Abstract
This section provides summary and light analysis of Supreme Court and federal Court of Appeals cases from 2023 addressing securities law issues.
The Business Lawyer
Caselaw Developments 2023
iStock.com/Mardoz
Jump to:
Caselaw Developments 2023
A. Overview
1. Supreme Court
The Court held that, where an issuer made a direct offering by a registration statement, purchasers of its already unrestricted shares that began trading on a national exchange but were not covered by the registration statement could not assert a claim under section 11 of the Securities Act of 1933 (the “Securities Act”). The Court also held that respondents in Securities and Exchange Commission (“SEC” or “Commission”) and Federal Trade Commission (“FTC”) administrative enforcement actions could sue in federal district court, before the administrative proceedings ran their course inside the agency, on the basis that the administrative law judges (“ALJs”) who would initially hear the proceedings were unconstitutionally insulated from removal by the President.
2. Courts of Appeals
SEC rulemaking and orders. The Fifth Circuit vacated an SEC order approving a regulation to require issuers to provide daily totals for stock that they repurchased and to explain the reasons for repurchases.
Proxy statements. The Seventh Circuit affirmed dismissal of a Rule 14a-9 claim based on a proxy statement’s failure to discuss an alternative transaction structure the plaintiffs alleged could have potentially produced a deal that was tax-free for the target’s shareholders.
Forum-selection clauses. The Ninth Circuit held that a forum-selection clause in an issuer’s bylaws was enforceable to dismiss, based on forum conveniens grounds, a derivative lawsuit filed in federal court asserting a Rule 14a-9 claim.
Materially false or misleading statements. The Ninth Circuit reversed dismissal of a Rule 10b-5 claim where the plaintiffs charged misallocation of company sales between business segments and supported that charge with a financial model. That circuit also held a plaintiff pled a Rule 10b-5 claim by alleging that risk factors misled by suggesting that problems could occur in the future when in fact the company had already experienced them. The Second Circuit affirmed dismissal of the Rule 10b-5 claim but reversed to some extent dismissal of Securities Act section 11 and section 12(a)(2) claims, where the case revolved around accounting errors in recognizing revenue from service contracts too soon and recording charges for bonuses too late.
Scienter and scienter pleading. The Tenth Circuit affirmed dismissal of a Rule 10b-5 case alleging fraud by statements that a principal customer would continue ordering from the issuer after that customer announced cutbacks in its own production.
Rebutting price impact on class certification. The Second Circuit brought to an end a long-running class certification fight in a Rule 10b-5 case against Goldman Sachs, (i) ruling that the defendants had carried their burden of proof in showing no price impact from the challenged statements, with the rationale turning largely on the generality of those statements and (ii) remanding with an instruction to decertify the class.
Life sciences. The Fourth Circuit affirmed dismissal of a Rule 10b-5 claim alleging that a pharmaceutical company misled by announcing favorable interim results without disclosing that, during the interim period, there were times when the company’s tested drug was outperforming the alternative treatment and times when the alternative treatment was outperforming the tested drug. The First Circuit affirmed dismissal of a complaint alleging Rule 10b-5 and Securities Act section 11 and section 12(a)(2) violations and charging that the issuer misled by statements that high dosages had proven efficacious—with the exception that the court of appeals reversed dismissal of the Rule 10b-5 claim to the extent it was based on the CEO’s statement that “all data” supported the high-dose efficacy conclusion, which plaintiffs alleged was false as to some patient subgroups.
SEC enforcement. Disagreeing with the Fifth Circuit, the Second Circuit held that (i) amendments to 15 U.S.C. § 78u(d) did not create a new type of disgorgement for SEC enforcement actions; (ii) all such disgorgements must conform to equitable principles; and (iii) the district court (a) failed to properly compute the “net profits” to be disgorged and (b) as to most assets, failed before ordering disgorgement by relief defendants to make a predicate finding that those defendants held bare legal title to the assets while the wrongdoer was the equitable owner. The Ninth Circuit reversed summary judgment on civil penalties, holding that the defendant raised issues of fact on the amount of his “pecuniary gain,” the degree of his scienter, and his acknowledgment of wrongdoing.
B. Supreme Court
The Supreme Court held that, where an issuer filed a registration statement to cover the sale of some 118 million shares but exchange rules permitted the simultaneous listing of 165 million additional outstanding shares, a purchaser who could not trace his shares to the 118 million shares covered by the registration sale could not bring a section 11 claim. The Court also held that a respondent in an administrative enforcement proceeding brought by the SEC could sue in federal district court under federal question jurisdiction to challenge the constitutionality of the ALJ’s tenure protection without waiting through the entire administrative proceeding and raising that question in court of appeals review.
1. Application of Section 11 to Direct Listing
Securities Act section 5 requires that every sale of a security either (i) be made pursuant to an effective registration statement or (ii) qualify for an exemption from registration. Securities Act section 11 provides that if a registration statement—when it becomes effective—contains an untrue statement of material fact or omits a material fact necessary to render a passage in the registration statement not misleading, then “any person acquiring such security” may sue a list of specified defendants, including the issuer on which liability is nearly absolute.
Slack Technologies, LLC engaged in a direct listing in 2019. Slack itself did not sell any stock by the registration statement it filed in connection with that listing. Instead, the registration covered the resale of 118 million Slack shares that were already outstanding.
However, the registration statement was also important to the remaining 165 million outstanding shares. Those shares could be sold without registration pursuant to their exemption by Securities Act section 4(a)(1) and Rule 144. But the practical ability to sell them depended on a national stock exchange listing, and a New York Stock Exchange Rule provided that all 283 million Slack shares would become tradeable through that exchange when the registration statement became effective—both the 118 million shares covered by the registration statement and 165 million that were exempt from registration.
Fiyyaz Pirani purchased 30,000 shares on the date when the Slack stock began trading on the exchange and 220,000 more over the following few months. After the price of Slack shares declined, he sued under Securities Act sections 11 and 12. Pirani did “not allege[] that he purchased shares traceable to the allegedly misleading registration statement. For all anyone could tell, he may have purchased unregistered shares unconnected to the registration statement and its representations about the firm’s business and financial health.”
Reversing a Ninth Circuit decision affirming the district court denial of a motion to dismiss, the Supreme Court held that Section 11 “requires a plaintiff to plead and prove that he purchased shares traceable to the allegedly defective registration statement.” While acknowledging that “there is no clear referent in § 11(a) telling us what ‘such security’ means,” the Court found “clues” in the statutory “context” in which section 11 resides. Section 11 itself speaks of liability based on statements or omissions in “‘the registration statement,’” rather than “a registration statement or any registration statement”—suggesting that a “plaintiff must ‘acquir[e] such security’ under that document’s terms.” Section 5 provides that “‘[u]nless a registration statement is in effect as to a security,’ it is unlawful ‘to sell such security’”—“clearly refer[ring] to shares subject to registration.” Section “6 provides that a ‘registration statement shall be deemed effective only as to the securities specified therein as proposed to be offered.’” The damages limitation for claims against underwriters in section 11(e) “ties the maximum available recovery to the value of the registered shares alone.”
While Pirani argued that his suit should go forward “because, but for the existence of Slack’s registration statement for the registered shares, its unregistered shares would not have been eligible for sale to the public,” that argument could not “be squared with the[se] various context clues.” Moreover, if Congress had intended the result Pirani urged, it “could have written § 11(a) to explain more clearly that liability attaches to ‘any security’ or ‘any security’ bearing some specified relationship to a registration statement.”
Significance and analysis. The Court noted that the Ninth Circuit explicitly tied its section 12 analysis to its section 11 analysis and therefore found “the best course is to vacate [the Ninth Circuit’s] judgment with respect to Mr. Pirani’s § 12 claim . . . for reconsideration in the light of our holding today about the meaning of § 11.”
2. Federal Court Injunction Against SEC Administrative Enforcement Proceeding
Both the Securities Exchange Act of 1934 (“Exchange Act”) and the Federal Trade Commission Act (“FTC Act”) permit agency staffs to bring enforcement actions either (i) in federal court or (ii) in administrative proceedings before the agency’s ALJs. If the agency chooses the second path, a party can contest a final order emerging from the administrative process by petitioning for review in a federal court of appeals.
Section 1331 of U.S. Code Title 28 gives federal district courts “original jurisdiction of all civil actions arising under the Constitution, laws, or treaties of the United States.”
In 2023, the Supreme Court considered whether, on the one hand, (i) 28 U.S.C. § 1331 permits a respondent in an administrative enforcement proceeding to sue in district court—before the agency has completed its adjudicatory process—for an injunction to stop the administrative proceeding on the ground that the agency’s process is structurally unconstitutional; or (ii) on the other hand, the respondent must litigate the constitutional issue all through the administrative process, then petition for review of an adverse agency order to obtain a court answer to the constitutional question.
The Court rendered a single decision in two cases—one from the Fifth Circuit (involving an SEC order) and one from the Ninth Circuit (involving an FTC order). In the Fifth Circuit case, Michelle Cochran had been named as a respondent in an SEC administrative enforcement proceeding before an SEC ALJ. Cochran filed a case in federal district court to stop the administrative proceeding, alleging in her complaint that the Merit Systems Protection Board (“MSPB”) can only remove an ALJ “for good cause” and the President can only remove an MSPB member for “cause, such as ‘neglect of duty’ or ‘malfeasance.’” Cochran argued that the ALJ’s double for-cause insulation from presidential removal unconstitutionally infringed on presidential supervisory power granted by Article II.
In the Ninth Circuit case, the FTC brought an administrative enforcement proceeding against Axon Enterprises (“Axon”). Axon sued in district court to enjoin that proceeding, making the same argument as Cochran as to ALJ tenure and, in addition, contending that “the combination of prosecutorial and adjudicative functions in the [FTC] renders all of its enforcement actions unconstitutional.” After the district courts dismissed both the Cochran and Axon suits, the Fifth Circuit en banc reversed and the Ninth Circuit affirmed.
The Supreme Court affirmed the Fifth Circuit and reversed the Ninth Circuit. Justice Kagan’s opinion for the Court employed the framework set out in Thunder Basin Coal Co. v. Reich to determine whether “by specifying” a particular method to resolve claims about agency action—“review in a court of appeals following the agency’s . . . process”—the securities laws and the FTC Act “implicitly” narrowed the unlimited jurisdiction granted by § 1331 to exclude claims of structural unconstitutionality from district court adjudication at the outset of administrative proceedings. Put another way, the Court sought to determine “whether the particular claims” here were “‘of the type Congress intended to be reviewed within [the] statutory structure’” and accordingly outside § 1331’s broad reach.
Thunder Basin identifies three critical factors. The first is “whether preclusion of district court jurisdiction ‘could foreclose all meaningful judicial review.’” Focusing on “the interaction between the alleged injury and the timing of [the] review,” the Court found the injury asserted by both respondents here consisted of “‘being subjected’ to ‘unconstitutional agency authority’—a ‘proceeding by an unaccountable ALJ.’” That harm could not be remedied by a court of appeals review after the administrative proceeding had run its course because “[a] proceeding that has already happened cannot be undone.” Therefore, “[j]udicial review of Axon’s (and Cochran’s) structural constitutional claims would come too late to be meaningful” if left to after-the-fact court of appeals consideration.
The second Thunder Basin factor considers whether “the claim [in the district court is] ‘wholly collateral’” to the review process built into the administrative process—i.e., court of appeals review after final agency adjudication. That factor, too, favored Cochran and Axon “because they are challenging the Commissions’ power to proceed at all, rather than actions taken in the agency proceedings” and both of them “object to the Commissions’ power generally, not to anything particular about how that power was wielded.” Their claims “in sum, have nothing to do with the enforcement-related matters the Commissions ‘regularly adjudicate[ ]’—and nothing to do with those they would adjudicate in assessing the charges against Axon and Cochran.”
The last of the three factors was whether Cochran and Axon’s claims were “‘outside the agency’s expertise.’” Completing the trifecta, this one also favored § 1331 district court jurisdiction because neither the SEC nor the FTC have any special competence or expertise in the constitutional questions of whether the ALJ tenure protections violate Article II of the Constitution or, added by Axon, whether combining the prosecutorial role and the adjudicative role in a single agency violates the Constitution.
With “‘[a]ll three Thunder Basin factors thus point[ing] in the same direction—toward allowing district court review of Axon’s and Cochran’s claims that the structure, or even existence, of an agency violates the Constitution”—the Court concluded that “[a] district court can therefore review” those claims and “resolve” them under § 1331.
Significance and analysis. The Court hastened to point out that “[n]othing we say today portends newfound enthusiasm for interlocutory review” and that the cases generating the decision were exceptions from the statutorily prescribed administrative route followed by court of appeals review, with the exceptions justified not by a run-of-the-mill objection that an administrative proceeding was imposing litigation and disruption costs on a respondent but by the different objection that the enforcement proceeding was—by its very nature, and without regard to outcome—structurally unconstitutional. Inventive counsel will test the boundaries of this limitation.
C. Courts of Appeals
1. SEC Rulemaking and Orders
In May 2023, the SEC adopted a rule requiring issuers to report, at the end of each quarter, the daily repurchases they made of their own stock during that three-month period, together with the reason for making the repurchases. The Fifth Circuit granted a petition for review of that rule and remanded to give the Commission thirty days in which to correct the defects the court of appeals found in the rule’s adoption. The SEC failed to meet that deadline, and the Fifth Circuit vacated the rule.
The Administrative Procedure Act (“APA”) requires a reviewing court to “hold unlawful and set aside agency action” that is “contrary to constitutional right” or “arbitrary [or] capricious,” or “without observance of procedure required by law.” The petition challenged the repurchase regulation on all three of these fronts.
On the constitutional front, the petition contended that the requirement to state the reason for repurchases violated the First Amendment by compelling speech. Employing the test set out by the Supreme Court for commercial speech, the Fifth Circuit rejected this challenge because the regulation (i) only required disclosure of “‘“purely factual and uncontroversial information”’”; (ii) that was “‘reasonably related’” to a legitimate state interest by seeking to “‘remedy a harm that is potentially real[,] not purely hypothetical’”; and (iii) was not “‘unduly burdensome’” because it was narrowly targeted on that harm. As to the first of these three requirements, “forcing a company to ‘explain the reason’ for its actions is a purely factual disclosure,” and “[i]f a social media company’s reason for removing user content was uncontroversial [as the Fifth Circuit had held in an earlier decision for purpose of applying the commercial speech test], then an issuer’s reason for repurchasing its own shares is uncontroversial” for that same purpose here. The repurchase regulation satisfied the second requirement because it is reasonably related to “a supposed asymmetry in information [between that known to the company and that known to shareholders] surrounding the reasons issuers repurchase their shares”—“whether a share repurchase was intended to increase the value of the issuer” or for other purposes, “such as a desire to achieve accounting metrics or impact executive compensation.” The rule met the third condition “because the only speech it compels relates directly to alleviating the information asymmetry.” In thus rejecting the constitutional challenge, the Fifth Circuit pointed out that the rationale for the rule, while sufficient for First Amendment purposes, might “not be enough to survive APA review.”
Turning to that very issue, the appellate court rejected one APA challenge to the rule but sustained two others. First, when it proposed the rule, the SEC “stated that [it] was unable to quantify most of the economic effects” of the proposal and “encourage[ed] commenters to provide information that could help quantify the costs and benefits.” In adopting the regulation, the Commission continued to “maintain[] that many of the effects of the daily-disclosure requirement could not be quantified” but “did . . . perform a cost-benefit analysis for both the rationale-disclosure requirement and the daily-disclosure requirement.” On review, the petitioner contended that “quantitative data is the ‘best’ data, so . . . the SEC cannot rely merely on qualitative analyses without first explaining why a rule’s costs and benefits ‘could not be quantified.’” The Fifth Circuit rejected this view, holding that the SEC “is not required to undertake a quantitative analysis to determine a proposed rule’s economic implications” because the relevant statutory text only requires the Commission to “‘consider’” certain factors—such as “‘whether [an] action will promote efficiency, competition, and capital formation[,’] . . . whether an action is ‘necessary or appropriate in the public interest,’ . . . or ‘consistent with the public interest.’” These words do “not restrict the universe of otherwise permissible methods by which the SEC can analyze the economic implications.” Instead, “[i]t is within the agency’s discretion to determine the mode of analysis.”
Second, and nevertheless, the Commission was “require[d] . . . to consider all relevant factors raised by the public comments and provide a response to significant points within.” Here the agency had itself asked commenters for assistance with quantification, and the petitioner had, in fact, made specific suggestions. The “SEC admits it never considered any of [them].” Moreover, the Commission’s proffered reasons for failing to do so were false. It claimed that the comments did not “identify any specific [data] already available that the Commission should have used” but the comments expressly identified databases that the SEC could access for some analysis, argued that the SEC could perform some analysis with data from existing required issuer disclosures, and suggested replicating a specific study from the U.K. The SEC’s second justification—that the quantitative analysis the comments suggested would not have impacted the proposed rule—was belied by the circumstance that the Commission itself had solicited help in performing such an analysis when it proposed the new rule. And all of the suggestions “address[ed] costs and benefits the SEC identified in the proposed rule”—including “the prevalence of improperly motivated buybacks” and “the strength of the incentives underlying such opportunistic repurchases.” As a result, “[t]he SEC—by continuing to insist that the rule’s economic effects are unquantifiable in spite of petitioners’ suggestions to the contrary—has failed to demonstrate that its conclusion that the proposed rule ‘promote[s] efficiency, competition, and capital formation’ is ‘the product of reasoned decisionmaking.’”
Third, the SEC “failed adequately to substantiate the rule’s benefits and costs.” Most importantly, the Commission failed to substantiate the existence of the condition motivating the rule—that “‘improperly motivated buybacks are actually a problem.’” The SEC attempted to sidestep this issue by contending that it was really aiming at the uncertainty that investors suffered from not being able to determine whether an issuer was repurchasing to help the shareholders or to benefit management. The Fifth Circuit responded that “[t]olerance of uncertainty varies depending on considerations of likelihood and severity” so that “[i]f opportunistic or improperly motivated buybacks are not genuine problems, then there is no rational basis for investors to experience any of the uncertainty the SEC now claims warrants the rule.”
2. Proxy Statements
Exchange Act section 14(a) prohibits solicitation of proxies in violation of rules adopted by the SEC. Rule 14a-9 prohibits the inclusion of false material statements in proxy solicitations and omission of material facts necessary so that the statements in the solicitations not mislead.
Johnson Controls, Inc. (“Johnson”) reverse-merged into an acquisition subsidiary of Tyco International plc (“Tyco”) in a deal that offered Johnson shareholders either cash or shares in a surviving company called Johnson Controls International plc (“Johnson International”). The merging companies structured the transaction—in order to avoid an inversion tax—so that Johnson shareholders would own about 56 percent of Johnson International. After the merger agreement was signed, the U.S. Treasury proposed regulations that would eliminate the inversion tax benefit, and the final joint proxy statement/prospectus (the “proxy statement”) reported this development.
The Johnson board nevertheless recommended shareholder approval because of “‘global tax synergies’ and ‘operational synergies.’” The proxy statement (i) included the opinions of two financial advisers that Johnson had retained, both of which concluded that the merger consideration was fair as a financial matter to the Johnson shareholders and (ii) set out the advisers’ analysis and their numerical conclusions. The proxy statement disclosed that—whether a Johnson shareholder elected to take the Johnson International shares or to take the cash consideration—the transaction would be a taxable event for that shareholder.
After Johnson shareholders voted to proceed with the merger, plaintiffs—who, just before the vote, had sued Johnson, Tyco, the Tyco acquisition sub, and Johnson executives and directors—filed an amended complaint alleging the proxy materials violated Exchange Act section 14(a) and SEC Rule 14d-9. Affirming dismissal, the Seventh Circuit described “[t]he crux of plaintiffs’ argument [as] . . . that ‘there was another way to structure the merger that would have potentially avoided’ the taxation of Johnson shareholders and that ‘the omissions regarding such an option were material.’” The court of appeals found “nothing in the Exchange Act that entitles investors to receive a list of alternative deal options that may provide a better return on their investment.” On a practical note, the court added: “Were we to adopt plaintiffs’ position, we would be creating a new rule requiring proxy statements to include a laundry list of potential merger options and disclose the potential benefits and drawbacks of each option.”
The Seventh Circuit then considered the possibility that the board’s endorsement of the merger in the proxy materials constituted a misleading opinion under Omnicare, Inc. v. Laborers District Council Construction Industry Pension Fund. As the court of appeals saw it, Omnicare—read together with Virginia Bankshares—meant that “a defendant’s subjective disbelief or hidden motivation in a stated opinion is not enough to create liability. The defendant must also misrepresent, affirmatively or by omission, either: (1) the underlying facts used to form the opinion; or (2) the scope of inquiry made prior to rendering the opinion.” The court applied this test to the plaintiffs’ contention “that the merger undervalued the price of their shares and was not ‘fair’ to and in the ‘best interests’ of shareholders, and that the directors knew this but did not include it in the prospectus because they had undisclosed motivations.” The panel found this argument failed the legal test because “the proxy statement not only disclosed the terms of the merger in detail, it also disclosed the facts used to form directors’ opinions and the scope of inquiry they conducted.”
Significance and analysis. Oddly, the court concluded its analysis of the plaintiffs’ case with the holding that discussion of an alternative tax-free transaction was not material. Perhaps such a discussion would be immaterial depending on the amount of tax that shareholders would have to pay, but unless that amount was insignificant, a reasonable investor could very well consider a tax-free alternative to be important when deciding whether to vote for a deal providing taxable consideration.
Two analytical avenues might have been better. First, the panel could have leaned on the general principle that there is no liability for failure to disclose even a material fact absent a duty to disclose it. While Rule 14a-9(b) requires inclusion of “any material fact necessary in order to make the statements [in the proxy solicitation] not false or misleading,” the Seventh Circuit might have reasoned that a proxy statement for a merger vote does not mislead—even as to a conclusion that the deal being offered is “fair” or in the “best interests” of shareholders—by failing to include a discussion of differently structured transactions because the proxy statement is, by its context, a document about the particular deal subject to the vote, not all other possible deals. Indeed, the court took a step in that direction by holding that the Exchange Act does not, by its terms, require disclosure of “alternative deal options.” The panel might have done well to leave it at that or to hold that the only alternative deals subject to Rule 14a-9(a) scrutiny are ones that the proxy statement itself discusses and identifies as in immediate competition with the one the board selected.
Second, since the language of Securities Act section 11 mimics that of Exchange Act Rule 14a-9, the Seventh Circuit appropriately turned to Omnicare in the Johnson proxy statement case. However, when attempting to fit Omnicare into its analysis, the panel might have done well to recall that Omnicare focused on how an opinion could be false or misleading, not how to determine whether a false or misleading opinion was material. True enough, whether an omission from an opinion misleads can depend on materiality, but the Court defined the issue in Omnicare as application of Security Act section 11’s “false-statement provision . . . to expressions of opinions.” The Court expressly noted that materiality, while related to whether an opinion was misleading, is a separate element.
3. Forum-Selection Clauses
The Gap, Inc.’s (“Gap”) bylaws provide that “[u]nless the Corporation consents in writing to the selection of an alternative forum, the Court of Chancery of the State of Delaware shall be the sole and exclusive forum for . . . any derivative action or proceeding brought on behalf of the Corporation.” Gap is a Delaware corporation.
A shareholder filed a derivative action against Gap in federal district court in California, alleging that the company and its directors had violated Exchange Act section 14(a) and SEC Rule 14a-9 by falsely stating in the company’s 2019 and 2020 proxy statements that Gap was committed to diversity and inclusion even though Gap discriminated in hiring and compensation, and high-level positions were not filled in a diverse way. Relying on Gap’s forum-selection clause, the district court granted Gap’s motion to dismiss based on forum non conveniens. In a six-to-five en banc decision, the Ninth Circuit affirmed, rejecting the three arguments the shareholder made against enforcement of the forum selection provision in this case.
First, the shareholder pointed to Exchange Act section 29(a), providing that “‘[a]ny condition, stipulation, or provision binding any person to waive compliance with any provision of [the act] or of any rule or regulation thereunder, . . . shall be void.’” She argued that the bylaw required her to waive Exchange Act section 27(a), which gives federal courts “exclusive jurisdiction of violations of ” that act or the rules adopted under it. The en banc majority responded that “§ 29(a) forbids only the ‘waiver of the substantive obligations imposed by the Exchange Act,’ not the waiver of a particular procedure for enforcing such duties.” The substantive right here was protection against false and misleading statements in proxy statements. Ultimately, this protects shareholders’ rights to a fully informed vote on the corporate matter addressed in the proxy statement. Under Delaware decisional law, the shareholder could bring her section 14(a)/Rule 14a-9 claim as a direct action because “[she] and other shareholders suffered the alleged harm—a proxy nondisclosure injury in violation of § 14(a) that interfered with their voting rights and choices—and would receive the benefit of the remedy—the equitable or injunctive relief sought in the complaint.” Since the bylaw did not “impose any limitation on direct actions,” she could bring that action in federal court. The bylaw therefore did not affect the shareholder’s right to sue in a federal court for violation of section 14(a) and Rule 14a-9 but only the form of that suit—direct versus derivative—and Exchange Act section 29(a)’s antiwaiver prohibition accordingly did not void Gap’s forum-selection clause.
Second, the shareholder contended that the “forum-selection clause cannot be enforced under the doctrine of forum non conveniens because doing so would violate the federal forum’s strong public policy of allowing a shareholder to bring a § 14(a) derivative action.” Structuring the analysis around the general enforceability of forum-selection clauses in federal courts, the majority recognized “a narrow exception to this general rule if the plaintiff can demonstrate ‘extraordinary circumstances unrelated to the convenience of the parties [that] clearly disfavor a transfer.’”
The shareholder asserted that the extraordinary circumstance here consisted of the “high importance” “that Congress placed . . . on corporate compliance with the Exchange Act,” as evidenced by the section 29(a) prohibition on waivers and the exclusive federal jurisdiction conferred by section 27(a). She further contended that the Supreme Court decision in J. I. Case Co. v. Borak—which found an implied right for shareholders to sue for violations of section 14(a)—“reflects a strong public policy to give shareholders a right to bring both a direct and a derivative action to enforce § 14(a). [She] conclude[d] by asserting that enforcing Gap’s forum-selection clause would contravene this policy.”
The Ninth Circuit en banc majority read Borak as “holding that a shareholder had the right to bring a direct action under § 14(a)” and “append[ing] a less well-reasoned statement that a shareholder could also bring a derivative § 14(a) action.” The majority ruled that the latter statement was “dicta.” Moreover, since Borak, Delaware courts had held that “an action asserting . . . ‘a duty of disclosure violation [that] impaired the stockholders’ right to cast an informed vote’” is a direct action, rather than a derivative one, thus “undermin[ing the plaintiff ’s] claim that there is a strong public policy of the federal forum to give shareholders a derivative § 14(a) action in this context.” Further, the majority interpreted Virginia Bankshares (decided more than two decades after Borak) as suggesting that—under federal law—a shareholder could not bring a derivative section 14(a) claim because “a person whose vote is not ‘legally required to authorize the [corporate] action proposed’ lacks standing to bring a § 14(a) claim,” and the corporation itself does not vote. Hence, since relevant state and federal law have evolved since Borak toward the position that a shareholder has no section 14(a) derivative claim, “there is no . . . public policy supporting a right to bring such actions derivatively” and “Borak does not help [the plaintiff] make a strong showing that enforcement of Gap’s forum-selection clause ‘would contravene a strong public policy’ of the federal forum.”
The shareholder’s contention that a second “extraordinary circumstance” weighed against the general accommodation of forum-selection clauses rested on the notion that the Gap clause “conflicts with the federal forum’s strong public policy of giving federal courts exclusive jurisdiction over Exchange Act claims under § 27(a).” The en banc majority responded that Supreme Court precedent holds “that there was ‘no specific purpose on the part of Congress in enacting § 27.’”
Third, the plaintiff contended that the Gap bylaw forum selection was “invalid as a matter of Delaware law under Section 115 of the DGCL [Delaware General Corporation Law].” That statute provides that bylaws “may require, consistent with applicable jurisdictional requirements, that any or all internal corporate claims shall be brought solely and exclusively in any or all of the courts in this State.” A legislative synopsis included the statement that section 115 was “‘not intended to authorize a provision that purports to foreclose suit in a federal court based on federal jurisdiction.’” The Seventh Circuit relied on this synopsis in 2022 to hold that a forum-selection clause in Boeing’s bylaws was unenforceable to preclude a section 14(a) suit because “applying the bylaw to this case would mean that plaintiff ’s derivative Section 14(a) action may not be heard in any forum.”
The Ninth Circuit en banc Gap majority, however, interpreted section 115 in light of DGCL section 109(b), which gives bylaws a wide scope by permitting them to “contain any provision, not inconsistent with law or with the certificate of incorporation, relating to the business of the corporation, the conduct of its affairs, and its rights or powers or the rights or powers of its stockholders, directors, officers or employees.” The majority concluded that the Seventh Circuit “failed to recognize [Delaware court] interpretation of Section 115 and Section 109(b) as being permissive statutes, rather than restrictive statutes defining ‘the outer limit of what is permissible’ or otherwise precluding federal claims.” Moreover, the synopsis “means only that Section 115 does not create a legislative safe-harbor for forum-selection clauses that requires claims to be brought in forums that lack jurisdiction over them.” The statute and the synopsis “neither authorize[d] nor prohibit[ed] a forum-selection clause that would preclude bringing an action in federal court.”
With this analytical framework and taking into account Delaware decisions that claims like the plaintiff ’s were direct rather than derivative and accordingly not within the “internal corporate claims” to which section 115 and its synopsis referred, Gap’s forum-selection clause fell within the broad range of bylaws permitted by section 109(b).
Significance and analysis. The Ninth Circuit en banc majority recognized that its decision conflicts with the Seventh Circuit decision. The Supreme Court may have to decide.
4. Materially False or Misleading Statements
Multiple federal securities statutes impose liability in private lawsuits for false statements of material facts or statements that mislead for failure to include material facts. Last year, the Ninth Circuit reversed dismissal of Rule 10b-5 claims where the plaintiff (i) claimed that company statements misreported the segments into which sales were placed and (ii) based the charge that the segment allocation misled on a model created by a financial expert. That circuit also reversed dismissal to the extent that a Rule 10b-5 claim alleged that risk factors warned of adversity that the company had already experienced. The Second Circuit affirmed dismissal of Rule 10b-5 claims but reversed as to some section 11 and section 12(a)(2) claims in an action based on restatement of financial numbers due to improper timing of revenue from service contracts and improper timing of charges for bonuses.
Allocation of sales between reported segments. NVIDIA Corporation (“NVIDIA”) makes graphics processing units (“GPUs”). While its GPUs are predominantly used in playing video games, they can also be used for crypto mining. NVIDIA shareholders brought a Rule 10b-5 action against the company, the CEO, and the CFO for statements during the period May 10, 2017, through November 14, 2018 (“the class period”) that misled investors as to the proportion of the company’s revenues derived from GPUs used for crypto mining. When the profitability of crypto mining declined, NVIDIA unexpectedly announced on November 15, 2018, that it had missed its quarterly revenue target and reduced its revenue projection for the next quarter, with the result that NVIDIA’s stock price dropped by 28.5 percent in two trading days.
Throughout the class period, NVIDIA produced two kinds of GPUs relevant to this case: GeForce GPUs and Crypto SKUs. Crypto SKUs were specifically designed for crypto mining, and were essentially GeForce GPUs without the video functionality for gaming. Although the Crypto SKUs had a dedicated purpose, both kinds of GPUs could be used for crypto mining and crypto miners bought both.
NVIDIA divided its sales by market segments, two of which were important here. NVIDIA reported revenue from GeForce GPU sales in its Gaming segment. The company reported revenue from Crypto SKU sales in its OEM segment.
The plaintiffs did not allege that any internal NVIDIA report showed the total revenue that the company garnered from all sales of both GPUs used for crypto mining. Instead, plaintiffs relied on two sources that purported to derive that number and contrast it with the revenue NVIDIA reported from sales of Crypto SKUs through NVIDIA’s OEM segment. First, the complaint pled that an RBC Capital Markets (“RBC”) report concluded that, from February 2017 to July 2018, NVIDIA’s total revenue from sales of GPUs used for crypto mining approximated $1.95 billion, as opposed to NVIDIA’s representation that the number was about $602 million because that was the number the company reported as sales in its OEM segment—an understatement of $1.35 billion over this eighteen-month period.
Second, the plaintiffs pointed to the report by a consulting firm that the plaintiffs had hired (Prysm Group (“Prysm”)), concluding that NVIDIA’s total crypto-related sales over five quarters (from and including the second quarter of fiscal year 2018 to and including the second quarter of fiscal year 2019) approximated $1.728 billion, whereas NVIDIA represented that the $602 million in Crypto SKU revenue from the OEM segment provided the crypto-related figure—an understatement of $1.126 billion.
Prysm had reached the $1.126 billion estimate by a complicated calculation based on gross industry statistics. Prysm (i) calculated the total increase in the computing power used by “the ‘three most popular GPU-mined cryptocurrencies during the Class Period’”; and (ii) divided that number by the computing power of a single GeForce GTX 1060 (NVIDIA’s least expensive GPU) to estimate that “a minimum of approximately 16.9 million GPU units would be required to make up for the difference in computing power during the Class Period.” Prysm then (iii) attributed 69 percent of those units to NVIDIA because that constituted NVIDIA’s share of the crypto mining GPU market, as derived from (a) a Jon Peddie Research study concluding that NVIDIA’s cryptocurrency-mining market share constituted 69.4 percent in August through October 2016 and 68.6 percent in November 2016 through January 2017, (b) the RBC study, and (c) an “internal NVIDIA study of market share in China.” After thus estimating the number of GPU units that NVIDIA sold to crypto miners, Prysm then multiplied that number times 67 percent of NVIDIA’s suggested retail price per GeForce GTX 1060 unit to find an estimated $1.728 billion in NVIDIA revenue from crypto-related sales over the period Prysm studied. Since NVIDIA reported only $602 million in Crypto SKU revenue through the company’s OEM segment during that time, the difference between the estimated $1.728 billion and the reported $602 million constituted $1.126 billion in crypto-related revenue not identified as such in NVIDIA’s financials.
The complaint bolstered these derived numbers by information from former NVIDIA employees, including three who said that crypto miners purchased large quantities of GeForce GPUs, with two of them referring to some months within the class period.
At bottom, the complaint alleged that the defendants hid NVIDIA’s dependence on GPU sales for crypto mining inside the GeForce GPU revenue numbers reported for the Gaming segment—deceiving investors into thinking that GPU sales for crypto mining were predominantly reported in the Crypto SKU revenue numbers for the OEM segment. As a result, investors could not appreciate that, if crypto-related GPU sales dropped (as they had at AMD when bitcoin prices collapsed in the second half of 2013), NVIDIA’s revenue could decline dramatically (as did AMD’s in 2013).
After holding that the plaintiffs had, in the way summarized above, adequately pled “a sufficient likelihood that a very substantial part of NVIDIA’s revenues during the Class Period came from sales of GeForce GPUs for crypto mining,” the Ninth Circuit panel majority found multiple statements by the NVIDIA CEO and CFO defendants that led “investors and analysts to believe that NVIDIA’s crypto-related revenues were much smaller than they actually were,” representing “that NVIDIA’s crypto-related revenues were either entirely or largely revenues from sales of Crypto SKUs, reported in the OEM segment,” and “fail[ing] to mention in their statements during the Class Period that the great majority of NVIDIA’s crypto-related revenues came from sales of GeForce GPUs, reported in the Gaming segment.”
Among the challenged statements made by the CEO were the following, together with the related allegations why they were false: In an August 10, 2017 second-quarter earnings call, he said “‘that the Company’s Crypto SKU accounted for just $150 million of second-quarter revenues, and that “we serve the vast . . . majority of the cryptocurrency demand out of that specialized product,”’” when in fact the company had also received an “additional $199 million in crypto-related revenues” included in reported Gaming revenues. The related Form 10-Q for the same quarter—signed by the CEO and CFO—represented that a 59 percent increase in GPU business year over year “‘was due primarily to increased revenue from sales of GeForce GPU products for gaming,’” but “failed to say on the form that about half of its Gaming-segment revenues . . . came from sales of GeForce GPUs to crypto miners rather than to gamers.” The CEO said, in an interview published on November 10, 2017, that “‘for NVIDIA, cryptocurrency [sales were] “small but not zero . . .” “[m]aybe $70 million”—the amount NVIDIA had attributed to the Crypto SKU the day before,’” failing to add that “during the quarter in question about $229 million of NVIDIA’s Gaming-segment revenues came from sales of its GeForce GPUs to crypto miners.” In an interview published in Barron’s on February 9, 2018, the CEO stated that “‘cryptocurrency represented a “small, overall” “part of our business this past quarter,”’” while actually “‘cryptocurrency-related revenues in fourth quarter fiscal 2018 comprised $541 million—nearly 20% of NVIDIA’s entire fourth quarter fiscal 2018 revenues across all business segments.’”
Included in the allegations of false or misleading statements by the CFO were these, together with the related allegations why they were false: During a September 6, 2017 technology conference, the CFO said that “we covered most of cryptocurrency with our cryptocards [Crypto SKUs] that we had developed,” failing to add that “in the fiscal quarter . . . end[ing] a week before, revenues from Crypto SKUs had been about $150 million, while revenues from GeForce GPUs sold to crypto miners had been about $199 million[] . . . ‘[with that $199 million] realized through the Gaming segment, not through the Crypto SKU.’” At another technology conference on November 29, 2017, when “‘asked about the impact of cryptocurrency-related demand on NVIDIA’s gaming revenues, [the CFO] stated that “there probably is some residual amount or small amount” but that “the majority does reside in terms of our overall crypto card [Crypto SKU], which is the size of about $150 million in Q2.”’” In fact, the complaint alleged, during the relevant quarter, “sales of GeForce GPUs to crypto miners, reported in the Gaming segment, far exceeded sales of Crypto SKUs” and were not a “small amount” but $199 million residing inside the Gaming-segment revenues.
To support the conclusion that the complaint pled the deception material, the panel majority noted the allegations that analysts and the financial press “believed during the Class Period” “[b]ased on statements by [the CEO and CFO]” “that NVIDIA was not vulnerable to the vicissitudes of crypto mining” and were “surprised when, during the earnings call on November 15, 2018, NVIDIA disclosed the degree to which its revenues during the Class Period had depended on sales to crypto miners.”
Turning to scienter, the majority found the plaintiffs pled sufficient facts to infer that the CEO made his false and misleading statements, “about the degree to which NVIDIA’s revenues were dependent on sales of GeForce GPUs to crypto miners,” either knowingly or recklessly. One former employee allegedly said that NVIDIA “‘kept meticulous track of who was buying its GPUs—not simply directly from the Company, but also from its partners and others down the distribution chain as well [and that the CEO and CFO] . . . had actual access to this data.’” He said that vice presidents met with the CEO every quarter to review company performance and that he learned from another employee that vice presidents “‘presented sales data reflecting GeForce sales to miners at the quarterly meetings with [the CEO] in 2017.’” Another former employee portrayed the CEO as “‘the most intimately involved CEO he had ever experienced.’” That same former employee stated that he had personally attended some quarterly meetings at NVIDIA headquarters at which the CEO was present and “‘recalled that when [the CEO] stated that miners were buying GeForce GPUs instead of the professional cards, the information came as no surprise’” to the other attending executives. The panel majority added that the CEO “showed himself to be familiar with specific revenue numbers attributable to particular categories of sales” “during earnings calls and in interviews with analysts.”
As the panel summarized, such facts alleged that “(1) [the CEO] had detailed sale reports prepared for him; (2) [he] had access to detailed data on both crypto demand and usage of NVIDIA’s products; (3) [he] was a meticulous manager who closely monitored sales data; and (4) sales data at the time would have shown that a large portion of GPU sales were being used for crypto mining.” The CEO’s “access and review of contemporaneous reports are the most direct way to prove scienter” and “[he had] admitted to closely monitoring sales data.” Holistically: “We state the obvious. A CEO who does not know the source of $1.126 billion in company revenues during [a] fifteen-month period, or $1.35 billion during an eighteen-month period, is unlikely to exist. Or if such a CEO does exist, he or she is not likely to remain CEO for very long.” All in all, “these allegations support a strong inference that [the NVIDIA CEO] reviewed sales data showing that a large share of NVIDIA’s GeForce GPUs sold during the Class Period were being used for crypto mining.”
Moving to the CFO, the panel majority found the “only concrete allegation” to be that she “had access to contradictory information during the Class Period” because of her ability to enter “NVIDIA’s centralized sales database,” and she “‘could direct VPs . . . to forward the data’ to her.” This was “insufficient to establish a strong inference that [she] personally accessed contradictory information during the Class Period.”
Accordingly, the panel reversed the lower court’s dismissal of the claim against the CEO and affirmed as to the CFO.
Significance and analysis. Permitting a plaintiff to successfully plead falsity and even scienter by hiring an expert to conduct a post hoc analysis estimating a market and a company’s share of that market, then backing those numbers into segment analysis within the company—with all the assumptions and inference that process includes—arguably erodes too greatly court-created prohibitions against piling inference upon inference. True, the purchased analysis was “supported” by an analyst report. But the pled description of that report did not provide the kind of detail permitting a judgment on its validity. The Supreme Court has granted certiorari and may well address this very issue.
The wide permission to employ inference that the majority opinion granted the plaintiff may have bled over to the court itself. To posit it is “obvious” the CEO must have known that so many of the GeForce GPUs were being used for crypto mining that OEM segment sales figures for Crypto SKUs chip provided a misleadingly reduced impression of NVIDIA’s total sales for crypto mining (even in light of executive comments that there was some leakage between segments) is to state a personal opinion. It is unclear whether judges, who may have had little personal experience in large and complicated corporations with segmented reporting numbers, should rely on such opinions in their decisions.
On a legal basis, the court’s scienter analysis in this regard simulates the “core operations” theory of scienter pleading—that executives know about key drivers in the parts of their companies that dominate the business overall. Perhaps the panel would have done well to acknowledge the restrictions that appellate decisions have imposed on use of the “core operations” assumption and to analyze whether the facts here fell within those limitations. Such an analysis might have discouraged a quick conclusion that the NVIDIA CEO should be charged with knowledge of a percentage difference in intersegment allocation of sales revenue in order to find a “strong inference” of intentional or reckless fraud.
Risk factors interpreted as statements that risks had not materialized. Facebook permits users to designate other users as “friends,” who share material that the users post. Users who enjoy a post by another Facebook user can “like” that post by clicking the “Like” button beneath the post.
Facebook CEO Mark Zuckerberg publicly stated in 2014 “that Facebook would no longer allow third parties to access and collect data from users’ friends.” In that same year, Zuckerberg and Facebook chief operating officer (“COO”) Sheryl Sandberg “created a ‘reciprocity’ system in which certain third-party apps that provided ‘reciprocal value to Facebook’ could be ‘whitelisted,’ meaning that those apps were exempt from the ban on third-party data access and collection.”
In September 2015, Facebook personnel “noticed that Cambridge Analytica [‘CA’] was ‘receiving vast amounts of Facebook user data,’” investigated, and “concluded that it was unlikely [CA] could use Facebook users’ data for political purposes without violating Facebook’s policies.” Facebook hired Aleksandr Kogan, who had assisted CA, “to give an internal presentation on the lessons he learned from collecting and working with the Facebook data.”
In December 2015, The Guardian reported that CA had generated a database containing information on U.S. voters via a personality quiz that Kogan had created and that was offered to Facebook users through a process that gave Kogan access not only to those users’ data but also obtained “data from their Facebook friends who had not taken the quiz”—including “name, gender, location, birthdate, ‘likes,’ and [a] list of [their own] Facebook friends.” While only some 250,000 Facebook users took the quiz, Kogan used this process to “harvest[] data from over thirty million users, most of whom did not consent to the data collection.” Kogan used this information to classify voters “on five personality traits: ‘openness to experience, conscientiousness, extraversion, agreeableness, and neuroticism (the “OCEAN scale”).’” The Guardian said that CA had used the data to help Ted Cruz in his primary challenge to Donald Trump. Facebook responded through a “spokesperson [who] stated that the company was ‘carefully investigating’ the situation, that misusing user data was a violation of Facebook’s policies, and that the company would ‘take swift action’ against third parties found to have misused Facebook users’ data.”
Still in December 2015, “a Facebook executive,” through “a private email exchange . . . told a [CA] executive that [CA] violated Facebook’s policies and terms by using data that Kogan ‘improperly derived’ from Facebook.” In January, CA agreed to delete “the personality score data harvested from Facebook.” Remaining concerned, Facebook continued to investigate and, in June 2016, obtained an agreement from Kogan to certify “that he had deleted the data in his possession derived from Facebook ‘likes’” and to identify “every entity with which he had shared raw Facebook user data.” Kogan revealed that he had shared both raw and derivative information with CA CEO Alexander Nix, who then—in response to a Facebook request—refused to certify that all the data from the personality quiz had been deleted.
The Washington Post reported in October 2016 that CA was still using the OCEAN scale data, this time to help Trump in his campaign in the general election. While the report did not say that the underlying data were derived from Facebook, “the use of the OCEAN scale suggested that [CA] may have been using the data originally harvested from Kogan’s personality quiz” on that social media platform.
A Guardian article in March 2017 reported CA use of data originating from Facebook and “quoted a [CA] spokesperson’s denial that the firm had access to Facebook ‘likes.’” The article also “quoted a Facebook spokesperson’s statement that Facebook’s investigation into [CA] had not yet uncovered any misconduct related to the firm’s work on political matters, specifically the Trump presidential campaign or the Brexit Leave campaign.” “A Facebook spokesperson made similar comments to journalists later that month.”
On March 12, 2018, both The New York Times and The Guardian advised Facebook that they were going to report that CA had not deleted improperly collected Facebook data. Before those articles appeared in print, Facebook “announced on its investor relations website that it was suspending [CA] for violating its policies by sharing Facebook users’ data without the users’ consent and for failing to delete” that data. The Facebook statement said the company had now learned that, contrary to the certification from Kogan, not all of the data had been erased. The announcement garnered widespread media coverage, calls for government investigations, and criticism that Facebook either knew about the misuse of user information and did not tell the public earlier or operated systems so deficient that the company did not know of the misuse until shortly before the March 2018 announcement. In the week following the investor relations post, Facebook stock declined by almost 18 percent.
In early June 2018, a New York Times article disclosed the “whitelisting” that Facebook had begun in 2014, with the Times saying that user and friends data had been shared “with dozens of whitelisted third parties like Apple, Microsoft, and Samsung without the users’ express consent.” The article stated that this practice was contrary to company statements that users controlled the distribution of their information and violated a 2012 FTC order to which Facebook had assented.
On July 25, 2018, Facebook announced disappointing second quarter revenue, profit, and user growth, blaming the results on both the company “‘putting privacy first,’” and implementation of the European Union’s General Data Protection Regulation (“GDPR”), which Zuckerberg said had reduced the number of European Facebook users. The next day, Facebook’s stock price dropped almost 19 percent, with market participants and commentators attributing the financial results to a combination of factors, including the CA and whitelisting revelations.
Investors brought a Rule 10b-5 action against Facebook, Zuckerberg, Sandberg, and CFO David Wehner. After the district court dismissed the complaint, the Ninth Circuit affirmed in part and reversed in part. The court of appeals separated the alleged misrepresentations into three categories: (i) risk statements, (ii) statements about CA, and (iii) statements that Facebook users controlled access to the information they posted on the platform. The class period during which the defendants made the challenged statements ran from February 3, 2017, through July 25, 2018.
The “risk” statements consisted of Risk Factors in Facebook’s Form 10-K, filed in February 2017 for the year ended December 31, 2016. The critical passages (with subject headings in bold) were:
If we fail to retain existing users or add new users, or if our users decrease their level of engagement with our products, our revenue, financial results, and business may be significantly harmed. . . . If people do not perceive our products to be useful, reliable, and trustworthy, we may not be able to attract or retain users or otherwise maintain or increase the frequency and duration of their engagement.
. . . .
Security breaches and improper access to or disclosure of our data or user data, or other hacking and phishing attacks on our systems, could harm our reputation and adversely affect our business. . . . Any failure to prevent or mitigate security breaches and improper access to or disclosure of our data or user data could result in the loss or misuse of such data, which could harm our business and reputation and diminish our competitive position. In addition, computer malware, viruses, social engineering (predominantly spear phishing attacks), and general hacking have become more prevalent in our industry, have occurred on our systems in the past, and will occur on our systems in the future. . . . Although we have developed systems and processes that are designed to protect our data and user data, to prevent data loss, and to prevent or detect security breaches, we cannot assure you that such measures will provide absolute security.
The district court concluded that the complaint failed to plead these statements false. The Ninth Circuit, however, applied the principle that warning a risk could occur is misleading when the risk has already materialized. Applying this standard, the court of appeals concluded that the lower court “correctly dismissed the challenged statements regarding the risk of security breaches and the risk of the public not perceiving Facebook’s products to be ‘useful, reliable, and trustworthy’” because “[t]hose statements do not relate to the misuse of Facebook user data by [CA], and the shareholders do not allege that those risks had materialized at the time of the 2016 10-K.”
But as to the other risk warnings, the Ninth Circuit panel majority found “that Facebook represented the risk of improper access to or disclosure of Facebook user data as purely hypothetical when that exact risk had already transpired.” The majority concluded that “[a] reasonable investor reading the 10-K would have understood the risk of a third party accessing and utilizing Facebook user data improperly to be merely conjectural.” And the complaint adequately alleged that was not true when the company filed the Form 10-K in February 2017 because (i) Facebook identified CA as potentially violating Facebook privacy policies in September 2015, (ii) Kogan made a presentation to the company in November 2015 describing what he had learned from Facebook data, (iii) a Facebook executive told CA in December 2015 that CA was violating Facebook policies, and (iv) Facebook understood in June 2016 that not all data Kogan used had been deleted and that the CA CEO refused to certify actual deletion. The panel majority found those “allegations, if true, more than support the claim that Facebook was aware of [CA’s] misconduct before February 2017, so Facebook’s statements about risk management ‘directly contradict[ed]’ what the company knew when it filed its 2016 10-K with the SEC.”
It did not matter to the two-judge panel majority that Facebook may not have known the extent of the harm CA’s conduct would wreak on Facebook; “[b]ecause Facebook presented the prospect of a breach as purely hypothetical when it had already occurred, such a statement could be misleading even if the magnitude of the ensuing harm was still unknown.” It was “the fact of the breach itself, rather than the anticipation of reputational or financial harm, that caused [the] anticipatory statements to be materially misleading.” And, while the Facebook warnings included that “‘computer malware, viruses, social engineering . . . , and general hacking have become more prevalent in our industry, have occurred on our systems in the past, and will occur on our systems in the future,’” “[c]ollapsing the risks of improper access to and use of Facebook users’ data in the same section as the risk of cyberattacks cannot rescue the risk statements from being false or materially misleading.”
Turning next to Facebook statements about its internal investigation of CA collection of user and friend data, the Ninth Circuit majority focused on the unnamed company spokesperson who told the press in March 2017 that Facebook “had ‘not uncovered anything that suggest[ed] wrongdoing’ related to [CA’s] work on the Brexit and Trump campaigns.” Agreeing with the district court’s conclusion that the complaint did not adequately plead scienter for this representation, the majority found that the plaintiffs “pleaded only that the Facebook spokesperson should have known that Facebook’s investigation into [CA] had uncovered misconduct, not that the spokesperson actually knew of any misconduct.”
The last category of challenged statements—(i) saying “that Facebook’s priorities of transparency and user control aligned with the GDPR framework” and (ii) “assur[ing] Facebook users that they had control over their information and content on Facebook”—included comments such as: “‘People can control the audience for their posts and the apps that can receive their data,’ ‘[e]very person gets to control who gets to see their content,’ and ‘[w]e respected the privacy settings that people had in place.’”
The panel majority agreed with the district court that the complaint did not allege that the “priorities” statements violated Rule 10b-5 because “those . . . ‘merely reiterated Facebook’s ongoing commitment to “transparency and control”’ rather than assuring users they controlled their Facebook data, and thus were not false when they were made.” But as to the assurances of user control, the court of appeals reversed dismissal, focusing not on falsity (which the opinion seems to assume) but loss causation—i.e., whether the complaint sufficiently alleged that corrective disclosures about users’ lack of control over their information caused the 18 percent decline in Facebook stock price in the week following the March 2018 investor relations announcement, and the 19 percent drop on the day after the July 25, 2018 report of disappointing second quarter results.
As to March, the Ninth Circuit concluded the plaintiffs sufficiently alleged that Facebook’s March 2018 announcement on its investor relations website—that Facebook was suspending CA from the platform because CA had violated Facebook policies by collecting, sharing, and failing to delete Facebook user data—“was the first time Facebook investors were alerted that Facebook users did not have complete control over their own data.” The 2015 and 2016 newspaper articles about CA “did not reveal that [CA] had misused Facebook users’ data,” Facebook’s response that it was investigating clouded the issue, and Facebook did not publicly acknowledge improper CA data use until its March 2018 revelation. The quick and steep decline of Facebook’s stock price after that acknowledgment “plausibly caus[ed] economic loss for the shareholders.”
The July stock drop was more complicated. After the March disclosure, The New York Times publicly disclosed on June 3, 2018, that Facebook was sharing user data with third parties via “whitelisting.” The Ninth Circuit nevertheless held that the complaint adequately alleged that both “the [CA] and whitelisting revelations, not any other factor, caused the July 2018 stock price drop.” Although the July 25 disclosure was simply that Facebook had had lamentable second quarter revenue, profit, and user growth and while the immediate stock drop occurred the next day—and although, by that time, some four months had elapsed since the March acknowledgment of CA’s misuse of user data and some eight weeks since the Times article in early June revealing whitelisting—the court of appeals ruled that the plaintiffs “plausibly plead ‘a causal relationship’ between the [CA] and whitelisting revelations and the dramatic drop in Facebook’s stock price.” More specifically, the temporal separation between the corrective disclosures and the price drop was plausibly overcome by the notion that the July 25 earnings announcement provided “new information to the market” that “allowed the public to ‘appreciate [the] significance’ of the [CA] and whitelisting scandals.”
Significance and analysis. Item 105 of Regulation S-K requires that an issuer “provide under the caption ‘Risk Factors’ a discussion of the material factors that make an investment in the registrant or offering speculative or risky.” Item 1A of Form 10-K requires that an issuer include such risk factors in that annual report. Item 1A of Form 10-Q requires that the issuer set forth in those quarterly reports “any material changes from risk factors as previously disclosed in the [company’s] Form 10-K.”
Facebook holds that an issuer including a risk factor in periodic filings incurs Rule 10b-5 liability if the risk it identifies has already “materialized,” and that such liability can be found “even if the magnitude of the ensuing harm [is] still unknown” because “[t]he mere fact that Facebook did not know whether its reputation was already harmed when filing the Form 10-K does not avoid the reality that it ‘create[d] an impression of a state of affairs that differ[ed] in a material way from the one that actually exist[ed].’” With respect, that reasoning is hard to square with either law or context.
As for law, Rule 10b-5(b) prohibits only misstatement of “material” facts. The circumstance that a risk has “materialized” does not mean that it is material. Probably the best way to determine whether the facts—here CA collecting and using Facebook users’ and friends’ data—were material for securities law purposes at the time Facebook filed its Form 10-Q in February 2017 would have been to apply the probability/magnitude test that the Supreme Court set out in Basic Inc. v. Levinson for analyzing the materiality of an event (there a particular step toward a merger) by considering both how the event affects the probability of future developments (such as a merger) and the magnitude of those developments should they occur (such as the importance of the merger). Here, the materiality question required weighing both the probability that CA’s unauthorized collection and analysis of user information would cause financial harm to Facebook and the magnitude of that harm—all based on the particular information that Facebook had at the time it filed the Form 10-K.
But the Ninth Circuit conducted no such analysis. This is not to say a probability/magnitude analysis would have determined that CA’s acquisition and use of Facebook data was immaterial when Facebook filed its 10-K. But a proper explanation of the court’s ruling, and guidance to companies going forward, arguably required that the court include this step.
As for context, it is strange—when considering an ongoing risk—to conclude that saying a risk could materialize in the future would lead a “reasonable investor” to conclude that it had not occurred already. The future risk of unauthorized acquisition and use of Facebook user data was always present. That was true regardless of whether an unauthorized use had occurred in the past or not—and that future risk was, by its terms, all the risk factors purported to concern. The notion that “a reasonable investor” would have concluded that risk factors affecting future operations were actually a guarantee of no risk at the time the company published the factors seems strained. The Supreme Court has granted review to consider this issue.
Accounting errors. AmTrust Financial Services, Inc. (“AmTrust”) sold retailers extended service plans (“ESPs”) to (i) cover extended warranty claims, and (ii) provide administrative services, including call center services, to deal with claims. In February and March 2017, AmTrust disclosed that, due to accounting errors, it would delay filing its Form 10-K for 2016. When the company filed that Form 10-K, it included restated financials for the full years 2012 through 2016, as well as each quarter during 2015 and 2016.
Two accounting issues caused the restatement. First, AmTrust had recognized revenue for administrative services in ESPs when the contracts began instead of recognizing that revenue over the life of the contracts, and the restatement revised that protocol to recognize that revenue “‘on a straight-line basis over the term of the ESP contracts.’” Second, AmTrust had delayed expensing discretionary employee bonuses until the company paid them, and the restatement revised that protocol to expense bonuses when they were earned.
After AmTrust disclosed the accounting errors, investors sued (i) under Securities Act sections 11 and 12(a)(2) based on November 2015 and September 2016 AmTrust stock offerings sold by documents that incorporated periodic filings containing financials that were later restated, and (ii) under Rule 10b-5—with AmTrust and officers or director defendants on all counts, underwriters also defendants on the claims arising out of the offerings, and AmTrust’s outside auditor an additional defendant on the section 11 and Rule 10b-5 claims. The district court dismissed the case in whole, and the Second Circuit affirmed in part, but vacated the dismissal in part as to the section 11 and section 12(a)(2) Securities Act claims against AmTrust, its officers and directors, and the underwriters.
The lower court based its dismissal on the ground that “for the most part . . . the alleged misstatements were nonactionable statements of opinion.” This prompted a lengthy Second Circuit discussion of whether the challenged accounting judgments were “subjective” opinions or “objective fact[s].” The court of appeals concluded that, in light of Omnicare, Inc. v. Laborers District Council Construction Industry Pension Fund, (i) “AmTrust’s financial statements relating to the warranty contract revenue reported in its historical consolidated financial statements were actionable statements of opinion under Section 11”; and (ii) it did “not need to decide whether these financial statements [were] statements of fact or . . . statements of opinion” insofar as they included the bonus charges because they would be actionable in either case.
Under Omnicare, an opinion is false if (i) the speaker or author does not believe the opinion when speaking or writing it; (ii) the opinion contains an embedded fact that is false; or (iii) the opinion omits information drawing the opinion into question (such as the inadequacy of the investigation leading to the formation of the opinion or some fact known to the speaker or author cutting against the opinion) that, under the circumstances, a reasonable investor would expect to be disclosed. Here, the accounting rule applicable to the recognition of revenue for ESP administration provided: “[R]evenue from separately priced extended warranty or product maintenance contracts shall be deferred and recognized in income on a straight-line basis over the contract period except in those circumstances in which sufficient historical evidence indicates that the costs of performing services under the contract are incurred on other than a straight-line basis.” The complaint alleged that AmTrust had “no historical evidence to support its [other-than-straight-line] approach,” and therefore under Omnicare “AmTrust’s representations about the warranty contract revenue reported in its historical consolidated financial statements misled investors to conclude that the company was aware of some historical evidence in support of recognizing the revenue on a non-straight-line basis, when in (alleged) fact it was not.” This constituted “‘saying one thing and holding back another’” and accordingly constituted a misleading opinion under the third alternative that Omnicare identified.
The accounting principle determining when AmTrust should have charged the discretionary bonuses as expenses provided that such a charge should be made “when it is ‘probable’ that a liability has been incurred and when ‘[t]he amount of [the liability] . . . can be reasonably estimated’ within a range.” As the Second Circuit understood it, the complaint argued that “AmTrust had a practice of paying bonuses [and] . . . there was no basis to conclude that the continued payment of earned bonuses was not ‘probable.’” The complaint therefore alleged that “there was no basis to conclude that the continued payment of earned bonuses was not ‘probable’ and that such bonuses therefore could not be expensed when earned.” As the company later acknowledged, “‘even though the bonuses were discretionary, the bonuses should have been estimated and expenses assigned to interim periods so that the interim periods [bore] a reasonable portion of the anticipated annual amount.’” Citing again the portion of Omnicare discussing the third way in which an opinion can mislead, the Second Circuit concluded that “because the Complaint adequately alleges that it was improbable that the earned bonuses would not be paid[,] . . . it [is] quite plausible that the AmTrust Defendants did not base the company’s statements of probability on a ‘meaningful . . . inquiry,’ that their statements did not ‘fairly align[] with the information in the issuer’s possession at the time,’ and that there was no basis for AmTrust to state that the bonuses should be expensed in the year they were paid rather than earned.”
After having in this way determined that the financials that AmTrust restated were misleading when included in the periodic filings incorporated by reference into the offering documents for the two AmTrust offerings and that the district court should not have dismissed the section 11 and section 12(a)(2) claims based upon the numbers in the financials, the Second Circuit turned to an additional section 11 claim: that the offering materials misled because the periodic reports they incorporated included misleading certifications by the CEO and CFO.
The court of appeals held that, insofar as the certifications related to the accuracy of the financial statements in the periodic reports, they were “opinions” expressly qualified by the statement that they were “based on [the] knowledge” of the signing officer. The court found, however, that the complaint failed to include any allegation that any of the certifications were “not based on the officer’s knowledge.” To the extent that the certifications addressed disclosure controls and internal control over financial reporting, they were also “opinions” that “convey[ed] management’s subjective judgments about the company’s internal controls.” The Second Circuit rejected the argument that, because AmTrust eventually changed its positions on the timing of ESP revenue recognition and discretionary bonus charges, the “certifications were not believed when made,” reasoning that a “change of opinion, standing alone, does not mean that the original certified opinions were disingenuous” and that “a genuinely held opinion that ‘turned out to be wrong’” is not “necessarily actionable.” While the plaintiffs argued that the opinions in the certifications misled by “falsely convey[ing] the existence of ‘some meaningful . . . inquiry’ conducted by the certifying executives,” the court of appeals found no facts pled in the complaint to “establish a lack of meaningful inquiry, other than the fact that the certification turned out to be wrong.” Accordingly, the panel affirmed the district court dismissal of the section 11 claims based on the certifications.
The panel similarly affirmed the lower court’s determination that the complaint failed to plead facts raising a strong inference that the AmTrust defendants made misleading statements with scienter (an element of the Rule 10b-5 claim) under either of the two alternatives that the Second Circuit favors. First, the plaintiffs failed to allege facts supporting both a motive to commit fraud and an opportunity to do so. While they “rel[ied] on the AmTrust Defendants’ financial incentives to keep share prices high and to fuel the company’s acquisition strategy,” such a “desire to sustain ‘the appearance of corporate profitability’ is not itself the kind of incentive or motivation that raises an inference of scienter.” Although the complaint also pointed to stock sales by the CFO and some other executives, it did not allege that either the CEO or the director defendants had sold and, in any event, the CFO’s sales “began several months before the AmTrust Class Period, a fact that renders his stock sales during this class period less unusual.” Second, the complaint failed to allege facts supporting conscious misbehavior or recklessness, as none of the allegations—not even the magnitude of the misstatements or the number of accounting periods they covered—provided the required strong inference of departure from standards that was so extreme that the defendants must have been aware that they would mislead investors.
Finally, as to scienter, the complaint pled that AmTrust had acquired Warrantech in 2010 and Warrantech had—in response to SEC pressure—changed its “time-of-sale” recognition of warranty-related revenue to straight-line recognition in 2006. This suggested to plaintiffs that AmTrust had known for years that it should not recognize all ESP revenue up front. But the panel concluded that, instead of manifesting fraud, “AmTrust’s . . . resort to a time-of-sale approach for the contracts [after buying Warrantech], though wrong, is more plausibly explained by the changes to the guiding accounting principles since 2006 to which AmTrust points us, or to AmTrust’s negligence.”
The Second Circuit, however, reversed the trial court’s dismissal of the Securities Act section 11 and section 12(a)(2) claims against the underwriters for the two offerings during the class period for the same reasons that it reversed those claims against the AmTrust defendants—the panel’s “disagree[ment] with the District Court’s conclusion that the reported income statements related to AmTrust’s warranty contracts and its employee bonuses were non-actionable opinions.” In doing so, the panel addressed a matter of first impression. Section 12 defines a proper plaintiff as “the person purchasing [the] security from [a seller.]” The Second Circuit held that, since the “seller” for section 12 purposes includes all who successfully solicited a plaintiff for the purchase, each purchaser in an offering could sue each underwriter in that offering because the complaint alleged that each class member purchased “‘pursuant to the Prospectuses.’”
Lastly, the court of appeals affirmed dismissal of the Rule 10b-5 claim against AmTrust’s outside auditor based on its audit reports for the years ending December 31, 2013, 2014, and 2015. The Second Circuit held that the plaintiffs “adequately alleged that BDO’s audit opinion contained potentially actionable misstatements of opinion because (i) the Complaint ‘render[s] it plausible that [the audit engagement partner],’ who signed the audit opinion, ‘disbelieved the statement that the audit was conducted in accordance with the relevant PCAOB standards’” and (ii) “BDO’s statement that it ‘believe[d] [its] audits provide a reasonable basis for [its] opinion,’ would lead a reasonable investor to conclude that BDO had conducted ‘some meaningful . . . inquiry,’ when in fact, according to the Complaint, BDO never conducted such an inquiry.” However, the Second Circuit held that the accounting firm’s statements in its audit letter “were ‘so general’ in this case ‘that a reasonable investor would not depend on [them] as a guarantee,’” and further that the complaint “in this case” “failed ‘to allege any facts relevant to the way or ways in which BDO’s failure to supervise, review, document, and perform in good faith the 2013 audit would have been significant to a reasonable investor in making investment decisions.’” The lower court had therefore properly dismissed the Rule 10b-5 claim against the auditor because the complaint failed to plead facts showing that the auditor’s challenged statements were material. The panel acknowledged, however, that it “might have come to a different conclusion had such facts been alleged.” Finally, the Second Circuit affirmed dismissal of the Securities Act section 11 claim against the auditor because the plaintiffs abandoned a challenge to that portion of the district court order on appeal.
Significance and analysis. The AmTrust opinion includes doctrinally questionable analysis. It flails about over “identif[ying a] statement as one of opinion rather than fact,” then turns to Omnicare to determine whether an opinion is “actionable.” The better analysis proceeds in two steps.
Multiple securities statutes and rules—including Rule 14a-9—prohibit false statements of material “fact” or statements that mislead by omitting a material “fact.” None of these employs the word “opinion.” In Virginia Bankshares, Inc. v. Sandberg, the Court held that opinions describing merger consideration as “high” value or “fair” were “facts” for purposes of Rule 14a-9 because they rested on “provable facts about the Bank’s assets, and about actual and potential levels of operation” that could “substantiate[] a value that was above, below, or more or less at the $42 figure [offered in the merger], when assessed in accordance with recognized methods of valuation.” Accordingly, the initial analysis in an opinion case—to determine whether the statement is “actionable”—should rest on similar analysis.
If, and only if, the opinion qualifies as a “fact” for purposes of the securities law should the analysis proceed to the second step. In that second step, the court should employ Omnicare to determine whether the opinion is false or misleading in any of the three ways that Justice Kagan identified there.
Perhaps even more important, the Second Circuit’s conclusion that the claim against the outside auditor could be dismissed at the pleading stage because no facts were specifically pled to show that investors would care whether the auditor violated auditing standards before issuing its opinion seems problematic. At a minimum, the court might have done better to order that the district court give plaintiffs an opportunity, if indeed this is necessary at all, to plead specific facts going to this materiality question.
5. Scienter and Scienter Pleading
The Exchange Act requires that a complaint in a private action seeking damages based on that statute must, “with respect to each act or omission alleged to violate [the act], state with particularity facts giving rise to a strong inference that the defendant acted with the required state of mind.” The required state of mind for a Rule 10b-5 violation is scienter, which means in a case in which the defendant is charged with making a false or misleading statement that the defendant must have made the statement either with an intent to deceive or with severe recklessness that, for this purpose, can be defined as making the statement in an extreme departure from ordinary care, presenting a danger of misleading investors of which the defendant was or must have been aware.
The Supreme Court elaborated this pleading standard by holding that a “strong inference” of scienter in a Rule 10b-5 case is one that is (i) “powerful or cogent”; (ii) “at least as compelling as any opposing inference one could draw from the facts alleged”; (iii) as determined by a process in which courts (a) “must consider plausible, nonculpable explanations for the defendant’s conduct, as well as inferences favoring the plaintiff ” and (b) consider not only individual facts bearing on the required state of mind but also “whether all of the facts alleged, taken collectively, give rise to a strong inference of scienter.”
Spirit AeroSystems, Inc. (“Spirit”) produced shipsets of components that Boeing used in constructing its 737 MAX airplanes. After two of those aircraft crashed, Boeing reduced its 737 MAX production per month from fifty-two to forty-two but continued to buy fifty-two shipsets per month from Spirit. In December 2019, Boeing announced that it was suspending production of this airplane altogether, and told Spirit to stop delivering shipsets. Spirit’s stock price dropped.
On January 30, 2020, Spirit disclosed that it had not properly accounted for contingent liabilities derived from customer complaints and, on February 28, 2020, the company said that the accounting errors traced to a material weakness in internal control over financial reporting relating to the “estimate at completion” (“EAC”) for these claims. Spirit’s stock price declined after each of these further disclosures.
Investors filed a Rule 10b-5 claim against Spirit, its CEO, its former CFO, its controller, and Spirit’s Vice President for the 737 NG and 737 MAX programs (“VP 737 MAX”) alleging that from October 31, 2019, through February 27, 2020, the defendants had made false and misleading statements about (i) Boeing’s continued purchases of shipsets for the MAX 737 and (ii) the reliability of the financial numbers Spirit published and the internal control systems generating those numbers. The district court dismissed the action, and the Tenth Circuit affirmed on the ground that the complaint failed to plead facts raising a strong inference that defendants made the challenged statements with scienter.
As to continued sales of the shipsets to Boeing after the two 737 MAX crashes, the complaint attacked (i) the Spirit CEO’s statement on October 31, 2019, “that Spirit would ‘be at 52 [shipsets of components produced per month] for an extended period of time’”; and (ii) the statement in an analyst report distributed on November 24, 2019 that “[t]he MAX is set to stay at a rate of 52/mo. until May 2020 w/ a potential rate decision at that time,” with the report saying that this information derived from a meeting with the Spirit CEO and the CFO.
The Tenth Circuit considered first whether the complaint pled the CEO’s scienter by allegations that (i) one former Spirit employee said that unidentified Boeing employees had told Boeing suppliers who told this informant employee that Boeing would cut 737 production and reduce orders for shipsets and (ii) a second former Spirit employee said that unidentified Boeing personnel told the Spirit VP 737 MAX and a second Spirit executive that Boeing planned to cut shipset purchases and that both of those Spirit executives told the second informant-employee. But neither of the former employees said he or she communicated this information to the CEO and neither claimed knowledge that anyone else did. Indeed, “the complaint [didn’t] allege that anyone at Spirit had informed [the CEO] about these conversations,” so these informant allegations could not show the CEO knew of the Boeing cuts at the time he made the supposedly false or misleading statements.
Aside from the unavailing allegations that the indirect information from unidentified sources inside Boeing found its way to the Spirit CEO, the plaintiffs also argued that the Spirit CEO knew of Spirit’s study of layoffs that could result from Boeing stopping 737 MAX shipset orders and that his knowledge of a resulting layoff projection showed the CEO’s scienter as to statements that those orders would continue at fifty-two per month. This, too, depended on information a former Spirit employee provided to plaintiffs—viz., that the informant and others had been told by a supervisor to gather data for such projection, that such a projection had been created, and that if the informant’s supervisor agreed with it, the projection would be passed to the Spirit CEO. But although the complaint included a “conclusory” allegation that the CEO actually saw the first-round projection, the plaintiffs pled no specific facts to show that the supervisor approved the projection, that it was ever delivered to the CEO, or that the CEO ever saw whatever input the informant provided for the projection. Moreover, the complaint failed to allege particularly what the final projection said—significant because “Spirit characterizes the final version as a compilation of various contingencies, including a drop in Boeing’s purchases” and “the plaintiffs have not questioned Spirit’s characterization.” Knowledge of “contingencies” would not “create a strong inference of [the CEO’s] knowledge of Boeing’s plan to reduce purchases of the shipsets.”
Even considering the scienter allegations in a holistic mosaic prepared by the one judge dissenting, the panel majority found the allegations inadequate. Instead, while the CEO “presumably knew, as the public did, that Boeing might reduce purchases,” the complaint failed to “show that [the CEO] knew of or consciously disregarded Boeing’s plans when he made the disputed statements.”
Turning to the November 24, 2019 analyst report, although the author said he based the key statement in it—that Spirit expected monthly sales of 737 MAX shipsets to continue at fifty-two until May 2020—on a meeting with the Spirit CEO and CFO, the Tenth Circuit held that the statement “could [be] attribute[d] [to those executives] only if they had controlled the contents or method of communication,” and “[n]othing in the complaint or the report suggests that [the CEO] or [CFO] . . . controlled” either. Moreover, since the majority had already concluded that the complaint failed to adequately allege the CEO’s scienter as of his challenged October 31, 2019 statement, his scienter with respect to the analyst report had to depend on events after that date. As to these, the plaintiffs contended that the CEO or the CFO “would have learned of Boeing’s plans at a staff meeting in mid-November 2019.” But the CEO/CFO meeting with the analyst would have occurred at some time before the analyst report—and could have been substantially before that date because the report ran to about thirty pages and would therefore have taken some time to prepare. Accordingly, the critical question was whether the mid-November staff meeting occurred before the meeting with the analyst (in which case the CEO and CFO might have known of Boeing’s plans from that staff meeting by the time they talked with the analyst) or whether the staff meeting had occurred after the meeting with the analyst (in which case the CEO and CFO could not, by the time they talked with the analyst, have known of Boeing’s plans from the staff meeting). The complaint, however, failed to plead which was the case, and “the district court couldn’t draw a strong inference of scienter from the general allegations that [the analyst] had met with [the CEO] and [CFO] at some unspecified date.”
Moving to the alleged accounting misstatements, the plaintiffs focused on certifications by the Spirit CFO and the Spirit controller attached to periodic filings, with the Tenth Circuit characterizing these generally as “certifying the adequacy of Spirit’s accounting controls.” The panel assumed “[f]or the sake of argument” that these certifications were “false,” since Spirit later acknowledged that it had failed to account properly for customer claims because of a material weakness in estimating the eventual cost of those claims. The plaintiffs’ informants characterized that weakness as loose control permitting the Spirit VP 737 MAX to understate the eventual costs of the claims in order “to make the 737 MAX program appear more profitable than it really was.”
To raise the required strong inference that the CFO and controller had scienter, the plaintiff relied, in part, on the fact that both of these officers were forced to resign. The court of appeals, however, reasoned that those resignations might have been forced either because the CFO and controller were “complicit in [the VP 737 MAX’s] understatements of contingent liabilities” or simply because the CFO and controller had, through their “inattention,” “allowed” the undervaluation. Since “[t]he inference of inattention is justifiable” and the plaintiffs failed to include “particularized allegations” connecting the resignations to fraud, the court could not infer scienter from these “personnel moves.”
The complaint also alleged that a former Spirit employee said that he or she had expressed “concerns” to the Spirit controller about “Spirit’s lack of appropriate training, delegation of too much control to [the VP 737 MAX], and his manipulation of the accounting.” That employee said that “[the controller] and other finance personnel had ‘shut [the former employee] down.’” As with the resignations, however, the Tenth Circuit saw two interpretations of this alleged interchange. On the one hand, “a factfinder might infer that [the controller] knew from [the former employee’s] expression of concern that the accounting controls were inadequate.” But the panel majority found “an even more plausible inference is that [the controller] disagreed with [the former employee] and maintained confidence in Spirit’s accounting controls.” And “[b]ecause there’s no indication that [the controller] believed [the former employee] or otherwise harbored these concerns, the complaint lacks the required particularity for a strong inference of scienter from [the former employee’s] communication with [the controller].” Indeed, since (i) the EACs were necessarily judgmental, (ii) their validity could only be ascertained after the claims were finally negotiated, and (iii) the plaintiffs affirmatively alleged that the controller did not monitor the VP 737 MAX’s estimates, the “complaint lacks a particularized reason to infer that [the controller] would have recognized an inadequacy in accounting controls as early as October 2019.”
Having found no scienter pled for the CEO, the CFO, or the controller, the Tenth Circuit considered whether the scienter of some other officer could be attributed to Spirit as a corporation. Specifically, the court of appeals applied the principle that a court “can impute scienter to a corporation if an official intentionally or recklessly makes a false statement or furnishes false information for inclusion in a statement” and considered whether applying this principle could attribute to Spirit the scienter of the VP 737 MAX.
Since the plaintiffs did not contend that the VP 737 MAX made any of the statements on which they sued, the plaintiffs were left to argue that the VP 737 MAX “furnished” information for inclusion in those statements. Although “plaintiffs allege[d] that when Spirit was reassuring investors, [the VP 737 MAX] knew about Boeing’s plan to cut purchases of the shipsets,” they did not “allege that [he] reported his information to anyone making the public disclosures or preparing a public statement” but “only that [he] had a chance to report what he knew.” The complaint similarly failed to “allege a basis to infer that [the VP 737 MAX] disclosed Spirit’s inadequate accounting controls,” pleading only that he “fudged the numbers to make himself look better.” There were no allegations that he “provided information about the adequacy of Spirit’s accounting controls to the other [individual] defendants or to anyone preparing a public statement.” And nothing “tie[d] the certifications to information from [him].”
Similarly, although the plaintiffs alleged that the VP 737 MAX “provided overly optimistic [EACs] to other executives,” none of those estimates “appear in any of the alleged statements by [the CEO], [the CFO], or [the controller].” As a final argument, the plaintiffs “tr[ied] to fill the gap” by imputing the VP 737 MAX’s scienter to Spirit simply because he was a “senior officer.” The majority of the panel acknowledged out-of-circuit authority imputing scienter to a company from an officer who ratified, approved, or tolerated a false statement, but here the complaint did not allege that the VP 737 MAX even knew of the challenged statements.
6. Rebutting Price Impact on Class Certification
Investors sued Goldman Sachs, Inc. (“Goldman”) in a Rule 10b-5 action, alleging that Goldman misrepresented (i) its business principles and (ii) how it managed conflicts of interest. The alleged misstatements of business principles (collectively the “business principles disclosure”) included: “We are dedicated to complying fully with the letter and spirit of the laws, rules and ethical principles that govern us. . . . Our clients’ interests always come first. . . . Integrity and honesty are at the heart of our business.” The challenged representations concerning conflicts of interest (collectively the “conflicts disclosure”) included: “We have extensive procedures and controls that are designed to identify and address conflicts of interest, including those designed to prevent the improper sharing of information among our businesses.” After the district court dismissed the case insofar as it challenged other statements on the ground that those statements were immaterial, the case proceeded on the business principles disclosure and the conflicts disclosure.
The plaintiffs contended that three corrective disclosures showed the market that the principles and conflicts representations were false: (i) an April 16, 2010 SEC complaint against Goldman alleging that Goldman sold the Abacus collateralized debt obligation (“CDO”) without disclosing that a hedge fund going short on the CDO had helped choose the vehicle’s assets (the “Abacus complaint”); (ii) an April 30, 2010 story in the Wall Street Journal reporting a DOJ investigation of Goldman concerning other investment products; and (iii) June 10, 2010 media reports that the SEC was probing Goldman’s role in the Hudson Funding 2006 transaction, in which Goldman created a CDO referencing unpromising assets in Goldman’s inventory then itself went short on the CDO.
The plaintiffs sought class certification under Federal Rule of Civil Procedure 23(b)(3). To grant such certification, a court must “find[] that the questions of law or fact common to class members predominate over any questions affecting only individual members.” The court cannot make such a finding if each member of the class must prove individual reliance on the allegedly fraudulent statements. The Goldman plaintiffs sought to surmount that problem by employing the fraud-on-the-market (“FOTM”) presumption—that all class members who purchased shares traded in an efficient market relied indirectly on all misrepresentations that the market impounded into the price of the stock.
The plaintiffs did not contend that the statements they challenged caused the market to inflate the price of Goldman shares but that those statements maintained an inflated price. Their argument on class certification accordingly depended on the efficient market for Goldman shares having used the information in the statements they contested to keep the Goldman stock higher than it would have been had Goldman not made those statements. As the plaintiffs saw it, while price impact could not be shown when the statements were made, their impact appeared when the efficient market dropped the Goldman share price in reaction to the corrective disclosures—by 12.79 percent upon the SEC filing its Abacus complaint, by another 9.39 percent after the Wall Street Journal article about the DOJ investigation, and by still another 4.52 percent after reports of the SEC investigating Goldman about the Hudson Funding.
Goldman opposed class certification, offering expert reports to rebut the FOTM presumption by showing that the challenged statements had had no price impact—not even to “maintain” an allegedly artificially inflated price. The class certification had a tortuous history, beginning with the district court’s certification of the class, moving to an appeal in which the Second Circuit reversed by holding inter alia that the district court had wrongly required Goldman to prove no price impact “conclusive[ly]” and should have employed a preponderance of the evidence standard, then progressing to the district court’s certification a second time after remand, followed by a Second Circuit affirmance, and then a trip to the Supreme Court.
The Court held that, in rebutting the FOTM presumption at the class certification stage, the defendant has the burden of proof and that the correct standard for that burden is preponderance of the evidence, as the Second Circuit had determined. But the Court identified a key issue as the generic nature of the statements underlying the plaintiffs’ claim.
Noting that the price maintenance theory depends on a price drop following corrective disclosures, Justice Barrett reasoned that the “final inference” necessary to apply that theory successfully “starts to break down when there is a mismatch between the contents of the misrepresentation and the corrective disclosure.” In particular, “when the earlier misrepresentation is generic” and “the later corrective disclosure is specific,” “there is less reason to infer front-end price inflation—that is, price impact—from the back-end price drop.” The Court therefore sent the case back to the Second Circuit for the court of appeals to “consider[] the generic nature of Goldman’s alleged misrepresentations in reviewing the District Court’s price impact determination.”
The Second Circuit responded by returning the case to the district court so that the lower court could make the initial effort to follow the Supreme Court’s guidance. The district court certified the class for the third time, specifically addressing Justice Barrett’s concern by concluding that “the alleged misstatements ‘are not so exceedingly more generic than the corrective disclosures that they vanquish the otherwise strong inference of price impact embedded in the evidentiary record.’” It is that decision that the Second Circuit reviewed in 2023.
The key question was whether the much more specific “back-end” disclosures actually “corrected” the much more general “front-end” statements that the plaintiffs attacked and, by that correction, brought about the resulting Goldman stock price declines. Vital to this analysis was an appreciation that “the ‘generic nature of a misrepresentation often will be important evidence of a lack of price impact’” and that a “mismatch” between the “genericness” of the front-end statements and the specificity of the back-end “corrective disclosures” “go[es] to the value of the back-end price drop as indirect evidence of a front-end, inflation-maintaining price impact.”
While the Second Circuit panel majority’s opinion wandered through one topic after another, it ultimately reasoned that “a searching price impact analysis must be conducted where (1) there is a considerable gap in front-end–back-end genericness, as the district court found here, (2) the corrective disclosure does not directly refer . . . to the alleged misstatement [as none of the three disclosures here referred to any of the challenged Goldman statements], and (3) the plaintiff claims, as plaintiffs claim here, that a company’s generic risk-disclosure was misleading by omission.” In conducting that inquiry, a court may refer to “case law bearing on materiality” and “determine not just whether the defendant spoke on topics generally important to investment decision-making, but instead whether the defendant’s generic statements on that topic were important in that regard.” A “useful indicator” in so gauging the “inflation-maintaining capacity” of those statements is whether there was “discussion in the market” of the generic statements the defendant made.
Because the panel majority concluded that the first category of statements on which the complaint focused (the business principles disclosure) were more generic than the second category (the conflicts disclosure), the panel centered its analysis on the conflicts disclosure. It reasoned that “if the district court’s mismatch analysis, centered as it is on the conflicts disclosure, cannot withstand scrutiny—and, as explained below, it cannot—then plaintiffs’ claim based on the business principles statements must also fail.”
As to the conflicts disclosure, the plaintiff ’s expert did not point to any market commentary that focused on the particular Goldman words about conflict identification and management that the complaint put at issue. Instead, the commentary on which the expert relied “might suggest that the market cared generally about how mismanaged conflicts could damage Goldman’s reputation, but they do not suggest that investors were misled by Goldman’s conflicts disclosure.” One of Goldman’s experts identified market reports that “touch[ed] on the subject of conflicts of interest . . . but [did] not expressly nor impliedly refer to the conflicts disclosure.” A second Goldman expert submitted “an analysis of 880 analyst reports published during the Class Period (both before and after the filing of the Abacus complaint), none of which reference[d] the conflicts disclosure.” Even the district court itself “found that investors would not consciously rely on the conflicts disclosure in making investment decisions.”
Accordingly, the Second Circuit reversed because its “review of the record leaves us with the firm conviction that there is an insufficient link between the corrective disclosures and the alleged misrepresentations”: Goldman had “demonstrated, by a preponderance of the evidence, that the misrepresentations did not impact Goldman’s stock price”; Goldman had therefore successfully “rebutted [the FOTM] presumption of reliance”; and “[t]he district court clearly erred in concluding otherwise, . . . therefore abus[ing] its discretion in certifying the shareholder class.” Because the parties had already “submitted a mountain of evidence,” there was no need for “further factfinding,” and the Second Circuit remanded “with instructions to decertify the class.”
Significance and analysis. Ironically, the first principal motion in Goldman was one to dismiss, which featured a defense argument that the challenged statements were too general to undergird a federal securities fraud suit. The long saga of the action’s class certification battle ends with the proof that indeed, the statements were too mushy to have affected trading and hence too mushy to have affected price. Perhaps the better course would have been for the district court to dismiss the case altogether at the outset.
7. Life Sciences
Because life sciences companies are frequent targets of securities class actions and because those lawsuits focus largely on reports of clinical test results as drugs or devices proceed toward Food and Drug Administration (“FDA”) approval, this survey treats decisions in life sciences separately, even though they involve topics addressed above, such as whether statements were false or misleading and whether defendants had scienter. Last year, the Fourth Circuit affirmed dismissal of a Rule 10b-5 claim alleging that defendants misleadingly omitted Kaplan-Meier curve graphs from descriptions of interim results from a clinical trial. The First Circuit affirmed dismissal of a Rule 10b-5 suit except for a statement that “all data” supported efficacy of high-dose treatment with a drug where the complaint alleged that was not true as to a subgroup of the patient population.
Omission from interim report of graphs showing comparative progress over trial period of the tested drug versus a control drug. MacroGenics, Inc. (“MacroGenics”) developed an antibody called Margetuxsimab to treat metastatic breast cancer. MacroGenics conducted a Phase 3 clinical trial (called “SOPHIA”) to compare the effect of Margetuximab plus chemotherapy, on the one hand, against Trastuzumab (the then current prevalent antigen) plus chemotherapy, on the other hand, in a population consisting of patients with metastatic breast cancer who had previously been treated with therapies attacking a protein (HER2), which occurs on some cancer cells and causes tumor growth. MacroGenics aimed to evaluate the trial along two endpoints: (i) progression-free survival (“PFS”), how long patients survived after enrollment in the study without any progression in their cancer; and (ii) overall survival (“OS”), how long patients lived after being diagnosed with breast cancer or after beginning treatment for it. Study design defined the PFS endpoint as when only 265 of the 536 enrolled patients remained without any disease progress and the OS endpoint as when 385 patients died.
The study reached the PFS endpoint on October 10, 2018. There followed a series of disclosures by MacroGenics and its officers about both the PFS results and the OS trends, consistently in concert with cautions that the OS endpoint had not been reached.
By a February 6, 2019 press release, MacroGenics announced that SOPHIA had reached the 265 PFS endpoint and that “‘[p]atients in the margetuximab arm experienced a 24% risk reduction in PFS compared to patients in the trastuzumab arm.’” The press release also said that, among the portion of the patient population, carriers of the CD16A 158F allele, “‘which has been previously associated with diminished clinical response to . . . [trastuzumab,] . . . patients in the margetuximab arm experienced a 32% risk reduction in PFS compared to patients in the trastuzumab arm.’”
The company added that “[f]ollow-up for determination of the impact of therapy on the sequential primary endpoint of overall survival (OS) is ongoing.” In a conference call on the same day, the MacroGenics president/CEO said that SOPHIA results would be presented at the June 2019 American Society of Clinical Oncology (“ASCO”) Conference. He added that, while the OS results were not needed for submission of a Biologic Drug Application to the FDA for margetuximab, which the company anticipated making in the last half of 2019, “the trending for OS has been positive in the direction of margetuximab, but we just don’t have enough events to be able . . . to have significance here.”
MacroGenics then made a public offering of its common stock on February 13, 2019. The offering documents cautioned that, while the company announced interim data from clinical studies, (i) “results and related findings and conclusions are subject to change following a more comprehensive review of the data related to the particular study or trial”; (ii) “audit and verification procedures . . . may result in the final data being materially different from the preliminary data we previously published”; (iii) “achievement of one primary endpoint for a trial does not guarantee that additional co-primary endpoints or secondary endpoints will be achieved”; and, specifically, (iv) “the achievement by margetuximab of its co-primary endpoint for progression-free survival events in the SOPHIA trial does not indicate whether the co-primary endpoint of overall survival will be achieved.” The company’s 10-K, filed on February 26, 2019, similarly warned that “[f]ollow-up for determination of the impact of therapy on the sequential second primary endpoint of overall survival (OS) is ongoing.”
On May 1, 2019, MacroGenics announced that “an ‘abstract containing data from SOPHIA was selected for presentation’” at ASCO. In a quarterly earnings conference call on the same day, the president/CEO said that this would likely include the OS data that MacroGenics had on October 10, 2018—the end-date for the SOPHIA data underlying the PFS results announced on February 6—and that the company was “very excited about being able to disclose additional data from the SOPHIA trial.” But he cautioned that “it is too early to evaluate the second sequential primary endpoint of overall survival as OS events continue to accrue in the study population.”
On May 15, 2019, MacroGenics issued a press release that, for the first time, published the OS results as of October 18, 2018, saying that “overall survival (OS) data based on 158 events [only 41% of the 385 required to reach the OS endpoint] were immature” but then stating that “[t]he median OS at that time was prolonged by 1.7 months in patients treated with margetuximab and chemotherapy.” The release added that “[f]or the exploratory subpopulation of patients carrying the CD16A 158F allele, the median OS was prolonged by 6.8 months in the margetuximab arm compared to the trastuzumab arm.” The company further said that “[f]or overall survival, we anticipate the preliminary positive trend in favor of Margetuximab to continue, although subsequent results could fluctuate as additional events accrue.”
On June 4, 2019, MacroGenics presented the SOPHIA data at ASCO, with the presentation stating that “the data depicted a ‘statistically significant improvement in PFS’ for the patients in the Margetuximab cohort compared to the Trastuzumab cohort of the [trial’s] population.” This presentation also provided Kaplan-Meier curve graphs for the interim OS results as of October 2018, which depicted the OS rate of the two arms of the study over time as patients died. This graph showed that, just as MacroGenics had announced on May 15, the Margetuximab arm did indeed show a 1.7 month advantage as of the date in October 2018 to which the displayed data ran. But it also showed that, as the study progressed from its beginning through the October 10, 2018 date, there were times when the Margetuximab OS rate was higher, times when the Trastuzumab OS rate was higher, and times when the two were tied. Similarly, the graph did indeed show that as of the end of the time graphed, those patients in the margetuximab arm who carried the CD16A 158F allele showed a median OS 6.7 months longer than such patients showed in the trastuzumab arm. But, as with the trial population overall, the graph showed that there were interim dates as of which one or the other was leading.
After presentation of interim OS data at a medical conference (including the Kaplan-Meier curve graphs) and resulting analyst reports, the price of MacroGenics declined by over 20 percent in two days.
Shareholders brought a lawsuit alleging violations by MacroGenics and its CEO and CFO of (i) Rule 10b-5 and (ii), on the basis of the offering documents for the company’s February 13 stock sale, violations of Securities Act sections 11 and 12(a)(2). Affirming dismissal, the Fourth Circuit reviewed de novo the district court’s analysis and examined the challenged statements in five ways in the context of the Rule 10b-5 claim.
First, plaintiffs argued that MacroGenics violated a duty to disclose the Kaplan-Meier curves when it made its statements about OS results in February and May because those statements put those OS results “in play” and misled without providing the adverse information in the curves. The court of appeals responded, as to the February 6 press release, that while its title (“MacroGenics Announces Positive Results from Pivotal Phase 3 SOPHIA Study of Margetuximab”) “did not qualify that the positive results were specifically related to PFS, that much was delineated by the text of the press release,” which provided numbers on PFS results and said, as to the OS, only that “determination of the impact of therapy on . . . (OS) is ongoing.” As to (i) statements in the February 6 conference call that the OS data had been “trending . . . positive in the direction of margetuximab,” (ii) statements in the May 1 call that MacroGenics was “very excited” about likely presenting the OS interim data at ASCO, and (iii) the numbers and comments in the May 15 press release together with the comment that the company anticipated that the OS trend would continue to be favorable—all those comments “accurately interpreted the OS interim data.” The “Kaplan-Meier curves graph[s were] not a fact contrary to Defendants’ positive statements” and “[d]efendants’ alleged awareness of the Kaplan-Meier curves graph did not conflict with their public statements.” Instead, those curves provided a vehicle for an alternative interpretation of the preliminary OS data. The company’s accurate report of OS numbers and its interpretation of those numbers did not mislead for failure to disclose—simultaneously—a presentation through a particular analytical prism that would support an alternative, less favorable interpretation.
Second, the court of appeals held that some statements the complaint attacked were inactionable puffery—“‘trending for OS has been positive in the direction of margetuximab,’ [February 6 call]; ‘we’re very excited about being able to disclose additional data from the SOPHIA trial,’ [May 1 call]; and ‘[t]he activity observed to date in SOPHIA is promising’ and ‘[f]or [OS] we anticipate the preliminary positive trend in favor of Margetuximab to continue,’ [May 15 press release].” The court found it particularly easy to reach this conclusion because the defendants “consistently qualified their expressions of optimism with warnings that the OS endpoint could still fail,” for example by the statement in the February 6 release that “determination of the impact” on OS “is ongoing”; and the caution during the February 6 call that the company did not “have enough events” to determine the “significance” of the positive “direction” of the OS interim results. Given that the optimistic generalities were “hedged with guarded and restrained positivity[,] . . . a reasonable investor could not have exclusively clung to [those generalities] when deciding to purchase MacroGenics’ stock.”
Third, the court of appeals reasoned that “[d]efendants’ positive statements about the interim OS data can also be classified as inactionable opinions, affording them further protection from suit.” This seemed to be in part motivated by the Fourth Circuit’s view that “[i]t would be a great disservice to stifle biopharmaceutical companies’ pursuit of medical advancements by failing to safeguard against an inundation of lawsuits alleging securities-law violations” based on differing interpretations of clinical test data.
Fourth, the court of appeals held that the statement in the May 15 press release—that MacroGenics “anticipate[s] the preliminary positive [OS] trend in favor of Margetuximab to continue”—fell within the Exchange Act’s forward-looking statement protection. While “[d]efendant’s statement uses ‘anticipate’ in the present tense, the verb itself has a forward-looking definition,” and “[t]he same analysis applies to the statement’s present tense use of ‘continue[,]’” particularly here, where “looking at the full context of the phrase allows us to infer that the word ‘continue’ is reliant upon the [d]efendants’ hope that the current positive results will remain the same as time elapses.” Thus, the statement fit within the protective statute’s coverage for projections of future performance.
The resulting statutory protection depended upon MacroGenics accompanying the statement with “meaningful cautionary statements identifying important factors that could cause actual results to differ materially from those in the forward-looking statement.” While the panel did not separately address MacroGenics’s many cautions in its analysis of protections for forward-looking statements, the Fourth Circuit effectively subsumed that issue within the fifth portion of its analysis that concluded, more generally, “[d]efendants’ statements [about OS] are immune from liability because they contained ample cautionary statements and Risk Warnings.” Reprising the various admonitions, the panel found them “‘extensive, specific and tailored’ to the ‘very complaints’ [p]laintiffs raise.”
With that exhaustive analysis of the Rule 10b-5 claim, the court of appeals quickly disposed of the causes of action based on Securities Act sections 11 and 12(a)(2). Since these were “inextricably intertwined with the alleged misstatements and omissions raised under their Exchange Act claims”—which the Fourth Circuit had already found wanting—the Securities Act fell as well.
Addressing two technical points, the Fourth Circuit noted, first, that the complaint alleged that MacroGenics failed to comply with Regulation S-K Item 303 because the offering documents for the February 13, 2019 offering did not disclose an “overall ‘trend’ that forecasted ‘uncertainty’ about the OS data’s final results.” In response, the court of appeals observed that an Item 303 violation requires, among other things, that an issuer “‘knew about [a trend, event, or uncertainty] before an offering.’” The court of appeals held that the complaint failed to plausibly plead this requirement, concluding that it could not “infer that Defendants believed SOPHIA’s interim OS data created an overall ‘trend’ that forecasted ‘uncertainty’ about the OS data’s final results” and that, as far as could be determined from the pleadings, the “Defendants did disclose the only trend of which they were aware,” i.e., that “per their analysis[,] the interim OS data did demonstrate a positive trend of longevity over Trastuzumab at that time.”
Turning to a second technical point, the panel likewise rejected the complaint’s contention that the offering documents offended Regulation S-K Item 105’s mandate to discuss “‘the material factors that make an investment in the [issuer] or offering speculative or risky.’” Here the court noted that the documents’ risk factors included the express caution that “‘the achievement by margetuximab of its co-primary endpoint for [PFS] events in the SOPHIA trial does not indicate whether the co-primary endpoint of [OS] will be achieved.’”
Statement that “all” data from a clinical trial supported efficacy where some subgroup data did not. Biogen, Inc. (“Biogen”) developed an antibody called aducanumab that Biogen tested in the hope that the treatment could slow Alzheimer’s disease by reducing amyloid plaque in patients’ brains. The treatment was particularly important because this antibody might thereby treat the cause of Alzheimer’s rather than just the symptoms.
Biogen undertook two Phase III aducanumab studies, named ENGAGE and EMERGE, with ENGAGE beginning one month before EMERGE. Because the APOE4 gene predisposed patients to developing Alzheimer’s, Biogen distributed the “carriers” of this gene (which constituted two-thirds of all test participants) equally over all three arms of each study—those receiving a placebo, those receiving a low dose of aducanumab, and those receiving a high dose. During the study periods, Biogen increased the dosage for APOE4 carriers twice. To prevent continuation of the studies if data from them showed it unlikely that they would prove aducanumab efficacious, Biogen conducted a futility analysis on data collected up to December 28, 2018, and on March 21, 2019, the company announced that it was terminating both ENGAGE and EMERGE. Nevertheless, the company had continued the studies during the period needed for that analysis and collected data after December 28, 2018 (during what is called below the “stub period”).
Biogen then proceeded to closely review all the data from the two studies—including from the stub period—and concluded that it supported efficacy in patients receiving a high dosage. After Biogen shared that analysis with the FDA, the agency and the company established a joint working group to further study the statistics. On October 22, 2019, Biogen’s then-Chief Medical Officer (“CMO”) stated in the third-quarter earnings call: “Our primary learning from these data is that sufficient exposure to high dose aducanumab reduced clinical decline across multiple clinical endpoints. This reduction in clinical decline was statistically significant in EMERGE, and we believe that patients . . . who achieved sufficient exposure to high dose aducanumab in ENGAGE support the findings of EMERGE. After consultation with the FDA, we believe that the totality of these data support a regulatory filing.” He explained that the two dosage changes and the inclusion of data from the stub period meant that “the larger dataset available after trial cessation included more patients with sufficient exposure to high dose aducanumab”; it was analysis of this larger data set that supported high-dose efficacy. In the same call, Biogen’s then-Vice President of Clinical Development and later Senior Vice President/Head of the Neurodegeneration Development Unit (“VPCD”) similarly said: “what we have learned clearly is that dose is very important, but that if individuals do receive 10 milligrams per kilogram then they do have an efficacious response.” Biogen’s then-CEO stated in a television interview the next day that aducanumab “is able to erode and eliminate the plaque leading to the benefits we see in terms of cognition for the patients. It reduces basically the decline, and we can see effects such as on memory orientation, language, but also functionally the ability to take care of oneself.”
In December 2019, Biogen released its “topline” results from the two studies, and its VPCD elaborated that “in EMERGE, the high dose reduced clinical decline as measured by the primary and secondary endpoints” and “a post-hoc analysis [of] data from a subset of patients exposed to the high dose of aducanumab” in ENGAGE supported the same conclusion. Biogen repeated these same conclusions on many occasions. Along the way, the CMO made what turned out to be a fatal representation during the July 22, 2020 earnings call for the second quarter: “[Y]ou really need to get to the higher dose” and “I think our data are all consistent with that” (the “all data” statement).
Biogen submitted an application for FDA approval in July 2020. On November 4, 2020, the FDA released on its website the briefing materials for the FDA Advisory Committee meeting on the aducanumab application. In them, “[t]he FDA’s commentary was overwhelmingly favorable, stating, among other things, that ‘the applicant has provided substantial evidence of effectiveness to support approval.’” However, the briefing materials included an analysis by the FDA’s statistical reviewer, Tristan Massie, providing Massie’s “belief that ‘the totality of the data does not seem to support the efficacy of the high dose[,]’ . . . that ‘there is no compelling substantial evidence of treatment effect or disease slowing[,] and that another study is needed to confirm or deny the positive study and the negative study.’” Her report then “detail[ed] subgroup-level analyses supporting” her overall conclusions.
Biogen’s stock price closed at $355.63 on November 4. It dropped to $328.90 the next day. Trading in Biogen stock was suspended on November 6, when the Advisory Committee met and “voted almost unanimously that it was unreasonable to consider EMERGE as ‘primary evidence of effectiveness of aducanumab for the treatment of Alzheimer’s disease,’ even in light of the post hoc analyses from ENGAGE and support from an aducanumab Phase I study.” The next trading day (November 9, a Monday), the stock price closed at $236.26.
Investors filed a Rule 10b-5 action against Biogen, the CEO, the CMO, and the VPCD, alleging that they misled the market during the period from the October 22, 2019 conference call to November 6, 2020, by statements that aducanumab was effective in high doses. Affirming dismissal as to every statement but the “all data” one, the First Circuit noted that the parties “d[id] not dispute . . . that the statements at issue constitute opinions.” As such, to plead falsity, the plaintiffs needed in this case to allege facts showing that the statements “did not ‘fairly align[ ]’ with the facts known to Biogen at the time.”
The complaint alleged “that subgroup data revealed . . . noncarriers, who received high dose aducanumab from the start and who had fewer treatment interruptions, did not achieve better clinical outcomes, and, when compared to placebo, the benefit for this group was ‘virtually nil’; carriers, who received high dose aducanumab following [increased dosage], did not experience better clinical outcomes after the dosing protocol change (in EMERGE, carriers’ CDR-SB scores got slightly worse); and carriers in ENGAGE are the only subgroup who ‘did better’ on high dose aducanumab.” This was sufficient for the First Circuit to conclude that “by failing to disclose the subgroup data, which would have contextualized their ‘all data’ claim, the complaint plausibly alleges that Defendants’ omission misled investors.” As to the materiality of the “all data” statement, since Biogen premised its FDA approval chances on the efficacy of “high” dosages of aducanumab, “[i]t logically follows . . . that whether all of Biogen’s data supported high dose aducanumab, or only some, was ‘a significant part of the mix of information [a reasonable investor would have] considered in evaluating [Biogen] as an investment,’ given that FDA approval of aducanumab had not yet occurred when the statement was made.”
Without evaluating whether the plaintiffs adequately pled that Biogen’s other challenged statements were materially misleading but assuming arguendo that they were, the court then turned to scienter. Again leading with the “all data” statement, the First Circuit reasoned that since the complaint alleged Biogen devoted extensive efforts to parse the Phase III data, “the logical inference is that Biogen analyzed aducanumab’s data based on carrier/noncarrier subgroups and therefore knew that at least some data did not support high dose aducanumab.” By further inference, “it follows that Defendants must have known that their failure to disclose said data risked misleading investors precisely because of what the ‘all data’ statement represented—that their ‘data [was] all consistent with’ ‘need[ing] to get to the higher dose’ of aducanumab.”
Failing to disclose subgroups that had not responded to high doses “was ‘an extreme departure from the standards of ordinary care’” and therefore reckless in a Rule 10b-5 sense given that Biogen had stated back in December 2019, while disclosing “topline” results, that the company “‘look[ed] forward to presenting all the data’ ‘in due time.’” Thereafter, as the court read the complaint, “Defendants then continued to withhold the subgroup data, despite publicly presenting aducanumab’s topline results, until . . . the Advisory Committee briefing materials were made public” on November 4, 2020, about two months after the CMO made the “all data” representation. All in all, “these allegations establish that Defendants knew they had subgroup data inconsistent with the ‘all data’ statement and consciously chose to hold back only the data that was inconsistent with their public claim.” The court therefore found adequate allegations from which to infer that the “failure to disclose said data was ‘a highly unreasonable omission,’ giving rise to a strong inference of scienter.”
However, the First Circuit immediately went on to find insufficient pleading that the defendants had scienter for any of the other challenged statements. Overall, “even if Defendants were aware of the subgroup data, it is not evident or inferable from the complaint that Defendants knew or believed that said data undermined their statements about aducanumab’s general efficacy.” Moreover, “despite the resources Defendants allegedly committed to reviewing the clinical data, there is no allegation that Defendants—or anyone else at Biogen for that matter—knew that the subgroup data undermined aducanumab’s effectiveness when Defendants made their public statements.” And the panel found “it difficult to say that the ‘incongruity’ between the subgroup data and Biogen’s conclusion as to aducanumab’s efficacy was ‘glaring’ where it involved the interpretation of significant amounts of data through complex statistical analysis.”
Notably, the plaintiffs did not claim that “at the time Biogen made the efficacy statements at issue, Biogen had been warned that the subgroup data undermined its conclusion about aducanumab’s clinical effect.” In fact, the complaint was “devoid of any allegation about how or when Defendants learned that the subgroup data potentially undermined their conclusion about aducanumab’s efficacy.” The FDA released the Massie report on November 4, 2022, after the defendants made the last of the allegedly fraudulent statements and, “even if the Defendants were on notice of Massie’s analyses at the time of their public statements, the complaint lacks any allegation that the Defendants honestly believed Massie’s interpretation of the data over their own.”
The court of appeals rejected all of what the First Circuit called the plaintiffs’ “indirect evidence” as supporting the required strong scienter inference. While plaintiffs alleged that Biogen “selective[ly] report[ed] . . . clinical data,” the First Circuit responded that “Biogen explained, during its first public statement about aducanumab after announcing futility, that the ‘details of subgroups is something that will come . . . later,’” then carefully said later that “it was presenting only aducanumab’s topline results publicly.” The company was therefore—except with respect to the “all data” statement—“transparent about what data it was withholding from investors.” Moreover, “scienter cannot be inferred from the failure to disclose the subgroup results at the time the general statements about aducanumab’s efficacy were made because Biogen’s own analysis of the data is not fully discredited by the subgroup data.”
Nor could scienter be strongly inferred from plaintiffs’ suggestions that “Biogen tasked its statisticians with reviewing the failed Phase III studies to ‘salvage any data that could support aducanumab’s approval’” or that, in doing so, the company “diverged from its prespecified analysis plan for evaluating the correlation between clinical outcomes and amyloid beta levels after unblinding the data.” Biogen repeatedly stated that it had engaged in post-study analysis and that that was the very work that provided the basis for the FDA application. While the plaintiffs contended Biogen’s selective reporting of the aducanumab results deviated from the company’s previous reporting of clinical test results, the court answered that—in the time period at issue here—Biogen was “facing an impending regulatory filing” and competing against other companies developing Alzheimer’s therapies. Taking these facts into account and that Biogen had turned over all its data to the FDA, any inference of scienter from “Biogen’s change in [public] reporting practices [was] less compelling when compared to the competing innocent explanation, particularly given the absence of any allegation that Defendants believed the subgroup data contradicted their efficacy statements.”
While the plaintiffs pointed to “irregularities in aducanumab’s FDA approval process as evidence of scienter”—e.g., “that Biogen and the FDA met or communicated almost daily for three months to analyze aducanumab’s data, that Biogen and the FDA worked together to prepare ‘highly atypical joint briefing materials’ for the Advisory Committee, [and] that the FDA submitted leading questions designed to support approval to said committee”—the First Circuit found these facts focused on possible FDA misconduct and “offer[ed] little meaningful insight into whether Defendants knew, or recklessly disregarded, that they would mislead investors about aducanumab’s efficacy by failing to disclose the subgroup data.”
Having in this way found that the complaint pled falsity and scienter for the “all data” statement but failed to plead scienter for all other statements, the First Circuit considered, last, whether the complaint adequately alleged loss causation. A plaintiff may do so by alleging that a stock price declined soon after a “corrective disclosure” revealing the truth of a falsehood or providing the information without which a defendant’s statement misleads where, considering other possible explanations for the price decline, “the factfinder can infer that it is more probable than not that it was the corrective disclosure[,] . . . as opposed to other possible depressive factors[,] . . . that caused at least a ‘substantial’ amount of the price drop.” Here, the Massie statistical analysis constituted the corrective disclosure to the falsehood that “all” Biogen data supported efficacy of aducanumab administered in high doses. That analysis debuted in public for the first time when the FDA published the briefing materials for the Advisory Committee on November 4, 2020. But those materials also included “overwhelmingly favorable” comments by the FDA. The Advisory Committee met virtually on November 6 and reached its negative conclusion toward the end of its meeting, which adjourned at 5:06 PM. The Committee revealed its vote “[l]ate that night.”
To surmount the conceptual difficulty that the stock price did not immediately fall after the November 4 publication of the Massie analysis, but only on November 9, the plaintiffs contended that Massie’s report “was ‘dense,’ ‘written for . . . world-renowned experts,’ followed 246 pages of effusive briefing material, and bore a ‘DRAFT’ watermark.” Accordingly, it “took time for the market to appreciate the merits of Massie’s report.” Finding “nothing [in First Circuit precedent] requiring that a stock’s price must drop immediately following a corrective disclosure for loss causation to be sufficiently pled,” the court of appeals held that “investors’ allegations cannot be per se implausible simply because a gap in time separates the price drop from the corrective disclosure.”
After all was said and done, the panel reversed dismissal as to the “all data” statement and otherwise affirmed.
Significance and analysis. The company made multiple statements that Phase III data supported the conclusion that high doses of aducanumab were efficacious in treating Alzheimer’s—without revealing that some subgroup data were inconsistent with this conclusion. Yet, the decision left the company exposed to possible liability only for the CMO’s statement that “you really need to get to the higher dose” and “I think our data are all consistent with that.”
Insofar as counseling is concerned, since post-test analysis frequently includes multiple statistical runs slicing the patient populations into subgroups, Biogen leaves pharmaceutical companies in the uneasy position of possibly facing lawsuits because an analysis somewhere in a mountain of computer runs shows a subgroup that did not respond well to treatment. The case also suggests that the old adage—never say never, never say always, never say none, and never say all—bears repeating to executives before they speak.
8. SEC Enforcement
In an action in which the SEC won summary judgment and the district court ordered more than $64 million in disgorgement, the Second Circuit held that (i) the 2021 amendments to 15 U.S.C. § 78u(d) did not create a new type of disgorgement; (ii) therefore all disgorgements under § 78u(d) are limited by equitable principles; and (iii) the district court (a) miscalculated the amount of the “net profits” to be disgorged and (b) improperly ordered disgorgement from relief defendants on the theory that they acted as nominees for the wrongdoer without—as to most assets—finding that the wrongdoer was the equitable owner of those assets. The Ninth Circuit reversed a summary judgment insofar as it included a civil penalty because the defendant had raised material issues of fact as to the amount of “gross pecuniary gain” to him from the wrongdoing, the degree of his scienter, and whether he had recognized that his conduct was wrong.
Whether the 2021 amendments to the disgorgement statute created a new type of disgorgement; calculations of net profit for disgorgement purposes; liability of relief defendants under nominee theory. Iftikar Ahmed worked at a venture capital firm named Oak Management Corporation (“Oak”). Ahmed identified and recommended portfolio companies in which Oak might invest. After an internal investigation found that Ahmed had misappropriated some $67 million dollars from Oak, the SEC filed a May 6, 2015 complaint against him in federal court, naming as relief defendants Ahmed’s wife, his three minor sons, and several entities “held in the Ahmeds’ names or for their benefit.” The district court entered a temporary restraining order freezing $55 million in assets and then a preliminary injunction freezing $118.3 million in assets.
The lower court then granted the Commission summary judgment on liability for violations of the Exchange Act and the Investment Advisers Act (“IAA”). It also imposed “a permanent injunction, $41,920,639 in disgorgement, $21 million in civil penalties, $1,520,953 in prejudgment interest for the period before the asset freeze at the IRS underpayment rate, and ‘actual returns on the frozen assets’ during the pendency of the asset freeze.” In doing so, the district court limited the backward reach of the disgorgement award by the five-year statute of limitations that the Supreme Court had held—in Kokesh v. SEC—applies to Commission disgorgement remedies constituting penalties.
While that Ahmed district court judgment was on appeal to the Second Circuit, the Supreme Court delivered its opinion in Liu v. SEC on June 22, 2020, holding that disgorgement in SEC enforcement actions was limited by equity principles to net profits. And in early 2021, Congress passed and the President signed the William M. (Mac) Thornberry National Defense Authorization Act for Fiscal Year 2021 (“NDAA”), which lengthened the statute of limitations applicable to disgorgement in SEC actions to ten years where the defendant has committed violations requiring scienter, and further provided that the new longer limitations period “shall apply with respect to any action or proceeding that is pending on, or commenced on or after, the date of enactment of this Act.”
In light of these developments, the Second Circuit granted an SEC motion to remand the Ahmed case so that the district court could consider recalculating the disgorgement award. The lower court then applied the ten-year limitations period and increased the disgorgement award to $64,171,646.14, with $9,755,798.34 in prejudgment interest. The Second Circuit’s 2023 opinion decided the appeal of that amended judgment, and the court of appeals affirmed it in part and vacated it in part.
At the outset, the Second Circuit declined to follow the Fifth Circuit’s 2022 decision that held that the NDAA amendments to the Exchange Act created a new type of disgorgement—legal disgorgement—which is free of equitable restraints. Instead, the Second Circuit held that disgorgement under the amended statute “must comport with traditional equitable limitations as recognized in Liu.” The court of appeals reached this conclusion because (i) “[t]he NDAA’s text evinces no intent to contradict Liu or to strip disgorgement of ‘limit[s] established by longstanding principles of equity’ in favor of an unbounded ‘legal’ form of disgorgement” and (ii) the NDAA’s legislative history suggests that the relevant provisions are “best read” simply as “clarification that equitable disgorgement is available under the Exchange Act.” With this ruling providing the intellectual framework, the court then proceeded to five topics.
First, Ahmed contended that the district court miscalculated his “net profit” to which Liu held his disgorgement must be limited. As to two companies in which he bought shares and then sold them to Oak or Oak-related entities while concealing his ownership of the shares, the district court had included in disgorgement the entire difference between his initial cost of those shares and his proceeds on sales of the shares. While Ahmed contended that the difference was due to an “‘increase in the market price of the shares’” rather than “Ahmed’s ‘unlawful activity,’” the Second Circuit held that his “misconduct with respect to these transactions was not in misrepresenting the purchase prices but in failing to disclose his conflicts of interest, which violated the [IAA].” These “transactions were thus entirely tainted, and Ahmed’s $14.4 million in profits from the transactions constituted his ‘net profits from wrongdoing’ under Liu.” The appellate court also rejected Ahmed’s additional argument—that the district court failed to reduce forfeiture by the amount of the “carried interest” he forfeited to Oak—because the carried interest constituted an “expectancy” and “disgorgement does not protect the wrongdoer’s expectancy interests.”
Second, Ahmed argued that the district court should not have applied the NDAA amendment increasing the limitations period during which he obtained property subject to disgorgement. He argued that the SEC had sued him for disgorgement under 15 U.S.C. § 78u(d)(5), not under the (d)(7) subsection, which was added by the 2021 amendment and contained the longer limitations period. But the court of appeals held that (i) since “[s]ection 78u(d)(7) did not exist at the time the SEC filed suit, . . . it would have been impossible to invoke that provision”; (ii) the SEC “brought the action ‘pursuant to the authority conferred upon it by . . . 15 U.S.C. § 78u(d)’” generally, to which section (d)(7) was added; and (iii) “the district court itself ‘did not rely solely on [15 U.S.C. § 78u(d)(5)] to authorize disgorgement in its initial ruling’ and instead exercised its inherent equitable power to do so.” Although Ahmed also argued that he had a vested interest in the five-year statute of limitations under which the district court entered its initial disgorgement order, the Second Circuit disagreed because the Supreme Court had handed down Kokesh v. SEC—which pronounced that the five-year penalty statute applied to SEC disgorgement relief—“over two years after the SEC brought this action,” and therefore “Ahmed could not have had a reliance interest in [the five-year] statute of limitations before the SEC brought this action.”
Ahmed further contended that retroactive application of the ten-year statute of limitations violated the Ex Post Facto clause in the U.S. Constitution. But that clause only applies to laws that impose penalties for crimes, and “in enacting 15 U.S.C. § 78u(d)(7), Congress clearly intended to provide a civil remedy.” While the Court ruled in Kokesh that disgorgement exceeding traditional equitable limitations is a penalty, the Second Circuit read Liu as finding that characterization inapplicable to disgorgement that does respect those limitations. Nor did the requirement for scienter as a condition for applying the ten-year limitations period change that result, because “scienter is an element of fraud, which may be harder to detect and investigate because fraud is usually committed with deception,” and the “more plausible inference” was that the longer limitations period derived from that practical point rather than from a motive to punish. Considering it all, the district court did not err in applying the longer retroactive limitations period in increasing the disgorgement amount.
Third, the relief defendants objected to awarding prejudgment interest to the disgorgement amount, computed at the IRS underpayment rate, for the period of time from the violations to the asset freeze imposed by the district court. The relief defendants contended that the interest should not be awarded because to the extent they held the assets “they did not act wrongfully or know of Ahmed’s wrongful actions.” But “the good faith of the Relief Defendants is immaterial because a prejudgment award concerns the amount that Ahmed, the primary defendant, must disgorge,” for which the relief defendants were derivatively liable. As to the interest rate used, the Second Circuit held that the district court did not abuse its discretion in using the IRS underpayment rate because it reflected the “unearned interest that the rightful owner of the funds could have received but for the fraud.”
Fourth, the Second Circuit considered the district court’s inclusion in the disgorgement award of “actual gains” from the money that Ahmed wrongfully acquired—i.e., “gains [that] ‘result from a profitable investment, use, or other disposition of the [plaintiff ’s] property, distinct from the transaction by which the defendant was originally enriched’”—during the period of the asset freeze. The Second Circuit held that such actual gains could be included in disgorgement provided that they were “not unduly remote from the fraud.” In this case, “the district court did not consider whether, or to what extent, consequential gains on frozen assets were remote from Ahmed’s fraud.” The court of appeals therefore remanded for the lower court to address that question, commenting that the district court could avoid doing so by “elect[ing] a different measure of supplemental enrichment consistent with ‘fair compensation,’ such as a fixed-interest rate for the period of the asset freeze.”
Fifth, the court addressed whether the relief defendants were liable for disgorgement at all. The district court premised their liability on the nominee theory. That theory rested on such defendants holding “bare legal” title to “all the frozen assets held in their names” and not being “the true equitable owner.” But that theory required an asset-by-asset determination of equitable ownership with the burden on the government to prove the facts underlying the theory. As to most assets, the “district court’s application of the nominee doctrine was inadequate . . . because it failed to determine whether the SEC proved that these particular assets (or groups of similar assets) were held by the Relief Defendants as mere nominees of Ahmed.” Accordingly, the Second Circuit vacated and remanded to the district court to make an asset-by-asset finding, except as to two assets that the court of appeals specifically identified and as to which the lower court had “weighed the SEC’s evidence[,] . . . considered the Relief Defendants’ counter-evidence[,] . . . and made findings on the record.”
The appellate court suggested that, if the assets were not forfeitable on the nominee theory, the district court might consider other theories, such as whether the relief defendants had (i) received the ill-gotten funds and could not prove a legitimate claim to them by (a) having given value for them (b) without notice of their “true provenance” or (ii) jointly owned the assets with Ahmed.
Significance and analysis. Ahmed’s particular importance lies in its disagreement with the Fifth Circuit’s 2022 decision that the NDAA amendments to 15 U.S.C. § 78u(d) created a new category—“legal” disgorgement—unrestricted by the limits on equitable disgorgement derived from “the teachings of equity treatises,” as Liu put it. Perhaps the Commission will propose additional amendments that will clarify.
Questions of fact precluding summary judgment imposition of civil penalties. Exchange Act section 21(d)(3) permits the SEC to seek imposition of civil penalties on defendants against whom the Commission files enforcement actions in federal court. The penalties are tiered, with the penalties for each tier increasing, so that a first-tier penalty can be levied for any violation, a second-tier penalty for any violation “involv[ing] fraud, deceit, manipulation, or deliberate or reckless disregard of a regulatory requirement” (scienter), and a third-tier penalty for any violation that involves second-tier scienter and “directly or indirectly resulted in substantial losses or created a significant risk of substantial losses to other persons.” The maximum penalty amount in any tier is the greater of (i) a set dollar amount per violation, with that amount increasing with each tier and increased by the SEC annually to account for inflation, or (ii) “the gross amount of pecuniary gain to such defendant as a result of the violation.”
The statute requires that a district court “determine[]” “the amount of a civil penalty . . . in light of the facts and circumstances.” In the Ninth Circuit, district courts should consider “(1) the degree of scienter, (2) the isolated or recurrent nature of the infraction, (3) the defendant’s recognition of the wrongful nature of his conduct, (4) the likelihood, because of the defendant’s occupation, that future violations might occur, and (5) the sincerity of his assurances against future violations”—the “Murphy factors,” so named after the defendant in the case in which the court of appeals listed them.
The Ninth Circuit applied these rules last year in SEC v. Husain. Husain and an attorney named Jaclin created shell companies, installed puppet CEOs, and engineered private placements by those companies to straw shareholders. Husain then arranged for shell companies to make a public offering “so that [their] shares could trade publicly.” A private company would then purchase one of Husain’s now-public shell companies by reverse merger and replace the puppet executives with the private company’s cadre—with the result that the private company became a public company without having to conduct a public offering.
During the implementation of this process, “in coordination with Jaclin and outside auditors, Husain directed the preparation of various SEC filings related to eight of the shells.” These “filings contained material misrepresentations and omissions regarding the business purposes of the shells, Husain’s role as the control person and promoter, the nature of the straw shareholders and puppet CEOs and, in two instances, the existence of merger plans.”
Husain made money on the sale of the shell companies to new owners. The proceeds of such sales were deposited into an escrow account “pursuant to an escrow agreement signed by nominee-representatives that Husain appointed.” After deduction of legal fees to be paid to Jaclin’s firm, the remaining proceeds were wired to a “nominee-representative’s bank account” or “paid to the offshore accounts of two entities, Liric and Ucino, that the SEC claimed were owned or controlled by Husain.” The proceeds of the sale of five of the shell companies—accomplished within the applicable statute of limitations—totaled $1,787,000.
Husain instructed a puppet CEO to testify falsely to the SEC and eventually pled guilty to obstruction of justice. In doing so, he admitted that shell companies he controlled submitted false filings to the SEC that failed to reveal his role and that he recruited puppet CEOs who did not control the shell companies. The district court sentenced Husain to three years of probation. Jaclin was also criminally punished by three years of probation, with partial home confinement for three months.
The SEC pursued both of them civilly. Jaclin settled by consenting to a judgment that included an injunction, disgorgement of $32,700.00 and interest in the amount of $7,773.80, for a total of $40,473.80—but no civil penalty.
The district court in Husain’s civil case granted summary judgment to the SEC on liability, finding that Husain violated the Securities Act, the Exchange Act, and Rule 10b-5. It also granted summary judgment on remedies—imposing a permanent injunction against violating securities laws, a seven-year bar from (i) holding the position of officer and director at public companies and (ii) participating in penny stock offerings—and a civil penalty of $1,757,000, which equaled the total amount paid for the shell companies sold during the limitations, minus an amount that approximated Jaclin’s disgorgement.
On appeal, the Ninth Circuit reversed the amount of that penalty. While holding that it was “review[ing] the district court’s choice of remedy for abuse of discretion,” the appellate court wrote that “on a summary judgment record, the district court can impose a civil penalty only after it has determined that no ‘genuine issues of material fact exist’ and all factual uncertainty is resolved in favor of the non-moving party.” Therefore, “if ‘material issues of fact are in dispute,’ the district court necessarily abuses its discretion in granting summary judgment on the amount of a civil penalty.”
Here the court of appeals found material fact issues on two matters relevant to the penalty: (1) calculation of Husain’s gross pecuniary gain and (2) scienter and contrition. First, if a district court elects to impose a penalty not in the set amount for a tier but instead in the amount of “the gross pecuniary gain,” it must limit the amount to the gain to “such defendant.” “The record indicate[d] that the proceeds [from the sales of shell companies] flowed to an escrow account, and the escrow agent then wired the proceeds to the nominee-representative’s bank account, less any amounts owed for Jaclin’s firm’s legal fees.” Husain submitted a declaration that those legal fees totaled $287,500. Thus, his several admissions—that he had “final authority” over the shell companies, he appointed representatives for them who received proceeds from the escrow account, and that the gross proceeds from the sales amounted to $1,757,000—“establish[ed] only that he and Jaclin received total sales proceeds of $1,787,000.” But as to Husain alone, his “declaration that legal fees of $287,500 were paid from the sales proceeds establishes a genuine issue of material fact whether such proceeds should be attributed to his—rather than Jaclin’s—gross pecuniary gain.”
Second, Husain “raised genuine issues of material fact on at least two [of the Murphy] factors—scienter and contrition [i.e., recognition of wrongful conduct]—that the district court cited in imposing the civil penalty on a summary judgment record.” Here, “[t]he SEC itself recognized that Husain had admitted at least part of his wrongdoing: (1) that he violated the registration requirements of Sections 5(a) and (c) of the Securities Act; (2) that the SEC filings of the shell companies contained material misrepresentations and omissions; (3) that he was liable as a control person under Section 20(a) of the Exchange Act; and (4) that he ‘did not act in good faith’ as to the Exchange Act violations.” Husain “acknowledged that he was ‘remorseful and there [was] virtually no chance of him repeating such conduct[,]’ . . . not[ing] that the Department of Justice had reached the same conclusion in his criminal case.”
While the district court had pointed out that Husain insisted his conduct had not victimized anyone and “‘refuse[d] to take responsibility for the impact of his illegal conduct on the market’s integrity,’” Husain “summarized the Assistant U.S. Attorney’s position in his criminal case, i.e. that ‘the shell corporations were legal entities and . . . Mr. Husain did not defraud investors.’” Moreover, the lower court had not “cite[d] undisputed evidence that Husain’s scheme [had] victimized individual investors.” Applying the rule that on summary judgment, the evidence must be “[v]iewed . . . in the light most favorable to the non-movant, Husain’s scheme did not victimize any member of the investing public and he took responsibility for deceiving the SEC.” Thus, he had raised a “genuine issue of material fact whether he recognized the wrongful nature of his conduct,” even though not admitting to investor harm.
The court of appeals held that Husain had also raised a material fact issue as to the degree of his scienter. While Husain did “not dispute that he acted with some level of scienter,” he argued that he had not “acted with a ‘high degree’ of scienter” because “he only intended to deceive the SEC, not investors,” and because Jaclin, his attorney for the transactions, advised him not to disclose his role in the shell companies. While he “‘now realize[d] that [Jaclin’s] advice was both wrong and illegal,’” Husain contended that the legal advice he received reduced the level of his scienter. All this, the Ninth Circuit held, was sufficient to “raise[] a genuine issue of material fact on the degree of scienter.”
Significance and analysis. Husain stands out because of its holding that the “gross amount of pecuniary gain” for purposes of computing penalty limits under 15 U.S.C. § 78u(d)(3)(A)–(C) may have to be computed defendant by defendant, depending on what each received. However, the Ninth Circuit simply reversed summary judgment while (i) stating: “provided . . . a district court properly views the evidence under Rule 56 of the Federal Rules of Civil Procedure, we do not suggest that it would abuse its discretion by imposing the maximum civil penalty on summary judgment”; and (ii) noting that “the district court did not . . . suggest that the amounts of [the defendants’] individual gains were not reasonably ascertainable.” The holding was accordingly narrow.
Husain was a two-to-one decision. The dissenter argued that the panel majority would permit any scoundrel to raise fact issues as to scienter and contrition simply by filing a declaration containing protestations that he had not wanted to hurt anyone, was deeply sorrowful for his wrongs, and that he would not stray into such error again—thereby avoiding summary judgment on the amount of penalties and forcing “wasteful evidentiary hearings.” But a district judge could avoid this issue by stating expressly that the penalty takes into account the defendant’s words of contrition and intention to forbear from future securities wrongdoing, but that the court’s penalty decision weights other factors more heavily—e.g., the time period that the scheme ran, the number of transactions, the dollar value of the transactions, and the number of misstatements. The court could conclude that, in light of these more compelling factors, it needed to impose the heaviest possible monetary penalty to prevent the currently contrite offender from yielding to future temptation.
9. Miscellaneous Cases
The Ninth Circuit held that a transaction could be exempted from computations under Exchange Act section 16(b)—on the basis that (i) the transaction was between an issuer and an officer or director and (ii) the issuer’s board approved the transaction—even though the board did not approve the transaction for the specific purpose that it be exempted from section 16(b). The Fourth Circuit reversed a defense judgment as a matter of law that a district court entered after completion of the SEC’s case in the trial of a tippee case, even though the Commission offered only circumstantial evidence that the alleged tipper had material inside information at the relevant time. The Fifth Circuit affirmed convictions for aiding and abetting and conspiring to commit securities fraud under 18 U.S.C. § 1348, where the defendants financed the distributions from one fund they were running with money from other funds they ran while representing that (i) the first fund was paying distributions with money derived from operations and (ii) one of the other funds would not engage in affiliated transactions.
Two opinions addressed 28 U.S.C. § 1658(b), which sets out a two-year discovery statute of limitations and a five-year statute of repose for civil actions in which plaintiffs allege federal securities fraud. The Ninth Circuit applied the statute of limitations in considering when the facts necessary to plead a Rule 10b-5 claim became reasonably available, which the court held to be when the SEC issued an order settling with the issuer and included in that order facts the private plaintiff needed to allege scienter. The Tenth Circuit held that a second amended complaint—that did not add any defendants and did not add any alleged misrepresentations—related back to a first amended complaint for purposes of applying the statute of repose.
Addressing economic loss resulting from revelation of the truth of a securities fraud, the Eleventh Circuit affirmed dismissal where (i) the plaintiff alleged that some corrective disclosures also continued the fraud; (ii) other corrective disclosures merely repeated already-public information; and (iii) the rest announced lawsuits or investigations that did not themselves show that a fraud had occurred, with the possible exception of a report by the issuer’s audit committee on an internal investigation, the corrective significance of which depended on later events (resignations and acknowledgment that investors could not rely on previously published financial numbers) occurring after the plaintiff had sold all its stock.
Two decisions extensively addressed imposition of sanctions against attorneys. The Third Circuit held a district court did not abuse its discretion in finding that attorneys committed violations of Rule 11(b) when representing plaintiffs in a case alleging violations of Rule 10b-5 and Security Act sections 11 and 12(a)(2); but the appellate court remanded because the lower court had not imposed sanctions, as required by 15 U.S.C. §§ 77z-1(c)(2), 78u-4(c)(2). Reviewing a case in which a district court imposed sanctions under 28 U.S.C. § 1927 on plaintiffs’ attorneys for pursuing an unmeritorious IAA section 36(b) case to trial, the Tenth Circuit held that the lower court’s sanctions did abuse discretion, with the court of appeals’ ruling resting largely on the circumstance that the lower court had recognized that a trial was needed by denying a defense motion for summary judgment and denying a defense motion to exclude the testimony of the plaintiffs’ expert.
The Second Circuit ruled that a plaintiff could not use IAA section 215(b) to sue for rescission of a contract—between a collateralized loan obligation from which the plaintiff bought a note and its portfolio manager—on the basis that the manager had violated fiduciary duties imposed by IAA section 206. That circuit also held that a bylaw amendment—providing that “control shares” bought in closed-end funds could not vote without authorization of other shareholders—violated Investment Company Act sections 2(a)(42) and 18(i).
Two courts of appeals addressed state law issues. Applying the criteria set out in Reves v. Ernst & Young. To state securities law claims—where the parties agreed that Reves governed whether a debt obligation fell within the term “note” for purposes of defining a “security” under California, Massachusetts, Colorado, and Illinois state securities laws—the Second Circuit held that notes given by issuers in a syndicated refinancing were not securities. The Eleventh Circuit applied California and Florida law to affirm dismissal of cases asserting a variety of state common law contract and tort claims against a broker that exercised its contractual right to decline buy orders when such orders generated large increases in the collateral the broker had to post with the National Securities Clearing Corporation.