Leventhal v. Chegg, Inc.

• Court: District Court, N.D. California
• Decided: March 4, 2024
• Docket: 5:21-cv-09953
• Status: Unknown

Opinion

1 2 3 4 UNITED STATES DISTRICT COURT 5 NORTHERN DISTRICT OF CALIFORNIA 6 7 STEVEN LEVENTHAL, Case No. 21-cv-09953-PCP

8 Plaintiff, ORDER DENYING DEFENDANTS’ 9 v. MOTION TO DISMISS AND DENYING PLAINTIFFS’ MOTION TO STRIKE 10 CHEGG, INC., et al., DDkktt.. NNooss.. 112222,, 112299 Defendants. 11

12 13 Lead plaintiffs Pompano Beach Police and Firefighters’ Retirement System and KBC 14 Asset Management NV brought this securities fraud class action lawsuit in December 2021 against 15 defendants Chegg, Inc., Chegg CEO Daniel Rosensweig, Chegg CFO Andrew Brown, and Chegg 16 President of Learning Services Nathan Schultz. Defendants move to dismiss the case and plaintiffs 17 move to strike certain portions of a declaration attached to defendants’ motion. For the reasons 18 that follow, the Court denies both motions. 19 BACKGROUND 20 Chegg is a publicly traded educational technology company that provides textbook rentals, 21 online tutoring services, and homework help to students. Chegg is comprised of two main business 22 segments: (1) Chegg Services, which includes online subscription products such as “Expert 23 Q&A,” wherein freelance experts answer students’ academic questions in nearly real-time; and (2) 24 Required Materials, which consists of the company’s print and electronic textbook offerings. In 25 their amended complaint, plaintiffs allege that Chegg defendants made material misrepresentations 26 during the class period (from May 5, 2020 to November 1, 2021) that inflated the company’s stock 27 price (which peaked in February 2021) and eventually caused a steep stock price decline in early 1 unprecedented growth during the COVID-19 pandemic to the growing trend of online learning 2 rather than rampant cheating on the platform. Plaintiffs contend that remote learning environments 3 during the pandemic accelerated Expert Q&A subscriptions, through which students cheated on 4 graded assignments and exams, and that Chegg failed to fully disclose the extent of this cheating 5 to protect its market valuation. In support of their argument, plaintiffs note that by the end of the 6 class period, Chegg Services accounted for nearly 90% of Chegg’s total revenues. Id. at 15. 7 Plaintiffs allege that many of defendants’ statements during the class period were false or 8 misleading. First, plaintiffs contend that defendants misrepresented the frequency of cheating on 9 Chegg’s platform. Defendant Brown purportedly said in a podcast that while there are “some very 10 isolated cases [of cheating] … the fact of the matter is this: [Chegg] is a learning site.” Dkt. No. 11 115, at 79. He also said that “user generated content where kids can upload papers and can upload 12 tests … doesn’t happen on Chegg.” Id. Further, Chegg’s Honor Code stated, “[M]isuse of our 13 platform represents an extremely small portion of the activity of our services.” Id. at 81. And 14 Chegg’s Communications Manager said only a “tiny fraction of users” cheat on Chegg. Id. at 86. 15 Second, plaintiffs allege that Chegg misrepresented the drivers of its revenue growth in its 16 public statements without ever mentioning cheating or remote learning during the pandemic as 17 possible factors. Chegg repeatedly insisted that growth was “primarily due to [Chegg’s] efforts to 18 reduce account sharing” as opposed to cheating. Id. at 96. Defendant Brown argued that “a 19 substantial increase in [Chegg’s] subscription services [was] driven by new US and International 20 subscribers to our platform, as well as increased success with our account sharing efforts.” Id. at 21 73. In multiple instances, defendants cited the reduction in account sharing and international 22 expansion as the key drivers for Chegg’s revenue growth during the class period, not cheating. 23 Further, plaintiffs argue that Chegg failed to acknowledge the role of at-home learning during the 24 pandemic in increasing subscriptions. Defendant Rosensweig said that Chegg is a “permanent 25 solution,” and that the platform has the same “take rates” for “those that are at school and not at 26 school.” Id. at 83. He further noted that “Chegg’s success in the US is not a result of people being 27 on campus or not being on campus.” Id. at 92. Rosensweig also said that whether “we’re a stay- 1 stated that “whether you are on-campus or not on-campus, [it doesn’t] matter to Chegg’s growth” 2 and “the issue for Chegg was never whether you’re physically on a campus or at home.” Id. at 95. 3 Rosensweig even unequivocally said in May 2021 that “we are not a COVID stock.” Id. at 21. 4 To support these allegations, plaintiffs rely on: (1) documents from universities nationwide 5 asserting that student cheating was rampant on Chegg during the COVID-19 pandemic; (2) 6 statements from high-level university officials that Chegg was informed about this rampant 7 cheating on several occasions; (3) accounts from former Chegg employees stating that Expert 8 Q&A fueled Chegg’s growth and that defendants were aware of student cheating on Chegg; and 9 (4) an empirical analysis by lead counsel showing surges in expert questions during the class 10 period and finding that approximately 25% of these questions exhibited indicia of cheating. 11 Plaintiffs further allege that defendants Rosensweig and Schultz profited substantially in 12 the class period through their misrepresentations, with Rosensweig purportedly selling more than 13 550,000 shares for $49.5 million and Schultz selling nearly 340,000 shares for $25 million. 14 Plaintiffs assert three claims in their complaint: (1) violation of Section 10(b) of the 15 Exchange Act and SEC Rule 10b-5 against all defendants for defrauding investors; (2) violation of 16 Section 20(a) of the Exchange Act against the three individual defendants as controlling persons; 17 and (3) violation of Section 20A of the Exchange Act against Rosensweig and Schultz for 18 contemporaneous trading. Plaintiffs request class certification, damages, and attorneys’ fees. 19 Defendants move to dismiss the amended complaint, arguing that plaintiffs fail to 20 adequately plead the Section 10(b) requirements of: (1) a false or misleading statement; (2) a 21 strong inference of scienter; and (3) loss causation. Since plaintiffs fail to state a primary violation 22 of Section 10(b), defendants argue the Sections 20(a) and 20A claims must also fail. Dkt. No. 122. 23 Plaintiffs separately move to strike paragraphs 34 through 39 of the Declaration of Heather 24 Speers (an associate at defendants’ law firm Cooley LLP) attached to the motion to dismiss. Dkt. 25 No. 122-1. In these paragraphs, Speers cites defendants Rosensweig and Schultz’s SEC Forms 4 to 26 show the number of Chegg shares that were sold by each before and during the class period, and 27 uses information about the individual defendants’ stock acquisitions during the class period to 1 the Speers Declaration violates Civil Local Rule 7-5(b) prohibiting “conclusions and argument,” 2 evades the 25-page limit governing motions to dismiss under Civil Local Rule 7-2, and improperly 3 argues disputed facts at the pleadings stage. Dkt. No. 129. 4 LEGAL STANDARDS 5 A Section 10(b) violation requires “a material misrepresentation or omission of fact, 6 scienter, a connection with the purchase or sale of a security, transaction and loss causation, and 7 economic loss.” Curry v. Yelp Inc., 875 F.3d 1219, 1224 (9th Cir. 2017). “A securities fraud 8 complaint under § 10(b) and Rule 10b-5 must satisfy the dual pleading requisites of Federal Rule 9 of Civil Procedure 9(b) and the PSLRA [Private Securities Litigation Reform Act].” In re 10 VeriFone Holdings, Inc. Securities Litigation, 704 F.3d 694, 701 (9th Cir. 2012). Rule 9(b) states: 11 “In alleging fraud or mistake, a party must state with particularity the circumstances constituting 12 fraud or mistake. Malice, intent, knowledge, and other conditions of a person’s mind may be 13 alleged generally.” Fed. R. Civ. P. 9(b). And “[t]he PSLRA significantly altered pleading 14 requirements in private securities fraud litigation by requiring that a complaint plead with 15 particularity both falsity and scienter.” Gompper v. VISX, Inc., 298 F.3d 893, 895 (9th Cir. 2002). 16 For plaintiffs in private securities fraud class actions, the PSLRA creates “formidable 17 pleading requirements to properly state a claim and avoid dismissal under Fed. R. Civ. P. 18 12(b)(6).” Metzler Inv. GMBH v. Corinthian Colleges, Inc., 540 F.3d 1049, 1055 (9th Cir. 2008). 19 To properly allege falsity a securities fraud complaint must “specify each statement alleged to 20 have been misleading, the reason or reasons why the statement is misleading, and, if an allegation 21 regarding the statement or omission is made on information and belief, … with particularity all 22 facts on which that belief is formed.” Zucco Partners, LLC v. Digimarc Corp., 552 F.3d 981, 990– 23 91 (9th Cir. 2009). To adequately plead scienter the complaint must “state with particularity facts 24 giving rise to a strong inference that the defendant acted with the required state of mind.” Id. at 25 991. 26 In Tellabs, Inc. v. Makor Issues & Rights, Ltd., 551 U.S. 308, 324 (2007), the Supreme 27 Court held that a “strong inference” of scienter exists “if a reasonable person would deem the 1 from the facts alleged.” In the Ninth Circuit, a court must conduct a dual inquiry in assessing 2 whether this standard is met: “[F]irst, it determines whether any one of the plaintiff’s allegations is 3 alone sufficient to give rise to a strong inference of scienter; second, if no individual allegations 4 are sufficient, it conducts a ‘holistic’ review to determine whether the allegations combine to give 5 rise to a strong inference of scienter.” Glazer Capital Management, L.P. v. Forescout 6 Technologies, Inc., 63 F.4th 747, 766 (9th Cir. 2023). The strong inference standard applies only 7 to the element of scienter, not to falsity. Id. (“Falsity is subject to a particularity requirement and 8 the reasonable inference standard of plausibility set out in Twombly and Iqbal, and scienter is 9 subject to a particularity requirement and a strong inference standard of plausibility.”). 10 ANALYSIS 11 I. Plaintiffs’ Complaint States Valid Securities Fraud Claims. 12 A. Plaintiffs Adequately Plead Falsity. 13 A statement is false or misleading “if it would give a reasonable investor the impression of 14 a state of affairs that differs in a material way from the one that actually exists.” Berson v. Applied 15 Signal Tech., Inc., 527 F.3d 982, 985 (9th Cir. 2008). Accordingly, to adequately plead falsity, 16 plaintiffs must demonstrate that cheating was prevalent on Chegg’s platform notwithstanding 17 statements from the company that there was minimal cheating occurring, and that Chegg’s 18 substantial growth is attributable to the at-home learning environment during the pandemic as 19 opposed to the reduction in account sharing and international expansion as publicized by Chegg. 20 While defendants complain that plaintiffs fail to “quantify the number or percentage of subscribers 21 that used Chegg to cheat before, during, or after the class period,” Dkt. No. 122, at 17, such 22 detailed quantification is unnecessary to demonstrate the existence of cheating on the platform at 23 the pleadings stage. Defendants also contend that plaintiffs assume that widespread cheating 24 substantially increased growth during the class period, but that “there is not a single well-pled fact 25 to support either proposition.” Id. But while plaintiffs must plausibly demonstrate that the remote 26 learning environment accelerated Chegg’s growth during the pandemic, the Court can draw 27 inferences from plaintiff’s circumstantial evidence to find that cheating on the platform in at-home 1 Applying this standard, the evidence cited in plaintiffs’ complaint, including plaintiffs’ 2 empirical analysis, former employee testimony, university interviews, and faculty statements, 3 provides sufficient factual support for plaintiffs’ allegations of falsity. 4 First, through the lead counsel’s empirical analysis, the complaint identifies “the number, 5 subjects, and timeframe of unique user questions reviewed; parameters used to identify cheating, 6 with examples; the resulting calculation of percentage of questions containing obvious signs of 7 cheating; and trends in the data before, during, and after the Class Period.” Dkt. No. 128, at 21. 8 This evidence provides a particularized showing of cheating on the platform. Further discovery 9 will allow for a larger sample size and more robust statistical analysis. 10 Defendants counter that the lead counsel’s empirical analysis concluding that cheating was 11 rampant is biased and unreliable because it was premised on a review of only 6,000 sample 12 questions on Expert Q&A (only 0.009% of the total 70 million questions on the platform). 13 Defendants further contend that the lead counsel is not an expert in statistics and it is not clear that 14 the sample was even random (e.g., only photographic question submissions were analyzed). 15 Additionally, defendants argue that plaintiffs’ search terms for cheating (“final exam” or “graded 16 assignment”) were overbroad. Finally, they contend that the number of questions that have indicia 17 of cheating does not necessarily correspond to the number of cheating subscribers, which is 18 ultimately the relevant metric pertaining to Chegg’s revenue growth. Dkt. No. 122, at 19. 19 But defendants inflate the requisite standard to demonstrate falsity here. Though plaintiffs 20 do not show the exact percentage of subscribers who were cheating on Chegg’s platform, they 21 present compelling empirical evidence of substantial cheating during the class period. This 22 analysis is sufficient to meet both the PSLRA’s particularity requirement and Twombly and Iqbal’s 23 reasonable inference plausibility standard on a Rule 12(b)(6) motion to dismiss. 24 Second, plaintiffs rely upon former employee (FE) testimony to support their allegations of 25 falsity. Plaintiffs allege that FE3 said that Expert Q&A was “the cheating business” and that FE1 26 said that it was the “main moneymaker” and “the main reason for people to subscribe, it was 27 driving [growth].” Dkt. No. 115, at 55. Further, FE2 purportedly said that Expert Q&A remained 1 Together, these statements support plaintiffs’ allegations that there was cheating on Expert Q&A, 2 and that such cheating was tied to both subscriber and revenue growth for Chegg. 3 Defendants respond that the statements by Chegg’s FEs are neither reliable nor 4 particularized. Specifically, defendants argue that “none of the FEs claim the Expert Q&A was 5 popular because students used it to cheat” and that “none of the FEs claim that cheating drove 6 subscriber or revenue growth.” Dkt. No. 122, at 21. But this mischaracterizes the complaint’s 7 allegations regarding the testimony. While defendants argue that no FEs stated that Expert Q&A 8 was used to cheat, testimony by FE3 directly contradicts this point. And as to whether cheating 9 drove growth, the Court can circumstantially infer from FE1 and FE2’s statements about Expert 10 Q&A’s popularity during the pandemic that cheating on Expert Q&A drove Chegg’s growth. 11 Third, plaintiffs provide comprehensive evidence from various universities and faculty to 12 support the allegation that cheating occurred nationwide using Chegg. For example, Dean 13 Blandizzi at UCLA tied “raining cases” of academic integrity violations to Chegg, stating that the 14 platform allowed students to post “actual questions of current homework assignments and exams.” 15 Dkt. No. 115, at 30. An Assistant Professor at UCLA further wrote “my midterms have been 16 compromised severely by Chegg.com. Nearly all questions have been posted, sometimes many 17 times.” Id. at 31. A Professor at Georgia Tech similarly stated that “the questions to my … take 18 home exam were on Chegg.” Id. at 34. Similar accounts of cheating via Chegg were noted by 19 faculty at the University of Nebraska, with students “submitting answers completely copied from 20 the solutions provided by Chegg,” id. at 38; at California State University, where a student “posted 21 [an] exam to ‘ask an expert’ during the exam window,” id.; and at the United States Air Force 22 Academy, where 249 cadets purportedly used Chegg to cheat, id. at 40. 23 While the defendants may be correct that the cheating represents only a small fraction of 24 the total students at these schools, the evidence demonstrates that a substantial number of students 25 subscribed to Chegg were cheating via Expert Q&A. Statements by high-level faculty at various 26 academic institutions similarly suggest that cheating on Chegg was quite common at these 27 schools. A Director at Texas A&M estimated that 80–85% of the massive increase in cheating 1 These extensive documents and interviews allege with particularity that cheating occurred 2 on Chegg, and in combination with the lead counsel’s empirical analysis and statements by 3 Chegg’s former employees, plausibly allege that this drove Chegg’s revenue growth. 4 Defendants respond that the documents from universities and interviews with faculty do 5 not support falsity because they are anecdotal and do not “tie Chegg’s class period revenue or 6 subscriber growth to cheating.” Dkt. No. 122, at 22. Defendants argue that this evidence merely 7 creates the impression that cheating was pervasive across Chegg’s platform, while in truth 8 establishing only that cheating was rampant at particular universities. Again, defendants contend 9 that the faculty grievances “provide no insight as to the number of subscribers who used the 10 platform to cheat,” and therefore fail to provide particularized facts to support the claims of falsity. 11 Id. at 23. As discussed above, however, plaintiffs need not show at this stage the number of 12 subscribers that were using Chegg to cheat; they need only demonstrate with particularity that 13 there was substantial cheating on the platform. They have met this burden through their empirical 14 analysis and documentation from universities and faculty nationwide. And from the FE testimony 15 the Court can reasonably infer that cheating on Expert Q&A drove Chegg’s revenue and 16 subscriber growth. 17 Finally, defendants counter that plaintiffs’ theory of the case is wrong because subscriber 18 and revenue growth continued even after the class period. Dkt. No. 133, at 13–14 (“If Plaintiffs’ 19 theory were true, one would expect Chegg subscriber base and revenue to crater after Q3 2021— 20 when Plaintiffs allege cheating stopped. But just the opposite happened.”). After markets closed 21 on November 1, 2021, Chegg “reported Q3’21 revenue growth of 23% and subscriber growth of 22 17% over Q3’20.” Dkt. No. 122, at 15. And in Q4 2021, both the number of subscribers and 23 revenue again purportedly increased. Dkt. No. 133, at 14. Chegg contends that “the fact that 24 Chegg’s subscriber growth continued beyond the class period undermines Plaintiffs’ claim that it 25 was driven entirely by pandemic-enabled cheating.” Dkt. No. 122, at 30. 26 But Chegg’s argument misses the mark. While revenue and subscriber growth may have 27 increased in absolute terms even after the class period, plaintiffs allege that Chegg nonetheless 1 estimates.” Dkt. No. 115, at 71. As Chegg itself admits, it lowered its Q4 2021 revenue guidance 2 because “enrollment slowed in fall 2021,” meaning that there was a decline in the company’s rate 3 of growth. Dkt. No. 122, at 15. Contrary to Chegg’s assertion that its “dramatic growth after 4 cheating allegedly stopped negates any inference that cheating drove growth, eviscerating 5 Plaintiffs’ entire theory of fraud,” Dkt. No. 133, at 14, Chegg concedes there was a relative decline 6 in the company’s growth rate. Further, Chegg overstates plaintiffs’ burden in establishing that 7 cheating drove Chegg’s growth. Plaintiffs need not show that cheating accelerated by remote 8 learning was the sole cause of the company’s subscriber and revenue growth (in which case Chegg 9 would be correct that its subscriber and revenue numbers should have dropped once students 10 returned to in-person learning in the fall of 2021), but only that cheating was a driver of growth 11 during the class period. Evidence of Chegg’s enrollment slowdown and lower growth rate after the 12 class period is sufficient to support plaintiffs’ theory of the case at the pleadings stage. 13 In short, plaintiffs have adequately pleaded falsity as to the challenged statements made by 14 defendants. Glazer, 63 F.4th at 76 (holding that courts should not “impose an impossibly high 15 burden on securities actions plaintiffs.”). 16 B. Defendants’ Allegedly False Statements Are Actionable. 17 Defendants respond that most of the challenged statements are inactionable as a matter of 18 law. They argue that most are either statements of historical fact about Chegg’s revenue, vague 19 expressions of optimism, opinions about Chegg’s value proposition, protected aspirational 20 statements, or expectations about the future.1 21 The Court disagrees. First, statements concerning the frequency of cheating on Chegg are 22 neither statements of historical fact nor protected aspirational or forward-looking statements. For 23 example, defendant Brown allegedly said that while there are “some very isolated cases [of 24 cheating] … the fact of the matter is this: [Chegg] is a learning site.” Dkt. No. 115, at 79. Further, 25 Chegg’s Communications Manager said that only a “tiny fraction of users” cheat on Chegg, id. at 26

27 1 The PSLRA’s safe harbor provision protects forward-looking statements that are either: (i) 1 86, and Chegg’s Honor Code stated, “[M]isuse of our platform represents an extremely small 2 portion of the activity of our services,” id. at 81. Unlike accurately reported historical information 3 about Chegg’s revenue and subscription growth, the unverified statements by defendant Brown 4 and Chegg’s Communication Manager about the amount of cheating on the platform do not 5 contain inactionable historical facts. And the statement in Chegg’s Honor Code is not aspirational, 6 as defendants contend, since it explicitly states that cheating is limited on the platform. 7 Second, while defendants argue that the statements about the drivers of Chegg’s growth 8 are vague expressions of optimism or opinions about Chegg’s value proposition, even opinions 9 can be misleading if the “facts going to the basis for the issuer’s opinion” are omitted. Omnicare, 10 Inc. v. Laborers Dis. Council Constr. Indus. Pension Fund, 575 U.S. 175, 194 (2015); In re Atossa 11 Genetics Inc. Sec. Litig., 868 F.3d 784, 802 (9th Cir. 2017) (holding that an “opinion statement did 12 not fairly align[] with the information in [the CEO’s] possession at the time” so it was “misleading 13 by omission”). Chegg insisted in its public statements that its growth was “primarily due to 14 [Chegg’s] efforts to reduce account sharing” as opposed to pandemic-related cheating. Dkt. No. 15 115, at 96. And defendant Brown argued that “a substantial increase in [Chegg’s] subscription 16 services [was] driven by new US and International subscribers to our platform, as well as 17 increased success with our account sharing efforts.” Id. at 73. But based on the plaintiffs’ evidence 18 (as discussed above), it is at least plausible that the challenged statements omit factual information 19 that was in Chegg’s possession regarding abuse of the platform and growth accelerated by the 20 pandemic. 21 Third, defendant Rosensweig made statements that “Chegg’s success in the US is not a 22 result of people being on campus or not being on campus,” id. at 92, and that whether “we’re a 23 stay-home company or not a stay-home company [is] completely irrelevant to Chegg.” Id. He also 24 unequivocally stated in May 2021 that “we are not a COVID stock.” Id. at 21. Defendants contend 25 that these are forward-looking statements that merely provide opinions about Chegg’s value 26 proposition and are thereby protected under the PSLRA’s safe harbor provision. But given the 27 plethora of evidence presented by plaintiffs about the company’s growth coinciding with at-home 1 defendants were aware of this purported misrepresentation (as discussed below), the safe harbor 2 provisions of the PSLRA would not apply because the defendants’ statements (even if forward- 3 looking) were not accompanied by “meaningful cautionary statements.” 4 Accordingly, the statements that plaintiffs allege to be false are actionable. 5 C. Plaintiffs Adequately Plead Scienter. 6 Scienter is “a mental state that not only covers intent to deceive, manipulate, or defraud, 7 but also deliberate recklessness.” Schueneman v. Arena Pharms., Inc., 840 F.3d 698, 705 (9th Cir. 8 2016). “The PSLRA’s ‘strong inference’ requirement has teeth. It is an ‘exacting’ pleading 9 obligation that ‘present[s] no small hurdle for the securities fraud plaintiff.’” Nguyen v. Endologix, 10 Inc., 962 F.3d 405, 414 (9th Cir. 2020) (citations omitted). Thus, to establish a strong inference of 11 deliberate recklessness, the plaintiff must plead “a highly unreasonable omission, involving not 12 merely simple, or even inexcusable negligence, but an extreme departure from the standards of 13 ordinary care, and which presents a danger of misleading buyers or sellers that is either known to 14 the defendant or is so obvious that the actor must have been aware of it.” Zucco, 552 F.3d at 991. 15 In determining whether each defendant had the requisite scienter, courts “must consider plausible, 16 nonculpable explanations” for defendants’ conduct, and determine “whether all of the facts 17 alleged, taken collectively, give rise to a strong inference of scienter.” Tellabs, 551 U.S. at 323. 18 Defendants argue that plaintiffs fail to provide particularized facts showing that defendants 19 were aware of rampant cheating on the platform driving Chegg’s revenue growth but nonetheless 20 omitted this information in their public statements. Plaintiffs respond that they properly pleaded 21 scienter by presenting evidence of the various reports sent to Chegg by universities and faculty 22 members about rampant cheating on the platform. For example, at UCLA, “Associate Dean Rush 23 sent Chegg more than ten letters in May 2020 reporting students’ use of Chegg to cheat” and 24 Professor Park told Chegg in an email that “all of the final exam problems (and solutions by the 25 3rd party?) were posted to [Chegg’s] website.” Dkt. No. 115, at 32. Chegg allegedly admitted 26 Park’s allegations in response. Id. at 33. Further, staff at the University of Nebraska “sent Chegg 27 at least five letters regarding students cheating using Chegg’s website, specifically identifying 1 Again, Chegg admitted these allegations in response. Faculty from California State University 2 similarly “sent Chegg multiple letters” about cheating, to which Chegg also responded. Id. at 39. 3 Notably, a Director at Texas A&M University was so “fed up with Chegg” that she “emailed 4 defendant Rosensweig directly,” and three high-level Chegg representatives responded to her 5 email. Id. at 44. Taken together, these reports and Chegg’s responses constitute particularized facts 6 that support a strong inference of scienter as to Chegg’s knowledge about cheating on its 7 platform.2 8 In further support of their argument, plaintiffs also present testimony from former 9 employees demonstrating that discussions about cheating on the platform were regularly held at 10 company meetings. For example, FE1 noted that “our CEO would sometimes talk about [student 11 cheating] during all-hands [meetings].” Id. at 57. These meetings were attended by individual 12 defendants Rosensweig, Brown, and Schultz. Another former employee noted that “executives 13 distributed a Forbes article discussing widespread student cheating using Chegg to employees.” Id. 14 at 58. Referring to the Forbes article, FE7 stated that “people weren’t surprised” because “the 15 [cheating] issues were absolutely well understood.” Id. FE4 even noted that Chegg “didn’t really 16 put any effort into stopping [cheating] because it was putting money into their pocket.” Id. at 56. 17 Ultimately, Tellabs requires courts to assess allegations of scienter holistically. Defendants 18 argue that the “cogent and compelling inference is that [Chegg] honestly and reasonably believed 19 that the pandemic had accelerated what they viewed as an inevitable transition to online learning.” 20 Dkt. No. 122, at 30. The extensive communications between university officials and Chegg about 21 rampant cheating on the platform combined with detailed former employee testimony, however, 22 meet the high bar for pleading scienter at this stage.3 Courts must consider “the totality of the 23 24 2 Plaintiffs separately argue that Chegg has access to user engagement data and should have seen blatant examples of cheating on the platform. But the Ninth Circuit has rejected the notion that a 25 defendant possesses scienter merely because it “could regularly track its sales data.” Lipton v. Pathogenesis Corp., 284 F.3d 1027, 1036 (9th Cir. 2002) (“[P]laintiffs merely assert in conclusory 26 terms that the defendants had access to internal data demonstrating a decline in sales of [product]. Plaintiffs do not identify any internal reports of ‘sales data,’ much less plead, in any detail, the 27 contents of any such report or the purported data.”). As defendants note, plaintiffs do not identify any specific user engagement data in the complaint or establish how it showed cheating. 1 circumstances” in making this determination, and “need not close their eyes to circumstances that 2 are probative of scienter viewed with a practical and common-sense perspective.” S. Ferry LP, No. 3 2 v. Killinger, 542 F.3d 776, 784 (9th Cir. 2008). Under this lens, the complaint’s allegations of 4 scienter satisfy both Rule 9(b)’s particularity requirement as well as the PSLRA’s strong inference 5 standard. 6 D. Plaintiffs Adequately Plead Loss Causation. 7 Defendants further argue that plaintiffs fail to adequately plead loss causation, which 8 requires particularized allegations that “the very facts about which the defendant [allegedly] lied” 9 caused injuries. Nuveen Mun. Fund v. City of Alameda, 730 F.3d 1111, 1120 (9th Cir. 2013). 10 “Rule 9(b) applies to all elements of a securities fraud action, including loss causation.” Oregon 11 Public Employees Retirement Fund v. Apollo Group Inc., 774 F.3d 598, 605 (9th Cir. 2014). 12 Plaintiffs seek to recover losses from a substantial stock price drop that occurred on 13 November 1, 2021, after Chegg announced that it was lowering its Q4 2021 revenue projections. 14 Defendants argue that the lowered guidance was driven by significantly fewer enrollments and 15 was unrelated to cheating or the end of remote learning. Dkt. No. 122, at 31. Plaintiffs counter that 16 the nearly 50% drop in stock price was caused by defendants’ misrepresentations about Chegg’s 17 growth—i.e., by not informing the public that the increase in subscribers was caused by remote 18 learning environments during the pandemic where students could more easily cheat through 19 Expert Q&A. Specifically, plaintiffs allege that Chegg’s misrepresentations artificially inflated the 20 company’s stock price, and when Chegg reported lower-than-expected growth in Q3 2021 after 21 students began returning to campus, the public markets corrected in response to this revelation. 22 Mineworkers’ Pension Scheme v. First Solar Incorporated, 881 F.3d 750, 754 (9th Cir. 2018) 23

24 period reinforce scienter. In re Daou Sys., Inc., 411 F.3d 1006, 1022 (9th Cir. 2005) (“Unusual trading or trading at suspicious times or in suspicious amounts by corporate insiders has long been 25 recognized as probative of scienter.”). As defendants point out, however, both individuals also made stock acquisitions, which reduces the percentage decrease in their stock ownership during 26 the class period. Defendant Rosensweig also sold three times as many shares in the 18-month period prior to the class period. See In re Apple Comput. Sec. Litig., 886 F.2d 1109, 1117 (9th Cir. 27 1989) (“Large sales of stock before the class period are inconsistent with plaintiffs’ theory that 1 (holding that a “stock price drop com[ing] immediately after the revelation of fraud can help rule 2 out alternative causes.”). As plaintiffs note, not only did Forbes respond after Chegg revealed its 3 lowered guidance that “the enrollment excuse makes no sense,” Dkt. No. 115, at 11, but Morgan 4 Stanley also stated that “as students returned to campus for in person learning, they did not return 5 to the Chegg platform [given] more graded assignments in class where Chegg would not be 6 helpful.” Id. The Ninth Circuit has held that when “analysts note[] the probable relationship 7 between” alleged misstatements and a stock price decline, plaintiffs have adequately pleaded loss 8 causation. Lloyd v. CVB Financial Corp., 811 F.3d 1200, 1202 (9th Cir. 2016). In light of the 9 evidence presented, plaintiffs here have met their burden to plead loss causation. In re 10 WageWorks, Inc., Sec. Litig., 2020 WL 2896547, at *8 (N.D. Cal. June 1, 2020) (noting a “low bar 11 of pleading proximate cause for purposes of loss causation”). 12 E. Plaintiffs’ Section 20(a) and 20A Claims Stand. 13 In moving to dismiss the Section 20(a) and 20A claims, defendants argue that those claims 14 necessarily fail because the complaint does not state a primary violation of Section 10(b). Because 15 plaintiffs have properly pleaded a Section 10(b) violation, the Section 20(a) claims against all 16 three individual defendants and the Section 20A claims against Rosensweig and Schultz likewise 17 will not be dismissed. In re NVIDIA Corp. Securities Litigation, 768 F.3d 1046, 1052 (9th Cir. 18 2014) (requiring “a primary violation of underlying securities laws, such as Section 10(b) or Rule 19 10b-5” to establish a cause of action under Section 20(a)); In re VeriFone Securities Litigation, 11 20 F.3d 865, 872 (9th Cir. 1993) (requiring the same to establish a cause of action under Section 21 20A). 22 In the alternative, defendants argue that the Section 20(a) claim against Schultz should be 23 dismissed because he did not exercise “actual power” over Rosensweig or Brown. Dkt. No. 122, at 24 31. Not only is this improbable given Schultz’s role as President of Learning Services, but 25 defendants also misunderstand the relevant test, which is whether Schultz “had the requisite power 26 and authority to direct the company” as a controlling person of Chegg, as plaintiffs have alleged. 27 Dkt. No. 128, at 32. Howard v. Everex Systems, Inc., 228 F.3d 1057, 1065 (9th Cir. 2000) (finding 1 II. Plaintiffs’ Motion To Strike Is Denied. 2 Plaintiffs separately move to strike paragraphs 34 through 39 of the Speers Declaration 3 Dkt. No. 122-1. The declaration provides updated stock ownership percentage decreases for 4 individual defendants Rosensweig and Schultz that account for reported stock acquisitions as well 5 as sales. The declaration also explains that Rosensweig and Schultz reported selling their shares 6 pursuant to Rule 10b5-1 trading plans and due to tax withholding. Plaintiffs argue that the 7 declaration violates Civil Local Rule 7-5(b), which prohibits “conclusions and argument,” because 8 introducing the reasons why Chegg shares were sold constitutes impermissible argument; that the 9 declaration evades the 25-page limit governing motions to dismiss under Civil Local Rule 7-2; and 10 that the declaration improperly argues disputed facts at the pleadings stage, Khoja v. Orexigen 11 Therapeutics, Inc., 899 F.3d 988, 998 (9th Cir. 2018). 12 Defendants respond that paragraphs 34 through 39 of the Speers Declaration “simply 13 explain the math underlying the figures Defendants cited in their Motion to Dismiss relating to 14 stock sales.” Dkt. No. 131, at 2. They contend that these paragraphs are in fact necessary to 15 explain the modified stock disposition percentages in their motion to dismiss under Civil Local 16 Rule 7-5(a), which requires that “[f]actual contentions made in support of or in opposition to any 17 motion must by supported by an affidavit or declaration and by appropriate references to the 18 record.” While Rule 7-5(b) states that “[a]n affidavit or declaration may contain only facts … and 19 must avoid conclusions and argument,” defendants further argue that a mathematical calculation is 20 a fact (not an argument) and the statement about shares being disposed for Rule 10b5-1 trading 21 plan or tax withholding purposes is also a fact based on the attached SEC Forms 4 in Exhibits 32– 22 33. Responding to plaintiffs’ argument about the 25-page limit in Rule 7-2, defendants note that 23 all the relevant legal arguments are contained in the motion to dismiss, and the affidavits only 24 provide background facts and calculations. Finally, regarding plaintiffs’ concern about competing 25 factual narratives, defendants argue that in determining scienter under the PSLRA’s heightened 26 requirements, “[c]ourts in this circuit have routinely taken judicial notice of Forms 4 to determine 27 whether insider stock sales raise an inference of scienter to support a § 10(b) action.” Azar v. Yelp, 1 The defendants’ position is persuasive. The calculations in the Speers Declaration simply 2 || provide factual background to support the legal arguments concerning Rosensweig and Schultz’s 3 stock sales in defendants’ motion to dismiss. The paragraphs at issue neither violate Rule 7-5(b) 4 || nor evade the 25-page limit under Rule 7-2. And the Court takes judicial notice of the Forms 4 in 5 support of defendants’ argument that stock trading plans negate an inference of scienter. But see 6 || Azar, 2018 WL 6182756, at *18 (“Without reviewing [the executive’s] actual trading plan, the 7 || Court ... cannot conclude that the plan negates any inference of scienter.”). Accordingly, the 8 || Court denies plaintiffs’ motion to strike paragraphs 34 through 39 of the Speers Declaration. 9 CONCLUSION 10 For the foregoing reasons, the Court denies defendants’ motion to dismiss and denies 11 plaintiffs’ motion to strike. IT IS SO ORDERED. 5 13 Dated: March 4, 2024

. P. Casey Pitts = 16 United States District Judge

18 19 20 21 22 23 24 25 26 27 28