On July 17, 2018, Judge Jon S. Tigar of the United States District Court for the Northern District of California granted plaintiffs’ motion to certify a class in a securities class action against Twitter, Inc. (the “Company”) and two of its officers. In re Twitter Inc. Securities Litigation, No. 3:16-cv-05314 (N.D. Cal. July 17, 2018). Plaintiffs allege that defendants violated Sections 10(b) and 20(a) of the Securities Exchange Act of 1934 (the “Exchange Act”) by making materially false and misleading statements regarding user growth and engagement, resulting in a 15 percent stock drop when the Company later disclosed that user engagement was “slowing quite dramatically.” The Court previously had granted in part and denied in part defendants’ motion to dismiss.
The Court first evaluated the proposed class definition, which included all persons and entities that purchased or otherwise acquired shares of the Company between February 6, 2015—the day after the Company filed a Form 8-K announcing an upward trend in its monthly average users—through July 28, 2015, when the Company disclosed on an analyst call that investors should not “expect a change in [the Company’s] growth rate . . . for a while.” Defendants argued that the class definition should be modified to exclude so-called “in-and-out traders” (i.e., investors who (i) purchased their shares after the start of the alleged class period but sold their shares prior to the first alleged corrective disclosure, or (ii) purchased their shares for the first time after the first alleged corrective disclosure and sold their shares before the second alleged corrective disclosure at the end of the class period). Defendants relied on cases interpreting the Supreme Court’s decision in Dura Pharmaceuticals, Inc. v. Broudo, 544 U.S. 336 (2005) to mean that in-and-out traders could not be included in a class because they would not be able to allege a cognizable loss. While noting that “[d]istrict courts in the Ninth Circuit have been in conflict with each other over whether in-and-out traders are appropriately included,” the Court found that Duraonly relates to how loss causation must be pled in the complaint, and does not affect the class definition at the class certification stage. Emphasizing that excluding in-and-out traders from the class prior to the completion of discovery would be “premature” because plaintiffs “may prove that some truth leaked out prior to the corrective disclosures and caused injury at a later date,” the Court held that in-and-out traders were appropriately included at this stage and approved plaintiffs’ proposed class definition. The Court noted, however, that it retained the power to modify the class definition at a later stage in the case and that defendants may renew their request after discovery is complete.
The Court then evaluated the four requirements for class certification under Federal Rules of Civil Procedure (“FRCP”) Rule 23(a)—numerosity, commonality, adequacy of representation, and typicality. The Company challenged only (1) whether the two proposed co-class representatives—an asset management company and a company that provides retirement benefits for members of a union—could adequately represent the class, and (2) whether it was appropriate to appoint co-class counsel. Regarding the proposed co-class representative structure, the Company argued that the appointment of co-representatives would undermine the lead plaintiff evaluation and appointment provisions of the Private Securities Litigation Reform Act of 1995 (the “PSLRA”), as the union representative was not previously selected as lead plaintiff in this case. According to the Company, the PSLRA lead plaintiff appointment process involves “a less searching inquiry” than the Rule 23 class certification process, and therefore because the union representative had withdrawn its motion for appointment of lead counsel it was not capable to serve as a co-class representative. The Court disagreed, finding that neither the text of the PSLRA nor its underlying purpose supported the Company’s argument, noting that just because one plaintiff might be “the mostcapable” does not mean that another plaintiff is not also capable, and that the PSLRA “does not in any way prohibit the addition of named plaintiffs to aid the lead plaintiff in representing a class.” Regarding the proposed co-class counsel, the Company argued that it was unnecessary to appoint an additional firm as co-counsel when the lead plaintiffs’ counsel could sufficiently represent the interests of the proposed class. The Court rejected this argument, noting that “courts in this district and circuit have appointed co-class counsel in PSLRA cases,” and emphasizing that the Company’s “interests in this litigation are not aligned with those of the class” and thus “it is unlikely that [the Company] has the class’s interests in mind when it argues against the appointment of additional counsel.” The Court therefore approved the two plaintiffs as class co-representatives as well as the two law firms as co-class counsel.
Finally, the Court considered the predominance requirement under FRCP Rule 23(b)(3). In contesting predominance, defendants argued that plaintiffs had failed to proffer a damages model tied to the facts of this case that could be applied on a class-wide basis, and that plaintiffs’ expert had not yet calculated damages for the alleged claims. The Court disagreed, noting that defendants were not attacking the methodology of a proposed damages model but rather the fact that the plaintiffs’ expert had not yet calculated damages for the class’s alleged claims. Citing Comcast Corp. v. Behrend, 569 U.S. 27 (2013), the Court held that plaintiffs’ burden at the class certification stage is only to demonstrate that “the questions of law or fact common to the class members predominateover any questions affecting only individual members,” not to require a class-wide damages model to demonstrate predominance. Finding all requirements under FRCP Rule 23 satisfied, the Court granted class certification.