsh, 2014 WL 554553 at *10 (finding that guilty plea to drug charge does not render class representative inadequate). Accordingly, the Court concludes that Knowles is a typical and adequate class representative.
b. Walton
Defendant next argues that Walton is not typical or adequate because he is subject to the unique defense that he sold his Modem on eBay on June 15, 2017, after filing this suit. Opp’n at 14-15. Defendant argues that Defendant was unable to examine Walton’s Modem as a result and so accuses Walton of spoliating evidence. Id. Defendant also contends that because Walton recouped $61.99 for his Modem, he may not be entitled to any damages. Id. at 15.
Taking the last argument first, as the Court explained above, whether Walton is entitled to restitution is a separate question from the amount of restitution to which he is entitled. See Pulaski & Middleman, 802 F.3d at 985. To the extent that Walton may be owed $0 in restitution, that is a matter of damage calculations, and the Ninth Circuit has held that "damage calculations alone cannot defeat certification." Yokoyama, 594 F.3d at 1094; see Pulaski & Middleman, 802 F.3d at 988 (holding that "Yokoyama remains the law of this court, even after Comcast ").
Turning to Defendant’s first argument, "[s]poliation of evidence is the destruction or significant alteration of evidence, or the failure to preserve property for another’s use as evidence, in pending or future litigation." Kearney v. Foley & Lardner, LLP, 590 F.3d 638, 649 (9th Cir. 2009) (internal quotation marks omitted). "Although the Ninth Circuit has not precisely defined when the duty to preserve is triggered, trial courts in this Circuit generally agree that, " ‘[a]s soon as a potential claim is identified, a litigant is under a duty to preserve evidence which it knows or reasonably should know is relevant to the action.’ " Apple v. Samsung Elecs. Co., 888 F.Supp.2d 976, 991 (N.D. Cal. 2012) (quoting In re Napster, Inc. Copyright Litig., 462 F.Supp.2d 1060, 1067 (N.D. Cal. 2006) ). Thus, "[a] party’s destruction of evidence qualifies as willful spoliation if the party has ‘some notice that the [evidence] w[as] potentially relevant to the litigation before [it] w[as] destroyed.’ " Leon v. IDX Sys. Corp., 464 F.3d 951, 959 (9th Cir. 2006) (quoting United States v. Kitsap Physicians Serv., 314 F.3d 995, 1001 (9th Cir. 2002) ).
Sanctions for spoliation arise where a litigant (1) had a duty to preserve evidence in its control; (2) destroyed evidence "with a culpable state of mind"; and (3) a reasonable trier of fact could find that the destroyed evidence was relevant. Apple, 888 F.Supp.2d at 996-97. The Second Circuit and district courts in this circuit have found that a plaintiff’s spoliation of evidence can render a putative class representative inadequate. See Falcon v. Philips Elecs. N. Am. Corp., 304 Fed.Appx. 896, 897 (2d Cir. 2008) ("Falcon’s disposal of the television set presented such a danger [of becoming the focus of litigation]— both with respect to (1) proving a design defect and (2) defending against a charge of spoliation of evidence— and it was far from an abuse of discretion for the District Court to conclude, in light of that danger, that Falcon could not serve as an adequate class representative."); Victorino v. FCA U.S. LLC, 2018 WL 2455432, at *9 (S.D. Cal. June 1, 2018) ("Tavitian will be unable to vigorously prosecute the action while he will be subject to questions arising from the spoliation of evidence and overcoming the mandatory presumption that the lost evidence is relevant and favorable to FCA."); Doyle v. Chrysler Grp. LLC, 2014 WL 7690155, at *3 (C.D. Cal. Oct. 9, 2014) ("Like all class members, Allen will be required to prove all elements of her claim, but unlike other class members, she would also be required to overcome any adverse inference a jury might draw as a result of her sale of the vehicle."), reversed on other grounds, 663 Fed.Appx. 576 (9th Cir. 2016).
Here, Defendant observes that the Court "may impose sanctions" on Walton and his counsel for the sale of the Modem, see Opp’n at 15, but Defendant does not actually move for sanctions. Plaintiffs respond that there is no allegation that Walton acted in bad faith. Reply at 8. Specifically, Plaintiffs state that Walton sold his Modem two weeks before Defendant "communicated that it intended to inspect plaintiffs’ Modems" because the Modem did not suit his needs. Id. (citing Walton Dep., ECF No. 122-8 at 3); see also Opp’n at 14 n.5 (citing June 28, 2017 filing for proposition that Defendant stated early in the litigation that it intended to inspect the Modems).
However, Plaintiffs do not dispute that Walton sold his Modem after filing the instant suit. Thus, at the time that Walton sold his Modem, Walton clearly bore a duty to preserve his Modem, because the Modem itself was obviously at least "potentially relevant" to his allegations that the Modem was defective. See Leon, 464 F.3d at 959. Thus, even if Plaintiffs are correct that Walton did not act in bad faith, sanctions may still be appropriate. See Lofton v. Verizon Wireless (VAW) LLC, 308 F.R.D. 276, 287 (N.D. Cal. 2015) ("The Court need not find bad faith by the offending part before issuing sanctions for destruction of evidence; willfulness or fault can suffice."). Moreover, even if the Court does not actually sanction Walton, the issue of spoliation can still render him an inadequate representative. For example, in Akaosugi v. Benihana Nat’l Corp., 282 F.R.D. 241 (N.D. Cal. 2012), another court in this district denied a motion for spoliation sanctions but nonetheless found that "[t]here is too much at risk ... that issues unique to plaintiff Nguyen, namely those addressed in the briefing on the motion for sanctions, will impede plaintiff Nguyen’s ability to vigorously represent the interests of the class." Id. at 257.
Plaintiffs argue that Defendant is not prejudiced by Walton’s sale of the Modem because Defendant never sought to test any of the other Plaintiffs’ Modems for defects. Id. at 9 (citing ECF No. 122-4 ¶ 2). However, Defendant did inspect the outer casings of the other Plaintiffs’ Modems to see whether Plaintiffs had opened or otherwise altered the Modems, which Defendant contends would have voided the warranties. ECF No. 122-4 ¶ 2; Opp’n at 14 n.5 (explaining purpose of visual inspections). Thus, inspection of Walton’s Modem would have at least been relevant to one of Defendant’s affirmative defenses, if not other issues as well.
Overall, the Court finds that at the time that Walton sold his Modem, he had a duty to preserve the Modem as evidence in the instant litigation. Thus, even in the absence of bad faith, the spoliation issue, which is unique to Walton, risks becoming a focus in the litigation. Accordingly, the Court finds that Walton is not an adequate class representative. See Akaosugi, 282 F.R.D. at 257.
c. Reyna
Defendant argues that Reyna, a software engineer, is not a typical or adequate class representative because some of his testimony allegedly contradicts the contentions of other named Plaintiffs. Opp’n at 15. Specifically, Defendant asserts that Reyna testified that Comcast’s provisioning of fewer than 32 download channels was not material to him because he did not subscribe to Comcast’s gigabit service, which Defendant contends undermines Plaintiffs’ theory of liability based on the alleged misrepresentation of the Modem’s compatibility with Comcast. Id. (citing Reyna Dep., ECF No. 93-13, at 56:14-22; 25:1-26:1). Defendant also argues that Reyna testified that he did not experience any issues with audio and video conferencing as a result of the alleged latency defect. Id. (citing Reyna Dep. at 67:5-22).
Plaintiffs respond that they did not propose Reyna as a class representative on the Comcast compatibility theory, which makes his testimony about the download channels irrelevant. Reply at 8. Plaintiffs’ argument might be persuasive if Plaintiffs had proposed a subclass related to the Comcast compatibility theory and Reyna was not proposed as a representative for that subclass. However, Plaintiffs have not proposed a subclass related to the Comcast compatibility theory. See Reply at 2 n.1. Instead, Plaintiffs frame the Comcast compatibility theory as one of two independent bases to support classwide liability under the UCL, FAL, and CLRA, the first being that the latency defects made Defendant’s representations about speed and reliability misleading, and the second being that Comcast’s refusal to allow the Modem to use all 32 download channels on its network made Defendant’s representation that the Modem was compatible with Comcast misleading. See Mot. at 2 n.2, 20-21. Thus, Plaintiffs appear to envision that Reyna would serve as a class representative for the entire class of California Modem purchasers, but Reyna would not represent that same class on one of the two theories of liability in this case. See Mot. at 20 n.17 (stating that "Plaintiffs seek certification of [the Comcast] issue only on behalf of proposed class representatives Alexander, Knowles, and Walton, who were each unaware of this limitation and who would not have purchased their Modems had they known that they did not support 32 channels or 1.4 Gbps speeds on Comcast.").
Plaintiffs cite no authority for such an arrangement. Nor do Plaintiffs otherwise address the typicality and adequacy issues presented by Reyna only pursuing one of the two theories of liability asserted by the rest of the class. See Rodriguez, 591 F.3d at 1124 (noting that typicality is satisfied when "each class member makes similar legal arguments to prove the defendants’ liability"); Mckinnon v. Dollar Thrifty Automotive Grp., Inc., 2016 WL 879784, at *7 (N.D. Cal. Mar. 8, 2016) (finding named plaintiff inadequate in part because her claim was based on a different legal theory than the rest of the class). Indeed, the fact that Plaintiffs did not propose Reyna as a class representative on the Comcast compatibility theory is effectively a concession that Reyna would not be a typical or adequate representative of other class members with respect to the Comcast compatibility. Similarly, Reyna will be unable to adequately defend the interests of the class because, by Plaintiffs’ own admission, Reyna will not represent the class on one of its two theories of liability.
Accordingly, because Reyna’s testimony about the Comcast compatibility issue conflicts with one of Plaintiffs’ two theories of liability and because Plaintiffs concede that Reyna would not represent the class on one of Plaintiffs’ two theories of liability, the Court finds that Reyna is not a typical or adequate representative.
d. Alexander
Finally, Defendant argues that Alexander is not typical or adequate because Alexander testified that he chose which modem to buy based on speed and the number of download channels. Opp’n at 15-16 (citing Alexander Dep., ECF No. 93-11 at 11:22-14:25). Defendant concludes that "Alexander cannot represent a class of individuals who claim ARRIS engaged in false advertising because he cannot identify any specific ARRIS representations that he relied upon in making his purchase of the SB6190." Id. at 16. Defendant’s argument is not supported by the record. Contrary to Defendant’s representation, Alexander testified that he spent ten minutes at the store comparing different modems’ boxes and, more importantly, Alexander identified specific statements from the Modem’s box upon which he relied in deciding to buy the Modem. See Alexander Dep. at 14:6-16, 32:1-13; Reply at 7. Indeed, the following exchange occurred during Alexander’s deposition:
Q: At the time that you made the purchase, was there anything in particular on the box that you relied on in making the purchase?
A: That the upload and download channels, that, and on the back— it really states how much— implies how much faster and better it is for online gaming and such.
Q: And what are you referring to specifically there?
A: Where it says "high performance online gaming." Where it also says "unbelievable download speeds," and "create a future-proof network capable of download speeds of up to 1.4 gigabytes per second."
Alexander Dep. at 32:1-13. Moreover, in Alexander’s sworn declaration Alexander states that he reviewed the Modem’s packaging, including the statements on the box that the Modem was optimized for "High-Performance Online Gaming" and would yield "unbelievable download speeds." Alexander says in his declaration that he relied on these statements. ECF No. 80-1 ¶ 4. Alexander also states in his declaration that he "believed that the Modem was a fast, reliable modem suitable for online gaming." ECF No. 80-1 ¶ 4. Nothing about these statements renders Alexander an atypical or inadequate plaintiff.
In sum, the Court finds that the proposed classes satisfy the numerosity and commonality requirements of Rule 23(a). In addition, the Court finds that Plaintiffs Knowles and Alexander satisfy the typicality and adequacy requirements, whereas the Court finds that Walton and Reyna are not typical or adequate class representatives. Because all the requirements of Rule 23(a) are thus satisfied, the Court proceeds to analyze Rule 23(b).
D. Rule 23(b) Analysis
1. Rule 23(b)(3) Predominance
Under Rule 23(b)(3), plaintiffs must show "that the questions of law or fact common to class members predominate over any questions affecting only individual members." Fed.R.Civ.P. 23(b)(3). The Rule 23(b)(3) predominance requirement is "even more demanding" than Rule 23(a)’s commonality counterpart. Comcast Corp. v. Behrend, 569 U.S. 27, 34, 133 S.Ct. 1426, 185 L.Ed.2d 515 (2013). Predominance "tests whether proposed classes are sufficiently cohesive to warrant adjudication by representation." Amchem Products, Inc. v. Windsor, 521 U.S. 591, 623, 117 S.Ct. 2231, 138 L.Ed.2d 689 (1997) (citation omitted). The Ninth Circuit has held that "there is clear justification for handling the dispute on a representative rather than an individual basis" if "common questions present a significant aspect of the case and they can be resolved for all members of the class in a single adjudication." Hanlon, 150 F.3d at 1022. As the U.S. Supreme Court recently explained, "[t]he predominance inquiry asks whether the common, aggregation-enabling, issues in the case are more prevalent or important than the non-common, aggregation-defeating, individual issues." Tyson Foods, Inc. v. Bouaphakeo, __ U.S. __, 136 S.Ct. 1036, 1045, 194 L.Ed.2d 124 (2016) (quoting 2 W. Rubenstein, Newberg on Class Actions § 4:49 (5th ed. 2012) ). "When ‘one or more of the central issues in the action are common to the class and can be said to predominate, the action may be considered proper under Rule 23(b)(3) even though other important matters will have to be tried separately, such as damages or some affirmative defenses peculiar to some individual class members.’ " Id. (quoting 7AA C. Wright, A. Miller, & M. Kane, Federal Practice and Procedure § 1778 (3d ed. 2005) ).
Plaintiffs assert that common questions predominate as to the warranty and fraud claims because those claims each turn on "objective questions that will resolve the heart of the litigation for all class members." Mot. at 16. Specifically, as to the Song-Beverly warranty claim, Plaintiffs contend that whether the Modem was suitable for its ordinary purpose will be answered based on classwide proof about the existence and effects of the alleged latency defects. Id. at 16-18. As to the claims under the FAL and fraudulent prong of the UCL, Plaintiffs argue that Defendant’s representations about the Modem’s speed and reliability were "remarkably uniform" across the Modems’ packaging, Defendant’s website, and third party sellers’ websites. Id. at 18-19. Because the representations were uniform, Plaintiffs contend that liability turns on whether a reasonable consumer would be misled by Defendant’s representations. In turn, the "reasonable consumer" standard is subject to common proof. Id. at 19-20. Similarly, with respect to the CLRA claim, Plaintiffs argue that "an inference of common reliance may be applied to a CLRA class that alleges a material misrepresentation consisting of a failure to disclose a particular fact." Id. at 22 (quoting McAdams v. Monier, Inc., 182 Cal.App.4th 174, 178, 105 Cal.Rptr.3d 704 (2010) ). As to the UCL’s unfair prong, Plaintiffs contend that the inquiry is objective and can be shown through common proof. Mot. at 21. Finally, with respect to the UCL’s unlawful prong, Plaintiffs argue that both the Song-Beverly and CLRA claims could serve as predicate acts, and because the Song-Beverly and CLRA claims are susceptible to common proof, so too is the UCL’s unlawful prong. Id. at 21.
Plaintiffs also assert that damages can be determined on a classwide basis because under any of Plaintiffs’ claims, the recovery "would be calculated by subtracting the actual value of the product at the time of purchase from the price paid." Id. at 23. To determine the actual value of the Modems at the time of purchase, Plaintiffs plan to use a choice-based conjoint study. Id. at 23-24.
Defendant challenges predominance on three grounds. First, Defendant argues that Plaintiffs have failed to establish proof of a common defect. Opp’n at 17-19. Second, Defendant asserts that the materiality and reliance elements of the CLRA, FAL, and UCL claims are not susceptible to common proof. Id. at 19-22. Third, Defendant argues that Plaintiffs’ damages model does not match Plaintiffs’ theory of liability and as a result fails under Comcast, 569 U.S. 27, 133 S.Ct. 1426. Opp’n at 22-25. The Court addresses these arguments in turn.
a. Common Defect
Defendant first argues that Plaintiffs’ liability theory is based on four alleged common defects that cause various performance issues, but Plaintiffs have not set forth sufficient evidence that such defects exist. Specifically, Defendant criticizes Plaintiffs’ computer networking expert, Dr. Newman, for failing to conduct any tests of the Modem and failing to suggest an alternate Modem design. Opp’n at 17-18. However, these arguments misapprehend both Plaintiffs’ burden at the class certification stage and the purpose of Newman’s report.
i. Plaintiffs’ Burden
"In order to satisfy the predominance requirement, the plaintiff need not prove the existence of the defect." Edwards v. Ford Motor Co., 603 Fed.Appx. 538, 540 (9th Cir. 2015) (citing Amgen, 568 U.S. at 459, 133 S.Ct. 1184). "Plaintiffs’ position [on the existence of defects] may or may not prevail, but that is a merits question not appropriately addressed at the class certification stage." Just Film, Inc. v. Buono, 847 F.3d 1108, 1122 (9th Cir. 2017). This is because "Rule 23(b)(3) requires a showing that questions common to the class predominate, not that those questions will be answered, on the merits, in favor of the class." Amgen, 568 U.S. at 459, 133 S.Ct. 1184. Here, that means that Plaintiffs’ burden at this stage is not to establish that the alleged common latency defects exist. Instead, Plaintiffs’ burden is to establish that the existence of the defects is subject to common proof, and that such common questions predominate over individual questions. Plaintiffs have carried that burden. Plaintiffs’ theory is that all Modems have the Puma 6 chip, and the Puma 6 chip causes the same four latency defects in all Modems. Whether Plaintiffs are correct that the latency defects exist is a question common to the class. See Amgen, 568 U.S. at 459, 133 S.Ct. 1184 ("The alleged misrepresentations and omissions, whether material or immaterial, would be so equally for all investors composing the class."). "As vital, the plaintiff class’s inability to prove [the existence of the defects] would not result in individual questions predominating. Instead, a failure of proof on the issue of [the existence of the defects] would end the case." Id. at 459-60, 133 S.Ct. 1184.
Fighting this conclusion, Defendant asserts that "it is impossible to determine— on a classwide basis— whether the SB6190 (or any cable modem) causes a consumer to experience performance problems (i.e., slow internet)." Opp’n at 18. Defendant’s expert, Allen Walston, who is also an employee of Defendant, explains that "[t]here are a number of other factors that can increase perceived delay on the Internet, including, but not limited to, a user’s Internet service provider, a user’s Internet service subscription, the number of individuals using the modem, the number and type of devices using the Internet through the modem, the types of applications used, the time of day for predominate use, level of network traffic, CMTS responsiveness, load on intervening servers, neighborhood factors, and load on home network." Walston Report, ECF No. 94-6 ¶ 10. Walston explains that "[o]utside of a laboratory setting where some contributing factors to latency may be isolated or measured, the source of network latency at any one time cannot be ascribed to any individual portion of the network. In other words, it is impossible for a user experiencing delay on the Internet to isolate the delay that each component of his/her network contributes to that delay." Id. ¶ 12. According to Walston, "[t]he plaintiffs in this case appear to have recognized that difficulty and chosen to seize on ARRIS’s and Intel’s own testing to try to prove that some deficiency existed, since they were unable to prove any concrete real-world impact." Id.
In other words, Defendant argues that because it would be difficult to determine whether any particular incidence of slowness that a user experiences is actually attributable to the Modem, as opposed to some other network element, common questions do not predominate. However, the Ninth Circuit’s decision in Wolin defeats this argument. The plaintiffs in Wolin alleged certain Land Rover vehicles were defective because the vehicles had a "defective alignment geometry" that caused the tires to wear more quickly than usual. Wolin, 617 F.3d at 1170. Land Rover argued that common issues did not predominate for the purposes of class certification because "the evidence w[ould] demonstrate that the prospective class members’ vehicles do not suffer from a common defect, but rather, from tire wear due to individual factors such as driving habits and weather." Id. at 1173. The Ninth Circuit rejected this argument, noting that the plaintiffs "assert[ed] that the defect exists in the alignment geometry, not in the tires, [and] that Land Rover failed to reveal material facts in violation of consumer protection laws." Id. Wolin therefore held that "[a]lthough individual factors may affect premature tire wear, they do not affect whether the vehicles were sold with an alignment defect" and thus that "all of [the plaintiffs’] allegations are susceptible to proof by generalized evidence." Id.
The same is true in the instant case. Individual factors may affect class members’ Internet performance, but these individual factors do not affect the ultimate question whether the Modems were sold with latency defects. Like Land Rover in Wolin, what Defendant argues here is "whether class members can win on the merits." Id. Plaintiffs argue that the Modems contain latency defects that cause performance issues. Mot. at 1, 4; ECF No. 76-1 at 33-45 (expert report of Newman identifying a HTTP latency defect, a TCP latency defect, a UDP latency defect, a DNS latency defect, and "instability/packet loss" issues). Defendant responds that there are no defects in the Modems and that performance issues could be caused by any number of other network elements. See Opp’n at 18-19.
The Court need not decide at this stage which party is correct. See Amgen, 568 U.S. at 466, 133 S.Ct. 1184 ("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."). In order to determine whether Rule 23(b)(3)’s predominance requirement is met, the Court need only consider Plaintiffs’ theory of the case and the type of evidence that Plaintiffs will use to prove this theory. Specifically, Plaintiffs allege that the Modems contain four common latency defects, and Plaintiffs point to Defendant’s and Intel’s own testing data as common proof of the alleged defects present in all Modems. Plaintiffs then rely on Newman’s report to interpret the testing data and to explain how the results of the testing data would likely manifest in terms of performance issues that users would notice.
Defendant faults Newman for failing to perform any independent testing of the Modem and for relying on the documents that Plaintiffs’ counsel identified as relevant. Defendant moves to exclude Newman’s testimony under Daubert as a result. See ECF No. 94-10 ("Newman Daubert Mot."). Accordingly, because the Court cites Newman in its analysis, the Court briefly addresses the Daubert challenge to Newman.
ii. Daubert Challenge to Newman
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, 509 U.S. at 589, 113 S.Ct. 2786. An expert witness may provide opinion testimony if: (1) the testimony is based upon sufficient facts or data; (2) the testimony is the product of reliable principles and methods; and (3) the expert has reliably applied the principles and methods to the facts of the case. Fed.R.Evid. 702. In Daubert, the Supreme Court identified four factors that "may" bear on such an analysis, including: (1) whether the theory can be and has been tested; (2) whether the theory has been peer reviewed and published; (3) what the theory’s known or potential error rate is; and (4) whether the theory enjoys general acceptance in the applicable scientific community. Murray v. S. Route Maritime SA, 870 F.3d 915, 922 (9th Cir. 2017) (citing Daubert, 509 U.S. at 593-94, 113 S.Ct. 2786).
However, the Ninth Circuit has acknowledged that these four factors "are not ‘equally applicable (or applicable at all) in every case.’ " Id. (quoting Daubert v. Merrell Dow Pharm., Inc., 43 F.3d 1311, 1317 (9th Cir. 1995) ). "Applicability ‘depend[s] on the nature of the issue, the expert’s particular expertise, and the subject of his testimony.’ " Id. (quoting Kumho Tire Co. v. Carmichael, 526 U.S. 137, 150, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999) ). "A district court may permissibly choose not to examine factors that are not ‘reasonable measures of reliability in a particular case.’ " Id. (quoting Kumho Tire Co., 526 U.S. at 153, 119 S.Ct. 1167).
"The duty falls squarely upon the district court to ‘act as a gatekeeper to exclude junk science that does not meet Federal Rule of Evidence 702’s reliability standards.’ " Estate of Barabin v. AstenJohnson, Inc., 740 F.3d 457, 463 (9th Cir. 2014) (quoting Ellis v. Costco Wholesale Corp., 657 F.3d 970, 982 (9th Cir. 2011) ). However, this duty "is to evaluate the soundness of the expert’s methodology, not the correctness of the expert’s conclusions." Fitzhenry-Russell, 326 F.R.D. at 599-600, 2018 WL 3126385 at *2 (citing Primiano v. Cook, 598 F.3d 558, 564 (9th Cir. 2010) ). 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, 598 F.3d at 564 (citing Daubert, 509 U.S. at 594, 596, 113 S.Ct. 2786). In other words, "[t]he Court has broad discretion and flexibility in assessing an expert’s reliability." Fitzhenry-Russell, 326 F.R.D. at 600, 2018 WL 3126385 at *2 (citing Estate of Barabin, 740 F.3d at 463).
Here, Newman, who holds a bachelor’s degree in mathematics and a master’s degree and Ph.D. in computer science, is well qualified in the area of computer networking. Newman has published more than thirty peer-reviewed papers on computer networking, including on the subject of latency reduction. ECF No. 76-1, ¶¶ 1-2. Newman also is a professor of computer science at the University of Florida. Id. ¶ 6.
Defendant argues that Newman’s opinion is unreliable and should be excluded because Newman did not perform any independent testing on the Modem. Newman Daubert Mot. at 7-8. This argument misconstrues both the law governing expert opinions and the purpose of Newman’s report. Specifically, nothing requires that an expert conduct independent testing in every case. Federal Rule of Evidence 703 provides that "[a]n expert may base an opinion on facts or data in the case that the expert has been made aware of or personally observed." Fed.R.Evid. 703. The U.S. Supreme Court recognized in Daubert that "[u]nlike an ordinary witness, an expert is permitted wide latitude to offer opinions, including those that are not based on firsthand knowledge or observation." 509 U.S. at 592, 113 S.Ct. 2786; see In re NJOY, Inc. Consumer Class Action Litig., 120 F.Supp.3d 1050, 1071 (C.D. Cal. 2015) ("An expert is of course permitted to testify to an opinion formed on the basis of information that is handed to rather than developed by him— information of which he lacks first-hand knowledge and which might not be admissible in evidence no matter by whom presented." (internal quotation marks omitted) ). For example, in Sementilli v. Trinidad Corp., 155 F.3d 1130 (9th Cir. 1998), the Ninth Circuit held that a doctor’s opinion in a slip and fall case was admissible even though the doctor did not personally examine the plaintiff. Id. at 1134. Citing Federal Rule of Evidence 703 and Daubert, the Ninth Circuit explained that the doctor’s "opinions and inferences were based on his review of [the plaintiff’s] medical records, as well as his knowledge, experience, training and education." Id. As such, the Ninth Circuit held that the expert would help the jury understand the evidence and that district court’s exclusion of the doctor’s testimony based on lack of firsthand knowledge was an abuse of discretion. Id.
To be sure, there may be circumstances, depending on the purpose and subject of the expert’s testimony, where independent testing would be necessary to support the expert’s opinion. However, this case is not one of those circumstances. Plaintiffs offer Newman to help the Court and the jury interpret Defendant’s and Intel’s own test results and technical data and to opine on the ways in which those test results would manifest as performance issues. Newman’s role is thus akin to the doctor’s role in Sementilli : to review Defendant’s and Intel’s records and interpret them based on his knowledge, experience, training, and education. See Sementilli, 155 F.3d at 1134; ECF No. 112-3 ("Newman Daubert Opp’n") ("Dr. Newman’s role as an expert in this case, based on his extensive experience in computer networking, is to educate this Court regarding what the extensive, confidential data produced by Arris and Intel concerning the Modem mean."). The fact that Newman relied on the results of tests that Defendant and Intel had already performed rather than running his own tests does not render his opinion unreliable, particularly because Defendant does not dispute the reliability of the underlying data. Of course, Defendant may ultimately persuade the factfinder that the conclusions Newman draws about the real-world effects of Defendant’s and Intel’s test results are wrong, potentially in part because Newman did not have enough information about the environments in which Defendant and Intel ran their tests. See Newman Daubert Mot. at 9-10, 14. However, "the appropriate way to discredit [Newman’s interpretation of the test results] [i]s through competing evidence and incisive cross examination," not exclusion. Murray, 870 F.3d at 925.
Defendant also attacks Newman for basing his opinion on documents that were "cherry-picked" from all of the discovery in this case and "highlighted" for him by Plaintiffs’ counsel. Newman Daubert Mot. at 4, 8-10. This argument is disingenuous. Defendant cites no authority for the proposition that an expert must independently sort through all of the discovery in a case in order to determine the relevant evidence. Such a rule would be incredibly costly and time consuming for the expert, who necessarily would be duplicating the work that the attorneys in the case had already performed. To the extent that Defendant believes that Newman failed to consider relevant evidence of record or formed opinions based on unreliable data, Defendant is free to cross-examine Newman at trial on the foundation for his opinions. See Murray, 870 F.3d at 925.
The Court finds that Defendant’s remaining arguments go to the weight of Newman’s opinions, not to their admissibility. Overall, considering the purpose for which Plaintiffs offer Newman’s testimony, the Court finds that Newman’s testimony would assist the Court and the factfinder in understanding the significance of the technical test result evidence in this case. Defendant’s motion to exclude Newman is DENIED.
iii. Conclusion as to Common Defect Argument
As in Wolin, whether Plaintiffs’ allegation of the existence of the defects is true can be determined on a common basis. Accordingly, Defendant’s argument that there are no common defects does not defeat predominance. See Davidson v. Apple, 2018 WL 2325426, at *20 (N.D. Cal. May 8, 2018) (reaching same conclusion in response to similar argument); Philips v. Ford Motor Co., 2016 WL 7428810, at *14 (N.D. Cal. Dec. 22, 2016) (same), aff’d, 726 Fed.Appx. 608 (9th Cir. 2018).
b. Materiality and Reliance Are Susceptible to Common Proof
Defendant’s next predominance argument is that Plaintiffs are "unable to show that materiality and reliance are susceptible to common classwide evidence to support certification of their proposed CLRA, FAL and UCL classes." Opp’n at 19. As a result, Defendant contends that individual questions of materiality and reliance predominate. In general, the Ninth Circuit has explained that "[u]nder California law, class members in CLRA and UCL are not required to prove their individual reliance on the allegedly misleading statements. Instead, the standard in actions under both the CLRA and UCL is whether ‘members of the public are likely to be deceived.’ " Bradach v. Pharmavite, LLC, 735 Fed.Appx. 251, 254, 2018 WL 2250508, at *2 (9th Cir. May 17, 2018) (quoting Kasky v. Nike, Inc., 27 Cal.4th 939, 951, 119 Cal.Rptr.2d 296, 45 P.3d 243 (2002) ). "For this reason, courts have explained that CLRA and UCL claims are ‘ideal for class certification because they will not require the court to investigate class members’ individual interaction with the product.’ " Id. (quoting Tait v. BSH Home Appliances Corp., 289 F.R.D. 466, 480 (C.D. Cal. 2012) ). The FAL employs a similar objective standard for misrepresentations: whether the public is likely to be deceived. See Ebner v. Fresh, Inc., 838 F.3d 958, 965 (9th Cir. 2016) ("[T]he California consumer protection statutes [including the FAL] are governed by the ‘reasonable consumer’ test. Under this standard, Plaintiff must ‘show that members of the public are likely to be deceived.’ " (quoting Williams v. Gerber Prods. Co., 552 F.3d 934, 938 (9th Cir. 2008) ) ); Stearns, 655 F.3d at 1020 ("[T]o state a claim under either the UCL or the false advertising law, based on false advertising or promotional practices, it is necessary only to show that members of the public are likely to be deceived." (quoting In re Tobacco II Cases, 46 Cal.4th at 213, 93 Cal.Rptr.3d 559, 207 P.3d 20) ).
Thus, "[e]ach statute allows plaintiffs to establish materiality and reliance (i.e., causation and injury) by showing that a reasonable person would have considered the defendant’s representation material." Fitzhenry-Russell v. Dr. Pepper Snapple Grp., Inc., 326 F.R.D. 592, 612, 2018 WL 3126385, at *15 (N.D. Cal. June 26, 2018) (quoting In re ConAgra Foods, Inc., 90 F.Supp.3d 919, 983 (C.D. Cal. 2015), aff’d sub nom. Briseno v. ConAgra Foods, Inc., 844 F.3d 1121 (9th Cir. 2017), and aff’d sub nom. Briseno v. ConAgra Foods, Inc., 674 Fed.Appx. 654 (9th Cir. 2017) ). UCL, CLRA, and FAL claims that rely on a theory of fraudulent omission are also susceptible to classwide proof because under California law, reliance can be inferred when an omission is material, and whether an omission is material is determined using a "reasonable consumer" standard. See Daniel v. Ford Motor Co., 806 F.3d 1217, 1225 (9th Cir. 2015); McAdams v. Monier, Inc., 182 Cal.App.4th 174, 178, 105 Cal.Rptr.3d 704 (2010).
Here, Defendant argues that Plaintiffs’ case "is centered around the concept of a ‘latency’ defect," but Defendant "made absolutely no representations regarding latency at all in connection with the SB6190 (and certainly not the 4 specific types of latency cited by Plaintiffs’ technical expert)." Id. at 20. As an initial matter, Defendant does not explain why the absence of any representations about latency would be a predominance problem. If, as Defendant asserts, there is a mismatch between the alleged defects and the alleged misrepresentations, individual issues would not predominate; instead, Plaintiffs’ claims would fail on a classwide basis. See Amgen, 568 U.S. at 459-60, 133 S.Ct. 1184; see also Briseno v. ConAgra Foods, Inc., 674 Fed.Appx. 654, 656-57 (9th Cir. 2017) (holding that whether a reasonable consumer would consider the challenged label misleading is a merits issue that "do[es] not bear on predominance").
Even if Defendant’s argument does go to predominance, Defendant’s argument still fails because it misunderstands Plaintiffs’ theory of the case. Plaintiffs’ theory is that the four types of latency defects cause performance issues that include decreased Internet speeds and glitches in online gaming, video streaming, and audio and video conferencing. See SACC ¶¶ 51-52; Mot. at 1-2, 4-9; Newman Report ¶¶ 90, 169-76. Thus, Plaintiffs contend that Defendant’s statements about speed and reliability were materially misleading. See, e.g., SACC ¶ 239, 257, 265; Reply at 12-13. Moreover, Plaintiffs allege that the failure to disclose the latency defects was a fraudulent omission. See Mot. at 20 ("Here, plaintiffs allege that Arris never disclosed to consumers ... that the SB6190 suffered from widespread network latency problems."); Reply at 12 (similar). The absence of any representations about latency in particular does not mean that individual issues predominate.
In addition, Defendant argues that consumer reliance cannot be inferred for the class because Plaintiffs have not shown that the alleged misstatements were material to the class. Specifically, Defendant argues that the putative class members were not exposed to uniform misrepresentations during the class period. Opp’n at 21-22. Defendant stresses that it "did little online advertising and no print or other media advertising at all." Id. at 21. If the alleged misrepresentations were contained in advertisements, then Defendant’s limited advertising might be relevant to the predominance analysis. See, e.g., Mazza, 666 F.3d at 596 (holding that "the limited scope of th[e] advertising makes it unreasonable to assume that all class members viewed it"). However, Defendant’s limited advertising is irrelevant in the instant case because the misrepresentations upon which Plaintiffs allegedly relied were on the Modem’s packaging, Defendant’s website, and the Modem’s product pages on retailers’ websites, not in advertisements. See Davidson, 2018 WL 2325426 at *19 (distinguishing Mazza from facts of Davidson, where the alleged omissions were on product packaging); Guido v. L’Oreal, USA, Inc., 2013 WL 3353857, at *11 (C.D. Cal. July 1, 2013) (same); SACC ¶¶ 10 (identifying packaging statements upon which B. Alexander relied), 11 (identifying statements on Amazon.com’s Modem product page upon which Knowles relied), 12 (identifying statements on Amazon.com’s Modem product page and the Modem’s packaging upon which Reyna relied), 13 (identifying statements on Defendant’s website upon which Walton relied), 45-49 (discussing statements and including photos of product packaging); Exhs. 130-31 to Kahan Decl., ECF Nos. 85-61, 85-63 (layout for Modem box). Indeed, Defendant concedes that it made representations about the Modem "on or in the box packaging, on the ARRIS website and in information provided to retailers for their product pages." Opp’n at 22.
Plaintiffs have offered evidence that the representations on the product’s packaging were substantially similar throughout the class period, and in turn that Defendant’s website and third party retailers’ websites contained statements that were the same as or similar to the statements on the packaging. See, e.g., Ex. 7 to Kahan Decl., ECF No. 78-5 at 31:5-32:3, 140:5-10 (excerpts from deposition of Jennifer Dale Hays stating that Defendant provided Amazon with language for the product page and that the language was the "same content that was used in other places"); id. at 162:1-16 (stating that primary message of packaging did not change throughout class period); Ex. 129 to Kahan Decl., ECF No. 85-59 (image of Defendant’s website from relevant time period). Such websites often included a picture of the product packaging as well. Defendant does not dispute that the language on the packaging, Defendant’s website, and third party retailers’ websites was substantially similar.
Finally, Defendant contends that the Court would be required "to conduct individualized inquiries to discover which class members read the box and warranty information." Opp’n at 22. However, Defendant offers no authority in support of this contention. As discussed above, the Ninth Circuit, this Court, and other courts have held that such an individualized inquiry is not required under California law. See, e.g., Bradach, 735 Fed.Appx. at 255, 2018 WL 2250508 at *2 (stating that CLRA and UCL claims do "not require the court to investigate class members’ individual interaction with the product" (quoting Tait, 289 F.R.D. at 480) ); Fitzhenry-Russell, 326 F.R.D. at 612-13, 2018 WL 3126385 at *15 (rejecting argument that inquiry into how each consumer interpreted an allegedly misleading claim was required); Davidson, 2018 WL 2325426 at *19 (rejecting argument that class members were not uniformly exposed to alleged omissions on product packaging); Butler v. Porsche Cars N. Am., Inc., 2017 WL 1398316, at *11 (N.D. Cal. Apr. 19, 2017) (finding that "all class members [who] purchased the product ... were necessarily exposed to the defendant’s omissions [on the packaging] prior to purchase in the same way" (internal quotation marks omitted) ).
Accordingly, the Court concludes that common questions predominate over individual questions with respect to the materiality and reliance elements of Plaintiffs’ CLRA, UCL, and FAL claims.
c. Plaintiffs’ Damages Model Satisfies Comcast
Defendant’s final predominance argument is that Plaintiffs’ damages model fails to satisfy the requirements set forth in Comcast . See Opp’n at 22-25. 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), the U.S. Supreme Court has held that the 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, 133 S.Ct. 1426. The damages model "must measure only those damages attributable to" the plaintiff’s theory of liability. Id. If the plaintiff does not offer a plausible damages model that matches her theory of liability, "the problem is not just that the Court will have to look into individual situations to determine the appropriate measure of damages; it is that Plaintiffs have not even told the Court what data it should look for." In re MyFord Touch Consumer Litig., 2016 WL 7734558, at *15 (N.D. Cal. Sept. 14, 2016) ("MyFord Touch ").
As discussed above, Plaintiffs’ first theory of liability is that the Modem contained latency defects that caused performance issues including slowed Internet connection speeds and glitches while streaming video, playing online games, and using video and audio conferencing applications. Plaintiffs contend that Defendant’s representations about fast and reliable Internet connections and Defendant’s failure to disclose the latency defects were therefore misleading. Plaintiffs’ second theory of liability relates to the Modem’s compatibility with Comcast’s network. Plaintiffs allege Defendant knew that Comcast would not allow the Modem to use all 32 download channels, which meant that the statement about compatibility with Comcast was misleading. Plaintiffs allege that they were injured by paying more for the Modem than they otherwise would have if they had known of the latency defects or the Comcast compatibility issue. SACC ¶¶ 51-52, 239, 257, 265; Mot. at 1-2, 4-9, 20; Reply at 12-13. Plaintiffs thus assert that for each of their claims, "class members’ recovery would be calculated by subtracting the actual value of the product at the time of purchase from the price paid." Mot. at 23.
Plaintiffs’ proposed damages model is a "choice-based conjoint study ... combined with an economic analysis based on this conjoint study." Id. One of Plaintiffs’ damages experts, Steven Gaskin, explains that "[t]he general idea behind conjoint analysis is that the market value for a particular product is driven by features or descriptions of features embodied in that product." ECF No. 77-1 ("Gaskin Report") at ¶ 13. Survey respondents are therefore asked to choose between different sets of product attributes, the responses are aggregated, and statistical methods are then used to determine the value (often termed "partworth") that consumers attach to each specific attribute. Id. at ¶¶ 13, 16-21, 44-45, 48-54; In re NJOY, Inc. Consumer Class Action Litig., 120 F.Supp.3d 1050, 1073 (C.D. Cal. 2015) (discussing conjoint analysis).
In the instant case, Gaskin proposes to ask survey respondents to choose between hypothetical modems that differ in price, brand, download and upload speed, compatibility with different cable network providers, warranty terms, security vulnerabilities, and level of latency. Gaskin Report ¶ 40. Gaskin defines three different levels of latency: widespread latency problems with normal Internet use, latency problems only when playing online games, and few or no latency problems. Id. Widespread latency is defined as "[e]ffects from latency result in frequent delays when loading web pages, playing online games, and using audio and video applications." Id. Latency specific to online games is defined as "[e]ffects from latency result in less responsible online games. Other applications are unaffected." Id. Few latency problems is defined as "[f]ew or no effects from latency. Achieved modem performance is consistent and up to potential." Id.
With respect to compatibility with different networks, Gaskin defines the three options as (1) compatible with all major cable internet providers except Verizon, AT & T, and Centurylink; (2) compatible with all major cable internet providers, but the maximum speed on Comcast is 1029 DL, 245 UL, even if the modem is designed for more; and (3) compatible with all major cable internet providers. Id.
Each survey respondent will be shown twelve "choice tasks," where each "choice task" involves choosing between "three different, hypothetical modems that will be described by the combinations of levels of the features" identified above. Id. ¶ 41. For each choice task, respondents will be asked, "If these were your only options and you had to choose a modem, which modem would you choose?" Id. ¶ 42. In addition, "[f]or each choice task, respondents will also be presented with a second question that reads, ‘Given your knowledge of the market, would you actually be willing to buy the modem that you chose above?’ " Id. (emphasis removed).
Gaskin states that, based on these survey inputs, the conjoint analysis will yield estimates of the "reduction in market value for an SB6190 Modem with a reduction in performance due to latency problems compared to an otherwise identical SB6190 Modem with performance as promised." Id. ¶ 46. The same analysis would apply to the Comcast compatibility feature. After determining the reduction in market value, Plaintiffs can calculate "the difference between the prices customers paid and the value of the [Modem] they bought— in other words, the price premium attributable to" the alleged misstatements and omissions. See Brazil v. Dole Packaged Foods, LLC, 660 Fed.Appx. 531, 534 (9th Cir. 2016) (internal quotation marks omitted). It is well-established that the "price premium attributable to" an alleged misrepresentation on product labeling or packaging is a valid measure of damages in a mislabeling case under the FAL, CLRA, and UCL. See id. ; see also Werdebaugh v. Blue Diamond Growers, 2014 WL 2191901, at *22 (N.D. Cal. May 23, 2014) ("The proper measure of restitution in a mislabeling case is the amount necessary to compensate the purchaser for the difference between a product as labeled and the product as received."). Accordingly, Plaintiffs assert that Gaskin’s proposed conjoint analysis satisfies Comcast’s requirement that a damages model must "measure only those damages attributable to" the plaintiff’s theory of liability, i.e. the amount Plaintiffs overpaid due to the misrepresentations and omissions related to the latency defects, those defects’ effects on performance, and the misstatement about Comcast compatibility. Mot. at 23; Reply at 14-15.
Defendant advances two sets of arguments against Gaskin’s proposed conjoint analysis. First, Defendant argues that the conjoint analysis fails to satisfy Comcast because it "fails to properly define the attribute of latency at issue in this case," and thus the measure of damages is not limited to the theory of liability. Opp’n at 24. Second, Defendant contends that the "conjoint design is flawed in at least several significant ways" that render it inadmissible under Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993), and explicitly references its motion to exclude Gaskin’s opinion testimony under Daubert . Opp’n at 23-24 (referencing ECF No. 96-4 ("Gaskin Daubert Mot.") ). The Court addresses these arguments in turn.
i. Whether the Conjoint Analysis Properly Defines the Alleged Latency Defects
Defendant argues that Plaintiffs’ proposed conjoint study does not measure only those damages attributable to Plaintiffs’ theory of liability because the conjoint survey "fails to properly define the attribute of latency at issue in this case." Opp’n at 24. Specifically, Defendant contends that the conjoint survey "would measure consumer[s’] reactions to ‘slow Internet’ generally." Id. at 23. Defendant also argues that the "few or no latency problems" option is not realistic because a modem is only one part of a network and so a modem provider could never promise problem-free connectivity. Id. Defendant adds that "[t]he proposed descriptions for latency were not based on the experiences described by the testimony of the Plaintiffs." Id. at 24. Separately, Defendant’s expert, Krista Holt, opines that (1) any latency is unlikely to be actually perceptible by users, and (2) the defined latency levels are "vaguely phrased and entirely subjective." ECF No. 97-6 ("Holt Report") at 24-25, 27-29. However, Defendant does not otherwise develop this definitional argument in its Opposition or more precisely identify why the conjoint study’s definition of latency is problematic in the context of Comcast’s requirements. See Opp’n at 23-25.
Plaintiffs respond that the latency level definitions in Gaskin’s survey "measure the negative performance effects directly resulting from the Modem’s latency defects." Reply at 14. Specifically, Gaskin explains several possible effects of latency, including "slow web browsing and file transfer, less responsive online gaming, glitches in video and audio streaming, and/or delays in [v]ideo and [a]udio [c]onferencing." Gaskin Report ¶ 40. Gaskin then defines the highest level of latency in the conjoint survey as "widespread latency problems with normal Internet use," which include "frequent delays when loading web pages, playing online games, and using audio and video applications." Id. Plaintiffs point out that these definitions are consistent with Newman’s opinion of the effects of the alleged latency defects. Reply at 14-15. Newman opines that the levels of latency evidenced in Defendant’s and Intel’s own testing would "have manifested to end users in the form of slow web browsing and file transfers; less responsive online gaming; glitches in video and audio streaming; and delays in video and audio conferencing." Newman Report ¶ 157; see also id. ¶¶ 158 (opining that a typical user may experience slow web browsing and file transfers resulting from the HTTP, TCP, and DNS latency defects), 165 (opining that a typical user may experience less responsive online gaming resulting from the UDP latency defect), 168 (opining that latency levels "could easily affect multiplayer online gaming, resulting in noticeable and frustrating delays"), 172 (opining that the "buffer bloat" issue "would likely affect Modem users who attempted to watch video or audio streams"), 177 (opining that the latency defects "resulted in some significant real-world effects on typical Internet users" that would likely compound as more users connected to the Internet).
In other words, the dispute between Defendant and Plaintiffs appears to center around two factual issues: whether users would perceive the effects of the alleged latency defects at all and, if so, whether the conjoint survey’s definition of the highest level of latency accurately captures the magnitude of those effects. As a result, Defendant’s objections to Plaintiffs’ conjoint survey present merits issues about the accuracy of the conjoint survey and the amount of damages, not a Comcast predominance issue.
In Comcast, the plaintiffs brought a class action antitrust suit on behalf of Comcast subscribers. The plaintiffs "proposed four theories of antitrust impact." 569 U.S. at 31, 133 S.Ct. 1426. The district court accepted one theory of antirust impact as capable of classwide proof but rejected the rest. Id. However, the expert’s damages model calculated antitrust impact generally; the model "did not isolate damages resulting from any one theory of antitrust impact." Id. at 32, 133 S.Ct. 1426. The U.S. Supreme Court held that at the class certification stage, "any model supporting a plaintiff’s damages case must be consistent with its liability case." Id. at 35, 133 S.Ct. 1426 (internal quotation marks omitted). Thus, the U.S. Supreme Court observed that "[i]t follows that a model purporting to serve as evidence of damages in this class action must measure only those damages attributable to that theory. If the model does not even attempt to do that, it cannot possibly establish that damages are susceptible of measurement across the entire class for purposes of Rule 23(b)(3)." Id. In the absence of any other damages model, the U.S. Supreme Court held that the plaintiffs had not established predominance. Id. at 34, 133 S.Ct. 1426.
Thus, this Court and other courts in this district have found that damages models fail under Comcast where the model is not consistent with the liability case. In Davidson, for example, this Court found that the liability theory was based on the failure to disclose an approximately 5% risk of a touchscreen defect manifesting in an iPhone, but the damages model measured how much consumers valued an iPhone in which the defect was certain to manifest. 2018 WL 2325426 at *22-23. In Opperman v. Kong Technologies, Inc., 2017 WL 3149295, at *12 (N.D. Cal. July 25, 2017), the district court rejected a conjoint survey that proposed to measure consumers’ valuation of "privacy" and "security" generally, rather than the specific features that formed the basis of the liability theory. In MyFord Touch, the district court declined to certify claims to the extent that they sought types of damages for which no damages model had been proposed. 2016 WL 7734558 at *16 (declining to certify claims to the extent they sought incidental or consequential damages due to "the lack of a model to calculate incidental or consequential damages in a manner amenable to class treatment").
Unlike in Comcast, Davidson, Opperman, and parts of MyFord Touch, where the damages models did not track the theories of liability, here "Plaintiffs propose measuring damages that are directly attributable to their legal theory of the harm." Just Film, 847 F.3d at 1121. Specifically, here the theory of liability is that latency defects in the Modem cause various performance issues, and the conjoint survey attempts to measure how much less consumers would pay for a Modem with those performance issues. To the extent that Defendant asserts that the conjoint survey overstates the magnitude or frequency of the performance issues, that is a merits argument about the proper amount of damages, not a mismatch between Plaintiffs’ damages model and theory of liability. Cf. MyFord Touch, 2016 WL 7734558 at *19 (declining to resolve factual dispute about whether damages theory properly accounted for a certain repair that mitigated damages— and thus affected the amount of damages— where the court had already established that the damages theory was consistent with the liability case and measured only damages attributable to the liability theory). Moreover, even though Defendant appears to have a strong argument that the "widespread latency problems" definition overstates the magnitude or frequency of the Modem’s performance issues, that flaw would not result in a failure to show that damages can be calculated on a classwide basis, nor would it result in individual damages issues predominating. See Lambert v. Nutraceutical Corp., 870 F.3d 1170, 1182 (9th Cir. 2017) ("Uncertainty regarding class members’ damages does not prevent certification of a class as long as a valid method has been proposed for calculating those damages.").
Thus, the Court concludes that Plaintiffs’ proposed conjoint survey satisfies Comcast . However, whether Gaskin’s proposed conjoint survey is sufficiently reliable from a methodological standpoint— and therefore admissible under Daubert — is a different issue from whether the conjoint analysis satisfies Comcast . See In re ConAgra Foods, 90 F.Supp.3d at 946 ("Admissibility [under Daubert ] turns on whether [an expert’s] methodology is sufficiently reliable; whether it satisfies Comcast and shows that a class should be certified is another question altogether."), aff’d sub nom. Briseno v. ConAgra Foods, Inc., 844 F.3d 1121 and 674 Fed.Appx. 654. Accordingly, the Court next analyzes Defendant’s Daubert challenge to Gaskin, because if Gaskin’s opinion testimony regarding his proposed conjoint analysis is found to be inadmissible, then Plaintiffs will lack a damages model and a class could not be certified.
ii. Daubert Arguments
In its motion to exclude Gaskin’s opinion testimony regarding his proposed conjoint analysis, Defendant does not appear to dispute "that conjoint analysis is a well-accepted economic methodology." In re Dial Complete Mktg. & Sales Practices Litig., 320 F.R.D. 326, 331 (D.N.H. 2017); see Miller v. Fuhu Inc., 2015 WL 7776794, at *21 (C.D. Cal. Dec. 1, 2015) (stating that "numerous courts ... have accepted" conjoint analyses "as reliable methodologies for calculating price premiums on a classwide basis in consumer class actions"); Odyssey Wireless, Inc. v. Apple Inc., 2016 WL 7644790, at *9 (S.D. Cal. Sept. 14, 2017) ("Further, a conjoint analysis is a generally accepted method for valuing the individual characteristics of a product."); Kurtz v. Kimberly-Clark Corp., 321 F.R.D. 482, 551 (E.D.N.Y. 2017) (stating that conjoint analyses "have previously been approved in consumer class actions as reliable methodologies available for calculating the price premium attributable to a product characteristic"). Nor does Defendant attack Gaskin’s qualifications as an expert.
Instead, Defendant asserts that the particular way in which Gaskin proposes to conduct the consumer surveys for his conjoint analysis renders the survey unreliable and inadmissible. Gaskin Daubert Mot. at 9-10. The Ninth Circuit has stated that as a general matter, "[c]hallenges to survey methodology go to the weight given the survey, not its admissibility." Wendt v. Host Int’l, Inc., 125 F.3d 806, 814 (9th Cir. 1997); see also Fortune Dynamic, Inc. v. Victoria’s Secret Stores Brand Mgmt., Inc., 618 F.3d 1025, 1038 (9th Cir. 2010) (holding that criticisms that "go to issues of methodology, survey design, reliability, ... [and] critique of conclusions" go to the weight of the survey rather than its admissibility (internal quotation marks omitted) (alterations in original) ). Here, Defendant argues that Gaskin’s conjoint survey is flawed because (1) latency is "too complex, ambiguous, and individualized" to be the subject of any conjoint study; (2) the study fails to reliably measure the specific latency defects alleged in this case; and (3) the study does not match Plaintiffs’ theory of liability. The Court addresses these arguments in turn and finds that none of them are persuasive.
A. Latency is an Acceptable Attribute for Conjoint Analysis
Defendant argues that latency as a concept is an inappropriate attribute to test using a conjoint survey, no matter how the survey is designed. Gaskin Daubert Mot. at 10-12. Defendant offers several reasons why this is so. First, Defendant contends, as it did above in its other predominance argument, that "the actual delay perceived by the consumer is the cumulative result of every part of the network from the data source to the user." Id. at 11. Thus, "[i]n a real world setting, users lack the ability to discern the specific cause of an episode of latency, i.e., which component of the network causes the user to experience slow Internet." Id. Defendant then concludes that "[b]ecause consumers do not have the capability to readily attribute network latency to the performance of cable modems or any of the myriad of other factors that can contribute to network latency, it is inappropriate to include ‘latency’ as an attribute in the choice tasks for the purchase of cable modems." Id. at 11-12. According to Defendant, this is because a "conjoint study should only include attributes which may impact the consumer’s overall judgment towards the focal product." Id. at 11.
Defendant’s conclusion does not follow from its premise. Whether a user is able to discern if any particular incidence of degraded performance is attributable to the alleged defects in the Modem is a separate issue from (1) whether the Modem in fact suffers from latency defects that cause performance issues, as Plaintiffs allege, and (2) whether a user understands what latency is and what its effects on performance are, such that a survey would be able to assess how much a user values different latency levels. Here, Gaskin’s conjoint survey, in reliance on Newman’s opinion, translates the alleged latency defects into the types of performance effects that users can understand. Moreover, Defendant does not dispute that consumers’ decisions may be influenced by the performance effects that Plaintiffs allege— Defendant instead disputes that any effects are in fact as severe as Plaintiffs allege. Thus, a user’s inability to correctly diagnose the cause of a particular incidence of latency does not render latency per se an inappropriate attribute to test using a conjoint survey.
Defendant also argues that Gaskin’s description of latency renders his methodology unreliable because Gaskin "devotes over 60% more words to his description of latency than to any other attribute description," thereby introducing focalism bias into the survey. Gaskin Daubert Mot. at 12 (quoting Holt Report at 20). Thus, Defendant argues that survey respondents are likely to pay more attention to the latency attribute and consider it more important because the survey describes the latency attribute in more detail. Id. However, the Ninth Circuit has held the critiques of survey design, including that a survey was unduly suggestive, typically go to the weight, not the admissibility, of a survey. See ThermoLife Int’l, LLC v. Gaspari Nutrition Inc., 648 Fed.Appx. 609, 613 (9th Cir. 2016) ("Although the district court faulted the survey’s biased questions and unrepresentative sample, neither defect was so serious as to preclude the survey’s admissibility."); Fortune Dynamic, 618 F.3d at 1038 ("To be sure, ... the Marylander survey has a number of shortcomings, including the fact that it ... may have been suggestive.... But these criticisms, valid as they may be, go to ... the weight of the survey rather than its admissibility." (internal quotation marks omitted) ); Southland Sod Farms v. Stover Seed Co., 108 F.3d 1134, 1143 (9th Cir. 1997) (holding that whether a survey’s questions were leading "go[es] only to the weight, and not the admissibility, of the survey").
In addition, with respect to focalism bias in particular, district courts have found that alleged focalism bias goes to the weight of the expert’s opinion, not its admissibility. See Fitzhenry-Russell, 326 F.R.D. at 603-04, 2018 WL 3126385 at *6 (declining to exclude expert opinion even after expressing concern with conjoint survey design); In re Scotts EZ Seed Litig., 2017 WL 3396433, at *9-10 (S.D.N.Y. Aug. 8, 2017) (finding that alleged biasing demand artifacts, focalism bias, and starting point bias went to the weight of the survey, not its admissibility); Morales v. Kraft Foods Group, Inc., 2017 WL 2598556, at *15-16 (C.D. Cal. June 9, 2017) ("Defendant also argues that Bodapati failed to conduct a blind survey and instead telegraphed to respondents the purpose of the survey.... This phenomenon is known as ‘focalism.’ ... This is another matter that goes to the weight, not the admissibility of the challenged survey."); Svenson v. Google, 2016 WL 8943301, at *6 (N.D. Cal. Dec. 21, 2016) (rejecting argument that biases, including focalism bias, rendered expert’s opinion inadmissible).
Accordingly, the Court finds that the alleged focalism bias of Gaskin’s survey goes to the survey’s weight and it is thus an issue more properly explored through "competing evidence and incisive cross-examination" than exclusion. See Murray, 870 F.3d at 925.
B. The Survey Reliably Measures Defects in This Case
Defendant next argues that Gaskin’s conjoint survey does not accurately measure the latency defects alleged in the instant case. Gaskin Daubert Mot. at 13. Specifically, Defendant asserts that consumers cannot relate to the magnitude of latency at issue here, which is measured in milliseconds, and faults Gaskin for "editorializ[ing]" by opting to define the alleged defects in terms of their "widespread" or "frequent" impact on performance. Id. Defendant contends that because the length of any delay or glitch is undefined, survey respondents will likely interpret the delays at issue to be longer than the alleged delays at issue in this case. Defendant also argues that Gaskin’s definition of the latency levels differs from Plaintiffs’ testimony about the performance problems that they experienced. Id. at 14-19.
These arguments amount to critiques of Gaskin’s survey design— specifically, the framing of the survey options. As discussed above, in the Ninth Circuit, criticisms that "go to the issues of methodology, survey design, reliability, ... [and] critique of conclusions" go to the weight of a survey, not its admissibility. Fortune Dynamic, 618 F.3d at 1038; see also Medlock v. Taco Bell Corp., 2015 WL 8479320, at *5 (E.D. Cal. Dec. 9, 2015) ("The Court finds that Defendants’ criticisms of Dr. Moore’s wording of the questions goes to the weight of the evidence, and not the admissibility of the survey.").
This Court addressed an analogous argument in Apple . See Apple, Inc. v. Samsung Elecs. Co., 2014 WL 794328, at *18 (N.D. Cal. Feb. 25, 2014). In Apple, Samsung argued that the conjoint survey "did not accurately capture the invention claimed by the patents, rendering the results unreliable." Id. at *13. The Court first noted that it "continue[d] to recognize that the framing of questions for purposes of surveys is generally an issue of weight, not admissibility." Id. at *18. "Nevertheless, the Court also recognize[d] that there must be outer limits to this principle. At some point, a description of a patent in a survey may vary so much from what is claimed that the survey no longer ‘relate[s] to any issue in the case’ and is ‘not relevant and, ergo, non-helpful.’ " Id. (quoting Daubert, 509 U.S. at 591, 113 S.Ct. 2786) (second alteration in original). Applied here, the Court also recognizes that at some point, the description of the attribute in a survey may vary so much from the allegations or evidence that the survey becomes untethered from the facts of the case "and, ergo, non-helpful." Daubert, 509 U.S. at 591, 113 S.Ct. 2786.
However, Defendant has not established that any deficiencies in Gaskin’s survey design are so severe as to rise to that level. As discussed above, Gaskin’s descriptions of latency and its effects, as well as the definitions of the different latency levels, are drawn from Newman’s report, which in turn is based on Newman’s analysis of Defendant’s and Intel’s internal test results. Moreover, even if each Plaintiff did not allege experiencing all of the performance effects that Gaskin includes in the survey, Plaintiffs do describe experiencing the same types of performance effects that Gaskin’s survey includes. See SACC ¶¶ 153 (B. Alexander allegedly experienced performance issues including stuttering and lag during online gaming that rendering gaming impossible at times), 154 (Knowles allegedly experienced slow Internet performance during online gaming and call degradation while using Skype), 155 (Reyna allegedly experienced stuttering and lag during online gaming and performance issues while using a virtual private network ("VPN") and while streaming videos on Netflix and Amazon), 156 (Walton allegedly experienced significantly slow Internet speeds and freezing during online gaming and performance issues while using a VPN and Windows Remote Desktop Connection). Thus, while Gaskin’s survey may overstate the frequency or magnitude of these effects, there is not so great a disconnect between Plaintiffs’ allegations and Gaskin’s survey as to render the survey inadmissible. "[Defendant] will be free to attack [Gaskin’s] conjoint study as inflating the price premium at trial." Fitzhenry-Russell, 326 F.R.D. at 604, 2018 WL 3126385 at *6 (citing Estate of Barabin, 740 F.3d at 463 ("Shaky but admissible evidence is to be attacked by cross examination, contrary evidence, and attention to the burden of proof, not exclusion.") ).
C. The Survey Matches the Theory of Liability
Defendant’s third Daubert argument repeats its other Comcast argument: Defendant contends that Gaskin’s conjoint survey is not limited to Plaintiffs’ theory of liability. For the reasons explained above, the Court finds this argument unpersuasive.
In sum, the Court finds that conjoint analysis is a generally reliable, well recognized method for estimating how consumers value different attributes of a product. The Court also finds that Gaskin is qualified as an expert in conjoint analysis. Defendant’s criticisms of Gaskin’s survey are either unpersuasive or go to the survey’s weight, not its admissibility. Accordingly, Defendant’s motion to exclude Gaskin’s testimony, ECF No. 97, is DENIED. Because the Court finds that neither Weir’s nor Holt’s expert opinions would alter the outcome of the Court’s analysis and the Court does not rely on either opinion, the Court DENIES AS MOOT the motions to exclude Weir and Holt, ECF Nos. 99, 124.
As a result, the Court finds that Rule 23(b)(3)’s predominance requirement is satisfied. Finally, the Court analyzes Rule 23(b)(3)’s superiority requirement.
2. Rule 23(b)(3) Superiority
Rule 23(b)(3) provides four factors that a court must consider in determining whether a class action is superior to other methods of adjudication. These factors are:
(A) the class members’ interests in individually controlling the prosecution or defense of separate actions;
(B) the extent and nature of any litigation concerning the controversy already begun by or against class members;
(C) the desirability or undesirability of concentrating the litigation of the claims in the particular forum; and
(D) the likely difficulties in managing a class action.
Fed. R. Civ. P. 23(b)(3). "[T]he purpose of the superiority requirement is to assure that the class is the most efficient and effective means of resolving the controversy." Wolin, 617 F.3d at 1175 (quoting 7A Charles Wright, Arthur Miller & Mary Kay Kane, Federal Practice and Procedure, § 1779 at 174 (3d ed. 2005) ). "In cases in which plaintiffs seek to recover relatively small sums and the disparity between litigation costs and the recovery sought may render plaintiffs unable to proceed individually, ‘class actions may permit the plaintiffs to pool claims which would be uneconomical to bring individually.’ " Moore v. Ulta Salon, Cosmetics & Fragrance, Inc., 311 F.R.D. 590, 622 (C.D. Cal. 2015) (quoting Local Joint Executive Bd. Of Culinary/Bartender Tr. Fund v. Las Vegas Sands, Inc., 244 F.3d 1152, 1163 (9th Cir. 2001) ).
Defendant’s only superiority argument, made in two sentences at the end of Defendant’s Opposition, is that Plaintiffs have failed to provide a trial plan or "articulate even a single fact in this case supporting superiority." Opp’n at 25. Neither argument is persuasive. First, with respect to a trial plan, this Court has previously observed "that no formal trial plan is required by Rule 23." See Dunbar v. Google, Inc., No. 12-CV-3305-LHK, 2012 WL 6202797, at *21 n.4 (N.D. Cal. 2012); see also Sullivan v. Kelly Services, Inc., 268 F.R.D. 356, 365 (N.D. Cal. 2010) ("Nothing in Rule 23 requires Plaintiff to submit a formal trial plan along with her motion for class certification."). Second, Defendant is incorrect that Plaintiffs have not addressed superiority. In their Motion, Plaintiffs argue that "[g]iven the large number of class members, each with relatively small claims for defective Modems, forcing individual purchasers to litigate their cases, particularly where common issues predominate for the Class, would be ‘an inferior method of adjudication.’ " Mot. at 25 (quoting Wolin, 617 F.3d at 1176). Plaintiffs go on: "In this case, a class action would therefore be ‘well suited to address the rights of groups of people who individually would be without effective strength to bring their opponents into court at all.’ " Id. (quoting Datta v. Asset Recovery Sols., LLC, No. 15-CV-188-LHK, 2016 WL 1070666, at *10 (N.D. Cal. Mar. 18, 2016) (internal quotation marks omitted) ). This Court has previously recognized that class actions are superior in cases such as this one, where many consumers have claims of relatively little value, such that it would be inefficient to bring individual claims. See, e.g., Datta, 2016 WL 1070666 at *7-10.
Specifically, with respect to the first superiority factor, the Ninth Circuit has observed that "[w]here damages suffered by each putative class member are not large," the "interest of each member in individually controlling the prosecution or defense of separate actions ... weighs in favor of certifying a class action." Zinser v. Accufix Research Inst., Inc., 253 F.3d 1180, 1190 (9th Cir. 2001) (internal quotation marks omitted). Such is the case here. The first factor thus weighs in favor of certification.
With respect to the second superiority factor, "the extent and nature of any [other] litigation concerning the controversy," the Court is not aware of any other actions against Defendant related to the claims at issue in the instant case. Thus, the second factor weighs in favor of certification. See Datta, 2016 WL 1070666 at *10.
The third factor, the desirability of concentrating the litigation in a particular forum, also weighs in favor of certification. The parties have stipulated to litigating the California claims first, and there is no indication that this particular forum is any less desirable than any other forum in California. See id.
Finally, the fourth factor, which concerns the difficulty of managing a class action, depends largely on whether Plaintiffs’ case "rises and falls [on] common evidence." In re High-Tech Emp. Antitrust Litig., 985 F.Supp.2d 1167, 1228 (N.D. Cal. 2013). This factor thus overlaps with the Court’s commonality, typicality, and predominance analysis. Accordingly, the Court finds that the fourth factor weighs in favor of certification.
Overall, after weighing the four superiority factors, the Court concludes that Plaintiffs have satisfied Rule 23(b)(3)’s superiority requirement.
Thus, because the Court finds that Rules 23(a) and 23(b)(3) have been satisfied, the Court GRANTS Plaintiffs’ motion for class certification.
IV. CONCLUSION
For the foregoing reasons, the Court DENIES Defendant’s motions to exclude Newman and Gaskin, DENIES AS MOOT Defendant’s motion to exclude Weir, DENIES AS MOOT Plaintiffs’ motion to exclude Holt, and GRANTS IN PART and DENIES IN PART Plaintiffs’ motion for class certification. The Court CERTIFIES the following class and subclass:
Class: All persons who purchased an Arris SURFboard SB6190 cable modem in California on or after October 1, 2015.
Subclass: All persons who purchased an Arris SURFboard SB6190 cable modem in California for personal, family, or household purposes on or after October 1, 2015.
The class seeks relief under the UCL and FAL. The subclass seeks relief under the Song-Beverly Consumer Warranty Act and CLRA.
The Court APPOINTS Greg Knowles and Brian Alexander as representatives of the class and subclass. As Defendant does not challenge the adequacy of proposed class counsel, the Court also APPOINTS Willem Jonckheer, Dustin Schubert, and Noah Schubert of Schubert Jonckheer & Kolbe LLP as class counsel.
IT IS SO ORDERED.