Securities and Exchange Commission v. AT&T Inc.

UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK SECURITIES AND EXCHANGE COMMISSION, Plaintiff, 21 Civ. 1951 (PAE) ~ OPINION & ORDER AT&T, INC., CHRISTOPHER C, WOMACK, KENT D. EVANS, and MICHAEL J. BLACK, Defendants. PAUL A, ENGELMAYER, District Judge: This case is a rare litigated enforcement action brought by the Securities and Exchange Commission (“SEC”) arising out of the SEC’s Regulation FD—Fair Disclosure (“Regulation FD” or “Reg FD”). See 17 C.F.R. § 243. Promulgated in 2000, Reg FD prohibits a public company from selectively disclosing material nonpublic information (“MNPI”) about itself or its securities to certain persons outside the company, unless it also discloses that information to the public. The SEC here sues the public telecommunications company AT&T, Inc. (“AT&T”), and three members of its Investor Relations (“IR”) Department: Christopher C. Womack, Kent D. Evans, and Michael J. Black (the “individual defendants” or the “IR defendants,” and, together with AT&T, “defendants”). The SEC alleges that in March and April 2016, AT&T embarked on a campaign to selectively disclose MNPI to analysts at 20 Wail Street firms. As alleged, AT&T’s goal was to “manage” those analysts to reduce their estimates of AT&T’s first quarter of 2016 (“Q1 2016”) total revenue, to enable AT&T to beat the consensus revenue estimate for that quarter. AT&T had missed consensus revenue estimates in two of the three preceding quarters, and by March 2016, analysts’ consensus revenue estimate exceeded AT&T’s internal estimates by more than $1 billion. The SEC alleges that, acting at the direction of AT&1’s chief financial officer and IR Director, defendants Womack, Evans, and Black, in calls to analysts, selectively disclosed MNPI that caused numerous analysts to significantly reduce their Q1 2016 revenue estimates. This scheme, as alleged, succeeded: AT&T’s total revenue, as announced, exceeded analysts’ final consensus revenue estimate by 0.1%. The internal data that AT&T selectively disclosed, as alleged, included the company’s projected or actual total revenue, and internal metrics bearing on total revenue, including wireless equipment revenue and wireless equipment upgrade rates. On this basis, the SEC brings a claim against AT&T under Section 13 of the Securities and Exchange Act of 1934, 15 U.S.C. § 78a ef seq., for violating Reg FD, and claims against Womack, Evans and Black for aiding and abetting that violation. Following extensive fact and expert discovery, all parties have now moved for summary judgment on all claims. Both have also moved to exclude evidence. These motions, although in the nature of motions in limine, were appropriately made at this stage, given the possibility that the exclusion of evidence might affect the summary judgment analysis. To this end, each side moves to exclude the testimony of the other’s proposed experts, under Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579 (1993). And the defense moves to exclude notes of analysts whom defendants contacted, arguing that these are inadmissible hearsay. For the reasons set out in this decision, the Court denies both sides’ summary judgment motions, To the extent defendants argue that Regulation FD is invalid—as violative of the First and Fifth Amendments; outside the SEC’s authority to promulgate; or logically inoperable— these challenges are unconvincing. To the extent defendants argue that the SEC has failed to come forward with sufficient evidence to support its claims, that, too, is wrong. The SEC has adduced sufficient evidence on each disputed element: to wit, that the information at issue was (1) material, (2) nonpublic, and selectively disclosed (3) with scienter. The evidence is, in fact, formidable that the information that the individual defendants selectively disclosed about AT&T in their calls to analysts was both material and nonpublic. And, although the balance of the evidence on the scienter element is closer, the SEC has adduced sufficient evidence on which a reasonable jury could find for the SEC on that element, too, to wit, that Womack, Evans, and Black knew that—or were at least reckless about whether-—the information they were selectively feeding analysts was material and nonpublic, in violation of Reg FD. At the same time, summary judgment cannot be entered for the SEC. A reasonable jury could find for the individual defendants, at a minimum, on the element of scienter. And because AT&T’s liability, as charged by the SEC, appears based on that of the individual defendants, summary judgment cannot be entered against AT&T, either. Barring settlement, the SEC’s claims therefore must be resolved at trial. As to the evidentiary motions, the Court denies defendants’ motion to globally strike the analysts’ notes. Subject to document- or excerpt-specific objections that the Court will take up closer to trial, these are, in the main, admissible as non-hearsay and under hearsay exceptions. The Court does not, and need not, resolve today the parties’ Daubert motions. Regardless of how these were resolved, there would be sufficient evidence of each element to reach a jury, and a material dispute of fact on at least the scienter element. The Court denies these motions, without prejudice to the parties’ right fo raise the same or similar motions closer to trial. L Factual Background! A. Regulation FD Reg FD was promulgated in 2000 to fill a gap in federal securities laws with respect to the selective disclosure by public companies of material nonpublic information. In promulgating the regulation, the SEC explained that where such information is selectively disclosed, it “leads to a loss of investor confidence in the integrity of our capital markets.” Final Rule: Selective Disclosure and Insider Trading, SEC Release No. 7881, 2000 WL 1201556, at *2 (Aug. 15, 2000) (“Adopting Release”). Selective disclosure, the agency added, “bears a close resemblance ordinary ‘tipping’ and insider trading,” in that it enables “a privileged few [to] gain an information edge—and the ability to use that edge to profit—from their superior access to corporate insiders, rather than from their skill, acumen, or diligence.” Jd. The regulation was also intended to prevent issuers from using “material information as a commodity to be used to gain or maintain favor with particular analysts or investors.” Id. Reg FD provides, in relevant part: Whenever an issuer, or any person acting on its behalf, discloses any material nonpublic information regarding that issuer or its securities to any person described in paragraph (b)(1) of this section, the issuer shall make public disclosure of that information .. . (1) Simultaneously, in the case of an intentional disclosure; and (2) Promptly, in the case of an non-intentional disclosure. The Court draws its account of the underlying facts from the parties’ respective submissions on the cross-motions for summary judgment, including the materials described infra in Part Il. Citations to a party’s 56.1 statement incorporate the evidentiary materials cited therein. When facts stated in a party’s 56.1 statement are supported by testimonial, video, or documentary evidence and not denied by the other party, or denied by a party without citation to conflicting admissible evidence, the Court finds such facts to be true. See $.D.N.Y. Local Civil Rule 56.1(c) (“Each numbered paragraph in the statement of material facts set forth in the statement required to be served by the moving party will be deemed to be admitted for purposes of the motion unless specifically controverted by a correspondingly numbered paragraph in statement required to be served by the opposing party.”); id. Rule 56.1(d) (“Each statement by the movant or opponent . . . controverting any statement of material fact[] must be followed by citation to evidence which would be admissible, set forth as required by Fed. R. Civ. P. 56(c).”). 17 C.F.R. § 243.100(a). Reg FD applies to a disclosure made to any person outside the issuer, including: a broker or dealer, investment adviser, institutional investment manager, and anyone who holds the issuer’s securities “under circumstances in which it is reasonably foreseeable that the person will purchase or sell the issuer’s securities on the basis of that information.” Jd. § 243.100(b)\(1). Reg FD does not apply, however, to disclosures made to “a person who owes a duty of trust or confidence to the issuer (such as an attorney, investment banker, or accountant),” “a person who expressly agrees to maintain the disclosed information in confidence,” or, in certain circumstances, when the disclosure is made in connection with a securities offering. Jd. § 243.100(b)(2). Reg FD’s Adopting Release addressed the material nonpublic information to which the regulation applied. The regulation, it stated, “does not define” such terms, but “relies on existing definitions of these terms established in the caselaw.” Adopting Release, 2000 WL 1201556, at *9, In the decades since Reg FD’s issuance, the SEC has enforced the regulation relatively sparingly, focusing enforcement action on more serious asserted violations of Reg FD. As the Director of the SEC’s Enforcement Division explained when the agency first issued the rule, the SEC did not intend “to test the outer limits of the rule by bringing cases that aggressively challenge the choices issuers are entitled to make regarding the manner in which a disclosure is made,” Richard H. Walker, Director, Div. of Enf't, S.E.C., Regulation FD-—An Enforcement Perspective (Nov. 1, 2000), available at https://www.sec.gov/news/speech/spch415.htm. Instead, the SEC intended to “be on the lookout for two types of violations. The first are egregious violations involving the intentional or reckless disclosure of information that is unquestionably material.....” The second are “cases against those who deliberately attempt to game the system.” /d. Consistent with that philosophy, the SEC has pursued claims under Reg FD against a relatively limited number of public companies and/or executives, with such actions almost invariably resulting in settlement. See, e.g., In re TherapeuticsMD, Inc., Exchange Act Release No. 86708, 2019 WL 3933685 (Aug. 20, 2019) ($200,000 civil penalty for Reg FD violation); SE.C. v. Brian Pappas et al., Litig. Release No. 23914, 2017 WL 3614292 (Aug, 22, 2017) (imposing officer-and-director and penny stock bars and requiring approximately $71,000 in disgorgerment, interest, and penalties); 7m re Lawrence D. Polizzotto, Exchange Act Release No. 70337, 2013 WL 4773958 (Sept. 6, 2013) ($50,000 penalty); S.#.C. v. David Ronald Allen et al., Litig. Release No, 22208, 2011 WL 10915927 (Dec. 22, 2011) (settlements and injunctions entered against numerous defendants); Jn re Fifth Third Bancorp, Exchange Act Release No. 65808, 2011 WL 5865859 (Nov. 22, 2011) (entering cease-and-desist order but not imposing penalties, in part based on cooperation with SEC investigation); S.C. v. Office Depot, Inc., Litig. Release No. 3199, 2010 WL 4134972 (Oct. 21, 2010) ($1 million civil penalty); Jn re Patricia A. McKay, Exchange Act Release No. 63154, 2010 WL 4134969 (Oct. 21, 2010) ($50,000 settlement for CFO in Office Depot action, supra); In re Stephen A. Odland, Exchange Act Release No. 63153, 2010 WL 4134968 (Oct. 21, 2010) ($50,000 settlement for CEO in Office Depot action, supra); S.E.C. v. Presstek, Inc. et al., Litig. Release No. 21443, 2010 WL 784231 (Mar. 9, 2010) ($400,000 civil penalty); S.E.C. v. Christopher A. Black, Litig. Release No. 21222, 2009 WL 3047574 (Sept. 24, 2009) ($25,000 penalty for CFO; declining to bring an enforcement action against corporation that promptly and publicly disclosed selective disclosure and cooperated extensively); S.Z.C. v. Srinivasan, Litig, Release No, 20296, 2007 WL 2778650 (Sept. 25, 2007) ($70,000 penalty for company’s former president; consent entry of cease-and- desist order against company, providing for disgorgement and prejudgment interest of $490,902); In re Flowserve Corp. et al., Exchange Act Release No. 51427, 2005 WL 677810 (Mar. 24, 2005) (final judgment in action in the United States District Court for the District of Columbia requiring company to pay $350,000 civil penalty and CEO to pay $50,000; see also S.E.C. v. Flowserve Corp. et al., No. 05 Civ. 612 (D.D.C. 2005)); In re Senetek PLC, Exchange Act Release No. 50400, 2004 WL 2076191 (Sept. 16, 2004) (cease-and-desist order); In re Schering-Plough Corp., Exchange Act Release No. 48461, 2003 WL 22082153 (Sept. 9, 2003) (final judgment in the United States District Court for the District of Columbia requiring company to pay $1 million civil penalty and CEO to pay $50,000; see also SLE.C. v. Schering- Plough Corp., No. 03 Civ. 1880 (D.D.C. 2003)); In re Raytheon Co., Exchange Act Release No, 46897, 2002 WL 31643026 (Nov. 25, 2002) (cease-and-desist order); S.E.C. v. Siebel Sys., Inc., Litig. Release No. 17860, 2002 WL 31643062 (Nov. 25, 2002) (consent judgment requiring company to pay $250,000); In re Secure Computing Corp. et al., Exchange Act Release No. 46895, 2002 WL 31643024 (Nov. 25, 2002) (cease-and-desist order); see also Report of Investigation Pursuant to Section 21(a) of the Securities Exchange Act of 1934: Netflix, Inc. and Reed Hastings, Exchange Act Release No. 69279, 2013 WL 5138514 (Apr. 2, 2013) (reporting investigation of Netflix, Inc.’s potential Reg FD violation, to guide issuers’ compliance with the law); Report of Investigation Pursuant to Section 21(a) of the Securities Exchange Act of 1934, Exchange Act Release No. 46898, 2002 WL 31650174 (Nov. 25, 2002) (“in re Motorola’) (same, for Motorola, Inc.). In only one reported action has an SEC action under Reg FD resulted in adversarial litigation. See S.E.C. v. Siebel Sys., Inc,, 384 F. Supp. 2d 694, 709 (S.D.N.Y. 2005) (dismissing action against software company and two officers on grounds that, as pled, the corporate disclosures were neither material nor nonpublic). B. The Defendants AT&T: AT&T, a telecommunications company, is a Delaware corporation headquartered in Dallas, Texas. Dkt. 84 (Joint Stipulation of Facts, or “JSF”) 41. It is publicly traded on the New York Stock Exchange under the ticker “T.” Jd. The products and services AT&T offered during the relevant period—2015 and 2016—included “wireless communications, data/broadband and Internet services, digital video services, wireline phone service, and telecommunications equipment.” Id. At all relevant times, Randall L. Stephenson was AT&T’s chairman and chief executive officer (“CEO”), id. 48, and John Stephens was AT&T’s chief financial officer (“CFO”), and reported to Stephenson, id. ff] 13-14. In this role, Stephens oversaw the company’s investor relations function. fd. 16. In 2015, John Stankey became CEO of AT&T’s Entertainment Group—one of four segments of AT&T’s business.” Jd 3, 10. Womack: Womack has worked for AT&T since 1997, and, in 2006, became a member of its IR Department. Jd § 26. In March and April 2016, the months the disclosures at issue were allegedly made, Womack was an executive director in the IR Department, id. § 25, reporting to Michael Viola, the IR Department’s supervisor,? id. {{[ 27, 62. Womack’s responsibilities included communicating with sell-side analysts who covered AT&T, and helping develop internal forecasts of results. fd. | 29; Dkt. 92 (“PL 56.1”) { 33. * The other segments, after AT&T acquired DIRECTV in July 2015, were Business Solutions, Consumer Mobility, and International. JSF ¢3. Since July 1, 2020, Stankey has been AT&T’s CEO. id. ¥ 12. 3 Viola retired from AT&T in June 2020. JSF 7 68. Black: Black has worked for AT&T since 2008, and, in 2011, joined the [R Department. JSF 435. In March and April 2016, he was a finance director in the IR Department. /d. { 34. He reported to Womack. fd. § 28. Black’s responsibilities included communicating with sell- side analysts who covered AT&T, Id. ff 35-37; Pl. 56.1 {] 36-37. He built relationships with analysts, with the goal of becoming their primary point of contact to understand AT&T’s business. PI]. 56.1 37. His responsibilities included extracting data from analysts’ models to build out a running document he maintained called the “Top Ten” average document. Among other metrics, it captured forecasts for wireless upgrade rates and wireless equipment revenue.* JSF 134; Pl. 56.1 4 41. Evans: Evans has worked for AT&T or its predecessor company, Cingular Wireless, in the IR Department since 2000. JSF {| 45-46. At all relevant times, he, like Womack, reported to Viola. id. | 47. Evans was one of the IR Department’s subject-matter experts about AT&T’s wireless business. /d. □ 48-49. His responsibilities included communicating with investors and analysts about the wireless business, and keeping others in the IR Department up to date about “key trends” within that business. fd § 53. Evans communicated with an AT&T financial group in Atlanta that tracked, inter alia, upgrade rates. fd. J] 50, 53. His job also included reviewing presentations that AT&T executives made to their teams for purposes of Reg FD compliance. PI. 56.1 4 59; Dkt. 137 (Def. Reply 56.1”) ¢ 59. C. AT&T’s Investor Relations Department At all relevant times, the AT&T IR Department’s “principal function was to provide investors and analysts, among others, with accurate information concerning the company” to “Wireless upgrade rates” refer to the rates at which AT&T’s subscribers upgraded their smartphones through AT&T. JSF “Wireless equipment revenue” refers to revenue AT&T receives from the sale of wireless phones. [d. § 87. OQ help investors and investment funds (i.¢., the “buy-side”) and brokerage or institutional analysts (i.e., the “sell-side”) make “informed decisions” about AT&T’s stock. Pl. 56.1 60; Def. Reply 56.1 4 60; see JSF J 124 (“It is the job of investor relations personnel at public companies to have private conversations with analysts as long as they do not disclose information which is both material and non-public . .. about their employer.”). AT&T’s CFO, Stephens, was in charge of the IR Department. JSF §[ 16,57. The department’s head, Viola, reported to Stephens; their offices were nearby at AT&T’s Dallas headquarters. /d. [] 62, 65-66. The IR Department had about 12 employees, including the individual defendants and Martin Sheehan, who reported directly to Viola.’ Jd. 59. The IR defendants typically spoke to analysts at about 30 investment firms at least twice a quarter: typically, right after AT&T published its quarterly earnings—usually the third week of the month after a quarter ends—and again before AT&T announced its earnings for the then-ending quarter. Dkt. 106 (ID 56.1”) 44 8, 11. In their roles in the IR Department, Womack, Black, and Evans were privy to nonpublic financial performance information about AT&T. That included AT&T’s “2+1” forecast (which includes two months of a quarter’s actual results), and its actual results for a full quarter before these were publicly announced, JSF ff 31, 44, 51. The IR Department tracked analyst commentary about AT&T for the entire company. Every morning, the IR Department sent an email digest summarizing recent such commentary to CEO Stephenson, CFO Stephens, Entertainment Group CEO Stankey, and IR Department employees, including Black, Evans, Womack, and Viola. JSF { 67; Pl. 56.1 4 73; Def. Reply > As reviewed below, in nearly all the relevant communications with an analyst, the AT&T IR Department participant was one of the individual defendants—Womack, Black, or Evans. The Court accordingly will refer to these calls generally as occurring with the “IR defendants,” while mindful that non-defendant Sheehan was responsible for one such call. 56.1 973. For the first quarter of 2016, Evans also prepared a “Q&A” document to track conversations with analysts about quarterly earnings reporting; Black testified that he used that document to prepare himself for such conversations. Pl. 56.1 4104; Def. Reply 56.1 { 104. Evans’s “Wireless Q&A” document contained both public and, sometimes, nonpublic information.® Pl. 56.1 { 105; Def. Reply 56.1 7 105. D. AT&T’s Training, Policies, and Procedures Regarding Reg FD AT&T’s internal policies and practices prohibited employees from disclosing AT&T’s internal actual or projected results. JSF 4116. Womack and Black were trained on Reg FD when they joined the IR Department, were periodically trained on Reg FD thereafter, and were generally familiar with its prohibition on the selective disclosure of MNPI. Id. 9] 30, 43. Since at least 2006, Evans regularly received Reg FD training. Jd. 55-56. Viola, the head of the IR Department, received training when he joined, periodic training thereafter, and was familiar with the regulation’s content. /d. 63. The training all three individual defendants received from AT&T instructed them that no metric or number may be discussed with analysts unless the metric was already public, regardless whether it was material, ID 56.1 73. Paula Anderson, AT&T’s in-house counsel with expertise about Reg FD, trained IR Department employees regularly about Reg FD. JSF ff] 60-61. On September 19, 2014, she sent Viola materials about Reg FD, including a 2007 legal memo from a law firm. Pl. 56.1 179; Def. Reply 56.1 4 179; see Dkt. 85 (Appendix A to JSF: Joint Exhibits, or “JX”), Ex, 141. Anderson also trained the employees on internal IR Department policies governing MNPI. For ® The document showed that AT&T’s revenue on a single $700 phone was $572, based on data provided by company accountants. Defendants contend that these numbers gave an incomplete picture of wireless sales’ impact on profit. Pl. 56.1 ff] 106-07; Def. Reply 56.1 4] 106-07. 11 instance, Anderson used the “IR Training Document,” which included a qualitative definition of materiality, to train the team. Pl. 56.1 {J 196-97; Def. Reply 56.1 ff 196-97; see JX 268. The IR Training Document included provisions addressing particular species of information, and reviewing Reg FD’s prohibition on selective disclosure. Pl. 56.1 ff] 198-201; Def. Reply 56.1 198-201; see JX 268. Anderson also used an April 7, 2005 memorandum from the law firm Davis Polk & Wardwell LLP to train the IR Department. JSF 4 1187; Pl. 56.1 202; Def. Reply 56.1 4 202; see JX 11 (the “Davis Polk Memo”), The memo made two “key points.” JX at 1. First, “(pjrivate meetings with analysts or institutional investors are inherently risky from a Regulation FD perspective.” Jd.. Second, “[i]t should still be acceptable to reaffirm guidance privately for a short time after guidance is announced publicly in the ordinary course, so long as there were no intervening events.” Jd. The Davis Polk Memo made other points relevant here. It warned that the SEC would review not only the words spoken but also the speaker’s tone, emphasis, demeanor, “winks and nods,” and code words, /d. at 2-3. It also contained a section titled “Managing the Street’s Estimates,” which noted that several Reg FD actions brought by the SEC had “resulted from company officials trying to manage analysts’ forecasts and bring them in-line with internal company projections.” fd. at 4. E. AT&T’s Wireless Business 1. AT&T’s Wireless Equipment Sales and Accounting AT&T does not manufacture its own phones; it sells phones manufactured by other companies, including Apple, Samsung, and LG. JSF 4 86. For several years, AT&T subsidized the cost of celiphones it sold to subscribers by discounting the cost of the phone at the time of 7 The JSF correctly dates the memo Apri! 7, 2005, whereas the parties’ 56.1 statements erroneously date the memo 20/3. See JSF 4 118; JX 11. 1? purchase. Jd. J 69; Pl. 56.1 9 75; Def. Reply 56.1 775. On or about July 26, 2013, AT&T changed its business model, adopting the “Next” program. Under it, “a subscriber would pay for a smartphone in installments over a set period.” Pl. 56.1 9 76; see JSF 70-71; Def. Reply 56.1 AT&T refers to these as “equipment installment plans” or “EIPs.” JSF 971. EIPs “are intended to encourage existing subscribers to upgrade their current services and/or add connected devices” and to generally “minimize subscriber churn.” Pl. 56.1 §{ 78, 175; Def. Reply 56.1 78, 175. A phone purchased under the Next program would also be eligible for discounted service plans. Pl. 56.1 | 79; Def. Reply 56.1 4 79. Although activation fees and discounts may apply, Next program phones would not be subsidized up front. JSF J 73. Under generally accepted accounting principles, the cost of a phone sold by AT&T under the Next program was higher than the revenue recognized on the phones.” Pl. 56.1 495; Dkt. 178 (attaching Plaintiff's Exhibits to SEC MSJ Mot., or “P. Ex.”), Ex. 86 (AT&T Rule 30(b)(6) deposition) at 23-24. AT&T also recognized imputed interest, with respect to the installment payments under the Next program and the trade-in value of the phones, as an offset to the equipment revenue in the quarter the phone was sold. Pl. 56.1 100-01; Def. Reply 56.1 100--01. Thus, the SEC submits, while a single $600 phone might bring in $500 of equipment revenue, when interest and discounts were taken into account, AT&T might incur a $320 loss on the sale.” Pl. 56.1 {| 102-03; Def. Reply 56.1 {J 102-03, § “Churn is a measure of how many subscribes were leaving a carrier.” JSF 4 111. Subscribers who do not buy a phone from AT&T and buy only services from AT&T do not generate any equipment revenue or expense. JSF § 90. '0 Defendants submit that this accounting treatment provides an incomplete picture of wireless sales’ impact on actual profit. Def. Reply 56.1 § 102. 13 In Q1 2016, AT&T stopped offering smartphones under subsidized plans to most customers. JSF € 76. That quarter, 4,135,000 phones were sold under the Next/EIP plans and 482,000 were sold under subsidized plans. PI. 56.1 4] 83. Eliminating subsidies reduced customers’ incentives to upgrade their phones every two years, as customers would pay significantly more for a new phone. Thus, the transition toward Next/EIP and away from subsidization plans meant that customers would upgrade their phones less frequently—therefore lowering the upgrade rate.'! JSF 78-79. AT&T’s phone renewal average or “cycle” rate was nonpublic information in Q1 2016. Pl. 56.1 § 147; Def. Reply 56.1 § 147. The overall impact on AT&T of this change is a source of some dispute. The parties agree that “[djeclining upgrade rates generally had the effect of lowering [AT&T’s] equipment revenue.” PL. 56.1 J 148; Def. Reply 56.1 § 148; see JSF 98 (‘Declining upgrade rates could correlate with lower equipment revenue.”). They also agree that “lower upgrade rates may... impact wireless margins.” P1. 56.1 J 168; Def. Reply 56.1 4 168. The parties disagree, however, about whether, especially over time, the cost of AT&T’s sales of smartphones was offset by the revenue gained from selling them.’? Pl. 56.1 4 164; Def. Reply 56.1 164, 168. Reduced upgrade rates may increase profitability by reducing commission expenses. Pl. 56.1 {| 165; see Def. Reply 56.1 § 165 (noting that decreases in commission rates contribute to savings). 2 For instance, on April 26, 2016, AT&T’s CEO told the Board that “lower handset upgrade volumes...... certainly impacts revenue growth rates,” but “we continue to see the offsetting, positive effect to earnings.” Pl. 56.1 § 166 (citing Dkt. 178-92). In an internal email, Stephens wrote, on January 16, 2016, “need to understand upgrade rte—500 [basis point] drop is a big deal for long term profits—as long as churn is stable.” Dkt. 178-48 at 1; see PL 56.1 4 146. Stephens also testified that upgrade rates did not affect churn in the last quarter of 2015 or QI 2016. See Def. Reply 56.1 4 146. 14 AT&T monitored customer upgrade rates more closely as customers transitioned away from the subsidy model for phone purchases.!? As part of his job, Evans monitored upgrade rates as one metric among many that “tells you the health of the business.” Pl. 56.1 144; Def. Reply 56.1 § 144. Upgrade rates could also be used to estimate portions of wireless equipment revenue, a concept Womack, Evans, and Black understood.'4 JSF □□ 99-101. Black testified that he would have discussions with analysts about phone renewal cycles, and that analysts would ask about the upgrade rate, and so he needed to be able to provide them with information about the upgrade rate. Pl. 56.1 § 145 (citing P. Ex. 66 (Black deposition) at 152, 314). In 2015 and 2016, Evans and Womack both discussed the handset renewal cycle with analysts. JSF 102, 103. On “various occasions” before March 2016, analysts emailed Evans to ask about the publicly disclosed upgrade rate for the previous fiscal quarter. Where it had been publicly disclosed, Evans provided it. JSF 4 105; Pl. 56.1 § 155; Def. Reply 56.1 4 155. 2. AT&T’s Relevant Reported Metrics AT&T reported its total consolidated revenue in the quarterly (10Q) and annual (10K) form reports it submitted to the SEC. JSF (ff 81-82. AT&T’s total consolidated revenue figure included the company’s total wireless revenue. Pi. 56.1 § 88; JSF 83. Total wireless revenue, in turn, “consisted of both wireless service revenue (e.g., revenues gained from customers’ payments for wireless service) and wireless equipment revenue (¢.g., revenues gained from 3 In 2016, some analyst firms covering AT&T did not include upgrade rate in their models, JSF q 132. Wireless equipment revenue includes revenue from non-phone devices such as tablets and phone accessories (¢.g., “stickers to make a device prettier”), in addition to revenue from phones. Upgrade rates pertain only to the phone aspects of wireless equipment revenue. Def. Reply 4 149-511, 1S customers’ payments for handsets, devices, or other hardware or accessories).”!> Pl. 56.1 88. AT&T’s consolidated equipment revenue included the company’s wireless equipment revenue. PI. 56.1 □ 88; JSF | 84. AT&T also disclosed the rate at which postpaid subscribers upgraded their smartphones through AT&T. This is known as the “postpaid upgrade rate” or “upgrade rate.” JSF 96, 112 (upgrade rates “reflect the percentage of existing AT&T customers who purchase new smartphones”), 115. In the relevant time periods, wireless equipment revenue and costs were as follows: Q12015. | $3.374 billion $4,280 billion Q32015 | $3.234 billion $4,082 billion Q42015. $4.071 billion $5,366 billion Q1-2016 $3.156 billion §3.991 billion 0% See ISF Ff 94-95, 114; Pl. 56.1 119-33; Def. Reply 56.1 4 119-33." In O1 2016, total wireless revenue was about 45% of AT&T’s $40.535 billion consolidated revenue. Pl. 56.1 4 131; Def. Reply 56.1 4 131. Consolidated equipment revenue was $3.434 billion, and consolidated equipment costs were $4.375 billion. Pl. 56.1 9j 121-22; Def. Reply 56.1 €§ 121-22. Wireless equipment revenue was about 92% of consolidated In its SEC filings, AT&T did not separately report wireless equipment costs. JSF Ff 85, 91. 16 The summary judgment record does not appear to contain comparable data as to Q2 2015. That is of no consequence, as it contains data as to relevant comparator quarters for Q1 2016, to wit, Q1 2015, which enables a year-over-year comparison to Q1 2016, and the two quarters immediately preceding Q1 2016, when AT&T missed consensus revenue. See infra Section IF. 16 equipment revenue ($3.156 billion of $3.434 billion), 17.6% of total wireless revenue ($3.156 billion of $17.954 billion), and 8.5% of total consolidated revenue ($3.156 billion of $40.535 billion). Pl. 56.1 (J 132-33; Def. Reply 56.1 {J 132-33. 3. Analyst Coverage of Wireless Revenue and Related Metrics Stock analysts have long published forecasts and other analyses regarding AT&T. JSF 127. Those forecasts cover certain individual financial metrics (e.g., earnings per share). By definition, such forecasts are publicly available where they appear in analysts’ published reports. Id, $128. Analysts prepare financial models that, among other things, set price targets for a stock and make buy, hold, and sell recommendations for companies they cover. /d. 129. Analysts base their research and modeling on a variety of public sources. These include a company’s financial and public statements in its quarterly, annual, and current reports, at earnings calls, and at investor conferences. These also include news reports, historical data, nonpublic investigation of facts or trends the analysts conduct on their own, and—-relevant here—discussions with AT&T’s IR Department. fd. { 130. Various analysts developed understandings of the relationship at AT&T between wireless revenue, upgrade rates, and other metrics, and opined on these matters in their reports. In testimony and public writings, many analysts and firms opined that wireless equipment revenue and upgrade rates mattered for AT&T’s revenue and other bottom-line metrics, including the company’s earnings per share (“EPS”). For example, Barry Sine, a Drexel Hamilton analyst, testified that he had asked Black if AT&T was still going to break out wireless revenue because he considered it “important and helpful in modeling and coming to . . . the conclusion on revenue and EBITDA and EPS for AT&T.” P. Ex. 83 (“Sine Depo.”) at 345-46. Sine also testified that upgrade revenue and equipment revenue “should reduce churn”: “So you’ ll have fewer customers leaving on the whole. You'll have more revenue, more—and greater earnings in future periods, unless, of course, the customer is free to pay off the contract and leave whenever they want. But... they tend to stay through the contract term.” /d. at 350-51. He further wrote in an email that “either bringing old iPhone 5s or just not upgrading” are “both... very profitable.” JX 236. Adam Ilkowitz, a Citi analyst, testified that upgrade rates were a topic of interest for investors because, if the upgrade cycle lengthened, “it would cause [AT&T] to lose some revenue because sales were down, but [its] profitability would improve because they would sell fewer loss making items,” P, Ex. 59 at 51. Ilkowitz testified that, all else equal, the lengthening of the upgrade cycle in April 2016 would cause revenue to decline. /d. at 52; see P. Ex. 77 at 186-87 (Ilkowitz testifying that “[a]ll else equal, if you sell less phones, you’ have less equipment revenue”). Brian Hyun, an RBC analyst, listed, as one of five questions “to focus on for [AT&T],” based on what clients asked him: “Will the upgrade rate go back higher or is the mid-5% the new norm?” JX 258. A January 26, 2016 Pacific Crest Research report premised a “Bull Case” for AT&T on lower upgrade rates and a “Bear Case” on increased upgrade rates “above our high-single-digit percentage rate estimates.” JX 244 at 3. An April 26, 2016 Moffett Nathanson report pointed to “lower-than-expected handset upgrades” as a basis for EBIDTA! growth. IX 126 at 4. On the other hand, other analysts viewed the wireless equipment revenue and upgrade rate metrics as apt to have a more neutral effect. Some testified that equipment revenue and 7 BBIDTA is earnings before interest, taxes, depreciation, and amortization. 18 upgrade rate would not affect earnings!* or profits.'? That opinion appeared in some analysts’ reports. For example, on March 21, 2016, Deutsche Bank published a report reflecting reduced revenue expectations, but described the change as “more optical in nature,” as it was due in part to lower handset upgrade volumes. IX 69. On March 28, 2016, Wells Fargo published a report describing the lower upgrade rates as causing a “negative for the headline revenue print” but opining that “this equipment revenue is margin neutral, or essentially a pass through. So while lower revenue is optically tough to see, it does not impact margin or EPS.” JX 70, On April 19, 2016, Pacific Crest published a report stating that it was lowering equipment revenue estimates based on upgrade rates, but that “equipment revenue is basically a pass-through account.” JX 94. Jennifer Fritzsche, then a Wells Fargo analyst, testified that even if there were a revenue miss because of reduced equipment revenues, that would not impact earnings. Dkt. 108 (“Krumholz Decl.”), Ex. 12 (“Fritzsche Depo.”) at 94. Walter Piecyk, then a BTIG analyst, testified that “because of phone payment plans, the upgrade rate was irrelevant to the EBITDA and earnings of the company.” Krumholz Decl., Ex. 18 (“Piecyk Depo.”) at 69. Sine, the Drexel Hamilton analyst, testified that lower upgrade rates “would have a beneficial impact on earnings if the phones are subsidized and a neutral impact if the phones are not.” Krumholz Decl., Ex. 19 (“Sine Depo.”) at 195. James Breen, a William Blair analyst, testified that he viewed revenue on phone purchases as profit neutral. Krumholz Decl., Ex. 10 (Breen deposition) at 85. John Hodulik, a UBS analyst, testified that, in his view, there were no margins or profit associated with equipment revenue because “from an installment plan standpoint, it’s roughly break even.” Krumholz Decl, Ex. 13 (“Hodulik Depo.”) at 58. He also testified that, more broadly, “nobody is buying or selling the stock based on the upgrade rate or the equipment revenue.” Dkt. 203 (“Stokes Decl.”), Ex. 26 at 93; see also Stokes Decl., Ex, 25 at 142-43 (Hodulik reading from previous testimony). Jeffrey Kvaal, a Nomura analyst, agreed that equipment revenue “either had no margin or very low margin attached to it.” _Krumholz Decl., Ex. 14 (‘“Kvaal Depo.”) at 28. Gregory McNiff, also a Nomura analyst, testified that “everything with equipment revenue is like what’s called like pass- through.” Krumholz Decl., Ex. 15 “(McNiff Depo.”) at 113. That is because AT&T is acting effectively as a “channel” for a phone manufacturer and then “rebating” the cost of the phone to the manufacturer; in McNiff’s words, “as I understand it, there is some breakage, but | think of it as like margin neutral, maybe slightly margin negative.” Id Matthew Niknam, a Deutsche Bank analyst, testified that the “driver” of revenue reductions within wireless was “entirely coming from equipment revenue, which given the industry changes towards no subsidy plans did not carry margin or profitability with it.’ Krumholz Decl., Ex. 17 (‘Niknam Depo.”) at 155. 19 In a similar vein, other analysts testified that, over time, AT&T’s profits would not be affected by the change,”® PF. AT&T’s Previous Misses on Revenue Consensus Financial research companies like Bloomberg and FactSet report “consensus estimates,” or “aggregations of various analyst forecasts for a company for certain metrics”-—typically the average of each forecast. JSF J 138. Consensus estimates are public information. Jd Analysts and some media outlets publish comparisons of AT&T’s actual results to Bloomberg and FactSet estimates. Jd. 9140. Missing consensus can cause concern among investors. See, e.g., Pl. 56.1 { 266 (citing P. Ex. 54 (Viola deposition in SEC investigation, or “Viola Inv. Depo.”) at 224-25; P. Ex. 58 (Fritzche deposition) at 194-95 (“Q. And why would a revenue miss not look good?....A, [A]ny time a company reports revenues lower than where the street consensus is, it’s not... a great data point for that company reporting. Q. Why not? A. Well, if—if people think they’re going to report a hundred dollars and they do 90, people worry that they haven’t hit—that there might be a flaw in the business or something like that.”)). But see Def. Reply 56.1 § 266 (citing Piecyk deposition at 36, denying considering consensus numbers in his analysis). AT&T internally tracked how its performance compared with consensus. Viola testified that AT&T did not want to miss consensus. Viola Inv. Depo. at 224~25. And in 2015 and 2016, senior AT&T executives were generally made aware of how AT&T performed with respect to the analyst consensus estimates; they, at times along with members of the board of directors, 20 Philip Cusick, a JP Morgan analyst, testified that, over a two-year period on an EIP plan, costs are offset by revenue. Krumholz Decl., Ex. 11 (“Cusick Depo.”) at 85. Batya Levi, a UBS analyst reporting to Hodulik, testified that under an EIP plan, “[i]t could be” profit-neutral “over time.” Krumholz Decl., Ex. 16 (“Levi Depo.”) at 66. 30 reviewed AT&T’s performance as compared to the Bloomberg and FactSet consensus estimates. JSF Jf 143-44. In Q3 2015, AT&T missed consensus revenue estimates. Jd. 164. On October 22, 2015, AT&T’s CEO wrote to the board of directors with a “note regarding our quarterly revenue of $39B,” P. Ex. 8 at 2. He said that analyst consensus revenue estimates as reported by the company First Call—-a reporting service that tracks analyst estimates—would likely be $41 billion, because a majority of the analysts First Call covered “modeled a full 3 months of revenue from DIRECTV,” even though AT&T “only had a little over 2 months of combined revenue, as the deal closed on July 24.” Jd. He further explained that “John Stephens and Lori Lee have been working with First Call and the press to educate them on the issue” and warned the board that “you may hear some in the media call it ‘a revenue miss.’” Jd. Ahead of that anticipated coverage, AT&T had tried to warn the market and analysts through a statement to the Wall Street Journal warning that revenue estimates were too high. Pl. 56.1 § 325 (citing 137, 250; P. Ex. 85 (“Stephens Depo.”) at 178). On October 22, 2015, the CEO emailed CFO Stephens, Lee, and another recipient a CNBC story with the headline “AT&T warns of revenue miss in Q3” and the message: “Sure glad we avoided a big headline about a revenue miss :).” JX 146. AT&T also missed Bloomberg and FactSet consensus revenue estimates in Q4 2015, falling $600 million short. JSF 4170. This was in large part due to analysts’ overestimates of AT&T’s wireless equipment revenue. /d. Ona January 26, 2016 earnings call, AT&T's CFO stated that “equipment revenues were down more than $700 million, mostly due to the lower upgrade volumes” and that “[t]otal wireless revenue was impacted by lower smartphone sales.” JX 46 at 6. Some media and analyst reports observed that AT&T had missed Bloomberg and FactSet consensus estimates for Q4 2015. JSF {174. A January 27, 2016 email to the CEO, 41 CFO, Lee, and Viola included a media digest noting that “most outlets emphasiz[ed] the in-line profits and revenue miss for the quarter.”?! JX 299 at 2. G. AT&T’s Concern About Missing Consensus in QI 2016 Undisputed evidence reflects that, in January 2016, AT&T executives and IR defendants tracked, among other metrics, consensus revenue estimates, and that, in February 2016, they escalated these efforts. It also reflects that, in March 2016, AT&T executives made public statements about such metrics. 1. January Emails and Tracking Activity On January 23, 2016, three days before AT&T’s Q4 2015 results were released, Viola emailed Womack: “[W]e have a tendency to focus on EPS and have recently missed the mark on consolidated revenue, We need to make sure our story gets consensus trued up for both EPS as well as revenue. Make sense?” JX 203 at 6. Later the same evening, after a request from Viola for the information, Womack emailed Viola with revenue estimates from the IR Department and the Top Ten consensus, writing: “[I]t looks like the street is too heavy first half and too light second half [of the year].” Jd. at 3. The next morning, Viola emailed Womack back: “We will have to nip 1Q in the bud otherwise we will be in the same spot we’ve been in the last few quarters, i.e. missing revenue. Need to figure out our story for body eps and revenue.” Jd. at 2, Womack replied: “Agreed—we need to help them figure out their equipment spread for the year. 21 Headlines of these news reports included: “Dow Jones: AT&T Misses on 4Q Top Line While Losing Phone Customers Again -- Market Talk,” “Fast? T: AT&T shares slip after results,” “Reuters: AT&T revenue below forecasts, shares fall,” “Associated Press: AT&T Misses Street 4Q Forecasts,” “MarketWatch: AT&T shares slip as revenue falls short of Wall Street’s estimates,” “Barron’s: AT&T Rising: Q4 Rev Misses, EPS In-Line,” “Zack’s: AT&T (T) Slips on Q4 Earnings Miss,” “24/7 Wall Street: AT&T Earnings Report Doesn’t Measure Up to Expectations,” “RTT News: AT&T Inc. (T) Is Losing Ground After Q4 Revenues Fell Short,” “RTT News: AT&T Swings to Profit, Revenues Up 22% But Fall Short of Street,” and “CED: AT&T Touts DirecTV Possibilities Despite Revenue Miss.” JX 299 at 3-4. 979 Equipment revenue is starting to become a real challenge.” Jd A few days later, on January 27, 2016, Black emailed Viola, Womack, and Evans with updates on EPS estimates from analysts for Q1 2016. JX 213 at3. Viola responded, “What about revenue? I’m equally concerned about revenue.” Id. 2. February Emails and Tracking Activity On February 7, 2016, CFO Stephens emailed CEO Stephenson with preliminary internal predictions for Q1 2016. Stephens wrote that the “highlights” included: “Revenue of $13.3B missed budget by $240M—wireless equipment missed by $250M-—EG offset pressure in Bus Sol & International,” and “[w]ireless missed volumes/equipment revenue but made contribution on expense side.” JX 153. On February 8, 2016, Womack emailed Black: “Please send me a revenue report today by analyst for each quarter, exactly like the EPS report that you generate on Fridays.” JX 212. Later that day, Black responded. He wrote: “[A]ttached is the revenue by analyst. I have included this as part of the consensus file that is on the shared drive.” JX 208 at 2. The attachment included a list of brokers, analysts, and their respective estimates, with a separate chart entitled “Top Ten Analysts.” The document listed both “Total Consensus” and “Top Ten” in summary columns. /d. at 3. On February 26, 2016, Black wrote to Viola with updated revenue estimates. He noted that revenue “is lower than total consensus analyst expectations by $193M for the quarter,” that the Top 10 consensus revenue estimate was consistent with “the IR view,” and that many analysts were continuing to update their quarterly estimates. JX 218 at 2. Viola responded to Black the same day, copying Womack: “Guys, what’s the plan to get first call numbers in line with 1Q and full year for revenue and eps? Let’s discuss .. . I don’t want to be in a mad dash in April,” a. 23 On February 27, 2016, Stephens emailed Black, asking for “a summary sheet of where we are on consensus estimates for the first quarter—top 10 and consensus. Need all the regular numbers—revenue, customer metrics, FCF, CAPEX, EPS etc. Debbie, once received let’s compare to 1Q outlook-—with updates for Feb actual—want to have info before I speak at DB [Deutsche Bank] conference on March 9th.” JX 219. Black responded the next day with charts showing the projected increases. JXs 220-22. 3. Early March Conference Statements by Executives, Reactions, and Internal Discussions On March 2, 2016, Stankey-—then head of AT&T’s Entertainment division-—-spoke at an investor conference, run by Morgan Stanley. JSF 181. An analyst there asked Stankey to comment on AT&T’s “handset utilization rates.” The analyst did not specifically ask for numerical upgrade rates, and Stankey did not provide them. Pl. 56.1 356; Def. Reply 56.1 4 356; see IX 397. Stankey replied “that many of AT&T’s customers ‘are no longer thinking that they want to upgrade it every 12 months or 18 months. They are taking care of it. They are going, wow, this isn’t a $199 device, this is a $600 device. I’d better care for it more carefully ... and as a result of that, renewal cycles on that very capable device are extending.” JX 3 4 7(c). Another analyst asked: “I wonder if you could put some data around the point you made on handset renewals.... [W]hat’s the average currently? How is that shifting up and how long do you think that could become?” Id. { 7(e). Stankey responded: “Yes, it’s not something we publicly disclose. I will tell you it’s shifting out and it’s getting longer and | think that is just a trend,” Jd. 4 7(f). In early March 2016, CFO Stephens and others, including Viola, considered issuing a Form 8-K to address, inter alia, lower equipment revenues. JSF 7 186. On March 3, 2016, Stephens emailed Viola and AT&T’s Controller, writing, in part: “After we see Feb Results and 94 before we present at DB conference next week let’s consider an 8k that states ... Wireless equipment unit sales down year over year—impacting equipment revenues.” JX 216, AT&T ultimately decided not to issue a Form 8-K. Instead, Stephens was to address the issue at the forthcoming Deutsche Bank conference. JSF { 188; PL. 56.1 | 363; Def. Reply 56.1 § 363. On March 2, 2016, Viola emailed Womack, Black, and an assistant for a meeting at 2 p.m. on March 4, 2016; the subject was “Consensus Discussion.” JSF 4 183. On March 4, 2016, at 6 p.m., Viola sent Womack and Black a calendar invitation to set up a “Weekly Consensus Discussion” starting on March 10, 2016. fd. 184. On March 7, 2016, AT&T’s Controller emailed Stephens with data on year-to-date and Q1 2016 metrics. JX 223, The email’s attachments showed, among other figures, an internal Q1 forecast of 21.6% total revenue, compared with First Call’s total analyst consensus of 26.8%, and an internal wireless equipment revenue projection of a 25.3% decline, compared with analyst consensus of a 4.5% growth. JX 224, Also on March 7, 2016, AT&T’s Controller emailed the IR Department to say that only 3.1% of AT&T’s postpaid subscriber base had upgraded their phones through February. P. Ex. 88; see Pl. 56.1 369; Def. Reply 56.1 § 369. On March 9, 2016, Stephens spoke at the Deutsche Bank conference. Relevant here, he said the following about upgrade rates and wireless equipment revenue: I think you saw in the fourth quarter, it was a slowdown in the handset upgrade cycle or the total sales, I wouldn’t be surprised to see that continue. . . . ['T]hose are impacts possible on revenues but very little impact at all on profitability because those are all hedged one way or another with the handset expenses. JSF § 191(a). Stephens later added, in response to a question about upgrade rates and cycles: I can only talk about up through the fourth quarter... . What we are seeing on an overall basis though is on average customers holding their phones longer and probably what I would suggest is a more important fact that I can point to is prior to Next [the EIP program], we had about 90,000 to 100,000 a quarter of customers bringing their own devices and saying can you hook up this device? ... Now most 35 recently last year for most of the year we were at $350,000 to $400,000 a quarter showing that customers are valuing that old device and reusing it. JSF € 191(b). After these remarks, Deutsche Bank published a research note. JSF 7195. It mentioned Stephens’s presentation and stated that “some revenue headwinds like lower handset volumes... appear to have continued in early 2016, [but] these do not impact margin/profitability improvements.” Dkt. 107 (Def. 56.1”) 32; Dkt. 194 (“SEC Resp, 56.1") J 32; see JX 59. Gabriella Brown, a UBS employee, emailed within UBS: “AT&T Corp.: CFO: Expect slowdown in phone sales to continue in FY 2016—conf comments-—what is he referencing?” Def. 56.1 § 34; SEC Resp. 56.1 34; see JX 57. UBS analyst Batya Levi responded to that email: “Slow volumes ie handset adds and upgrades. Installment plans and lack of iconic device launches are keeping volumes low. VZ [Verizon] said the same yesterday.” Def. 56.1 | 36; SEC Resp. 56.1 4 46; see JX 57. Also on March 9, 2016, Womack emailed Viola and Black, in separate emails. Womack wrote first to Viola, noting that “the combination” of the EIP program and the lack of a new iPhone model in which customers would be interested had “the potential to drive historically-low upgrade rates in the first half of this year.” P. Ex. 12. Asa result, Womack wrote that the difference could be “$1.5B less in equipment revenue; a substantial reduction in no-margin equipment revenue YoY.” Jd) Womack forwarded the message to Black. See JX 161. As of March 9, 2016, no analyst covering AT&T had published a report reflecting a 5% upgrade rate year-over-year. JSF 4 135. H. March 9, 2016 to April 25, 2016: Defendants Discuss Q1 2016 with Analysts On March 9, 2016, the IR Department, via Womack, Viola, and Black, began communicating with analysts about AT&T’s Qi performance. These communications-—-which 26 continued through the eve of AT&T’s 1Q16 earnings announcement, made on April 26, 2016— are the basis for the SEC’s allegation that the defendants disclosed MNPI to “walk down” the analysts’ estimates for that quarter. It is undisputed that, before AT&T announced its 1Q16 financial results, the following data about AT&T’s Q1 performance was not publicly available: total consolidated revenue, wireless equipment revenue, and the wireless equipment upgrade rate. JSF § 97; Pl. 56.1 4 98; Def. Reply 56.1 4 98. Throughout this period, there were also internal communications in the IR Department about such outreach. These stressed the importance of “walking down” analysts’ estimates for Q1. For instance, although Black has testified that he does not remember the specific analyst calls at issue here, he acknowledged that “the focus in [Q1 2016] was to ensure that the analysts” were “aware of the comments from John [Stephens], and underst[ood] the impact of the equipment revenue.” Pl. 56.1 J] 382, 473 (citing P. Ex. 52 (Black deposition in SEC investigation) at 100-01, 137). But see Def. Reply 56.1 {382 (providing additional context). The following summarizes the record evidence as to (1) each individual defendant’s calls to analysts and (2) internal communications within AT&T about the emerging Q1 data and the effort to “walk down” the analyst consensus. 1. March 9, 2016 — March 15, 2016 On March 9, 2016, Womack spoke to JP Morgan analysts Phil Cusick and Richard Choe, and UBS analyst John Hodulik. Cusick and Choe’s respective notes from these conversations contained similar figures: an upgrade rate of 5%, that equipment revenue could be down more than JP Morgan’s estimate of 15% year-over-year, and that consolidated revenue would be closer 27 to $40b than $40.7b,”* Cusick testified that it was “very likely” that Womack told him the 5% figure “[b]ecause I wouldn’t have written two exclamation points if there hadn’t been some statement from him that that was the case.” P. Ex. 51 at 131-32. Cusick and Hodulik’s notes also refer to the upgrade rate as “historically low” or “record low.” JP Morgan would eventually (on March 22, 2016) update its Q1 estimates to bring them closer to these numbers—among other things, lowering its upgrade rate from 6.5% to 5%, and its total revenue from $40,665b to $40.023b. See JX 305. After the call, Hodulik emailed UBS analysts: “AT&T (which sells the most iPhones in the US) is saying it will prob see record low upgrade rates for the next 3 qtrs. It’s due to the move to installment phone sales.” P, Ex. 13. The next day, March 10, 2016, within AT&T, Evans emailed Womack, “The upgrade rate will not be in the 3% range... it could be as low as having a 4.X% look but more in the mid to high 4%’s.” JX 197. Womack also emailed Viola, Black, and Evans that Evans had “confirmed the forecasted upgrade rate for the quarter is in the 4.5% range,” which would “reduce equipment revenue by ~1B YoY for the quarter.” TX 181, Evans replied to all: “Iam guessing the rate will be closer to 5% but [| Womack’s] math works better.” Jd. On March 14, 2016, Womack emailed Viola, Evans, and Black with projections of equipment revenue based on a range of upgrade rates. He noted that Evans “believes that the rate most likely will fall in the 5-5.25% range which drives a revenue decline of ~625 YoY.” JX 22 Cusick’s notes stated: “Historically low u/g next couple of quarters. > The Court does not have occasion to consider whether other provisions of the Exchange Act could have authorized the rulemaking. See, e.g., 15 U.S.C. § 78w(a)(1) (empowering SEC to “make such rules and regulations as may be necessary or appropriate to implement the provisions of this chapter for which they are responsible or for the execution of the functions vested in them by this chapter”). 83 agent of an issuer who discloses material nonpublic information in breach of a duty of trust or confidence to the issuer shall not be considered to be acting on behalf of the issuer.” Jd. (emphasis added). Seizing on that exception, defendants argue that an intentional or reckless disclosure of MNPI inherently bespeaks a breach of a duty of trust or confidence to the issuer. The exception, they argue, thus does not leave Reg FD any room to operate. Defendants are mistaken. An intentional or reckless disclosure of MNPI, although breaching Reg FD if not remedied with the required dispatch, does not inherently breach a duty of trust or confidence to the issuer. That is because Reg FD—unlike the prohibition on insider trading under Section 10b and Rule 10b-5 as developed by the case law—is not premised on a theory of a breach of a preexisting duty fo the issuer, Defendants do not cite any authority that corporate officers, employees, and agents have an ex officio, inherent, duty to their employers to make disclosures of MNPI on a market-wide basis only. The obligation to refrain from selective disclosures of MNPI instead is a regulatory one, deriving from Reg FD, As the SEC explained in promulgating Reg FD, the exception for disclosures of MNPI in breach of a duty or confidence is instead aimed at a narrower class of selective disclosures: those in breach of an existing duty to the issuer, with the obvious example being selective disclosures made for personal gain, such as in exchange for money or in connection with insider trading. See Adopting Release, 2000 WL 1201556, at *9 (“Thus, an issuer is not responsible under Regulation FD when one of its employees improperly trades or tips.”)..”) Thus understood, Reg FD’s exception for selective disclosures in breach of a duty to the issuer does not swallow the rule or make it inoperable. Accord Kisor v. Wilkie, 139 5. Ct. 2400, 2414 (2019) (“The drafters will know what [the regulation] was supposed to include or exclude or how it was supposed to R4 apply to some problem.”); Forest Watch v. U.S. Forest Serv., 410 F.3d 115, 117 (2d Cir. 2005) (according deference to agency interpretation of its own rules). As the SEC further explained when it proposed Reg FD, the rule was designed to reach selective disclosure of MNPI by corporate officers and employees “acting within the scope of their authority.” See Reg FD Proposing Release, 64 Fed. Reg. 72,590, at 72,594, 1999 WL 1255550 (Dec. 28, 1999) (“By focusing on employees and agents acting within the scope of their authority, the Rule would make an issuer responsible only for the disclosures of company officials, employees, or agents who are properly authorized to speak to the media, the analyst community, and/or investors.”). After public comment, the SEC modified the definition “to make it more precise,” specifying in the final rule that it applies to “any senior official of the issuer... or any other officer, employee, or agent who regularly communicates with .. securities market professionals or security holders.” Adopting Release, 2000 WL 1201556, at *9; see 17 C_F.R. § 243.101(c). Defendants do not—-and cannot seriously--argue that all selective disclosures within this definition breach an inherent duty of trust and confidence to the issuer. This case, in fact, supplies an excellent example of the contrary. As alleged, the selective disclosures in this case were by IR defendants authorized to speak with the analyst community, and acting not only within the scope of their employment but at the instigation—f not the direction—of corporate superiors at AT&T, including the CFO and IR head, for the company’s ostensible (if misbegotten) benefit. See Adopting Release, 2000 WL 1201556, at *9 (“[T]o the extent that another employee had been directed to make a selective disclosure by a member of senior management, that member of senior management would be responsible for having made 85 the selective disclosure.” (emphasis in original)). No party argues that, if liable, Womack, Evans and Black breached a duty of trust and confidence to AT&T, The Court therefore rejects defendants’ claim that Reg FD is logically inoperable. V. Cross-Motions for Summary Judgment The Court next turns to the summary judgment motions directed to the evidence adduced as to SEC’s claims. As to each of the three elements in dispute—that the information in question was (1) material; (2) nonpublic; and (3) selectively disclosed with the requisite scienter—the parties argue that a reasonable jury evaluating the evidence could find only in their favor. For the reasons that follow, the Court finds that although the evidence overwhelmingly supports the SEC as to the materiality and nonpublic elements, a reasonable jury could find for either side on the third element, scienter. These findings compel denial of both sides’ summary judgment motions. A. Legal Standard for Summary Judgment To prevail on a motion for summary judgment, the movant must “show[] that there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law.” Fed. R. Civ. P. 56(a); see Celotex Corp. v. Catrett, 477 U.S. 317, 322-23 (1986), The movant bears the burden of proving the absence of a question of material fact. In making this determination, the Court must view all facts “in the light most favorable” to the non-moving party. Holcomb v. Jona Coll., 521 F.3d 130, 132 (2d Cir. 2008). If the movant meets its burden, “the nonmoving party must come forward with admissible evidence sufficient to raise a genuine issue of fact for trial in order to avoid summary judgment.” Jaramillo v. Weyerhaeuser Co., 536 F.3d 140, 145 (2d Cir. 2008). “[A] party may not rely on mere speculation or conjecture as to the true nature of the facts to overcome a motion for summary judgment.” Hicks v. Baines, 593 F.3d 159, 166 (2d Cir. 2010) (citation omitted). 86 Rather, to survive a summary judgment motion, the opposing party must establish a genuine issue of fact by “citing to particular parts of materials in the record.” Fed. R. Civ. P. 56(¢)(I)(A); see also Wright v. Goord, 554 F.3d 255, 266 (2d Cir. 2009). “Only disputes over facts that might affect the outcome of the suit under the governing law” will preclude a grant of summary judgment. Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248 (1986). In determining whether there are genuine issues of material fact, a court is “required to resolve all ambiguities and draw all permissible factual inferences in favor of the party against whom summary judgment is sought.” Johnson v. Killian, 680 F.3d 234, 236 (2d Cir, 2012) (quoting Terry v. Ashcroft, 336 F.3d 128, 137 (2d Cir. 2003)); see Donoghue y. Oaktree Specialty Lending Corp., No. 21 Civ. 4770 (PAE), 2022 WL 1213639, at *7 (S.D.N.Y. Apr. 25, 2022). B. Regulation FD’s Elements Regulation FD states in relevant part: Whenever an issuer, or any person acting on its behalf, discloses any material nonpublic information regarding that issuer or its securities to any person described in paragraph (b)(1) of this section, the issuer shall make public disclosure of that information .. . (1) Simultaneously, in the case of an intentional disclosure; and (2) Promptly, in the case of an non-intentional disclosure. 17 C.E.R. § 243.100(a). Reg FD applies to a disclosure made to any person outside the issuer, including: a broker or dealer, investment adviser or institutional investment manager, and anyone who holds the issuer’s securities “under circumstances in which it is reasonably foreseeable that the person will purchase or sell the issuer’s securities on the basis of that information.” Id. § 243.100(b)(1). The parties do not dispute that (1) Womack, Evans, and Black were persons “acting on behalf of’ AT&T under Reg FD, see id. § 243.101(c); (2) the analysts with whom Womack, Evans, and Black communicated were covered by Reg FD, see id. § 243.100(b){1); or (3) after QF the defendants’ alleged selective disclosures, AT&T did not “publicly disclose” the information at issue, see id. § 243.101(e) (“public disclosure” requirement means furnishing or filing with the Commission a Form 8-K disclosing the information or disseminating the information through another method or combination of methods reasonably designed to provide broad distribution to the public). The parties’ dispute on summary judgment centers on the materiality, nonpublic, and scienter elements. As noted, as to the terms “material” and “nonpublic,” Reg FD “relies on existing definitions of these terms established in the caselaw.” Adopting Release, 2000 WL 1201556, at *9 & nn. 38-40 (citing cases). C. Materiality 1. Legal Standards “Although the SEC declined to set forth an all-inclusive list of what matters are to be deemed material, it did provide seven categories of information or events that have a higher probability of being considered material.” Siebel, 384 F. Supp. 2d at 702. The categories are: (1) [elarnings information; (2) mergers, acquisitions, tender offers, joint ventures, or changes in assets; (3) new products or discoveries, or developments regarding customers or supplies (¢.g., the acquisition or loss of a contract), (4) changes in control or in management; (5) change in auditors or auditor notification that the issuer may no longer rely on an auditor's audit report; (6) events regarding the issuer’s securities—e.g., defaults on strict securities, calls of securities for redemption, repurchase plans, stock splits or changes in dividends, changes to rights of security holders, public or private sales of additional securities; and (7) bankruptcies and receiverships. Adopting Release, 2000 WL 1201556, at *10.°° 56 Although leaving open that other types of information could qualify as material under Reg □□□ the SEC has signaled that it would not pursue “close calls” in this area. As Judge Daniels recounted in Siebel: RR Under the existing precedents for materiality incorporated by Reg FD, information is material when there is “a substantial likelihood that the disclosure of the omitted fact would have been viewed by the reasonable investor as having significantly altered the total mix of information made available.” Basic, 485 U.S. at 231-32; see also Matrixx Initiatives, Inc. v. Siracusano, 563 U.S. 27, 38 (2011). The “total mix” of information includes all information that is reasonably available to the public. Starr v. Georgeson S’holder, Inc., 412 F.3d 103, 110 Cir, 2005) (quoting Press v. Quick & Reilly, Inc., 218 F.3d 121, 130 (2d Cir. 2000)). As the Supreme Court has explained, a lower standard—-such as defining a “material fact” as any “fact On November 1, 2000, approximately a week after Regulation FD became effective, then-Director of the SEC Division of Enforcement, Richard H. Walker, gave a speech before the Compliance and Legal Division of the Securities Industry Association regarding the enforcement perspective of Regulation FD. Former- Director Walker indicated that in enforcing Regulation FD, the SEC was “not going to second-guess close calls regarding the materiality of a potential disclosure” noting that “[aJn issuer’s incorrect determination that information is not material must represent an extreme departure from standards of reasonable care in order for [the SEC] to allege a violation of FD.” [See] 2000 WL 1635668, at *3. He further stated, “Regulation FD’s adopting release . . . spells out seven items that should be reviewed carefully to determine whether they are material... This list puts the world on notice that an intentional or reckless selective disclosure of information falling into one of these categories is likely to draw the attention of the Enforcement Division.” fd. The views expressed in Mr. Walker’s speech were his own and did not necessarily reflect the views of the SEC or its staff. Jd. at *1. However, in May of 2001, the SEC quoted and summarized portions of Mr. Walker’s speech, wherein he indicated “that Regulation FD was not designed as a ‘trap for the unwary’ and that enforcement cases will not be based on second-guessing reasonable judgments made in good faith by issuers, including judgments about materiality.” Written Statement Concerning Regulation Fair Disclosure of the U.S. Securities and Exchange Commission Before the Subcommittee on Capital Markets, Insurance and Government Sponsored Enterprises Committee on Financial Services United States House of Representatives, 2001 WL 634672, at *7 (May 17, 2002). The SEC noted that “[t]hese remarks .. . have indicated that Commission enforcement of the regulation will be focused on clear violations.” Jd. Siebel, 384 F. Supp. 2d at 702 n.8. 89 which a reasonable shareholder might consider important”—would lead corporations to “bury the shareholders in an avalanche of trivial information[,] a result that is hardly conducive to informed decisionmaking.” TSC Indus., 426 U.S. at 448-49. “Information that would affect the probable future of the company and which may affect an investor’s desire to buy, sell, or hold the company’s securities is material.” Siebel, 384 F. Supp. 2d at 703 (citing Tex. Gulf Sulphur, 401 F.2d at 849). A reasonable investor need not have necessarily have changed his or her investment decision as a result of the information, for that information to qualify as material. TSC Indus., 426 U.S. at 449; SE.C. vy. Mayhew, 121 F.3d 44, 52 (2d Cir. 1997) (citations omitted); see also Folger Adam Co. v. PMI Indus., Inc., 938 F.2d 1529, 1533 (2d Cir. 1991) (a “material fact need not be outcome-determinative”; it is sufficient that it would assume actual significance to a reasonable investor), The “materiality hurdle” is, thus, “a meaningful pleading obstacle.” Jn re ProShares Tr. Sec. Litig., 728 F.3d 96, 102 (2d Cir. 2013). The Second Circuit has “consistently rejected a formulaic approach to assessing... materiality.” Litwin v. Blackstone Grp., L.P., 634 F.3d 706, 717-18 (2d Cir. 2011) (quoting Ganino vy. Citizens Utils. 228 F.3d 154, 162 (2d Cir, 2000)); see also ECA & Loc, 134 IBEW Joint Pension Tr. v. JP Morgan Chase Co., 553 F.3d 187, 204 (2d Cir. 2009) (ECA & Local 134”) (“While Ganino held that bright-line numerical tests for materiality are inappropriate, it did not exclude analysis based on, or even emphasis of, quantitative considerations.”). The Second Circuit has, however, cited with approval the SEC Staff Accounting Bulletin No. 99 (“SAB No, 99”), which offers guidance as to assessing materiality. See ECA, 553 F.3d at 197~ 98; Ganino, 228 F.3d at 163-64. Relevant here, it advises: The use of a percentage as a numerical threshold, such as 5%, may provide the basis for a preliminary assumption that... deviation of less than the specified percentage with 90 respect to a particular item . .. is unlikely to be material . . . . But quantifying, in percentage terms, the magnitude of a misstatement . . . cannot appropriately be used as a substitute for a full analysis of all relevant considerations. SAB No. 99, 1999 WL 1123073, at *2; see also ECA, 553 F.3d at 204 (a “five percent numerical threshold is a good starting place for assessing . . . materiality”). In the end, “a court must consider ‘both “quantitative” and “qualitative” factors in assessing an item’s materiality,’ and that consideration should be undertaken in an integrative manner.” Litwin, 634 F.3d at 717 (quoting SAB No. 99, 1999 WL 1123073, at *2) (citing Ganino, 228 F.3d at 163; and in re Kidder Peabody Sec. Litig., 10 F. Supp. 2d 398, 410-11 (S.D.N.Y. 1998)); SAB No, 99, 1999 WL, 1123073, at *3 (“Qualitative factors may cause misstatements of quantitatively small amounts to be material.”). “(Since the importance of a particular piece of information depends on the context in which it is given, materiality has become one of the most unpredictable and elusive concepts of the federal securities laws.” S.E.C. v. Bausch & Lomb Inc., 565 F.2d 8, 10 (2d Cir, 1977). “Summary judgment on matters of materiality in a securities fraud case is appropriate when the omissions and misrepresentations in question are ‘so obviously important to the investor, that reasonable minds cannot differ on the question of materiality.’” S.E.C. v. Credit Bancorp, Ltd., 195 F, Supp. 2d 475, 492 (S.D.N.Y. 2002) (SEC. v. Rsch, Automation Corp., 585 F.2d 31, 35 (2d Cir. 1978)). Whether information is material is a mixed question of law and fact that is especially well suited for jury determination. See United States v. Bilzerian, 926 F.2d 1285, 1298 (2d Cir. 1991) (citing TSC Indus., 426 U.S. at 450 (“The determination requires delicate assessments of the inferences a ‘reasonable shareholder’ would draw from a given set of facts... and these assessments are peculiarly ones for the trier of fact.”)). 9] 2. Application The SEC argues that the information selectively disclosed here was material for multiple reasons, considered separately and together: the information (1) enabled AT&T to meet (and exceed) analysts’ consensus expectations for Q1 2016; and (2) alerted analysts in advance about three important performance metrics—all unexpectedly low—~for that quarter: AT&T’s consolidated total revenue, wireless equipment revenue, and wireless upgrade rates, Although the materiality inquiry is holistic, for clarity, the Court analyzes these factors serially: a. Meeting analysts’ consensus expectations As chronicled above, there is overwhelming evidence, documentary and testimonial, on which a jury could find that in March and April 2016, AT&T undertook a campaign, choreographed at high levels of the company, to walk down analysts’ estimates. The campaign’s purpose was to ensure that AT&T’s actual financial results did not underperform the consensus estimates for a third consecutive quarter. As IR Department head Viola wrote, AT&T had “recently missed the mark on consolidated revenue[ and]... need[ed] to make sure [its] story [got] consensus trued up for both EPS as well as revenue.” JX 203 at 6. The implicit assumption underlying the campaign was thus that, unless the analysts could be induced to lower their estimates, AT&T would miss consensus, and potentially badly. Insofar as the IR defendants’ disclosures were found to convey to analysts the fact that AT&T was on track to miss the then-consensus, ample authority—in both accounting literature and case law—supported that such information was material. See, e.g., SAB No. 99, 1999 WL, 1123073, at *3 (“Among the considerations that may well render material a qualitatively small misstatement of a financial statement are... whether the misstatement hides a failure to meet analysts’ consensus expectations for the enterprise.”); ECA, 553 F.3d at 204 (quantitative assessment of materiality “must be supplemented” by evaluation of SAB No. 99’s qualitative 0? factors); Ganino, 228 F.3d at 163 (“[W]hether the misstatement hides a failure to meet analysts’ consensus expectations” is a SAB No. 99 qualitative factor.); see also SE.C. v. Huang, 684 F. App’x 167, 172-73 (3d Cir. 2017) (upholding finding of materiality where “insider information gave [defendant] early and nonpublic insight into whether the companies were likely to under- or over-perform expert predictions”); S.C. v. Todd, 642 F.3d 1207, 1225 (9th Cir. 2011) (“The misrepresentations were material because by including the transactions, Gateway was able to meet analysts’ expectations for the third quarter of 2000.”); Backman vy. Polaroid Corp., 893 F.2d 1405, 1422 (1st Cir. 1990) (“Hard facts indicating that lower than projected sales actually are occurring clearly constitute material information vis-A-vis mere predictions by analysts of the likelihood of sales difficulties.”); Fresno Cnty. Emps.’ Ret. Ass’n v. comScore, Inc., 268 F. Supp. 3d 526, 549-50 (S.D.N.Y. 2017) (denying motion to dismiss for failure to plead materiality where “misstatements also plausibly masked a change in earnings or other trends and hid a failure to meet analysts’ consensus expectations for the enterprise”) (cleaned up); SAAC. v. Winemaster, 529 F, Supp. 3d 880, 909-10 (N.D. IIL. 2021) (finding “‘analysts’ consensus net revenue estimate an important metric for a company’s performance and [that] failure to meet that projection would provoke a negative reaction from investors .... Thus, although the amount of revenue recognized from any particular subject transaction may have been quantitatively small, it played an important role . . . in demonstrating that its financial performance was meeting expectations”).°’ In moving for summary judgment on the grounds that the SEC cannot prove this element, defendants argue that the negligible movement in AT&T’s stock price in prior quarters when it >? These authorities rebut defendants’ attempt to minimize the “shock of a miss” as adding nothing to the “information itself.” AT&T MSJ Opp’n at 33. 93 missed consensus establishes the immateriality of the company’s Q1 2016 miss. In the fourth quarter of 2015, for example, AT&T announced that it had missed equipment revenue by “more than $700 million” year-over-year, “mostly due to lower upgrade volumes.” JSF 171. This event, a defense expert opines, was a “natural experiment” to test whether missing consensus is material. Dkt. 111 (Allen) 971. Defendants argue that it was not, because the stock price was barely different—it closed only $0.08 higher (0.225%)—-the day after the Q4 2015 announcement. JSF 4] 173. Although an event’s stock price impact can surely be germane to the materiality inquiry, under the case law, it is no more than a relevant factor and is rarely dispositive of materiality, particularly to the extent that it would mandate entry of summary judgment. See Bilzerian, 926 F.2d at 1298 (“[W]hether a public company’s stock price moves up or down or stays the same... does not establish the materiality of the statements made, though stock movement is a factor the jury may consider relevant.”); S.E.C. v. Monterosso, 768 F. Supp. 2d 1244, 1265 (8.D. Fla. 2011) (“[T]he movement of a company’s stock price, or lack thereof, is not dispositive of whether a given statement is material. Rather, whether a public company’s stock price moves up or down is simply a factor that may be relevant to materiality.”) (citing Bilzerian, 926 F.2d at 1298), aff'd, 756 F.3d 1326 (ilth Cir. 2014); Veleron Holding, BV. v. Morgan Stanley, 117 □□ Supp. 3d 404, 434 (S.D.N.Y. 2015) (“Nor can the lack of a significant drop in the price of [a company’s] stock after disclosure by itself establish immateriality as a matter of law.” (internal citations and quotation marks omitted)); S.E.C. v. Stanard, No. 06 Civ. 7736 (GEL), 2009 WL 196023, at *23 (S.D.N.Y. Jan. 27, 2009); SEC. v. Penthouse Int'l, Inc., 390 F. Supp. 2d 344, 353 (S.D.N.Y. 2005) (“[T]here is no requirement to allege or demonstrate any particular movement in a company’s stock price in order to sustain the element of materiality.”); S.E.C. v. 94 DCT Telecomms., 122 F. Supp. 2d 495, 499 (S.D.N.Y. 2000) (“There is no requirement that stock prices fluctuate as a result of a defendant’s misstatements or omissions in order for them to be material.”); Manavazian vy. Atec Group, Inc., 160 F. Supp. 2d 468, 483-84 (E.D.N.Y. 2001). Further, as the SEC points out, here, the record contains evidence refuting the defense’s interpretation of the lack of stock-price movement after the Q4 2015 earnings announcement. This would support a jury in finding that this lack of movement did not signify a lack of market concern about AT&T’s latest miss. As the SEC notes, there is contemporaneous evidence that positive factors the same day—in particular, favorable news in the same announcement about AT&T’s then-recent merger with DirecT V—overshadowed the bad news, causing AT&T’s stock ptice to move up slightly. See, e.g, P. Ex. 106 at 2 (Drexel Hamilton report describing doubled customer acquisitions across international wireless business and that, due to the DirecTV merger, AT&T’s dividend payout ratio was historically high); P. Ex. 110 at 2 (Nomura report describing record-high EBITDA service margins); P. Ex. 119 at 3 (JP Morgan report noting churn rate beating expectations). Cf In re Eaton Corp. Sec. Litig., 318 F. Supp. 3d 659, 670 (S.D.N_Y. 2018) (“[T}here is reason to discount the relevance of the stock drop to the materiality of the defendants’ statements because the drop occurred on a day when other negative news regarding the Company’s finances was published.”). In sum, although a finder of fact could reach different conclusions as to the significance, ifany, of AT&T’s stock-price inactivity after the Q4 2015 announcement, the evidence as to this point, even if construed in the defense’s favor, would not come close to securing summary judgment. The evidence that AT&T’s disclosures enabled it to avoid again missing consensus supplies a sufficient basis for the SEC’s claims to reach a jury. 95 b. Total consolidated revenue The Court next turns to the materiality of the individual performance metrics the SEC claims AT&T selectively disclosed, beginning with total consolidated revenue. There is ample evidence on which the jury could find these selective disclosures material, which independently precludes summary judgment for the defense on this element. As chronicled above, there is overwhelming evidence, including identical data appearing in multiple analysts’ notes of calls with defendants, that in the March and April 2016 campaign to walk down analyst consensus, AT&T disclosed its estimated or eventual total revenue for Q1 2016. See SEC 56.1 App. C (chart depicting Q1 2016 quarterly total revenue estimate for each analyst firm before and after speaking with an IR defendant). Reg FD’s Adopting Release identifies “earnings information” as an area—the first it lists—carrying a heightened probability of being considered material. See 2000 WL 1201556, at *10. In contending that AT&T’s total revenue was immaterial, defendants seize on the absence of the word “revenue” in the Adopting Release’s list. They argue that while “earnings” data is material, total revenue data is not. That argument is more semantic than persuasive. The Adopting Release cannot credibly be read to imply the general non-materiality of total revenue data. To state the obvious, total revenue is half the equation by which “earnings” are tabulated. And the Adopting Release (whose other listed examples concern events such as mergers or new products, as opposed to purporting to comprehensively list potentially material financial metrics) is explicit that “it is not possible to create an exhaustive list” of potentially material areas of information, See id Defendants’ notion that a reasonable investor would be unconcerned with such data is also impeached by a 2012 AT&T internal guidance memo, prepared by its legal department. It advised that “[e]ven information that might usually be considered non-material 96 may be treated differently close to quarterly earnings announcements. This is particularly true for data that might be used to help analysts or investors calculate expected results.” See JX 286 at 2. That a reasonable investor could consider an issuer’s total revenue, and particularly a year-over-year drop in total revenue, in making an investment decision is underscored by both the accounting literature and the case law. Material misstatements of “a registrant’s revenues” have been identified as “an obvious example” of a materially misleading statement, “even if the effect on earnings is completely offset by an equivalent overstatement of expenses,” SAB No. 99, 1999 WL 1123073, at *5; see SEC. v. Reyes, 491 F. Supp. 2d 906, 910 (N.D. Cal. 2007) (“{R]evenues and cash flow .. . are widely, if not universally, regarded as the best indicators of a company’s financial health.”); Monterosso, 768 F. Supp. 2d at 1264 (“[R]evenue is generally considered an important indicator of a company’s financial health.”); of Panther Partners Inc. v. Ikanos Comm’ns, Inc., 681 F.3d 114, 116 (2d Cir. 2012) (holding that Regulation S-K compels disclosure of “defects [that] constituted a known trend or uncertainty that the Company reasonably expected would have a material unfavorable impact on revenues” and comparing disclosure obligation to “materiality under the federal securities laws’ anti-fraud provisions”). In this case, the SEC has adduced voluminous evidence that total revenue demonstrably mattered within AT&T, in general and during the events at issue. This context, too, is indicative of the materiality of this metric. See Mayhew, 121 F.3d at 52 (“[A] major factor in determining whether information was material is the importance attached to it by those who knew about it.”) (citing Tex, Gulf Sulphur, 401 F.2d at 851). AT&T internally identified revenue as a “key short term financial metric[] for the operation of our business” and an “important key metric[] to our stockholders.” JX 251 at 43. It O7 keyed its executive incentive compensation to revenue. See id. at 47. And AT&T’s policies and training documents instructed employees that revenues and sales were material to investors. See, e.g., JX 143 at 4 (guidelines for internal dissemination of financial and operating information, stating that “[m]aterial information can be quantitative—such as revenue or access line data’); JX 268 at 4 (training document giving as an example of material information “[p]redictions of future revenues”); JX 286 at 2 (policies stating that material information can include “[f]inancial results such as revenue”); JX 267 at 13 (‘Complying with Reg FD” training document identifying “[rJevenue or income data, including for business units,” and “[s]ales for a significant product or service (e.g. U-verse, iPhone)” as examples of material information); FX 269 at 5 (listing revenues first in a list of items that “are important for AT&T as a whole”). The documentary evidence adduced by the SEC of the events surrounding Q1 2016 further supports that AT&T’s March and April 2016 campaign was specifically undertaken to avoid missing analysts’ consensus as fo total revenue, During early 2016, AT&T IR department officials repeatedly emphasized the importance of total revenue to the campaign to sway analysts. For example, on January 23, 2016, IR head Viola emailed colleagues that AT&T “need[ed] to make sure our story gets consensus trued up for both EPS as well as revenue.” JX 203 at 6. On January 24, 2016, Viola emailed Womack, identifying as a goal to “nip 1Q in the bud” to avoid “befing] in the same spot we’ve been in the last few quarters, i.e. missing revenue.” /d, at 2; see also infra Sections LF-G (detailing AT&T’s previous revenue misses and concerns about missing revenue again); JX 203 at 2~3 (discussing importance of “figur[ing] out our story for body eps and revenue” in light of “the street [being] too heavy first half and too light second half’), On January 27, 2016, Black emailed Viola, Womack, and Evans with updates on earnings estimates, in response to which Viola asked, ““What about revenue? I’m 98 equally concerned about revenue.” JX 213 at 3. On February 8, 2016, Womack requested that Black send over “a revenue report today by analyst for each quarter, exactly like the EPS report that you generate on Fridays.” JX 212. On February 26, 2016, Viola emailed Black, ce’ing Womack, about “the plan to get first call numbers in line with 1Q and full year for revenue and eps” so as to avoid being “in a mad dash in April.” JX 218, And, on the eve of the earnings announcement, CFO Stephens emailed CEO Stephenson, after a pair of analysts revised their estimates, stating, “[t]hese two updates may do it for us—we may beat revenue consensus—not by much but a beat nonetheless”—to which Stephenson responded, “Good.” JX 13 13(b). A jury could easily find the fact that the company’s top leaders effectively “spiked the football” upon learning that AT&T had beat analysts’ total revenue estimates to be memorable confirmation of the importance of that data.*® Given the assembled documentary record of the importance that the executive suite at AT&T and the defendants attached to the company’s total revenue in Q1 2016, the evidence that the metric was material, far from being insufficient, is overwhelming. To a reasonable juror, the demonstrated salience of this performance metric to the architects and implementers of AT&T's campaign to work down the analysts’ consensus could easily obliterate defendants’ abstract notion that only earnings data—not revenue—can be material. See Mayhew, 121 F.3d at 52 (“[A] major factor in determining whether information was material is the importance attached to it by those who knew about it.”); Lilly v. State Tchrs. Ret. Sys. of Ohio Pension Fund, 608 F.2d °8 For the same reasons, a reasonable juror could easily put aside defendants’ semantic claim that even though total revenue was “important,” it was not necessarily “material.” See AT&T MSJ Opp’n at 35 n.119 (citing Kuebler v. Vectren Corp., No. 18 Civ. 113, 2018 WL 4003626, at *3 (S.D. Ind. Aug. 22, 2018) (“[Jjust because a piece of information is important does not necessarily mean it is material.”); and then citing Himmel v. Bucyrus Intern., Inc., No, 10 Civ. 1104, 2014 WL 1406279, at *17 (E.D. Wisc. Apr. 11, 2014)). O90 55, 58 (2d Cir. 1979) (“[T]he manner in which the information was regarded by those privy to it and the importance attached to the information by the recipients . . . were entirely consistent with a conclusion that the information was material information.”); S.E.C. v. Geon Indus., Inc., 531 F.2d 39, 48 (2d Cir. 1976) (Friendly, J.) (noting that “fi]n cases of the disclosure of inside information to a favored few, determination of materiality has a different aspect” than in other cases, as “the information takes on an added charge just because it is inside information”; and, in upholding trial court’s determination that a tip violated Rule 10b-5, noting the “demonstrated . . . importance [insiders] attached to the [disclosed] information”); S.Z.C. v. Shapiro, 494 F.2d 1301, 1307 (2d Cir. 1974) (“The behavior of appellant, his partner Shapiro, and others who knew of the merger, all of whom were sophisticated investors, demonstrates empirically that the information was material.”); S.E.C. v. Drescher, No. 99 Civ. 1418 (SAS), 1999 WL 946864, at *5 (S.D.N.Y. Oct. 19, 1999) (“Information coming from an insider takes on special importance.”); see also United States v. Contorinis, 692 F.3d 136, 143 (2d Cir. 2012) (‘Information also comes in varying degrees of specificity and reliability, and the extent to which a newly reported item of information alters the total mix may depend on the specificity or reliability of that information. ... To the extent that appellant’s suggested charges focused entirely on the content of reports or tips, excluding from consideration the reliability of the source, they misstated the law.”). c. Equipment revenue and upgrade rates Defendants next attempt to minimize the materiality of AT&T’s wireless equipment revenue and wireless upgrade rates. The analyst notes and other evidence, reviewed above, are plentiful that the IR defendants selectively disclosed data as to these metrics. Although these metrics are idiosyncratic and narrow such that they ordinarily would not form the basis of a 100 finding of materiality, the evidence the SEC has adduced in this case would enable a jury to so find here. This supplies an additional basis to deny defendants summary judgment to the extent they move based on materiality. The parties’ arguments on these metrics fall into four categories. Quantitative significance: Defendants emphasize what they term the quantitative insignificance of AT&T’s equipment revenues and upgrade rates. They (1) note that wireless equipment revenue accounted for less than 10% of AT&T’s total revenue in 1Q15, see JSF {| 93-95; (2) argue that the information allegedly disclosed during the calls implicated less than 5% of AT&T’s earnings, see AT&T MSJ Opp’n at 31 (citing, inter alia, SEC expert DiBucci, who opined that the difference in upgrade rates and equipment revenues as disclosed would impact AT&T’s total revenues and net incomes by less than 5%, see Dkt. 147-4 at 120-29, 203~ 05); and (3) specific to upgrade rates, argue that “the difference between a 5.7% upgrade rate (the consensus estimate before the alleged campaign began, see JSF { 114, and a 5.0% upgrade rate” is not material. Although these arguments potentially could gain traction with a jury, as a basis for granting defendants summary judgment, they are unpersuasive. As a matter of law, a 5% impact on a financial metric is not mechanically dispositive as to materiality.’ And in the context of the package of disclosures that enabled AT&T to lower analysts’ revenue consensus so as to enable it to avoid a third miss in four quarters, a jury could See Basic, 485 U.S. at 236 n.14 & accompanying text; Ganino, 228 F.3d at 162 (“Following Basic, we have consistently rejected a formulaic approach to assessing the materiality of an alleged misrepresentation.”); ECA, 553 F.3d at 197 (“[B]right-line numerical tests for materiality are inappropriate.” (internal citations and quotations omitted)); Kidder Peabody, 10 F. Supp. 2d at 410 (declining to hold as a matter of law that misstatements impacting profits by no more than 2.54% were immaterial). 101 find these performance metrics influential. The evidence would support the finding that analysts’ outsize projections of AT&T’s equipment revenue and upgrade rates largely drove the consensus as to total revenue for Q1 2016 before the March and April 2016 campaign. The analysts’ downward estimates of AT&T’s equipment revenue in particular appear to have driven most of the lowered estimates of total revenue that enabled AT&T to meet and beat consensus. See Pl. 56.1 App. B; Adopting Release, 2000 WL 1201556, at *11 (“[A]n issuer cannot render information immaterial simply by breaking it into ostensibly non-material pieces.”). Accordingly, viewing equipment revenue or upgrade rates in context and not in a vacuum, a jury could find them highly consequential to the analysts’—and hence investors’ assessment of AT&T’s quarterly performance. See Litwin, 634 F.3d at 720 (“Even where a misstatement or omission may be quantitatively small compared to a registrant’s firm-wide financial results, its significance to a particularly important [business] segment . . . tends to show its materiality.”); S.E.C. v. DiMaria, 207 F. Supp. 3d 343, 353 (S.D.N.Y. 2016) (“Courts are required to consider qualitative factors, which ‘can turn a quantitatively immaterial statement into a material statement.’” (quoting JBEW Loc. Union No. 58 Pension Tr. Fund & Annuity Fund v. Royal Bank of Scot, Grp., PLC, 783 F.3d 383, 391 (2d Cir. 2015))); United States v. Ferguson, 553 F. Supp. 2d 145, 153~55 (D. Conn. 2008) (finding, on summary judgment, that “quantitatively insignificant” losses did not warrant entering summary judgment for defendant where, inter alia, analysts believed company’s misstatements to be important and misstatements were designed, in part, “to hide its failure to meet analyst expectations”). 69 For similar reasons, defendants’ claim that only revenue from unsubsidized sales of handsets should be considered, see AT&T MSJ Mot. at 25-27, does not establish immateriality. 10 Significance within AT&T: Bolstering the SEC’s claim that these metrics were material is the significance AT&T attached to them. Equipment revenue featured in the company’s earnings calls and in emails from C-suite executives. See JX 46 at 4 (Jan. 26, 2016 earnings call in which CFO states, “Consolidated revenues grew to $42.1 billion. That’s up more than 22% year over year, mostly due to our acquisition of DIRECTV. That growth comes even with lower equipment sales, as customers chose to hold onto their smartphones for a longer period of time”); DiMaria, 207 F, Supp. 3d at 354 (company highlighting data in opening paragraphs of earnings release tended to show materiality); see also JX 46 at 6 (“Total wireless revenue was impacted by lower smartphone sales.”); P. Ex. 48 (January 16, 2016 email from Stephens to Viola stating, “need to understand upgrade rte — 500 bp [basis point] drop is a big deal for long term profits — as churn is stable”), Significantly, too, AT&T discussed wireless equipment revenue as part of its disclosures under SEC Regulation S-K-—-which, as relevant here, obligates disclosure of “any known trends .. . reasonably likely to have a material favorable or unfavorable impact on net sales or revenues or income from continued operations.” 17 C.F.R. § 229.303(b)(2)(ii). In a paragraph titled “2016 Revenue Trends” of AT&T’s FY-2015 Form 10-K, AT&T stated, “[oJur AT&T Next [EIP] program is expected to generate continued growth in equipment revenue, which has the corresponding impact of lowering service revenues.” JX 1 at 48; see also id. at 14 (“Wireless data services continue to be a growing area of Consumer Mobility’s business, representing an increasing share of overall subscriber revenue. Subscribers continue to upgrade their handsets to more advanced integrated devices, contributing to growth from wireless data services.”); JX 6 at 15 (FY-2016 Form 10-K stating the same). 103 AT&T also disclosed upgrade rates in investor presentations released contemporaneously with its earnings releases, see, e.g., JSF { 96; JX 47 at 12 (“Total wireless revenues were down 4.9% year over year to $18.9 billion, largely due to decreases in equipment revenue. . . . Wireless operating income was $4.4 billion, up 33.5% year over year largely due to stabilizing service revenues, lower smartphone upgrade volumes, and lower expenses driven by efficiencies.”). The company there touted decreased upgrade rates as contributing to more than $1 billion in savings in sales and commissions expenses. See JX 1 at 39; JX 2 at 26, 30, A reasonable juror could find AT&T’s repeated disclosures about these subjects alongside its financial releases indicative of materiality. See Kidder Peabody, 10 F. Supp. 2d at 410 (“[Bly their very nature, financial reports are relevant to investment decisions.”). A reasonable jury, too, could view AT&T’s treatment of declining equipment revenues during March and April 2016 as indicating the acute significance the company attached to these metrics. In early March 2016, CFO Stephens and others, including Viola, considered issuing a Form 8-K. to address, among other subjects, lower equipment revenues. JSF { 186; JX 216 (email from Stephens to Viola and AT&T’s Controller, stating, “After we see Feb Results and before we present at DB conference next week let’s consider an 8k that states ... Wireless equipment unit sales down year over year—impacting equipment revenues.”), AT&T ultimately decided not to do so. Instead, after internal discussion, Stephens opted to address declining equipment revenue at the March 9, 2016 Deutsche Bank conference. Stephens Depo. at 211-12; JSF § 188. Stephens’s statements about wireless equipment revenue were intended as part of the company’s efforts to walk down analysts’ estimates. See AT&T MSJ Mot. at 19 (“Following Stephens’ public statements, AT&T’s IR employees had follow-up conversations with individual analysts between March 9, 2016, and April 21, 2016, concerning what Stephens told the market.”); see also TX 203 at 2-3, 6 (Womack, in response to email from Viola explaining that AT&T “need[ed] to make sure our story gets consensus trued up for both EPS as well as revenue,” stating that “we need to help them figure out their equipment spread for the year. Equipment revenue is starting to become a real challenge”). Internally, too, AT&T executives’ communications bespoke concern about equipment revenues and upgrade rates. The issue was elevated to the company’s board.*! CFO Stephens alerted CEO Stephenson that a “highlight” of the first quarter’s preliminary results was that “Revenue of $13.3B missed budget by $240M-—-wireless equipment missed by $250M” and that “Wireless missed volumes/equipment revenue but made contribution on expense side.” JX 153. And he internally described the decline in upgrade rates as “a big deal for long term profits.” P. Ex. 48 at 1. See DiMaria, 207 F. Supp. 3d at 354 (“[A] fter all, if these two measures were so insignificant to the reasonable investor, why was [the defendant] so focused on them?”). Significance to the market. There is also ample record evidence that wireless equipment revenue and upgrade rates mattered to analysts. Many addressed these metrics in their reports to investors, presenting them as relevant to AT&T’s profitability. See, e.g., JX 167 at 24 (Nomura analyst testifying that “for every dollar that they sell on a phone, they lose 15 cents”); JX 64 at 3 (UBS extolling importance of longer upgrade cycles, which “improve[] churn and lower gross adds while boosting carrier profitability”); JX 111 at 9-10 (Moffett Nathanson describing declining upgrade rates as a “critical factor” in assessing profitability outlook); JX 76 (Citigroup increasing earnings estimates based on lower upgrades); JX 305 (similar, for JP Morgan). 51 See Stephens Depo. at 84-86; see also P. Ex. 92 (CEO Stephenson telling board that “lower handset upgrade volumes... certainly impacts revenue growth rates,” but “we continue to see the offsetting, positive effect to earnings”). TOS To be sure, there is evidence on this point on which a jury could rely to find AT&T’s performance on these metrics immaterial. There was analyst testimony that they did not view declining equipment revenue and/or upgrade rates as likely to affect AT&T’s profit margins. See sources cited at AT&T MSJ Mot. at 22-23; AT&T MSJ Opp’n at 15 n.44 (citations to analyst testimony on the subject),16 n.45; AT&T MSJ Mot. at 24 (citations to updated reports on the subject, published after calls with IR defendants). In the end, however, there is sufficient evidence on which a jury could find the data that the [R defendants disclosed as to these metrics material. Wireless equipment revenue’s impact en earnings: Defendants, finally, emphasize the evidence that wireless equipment revenues were profit neutral, or “pass through.” One analyst’s deposition testimony explained this thesis: Q., “Explain to the ladies and gentlemen of the jury why you, as an analyst of 20- plus years covering AT&T, wouldn’t think this is material?” A. “Because I would, that is probably going to have if, if wireless equipment revenue is down 20 percent, probably my cost of equipment is going to be down a similar amount, and there will be no change to my EBITDA or EPS estimates that really drive my price target and recommendation on the day.” Q. “In your mind, as an analyst working the numbers, doing the analysis here, even if someone told you, “I think equipment revenue is going to be down 20 percent,” to you that wouldn’t be material, because it’s going to be offset by declining costs?” A. “That’s right.” 106 Levi Depo, at 97, Toward this same end, defendants supply a declaration from CFO Stephens, attesting that wireless equipment revenue had little impact on profitability. See Dkt. 109 (“Stephens Decl.”), Although a jury could rely on this testimony as a basis for finding wireless equipment revenues not to be material, this evidence cannot secure summary judgment for the defense, given the weighty contrary evidence set out above. And it is established that revenues can be material even if not influencing earnings. See, e.g., SAB No. 99, 1999 WL 1123073, at *5 “To take an obvious example, if a registrant’s revenues are a material financial statement item and if they are materially overstated, the financial statements taken as a whole will be materially misleading even if the effect on earnings is completely offset by an equivalent overstatement of expenses.”); see also S.E.C. v. Huang, 186 F. Supp. 3d 380, 383 (E.D. Pa. 2016) (denying defendant’s motion for judgment as a matter of law following jury finding that nonpublic information regarding credit card revenue data was material), aff'd, 684 F. App’x 167. 82 See also sources cited at AT&T MSI Mot. at 22-23 and AT&T MSJ Opp’n at 15 n.44 (citations to analyst testimony that equipment revenue does not impact earnings); JX 69 at 5 (Deutsche Bank report describing decreased equipment revenue as “optical in nature”); JX 70 at 1 (Wells Fargo report describing “equipment revenue [a]s margin neutral, or essentially a pass through”); JX 94 at 7 (Pacific Crest report describing “equipment revenue [a]s basically a pass- through account”); AT&T MSJ Mot. at 22--23 (citations to similar analyst testimony). 63 The SEC has indicated that it will move before trial to exclude Stephens’s declaration. See SEC MSJ Opp’n at 19-20 (terming it an “argumentative and uncorroborated declaration that conflicts with the record evidence and should be stricken and/or disregarded”). For purposes of the instant motions, the Court assumes arguendo that testimony along these lines would be received. See Macuba v. Deboer, 193 F.3d 1316, 1323 (11th Cir. 1999) (court may “consider a hearsay statement in passing on a motion for summary judgment if the statement could be reduced to admissible evidence at trial or reduced to admissible form”); Schmidt v. DIRECTY, LLC, No. 14 Civ. 3000, 2017 WL 3575849, at *1 (D. Minn. Aug. 17, 2017) (holding, on summary judgment, that a declaration “need not comply with Fed. R. Evid. 1006 at this stage of the proceedings”). ET The SEC has, in any event, mustered contrary evidence on this point, drawn from sources including AT&T financial statements, board presentations, and witness testimony. See sources cited at Pl. 56.1 99 91, 99-111, 117, 119-30, 165; JX 47 at 12 (investor presentation released with its earnings release stating that the mobility division enjoyed a 22.5% year-over-year improvement in operating income “largely due to. . . lower handset upgrade volumes”); P. Ex. 97 at 5—6 (1Q16 earnings call, during which CFO explained, inter alia, that “laser focus on cost efficiencies and fewer upgrades drove our best ever first-quarter wireless EBITDA margins”); see also SEC MSJ Opp’n at 16~18 (record evidence of analysts atiributing changes in estimates to lower upgrade rates). d. Materiality: overall assessment For the reasons canvassed above, the SEC has amassed overwhelming evidence supporting the materiality of the data it alleges AT&T selectively disclosed in March and April 2016. To the extent that defendants have moved for summary judgment based on this element, that motion is meritless, D. Nonpublic The allegedly nonpublic information disclosed on defendants’ calls with analysts falls into two categories: (1) the three specific metrics addressed above (AT&T’s total revenue, equipment revenue, and upgrade rates); and (2) statements amplifying on the public remarks Stephens had given at the March 9, 2016 Deutsche Bank conference. Each side moves for summary judgment on this element. Black makes a separate argument why summary judgment, particular to him, is warranted. TOR For the reasons reviewed below, there is overwhelming evidence on which a jury could find that the IR defendants selectively disclosed nonpublic information to analysts. Defendants’ motions for summary judgment based on this element, including Black’s, therefore lack merit. 1, Legal Standard Reg FD’s Adopting Release provides that information “is nonpublic if it has not been disseminated in a manner making it available to investors generally.” Adopting Release, 2000 WL 1201556, at *9 (citing Tex. Gulf Sulphur, 401 F.2d at 854). The rule “relies on existing definitions of the[] term[] established in the case law.” Jd. Under that case law, “[i]nformation becomes public when disclosed ‘to achieve a broad dissemination to the investing public generally and without favoring any special person or proup.”” Mayhew, 121 F.3d at 50 (quoting Dirks, 463 U.S. at 653 n.12), “[I]nformation is nonpublic if it is not available to the public through such sources as press releases, Securities and Exchange Commission filings, trade publications, analysts’ reports, newspapers, magazines, rumors, word of mouth or other sources,” United States v. Cusimano, 123 F.3d 83, 89 n.6 (2d Cir, 1997) (internal quotation marks omitted); see Contorinis, 692 F.3d at 142-44. Information also becomes public “when, although known only by a few persons, their trading on it ‘has caused the information to be fully impounded into the price of the particular stock.’” Mayhew, 121 F.3d at 50 (quoting United States v. Libera, 989 F.2d 596, 601 (2d Cir, 1993)); see Contorinis, 692 F.3d at 143. Reg FD’s Adopting Release further instructs that selective disclosures revealing that an issuer’s results will be “higher than, lower than, or even the same as what analysts have been forecasting ... will have likely violated Regulation FD.” Adopting Release, 2000 WL 1201556, at *11. An issuer and its agents cannot privately disclose this information “through indirect 109 ‘suidance,’ the meaning of which is apparent though implied,” id., or “by breaking it into ostensibly non-material pieces,” id.; see also Proposing Release, 64 Fed. Reg, 72,590, at 72,595 (“[I]ssuers should avoid giving guidance or express warnings to analysts .. . about important upcoming earnings or sales figures.”). 2 Application Before April 26, 2016, AT&T undisputedly had not publicly revealed the Q1 2016 results at issue—its projected or actual total revenue, wireless equipment revenue, and wireless upgrade rates, See JSF 97, 116, P. Ex. 66 (Black Deposition) at 64-65; P. Ex. 71 (Evans Deposition) at 146; P. Ex. 88 (Womack Deposition) at 104—05. Defendants nonetheless make a series of arguments why they are entitled to summary jadgment on this element. Defendants first dispute that a reasonable jury could find that the IR defendants’ communications with analysts included such information. In fact, based on the analyst notes and testimony, there is abundant evidence on which a reasonable juror could so find. As reviewed above, nearly identical numbers regarding AT&T’s revenues and upgrade rates appear across different analysts’ notes of their calls with IR defendants. And numerous analysts testified that an IR defendant did, expressly or impliedly, reveal such metrics to them. See, e.g., P. Ex. 51 at 131-32 (JP Morgan analyst testifying that it was “very likely” that Womack told him the 5% upgrade rate figure because he wouldn’t have written two exclamation points [in his notes] if there hadn’t been some statement from [Womack] that that was the case’’); P. Ex. 103, Sine Decl. 12 (“The ‘consensus figures’ that Michael Black emphasized to me on the March 24 call signaled to me that my revenue estimates for AT&T’s first quarter 2016 might be too high and played an important role in my decision to reduce revenue estimates for AT&T.”), 15 (updating 10H estimated quarterly upgrade rate to 5% after speaking with Black); P. Ex. 56 at 225 (“They were — you know, clearly they were working toward a number.”); P. Ex. 67 at 20 (“I would say it is highly unusual to have gotten a call like that, with that information, at that juncture before earnings.”’). Defendants next argue that although AT&T had not publicly disclosed its preliminary or actual results, analysts and investors could have extrapolated these from public information. See Siebel, 384 F. Supp. 2d at 706 (holding fact public where “the information available to the public provides a sufficient factual basis for a reasonable investor” to infer it). Defendants’ arguments to this effect center on the 5% upgrade rate. As to it, defendants contend that (1) the downward trends in upgrade rates and equipment revenues were known to analysts, and (2) “simply applying the same 1.4% year-over-year drop that occurred from 3Q 2014 to 3Q 2015 (from 7.1% to 5.7%) to the 6.6% upgrade rate for the first quarter of 2015 would yield a project[ed] 5.2% upgrade rate for the first quarter of 2016—-substantially the same as what the analysts projected after the purported selective disclosures.” AT&T MSJ Mot. at 36 (internal footnotes omitted), This argument is based largely on the constructs and opinions of defendants’ experts, which the Court, for purposes of this motion, treats as admissible. But for two reasons, it cannot secure summary judgment for defendants on this element, and does not negate, or meaningfully offset, the SEC’s evidence. First, this argument applies only to one item of data: the 5% upgrade rate. Defendants, critically, do not argue that the other data (including AT&T’s total revenue) that the defendants allegedly disclosed was already in the public domain. Second, insofar as defendants suggest that nothing nonpublic was conveyed in the calls, and that an analyst who made the right assumptions could have deduced AT&T’s Q1 results at issue, that claim is at odds with the evidence of the lil analysts’ behavior. The evidence does not reflect that any analyst, or other AT&T outsider, had performed such an exercise and arrived at this result. On the contrary, it reflects that, after their conversations with the defendants, the analysts—far from finding “nothing to see here” repeatedly adjusted their estimates downward. See generally supra Section I.H-I, Pl. 56.1, App’x A-B (chart detailing analysts’ adjustment to equipment revenue and upgrade rates shortly after contact with AT&T IR defendants). The record also does not supply a nonspeculative basis to claim that these analysts each independently and serendipitously calculated the company’s Q1 2016 results, as opposed to being told them by Womack, Evans, and Black. On the evidence adduced, a reasonable juror would not have any evidentiary basis to so conclude. Defendants next argue that the IR defendants did little more than direct analysts to the information CFO Stephens had publicly disclosed on March 9, 2016 at the Deutsche Bank conference. There, Stephens suggested, at a high level of generality, that declines in equipment revenues and wireless upgrade rates were likely to continue. His words were: I think you saw in the fourth quarter [of 2015], it was a slowdown in the handset upgrade cycle or the total sales. I wouldn’t be surprised to see that continue... . [T]hose are impacts possible on revenues but very little impact at all on profitability because those are all hedged one way or another with the handset expenses. . . . I can only talk about [upgrade rates and cycles] up through the fourth quarter [of 2015].... What we are seeing on an overall basis though is on average customers holding their phones longer and probably what I would suggest is a more important fact that I can point to is prior to Next [the EIP program], we had about 90,000 to 100,000 a quarter of customers bringing their own devices and saying can you hook up this device? ., . Now most recently last year for most of the year we were at $350,000 to $400,000 a quarter showing that customers are valuing that old device and reusing it. 191(a)-(b). As a basis for summary judgment for the defense, this argument fails. Although Stephens’s remarks were public, they self-evidently did not disclose the data which the SEC contends—-and the analysts’ notes corroborate-—-was later selectively disclosed to individual 117 analysts, They did not come close to doing so. Stephens’s remarks did not disclose AT&T’s Qi 2016 total revenue, its equipment revenue, or its equipment upgrade rate. On the contrary, they were statistically non-specific (e.g., “I wouldn’t be surprised to see [a slowdown in the handset upgrade cycle or the total sales] continue.”). And, to the extent Stephens recited specific data, this data concerned the prior quarter, whose results AT&T had already publicly reported (e.g., “I can only talk about [upgrade rates and cycles] up through the fourth quarter [of 2015].”). Contrary to defendants’ portrayal, this was thus emphatically not an instance in which “simple arithmetic computation based on [quantitative] information [previously] disclosed” would reveal the same information. See Nguyen v. MaxPoint Interactive, Inc., 234 F. Supp. 3d 540, 547 (S.D.N.Y. 2017). The analysts’ ensuing actions further undermines the theory that Stephens, on March 9, 2016, effectively disclosed the data the SEC contends Womack, Evans, and Black selectively disclosed over the next six weeks. As the SEC notes, many analysts did update their models following the conference.© But they updated these again—with materially worsened forecasts 64 The record, in fact, suggests that some analysts had already incorporated into their models the general trends noted by Stephens. See, e.g., P. Ex. 59 at 112 (‘Q. Would this be the working model as of... March 11, 2016? A. Yes. Q. And does it incorporate the most recent updates as reflected in [a March 10, 2016 research note]? A. Yes.”); P. Ex. 150 at 72-73, 87-88 (analyst testifying that they forecasted year-over-year declines in upgrade rates before Stephens’s remarks and that his comments were “not new news”); P. Ex. 118 (Oppenheimer report noting low upgrade rates and reflecting changes to estimates following 4Q15 earnings announcement); P. Ex. 119 (similar, in JP Morgan report). As reviewed above, documentary and other evidence suggests that the failure of Stephens’s remarks to cause the analyst community to downwardly adjust projections for Q1 2016 to the degree necessary to meet AT&T’s internal forecasts catalyzed the campaign of calls by Womack, Evans, and Black beginning in mid-March. 65 Stephens’s remarks undisputedly conveyed some news. Three weeks beforehand, the company’s 2015 Form 10-K had stated that: “Our Next program is expected to generate continued growth in equipment revenue in 2016.” See JX 1 at 48 (2015 Form 10-K) (emphasis added). 113 for Q1 2016’s revenue—after their conversations with an IR defendant. See, e.g., JSF | 240, JXs 76, 228-31, 234-35 (Citi analysts updated their model on March 11, 2016, and updated their model again to lower specific estimates following a March 31, 2016 call with Black); JSF [4 245-46, P. Exs. 23-24, 28, Xs 82, 186 (similar, for Moffett Nathanson analyst who updated model on April 1, 2016, spoke with an AT&T IR employee on April 6, 2016, and updated estimates on April 7, 2016); see also sources cited at Pl. 56.1 [| 706-21 (similar evidence as to William Blair), 722-45 (similar evidence as to Pacific Crest). Defendants downplay these analysts’ actions, noting that more “influential” analysts from UBS and Deutsche Bank “immediately concluded” from Stephens’s remarks that the downward trend in upgrade rates was to continue into Q1 2016, and updated their reports soon thereafter (March 18, 2016 for UBS; March 21, 2016 for Deutsche Bank), AT&T MSJ Opp’n at 39-40; see IXs 64, 69; see also JX 305 (JP Morgan updating report on March 22, 2016). But that does not advance defendants’ claim that the ensuing disclosures by the IR defendants covered only information Stephens had already made public. On the contrary, each of those “influential” firms updated its model for AT&T after later hearing from Womack. See supra Sections LH.1— 2. And a JP Morgan analyst to whom Womack spoke testified it was “very likely” that Womack told him that AT&T’s upgrade rate was 5%. P. Ex. 51 at 131-32. In light of this evidence, a reasonable juror could easily infer that, consistent with the analyst notes, fresh nonpublic information was disclosed on the calls, whether explicitly or by indirect but equally 6° The IR employee in question, Martin Sheehan, is not a defendant here. Documentary evidence supports, however, that he acted at Viola’s direction. See JX 186 (email from Viola to Sheehan instructing Sheehan, among other things, to “remind” the Moffett Nathanson analyst that their “1Q16 revenue numbers are way off due to his assumptions re wireless equipment revenue”). 114 impermissible “guidance,”®’ leading the analysts to more fully grasp the magnitude of AT&T’s forthcoming bad news to which Stephens had only generally gestured on March 9, 2016. And it is undisputed that between Stephens’s public remarks on March 9, 2016 and the company’s earnings release on April 26, 2016, “AT&T did not publicly quantify, reiterate or otherwise amplify [on] Stephens’s remarks to the market at large.” SEC MSJ Opp’n at 24. Defendants next argue that they are entitled to summary judgment for claims based on calls to analysts after UBS—the first analyst firm to revise its estimates after a call with an IR defendant—-published its updated report on March 18, 2016. That argument presupposes that upon UBS’s report, AT&T’s internal business metrics, to the extent reflected therein, became public. That is incorrect. The evidence, including UBS’s report, does not show that UBS anywhere attributed its March 18 forecasts or analyses to AT&T. UBS did not state that its updated estimates derived from information from within AT&T, however obtained. See JXs 274, 278. Under these circumstances, the public did not have any way to know that UBS, in publishing its estimates, was tacitly channeling as its estimates data that had selectively and secretly been fed to it by AT&T’s Womack. And UBS, for its part—and the other analyst firms to whom IR defendants thereafter allegedly shared nonpublic Q1 2016 projections and results— had material information the public lacked: that AT&T had revealed important data as to its Q1 performance and what that data was. See Backman, 893 F.2d at 1422 (“Hard facts indicating that 5? See Adopting Release, 2000 WL 1201556, at *11 (nonpublic information cannot be privately disclosed “through indirect ‘guidance,’ the meaning of which is apparent though implied”); Jn re Motorola, 2002 WL 31650174, at *2 (“Issuers may not evade the public disclosure requirements of Regulation FD by using ‘code’ words or ‘winks and nods’ to convey material nonpublic information during private conversations. What is particularly troubling about this case is that Motorola communicated to the public using general terms such as ‘significant,’ and then engaged in private discussions with analysts to provide a more detailed quantitative definition of the code word ‘significant.’”). 114 lower than projected sales actually are occurring clearly constitute material information vis-a-vis mere predictions by analysts of the likelihood of sales difficulties.”); Contorinis, 692 F.3d at 143 (“Information also comes in varying degrees of specificity and reliability, and the extent to which a newly reported item of information alters the total mix may depend on the specificity or reliability of that information.”); DeMarco v. Lehman Bros. Inc., 222 F.R.D, 243, 246 (S.D.N.Y. 2004) (“{T}here is a qualitative difference between a statement of fact emanating from an issuer and a statement of opinion emanating from a research analyst.”). In a final argument on this element, Black argues that he “did nothing more than convey to analysts consensus figures compiled by him” pursuant to a “practice he had employed for many years and had no reason to believe was improper.” ID MSJ Mot. at 21; see also id. at 20— 24. Black asserts that data he disclosed as AT&T’s upgrade rates in fact consisted of consensus figures he had compiled on his “Top Ten” charts, which contained the published estimates of 10 top analyst firms covering AT&T. See Pl. 56.1 § 281; P. Ex. 66 at 128-29, P. Ex. 85 at 76, Black argues that, even though his Top Ten charts were not publicly disclosed, insofar as his calculations of the “Top Ten” consensus derived from publicly available estimates by analyst firms, his disclosures were not of nonpublic information. 68 Defendants cite Contorinis for the proposition that “sometimes a corporation is willing to make information available to securities analysts ... even though it may never have appeared in any newspaper publication or other publication. Such information would be public.” 692 F.3d at 142. But the jury instruction to that effect, which the Second Circuit upheld, emphasized that, assessing whether information is nonpublic, the keyword is ‘available.’ . . . For example, if UBS policy was to give out certain information to people who ask for it, that information is public information.” Id. The evidence here is that defendants reached out to select analyst firms and disclosed nonpublic information, to induce them to lower estimates. There is no evidence that any analyst publicly sourced the disclosed data to AT&T or that AT&T or the individual defendants expected the analyst to do so. Under such circumstances, the information at issue-— that AT&T had reported Q1 2016 results—would not have been publicly available until April 26, 2016, when AT&T released those results. 114 Black’s argument cannot win him summary judgment, even assuming his contention— that the data he disclosed was actually his tabulation of publicly available consensus estimates of analysts—was found by a jury to apply to every disclosure he made to an analyst during the March and April 2016 campaign. That is because the SEC has adduced contrary evidence: that what Black conveyed were AT&T’s internal projections of total revenue, equipment revenue, and upgrade rates, and that, presumably to give him plausible deniability, Black disguised these as “consensus” figures. See Pl. 56.1 □□ 606-21; id. J] 588-602. A reasonable juror could so find based on record evidence. Black’s March 22, 2016 calls to analysts from Wells Fargo and RBC supply a useful illustration. At that time, Black’s “Top Ten” charts showed consensus figures of -1.6% (wireless equipment revenue year-over-year decline), 6.1% (upgrade rate), and 25.3% (total revenue growth), whereas AT&T’s internal estimates showed -11.5% (wireless equipment revenue year-over-year decline), 4.7% (upgrade rate), and 23.7% (total revenue growth). See sources cited supra Section 1.11.2. The Wells Fargo and RBC analysts’ notes reflect that Black disclosed figures that did not match his “consensus” numbers, but were figures closer to AT&T’s internal estimates: -18—21% (wireless equipment revenue year-over-year decline, Wells Fargo), -15—18% (wireless equipment revenue yeat-over- year decline, RBC); 5% (wireless equipment upgrade rate); and 22.5--23% (total revenue growth). Jd. Black’s March 24, 2016 call to a Drexel Hamilton analyst is similarly instructive. At the time, the consensus upgrade rate was 6.4%, whereas AT&T’s internal projected upgrade rate was 4.7%, See sources cited id. Section LH.3. After speaking with Black, the Drexel Hamilton analyst emailed a colleague, writing, “they are proactively reaching out to analysts to reinforce comments CFO made at conferences[.] Consensus phone upgrade rate in 1Q around 117 5%” P. Ex. 103 8. A juror could reasonably infer that Black, sub silentio, was disclosing AT&T’s internal forecasts, and dressing these up as analyst “consensus.” Black counters that the upgrade rate the SEC alleges that he conveyed to analysts in these calls, 5%, did not precisely match—it was marginally above—AT&T’s actual internal projection at that time of 4.7%, weakening the claim that Black was disclosing internal AT&T data. See JX 132 3. That, however, does not preclude the SEC from reaching a jury as to its claim against Black, For one, the evidence reflects that Black revealed revenue data—total and equipment—in addition to upgrade rates. For another, there is evidence that Black spoke with analysts and reported a 5% upgrade rate at a time when AT&T internally estimated its upgrade rate at 5% or between 5—5.25%, see JXs 165,171~72—close to (or exactly) AT&T’s internal numbers.® For yet another, Black’s Top Ten consensus figure during the relevant time period was 6.1%, see JX 132 at 3, much farther afield than the number Black disclosed as, ostensibly, the Top Ten consensus figure. As for why Black disclosed an upgrade rate close to, but not precisely matching, AT&T’s internal calculations, a jury could reasonably conclude that Black did so strategically, to avoid the heightened risk that would come from disclosing AT&T’s precise internal forecast. As support, the SEC can point to Black’s arguably disingenuous response to an email from a Nomura analyst asking “where is consensus on 1Q handset upgrade rate?”, to which Black responded that “folks that have update[d] recently [project] 5%,” JX 256. In fact, his tabulation of consensus at the time was 6.1%, and AT&T’s actual projection at the time was 4.7%. And the ® Analysts’ notes during Black’s conversations reflect, in general, numbers strikingly similar to AT&T’s projections during those periods. See supra Sections [.H.2, 1H.5 (detailing evidence surrounding Black’s communications with Wells Fargo, RBC, Scotiabank, Buckingham, and Bank of America analysts). 118 “folks that have update[d] recently” to whom Black argues he was referring to in the email-— analysts from JP Morgan and UBS—had updated their upgrade rate estimations to an average of 5.05%—after speaking privately with Womack. See supra Sections I.H.1—2. The SEC’s theory is thus that Black, in the guise of reporting consensus, directed analysts to the reports of analysts who had already shaped their public reports in response to his colleagues’ selective disclosures. See SEC MSJ Opp’n at 35-36. In sum, there is ample evidence—circumstantial as well as direct—on which a jury could find that Black knowingly revealed AT&T’s nonpublic estimates of material aspects of its results, while falsely holding those out as market consensus data. In sum, there is overwhelming evidence on which a reasonable jury could find that, in March and April 2016, the individual defendants revealed nonpublic data about AT&T’s QI 2016 performance to analysts. The Court thus denies defendants’ motions for summary judgment to the extent that they are based on this element. E. Scienter Unlike the preceding two elements, as to which the evidence lopsidedly supports the SEC’s claim, there is substantial evidence on which a jury could find for either side as to the scienter element. That a reasonable jury could find for defendants on this element requires denial of the SEC’s motion for summary judgment. 1L, Legal Standard Under Reg FD, an issuer must “[s]imultaneously, in the case of an intentional disclosure,” publicize the MNPI selectively disclosed. 17 C.F.R. § 243.100(a)(1).” “A selective disclosure of material nonpublic information is ‘intentional’ when the person making the 7 Reg FD alternatively provides that an issuer must “[p]romptly, in the case of a non-intentional disclosure,” publicize the MNPI selectively disclosed. 17 C.F.R. § 243.100({a)(2). The Court understands the SEC solely to be alleging intentional disclosures here. 119 disclosure either knows, or is reckless in not knowing, that the information he or she is communicating is both material and nonpublic.” Jd § 243.101(a). Recklessness, in turn, is “conduct which is highly unreasonable and which represents an extreme departure from the standards of ordinary care.” S.E.C. v. Frohling, 851 F.3d 132, 136 (2d Cir. 2016) (quoting SEC. v. Obus, 693 F.3d 276, 286 (2d Cir. 2012) (internal quotation marks omitted)). Giving content to this element, Reg FD’s Adopting Release “emphasize[s] that the definition of ‘intentional’ .. . requires that the individual making the disclosure must know (or be reckless in not knowing) that he or she would be communicating information that was both material and nonpublic.” Adopting Release, 2000 WL 1201556, at *12 (emphasis in original). It adds that, in assessing whether a defendant acted with that state of mind, “the circumstances in which a selective disclosure is made may be important.” Jd. The Adopting Release also addresses the implications of an issuer’s efforts to comply with the regulation. “[I]n view of the definition of recklessness that is prevalent in the federal courts, it is unlikely that issuers engaged in good-faith efforts to comply with the regulation will be considered to have acted recklessly.” /d. 2. Application The element of scienter here turns on whether the IR defendants knew, or were reckless in not knowing, that the data about AT&T’s Q1 2016 performance which they communicated to analysts in March and April 2016 was both material and nonpublic. For this discussion, the Court assumes arguendo that a jury would find each selective disclosure alleged by the SEC (including as to total revenue, wireless equipment revenue, and wireless upgrade rates) to have been made, and that the information disclosed was material and nonpublic. For the reasons below, there is a basis in material fact on which a reasonable jury could find for either side as to whether those disclosures were made with the requisite scienter. 120 A brief summary of the central evidence and inferences on which a jury could so find, first for the SEC and then for defendants, follows. Evidence and inference supporting scienter: A jury—tfinding the facts for the SEC could find scienter based on the following: the sheer number of times the defendants disclosed confidential AT&T performance data relating to Q1 2016; the variety of internal data that the defendants disclosed; the sustained (six-week) duration of these disclosures; the persistence of these numeric disclosures until AT&T had finally brought the analyst consensus into line with its forthcoming earnings announcement; defendants’ numerous violations of internal rules relevant to Reg FD, which provided, among other things, that nonpublic financial metrics were not to be disclosed; and the extensive and explicit training the defendants had received as to Reg FD and the duty not to disclose MNPI. Representative of this training was a memorandum from the Davis Polk law firm. Received by Womack, Evans, and Black, it instructed the IR Department that it is improper to “[m]anag[e] the Street’s [e]stimates” by “contact[ing] analysts because the company believed that their estimates were incorrect.” The Davis Polk Memo further advised that “the SEC has indicated that companies that believe that analysts did not understand or absorb the full scope of the message they were trying to convey through their public disclosures should not attempt to correct those analysts through private conversations.” JX 11 at 3—5 (citing Raytheon Co., 2002 WL 31643026; and re Motorola, 2002 WL 31650174). In further support of scienter, a jury could rely on guidance memoranda within the IR Department, whose admonitions appear squarely to prohibit the conduct in which the individual defendants engaged. See, e.g., JX 143 (2013 email from Womack to Black, attaching AT&T IR guidelines emphasizing need to adhere to “laws regarding selective disclosure,” and advising that material information “can be quantitative—such as revenue or access line data,” as well as 121 “qualitative-—for example, statements that sales are ‘very strong’ or that a new initiative ‘is not going well’”). A reasonable jury could find the defendants to have done precisely what their training and guidance had instructed was prohibited.”! A jury could further find—including based on the evidence canvassed above—that the data disclosed was so clearly material, and so clearly nonpublic, that AT&T’s violation of Reg FD was open-and-shut, and that any trained IR professional at AT&T would clearly have appreciated this. That the calls disclosing the alleged MNPI were part of a systematic campaign initiated by AT&T—and not one-off, impromptu, or generally analyst initiated—could also be taken to support scienter. See Adopting Release, 2000 WL 1201556, at *12 (“[A] materiality judgment that might be reckless in the context of a prepared written statement would not necessarily be reckless in the context of an impromptu answer to an unanticipated question.”). A jury could also find incredible and abundantly contradicted by the record—and so risible as to support drawing negative consciousness-of-guilt inferences—-defendants’ denials in testimony that they had ever disclosed quantitative metrics to the analysts. Similarly, a jury could discredit as clearly false, and draw negative inferences from, defendants’ claims that their discussions with the analysts merely recapitulated Stephens’s public statements at the Deutsche Bank conference. Finally, although each defendant would be entitled to an individualized determination of his liability, the jury could consider the big picture as reinforcing each’s scienter. A jury could find that a defendant who was savvy to the urgent campaign directed by top executives to move ”| The IR defendants acknowledged that they knew that certain information could violate Reg FD, such as “signaling an analyst without explicitly telling them AT&T’s own numbers.” P. Ex. 66 at 337-38 (Black Tr.); see also, e.g., P. Ex. 88 at 106-07 (Womack Tr.). 122 consensus down before the Q1 2016 earnings announcement, and who was aware of the actions and words of the other two defendants and their supervisors in service of that campaign, would have understood that these overall efforts were highly irregular and that it had been reckless (at a minimum) to participate without confirming that these complied with Reg FD. Evidence and inference opposing scienter: On the other hand, a jury could find that the SEC had not carried its burden to establish scienter. A defendant’s subjective state of mind— including that he took action knowing that he was violating a legal standard or with recklessness as to that point—is a determination classically and commonly made by juries. See, e.g., Wechsler v. Steinberg, 733 F.2d 1054, 1058-59 (2d Cir. 1984) (“Issues of motive and intent are usually inappropriate for disposition on summary judgment.”); Pearlstein v. BlackBerry Ltd., No. 13 Civ. 7060 (CM) (KHP), 2022 WL 19792, at *9 (S.D.N_LY. Jan. 3, 2022) (“The fact-intensive disputes over falsity, materiality, and scienter will likewise be decided by the jury.”); dm re Celestica Inc. Sec. Litig., No. 07 Civ. 0312 (GBD), 2014 WL 4160216, at *10 (S.D.N.Y. Aug. 20, 2014) (“The fact-intensive nature of a scienter inquiry often militates against granting judgment as a matter of law.”). And the showing as to the defendants’ states of mind required of the SEC here, while short of criminal intent, is formidable, See, e.g., Malik v. Network I Fin. Sec., Inc., No. 20 Civ. 2948, 2022 WL 453439, at *3 (2d Cir. Feb. 15, 2022) (summary order) (affirming dismissal “where allegations sound|ed], at most, in negligence and thus [were] insufficient to support an inference of recklessness that satisfies the scienter requirement”) (citing S. Cherry Si, LLC y, Hennessee Grp. LLC, 573 F.3d 98, 109 (2d Cir. 2009} (describing recklessness as requiring a “state of mind approximating actual intent, and not merely a heightened form of negligence”)); Kalnit y. Eichler, 264 F.3d 131, 144 (2d Cir, 2001) (same where allegations did “not provide strong evidence of conscious misbehavior or recklessness”); 123 see also Adopting Release, 2000 WL 1201556, at *12 (“[I]n the case of a selective disclosure attributable to a mistaken determination of materiality, liability will arise only if no reasonable person under the circumstances would have made the same determination.”). On multiple grounds, a jury here could find this state of mind not established. Most obviously, the jury could credit defendants’ uniform testimony that, in real time, they had not appreciated that the information they were disclosing was material and nonpublic. See Evans Decl. 4—6, 8, 16, 18, 20; Womack Decl. 4—6, 8, 16, 18, 20; Black Decl. #5, 8, 17, 20, 28— 32; AT&T 56.1 IF 17, 63; see also In re Columbia Sec. Litig., 155 F.R.D. 466, 479 (S.D.NLY. 1994) (“Resolution of the question of scienter, as with any issue of motive or intent, generally requires examination of a witness’s demeanor and credibility and is thus . . . inappropriate for disposition on summary judgment.”). A jury could also note the absence of evidence that, in real time, any person within AT&T—including the defendants or the supervisors who had instigated the campaign to lower consensus-—-raised an alarm or expressed any hesitation or reservation about the legality of the ongoing disclosures. At argument, the SEC conceded that its review of emails and other documents did not uncover a single written communication indicative of consciousness of wrongdoing on the part of any defendant or supervisor. See July 6 Tr. at 98 (counsel for the SEC, conceding the absence of “any communication of any sort by any of [the individual defendants] that evinces any discomfort with any aspect” of the campaign to walk down analyst estimates, as well as the lack of “any indication that anybody ever raised any concern about [the plan] internally in AT&T”); see, e.g., Pehlivanian v. China Gerui Advanced Materials Grp., Ltd., 153 F. Supp. 3d 628, 652-54 (S.D.N.Y. 2015) (scienter in securities fraud case established based on evidence including circumstantial evidence of conscious misbehavior). A jury could read the 124 absence of such evidence, particularly given the duration of the campaign and the number of persons involved, to reflect that any selective disclosure of MNPI here was non-obvious and that defendants therefore acted no worse than negligently, not intentionally or recklessly. A jury could also note the absence of evidence that any analyst during March or April 2016 expressed concern that information that was being disclosed to him or her was 1n violation, or possible violation, of Reg FD. See AT&T MSJ Mot. at 33; see also sources cited at AT&T MSJ Opp’n at 15 n.43 (analysts “affirmatively testif[ying]” that they did not believe they had received MNPI); AT&T 56.1 § 51 n.57 (similar). There was evidence that the analysts, like the IR defendants, had been trained on Reg FD and MNPI.” A jury could discount the analysts’ failure to alert to a possible Reg FD violation, including because of analysts’ self-interest in not deterring the flow of information from issuers. See July 6 Ty. at 17 (defense counsel, conceding at argument that the analysts in this case had rarely reported Reg FD violations in their careers), But a jury could find that the wholesale lack of expressed concern or alarm by the analysts to whom Womack, Evans, and Black spoke support that any violations of Reg FD here were non- obvious. Cf in re Bank of Am. AIG Disclosure Sec. Litig., 980 F. Supp. 2d 564, 586 (S.D.N.Y. 2013) (scienter finding precluded where plaintiff conceded that “reasonable minds could differ” on practice’s legality, because reckless conduct must be ‘highly unreasonable’ and constitute ‘an See, e.g., AT&T MSI Opp’n at 16 n.46 (citations to analyst firms’ policies regarding receipt of MNPI).. These included, for example, a Wells Fargo Compliance Bulletin directing, “If an Analyst inadvertently receives material non-public information, the Analyst should immediately contact Research Compliance/Legal for further guidance and should not use such information further.” See Dkt. 134, Ex. 9 at 4. 125 extreme departure from ordinary standards of care’” (quoting Chill v. Gen. Elec. Co., 101 F.3d 263, 269 (2d Cir. 1996)) (emphasis added in Bank of America)).” In sum, a jury could find for either side on scienter. That precludes entry of summary judgment in favor of the SEC on its claims.” It also precludes entry of summary judgment for the defendants to the extent their motions are based on this element. See Wechsler v. Steinberg, 733 F.2d 1054, 1058-59 (2d Cir. 1984) (“Issues of motive and intent are usually inappropriate for disposition on summary judgment.”); Pearlstein, 2022 WL 19792, at *9-10 (denying summary judgment where parties took competing views on what the “evidence produced in discovery tends to show about materiality, falsity, and scienter,” and noting that “[s]uch fact- intensive inquiries are always reserved for the jury”); Jn re Celestica, 2014 WL 4160216, at *10 (stating that the “fact-intensive nature of a scienter inquiry often militates against granting judgment as a matter of law” and citing cases). ® Although defendants’ training would likely be most decisive as to their understanding as to what Reg FD forbade, to the extent a defendant could credibly testify that he was aware of the rule’s enforcement history, Reg FD’s sparse such history as of March and April 2016 and the absence of any judicial applications of the rule, see supra pp. 4-5, potentially could inform an analyst’s claims not to have appreciated what the rule forbade so as to make the disclosures here not an “extreme departure from the standards of ordinary care,” Frohling, 851 F.3d at 136. The summary judgment record does not make clear whether any individual defendant is positioned to so testify. 74 Tn light of this determination, the Court does not have occasion now to resolve whether the evidence bearing on the other two disputed elements (disclosures of material and nonpublic information), if viewed in the light most favorable to the defense, could reasonably be found for the defense, or whether the indisputable evidence on these points is conclusive for the SEC. In the event the case proceeds to trial, the Court will entertain briefing on these issues, the resolution of which may be affected by the Court’s resolution of Daubert challenges to proposed expert testimony. 126 F. Motions Not Resolved In the course of this decision, the Court has deferred resolution of several motions that it has determined are unnecessary to resolving the pending cross-motions for summary judgment. These include both sides’ Daubert motions. They also include, to the extent that defendants’ global challenge to analyst notes can be read to include motions directed to several particular such notes, such motions. The Court similarly defers resolution of a motion by the defendants relating to calls that the defendants made to a subset of analyst firms as to which the SEC did not obtain analyst notes or testimony. The SEC has relied here on notes by analysts at 10 of the 20 analyst firms the IR defendants contacted.” The parties deposed analysts from these 10 firms, plus analysts from two others as to which the SEC has not obtained notes.” The SEC did not, however, secure analyst notes, and no party took testimony, from the remaining eight firms.’? The individual defendants make a motion—which they style as one for summary judgment—with respect to those eight firms. See ID MSJ Mot. at 25~28. The Court denies that motion as not appropriate for summary judgment. Each defendant is charged in a single unitary count with violating Section 13(a) of the Exchange Act and Reg FD in connection with disclosures of MNPI to analysts in March and April 2016. See Compl. AT&T is charged as a principal, id. at 27; and Womack, Evans, and Black are charged as aiders and abetters of AT&T, id. at 28. The Court has denied the parties’ summary judgment motions These are JP Morgan, UBS, Deutsche Bank, RBC, Wells Fargo, Drexel Hamilton, Citi, Nomura, Pacific Crest, and Scotia Bank. SEC 56.1 App. D; P. Ex. 32. 76 These are BTIG and William Blair. ™ These are R.W. Baird, Oppenheimer, Evercore, Moffett Nathanson, Jefferies, DA Davidson, Bank of America/Merrill Lynch, and Buckingham. See SEC Resp. to ID 56.1 § 70. 127 on these counts, leaving a jury to determine liability. The Complaint does not contain any count keyed to an individual analyst firm. Accordingly, there is thus no count against which to move for summary judgment relating to these eight firms. The individual defendants’ motion is instead properly understood as a pretrial motion in limine, aimed at excluding evidence the SEC has indicated it may offer relating to these firms, in particular, evidence that, like the firms as to which analyst testimony has been secured, these firms lowered their public forecasts as to AT&T’s upgrade rates and/or revenues shortly after calls with the IR defendants.’* The Court will take up such a motion closer to trial, alongside other evidentiary motions. G. Dispositions As to the claims against Womack, Evans, and Black, in light of tts determination that material disputes of fact preclude granting summary judgment to the SEC or these defendants, the Court denies all motions for summary judgment. As to the SEC’s claim against AT&T, the Complaint appears to seek to hold AT&T liable solely based on the conduct of the individual defendants. See Compl. 9 139 (Defendant AT&T, through Womack, Evans, and Black, intentionally disclosed material nonpublic information during private phone calls with analyst firms as alleged above without making simultaneous disclosures of that information to the public.”). In moving for summary judgment, the SEC has not identified an alternative basis on which AT&T could be held liable. Accordingly, the denial 78 See SEC Resp. to ID 56.1 4 70; SEC 56.1 App. A (charting lower upgrade rates following calls with IR defendants, for Oppenheimer (6%), Evercore (5%), Moffett Nathanson (5.7%), Jefferies (5%)}, DA Davidson (5%), Bank of America/Merrill Lynch (5.1%), and Buckingham (5.5%)); id. App. B (same regarding equipment revenue decline, for R.W. Baird (-21.3%), Oppenheimer (- 20.1%), Evercore (-16.3%), Moffett Nathanson (-14,5%), Jefferies (-10.4%), DA Davidson (- 24.1%), Bank of America/Merrill Lynch (-14.6%), and Buckingham (-17%); id. App. C (same regarding total revenue adjustments). [28 of the summary judgment motions on the claims against the individual defendants requires denial of the motions directed to the claim against AT&T. CONCLUSION For the foregoing reasons, the Court denies the individual defendants’ motion to strike analysts’ notes; denies without prejudice the SEC’s and AT&T’s Daubert motions; and denies all parties’ motions for summary judgment. This case will now proceed toward trial, barring settlement. The Court will, by separate order, issue an order as to next steps in the case. The Clerk of Court is respectfully directed to terminate the motions pending at docket entries 88, 90, 97, 99, 101, and 138. SO ORDERED. Punk A Ln UL Paul A, Engelmayer United States District Judge Dated: September 8, 2022 New York, New York 1379