From Casetext: Smarter Legal Research

Sec. & Exch. Comm'n v. AT&T Inc.

United States District Court, S.D. New York
Sep 8, 2022
626 F. Supp. 3d 703 (S.D.N.Y. 2022)

Opinion

21 Civ. 1951 (PAE)

2022-09-08

SECURITIES AND EXCHANGE COMMISSION, Plaintiff, v. AT&T, INC., Christopher C. Womack, Kent D. Evans, and Michael J. Black, Defendants.

Alexander Mircea Vasilescu, Victor Edward Suthammanont, Joy Guo, Richard R. Best, Amy Rublin Mayer, Jacob David Zetlin-Jones, U.S. Securities and Exchange Commission, New York, NY, for Plaintiff. Randall Wade Jackson, Brittany Morgan Wagonheim, Willkie Farr & Gallagher LLP, New York, NY, Katey Lynn Fardelmann, Seth Michael Kruglak, Victoria Vance Corder, Norton Rose Fulbright US LLP, New York, NY, Richard Samson Krumholz, Norton Rose Fulbright US LLP, Dallas, TX, Peter Andrew Stokes, Norton Rose Fulbright US LLP, Austin, TX, Rachel Roosth, Norton Rose Fulbright US LLP, Houston, TX, for Defendant AT&T Inc. Jeffrey M. Tillotson, Jonathan Robert Patton, Tillotson Law, Dallas, TX, Leslie C. Thorne, Haynes and Boone LLP, New York, NY, for Defendants Christopher C. Womack, Kent D. Evans, Michael J. Black.


Alexander Mircea Vasilescu, Victor Edward Suthammanont, Joy Guo, Richard R. Best, Amy Rublin Mayer, Jacob David Zetlin-Jones, U.S. Securities and Exchange Commission, New York, NY, for Plaintiff.

Randall Wade Jackson, Brittany Morgan Wagonheim, Willkie Farr & Gallagher LLP, New York, NY, Katey Lynn Fardelmann, Seth Michael Kruglak, Victoria Vance Corder, Norton Rose Fulbright US LLP, New York, NY, Richard Samson Krumholz, Norton Rose Fulbright US LLP, Dallas, TX, Peter Andrew Stokes, Norton Rose Fulbright US LLP, Austin, TX, Rachel Roosth, Norton Rose Fulbright US LLP, Houston, TX, for Defendant AT&T Inc.

Jeffrey M. Tillotson, Jonathan Robert Patton, Tillotson Law, Dallas, TX, Leslie C. Thorne, Haynes and Boone LLP, New York, NY, for Defendants Christopher C. Womack, Kent D. Evans, Michael J. Black.

OPINION & ORDER

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 Wall 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 & T'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 et 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, 113 S.Ct. 2786, 125 L.Ed.2d 469 (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 to raise the same or similar motions closer to trial.

I. Factual Background

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 II.
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 S.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).").

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 ... to 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." Id. 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.

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." Id. § 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. Id. § 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." Id. 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); S.E.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 disgorgement, interest, and penalties); In re Lawrence D. Polizzotto, Exchange Act Release No. 70337, 2013 WL 4773958 (Sept. 6, 2013) ($50,000 penalty); S.E.C. v. David Ronald Allen et al., Litig. Release No. 22208, 2011 WL 10915927 (Dec. 22, 2011) (settlements and injunctions entered against numerous defendants); In 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.E.C. v. Office Depot, Inc., Litig. Release No. 3199, 2010 WL 4134972 (Oct. 21, 2010) ($1 million civil penalty); In 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.E.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 S.E.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") ¶ 1. It is publicly traded on the New York Stock Exchange under the ticker "T." Id. 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. ¶ 2. At all relevant times, Randall L. Stephenson was AT & T's chairman and chief executive officer ("CEO"), id. ¶ 8, and John Stephens was AT & T's chief financial officer ("CFO"), and reported to Stephenson, id. ¶¶ 13-14. In this role, Stephens oversaw the company's investor relations function. Id. ¶ 16. In 2015, John Stankey became CEO of AT & T's Entertainment Group—one of four segments of AT & T's business. Id. ¶¶ 3, 10.

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.

Womack: Womack has worked for AT & T since 1997, and, in 2006, became a member of its IR Department. Id. ¶ 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. Id. ¶ 29; Dkt. 92 ("Pl. 56.1") ¶ 33.

Viola retired from AT & T in June 2020. JSF ¶ 68.

Black: Black has worked for AT & T since 2008, and, in 2011, joined the IR Department. JSF ¶ 35. In March and April 2016, he was a finance director in the IR Department. Id. ¶ 34. He reported to Womack. Id. ¶ 28. Black's responsibilities included communicating with sell-side analysts who covered AT & T. Id. ¶¶ 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. Pl. 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 ¶ 41.

"Wireless upgrade rates" refer to the rates at which AT & T's subscribers upgraded their smartphones through AT & T. JSF ¶ 96. "Wireless equipment revenue" refers to revenue AT & T receives from the sale of wireless phones. Id. ¶ 87.

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. Id. ¶¶ 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. Id. ¶ 53. Evans communicated with an AT & T financial group in Atlanta that tracked, inter alia, upgrade rates. Id. ¶¶ 50, 53. His job also included reviewing

presentations that AT & T executives made to their teams for purposes of Reg FD compliance. Pl. 56.1 ¶ 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 help investors and investment funds (i.e., 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 ¶ 60; see JSF ¶ 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. Id. ¶¶ 62, 65-66. The IR Department had about 12 employees, including the individual defendants and Martin Sheehan, who reported directly to Viola. Id. ¶ 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") ¶¶ 8, 11.

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.

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 ¶¶ 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 ¶ 73; Def. Reply 56.1 ¶ 73. 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 ¶ 104; 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 ¶ 105.

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 ¶¶ 106-07; Def. Reply 56.1 ¶¶ 106-07.

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 ¶ 116. 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. ¶¶ 30, 43. Since at least 2006, Evans regularly received Reg FD training. Id. ¶¶ 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. Id. ¶ 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 ¶ 3.

Paula Anderson, AT & T's in-house counsel with expertise about Reg FD, trained IR Department employees regularly about Reg FD. JSF ¶¶ 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 ¶ 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 instance, Anderson used the "IR Training Document," which included a qualitative definition of materiality, to train the team. Pl. 56.1 ¶¶ 196-97; Def. Reply 56.1 ¶¶ 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 ¶¶ 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 ¶ 118; Pl. 56.1 ¶ 202; Def. Reply 56.1 ¶ 202; see JX 11 (the "Davis Polk Memo"). The memo made two "key points." JX at 1. First, "[p]rivate meetings with analysts or institutional investors are inherently risky from a Regulation FD perspective." Id. 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." Id. 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. Id. 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." Id. at 4.

The JSF correctly dates the memo April 7, 2005, whereas the parties' 56.1 statements erroneously date the memo 2015. See JSF ¶ 118; JX 11.

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 ¶ 86. For several years, AT & T subsidized the cost of cellphones it sold to subscribers by discounting the cost of the phone at the time of purchase. Id. ¶ 69; Pl. 56.1 ¶ 75; Def. Reply 56.1 ¶ 75. 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 ¶ 76; see JSF ¶¶ 70-71; Def. Reply 56.1 ¶ 76. AT & T refers to these as "equipment installment plans" or "EIPs." JSF ¶ 71. 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

"Churn is a measure of how many subscribes were leaving a carrier." JSF ¶ 111.

under the Next program would also be eligible for discounted service plans. Pl. 56.1 ¶ 79; Def. Reply 56.1 ¶ 79. Although activation fees and discounts may apply, Next program phones would not be subsidized up front. JSF ¶ 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 ¶ 95; 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 ¶¶ 102-03.

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.

Defendants submit that this accounting treatment provides an incomplete picture of wireless sales' impact on actual profit. Def. Reply 56.1 ¶ 102.

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. Pl. 56.1 ¶ 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.

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).

The overall impact on AT & T of this change is a source of some dispute. The parties agree that "[d]eclining upgrade rates generally had the effect of lowering [AT & T's] equipment revenue." Pl. 56.1 ¶ 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." Pl. 56.1 ¶ 168; Def. Reply 56.1 ¶ 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 ¶ 164; Def. Reply 56.1 ¶¶ 164, 168.

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 ¶ 146. Stephens also testified that upgrade rates did not affect churn in the last quarter of 2015 or Q1 2016. See Def. Reply 56.1 ¶ 146.

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

In 2016, some analyst firms covering AT & T did not include upgrade rate in their models, JSF 1132.

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. 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 ¶ 105; Pl. 56.1 ¶ 155; Def. Reply 56.1 ¶ 155.

Wireless equipment revenue includes revenue from non-phone devices such as tablets and phone accessories (e.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 ¶¶ 149-511.

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 ¶¶ 81-82. AT & T's total consolidated revenue figure included the company's total wireless revenue. Pl. 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 (e.g., revenues gained from 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 its SEC filings, AT & T did not separately report wireless equipment costs. JSF ¶¶ 85, 91.

In the relevant time periods, wireless equipment revenue and costs were as follows:

Wireless equipment revenue Wireless equipment costs Upgrade rate Q1 2015 $3.374 billion $4.280 billion 6.6% Q3 2015 $3.234 billion $4.082 billion 5.7% Q4 2015 $4.071 billion $5.366 billion 7.6% Q1 2016 $3.156 billion $3.991 billion 5.0%

See JSF ¶¶ 94-95, 114; Pl. 56.1 ¶¶ 119-33; Def. Reply 56.1 ¶¶ 119-33. In Q1 2016, total wireless revenue was about 45% of AT & T's $40.535 billion consolidated revenue. Pl. 56.1 ¶ 131; Def. Reply 56.1 ¶ 131. Consolidated equipment revenue was $3.434 billion, and consolidated equipment costs were $4.375 billion. Pl. 56.1 ¶¶ 121-22; Def. Reply 56.1 ¶¶ 121-22. Wireless equipment revenue was about 92% of consolidated 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 ¶¶ 132-33; Def. Reply 56.1 ¶¶ 132-33.

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 I.F.

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. Id. ¶ 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. Id. ¶ 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." Id. 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. Id. at 52; see P. Ex. 77 at 186-87 (Ilkowitz testifying that "[a]ll else equal, if you sell less phones, you'll 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 EBITDA growth. JX 126 at 4.

EBITDA is earnings before interest, taxes, depreciation, and amortization.

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 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. JX 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. In a similar vein, other analysts testified that, over time, AT & T's profits would not be affected by the change. F. AT & T's Previous Misses on Revenue Consensus

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 I 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.

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.

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 ¶ 138. Consensus estimates are public information. Id. Analysts and some media outlets publish comparisons of AT & T's actual results to Bloomberg and FactSet estimates. Id. ¶ 140. 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, reviewed AT & T's performance as compared to the Bloomberg and FactSet consensus estimates. JSF ¶¶ 143-44.

In Q3 2015, AT & T missed consensus revenue estimates. Id. ¶ 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 24th." Id. 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.'" Id. 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 JXs 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 ¶ 170. This was in large part due to analysts' overestimates of AT & T's wireless equipment revenue. Id. On a 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, 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.

Headlines of these news reports included: "Dow Jones: AT & T Misses on 4Q Top Line While Losing Phone Customers Again — Market Talk," "FastFT: AT & T shares slip after results," "Reuters: AT & T revenue below forecasts, shares fall," "Associated Press: AT & T Misses Street 4Q Forecasts," "Market-Watch: 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.

G. AT & T's Concern About Missing Consensus in Q1 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]." Id. 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." Id. at 2. Womack replied: "Agreed—we need to help them figure out their equipment spread for the year. Equipment revenue is starting to become a real challenge." Id. 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 at 3. 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. Id. 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." Id.

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 ¶ 356; see JX 3 ¶ 7. 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 ¶ 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 I think that is just a trend." Id. ¶ 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 ¶ 186. On March 3, 2016, Stephens emailed Viola and AT & T's Controller, writing, in part: "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." 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 ¶ 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. Id. ¶ 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 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 ¶ 195. 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") ¶ 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 ¶ 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. As a result, Womack wrote that the difference could be "$1.5B less in equipment revenue; a substantial reduction in no-margin equipment revenue YoY." Id. 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 ¶ 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 Q1 performance. These communications—which 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 ¶ 98; Def. Reply 56.1 ¶ 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 ¶¶ 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 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.

Cusick's notes stated: "Historically low u/g next couple of quarters. <5%!!" Pl. 56.1 ¶ 480; P. Ex. 14. They added: "Eqpt revenue could be down more than our 15% y/y." P. Ex. 14. After the call, Cusick sent an internal email stating: "AT & T telling us that handset upgrades this quarter will be historic lows. <5% vs 6.5% a year ago." P. Ex. 11. Choe's notes included: "Upgrade rate of 5%" and "Equipment revenue could be fairly negative, JPM down 15% y/y and could be worse ... Could be down 20% y/y." P. Ex. 15. Cusick's notes include the language, "closer to $40b from $40.7," and Choe's read, "$40 vs. $40.7." P. Ex. 14.

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." JX 181. Evans replied to all: "I am guessing the rate will be closer to 5% but [Womack's] math works better." Id.

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 165. The chart in Womack's email projected that a 5% upgrade rate would lead to a 19.4% decline in equipment revenue year-over-year. See id. That same day, Evans spoke to analysts from R.W. Baird ("Baird"), for about 14 minutes. JSF ¶¶ 220-21. Evans testified that he does not remember the order of any discussion with analysts on any call. P. Ex. 93. But he testified that there was "one key point" from Stephens's comments at the Deutsche Bank conference, "and that was on the upgrade trends." P. Ex. 53 (Evans deposition in SEC investigation) at 37, 56-57. Evans testified that his conversation with Baird "was related to making sure that he saw what or had heard what John Stephens had said at the Deutsche Bank conference." P. Ex. 71 (Evans deposition) at 169. He admitted that it would not have taken 14 minutes solely to point an analyst to Stephens's comments. Id. at 170. Baird would later (on March 31, 2016) issue a note revising downward its estimate of AT & T's revenue. JX 67 at 2, 5; see JSF ¶ 222. Baird's note stated: "Driven by fewer handset upgrades, we are lowering our wireless equipment revenue forecast, resulting in Q1 consolidated revenue dropping from $41.1 billion to $40.5 billion." JX 67.

The next day, on March 15, 2016, Womack emailed Viola: "I think the key takeaway is its looking like a ~20% decline YoY for equipment revenue." Viola responded the same day: "Yes, based on what Kent says probably a high-teen handle. Still huge percentage decline given some of the estimates (e.g. Batya at + 8%)." JX 194.

2. March 16, 2016 - March 22, 2016

On March 16, 2016, Womack spoke to UBS's Batya Levi. Levi testified that, to the extent she took notes on the call, she would have typed them directly into her model. P. Ex. 79 (Levi deposition) at 126. After the call, she wrote "down 20%" as a note in the cell in her model reflecting AT & T's equipment revenues for the first quarter of 2016. JX 278. On March 18, 2016, UBS published revised estimates, dropping its revenue projections, equipment revenue projections from 8.1% to -18.2%, and its upgrade rate estimate from 6.2% to 5.1%. JXs 274, 278.

On March 17, 2016, Womack spoke to Deutsche Bank analysts Matthew Niknam and Whitney Fletcher for about 56 minutes. Pl. 56.1 ¶ 527. Later that day, Fletcher sent an email to Niknam with notes from the call, including the statement: "High teens-20% yoy decline in equipment revenue." Pl. Ex. 16. On March 21, 2016, Deutsche Bank updated its estimates to a 19.8% year-over-year decline (down from a 10.8% increase) in equipment revenue, a reduced upgrade rate estimate of 5.2% (down from 6.5%), and a total revenue estimate of $39.969 billion (down from $41.035 billion). JX 279.

This followed Womack and Viola's March 15, 2022 email exchange in which Womack told Viola "it's looking like a ~20% decline YoY in equipment revenue" and Viola responded that Evans "says probably a high-teen handle." JX 194.

Also on March 17, 2016, Womack called Wells Fargo analyst Jennifer Fritzsche. JSF ¶ 228. After the call, Fritzsche emailed junior analyst Caleb Stein and another analyst to discuss "handset revenues"; five days later, Fritzsche directed Stein to contact Black regarding "equipment revenue thoughts." See JXs 316, 319, 321.

On March 21, 2016, Womack spoke to Oppenheimer analyst Tim Horan. JSF ¶ 234. The conversation followed Horan's having emailed a copy of his model to Womack and Evans, asking them to "let me know if you have thoughts on the model"—an email Evans forwarded to Black. Black responded, "I have worked through their model and believe it would be a good idea to schedule time to discuss consensus." JX 199. The next day, after Womack's call with Horan, Oppenheimer reduced its 1Q16 revenue estimates to

$40.510 billion (down from $41.376 billion), moved its equipment revenue estimates to a 15% decrease (down from a 5% increase), and revised its upgrade rate estimate to 6% (down from 7%). P. Ex. 89.

On March 22, 2016, within AT & T, Stephens requested from the IR Department "a comparison of AT & T's 1q outlook to current analyst estimates," for AT & T's board. JX 162. Black emailed Womack a comparison of the requested numbers. Black's chart reflected that the Top Ten wireless equipment revenue consensus was -1.6%; that the consensus upgrade rate was 6.1%; and that total consensus revenue was 25.3% growth. By contrast, AT & T then was forecasting, for the full quarter, a -11.5% change in equipment revenue; a 4.7% upgrade rate; and 23.7% revenue growth. JX 163.

See JSF ¶ 89 ("When AT & T reported its wireless equipment revenue in its quarterly earnings releases, it provided as a comparison the percentage increase or decrease from the same quarter of the prior year, e.g., 1Q16 against the first quarter of 2015 ('1Q15').").

The same day, Black spoke to two analysts: Caleb Stein, of Wells Fargo, and Brian Hyun, of RBC. JSF ¶¶ 229, 236. Analyst notes from those calls with Black reflect a consensus upgrade rate of at or around 5% for both Hyun and Stein; consolidated revenue growth of 22.5% (Hyun) or 23% (Stein); and equipment revenue change of -15% to -18% (Hyun) or -18% to -21% (Stein). See P. Exs. 20, 21 at 2-3; Pl. 56.1 ¶¶ 552-54.

Hyun took notes during the call. Dkt. 178-42 ¶ 5. In a declaration, he attested that the notes "are an accurate summary of what Michael Black said to me in the phone call." Id. ¶ 6.

3. March 23, 2016 - March 30, 2016

On March 23, 2016, Evans emailed Viola, Womack, and Black: "When we were previously discussing the upgrade rate, I had been talking about a low 5% range. Volumes continued to be very low in March and the new estimate being used in the 2 + 1 is 4.7%. [Womack] had been initially using something closer to this number to explain revenue softness and I moved him up but should not have." JX 201. The same day, while Viola was out of the office, a colleague emailed him with the subject line "JJS was looking for you" (referring to AT & T's CFO), writing in the body, "Mike, FYI—JJS stopped by today.... I also offered to have you call him but he asked to speak to Jeston [Dumas, another IR employee]." JX 178. Viola responded: "Thanks Di. I just tried calling. Jeston, let me know what he wants if you speak to him before me. Thanks." JX 178. Dumas replied, writing: "Just talked to him. He wants to make sure we are working the analysts that still have equipment revenue too high. I told him you just told me that was your top priority over the next few weeks." JX 178.

On March 24, 2016, Viola, Womack, Black, and Evans received, via an attachment to an email, AT & T's actual results for January and February 2016 and projections for March 2016 (i.e., AT & T's 1Q16 2 + 1 numbers). JX 133. The attachment stated that, whereas the consensus upgrade rate was 6.4%, AT & T's projected upgrade rate was 4.7%, and that wireless equipment revenue was $2.987 billion against a cost of $3.798 billion. Id. at 6-7, 9. That same day, Viola responded to an email from Black showing that the Top Ten average EPS had increased by $0.01, stating "EPS going in the wrong direction"; Womack replied, "Phil [Cusick, of JP Morgan] adjusted his upgrade rate which brought up his EPS." JX 276. Later that day, Womack confirmed to Viola that the two of them and Black would "circle the wagons" and continue reaching out to analysts. JX 191. Meanwhile on March 24, 2016, Black spoke to Barry Sine, a Drexel Hamilton analyst. JSF ¶ 238. After the call, Sine emailed a colleague with the subject line: "[J]ust got off the phone with AT & T," P. Ex. 103 ("Sine Decl.") ¶ 8; id. Ex. A. In the body of the email, he wrote: "[T]hey are proactively reaching out to analysts to reinforce comments CFO made at conferences ... Consensus phone upgrade rate in 1Q around 5%, versus 7.6% actual in 4Q. Consensus equipment revenue down 15% to 20%." Sine Decl. ¶ 8; id. Ex. A. Sine testified that he had "vague recollections of the call" but that his recollection "really is solely what is in this email." P. Ex. 33 at 278-79. Sine has attested that the conversation "played an important role" in his decision to reduce his quarterly revenue, equipment revenue, and upgrade rate estimates for AT & T, including to a 5% handset upgrade rate. Sine Decl. ¶¶ 12, 15.

On March 30, 2016, after RBC revised its estimates downward, Viola emailed Black and Womack (along with another IR employee), "Nice job with Jonathan [Atkin, an RBC analyst], especially getting his revenue number down to $40.3B." JX 183. That day, Black emailed Viola that Evercore would publish a preview note "shortly based on a deep dive discussion I had with them this morning. In addition, we should see Drexel Hamilton update as well as I had the opportunity speak with them just prior to Easter." Viola responded, inter alia: "Awesome work." JX 167. That day, Viola emailed Black to ask for his "schedule with the latest status of where everyone is with respect to consolidated revenue and EPS." JX 168. Black sent that information to Viola, who forwarded it to Evans. JX 185; P. Ex. 98.

4. March 31, 2016 - April 7, 2016

On March 31, 2016, Viola, Womack, and Black met to discuss "Consol Rev and EPS Consensus." JX 160. After the meeting, Black sent to Viola and Womack an email titled, "Latest view of revenue — updated through this morning." JX 190. Black's latest data (calculated the day before) showed a Top Ten and First Call consensus reflecting a 6.1% upgrade rate, a decline in wireless equipment revenue by 6.4%, and consolidated revenue growth of 24.8%. JX 132 at 3. In contrast, AT & T's actual 2 + 1 results showed a 4.7% upgrade rate and decline in equipment revenue of 11.5%. JX 132 at 3.

The individual defendants' calls with analysts continued. On March 31, 2016, Black spoke with Citi's Ilkowitz. JSF ¶¶ 240, 250, 268. Before the call, Citi's estimates showed a 1Q16 total revenue of $40.944 billion (a 26% increase year over year), a 12% increase in wireless equipment revenue, and a 6.8% upgrade rate. Pl. 56.1 ¶¶ 607-09; JX 228-31, 234-35. Ilkowitz took notes of their conversation. P. Exs. 22, 59 (Ilkowitz deposition in SEC investigation) at 135. Ilkowitz testified that on the call, he "would [have been] asking for consensus and that's what I would be attempting to record" in his notes. P. Ex. 77 (Ilkowitz deposition) at 116-17. The notes reflect, "5% upgrade rate," JX 183; equipment revenue being down 10-15%, P. Ex. 22; and, on revenue, "[c]onsolidated rev 23%," id. Citi would later (on April 7, 2016) lower its estimates to reflect consolidated revenue of $40.566 billion, an 8.5% decline in wireless equipment revenue, and a 5% upgrade rate. JX 76 at 7-8, 11.

On April 1, 2016, Viola emailed Sheehan (an AT & T IR employee), "when you talk to Moffett [Nathanson], remind him his 1Q'16 revenue numbers are way off due to his assumptions re wireless equipment revenue ... it would be great if he reduced his number closer to Cusick [of JP Morgan], Hodulik [of UBS], etc to $40.0B." JX 186.

On April 5, 2016, Black spoke to Greg McNiff, a Nomura analyst. JSF ¶¶ 240, 250, 268. McNiff took notes on the conversation

—although he testified that the words he wrote down were not necessarily words Black used. Pl. 56.1 ¶ 656; P. Exs. 26, 62; Stokes Decl., Ex. 54 at 79-80. McNiff's notes state, "upgrade rate — 5%," with equipment revenue being between 10 and 15%, "~12 Y/Y." P. Ex. 26. Shortly after the conversation, McNiff emailed other Nomura analysts: "Based on [Black's] subset of analysts, consensus for AT & T's 1Q rev is $40.3bn vs our $41.4bn and FactSet's $40.9bn," "~50% of the delta ... is related to equipment rev." JX 294 at 2-3.

Black also emailed McNiff after the conversation, responding to a question by writing: "That is where the folks that have updated recently 5%." JX 256.

5. April 8, 2016 - April 15, 2016

On April 8, 2016, AT & T's finance team emailed Viola and Black a sheet with preliminary calculations of the company's actual first quarter results; Black appears to have forwarded it to Womack minutes later. Those results, along with consensus estimates at the time, follow:

Top Ten average estimate AT&T preliminary actual results $40.54 billion (Top Ten); Total consolidated $40.54 billion revenue $40.75 billion (First Call) Wireless equipment -11.3% -6.5% revenue growth Upgrade rate 5.8% 5%

JX 171-72; Pl. 56.1 ¶¶ 448-49; see PI. Ex. 31. But see Def. Reply 56.1 ¶ 449 (suggesting Black forwarded Womack a different email).

That same day, April 8, 2016, Evans spoke to Michael Bowen, a Pacific Crest analyst. JSF ¶ 200. The day before, Pacific Crest had issued a note showing Q1 estimates of AT & T's total revenue as $41.691 billion. JX 265; P. Ex. 93. After the conversation, Bowen emailed a colleague: "Just heard from Kent Evans, and given John Stevens [sic] commentary last week at Deutsche (not sure which conference), upgrade rate will be at a record low .... So we need to move numbers lower. We will likely have to call kent next week to work through it." JX 242.

The Deutsche Bank conference at which Stephens made these comments had in fact occurred on March 9, 2016. JSF ¶ 191.

Also on April 8, 2016, Black spoke with Matthew Lee, a Scotiabank analyst. Lee emailed colleagues after the call, writing: "Below are my notes from my call with Mike on Friday." P. Ex. 32. Lee's notes state: "Not upgrading as fast—devices being used for longer.... Equipment revenue down -10%," and, "Revenue between 40.1B and 40.5B." Id.

On April 11 and 12, 2016, respectively, Black and Womack spoke with a Bank of America analyst. JSF ¶ 274. On April 18, 2016, Bank of America revised its estimates, lowering revenue to $40.41 billion (from $41.798) and listing equipment revenue as a 14.6% decline and an upgrade rate of 5.1%. P. Ex. 38; see JSF ¶ 275.

Also on April 11, 2016, Womack spoke with James Moorman at D.A. Davidson. JSF ¶ 279. On April 13, 2016, D.A. Davidson adjusted certain metrics in its AT & T estimates—to now reflect total equipment

revenue of $40.044 billion, a decline in wireless equipment revenue of 24%, and an upgrade rate of 5%. JX 101; P. Ex. 90; see JSF ¶ 280.

On April 12, 2016, Black spoke to James Ratcliffe, a Buckingham analyst, JSF ¶ 271, and on April 20, 2016, one of Ratcliffe's junior analysts, id. ¶ 764. On April 21, 2016, Black emailed Viola writing: "I spoke with [Ratcliffe] this past Friday as well as his junior associate just yesterday afternoon and they are aware that they are out of range of consensus and both indicated that they were going to adjust their estimates." JX 189. On April 25, 2016, Buckingham revised certain AT & T estimates—to total operating revenue to $40.141 and a wireless segment revenue of $17.660 billion (a 23,2% increase), JX 97; see JSF ¶ 272.

On April 15, 2016, Evans spoke to Bowen of Pacific Crest for a second time (the first conversation had been on April 8, 2016). JSF ¶ 266. Bowen testified that he remembered the conversation. P. Ex. 67 (Brown deposition) at 69-70. His notes state: "What caught his attention was that John Stevens [sic] was at Deutche [sic] talking about euqipment [sic] revenue.... Upgrade rates will be a record low. Record before was 5.6%-5.7%. Postpaid upgrade rate down to around 5% .... Equipment revenues are lower y/y b/c of lower upgrade rates.... Consensus 40.6 and it is going down. 40.3 is a good number." P. Ex. 34 at 3; see also P. Ex. 96 (handwritten notes). On April 19, 2016, Pacific Crest lowered its total revenue estimate to $40.5673 billion (down from $41.691 billion) and had a revised upgrade rate estimate of 4.8%. JX 243; see JSF ¶ 267.

Also on April 15, 2016, Womack spoke to McNiff and Jeffrey Kvaal, from Nomura. JSF ¶¶ 204, 251. McNiff and Kvaal took notes on the call. Both sets of notes refer to a 5% upgrade rate. After his call with Womack, Kvaal emailed colleagues: "I spoke to AT & T this morning. This is a very low quarter for handset sales. The[ ] upgrade rate, which has been 8-9% historically, is at an all-time low this quarter at 5%." JX 296. On April 19, 2016, Nomura updated its estimates, reducing total revenue to $40.535 billion and equipment revenue to $2.868 billion, noting, "[w]e have heightened confidence in our metrics, including ... a record low upgrade rate of ~5%." JX 84; see JSF ¶ 252.

"Record low upgrade rates — low 5% range" (Kvaal) and "Upgrade rate: 5% levels" (McNiff). Dkts. 178-35, 178-36.

6. April 16, 2016 - April 25, 2016

On April 18, an IR employee emailed Viola and Womack, "Should we be concerned about consensus EPS creeping up? ... 2. Better wireless margins (fewer upgrades) over next few quarters.... Wonder if there are headwinds we should talk about to keep consensus down." JX 277. Viola responded, "Agree we need a script as to what we should say." Id.

On April 19, 2016, Nomura and Cowen & Co. published revised estimates. That same day, Viola emailed Womack and Black: "Kvaal [from Nomura] came thru.... We really need Breen [from William Blair] and Bowen [from Pacific Crest] to adjust. Can we somehow get them excluded because they haven't refreshed their forecasts?" Womack responded, "Bowen is updating his numbers early this week from what I understand. If one or two are way out of trend, first call will exclude them based on stale estimates." JX 175. After this exchange, on April 21, 2016, Womack emailed James Breen, the William Blair analyst, writing: "[W]e looked through everything and not sure we need a preview as our numbers are close to consensus, let me know if you want to

discuss." JX 304. Womack called Breen the same day; they spoke for 17 minutes. JSF ¶ 263; P. 56.1 ¶ 463; Def. Reply 56.1 ¶ 463.

Later that day, Black told Viola that Womack had spoken with Breen "and he indicated that they are planning on adjusting their estimates—they report in both First Call and Bloomberg." JX 300. On April 24, 2016, William Blair reduced its total revenue estimates to $40.51 billion (down from $41.55 billion), based on lower equipment revenues. JX 91-92; see JSF ¶ 264. Breen testified that he understood Womack's April 21 call to signal that his revenue forecasts were too high and needed to be reduced. P. Ex. 68 at 134-37; P. Ex. 56 at 197, 223-25; Pl. 56.1 ¶ 719. But see Def. Reply 56.1 ¶ 719 (providing additional context for Breen's testimony).

Breen testified equivocally about whether he had reduced the estimates because of the call with Womack. See P. Ex. 68 at 137 (Breen, testifying that "[c]learly there was some discussion around that [revenue estimates] because I ultimately decided to do it," but that he did not recall anything about the conversation).

As noted, on April 25, 2016, Buckingham published revised estimates, bringing its estimated AT & T total operating revenue to $40.141 and wireless segment revenue to $17.660 billion (a 23.2% increase year over year). JX 97; see JSF ¶ 272.

I. April 26, 2016: AT & T Releases Q1 Results and Makes Consensus

On April 25, 2016, the day before AT & T released its results for Q1 2016, Stephens emailed Viola, "May just beat revenue consensus." Viola responded: "I think we will :-)." JX 179. That day, Blair and Buckingham issued their revised estimates, driving the First Call consensus revenue to $40.468 billion (down from $41.311 billion, as of March 1, 2016), and the Bloomberg consensus revenue also to $40.468 billion (down from $41.309 billion, as of March 1, 2016). Pl. 56.1 ¶¶ 789-91. But see Def. Reply 56.1 ¶¶ 789-91 (disputing admissibility of expert testimony establishing consensus figures). After these two revisions, CFO Stephens emailed CEO Stephenson, writing: "These two updates may do it for us—we may beat revenue consensus—not by much but a beat nonetheless." Stephenson responded: "Good." JX 3 ¶ 13(b).

On April 26, 2016, AT & T released its Q1 2016 results. The company's consolidated revenue of $40.535 billion beat consensus revenue by less than 1%. See JSF ¶ 5 (reported revenue of $40.535 billion); Dkt. 93-4 ¶ 29 (consensus revenue of $40.468 billion). AT & T announced $3.156 billion in wireless equipment revenue. JSF ¶ 95. And it announced a 5% upgrade rate. JSF ¶ 114. The SEC notes that, of the 25 quarters since 2006 in which analysts' estimates had initially been more than 0.5% higher than AT & T's ultimate results, Q1 2016 was the only quarter in which these estimates were eventually revised to a figure lower than AT & T's revenue, enabling it to exceed the consensus revenue forecast. Pl. 56.1 ¶ 794.

II. Procedural History

Many documents below were submitted both under seal and publicly (with appropriate redactions). The Court cites to the sealed versions of the documents. On June 10, 2022, the parties filed public versions of the documents at issue. See generally Dkts. 164-220.

On March 5, 2021, the SEC filed the Complaint. Dkt. 1 ("Compl."). On May 7, 2021, defendants filed answers. Dkts. 26-28, 30. On May 28, 2021, the Court held an initial pretrial conference. See Dkt. 44 (transcript). On March 17, 2022, the Court held a case management conference, during which the parties previewed their anticipated cross-motions for summary judgment. See Dkt. 80 (transcript). On April 26, 2022, the parties filed their joint statement of material facts, Dkt. 82, which they amended on April 29, 2022, Dkt. 84. The parties attached to the JSF their joint exhibits as Appendices A (public exhibits), Dkt. 85, and B (sealed exhibits), Dkt. 86.

As to the individual defendants' motion to strike analysts' notes: On May 20, 2022, the individual defendants filed that motion, Dkt. 138, and a memorandum of law in support, Dkt. 139 ("ID Strike Mot."). On June 10, 2022, the SEC filed an opposition to the motion to strike, and a memorandum of law in support. Dkt. 222 ("SEC Strike Mot. Opp'n").

The Court had instructed counsel that it would not entertain separate evidentiary motions alongside summary judgment motions, and that counsel were to make arguments regarding the admissibility of the analysts' notes in the course of their summary judgment briefs. Defense counsel flouted that direction, and the SEC moved to strike the brief. The Court instead reproved defense counsel and authorized the SEC to file a separate brief on this point, as to which the Court did not authorize a reply. See Dkt. 145.

As to the parties' Daubert motions: On May 3, 2022, the SEC filed a motion to strike defendants' expert witnesses, Dkt. 88, a memorandum of law in support, Dkt. 89, and a declaration (and attached exhibits) from the SEC's counsel in support of the SEC's Daubert motion, Dkt. 93. That same day, defendants filed a motion to strike the SEC's experts, Dkt. 97, and a memorandum of law in support, Dkt. 98. On May 4, 2022, defendants filed expert declarations and supporting exhibits. Dkts. 111-15. On May 20, 2022, the parties filed oppositions to the Daubert motions. Dkts. 127 (SEC opposition to defendants' Daubert motion), 132 (defendants' opposition to SEC's Daubert motion). The SEC also submitted a declaration (and attached exhibits) in support of the SEC's opposition to defendants' Daubert motion. Dkt. 126. And counsel for the individual defendants filed a declaration (and attached exhibits) in support of their opposition to the SEC's Daubert motion. Dkt. 131. On May 31, 2022, the parties filed replies. Dkts. 150 (SEC's reply to defendants' opposition), 154 (defendants' reply to SEC's opposition). AT & T filed a declaration (and attached exhibits) in support of defendants' reply, Dkt. 147, and the SEC did the same, in support of its reply, Dkt. 151.

As to the parties' motions for summary judgment as to liability: On May 3, 2022, the SEC filed its motion, Dkt. 90, and a memorandum of law in support, Dkt. 91 ("SEC MSJ Mot."). The SEC also filed its Rule 56.1 statement, Dkt. 92, and declarations (with attached exhibits) of three attorneys on behalf of the SEC, Dkts. 93-95. AT & T filed its motion for summary judgment, Dkt. 99, and a memorandum of law in support, Dkt. 102 ("AT & T MSJ Mot."). The individual defendants filed a motion for summary judgment, Dkt. 101, and a memorandum of law in support on May 4, 2022, Dkt. 104 ("ID MSJ Mot."). On May 4, 2022, the individual defendants submitted an "appendix" to their motion for summary judgment, which contained evidence they relied on in support of their arguments, Dkt. 105, and their Rule 56.1 statement, Dkt. 106. AT & T also submitted its Rule 56.1 statement. Dkt. 107. AT & T's counsel filed a declaration with attached exhibits, Dkt. 108, and non-parties Stephens and Viola (AT & T's CFO and IR Department head during the relevant time period) filed declarations in support of AT & T's motion for summary judgment. Dkts. 109 (Stephens), 110 (Viola). Finally, the individual defendants filed declarations in support of their motion for summary judgment.

AT & T also styled its motion as one for summary judgment and for judgment on the pleadings.

Dkts. 116 (Black), 117 (Evans), 118 (Womack).

On May 20, 2022, the parties filed numerous documents germane to the cross-motions for summary judgment: The SEC filed in a single document an opposition to defendants' motions for summary judgment, Dkt. 128 ("SEC MSJ Opp'n"), along with its responses in opposition to defendants' Rule 56.1 statements, Dkts. 129 (response to AT & T's Rule 56.1 statement), 130 (response to individual defendants' Rule 56.1 statement). The SEC filed a declaration with attached exhibits in support of its opposition to defendants' motions for summary judgment. Dkt. 133. AT & T and the individual defendants filed oppositions to the SEC's motion for summary judgment, Dkt. 136 ("AT & T MSJ Opp'n"), Dkt. 143 ("ID MSJ Opp'n"), along with supplemental declarations (and attached exhibits) in support from AT & T's counsel, Dkt. 134, and Stephens, Dkt. 135. Defendants filed their oppositions to the SEC's Rule 56.1 statement. Dkt. 137. The individual defendants filed a supplemental Rule 56.1 statement, Dkt. 141, along with a supplemental appendix to that document, Dkt. 142.

The individual defendants initially filed their opposition motion on May 20, 2022, Dkt. 140, and an unobjected-to amended memorandum of law on May 22, 2022, Dkt. 143. The Court considers the amended memorandum of law in assessing the individual defendants' arguments.

On May 31, 2022, the parties again filed numerous documents germane to the cross-motions for summary judgment: Defendants filed a joint reply to the SEC's opposition to their motions, Dkt. 148, along with a declaration from counsel (and attached exhibits), Dkt. 152. The SEC filed in a single document a reply to defendants' oppositions to the SEC's motion. Dkt. 149. Defendants filed a response to the SEC's response to AT & T's Rule 56.1 statement. Dkt. 153. And the SEC filed responses to the Rule 56.1 statements by AT & T, Dkt. 155, and individual defendants, Dkt. 156. The SEC also submitted the declaration (and attached exhibits) of Jordan Baker, an SEC securities compliance examiner, in support of the SEC's motion for summary judgment, Dkt. 157, and a reply declaration (and attached exhibits), Dkt. 158.

On June 7, 2022, the Court scheduled, Dkt. 162, and on July 6, 2022 heard, argument. See Dkt. 278 (transcript, or "July 6 Tr."). On August 3, 2022, the parties filed a joint letter requesting referral to a magistrate judge to discuss potential settlement, Dkt. 269, which the Court granted, Dkt. 270.

III. Evidentiary Motions

The Court first addresses the motions to exclude two categories of evidence cited in the summary judgment motions. First, the individual defendants move, on a global basis, to strike the analyst notes of the calls chronicled in Section I.H. Second, both sides move to exclude, in whole or part, the other's proposed experts.

A. Motion to Strike Analyst Notes

Specifically, defendants seek to strike exhibits 13-16, 20-22, 26, 32, and 34-36 to the SEC's motion for summary judgment; joint exhibits 96, 242, 278, and 308; and notes from Scotiabank analyst Lee, Dkt. 95-32, who was deposed after the SEC filed its motion for summary judgment.

Defendants assail the analyst notes as "an amalgam of casual notations regarding calls that often lasted an hour or more and ... that represent some unknown blend of what may have been said by the various participants on the call." ID Strike Mot. at 6. Defendants reject the SEC's characterization of the notes as "concrete and contemporaneous record[s] establishing what

Defendants said" to the analysts. See SEC MSJ at 31.

Defendants are descriptively correct that the analysts' notes, in general, do not reflect studied accounts of a conversation in the manner, for example, of a formal interview memorandum. Different analysts described their note-taking processes differently. Some took notes in real time and others shortly after the conversation with their AT & T IR interlocutor; some took notes by hand and others entered information electronically; some described their note-taking as aspiring to capture factual statements by the IR representatives and others as also capturing the analysts' mental impressions; and some recalled the March and April 2016 conversations to some degree, while others claimed not to recall these at all. Nearly every analyst deposed testified that he or she did not have a specific recollection of the statements at issue, or that the notes reflected both the speaker's statements and the analyst's thought process.

See, e.g., Cusick Depo. at 68; Choe Depo. at 29-30; Niknam Depo. at 116-17; Fletcher Depo. at 53-55; Kvaal Depo. at 87-90; Stein Depo. at 35-37; Lee Depo. at 104; Levi Depo. at 98-99.

These circumstances would give the defendants a basis—before the jury—on which to attempt to distance themselves from particular note excerpts. For example, to the extent the SEC were to argue that the appearance in a set of notes of a Q1 2016 performance statistic matching internal AT & T numbers (e.g., total revenue, wireless equipment revenue, or wireless upgrade rate) bespoke a disclosure by a defendant, the defendant could argue that the statistic was the product of the analyst's thought process and not a disclosure of fact. The persuasiveness of such an argument would, of course, turn on the excerpt and circumstances at issue.

To the extent, however, that defendants mean by their broadsides about the notes' unreliability to contend that the analyst notes are categorically excludable—such that they would never reach a jury and thus cannot be considered on summary judgment—that critique is wholly unpersuasive. Such an argument is evaluated under Federal Rule of Evidence 403. Under it, relevant evidence is to be excluded only where its probative value as to matters at issue is substantially outweighed by the risks of unfair prejudice, confusion, misleading the jury, or delay.

Here, the analyst notes have potentially formidable probative value. Provided that an individual set of notes were authenticated as having been taken by the analyst in connection with his or her discussion with Womack, Evans, or Black, the notes have the capacity to fortify the SEC's claim that the IR defendant communicated to the analyst non-public information about AT & T contained in the notes. The global nature of defendants' motion precludes an excerpt- or document-specific assessment. But, in general, the inference that an analyst disclosed nonpublic AT & T information would appear particularly strong as to precise numeric data (e.g., revenue figures or upgrade rates) contained in an analyst's notes. Where such data matched nonpublic data available to the IR defendant, or where the same data appears in the other analysts' notes, a strong—if not compelling—inference could be drawn that the IR defendant disclosed such data to the analyst. And indeed, nearly identical language or numbers appears repeatedly across different analysts' notes. See, e.g., P. Exs. 35 ("Record low upgrade rates — low 5% range"), 36 ("Upgrade rate: 5% levels") (emphasis in original), 14 ("Historically low u/g next couple of quarters. <5%!! .... Eqpt revenue could be down more than our 15% y/y.... closer to $40b from

$40.7"), 15 ("Upgrade rate of 5% .... Equipment revenue could be fairly negative, JPM down 15% y/y and could be worse ... Could be down 20% y/y." ... "$40 vs. $40.7"); JX 242 ("[U]pgrade rate will be at a record low."); 20 (5% consensus upgrade rate; equipment revenue down 18% "y/y b/c of lower upgrade and smartphone sales" and down 15 to 18% "this quarter"), 21 (5% consensus upgrade rate); Sine Decl. ¶¶ 12, 15 (conversation with AT & T IR Department "played an important role" in decision to reduce estimates, including to a 5% upgrade rate). The SEC thus may argue that, given the sheer number of times the same data matching AT & T's internal data appears in analyst notes, the only credible explanation is that Womack, Evans, and Black engaged in a coordinated campaign to selectively disclose it so as to move the analyst consensus in AT & T's favor. In such circumstances, a jury could find the circumstantial logic compelling that data in particular notes reflected, in shorthand form, data revealed by the IR defendant.

To be sure, as to any excerpt, the jury could draw the opposite inference. In testimony, the defendants have denied disclosing such information. The jury could credit such denials. Or it could find that the SEC has not carried its burden of proof based on an analyst's inability to source specific data in a set of notes to disclosures by a defendant, as opposed to the analyst's thought process. But that the jury's task may be difficult and its assessment unknowable does not make the notes unreliable or inadmissible under Rule 403. On the contrary, the exercise of determining based on testimony, notes, and the surrounding circumstances whether the analyst disclosed particular data in a particular conversation would call upon familiar jury competencies of assessing credibility and evaluating competing inferences. There is nothing about this exercise that would unfairly prejudice a defendant, mislead or confuse a jury, or cause undue delay, let alone to an extent that would substantially outweigh the probative value of the notes. Therefore, to the extent defendants move to exclude the notes based on global pronouncements about their unreliability, the Court rejects that argument. Subject to arguments that may be made on an excerpt- or document-specific basis, the analyst notes' probative value is not substantially outweighed—or outweighed at all—by the countervailing factors cited in Rule 403.

Defendants alternatively argue—and make their predominant challenge—that the analyst notes categorically are inadmissible as "double hearsay." Although defendants are correct that the admission of certain such notes will require a hearsay analysis of two layers of statements, this global critique also fails.

For each factual representation contained in a set of analyst notes, there are potentially embedded two statements: a representation of fact by the IR defendant ("the Q1 upgrade rate is 5%"), and a representation by the analyst that the IR defendant so stated in their conversation ("Womack told me the Q1 upgrade rate is 5%"). As a result, for the notes to be received for the truth of the matter asserted, each layer must satisfy the hearsay rules. Rodriguez v. Mod. Handling Equip. of N.J., Inc., 604 F. Supp. 2d 612, 622 (S.D.N.Y. 2009) ("Double hearsay is not admissible unless each level of hearsay is covered by an exception to the hearsay rule." (quoting Agric. Ins. Co., Inc. v. Ace Hardware Corp., 214 F. Supp. 2d 413, 416 (S.D.N.Y. 2002))); see generally Fed. R. Evid. 805 ("Hearsay within hearsay is not excluded by the rule against hearsay if each part of the combined statements conforms with an exception to the rule.").

As to the first layer, the factual representation by the IR defendant, if offered

by the SEC, would be admissible at trial as the non-hearsay statement of a party opponent under Federal Rule of Evidence 801(d)(2). See, e.g., United States v. Meskini, 319 F.3d 88, 93 (2d Cir. 2003) ("The statement was not hearsay, because it was admissible as a statement by a party opponent ... under Fed. R. Evid. 801(d)(2)(A)."); Pappas v. Middle Earth Condo. Ass'n, 963 F.2d 534, 537 (2d Cir. 1992); United States v. Malka, No. S3 19 Cr. 497, 2022 WL 1488568, at *28 (S.D.N.Y. May 11, 2022) ("[I]t is well-settled that a defendant's own statements are admissible as statements of a party opponent."). That is because the analyst notes at issue on this motion all derive from conversations between an analyst and Womack, Evans, or Black—each a named defendant.

To the extent, of course, that a jury rejects the SEC's contention that a factual representation in an analyst's notes reflected a statement by the IR defendant, finding it instead to reflect a mental impression of the analyst, the hearsay issue would disappear.

The Court has no occasion here to consider whether, if offered by the SEC, a statement to an analyst by a non-defendant member of the IR department would be admissible non-hearsay under Rule 801(d)(2)(C) as a statement of an authorized agent of party-opponent AT & T.

The question then becomes whether the second layer, the statement by the analyst to the effect that the IR defendant made the factual representation to him or her about AT & T, is inadmissible hearsay. See Restivo v. Hessemann, 846 F.3d 547, 574 (2d Cir. 2017) ("The fact that Restivo made this statement was admissible against him, as a statement by a party opponent under Rule 801(d)(2)(A). But there is a second level of hearsay incorporated into this statement—the statements that Restivo attributed to Halstead. No exception to the hearsay rule was shown to apply that would permit admitting this evidence for its truth against Restivo, and the jury could therefore not consider the statements attributed to Halstead for this purpose."). Provided that an analyst testified at trial, the analyst's testimony about his or her own statement as to the source of information in the notes would not constitute hearsay under Rule 801(c)(1). That is so regardless whether such testimony (1) explicitly sourced the factual representation in the notes to the IR defendant; or (2) implicitly did so, by authenticating notes which themselves sourced the statement to the IR defendant, as many of the analyst notes do. See, e.g., P. Ex. 11 ("AT & T telling us that handset upgrades this quarter will be historic lows. <5% vs 6.5% a year ago."); JX 242 ("Just heard from Kent Evans, and given John Stevens [sic] commentary last week at Deutsche (not sure which conference), upgrade rate will be at a record low .... So we need to move numbers lower."); P. Ex. 13 ("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. 103 ¶ 8 ("[J]ust got off the phone with AT & T.... [T]hey are pro actively reaching out to analysts to reinforce comments CFO made at conferences .... Consensus phone upgrade rate in 1Q around 5%, versus 7.6% actual in 4Q."); JX 186 (Viola instructing AT & T IR employee: "[W]hen you talk to Moffett [Nathanson], remind him his 1Q'16 revenue numbers are way off due to his assumptions re wireless equipment revenue."); JX 296 ("I spoke to AT & T this morning. This is a very low quarter for handset sales. The[ ] upgrade rate, which has been 8-9% historically, is at an all-time low this quarter at 5%.").

And the hearsay rules would not even come into play as to a second-layer statement—let alone be offended—where the analyst at trial authenticated the notes as

taken in connection with a conversation with an IR defendant, but could not source the statements therein to statements of the analyst. That is because the repetition of data across analysts' notes may give the jury a strong alternative basis to infer that such data came from the defendant. In this situation, there would be no second-layer statement offered in evidence to trigger hearsay analysis. Instead, that the analyst was the source of the data in particular notes would be proven circumstantially. See Bourjaily v. United States, 483 U.S. 171, 179-80, 107 S.Ct. 2775, 97 L.Ed.2d 144 (1987) ("[I]ndividual pieces of evidence, insufficient in themselves to prove a point, may in cumulation prove it. The sum of an evidentiary presentation may well be greater than its constituent parts. Taken together, these two propositions demonstrate that a piece of evidence, unreliable in isolation, may become quite probative when corroborated by other evidence."); Parhat v. Gates, 532 F.3d 834, 849 (D.C. Cir. 2008) ("[M]any of those assertions are made in identical language, suggesting that ... all may ultimately derive from a single source."); cf. United States v. Benedetto, 571 F.2d 1246, 1249 (2d Cir. 1978) (explaining admissibility of modus operandi evidence "to prove other like crimes by the accused so nearly identical in method as to earmark them as the handiwork of the accused") (quoting McCormick, Evidence § 190, at 449 (2d ed. 1972)).

Thus, although viable hearsay challenges may be available as to particular excerpts or notes, defendants' global hearsay challenge fails.

The analysis herein presupposes that the analyst's testimony will be received at trial—whether in the form of live testimony or, were the witness unavailable, deposition testimony, pursuant to Federal Rule of Evidence 804(b)(1)—and would authenticate the notes as having been made in connection with a discussion with an IR defendant. The Court does not understand the SEC to propose an alternative means to authenticate the notes of non-testifying analysts, or to propose to offer such notes at trial.

In any event, the SEC would have colorable arguments, on top of those above, why the admission of subsets of notes would comport with the hearsay rules.

First, the SEC contends that certain statements attributed to analysts were false. These statements—independent of being by a party opponent—may separately qualify as non-hearsay insofar as, to be hearsay, a statement, must be offered for the truth of the matter asserted. See Fed. R. Evid. 801(c)(2). To be sure, for the most part, given the nature of its Reg FD claims, the SEC offers the statements in analysts' notes for the truth of the matters asserted: as evidence of the IR defendants' selective disclosures of truthful non-public facts known within AT & T. But, as to some, the SEC contends that individual defendants made deliberately untrue statements to analysts. In particular, it alleges that Black aimed to disclose AT & T's actual internal revenue and upgrade rate forecasts, but, to whitewash the violation of Reg FD or to induce herd-following analysts to lower their forecasts, falsely cast these numbers as analyst "consensus" numbers. See infra Section V.D.2. False statements of this nature would not be offered to prove the truth of the IR defendant's statement, but to reveal his state of mind and/or the effect on the listener, to wit, to induce the analysts to lower their Q1 2016 estimates as part of the broader plan within AT & T to induce such revisions. And "[i]t is well established ... that statements offered for their effect on the listener are non-hearsay.'" United States v. Gotti, 457 F. Supp. 2d 395, 397 (S.D.N.Y. 2006) (quoting Smith v. City of New York, 388 F. Supp. 2d 179, 182 n.2 (S.D.N.Y. 2005)); United States v. Garcia, 900 F.2d 571, 576 (2d Cir. 1990) (same). Second, the SEC credibly argues that some excerpts of analysts' notes could be admissible under hearsay exceptions. Statements by a defendant, for example, about AT & T's Q1 2016 revenues or upgrade rates, might be admissible for purposes including (1) to demonstrate the operation of the scheme alleged within AT & T to divulge data to cause analysts to reduce their forecasts, (2) to show an IR defendant's state of mind, or (3) to explain the analyst's actions after a discussion with a defendant. Whether and to what extent a hearsay exception would apply cannot be determined on a global basis. That determination would require a document-by-document, and likely excerpt-by-excerpt, analysis. It would depend on the circumstances and the purpose articulated for the admission of a particular excerpt. At this stage, it is enough, in ruling on defendants' global bid to preclude these notes, to find that the SEC has persuasively argued that hearsay exceptions will likely permit receipt of some excerpts. In particular, the Court is persuaded that some excerpts are likely to be received under the exceptions for present-sense impressions, see Fed. R. Evid. 803(1), and for statements of the declarant's then-existing state of mind, such as his motive, intent, or plan, see id. 803(3). As to the former exception, to the extent the SEC offers a statement to show the analyst's impressions following a conversation with an IR defendant, multiple analysts have attested that they took notes during, or right after, conversations with defendants to enable them to capture their takeaways. The SEC also argues that excerpts may be admissible as evidence of an analyst's recorded recollections, under Fed. R. Evid. 803(5), On the limited briefing to date, it is plausible that, for some excerpts, this hearsay exception, too, might apply. For the above reasons, the Court denies defendants' global motion to strike the notes of the analysts. In the main, the analysts' notes appear admissible, and certain notes or portions thereof may be admissible under multiple theories.

See Okoroafor v. City of New York, No. 07 Civ. 9387 (DAB), 2013 WL 5462284, at *2 n.3 (S.D.N.Y. Sept. 25, 2013) ("Handwritten notes are admissible hearsay as a present-sense impression under Rule 803(1) ... where the note describes an event personally perceived by the declarant and which is made contemporaneously or immediately after the event described."); see generally United States v. Jones, 299 F.3d 103, 112 n.3 (2d Cir. 2002) ("[T]he hearsay exception for present sense impressions focuses on contemporaneity as the guarantor of reliability.").

See United States v. Harris, 733 F.2d 994, 1004 (2d Cir. 1984) ("These statements were admissible, not for their truth, but instead as circumstantial evidence of Harris's state of mind—his knowledge of Steward's cooperation.") (citing United States v. Parry, 649 F.2d 292, 295 (5th Cir. 1981)); see also, e.g., United States v. Enstam, 622 F.2d 857, 865 (5th Cir. 1980) ("[T]he judge appears to have based his ruling on the ground that the book would only be admitted to prove that Oldham had made a certain written statement to Enstam and not to prove that the statement was true. Given this limited basis for admission, the minute book was not hearsay, and most of the defendants' objections lack merit."). But see United States v. Dawkins, 999 F.3d 767, 790 (2d Cir. 2021) ("Broughton's proffered testimony did not reflect Code's state of mind within the meaning of Rule 803(3). Rather, Code's statement regarding his relationship with LOYD was an assertion of fact—that is, an archetypal hearsay statement—not a statement of motive or intent.").

See, e.g., McNiff Depo. at 119-21 ("Q. Why would you take notes of these calls? A. I just have a bad memory, so just to keep track of — you know, the calls and what's on them. Q. So you could refer to it at a later point? Yeah ... yeah, exactly."); Sine Decl. ¶ 8; id. Ex. A ("[J]ust got off the phone with AT & T.... Consensus phone upgrade rate in 1Q around 5%, versus 7.6% actual in 4Q. Consensus equipment revenue down 15% to 20%."); see also Okoroafor, 2013 WL 5462284, at *2 n.3; United States v. Kehoe, 562 F.2d 65, 70 (1st Cir. 1977) ("The rough notes represented a 'present sense impression' recorded close to the time of the event and as such are not excluded under the hearsay rule.").

See Parker v. Reda, 327 F.3d 211, 213 (2d Cir. 2003) ("To constitute such a recorded recollection, a memorandum must concern a matter about which a witness once had knowledge but now has insufficient recollection to enable the witness to testify fully and accurately, and be shown to have been made or adopted by the witness when the matter was fresh in the witness' memory and to reflect that knowledge correctly.") (cleaned up). The SEC also reserves the right to argue that the notes are admissible under the business records, Fed. R. Evid. 803(6), and the residual, Fed. R. Evid. 807, hearsay exceptions. The Court has not assessed these arguments.

The Court will therefore consider the analyst notes in resolving the motions for summary judgment. See Major League Baseball Props., Inc. v. Salvino, Inc., 542 F.3d 290, 309 (2d Cir. 2008) (court may rely on "any material that would be admissible or usable at trial" on summary judgment). This ruling is without prejudice to defendants' right, at trial, to object to particular notes or excerpts.

B. Daubert Motions

Each side moves to exclude its adversary's proposed expert witnesses under Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993). The five SEC experts are John Griffin, Paul Blalock, Marianne Wolk, Lauren Cohen, and Dominic DiBucci. The five defense experts are Paul Atkins, James Valentine, Roger Entner, Allen Ferrell, and Lucy Allen. The Court briefly summarizes each expert's testimony and the challenge to it, and then explains its determination not to resolve these challenges at the summary judgment stage.

1. The SEC's Experts and Defendants' Challenges

John Griffin : Griffin is a finance professor at the University of Texas McCombs School of Business and the owner of an expert consulting company. The SEC retained him to analyze the relationship between consensus estimates for Q1 2016 total revenue and the IR Department's communications with analysts, and how analysts' estimates of upgrade rates, equipment revenue, and total revenue changed after communications with the defendants. Griffin determined that analysts were significantly more likely to revise estimates after such communications, and that AT & T's stock responded negatively to intra-quarter updated analyst reports. He would also opine that the changes in analysts' estimates were material and that defendants acted with scienter. Defendants challenge these conclusions as methodologically unsound; rooted in vague, unreliable, cherry-picked data; or inappropriate for expert testimony.

Paul Blalock : Blalock has long experience in the investor relations field. He was retained to provide background information on Reg FD and best practices for IR professionals to comply with the law and applicable ethics codes. He would opine that AT & T's communications with analysts here were inconsistent with those standards, and that professionals are on notice of Reg FD's strictures. Defendants challenge his opinions as improper legal conclusions.

Marianne Wolk : Wolk has long experience as an equity analyst covering the technology industry, and is a professor of finance and economics at Baruch College and City College of New York. Wolk would review the role of sell-side analysts, the relationship between analysts and public companies, and how Reg FD bears on these. She would also opine that allegedly communicated information was both material and nonpublic. Defendants challenge

her conclusions as unsupported by facts specific to this case and as improper legal conclusions.

Lauren Cohen : Cohen is a professor of finance and asset management at Harvard Business School. He was retained to conduct an empirical analysis of the behavioral patterns of analysts covering AT & T in Q1 2016 as compared to other quarters. He determined that the IR Department's communications were statistically more likely in that quarter to result in an analyst changing her estimate, He would also opine on the materiality to the market of AT & T's having met or missed revenue consensus, and on studies about the interactions between analysts and the IR departments of public companies. Defendants challenge his opinions as based on arbitrary and improper handling of data. They argue that his conclusions about materiality are irrelevant.

Dominic DiBucci : DiBucci has long experience in the telecommunications industry, including in investor relations posts, and with accounting principles in the context of smartphone costs and revenues. He would opine on the impact of wireless equipment revenues and upgrade rates on AT & T's financials and that trends on those metrics were material. Defendants challenge his conclusions as based on insufficient evidence and unreliable methods.

2. AT & T's Experts and the SEC's Challenges

Paul Atkins : Atkins is a former SEC commissioner and partner at a Big Four accounting firm, and is CEO of his own consulting firm. Defendants retained him to rebut Blalock's report and to provide "insights into the application, adoption and purpose behind Regulation Fair Disclosure." Dkt. 114 ¶ 2. He would opine that Blalock wrongly interpreted Reg FD, that there are industry standards of conduct in this area, and that the disclosures in this case involved MNPI. Id. ¶¶ 68-71. The SEC challenges Atkins' testimony, on the ground that he (i) lacks experience as an IR professional, and improperly (2) opined that there are no industry standards for IR professionals, (3) opined on the interpretation of Reg FD, and (4) offered legal conclusions as to materiality.

James Valentine : Valentine is an experienced former sell-side analyst. He would opine that equipment revenue and handset upgrade rates were immaterial because they did not impact earnings or stock prices, and that the information defendants allegedly disclosed was publicly available. The SEC challenges these opinions as improper legal conclusions and as based on suspect data.

Roger Entner : Entner is a telecommunications professional. He would opine that the SEC's experts are wrong to opine that the information allegedly disclosed by defendants was material and nonpublic. The SEC seeks to exclude his opinions on the grounds that he is unqualified to opine on analysts' findings based on their models, and that his means of computing upgrade rates was methodologically unsound.

Allen Ferrell : Ferrell is a law professor and a proposed rebuttal expert. He performed an event study to analyze whether analysts' changes to revenue estimates impacted AT & T's stock price. The SEC seeks to exclude his opinions as irrelevant, lacking valid theoretical grounding, and based on bad data.

Lucy Allen : Allen is an economist. She would opine that defendants' alleged disclosure of handset upgrade rates and equipment revenues did not impact AT & T's stock price, based on an event study she conducted. The SEC challenges her testimony as irrelevant and based on misleading assumptions.

3. Disposition

The parties' Daubert motions to exclude expert testimony are in the nature of motions

ordinarily made shortly before trial, as motions in limine. In this case, however, such motions were appropriately made at the summary judgment stage. That is because the proposed experts each opine on at least one of the three elements on which the summary judgment motions turn: whether the information selectively disclosed was (1) material and (2) nonpublic, and (3) whether defendants acted with scienter in disclosing it. It thus was reasonable for counsel to conclude that, depending on the Court's assessment of the lay evidence, the exclusion of expert testimony could have tipped the scales on a motion for summary judgment.

The Court's assessment of the lay evidence, however, makes it unnecessary to resolve the Daubert motions at the summary judgment stage. As explained below, the Court finds sufficient lay evidence as to each element for the SEC's claim to reach a jury. Thus, defendants' summary judgment motions would be denied even if the SEC's expert testimony were excluded. But the Court also finds sufficient lay evidence on which a jury could find for the defendants on—at least—the element of scienter. Thus, the SEC's summary judgment motion would be denied even if the defense's expert testimony was excluded.

Accordingly, this case—for all its complexity—is unlike the cases (often scientific) in which the ability of a party to survive summary judgment turns on the admissibility of expert testimony. See, e.g., In re Mirena IUS Levonorgestrel-Related Prod. Liab. Litig. (No. II), 387 F. Supp. 3d 323 (S.D.N.Y. 2019) (granting defense motion for summary judgment, after decision excluding under Daubert all seven of plaintiffs' experts on general causation, on the ground that lay evidence was insufficient and incompetent to establish general causation), aff'd, 982 F.3d 113 (2d Cir. 2020). The expert testimony proposed in this case would certainly have the capacity to guide the jury's assessment of certain elements, such as materiality. But the plaintiff SEC can establish every element of its case by lay evidence alone, and defendants can reach a jury on at least the element of scienter on the strength of lay evidence alone.

For that reason, the Court does not have occasion now to resolve the Daubert motions. And resolving those 10 motions would have delayed the resolution of the summary judgment motions. The Court accordingly denies the Daubert motions without prejudice to either side's right to move anew under Daubert should the case approach trial.

See Doe No. 1 v. Putnam Cnty., No. 16 Civ. 8191 (PMH), 2020 WL 7027596, at *9 (S.D.N.Y. Nov. 30, 2020) ("[W]here a court determines that consideration of an expert's testimony in support of or opposition to a motion for summary judgment is unnecessary to the determination of the summary judgment motion itself, it may deny the motion to preclude without prejudice.") (citing cases); In re Scotts EZ Seed Litig., No. 12 Civ. 4727 (VB), 2017 WL 3396433, at *15 (S.D.N.Y. Aug. 8, 2017) ("[B]ecause deciding plaintiffs' Daubert motion.... is not necessary for deciding the parties' cross-motions for summary judgment, plaintiffs' motion to preclude the testimony of [the experts] is denied without prejudice."); Kendall v. Metro-N. Commuter R.R., No. 12 Civ. 6015 (DLC), 2014 WL 1885528, at *8 (S.D.N.Y. May 12, 2014) ("In sum, Kendall has raised a genuine question of material fact as to both causation and negligence, even without relying on the contested expert report of Dr. Morrisey. Accordingly, Metro-North's summary judgment motion is denied. The Daubert motion to Dr. Morrisey's expert report is denied without prejudice to renewal as a motion in limine.").

IV. Threshold Challenges to Regulation FD

Defendants next make a series of perfunctory challenges to Regulation FD itself.

They argue that it (1) is unconstitutional under the First and Fifth Amendments; (2) was promulgated without statutory authority; and (3) is logically inoperable, based on an exception to Reg FD that defendants interpret to swallow the rule.

For the reasons below, the Court does not find merit to any of these challenges.

A. First Amendment Challenge

In a three-page portion of their brief, defendants argue that Reg FD violates the First Amendment. They argue that Reg FD must be evaluated under strict scrutiny and that it fails that standard. The SEC defends Reg FD as constitutional.

Defendants have oscillated whether this is exclusively an as-applied challenge or also a facial one. At argument, however, counsel stated that their summary judgment motion makes only an as-applied challenge. See AT & T Mot. at 46; July 6 Tr. at 39-42.

1. Analytic Framework

The Court at the outset must determine the standard of review to apply to Reg FD. No court has yet ruled, or opined, on Reg FD's constitutionality: Although it has been the basis of occasional enforcement actions, the regulation has been little litigated, and in the one case in which a First Amendment challenge to Reg FD was raised, the district court did not reach that issue. See Siebel, 384 F. Supp. 2d at 709 n.16. Lacking such guidance, and perhaps due to the regulation's unusual nature, the parties propose diametrically different analytic frameworks—and consequent levels of scrutiny—applicable to it.

Defendants, terming Reg FD "content based," urge that it, like the content-based laws at issue in a line of Supreme Court precedents, must be narrowly tailored and advance a compelling government interest to satisfy the First Amendment. See AT & T Mem. at 46 & n.183 (citing First Nat'l Bank of Bos. v. Bellotti, 435 U.S. 765, 786, 98 S.Ct. 1407, 55 L.Ed.2d 707 (1978)); AT & T Reply Mem. at 24 & n.71 (citing Reed v. Town of Gilbert, 576 U.S. 155, 163, 169, 135 S.Ct. 2218, 192 L.Ed.2d 236 (2015)). As an alternative reason to apply strict scrutiny, defendants liken Reg FD to laws that compel speech. See AT & T Mem. at 46 & n.184 (citing Pac. Gas & Elec. Co. v. Pub. Utils. Comm'n of Cal., 475 U.S. 1, 5-21, 106 S.Ct. 903, 89 L.Ed.2d 1 (1986)). They argue that Reg FD fails that standard, and would also fail intermediate scrutiny.

The SEC counters that Reg FD is content neutral and governs only the selective manner, rather than the content, of commercial speech. Thus, it argues, the regulation is subject to rational basis scrutiny, such that it need only reasonably relate to an adequate interest and not be unduly burdensome. See SEC MSJ Opp'n at 42-43 (citing Ward v. Rock Against Racism, 491 U.S. 781, 791, 109 S.Ct. 2746, 105 L.Ed.2d 661 (1989); Zauderer v. Off. of Disciplinary Counsel, 471 U.S. 626, 650, 105 S.Ct. 2265, 85 L.Ed.2d 652 (1985); and Cent. Hudson Gas & Elec. Corp. v. Pub. Serv. Comm'n, 447 U.S. 557, 561, 100 S.Ct. 2343, 65 L.Ed.2d 341 (1980)). The SEC argues that Reg FD clears that standard, as it would intermediate scrutiny.

Although each proposed framework has some resonance, none provides a satisfying match for Reg FD.

Defendants' content-based framework : The Court first addresses defendants' argument that Reg FD should be evaluated as a "content based" law. Whether Reg FD is content based presents a substantial question. The SEC fairly notes the Supreme Court's teaching that "[t]he principal inquiry in determining content neutrality ... is whether the government has adopted a regulation of speech because of disagreement with the message it conveys," Ward, 491 U.S. at 791, 109 S.Ct.

2746, which is not so with respect to the content governed by Reg FD (selectively disclosed material nonpublic information). Nonetheless, on balance, defendants are correct that, in a strictly denotative sense, Reg FD draws a distinction on the basis of the content of a corporate disclosure. That is because the regulation applies only to material non-public information, and not to disclosures of other information. See Picard v. Magliano, 42 F.4th 89, 102 (2d Cir. 2022) ("A regulation is content-based if it applies to speech 'because of the topic discussed or the idea or message expressed,' or in other words, if the regulation has a content-based purpose or justification." (quoting Reed, 576 U.S. at 165-66, 135 S.Ct. 2218)). As such, although the content distinction Reg FD draws is at a very high level of generality, the regulation requires an issuer to promptly or simultaneously disclose to the public only some selectively disclosed information—that whose content, viewed in context of the then-surrounding circumstances, qualifies as MNPI.

But defendants' suggestion that all content-based regulations must satisfy strict scrutiny overlooks the significant body of decisions involving laws and regulations mandating affirmative disclosures of information, particularly for public issuers and other participants in the securities industry. These disclosure provisions are explicitly content based. They require the disclosure of certain content (e.g., corporate earnings or executive compensation) and do not require the disclosure of others (e.g., future acquisition targets or forthcoming product lines). Yet these have routinely withstood First Amendment challenges without any suggestion that strict, or even intermediate, scrutiny applied. As the District of Columbia Circuit has recognized, "Speech relating to the purchase and sale of securities ... forms a distinct category of communications." S.E.C. v. Wall St. Publ'g Inst., Inc., 851 F.2d 365, 373 (D.C. Cir. 1988). "If speech employed directly or indirectly to sell securities were totally protected, any regulation of the securities market would be infeasible—and that result has long since been rejected." Id. (citing Paris Adult Theatre I v. Slaton, 413 U.S. 49, 61, 93 S.Ct. 2628, 37 L.Ed.2d 446 (1973) and cases cited therein); see S.E.C. v. Cap. Gains Rsch. Bureau, Inc., 375 U.S. 180, 186-95, 84 S.Ct. 275, 11 L.Ed.2d 237 (1963); Am. Power & Light Co. v. S.E.C., 329 U.S. 90, 99-103, 67 S.Ct. 133, 91 L.Ed. 103 (1946); N. Am. Co. v. S.E.C., 327 U.S. 686, 705-07, 66 S.Ct. 785, 90 L.Ed. 945 (1946).

Two sets of examples illustrate the point. Section 17(b) of the Securities Act of 1933, 15 U.S.C. § 77q(b), obliges publicists to review the content of their radio broadcasts, newsletters, and articles touting stocks, and to publicly disclose those as to which they received consideration from an issuer for the publicity. That statute draws explicit content lines. Yet courts have consistently upheld it against First Amendment challenges, without resort to strict scrutiny. See United States v. Wenger, 292 F. Supp. 2d 1296, 1306, 1309 (D. Utah 2003) (applying rational basis to § 17(b) disclosure requirement and noting that in "many other securities cases, reaching back decades, ... the courts have repeatedly found that nondisclosure of relevant information is fraudulent or deceptive") (emphasis in original) (citing cases), aff'd, 427 F.3d 840 (10th Cir. 2005); Wall St. Publ'g, 851 F.2d 365 (upholding constitutionality of § 17(b) regardless of whether rational basis or intermediate scrutiny applied); S.E.C. v. Huttoe, No. 96 Civ. 2543, 1998 WL 34078092 (D.D.C. Sept. 14, 1998) (same, applying rational basis scrutiny).

Similarly, public issuers of securities must file current, 17 C.F.R. § 249.308, quarterly, id. § 249.308a, and annual, id.

§ 249.310, reports. These disclose a wide array of information, from corporate officers' business backgrounds, id. § 249.401(e), and compensation, id. § 249.402, to the identity of underwriters, id. § 249.701(b), and a company's oil drilling activities, § 249.1205. See generally 15 U.S.C. § 78l ; 17 C.F.R. § 229 et seq. The information that must be disclosed is overtly defined by its content. Yet the Supreme Court has upheld these obligations, recognizing that "the exchange of information about securities" and "corporate proxy statements" may be regulated without offending the First Amendment. Ohralik v. Ohio State Bar Ass'n, 436 U.S. 447, 457, 98 S.Ct. 1912, 56 L.Ed.2d 444 (1978). Notably, in support of this proposition, the Court cited cases that did not apply heightened scrutiny at all. See id. (citing Mills v. Elec. Auto-Lite Co., 396 U.S. 375, 90 S.Ct. 616, 24 L.Ed.2d 593 (1970); and S.E.C. v. Tex. Gulf Sulphur Co., 401 F.2d 833 (2d Cir. 1968), cert. denied, 394 U.S. 976, 89 S.Ct. 1454, 22 L.Ed.2d 756 (1969)).

Driving the point home, the Court has since distinguished the regulatory disclosure requirements involving "the exchange of information about securities" and "corporate proxy statements" upheld in Ohralik from "similar regulation of political speech [that] is subject to the most rigorous scrutiny." See Dun & Bradstreet, Inc. v. Greenmoss Builders, Inc., 472 U.S. 749, 759 n.5, 105 S.Ct. 2939, 86 L.Ed.2d 593 (1985). In the heavily regulated securities industry context, the case law thus leaves a proper place for disclosure regulations drawn on the basis of content. As the Ohralik Court put the point, it "has never been deemed an abridgment of freedom of speech or press to make a course of conduct illegal merely because the conduct was in part initiated, evidenced, or carried out by means of language, either spoken, written, or printed." 436 U.S. at 456, 98 S.Ct. 1912 (quoting Giboney v. Empire Storage & Ice Co., 336 U.S. 490, 502, 69 S.Ct. 684, 93 L.Ed. 834 (1949)).

For this reason, where courts have struck down securities disclosure laws under the First Amendment, they have largely done so because the disclosure law compelled "otherwise protected speech using the guise of securities laws." Nat'l Ass'n of Mfrs. v. S.E.C., 748 F.3d 359, 372 (D.C. Cir. 2014) ("NAM I"), subsequent opinion after reh'g, 800 F.3d 518 (D.C. Cir. 2015).

Justice Breyer emphasized much the same point this past Term: valid laws drawn on the bases of the content of communications are common, including in the securities industry. In City of Austin v. Reagan Nat'l Advertising of Austin, LLC, — U.S. —, 142 S. Ct. 1464, 212 L.Ed.2d 418 (2022), the Supreme Court upheld a municipal signage law that differentiated between signs that advertised on- versus off-premises services and activities. The majority found this distinction facially content-neutral, noting that "restrictions on speech may require some evaluation of the speech and nonetheless remain content neutral," id. at 1473—an observation that the SEC can fairly muster in support of its reading of Reg FD as not content-based. In his concurrence, Justice Breyer pointed out the ubiquity of valid laws that differentiate on the basis of communicative content:

The U.S. Code (as well as its state and local equivalents) is filled with regulatory laws that turn, often necessarily, on the content of speech. Consider laws regulating census reporting requirements, e.g., 13 U.S.C. § 224; securities-related disclosures, e.g., 15 U.S.C. § 78l ; copyright infringement, e.g., 17 U.S.C. § 102; labeling of prescription drugs, e.g., 21 U.S.C. § 353(b)(4)(A), or consumer electronics, e.g., 42 U.S.C. § 6294;

highway signs, e.g., 23 U.S.C. § 131(c); tax disclosures, e.g., 26 U.S.C. § 6039F; confidential medical records, e.g., 38 U.S.C. § 7332; robocalls, e.g., 47 U.S.C. § 227; workplace safety warnings, e.g., 29 C.F.R. § 1910.145 (2021); panhandling, e.g., Ala. Code § 13A-11-9(a) (2022); solicitation on behalf of charities, e.g., N. Y. Exec. Law Ann. § 174-b (West 2019); signs at petting zoos, e.g., N. Y. Gen. Bus. Law Ann. § 399-ff(3) (West 2015); and many more.

Id. at 1477 (Breyer, J., concurring) (emphasis added); see also Reed, 576 U.S. at 177, 135 S.Ct. 2218 (Breyer, J., concurring in the judgment) (describing "governmental regulation of securities" as laws for which "a strong presumption against constitutionality," that is, strict scrutiny, "has no place").

See also Nat'l Elec. Mfrs. Ass'n v. Sorrell, 272 F.3d 104, 116 (2d Cir. 2001) ("Innumerable federal and state regulatory programs require the disclosure of product and other commercial information. To hold that the Vermont [labelling] statute is insufficiently related to the state's interest in reducing mercury pollution would expose these long-established programs to searching scrutiny by unelected courts. Such a result is neither wise nor constitutionally required.") (citing 2 U.S.C. § 434 (reporting of federal election campaign contributions); 15 U.S.C. § 78l (securities disclosures); 15 U.S.C. § 1333 (tobacco labeling); 21 U.S.C. § 343(q)(1) (nutritional labeling); 33 U.S.C. § 1318 (reporting of pollutant concentrations in discharges to water); 42 U.S.C. § 11023 (reporting of releases of toxic substances); 21 C.F.R. § 202.1 (disclosures in prescription drug advertisements); 29 C.F.R. § 1910.1200 (posting notification of workplace hazards); Cal. Health & Safety Code § 25249.6 ("Proposition 65"; warning of potential exposure to certain hazardous substances); N.Y. Envt'l. Conserv. Law § 33-0707 (disclosure of pesticide formulas)).

Reg FD is of a broadly similar character to this body of securities laws and regulations. Although Reg FD's disclosure command is not absolute but conditional (it depends on the issuer's prior selective disclosure of the MNPI), and although it applies outside the marketing process governed by Section 17(b) and the quarterly and annual reporting process, Reg FD is fairly viewed as a component of a broad regulatory disclosure scheme aimed at assuring that issuers inform investors on an evenhanded basis of material information about an issuer, and protecting investors against sharp practices, which here include slipping MNPI to favored audiences. The disclosure components of this body of law and regulations have not, historically, been subject to strict scrutiny. See Wall St. Publ'g, 851 F.2d at 373 ("If speech employed directly or indirectly to sell securities were totally protected, any regulation of the securities market would be infeasible—and that result has long since been rejected."); Dun & Bradstreet, 472 U.S. at 758 n.5, 105 S.Ct. 2939 (citing Ohralik, and recognizing that "certain kinds of speech are less central to the interests of the First Amendment than others").

The Court, accordingly, is unpersuaded by defendants' argument that, because, at a high level, Reg FD bases its obligation of public disclosure on whether an earlier private disclosure contained content qualifying as MNPI, it is subject to strict scrutiny.

Defendants' compelled speech framework : Defendants alternatively argue that Reg FD should be subject to strict scrutiny because it compels an issuer to speak. As the Supreme Court has synopsized this line of authority: "[T]he right of freedom of thought protected by the First Amendment against state action includes both the right to speak freely and the right to refrain from speaking at all." Wooley v. Maynard, 430 U.S. 705, 714, 97 S.Ct. 1428, 51 L.Ed.2d 752 (1977) (holding that New Hampshire could not constitutionally require citizens to display the state

motto on their license plates); Turner Broad. Sys. v. F.C.C., 512 U.S. 622, 642, 114 S.Ct. 2445, 129 L.Ed.2d 497 (1994) ("Laws that compel speakers to utter or distribute speech bearing a particular message are subject to the same rigorous scrutiny" as laws that "suppress, disadvantage, or impose differential burdens upon speech because of its content"); Burns v. Martuscello, 890 F.3d 77, 84 (2d Cir. 2018) (stating that, as a general matter, the First Amendment protects the "right to decide what to say and what not to say" (quoting Jackler v. Byrne, 658 F.3d 225, 241 (2d Cir. 2011))). For this reason, "laws mandating speech" are subject to "strict or exacting scrutiny." Evergreen Ass'n, Inc. v. City of New York, 740 F.3d 233, 244 (2d Cir. 2014) (citations omitted).

Although this line of authority has its roots in cases like Wooley, which involved a law compelling individuals' speech, it applies to corporate speech as well. And, as defendants note, Reg FD unavoidably compels corporate speech. But it is a far cry from the law at issue in the one compelled corporate speech case on which defendants rely for their claim that strict scrutiny is warranted. At issue in Pacific Gas & Electric Co. v. Public Utilities Commission of California was a California law requiring a private utility company ("PG & E"), in the newsletter it included with monthly billing statements, to include materials submitted by third parties. The newsletter contained a mix of commercial and political speech: "political editorials, feature stories on matters of public interest, tips on energy conservation, and straightforward information about utility services and bills." 475 U.S. at 5, 106 S.Ct. 903. Applying strict scrutiny and invalidating the law, the Supreme Court held that the PG & E newsletter merited "the full protection of the First Amendment." Id. at 8-9, 106 S.Ct. 903. The newsletter's content extended, the Court noted, "well beyond speech that proposes a business transaction, and includes the kind of discussion of 'matters of public concern' that the First Amendment both fully protects and implicitly encourages." Id. (internal citations omitted) (quoting Thornhill v. Alabama, 310 U.S. 88, 101, 60 S.Ct. 736, 84 L.Ed. 1093 (1940)).

Defendants rely on other cases, but they are inapposite because they involved outright prohibitions on speech. See Reed, 576 U.S. 155, 135 S.Ct. 2218 (striking down content-based speech restriction); Bellotti, 435 U.S. 1, 98 S.Ct. 906 (invalidating prohibition on certain corporate speech); Vugo, Inc. v. City of New York, 931 F.3d 42 (2d Cir. 2019) (applying intermediate scrutiny to content-based restriction on commercial speech).

Reg FD, however, is very far afield, as to the content it concerns and the conduct it compels. As to content, there is no claim that Reg FD implicates political speech. It facially does not. Unlike the California law that added political content and viewpoints to PG & E's newsletter, Reg FD does not facially implicate the "profound national commitment to the principle that debate on public issues should be uninhibited, robust, and wide-open." N. Y. Times Co. v. Sullivan, 376 U.S. 254, 270, 84 S.Ct. 710, 11 L.Ed.2d 686 (1964). The interest Reg FD aims to advance in preventing issuers from favoring select audiences with potentially market-moving information is more akin to the interest in avoiding consumer deception that underlies numerous statutory and regulatory disclosure requirements. These historically have been upheld provided they are reasonably related to preventing the deception of consumers. See Zauderer, 471 U.S. at 651, 105 S.Ct. 2265; NAM I, 748 F.3d at 372. The information that Reg FD obliges an issuer to publicly disclose is limited to "material nonpublic information regarding [an] issuer" that the issuer has

already selectively disclosed. 17 C.F.R. § 243.100(a). Focused as it is on nonpublic factual information bearing on the issuer, this definition effectively assures that—save in highly improbable hypotheticals—the MNPI that an issuer must disclose publicly will not contain political content or viewpoints.

See Adopting Release, 2000 WL 1201556, at *9 ("Information is material if 'there is a substantial likelihood that a reasonable shareholder would consider it important' in making an investment decision.... Information is nonpublic if it has not been disseminated in a manner making it available to investors generally." (quoting TSC Indus., Inc. v. Northway, Inc., 426 U.S. 438, 449, 96 S.Ct. 2126, 48 L.Ed.2d 757 (1976)) (citing Basic Inc. v. Levinson, 485 U.S. 224, 231, 108 S.Ct. 978, 99 L.Ed.2d 194 (1988); Texas Gulf Sulphur, 401 F.2d at 854)).

This case, in fact, supplies an excellent illustration. (And, insofar as defendants' First Amendment challenge on this motion is solely an as-applied one, the Court's focus in resolving it is on the MNPI that the SEC claims AT & T selectively disclosed here, not on hypothetical applications of Reg FD.) AT & T's selective disclosures, as alleged, related narrowly to its Q1 2016 performance. The SEC claims that AT & T privately shared with select analysts internal metrics, including the percentage of customers who had upgraded their smartphones in Q1 2016 and AT & T's anticipated quarterly wireless equipment and total revenue. Thus, as the SEC would apply it here, Reg FD mandated the non-selective disclosure only "of accurate, factual, commercial information [which would] not offend core First Amendment values." See Sorrell, 272 F.3d at 113-14; cf. Pac. Gas, 475 U.S. at 8, 106 S.Ct. 903 (invalidating law that shaped content of newsletter akin to a "small newspaper," which addressed subjects "from energy-saving tips to stories about wildlife conservation, and from billing information to recipes"); Commodity Trend Serv., Inc. v. Commodity Futures Trading Comm'n, 149 F.3d 679, 686 (7th Cir. 1998) (holding that independent investment advice in newsletters is protected political speech). From a First Amendment perspective, the content at issue here thus merits less protection than the political speech that was at issue in Pacific Gas.

As to the conduct it compels, Reg FD, as applied, presents a far lesser intrusion than did the mandate in Pacific Gas. As described by the Supreme Court, California's law "[c]ompelled access" to PG & E's newsletter, "penalize[d] the expression of particular points of view," and forced the utility to alter its mix of speech "to conform with an agenda [it] d[id] not set." 475 U.S. at 9, 106 S.Ct. 903. Here, in contrast, AT & T, by definition, had already embraced the identical factual message that the SEC would have it share. Reg FD merely required AT & T to disclose to the market the MNPI it had privately shared with analysts, to prevent "a privileged few [from gaining] an informational 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." Adopting Release, 2000 WL 1201556, at *2; see also 17 C.F.R. § 243.100(a). Unlike in Pacific Gas, Reg FD thus did not require AT & T to publish political speech, to associate with an unwanted message, or to allow a third party to gain access to an internal circular or any other proprietary mode of communication. In compelling public disclosure of selectively disclosed MNPI, Reg FD is instead far closer to the securities laws mandating public issuers to periodically disclose select internal data, such as the laws upheld in Ohralik.

The Court, accordingly, is unpersuaded by defendants' argument that, because

Reg FD compels issuers to make factual disclosures, it is subject to strict scrutiny.

The SEC's commercial speech framework: The SEC, for its part, likens the speech at issue to compelled commercial speech—"expression related solely to the economic interest of the speaker and its audience," Cent. Hudson, 447 U.S. at 561, 100 S.Ct. 2343—subject to rational basis review. But although there are points of similarity between speech regulated by the securities laws and commercial speech—including insofar as each may compel public disclosure of data in the interest of protecting consumers or investors—the case law to date has stopped short of equating the two.

The Supreme Court has noted, in general, the "difficulty of drawing bright lines that will clearly cabin commercial speech in a distinct category," City of Cincinnati v. Discovery Network, Inc., 507 U.S. 410, 419, 113 S.Ct. 1505, 123 L.Ed.2d 99 (1993). In 1988, in Wall Street Publishing Institute, the D.C. Circuit addressed the SEC's argument that that the speech it regulated was commercial speech, entitled only to limited protection. The D.C. Circuit declined to find an industry publication charged with violating the Exchange Act's anti-touting provisions to be commercial speech, noting that Central Hudson had defined such speech as "expression related solely to the economic interests of the speaker and its audience." Wall St. Publ'g, 851 F.2d at 372 (quoting Cent. Hudson, 447 U.S. at 561, 100 S.Ct. 2343) (emphasis in Wall St. Publ'g). The Court noted that although speech "that is concededly an advertisement, refers to a specific product, and is motivated by economic interest may properly be characterized as commercial speech," id. (citing Bolger v. Youngs Drug Prods. Corp., 463 U.S. 60, 66-67, 103 S.Ct. 2875, 77 L.Ed.2d 469 (1983)) (emphasis in Wall St. Publ'g), the circular at issue was not a "clear fit" for those criteria, id. Canvassing the law more broadly, the D.C. Circuit noted that the Supreme Court's decisions indicate "that securities regulation is a form of regulation distinct from the more general category of commercial speech," and that "the First Amendment protections provided by the commercial speech doctrine do not detract from the government's regulatory power over the securities market." Id. at 373 (citing Dun & Bradstreet, Inc., 472 U.S. 749, 105 S.Ct. 2939, 86 L.Ed.2d 593, and Ohralik, 436 U.S. at 456, 98 S.Ct. 1912); (emphasis in Wall St. Publ'g); cf. Riley v. National Federation of the Blind of North Carolina, Inc., 487 U.S. 781, 796 n.9, 108 S.Ct. 2667, 101 L.Ed.2d 669 (1988) ("Of course, the dissent's analogy to the securities field entirely misses the point. Purely commercial speech is more susceptible to compelled disclosure requirements.")); Lowe v. S.E.C., 472 U.S. 181, 234, 105 S.Ct. 2557, 86 L.Ed.2d 130 (1985) (White, J., concurring).

The commercial speech doctrine, although a closer fit than the lines of First Amendment doctrines invoked by the defense, is ultimately also a mismatch for the speech covered by Reg FD, in general and as at issue here. At its core, the commercial speech doctrine has centered on advertisements or speech otherwise proposing a commercial transaction. See 44 Liquormart, Inc. v. Rhode Island, 517 U.S. 484, 496-500, 116 S.Ct. 1495, 134 L.Ed.2d 711 (1996) (reviewing development of the doctrine); see also Cent. Hudson, 447 U.S. at 563, 100 S.Ct. 2343 ("The First Amendment's concern for commercial speech is based on the informational function of advertising."); Discovery Network, 507 U.S. at 422, 113 S.Ct. 1505 ("[The] core notion of commercial speech" is that it does "no more than propose a commercial transaction.") (quoting Bolger, 463 U.S. at 66, 103 S.Ct. 2875). Reg FD, however, reaches

issuer speech made in a wide array of contexts that do not involve a proposed commercial transaction. These include communications of any sort—including in meetings, phone calls, and writings—with analysts, investment advisors, broker-dealers, or individual investors, provided that the circumstances make it "reasonably foreseeable that the person will purchase or sell the issuer's securities on the basis of" the MNPI disclosed. 17 C.F.R. § 243.100(b)(1). As to the speech at issue here, the SEC does not claim that AT & T's IR defendants communicated with any analysts in order to sell a product or security. The calls also differed in length and subject matter from a paradigmatic sales pitch, with some lasting up to an hour and covering a range of subjects relating to AT & T.

For these reasons, Zauderer, on which the SEC principally relies, is inapposite. Relevant here, at issue there were Ohio rules with respect to attorney advertising that required disclosures of fees and costs. Although the boundaries of "commercial speech" were "subject to doubt," the Supreme Court held, it was "clear enough" that "advertising pure and simple" "falls within those bounds." 471 U.S. at 637, 105 S.Ct. 2265. The ads "undeniably propose[d] a commercial transaction," id., and as such only minimally implicated "constitutionally protected interest[s]," id. at 651, 105 S.Ct. 2265. Upholding the disclosure rules, the Court held that an advertiser's First Amendment rights "are adequately protected as long as disclosure requirements are reasonably related to the State's interest in preventing deception of consumers," and that requiring attorneys to disclose potential costs and fees met this standard, as it furthered the state's interest in the prevention of deception of clients. Id. at 651-52, 105 S.Ct. 2265. Zauderer assists the SEC insofar as it underscores that laws compelling disclosures of factual information in connection with commercial transactions may implicate only minimal First Amendment interests of the merchant, are judged under a rational basis standard, and do not implicate the traditional concerns associated with compelled speech. Id. at 650-51, 105 S.Ct. 2265 (citing, inter alia, Wooley, 430 U.S. 705, 97 S.Ct. 1428). But, because Zauderer speaks to a context (advertising) outside that addressed by Reg FD, it is no more than lightly instructive as to the governing standard of First Amendment review. The Court accordingly is also unpersuaded by the SEC's argument that the paradigm of commercial speech applies here.

As the Second Circuit has explained, in the area of commercial speech, different standards apply to laws that restrict speech as opposed to imposing mandatory disclosure obligations:

Commercial disclosure requirements are treated differently from restrictions on commercial speech because mandated disclosure of accurate, factual, commercial information does not offend core First Amendment values of promoting efficient exchange of information or protecting liberty interests.... Zauderer, not Central Hudson..., describes the relationship between means and ends demanded by the First Amendment in compelled commercial disclosure cases. The Central Hudson test should be applied to statutes that restrict commercial speech.


Sorrell, 272 F.3d at 113-14; see also Safelite Grp., Inc. v. Jepsen, 764 F.3d 258, 264 (2d Cir. 2014) ("In interpreting these Supreme Court precedents, our previous cases have drawn a distinction between standards of review to be applied to laws mandating commercial speech disclosures and laws restricting commercial speech.... [A]s far as we know, all federal cases applying Zauderer [review to factual, commercial disclosure]—ha[ve] dealt with disclosure requirements about a company's own products or services. This distinction is important ....") (internal citations omitted) (cleaned up). But cf. NAM I, 748 F.3d at 372 (declining to decide level of scrutiny for First Amendment challenge to corporate disclosure law, where court held that rational basis did not apply and rule would fail even intermediate scrutiny).

Conclusion : Reg FD's idiosyncratic quality makes it an imperfect fit for any existing familiar First Amendment framework, including those suggested by the parties. The Court is unaware of any law remotely similar to Reg FD's requirement that public companies who choose to disclose material information to the market do so on an evenhanded basis.

First Amendment frameworks not raised by the parties prove inapposite, too. For example, the constitutionality of the (since-repealed) "fairness doctrine," under which broadcasters were once required to air different viewpoints on controversial issues of public importance, was premised, in large part, on the scarcity of the broadcast spectrum (and the absence at the time of alternative channels to reach the public). See Red Lion Broad. Co. v. F.C.C., 395 U.S. 367, 89 S.Ct. 1794, 23 L.Ed.2d 371 (1969). The perceived need for a regulation compelling an issuer to provide MNPI disclosures to the entire market, however, does not derive from scarce media, but from the history, chronicled by the SEC, of issuers and executives favoring select audiences with such disclosures.

It is clear to the Court that strict scrutiny does not apply to Reg FD, facially, or, relevant here, as applied. The regulation does not have any hallmark indicative of an occasion to apply strict scrutiny. Reg FD does not implicate political speech; it does not restrict speech; and it does not implicate expressions of opinion or points of view, but concerns only factual information (that qualifies as MNPI). Insofar as Reg FD imposes a disclosure obligation, it does not compel a company to make a disclosure that it has not already chosen to make to a subset of the market; and the entities to which it applies (public issuers of securities) are already subject to broad-ranging disclosure obligations with respect to internal corporate data that are likely in due course to pick up much of the information covered by Reg FD. Existing First Amendment doctrine does not afford any charter to apply a standard of review to Reg FD that does not presume constitutionality and that often proves fatal in fact. And on the present, as-applied, challenge, there is nothing about the prosaic performance data allegedly disclosed by the IR defendants on behalf of AT & T that would implicate heightened First Amendment concerns.

Less clear is whether rational basis review or intermediate scrutiny applies to Reg FD. Favoring rational basis review (in addition to the factors above) are (1) the long line of cases upholding disclosure and reporting requirements for public issuers without subjecting them to heightened scrutiny; and (2) the use of rational basis review for disclosure laws for commercial speech, the consumer-protective purpose of which echoes Reg FD's investor-protection purpose. Favoring scrutiny above this level is (1) the formal content-specificity of Reg FD; and (2) that Reg FD compels a potentially wide array of information whose boundaries by definition are contextual and cannot be set in advance with precision. See Safelite, 764 F.3d at 264 (applying intermediate scrutiny to law compelling disclosure of commercial speech, on ground that the disclosure law compelled "speech that goes beyond the speaker's own product or service").

In the interest of completeness, the Court will evaluate Reg FD under both standards.

2. Assessment

Under intermediate scrutiny, a court asks whether "(1) the expression is protected by the First Amendment; (2) the asserted government interest is substantial; (3) the regulation directly advances the government interest asserted; and (4) the regulation is no more extensive than

necessary to serve that interest." Vugo, 931 F.3d at 44. Rational basis review requires merely a rational connection between the purpose of a commercial disclosure requirement and the means employed to realize that purpose. Sorrell, 272 F.3d at 115 (citing Zauderer, 471 U.S. at 651, 105 S.Ct. 2265). The Court first applies the more demanding test of intermediate scrutiny. Because Reg FD satisfies this standard, it necessarily satisfies rational basis review. See Competello v. LaBruno, No. 02 Civ. 664, 2005 WL 1637907, at *12 (D.N.J. July 12, 2005) ("Since [the] policies satisfy the intermediate scrutiny standard, they also necessarily satisfy the rational basis standard, should the Court of Appeals eventually utilize the rational basis standard.").

First factor : The first factor inquires whether the speech at issue is lawful, because the Government's latitude to regulate is greater with respect to speech related to illegal activity. See Safelite, 764 F.3d at 264; Cent. Hudson, 447 U.S. at 563-64, 100 S.Ct. 2343 ("If the communication is neither misleading or related to unlawful activity, the government's power is more circumscribed.") (internal citations omitted). The SEC does not claim that the information Reg FD covers in general—or that was selectively disclosed here—was illegal. The Court therefore treats the regulated speech as lawful. See Centro de la Comunidad Hispana de Locust Valley v. Town of Oyster Bay, 868 F.3d 104, 113-16 (2d Cir. 2017) (ordinance concerned lawful activity where consummation of the commercial transaction did not necessarily constitute an illegal act).

Second and third factors : The Court considers the second and third factors together as they are closely related. For these reasons, which largely track those the SEC articulated when it promulgated Reg FD, the Court finds the Government's interest in combatting selective MNPI disclosures by public companies substantial, and that Reg FD directly advances that interest.

In promulgating Reg FD, the SEC persuasively explained that barring issuers from selectively disclosing MNPI—and requiring an issuer that discovers such a disclosure promptly to act to disclose it to the public—advances market integrity and protects investors. Selective disclosures, it explained, are deleterious for much the same reasons as is insider trading:

Issuer selective disclosure bears a close resemblance in this regard to ordinary "tipping" and insider trading. In both cases, a privileged few gain an informational 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. Likewise, selective disclosure has an adverse impact on market integrity that is similar to the adverse impact from illegal insider trading: investors lose confidence in the fairness of the markets when they know that other participants may exploit "unerodable informational advantages" derived not from hard work or insights, but from their access to corporate insiders. The economic effects of the two practices are essentially the same.

Adopting Release, 2000 WL 1201556, at *2 (internal footnotes omitted).

Reg FD, the SEC explained, responds to this problem by obliging corporations to release MNPI on an evenhanded basis. This, the SEC added, in turn promotes market integrity, insofar as it reduces the risk that issuers will attempt to use nonpublic information to sway or corrupt analysts. Otherwise, "corporate management" may treat "material information as a commodity to be used to gain or maintain favor with particular analysts or investors." Id. Further, "analysts may feel pressured

to report favorably about a company or otherwise slant their analysis in order to have continued access to selectively disclosed information." Id. (citing reports to this effect).

Although the criminal and regulatory prohibitions against insider trading play important deterrent and punitive roles, they are unequal to the challenge of protecting against selective disclosure by issuers of their MNPI. The SEC recognized this point, id. at *2 & n.7, and it is clearly correct. The United States lacks an omnibus insider trading statute. Only in two narrow contexts does a federal statute ban insider trading. Instead, as commentators including this Court have chronicled, the federal prohibition on insider trading largely has been developed by case law, effectively as a matter of federal common law, under the anti-fraud provision of the Securities Exchange Act of 1934, Section 10b, and its implementing rule 10b-5. And the requirements for liability for insider trading that this case law has developed, anchored in corporate law concepts of fiduciary duty, are demanding to establish. The Government must prove, among other things, scienter (intent to defraud) and intended personal benefit. And there is no liability for insider trading for a person who receives MNPI and trades on it unaware that the information had been released in breach of a duty. See Salman, 137 S. Ct. at 423 ("The tippee acquires the tipper's duty to disclose or abstain from trading if the tippee knows the information was disclosed in breach of the tipper's duty, and the tippee may commit securities fraud by trading in disregard of that knowledge."); see also id. at 427 ("In Dirks, we explained that a tippee is exposed to liability for trading on inside information only if the tippee participates in a breach of the tipper's fiduciary duty.").

See Williams Act of 1968, Pub. L. 90-439, 82 Stat. 454 (1968), as amended, 84 Stat. 1497 (1970) (prohibiting trading based on an undisclosed tender offer); Stop Trading on Congressional Knowledge (STOCK) Act of 2012, Pub. L. 112-105, 126 Stat. 291 (2012) (prohibiting Congress and its staff from trading on inside knowledge), Only the first of these was in place at the time Reg FD was adopted.

See Paul A. Engelmayer, Congress: U.S. Needs an Insider Trading Law, at 1 (Oct. 16, 2015), reprinted in New York L.J. (Oct. 23, 2015) (speech to Securities and Litigation Enforcement Institute of the New York City Bar) (tracing doctrinal development of insider trading law and urging enactment of a statute to clarify standards for tippee liability).

The case law has developed two complementary theories of insider trading. Under the classical theory, a corporate insider or temporary fiduciary of a corporation (e.g., an attorney, accountant, or consultant) breaches a fiduciary duty to shareholders by trading on MNPI for her personal benefit, unfairly advantaging her vis-à-vis uninformed stockholders. See Chiarella v. United States, 445 U.S. 222, 228-29, 100 S.Ct. 1108, 63 L.Ed.2d 348 (1980); see also United States v. O'Hagan, 521 U.S. 642 652-53, 117 S.Ct. 2199, 138 L.Ed.2d 724 (1997); Dirks v. S.E.C., 463 U.S. 646, 655 n.14, 103 S.Ct. 3255, 77 L.Ed.2d 911 (1983). Under the misappropriation theory, a corporate outsider breaches a duty of trust owed to the source of confidential information by trading on that information. See O'Hagan, 521 U.S. at 652-53, 117 S.Ct. 2199 (lawyer who traded on confidential information entrusted to his law firm). Under both theories, where the trading is by a downstream "tippee," for the "tipper" source of the information to be liable, she must have had the purpose to personally benefit, directly or indirectly, from the disclosure. See Salman v. United States, 580 U.S. 39, 137 S. Ct. 420, 427, 196 L.Ed.2d 351 (2016) (citing Dirks, 463 U.S. at 662-64, 103 S.Ct. 3255, and holding that "the disclosure of confidential information without personal benefit is not enough").

As a result, insider trading law reaches only a small subset of the circumstances under which a corporation's MNPI may be selectively disclosed. Critically, absent proof of intended personal benefit, insider trading law does not reach a corporate

insider's selective disclosures of MNPI aimed at benefitting the corporation. For this reason, as the SEC reasoned, insider trading laws are unlikely to reach or deter stratagems—such as alleged here—to selectively divulge MNPI for corporate benefit, notwithstanding the capacity of such disclosures to enable recipients (or persons further downstream) to profit from this "informational edge." See Adopting Release, 2000 WL 1201556, at *2 & n.7 (noting that "as a result of judicial interpretations, tipping and insider trading can be severely punished under the antifraud provisions of the federal securities laws, whereas the status of issuer selective disclosure has been considerably less clear" and that "in light of the 'personal benefit' test [in Dirks], ... many have viewed issuer selective disclosures to analysts as protected from insider trading liability").

The SEC further found that it is common for corporations to selectively disclose MNPI for their benefit. It stated:

As reflected in recent publicized reports, many issuers are disclosing important nonpublic information, such as advance warnings of earnings results, to securities analysts or selected institutional investors or both, before making full disclosure of the same information to the general public. Where this has happened, those who were privy to the information beforehand were able to make a profit or avoid a loss at the expense of those kept in the dark.
We believe that the practice of selective disclosure leads to a loss of investor confidence in the integrity of our capital markets. Investors who see a security's price change dramatically and only later are given access to the information responsible for that move rightly question whether they are on a level playing field with market insiders.

Adopting Release, 2000 WL 1201556, at *2.

As the SEC also recognized, a rule aimed at assuring that MNPI disclosures reach the public and not a privileged few advances purposes recognized in the case law. See id. at *2-3, 20-22. Such a rule would track laws requiring the disclosure and exchange of information about securities. See Ohralik, 436 U.S. at 456, 98 S.Ct. 1912 ("Numerous examples could be cited of communications that are regulated without offending the First Amendment, such as the exchange of information about securities."). It would advance the "animating purpose of the Exchange Act: to insure honest securities markets and thereby promote investor confidence." See O'Hagan, 521 U.S. at 658, 117 S.Ct. 2199 (citing 45 Fed. Reg. 60,412 (1980) (recognizing that trading on misappropriated information "undermines the integrity of, and investor confidence in, the securities markets")); see also id. at 659, 117 S.Ct. 2199 (citing Victor Brudney, Insiders, Outsiders, and Informational Advantages Under the Federal Securities Laws, 93 HARV. L. REV. 322, 356 (1979) ("If the market is thought to be systematically populated with ... transactors [trading on the basis of misappropriated information,] some investors will refrain from dealing altogether, and others will incur costs to avoid dealing with such transactors or corruptly to overcome their unerodable informational advantages.")). And it would promote the integrity of the financial markets.

For these reasons, the Court finds, Reg FD directly and materially advances substantial governmental interests. See, e.g., Centro, 868 F.3d at 115; Am. Meat Inst. v. U.S. Dep't of Agric., 760 F.3d 18, 23-27 (D.C. Cir. 2014) (en banc); Vugo, 931 F.3d at 51-53; Clear Channel Outdoor, Inc. v.

City of New York, 594 F.3d 94, 106 (2d Cir. 2010).

Fourth factor : As to the fourth Central Hudson factor, the SEC need not "show that it chose the least restrictive means of advancing its asserted interests, nor that there was no conceivable alternative. Rather, its obligation [is] to establish that the regulation not burden substantially more speech than necessary to further its legitimate interests." Centro, 868 F.3d at 115 (citing Safelite, 764 F.3d at 265; and Clear Channel Outdoor, 594 F.3d at 104).

Reg FD meets that standard. It does not prohibit speech. It does not compel speech containing any new content; its sole requirement, that MNPI be disseminated publicly, is limited to circumstances in which a corporation has already disclosed—selectively—the same MNPI. And, as the SEC explained, it is simple and inexpensive for a corporation to disseminate such information to the market. Defendants do not argue to the contrary.

Adopting Reg FD in 2000, the SEC noted that "technological developments have made it much easier for issuers to disseminate information broadly," there being "a variety of methods to communicate directly with the market[] .... includ[ing], among others, Internet webcasting and teleconferencing." Adopting Release, 2000 WL 1201556, at *3. It was thus not the case that "technological limitations" would "provide an excuse for abiding the threats to market integrity that selective disclosure represents." Id. Ensuing advances in technology have made the process of market-wide disclosure, if anything, easier.

Reg FD thus does "not burden substantially more speech than is necessary to further its legitimate interests." Centro, 868 F.3d at 115. See, e.g., Vugo, 931 F.3d at 58-59; Clear Channel Outdoor, 594 F.3d at 106; Anderson v. Treadwell, 294 F.3d 453, 465 (2d Cir. 2002).

The Court accordingly finds that Reg FD, as applied, satisfies intermediate scrutiny—the most rigorous level potentially applicable to it—and does not violate the First Amendment.

B. Fifth Amendment Challenge

Defendants make a brief challenge to Reg FD under the Fifth Amendment. In full, it consists of these three sentences (and a footnote) on the final page of AT & T's brief:

Regulation FD also violates the Fifth Amendment's due process clause. The government must provide clear notice of what conduct is prohibited. As set forth above, the text of Regulation FD, as well as the SEC's conflicting public positions, failed to provide AT & T with fair notice of what conduct is actually prohibited and has a chilling effect on protected speech.

AT & T MSJ Mot. at 49 (citing City of Chicago v. Morales, 527 U.S. 41, 56, 119 S.Ct. 1849, 144 L.Ed.2d 67 (1999); Grayned v. City of Rockford, 408 U.S. 104, 108-09, 92 S.Ct. 2294, 33 L.Ed.2d 222 (1972); and Baggett v. Bullitt, 377 U.S. 360, 372, 84 S.Ct. 1316, 12 L.Ed.2d 377 (1964)). This challenge based on a lack of clear notice fails.

Insofar as defendants refer to what they had "set forth above," see AT & T Mot. at 47, they cited there to the 2005 decision in Siebel, the one prior reported decision involving Reg FD. Judge Daniels there noted, in dicta, the "potential chilling effect" of Reg FD, were it applied as the SEC proposed there. But that observation was based on his immediately preceding determination that "none of the statements challenged by the SEC" in that case fell "squarely within the seven enumerated categories listed by the SEC, in the Adopting Release, as being more likely to be

considered material." Siebel, 384 F. Supp. 2d at 708. As a result, Judge Daniels reasoned, Reg FD had not given Siebel Systems and the two individual executive defendants "clear guidance for companies to conform their conduct in compliance" with the law. Id. In the end, however, Judge Daniels "decline[d] to opine on the constitutional challenges raised," including under the Fifth Amendment. Id. at 709 n.16. He instead dismissed on the ground that the facts pled did not make out a violation of Reg FD, as "the statements relied upon in the complaint fail[ed] to support its conclusory allegation that material information disclosed by Mr. Goldman in private, had not already been publicly disclosed by Siebel Systems." Id. at 709; see also id. at 701 ("The complaint relies on no statements regarding specific earnings or sales figures.").

The SEC's present case against AT & T and the IR defendants is not subject to this critique. The SEC alleges that defendants selectively disclosed information that, unavoidably, had not been publicly available: specific metrics including AT & T's 1Q16 total revenue, wireless equipment revenue, and wireless upgrade rates, and signals that analysts' estimates of those metrics were higher than AT & T's actual internal projections. See, e.g., Compl. ¶¶ 1, 5-7, 46-50, 64-130.

Insofar as defendants contend that the SEC has made "conflicting public positions" as to what Reg FD prohibits, AT & T Mot. at 48, that argument is baseless. Defendants rely on the SEC's statement in Reg FD's Adopting Release that it would pursue only clear-cut violations of the rule. They argue that that statement is betrayed by the agency's having brought Siebel. But that single deficiently pled 2005 case does not prove that the SEC has "inexplicably departed from [its] assurances," id., and defendants do not cite any other enforcement action under Reg FD that has been challenged and found deficient, see supra pp. 712-14 (listing 17 Reg FD actions in the rule's 20+ years in the Federal Register). And, in this case, the Court has today found sufficient evidence on which a reasonable jury could find a violation—by AT & T and the individual defendants—of Reg FD. See infra Part V.

Finally, insofar as defendants assail Reg FD as "so vague and standardless that it leaves the public uncertain" as to the conduct prohibited, Morales, 527 U.S. at 56, 119 S.Ct. 1849, that salvo misses the mark. The regulation, in fact, is short and exceedingly clear: it requires covered entities to promptly disclose material, nonpublic communications that it disclosed on a selective basis. 17 C.F.R. § 243.100. The Adopting Release explains and amplifies the rule. And, far from using elusive standards, Reg FD incorporates the existing definitions of "material" and "nonpublic"—terms familiar to securities lawyers. See Adopting Release, 2000 WL 1201556, at *9 nn.38-40 & accompanying text (citing TSC Industries, 426 U.S. at 449, 96 S.Ct. 2126; Basic, 485 U.S. at 231, 108 S.Ct. 978; Tex. Gulf Sulphur, 401 F.2d at 854; and SEC Staff Accounting Bulletin No. 99, 64 Fed. Reg. 45,150 (Aug. 12, 1999), 1999 WL 1123073 ("SAB No. 99") (discussing materiality for purposes of financial statements)).

Reg FD, as applied here, thus does not offend the Fifth Amendment.

C. Challenge to Statutory Authority

Defendants next argue that the SEC lacked statutory authority to promulgate Reg FD. The SEC primarily responds that Section 13(a) of the Exchange Act empowered it to do so. The SEC is correct.

Section 13(a) of the Exchange Act provides, in relevant part:

Every issuer of a security registered pursuant to section 78l [Section 12(b)] of this title shall file with the Commission, in accordance with such rules and regulations as the Commission may prescribe as necessary or appropriate for the proper protection of investors and to insure fair dealing in the security—
(1) such information and documents (and such copies thereof) as the Commission shall require to keep reasonably current the information and documents required to be included in or filed with an application or registration statement filed pursuant to section 78l of this title, except that the Commission may not require the filing of any material contract wholly executed before July 1, 1962[; and]
(2) such annual reports (and such copies thereof), certified if required by the rules and regulations of the Commission by independent public accountants, and such quarterly reports (and such copies thereof), as the Commission may prescribe.

15 U.S.C. § 78m(a).

Defendants' argument is based on the text of Sections 13(a)(1) and (2). As to Section 13(a)(1), defendants argue that the information Reg FD regulates does not bear on any "'application' or 'registration statement,'" and they dispute that "the filing of such information would 'keep current' the information required in any AT & T 'application' or 'registration,'" and thus exceeds the authority conferred by Section 13(a)(1). AT & T MSJ Mot. at 43. As to Section 13(a)(2), they argue that Reg FD's "continuous disclosure" requirement provides for disclosure more often than the annual and quarterly reports addressed by that provision. Id. Finally, defendants argue that because the Supreme Court has held that Section 10(b)—pursuant to which the SEC proscribes insider trading—does not embody "an absolute equal information rule," id. (citing Dirks, 463 U.S. at 657 n.16, 103 S.Ct. 3255), Section 13(a) cannot either. Defendants characterize Reg FD as such a rule.

Defendants overlook the aspects of the SEC statutory and regulatory framework that authorized the agency to promulgate Reg FD. As the SEC notes, Section 13(a)(1) of the Exchange Act, considered in tandem with the text immediately preceding it in Section 13, expressly authorizes the agency to promulgate regulations germane to Section 12 of the Act. It provides that the SEC may promulgate regulations governing "such information and documents ... as the Commission shall require to keep reasonably current the information and documents required to be included in or filed with an application or registration statement filed pursuant to [Section 12]." 15 U.S.C. § 78m(a)(1). Section 12(b), in turn, empowers the SEC to issue regulations governing the information that must be disclosed for a security to be registered on an exchange. See 15 U.S.C. § 78l(b). Pursuant to that authority, the SEC has established a broad array of disclosure requirements, including those required by SEC Form 10. See 17 C.F.R. § 249.210. And SEC Form 10, in relevant part, requires the disclosure of the information required by Regulation S-K. See, e.g., SEC Form 10 Item 1 ("Furnish the information required by Item 101 of Regulation S-K."). As for the content that Reg FD requires an issuer to publicly disclose where it has been selectively disclosed, it generally tracks the content that must be reported on the non-financial statement portion of registration statements and on Form 8-K, used to report major corporate events. See 17 C.F.R. § 229.10 et seq.; id. § 229.101(e)(4)(i); id. § 249.308. Section 7 of Form 8-K, in fact, provides for issuers to disclose on that Form information covered by Reg FD. In light of this framework, defendants' arguments based on Section 13(a)(1) fail. The text prefacing Section 13(a)(1) authorizes the SEC to promulgate regulations in service of "information and documents ... the Commission shall require to keep reasonably current the information" included in issuers' registration statements. Reg FD—which sets out some of the information issuers are to disclose on their "current reports" to the SEC—is such a regulation. And for the reasons reviewed above as to Reg FD's purposes, Reg FD implements Section 13(a)'s command that the SEC "prescribe as necessary or appropriate for the proper protection of investors and to insure fair dealing in the security." See supra Section IV.A.2.

Defendants are also wrong to claim that Reg FD is in conflict with Dirks. Dirks stated that "formulation of an absolute equal information rule 'should not be undertaken absent some explicit evidence of congressional intent.'" 463 U.S. at 657 n.16, 103 S.Ct. 3255 (quoting Chiarella, 445 U.S. at 233, 100 S.Ct. 1108). Reg FD does no such thing. It requires public access only to that MNPI that an issuer has selectively disclosed, so as to advance Section 13(a)'s goals of investor protection and fair dealing. It does not create a freestanding right to corporate information or oblige an issuer to redress information asymmetries in the market other than those brought about by its own selective disclosures.

The Court, therefore, finds that Section 13(a) of the Exchange Act empowered the SEC to promulgate Reg FD, and rejects defendants' argument to the contrary.

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").

D. Regulatory Operability

Defendants' final threshold argument is that Reg FD is void. They argue that Reg FD's exception for disclosures made in breach of a "duty of trust or confidence to the issuer" swallows the rule, making the regulation "logically inoperable."

Reg FD applies to "an issuer, or any person acting on its behalf." 17 C.F.R. § 243.100(a). And, as noted, it defines a "[p]erson acting on behalf of an issuer" as "any senior officer of the issuer ... or any other officer, employee, or agent of an issuer" who communicates with certain outsiders. Id. § 243.101(c). But it contains an exception: "An officer, director, employee, or 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." Id. (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 to 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, — U.S. —, 139 S. Ct. 2400, 2414, 204 L.Ed.2d 841 (2019) ("The drafters will know what [the regulation] was supposed to include or exclude or how it was supposed to 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—if 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 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, 106 S.Ct. 2548, 91 L.Ed.2d 265 (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. Iona 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). 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(c)(1)(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, 106 S.Ct. 2505, 91 L.Ed.2d 202 (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 v. Oaktree Specialty Lending Corp., No. 21 Civ. 4770 (PAE), 600 F.Supp.3d 463, 475-76 (S.D.N.Y. 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.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 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 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) [e]arnings information; (2) mergers, acquisitions, tender offers, joint ventures, or changes in assets; (3) new products or discoveries, or developments regarding customers or supplies (e.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 MO. 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, 108 S.Ct. 978; see also Matrixx Initiatives, Inc. v. Siracusano, 563 U.S. 27, 38, 131 S.Ct. 1309, 179 L.Ed.2d 398 (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 (2d 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 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, 96 S.Ct. 2126.

Although leaving open that other types of information could qualify as material under Reg FD, the SEC has signaled that it would not pursue "close calls" in this area. As Judge Daniels recounted in Siebel:

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 "[a]n 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." Id. The views expressed in Mr. Walker's speech were his own and did not necessarily reflect the views of the SEC or its staff. Id. 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." Id.


Siebel, 384 F. Supp. 2d at 702 n.8.

"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, 96 S.Ct. 2126; S.E.C. v. 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." In 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 v. Citizens Utils. Co., 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 ... a deviation of less than the specified percentage with 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.").

"[S]ince 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) (S.E.C. 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, 96 S.Ct. 2126 ("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.")).

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."); EGA, 553 F.3d at 204 (quantitative assessment of materiality "must be supplemented" by evaluation of SAB No. 99's qualitative 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 S.E.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.E.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 v. 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-à-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); S.E.C. v. Winemaster, 529 F. Supp. 3d 880, 909-10 (N.D. Ill. 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").

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.

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 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) ¶ 71. 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 ¶ 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 (S.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 (11th Cir. 2014); Veleron Holding, B.V. v. Morgan Stanley, 117 F. 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); S.E.C. 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. DCI 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 v. 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 DirecTV—overshadowed the bad news, causing AT & T's stock price 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, if any, 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.

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 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 S.E.C. 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."); cf. 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 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 "[r]evenue 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); JX 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 to 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 "be[ing] in the same spot we've been in the last few quarters, i.e. missing revenue." Id. at 2; see also infra Sections I.F-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 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, cc'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.

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) ("[J]ust 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)).

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 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 "[i]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.E.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 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 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

See Basic, 485 U.S. at 236 n.14, 108 S.Ct. 978 & 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).

immaterial simply by breaking it into ostensibly non-material pieces.").

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.

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 IBEW 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").

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, "[o]ur 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).

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 ("[B]y 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 JX 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?").

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").

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).

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 IR defendants disclosed as to these metrics material.

Wireless equipment revenue's impact on 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."

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.").

See also sources cited at AT & T MSJ 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"); IX 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).

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. DIRECTV, 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").

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 non-public information regarding credit card revenue data was material), aff'd, 684 F. App'x 167.

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 ¶¶ 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 attributing 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.

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." Id.

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 group.'" Mayhew, 121 F.3d at 50 (quoting Dirks, 463 U.S. at 653 n.12, 103 S.Ct. 3255). "[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 'guidance,' 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 judgment 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 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 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.

JSF ¶ 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 analysts. They did not come close to doing so. Stephens's remarks did not disclose AT & T's Q1 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 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.

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 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 ¶¶ 245-46, P. Exs. 23-24, 28, JXs 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).

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).

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").

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 JXs 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 I.H.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 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.

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"); In 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.'").

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 lower than projected sales actually are occurring clearly constitute material information vis-à-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 MS J 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 non-public information.

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, "[i]n 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.

Black's argument cannot win him summary judgment, even assuming Ms. 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. ¶¶ 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 I.H.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 year-over-year decline, RBC); 5% (wireless equipment upgrade rate); and 22.5-23% (total revenue growth). Id. 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 I.H.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 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 at 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.

Analysts' notes during Black's conversations reflect, in general, numbers strikingly similar to AT & T's projections during those periods. See supra Sections I.H.2, I.H.5 (detailing evidence surrounding Black's communications with Wells Fargo, RBC, Scotiabank, Buckingham, and Bank of America analysts).

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 "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 Q1 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.

1. 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 disclosure either knows, or is reckless in not knowing, that the information he or she is communicating is both material and nonpublic." Id. § 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 S.E.C. 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." Id.

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.

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." Id.

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.

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—finding 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 In re Raytheon Co., 2002 WL 31643026; and In 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 "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.").

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.).

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 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.Y. Jan. 3, 2022) ("The fact-intensive disputes over falsity, materiality, and scienter will likewise be decided by the jury."); In 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 1 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 St., LLC v. 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 v. Eichler, 264 F.3d 131, 144 (2d Cir. 2001) (same where allegations did "not provide strong evidence of conscious misbehavior or recklessness"); 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 ¶¶ 17, 63; see also In re Columbia Sec. Litig., 155 F.R.D. 466, 479 (S.D.N.Y. 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 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 in 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 Tr. at 17 (defense counsel, conceding at argument that the analysts in this case had rarely reported Reg

See, e.g., AT & T MSJ 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.

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 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)).

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. 711-12, 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.

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"); In 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).

In 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.

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

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.

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.

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.

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 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.

See SEC Resp. to ID 56.1 ¶ 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).

G. Dispositions

As to the claims against Womack, Evans, and Black, in light of its 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. ¶ 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 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.

SO ORDERED.


Summaries of

Sec. & Exch. Comm'n v. AT&T Inc.

United States District Court, S.D. New York
Sep 8, 2022
626 F. Supp. 3d 703 (S.D.N.Y. 2022)
Case details for

Sec. & Exch. Comm'n v. AT&T Inc.

Case Details

Full title:SECURITIES AND EXCHANGE COMMISSION, Plaintiff, v. AT&T, INC., CHRISTOPHER…

Court:United States District Court, S.D. New York

Date published: Sep 8, 2022

Citations

626 F. Supp. 3d 703 (S.D.N.Y. 2022)

Citing Cases

United States v. Patel

While there are hearsay exceptions that might apply, such as 803(1) or 803(5), see, e.g., Sec. & Exch. Comm'n…

Sec. & Exch. Comm'n v. City of Rochester

And Rule G-42 is one of many regulations and laws governing disclosures related to securities-regulations and…