City of Warren Police and Fire v. Prudential Financial Inc

70 F.4th 668

• Court: Court of Appeals for the Third Circuit
• Decided: June 13, 2023
• Docket: 21-1147
• Status: Published

Opinion

PRECEDENTIAL

UNITED STATES COURT OF APPEALS FOR THE THIRD CIRCUIT ____________

No. 21-1147 ____________

CITY OF WARREN POLICE AND FIRE RETIREMENT SYSTEM, Individually and on behalf of all others similarly situated, Appellant

v.

PRUDENTIAL FINANCIAL, INC.; CHARLES F. LOWREY; KENNETH Y. TANJI; ROBERT M. FALZON ____________

On Appeal from the United States District Court for the District of New Jersey (D.C. No. 2-19-cv-20839) District Judge: Honorable Stanley R. Chesler ____________

Argued: October 27, 2021

Before: GREENAWAY, JR., KRAUSE, and PHIPPS, Circuit Judges.

(Filed: June 13, 2023) ____________

Joseph D. Daley ROBBINS GELLER RUDMAN & DOWD 655 West Broadway, Suite 1900 San Diego, CA 92101

Peter S. Pearlman COHN LIFLAND PEARLMAN HERRMANN & KNOPF Park 80 West, Plaza One 250 Pehle Avenue, Suite 401 Saddle Brook, NJ 07663

Daniel J. Pfefferbaum [Argued] Shawn A. Williams ROBBINS GELLER RUDMAN & DOWD One Montgomery Street, Suite 1800 San Francisco, CA 94104

Douglas Wilens ROBBINS GELLER RUDMAN & DOWD 225 North East Mizner Boulevard, Suite 720 Boca Raton, FL 33432

Counsel for City of Warren Police and Fire Retirement System

2 David D. Cramer Tricia B. O’Reilly WALSH PIZZI O’REILLY & FALANGA Three Gateway Center 100 Mulberry Street, 15th Floor Newark, NJ 07102

Maeve L. O’Connor [Argued] Susan R. Gittes Aasiya F.M. Glover DEBEVOISE & PLIMPTON 66 Hudson Boulevard New York, NY 10001

Counsel for Prudential Financial, Inc.; Charles F. Lowrey; Kenneth Y. Tanji; and Robert M. Falzon

_______________________

OPINION OF THE COURT _______________________

PHIPPS, Circuit Judge. Insurance companies typically set aside funds, known as reserves, to pay for anticipated benefit claims by their policyholders. As an exercise of actuarial judgment, a wide range of considerations bear on the determination of the amount to hold in reserves. And because circumstances change, an insurer’s reserves may vary over time. But in this case, one of the country’s largest publicly traded life insurance companies suddenly announced that it would need to increase

3 its reserves by $208 million and that, in addition to a one-time charge in that amount, its earnings would be reduced by $25 million per quarter for the foreseeable future. After that news, the company’s stock price dropped by more than twelve percent over two days.

A municipal retirement system that had purchased the company’s common stock before the announcement now alleges that the company knew beforehand of problems with its reserves and misled investors about those issues. On that premise, the retirement system filed this putative class action against the company and three of its corporate executives, alleging securities fraud under § 10(b) and § 20(a) of the Securities Exchange Act of 1934.

In response to the retirement system’s amended complaint, the insurance company and the executives moved to dismiss for failure to state a claim for relief. They argued that, under the heightened pleading standard for securities-fraud claims, the retirement system’s complaint failed to plausibly allege three necessary elements of its claims: false or misleading statements; loss causation; and scienter.

The District Court granted that motion and dismissed the complaint with prejudice. It determined that the retirement system did not adequately plead falsity, and for that reason, it did not evaluate the sufficiency of the complaint’s loss causation or scienter allegations. The retirement system then brought this appeal.

While most of the District Court’s judgment holds up on de novo review, the retirement system’s amended complaint does contain particularized and plausible allegations of falsity with respect to one set of statements by the insurance company. On a conference call with investors eight weeks before the company adjusted its reserves, its Chief Financial Officer stated that the recent mortality experience of the company’s life insurance business was within the “normal” range of

4 volatility or, at worst, only “slightly negative.” App. at 76–77 (Am. Compl. ¶ 54 (emphasis removed)). But based on information from a confidential former employee, who qualifies as credible at the pleading stage, the complaint alleges that the insurance company was already contemplating a significant increase in reserves due to negative mortality experience at the time of the CFO’s statements. And the magnitude of the company’s reserve charge and its temporal proximity to the CFO’s statements further undercut the CFO’s assertion that recent mortality experience was within a normal range. Those particularized allegations satisfy the heightened standard for pleading falsity, and they plausibly allege the falsity of the CFO’s statement.

Accordingly, we will partially vacate the District Court’s judgment and remand the case to the District Court to consider in the first instance the adequacy of the amended complaint’s allegations of loss causation and scienter with respect to the CFO’s statement. I. FACTUAL BACKGROUND (AS ALLEGED IN THE AMENDED COMPLAINT) Founded over 140 years ago in Newark, New Jersey, Prudential Financial, Inc. offers a wide range of financial products and services. Those products and services include mutual funds, annuities, investment management, and life insurance. About ten percent of Prudential’s revenue comes from its Individual Life business segment, which offers term, variable, and universal life insurance policies. As part of its life insurance business, Prudential sets aside funds – reserves – to pay death-benefit claims under its policies. The amount of those reserves represents a liability for future policy benefits on its balance sheet, which Prudential publishes in its annual and quarterly reports with the Securities and Exchange Commission. To determine the amount to hold in reserves, Prudential exercises actuarial judgment in consideration of many factors, including policyholder

5 mortality rates. Typically, during the second quarter of each fiscal year, Prudential reevaluates and, if necessary, updates the actuarial assumptions underlying those calculations. If the amount held in reserves will not cover anticipated death benefits, then Prudential increases that amount, and the corresponding charge reduces its income. In January 2013, Prudential expanded its life insurance portfolio by acquiring 700,000 life insurance policies that were underwritten by another insurance company, The Hartford. Prudential paid $615 million for those policies, referred to as the ‘Hartford Block.’ Prudential was then able to collect premiums from the Hartford Block’s policyholders, but it also assumed the obligation to pay the approximately $141 billion in death benefits owed under the policies as they came due. By 2015, Prudential had fully integrated the Hartford Block into its Individual Life business segment.

The Hartford Block proved problematic for Prudential. Those policies experienced negative mortality development, meaning that policyholders were not living as long as predicted, obligating Prudential to pay death benefits sooner than expected. As a result of that negative mortality development, the Hartford Block “regularly missed internal performance expectations” from the time Prudential acquired it in 2013. App. at 73 (Am. Compl. ¶ 53(a)). In 2016 and 2017, Individual Life reported poor results due in large part to one- time adjustments made to integrate the Hartford Block. And, following its annual assumptions review in the second quarter of 2018, Prudential announced a $65 million reserve increase (and corresponding charge against Individual Life’s income), which the company attributed, in part, to updated mortality- rate assumptions. The following year, Prudential made several public statements that disavowed any serious problems with Individual Life. In its 2018 Form 10-K annual report, filed with the SEC on February 15, 2019, Prudential explained its

6 general methodology and procedure for calculating reserves. That annual report further suggested that the amount of its reserves was adequate, if not excessive, in light of low interest rates. The Form 10-K also reported Prudential’s liability for future policy benefits as well as its net income. The next month, in a meeting with analysts from the Credit Suisse investment bank, the Vice Chairman of Prudential Financial and Prudential Insurance, Robert M. Falzon, similarly assured investors that there were no systemic issues with underwriting or mortality assumptions in Individual Life. On May 3, Prudential filed its Form 10-Q for the first quarter of 2019 with the SEC, and that document, in reporting the company’s net income and providing its end-of-quarter balance sheets, disclosed no problems with the reserves for Individual Life. And on June 5, during an Investor Day conference call, Prudential’s CFO, Kenneth Y. Tanji, described Individual Life’s recent mortality experience as within the range of “normal volatility” or, at worst, only “slightly negative.” Id. at 76–77 (Am. Compl. ¶ 54 (emphasis removed)). But eight weeks after that Investor Day call, Prudential disclosed a significant adjustment to its reserves. In a July 31 press release – issued after the stock market had closed – Prudential announced that, due to unfavorable updates to its mortality assumptions, it would charge $208 million to Individual Life’s income to supplement its reserves. By Prudential’s own benchmarks, a reserve charge of that size was unusual. In a Form 8-K that Prudential had previously filed with the SEC in December 2018, the company reported that negative mortality within one standard deviation from expectation would reduce Individual Life’s annual pre-tax adjusted operating income by a comparatively smaller amount – between $55 million and $80 million. The $208 million adjustment to reserves, along with other unfavorable developments, drove Individual Life to report an adjusted operating loss of $135 million for the second quarter of 2019 – far below the company’s expected quarterly income of $108 million.

7 The analyst community immediately criticized the timing of Prudential’s announcement. In a report issued the evening of July 31, one investment bank opined that Prudential “should have used its June investor day to lay out the new disclosure and reset the bar at that point.” Id. at 81–82 (Am. Compl. ¶ 61 (emphasis removed)). Another report issued that same night predicted that “investors will most likely be surprised since this came so close to [Prudential’s] investor day in June.” Id. at 82 (Am. Compl. ¶ 62 (emphasis removed)).

The next day, three of Prudential’s officers held a conference call with analysts to discuss the company’s quarterly results. Those officers explained that the company’s revised mortality assumptions “related to the longer-dated vintages” in Individual Life, a seeming reference, at least in part, to the Hartford Block. Id. at 84 (Am. Compl. ¶ 66 (emphasis removed)). They also previewed that the updated mortality assumptions, which led to the $208 million charge, would continue to reduce Individual Life’s earnings by “about $25 million a quarter . . . for the foreseeable future.” Id. (Am. Compl. ¶ 65 (emphasis removed)).

On August 2, in its Form 10-Q for the second quarter of 2019, Prudential also commented on the reserve charge. Specifically, it stated that the reserve charge was “mainly driven by unfavorable impacts related to mortality rate assumptions.” Id. at 88 (Am. Compl. ¶ 72 (emphasis removed)).

In line with the analyst community’s reaction, the market did not respond favorably to Prudential’s announcements. On July 31, the price of Prudential’s common stock closed at $101.31 per share. But on August 1, following Prudential’s after-hours press release the evening before, the stock closed at $91.09 per share on heavy trading volume. The stock continued to drop the next day, closing at $88.56 per share on August 2. In those two days of trading, Prudential lost about one-eighth of its market capitalization.

8 II. PROCEDURAL HISTORY After the drop in stock price, two of Prudential’s shareholders – the City of Warren Police and Fire Retirement System (the ‘Warren Retirement System’) and Donald P. Crawford – separately initiated class actions against Prudential for making false or misleading statements related to the company’s life insurance reserves. They sued under the Securities Exchange Act of 1934, which regulates the trading of securities on the secondary market. See Pub. L. No. 73-291, 48 Stat. 881 (codified as amended at 15 U.S.C. § 78a et seq.). The centerpiece of the 1934 Act’s antifraud framework, § 10(b), prohibits the use of “any manipulative or deceptive device or contrivance” in violation of regulations promulgated by the SEC. 15 U.S.C. § 78j(b). And under one of the SEC’s regulations, Rule 10b-5, it is illegal “[t]o make any untrue statement of a material fact or to omit to state a material fact necessary in order to make the statements made, in the light of the circumstances under which they were made, not misleading.” 17 C.F.R. § 240.10b-5(b).

Together, § 10(b) and Rule 10b-5 imply a private cause of action for securities fraud. See Basic Inc. v. Levinson, 485 U.S. 224, 230–31 (1988) (“Judicial interpretation and application, legislative acquiescence, and the passage of time have removed any doubt that a private cause of action exists for a violation of § 10(b) and Rule 10b-5 . . . .”). That claim has six elements: (i) a misrepresentation or omission of material fact; (ii) scienter; (iii) a connection with the purchase or sale of a security; (iv) reliance; (v) economic loss; and (vi) loss causation. See Dura Pharms., Inc. v. Broudo, 544 U.S. 336, 341–42 (2005); Fan v. StoneMor Partners LP, 927 F.3d 710, 714 (3d Cir. 2019). Both the Warren Retirement System and Crawford pursued such a claim against Prudential, as well as against its Chief Executive Officer, Charles F. Lowrey, and its Chief Financial Officer, Kenneth Y. Tanji. They also brought claims under § 20(a) of the 1934 Act, which imposes joint and several liability on persons who control an individual or entity that violates § 10(b) and Rule 10b-5. See 15 U.S.C. § 78t(a);

9 Williams v. Globus Med., Inc., 869 F.3d 235, 246 (3d Cir. 2017) (“Section 20(a) of the Securities Exchange Act permits plaintiffs to bring a cause of action against individuals who control a corporation that has violated Section 10(b).” (citing 15 U.S.C. § 78t(a))).

Exercising jurisdiction over those suits, see 15 U.S.C. § 78aa(a), the District Court consolidated the actions and appointed the Warren Retirement System as lead plaintiff, see id. § 78u-4(a)(3). After that appointment, the Warren Retirement System filed an amended complaint, which, among other things, added § 10(b) and § 20(a) claims against Prudential’s Vice Chairman, Robert M. Falzon. The Warren Retirement System also sought to represent a class of persons who, like it, bought Prudential’s common stock between February 15, 2019, the date of the first alleged misrepresentation, and August 2, 2019, the second day that the company’s stock price dropped following Prudential’s corrective disclosures. In response, Prudential and the individual defendants moved to dismiss the amended complaint for failure to state a claim for relief. In that motion, they argued that the Warren Retirement System failed to adequately allege three essential elements of its securities-fraud claims: a misrepresentation or omission of material fact, scienter, and loss causation.

The District Court granted the motion on the ground that the complaint did not plausibly allege that any of Prudential’s class-period statements were false or misleading. See In re Prudential Fin., Inc. Sec. Litig., 2020 WL 7706860, at *15 (D.N.J. Dec. 29, 2020). Reasoning that any attempt to further amend would be futile, the District Court dismissed the amended complaint with prejudice without addressing scienter or loss causation. See id. The Warren Retirement System timely appealed the resulting order of dismissal, bringing the case within this Court’s appellate jurisdiction. See 28 U.S.C. § 1291.

10 III. DISCUSSION By virtue of Civil Rule 9(b) and the Private Securities Litigation Reform Act of 1995 (the ‘PSLRA’), heightened pleading standards govern securities-fraud claims brought under § 10(b) and Rule 10b-5. See Fed. R. Civ. P. 9(b); Pub. L. No. 104-67, 109 Stat. 737 (codified as amended in various sections of Title 15 of the United States Code). These heightened standards share a common foundation with the ordinary pleading standards in that speculative or threadbare allegations along with legal conclusions are disregarded, and the remaining allegations are generally taken as true. See Tellabs, Inc. v. Makor Issues & Rts., Ltd., 551 U.S. 308, 322 (2007); Lutz v. Portfolio Recovery Assocs., LLC, 49 F.4th 323, 327–28 (3d Cir. 2022). But due to Rule 9(b) and the PSLRA, a securities-fraud complaint must contain more than “a short and plain statement of the claim showing that the pleader is entitled to relief,” Fed. R. Civ. P. 8(a)(2); it must include particularized allegations of fraud. See Tellabs, 551 U.S. at 319; Fed. R. Civ. P. 9(b); 15 U.S.C. § 78u-4(b)(1)–(2).

To plead falsity, Rule 9(b) and the PSLRA each demand specificity. Rule 9(b) requires that a fraud plaintiff “state with particularity the circumstances constituting fraud.” Fed. R. Civ. P. 9(b). Under that standard, the complaint must describe the time, place, and contents of the false representations or omissions, as well as the identity of the person making the statement and the basis for the statement’s falsity. See Institutional Invs. Grp. v. Avaya, Inc., 564 F.3d 242, 253 (3d Cir. 2009) (explaining the Rule 9(b) pleading standard as requiring “the who, what, when, where and how” (quoting In re Advanta Corp. Sec. Litig., 180 F.3d 525, 534 (3d Cir. 1999), abrogated on other grounds by Tellabs, 551 U.S. at 323–34)); 5A Charles Alan Wright, Arthur R. Miller & A. Benjamin Spencer, Federal Practice & Procedure § 1297 (4th ed. 2022). Like Rule 9(b), the PSLRA requires the pleadings to identify “each statement alleged to have been misleading” and to specify “the reason or reasons why the statement is misleading.” 15 U.S.C. § 78u-4(b)(1); see also City of

11 Cambridge Ret. Sys. v. Altisource Asset Mgmt. Corp., 908 F.3d 872, 881 n.7 (3d Cir. 2018) (“In general, a complaint that satisfies the PSLRA’s heightened pleading standards will also satisfy Rule 9(b)’s requirements.”). And if allegations of falsity are based on information and belief, instead of on “evidentiary support,” Fed. R. Civ. P. 11(b)(3), the PSLRA requires the complaint to plead, with particularity, facts “sufficient to support a reasonable belief as to the misleading nature of the statement or omission” before the allegations can be accepted as true. Cal. Pub. Emps.’ Ret. Sys. v. Chubb Corp., 394 F.3d 126, 147 (3d Cir. 2004) (quoting Novak v. Kasaks, 216 F.3d 300, 314 n.1 (2d Cir. 2000)); see also 15 U.S.C. § 78u-4(b)(1). For example, even at the pleading stage, information provided by confidential witnesses is subject to a steep discount when the source is not credible or reliable. See Rahman v. Kid Brands, Inc., 736 F.3d 237, 244 (3d Cir. 2013); Chubb, 394 F.3d at 147; see also Avaya, 564 F.3d at 263 (recognizing that anonymity alone is not a basis for rejecting allegations by confidential witnesses that are otherwise reliable). 1 Upon a motion by any defendant, a claim for securities fraud under § 10(b) and Rule 10b-5 that lacks particularized allegations of falsity must be dismissed. See 15 U.S.C. § 78u- 4(b)(3)(A) (requiring the dismissal of a complaint that fails to 1 Although the pleading standards in Rule 9(b) and the PSLRA can be generally reconciled harmoniously for allegations of falsity, the PSLRA’s requirements for allegations of scienter control over the more lenient standard in Rule 9(b) for mental- state allegations. See Avaya, 564 F.3d at 253 (“The PSLRA’s requirement for pleading scienter . . . marks a sharp break with Rule 9(b).”); Tellabs, 551 U.S. at 323–24. Compare Fed. R. Civ. P. 9(b) (permitting “[m]alice, intent, knowledge, and other conditions of a person’s mind [to] be alleged generally”), with 15 U.S.C. § 78u-4(b)(2)(A) (requiring a particularized statement of the “facts giving rise to a strong inference that the defendant acted with the required state of mind”).

12 satisfy the PSLRA’s pleading rules); In re Rockefeller Ctr. Props., Inc. Sec. Litig., 311 F.3d 198, 224 (3d Cir. 2002) (recognizing that Rule 9(b) and the PSLRA “impose independent, threshold pleading requirements that, if not met, support dismissal apart from Rule 12(b)(6)”). But even if falsity is pleaded with the requisite particularity, those factual allegations must still satisfy the plausibility standard, which requires a “reasonable expectation that discovery will reveal evidence” of the necessary elements of the claim. Bell Atl. Corp. v. Twombly, 550 U.S. 544, 556 (2007); see also In re Synchrony Fin. Sec. Litig., 988 F.3d 157, 161, 169 (2d Cir. 2021) (examining whether the complaint’s “particularized allegations plausibly allege” a false or misleading statement); United States ex rel. Grubbs v. Kanneganti, 565 F.3d 180, 186 (5th Cir. 2009) (“Rule 9(b) . . . requires only ‘simple, concise, and direct’ allegations of the ‘circumstances constituting fraud,’ which after Twombly must make relief plausible . . . .”); see generally 5A Charles Alan Wright, Arthur R. Miller & A. Benjamin Spencer, Federal Practice & Procedure § 1298 (4th ed. 2022) (explaining that the requirement in Rule 9(b) to plead with particularity “must be read in conjunction with” the plausibility pleading standard). Thus, Rule 9(b) and the PSLRA do not insist upon irrefutable evidence of a statement’s falsity at the pleading stage; rather, a complaint must contain particularized factual allegations that plausibly allege that a statement was misleading.

On appeal, the Warren Retirement System argues that the District Court erred in dismissing its securities-fraud claims for failing to plead falsity. It identifies four sets of statements that it contends are pleaded with particularity and are plausibly false or misleading: (i) the statements in Prudential’s 2018 Form 10-K regarding company’s methodology for updating its reserves, (ii) the statements about the adequacy of Prudential’s reserves, which render false or misleading the financial disclosures in its 2018 Form 10-K and in its first-quarter 2019 Form 10-Q, (iii) the statement by Prudential’s Vice Chairman, Robert M. Falzon, to Credit Suisse analysts in March 2019,

13 which declared that there were no “systemic issues” with the company’s underwriting practices or mortality assumptions, App. at 67–68 (Am. Compl. ¶ 43), and (iv) the statements by Prudential’s CFO, Kenneth Y. Tanji, on June 5, 2019, during the company’s Investor Day conference regarding Prudential’s recent mortality experience being within a normal range or at worst, only slightly negative. As a fallback, the Warren Retirement System argues that the District Court abused its discretion by not permitting it to amend its complaint for a second time to augment its allegations of falsity.

A. Statements Regarding Prudential’s Reserve- Setting Methodology The Warren Retirement System argues that it adequately pleaded the falsity of two statements concerning Prudential’s methodology for determining and updating its reserves. While the amended complaint identifies the statements with particularity and provides reasons for their falsity, those allegations fail to plausibly demonstrate that Prudential misrepresented its methodology for setting reserves. In satisfaction of Rule 9(b) and the PSLRA, the amended complaint alleges with particularity the circumstances surrounding the statements about Prudential’s reserve-setting methodology. Prudential made the two statements in its 2018 Form 10-K, which was filed with the SEC on February 15, 2019. One of the statements in that annual report disclosed that Prudential’s actuarial “assumptions used in establishing reserves are generally based on [its] experience, industry experience, and/or other factors, as applicable.” App. at 64 (Am. Compl. ¶ 37 (emphasis removed)). Another statement was an assurance that although Prudential typically updates its actuarial assumptions (including those relating to mortality) once a year, the company would make an earlier adjustment if “a material change is observed in an interim period that [it] feel[s] is indicative of a long-term trend.” Id. at 65 (Am. Compl. ¶ 38 (emphasis removed)).

14 The Warren Retirement System also articulates the reason for each statement’s falsity. The amended complaint alleges that Prudential’s description of its reserve-setting methodology was misleading because Prudential ignored the negative mortality experience in the Hartford Block when setting its reserves. Similarly, the amended complaint alleges that Prudential’s assurance that it would revisit its mortality assumptions in an interim period to reflect any material changes created the false impression that no such changes had occurred since its 2018 annual review.

Although the Warren Retirement System identifies the allegedly false statements with particularity, that is not enough to survive a motion to dismiss. The allegations in the amended complaint must still be plausible. See City of Cambridge Ret. Sys., 908 F.3d at 879 n.6; In re Synchrony, 988 F.3d at 161, 169. Under that standard, the Warren Retirement System must demonstrate that discovery would be reasonably likely to reveal evidence of the falsity of the two statements. See Twombly, 550 U.S. at 556; Lutz, 49 F.4th at 328. And the allegations in the amended complaint fail to do so.

The first statement was that Prudential’s assumptions are generally based on a variety of factors, one of which is its applicable experience. To allege the plausible falsity of this statement, the Warren Retirement System must provide facts showing that Prudential did not generally consider a variety of factors, including its applicable experience, in updating its mortality assumptions. But by its own terms, the amended complaint undermines that effort. It alleges that Prudential used its experience – including with the Hartford Block – to update mortality assumptions. See App. at 74 (Am. Compl. ¶ 53(c)) (“Prudential evaluated mortality experience at least quarterly as it was a key component of Individual Life’s business performance.” (emphasis added)); id. at 64, 73–74 (Am. Compl. ¶¶ 36, 53(a), (d)) (alleging that the “Hartford [B]lock was closely monitored,” separately forecasted, and that updates to mortality assumptions had prompted a $65 million

15 reserve charge following Prudential’s 2018 annual review). The pleading also alleges that Prudential “was particularly focused on the financial impact of quarterly mortality experience,” including the Hartford Block’s experience, and was willing to update its mortality assumptions, and by extension its reserves, based on that experience. Id. at 74 (Am. Compl. ¶ 53(c)). These allegations undercut any reasonable likelihood that discovery would reveal that Prudential did not consider its experience with the Hartford Block, or its experience generally, in calibrating its mortality assumptions.

The amended complaint likewise fails to allege the plausible falsity of Prudential’s assurance about updating its mortality assumptions on an interim basis. By its terms, that statement conditioned the promise of interim updates on Prudential’s observation of a “material change” that it perceived as “indicative of a long-term trend.” Id. at 65 (Am. Compl. ¶ 38 (emphasis removed)). The Warren Retirement System, however, does not allege that between mid-2018 (when Prudential took a $65 million reserve charge based on updated mortality assumptions) and February 2019 (when it filed the Form 10-K), the company observed additional negative mortality developments within the Hartford Block that were so severe as to constitute a material change in the assumptions about Individual Life’s reserves, which had been increased less than a year earlier. The amended complaint contains even less support for the proposition that Prudential perceived the pre-February 2019 mortality experience within the Hartford Block as indicative of a long-term trend in Individual Life requiring an immediate assumptions update. Instead, as the District Court succinctly observed, the Warren Retirement System’s theory of falsity “elides the difference between short-term mortality experience in one group of policies and long-term trends for Individual Life as a whole.” In re Prudential, 2020 WL 7706860, at *13. For these reasons, the amended complaint does not plausibly allege falsity with respect to Prudential’s statements

16 about its reserve-setting methodology. And because the Warren Retirement System did not propose any additional facts that would demonstrate the plausible falsity of Prudential’s stated methodology, either before the District Court or on appeal, the District Court did not abuse its discretion in denying leave to amend this part of the claim. See City of Cambridge Ret. Sys., 908 F.3d at 879 n.6 (explaining that denial of leave to amend is proper when “the complaint ‘would not survive a Rule 12(b)(6) motion even if pled with more particularity’” (quoting In re Burlington Coat Factory Sec. Litig., 114 F.3d 1410, 1435 (3d Cir. 1997))); In re Allergan ERISA Litig., 975 F.3d 348, 357 (3d Cir. 2020) (affirming a denial of leave to amend because the plaintiffs gave the district court “no hint as to how they would further amend their complaint”); In re Adams Golf, Inc. Sec. Litig., 381 F.3d 267, 280 (3d Cir. 2004) (affirming a denial of leave to amend because the “proposed amendments would not have remedied the pleading deficiencies”).

B. Prudential’s Statements Regarding the Adequacy of its Reserves The Warren Retirement System also contends that it properly pleaded the falsity of a group of Prudential’s statements related to the adequacy of its reserves. Although the amended complaint provides the particularity required by Rule 9(b) and the PSLRA about the circumstances of those statements, the challenged statements are opinions that also are not actionable.

The Warren Retirement System identifies with particularity Prudential’s statements about the adequacy of its reserves. Prudential made those statements in two public filings with the SEC: its 2018 Form 10-K, filed on February 15, 2019, and its Form 10-Q for the first quarter of 2019, filed on May 3, 2019. Those reports disclosed, among other things, the company’s earnings per share and net income for the prior quarter, and they included the company’s end-of-period balance sheets. The balance sheets reported the company’s liabilities, which

17 included the amount held in reserves. Without any qualifiers or annotations, the identification of those amounts implied their adequacy as reserves. And in its Form 10-K, Prudential stated that in light of low interest rates, there was “an increased likelihood that the reserves determined based on best estimate assumptions may be greater than the net liabilities.” App. at 66 (Am. Compl. ¶ 39 (emphasis removed)). As far as the basis for the falsity of those statements, the amended complaint relies on information from confidential former employees. Those sources report that at the same time as Prudential made those filings with the SEC, the Hartford Block was experiencing consistently negative mortality. On that ground, the Warren Retirement System postulates that Prudential’s reserves were inadequate. At the very least, the Warren Retirement System contends, Prudential’s omission of the Hartford Block’s negative mortality gave investors false confidence in the company’s reserves and rendered misleading the suggestion in its Form 10-K that it was over-reserved. Those particularized statements about the adequacy of Prudential’s reserves are opinions. As the Supreme Court has explained, “[a]n opinion is ‘a belief[,] a view,’ or a ‘sentiment which the mind forms of persons or things,’” whereas a fact is a “‘thing done or existing’ or ‘[a]n actual happening.’” Omnicare, Inc. v. Laborers Dist. Council Const. Indus. Pension Fund, 575 U.S. 175, 183 (2015) (quoting Webster’s New International Dictionary 782, 1509 (1927)). And the setting of reserves reflects an insurer’s actuarial judgment, based on a variety of complex assumptions and considerations, of the amount that it must set aside to pay claims by policyholders. See Shapiro v. UJB Fin. Corp., 964 F.2d 272, 281 (3d Cir. 1992) (observing that there is “no single method of evaluating and setting loan loss reserves” and cautioning that “the economic judgments made in setting [such] reserves can be validated only at some future date”). Thus, when the stated amount of reserves is challenged, not on the factual ground that the indicated amount is not actually set aside, but

18 for the sufficiency of that set-aside to pay claims by policyholders, the stated reserve amount, as a manifestation of actuarial judgment, functions as an opinion. See Fait v. Regions Fin. Corp., 655 F.3d 105, 112–13 (2d Cir. 2011) (analyzing statements regarding the adequacy of loan loss reserves as opinions), abrogated on other grounds by Omnicare, 575 U.S. at 184–89. Similarly, when a claim against an insurance company for false or misleading financial statements hinges on an opinion about the adequacy of reserves, those financial statements should be treated as opinions too. 1. Opinion Falsity Under § 10(b) and Rule 10b-5. Although the challenged statements are opinions, they may still be false or misleading. In a case arising under § 11 of the Securities Act of 1933, 2 Omnicare, Inc. v. Laborers District Council Construction Industry Pension Fund, 575 U.S. 175 (2015), the Supreme Court identified three scenarios in which an opinion may be false or misleading. First, because every statement of opinion “explicitly affirms one fact: that the speaker actually holds the stated belief,” an insincere statement of the speaker’s opinion qualifies as a misrepresentation. Id. at 184. Second, opinion statements that contain expressly “embedded” factual assertions are misleading if any of the embedded factual assertions are untrue. Id. at 185–86. The third scenario involves omissions that render opinion statements misleading: if, under the circumstances in which it is given, an opinion reasonably implies facts that are untrue, then, without a qualifying statement regarding those facts, the opinion is misleading. See id. at 188–89.

Unlike Omnicare, this case involves claims under § 10(b) of the 1934 Act and Rule 10b-5. And this Circuit has not

2 Pub. L. No. 73-22, 48 Stat. 74 (codified as amended at 15 U.S.C. § 77a et seq.).

19 precedentially addressed the applicability of Omnicare’s opinion-falsity framework to claims under § 10(b) for violations of Rule 10b-5. But this Court has assumed that Omnicare applies to proxy-fraud claims under § 14(a) of the 1934 Act. See Jaroslawicz v. M&T Bank Corp., 962 F.3d 701, 717 & n.16 (3d Cir. 2020) (observing that “[t]he Supreme Court’s decision in Omnicare provides the relevant framework” for assessing the falsity of opinion statements under § 14(a)). More broadly, every other Court of Appeals to encounter the issue has applied the Omnicare framework for opinion falsity to claims for Rule 10b-5 violations. See Emps.’ Ret. Sys. of the City of Baton Rouge & Par. of E. Baton Rouge v. MacroGenics, Inc., 61 F.4th 369, 386–91 (4th Cir. 2023); Constr. Indus. & Laborers Joint Pension Tr. v. Carbonite, Inc., 22 F.4th 1, 7 (1st Cir. 2021); Carvelli v. Ocwen Fin. Corp., 934 F.3d 1307, 1322 n.7 (11th Cir. 2019); City of Dearborn Heights Act 345 Police & Fire Ret. Sys. v. Align Tech., Inc., 856 F.3d 605, 616 (9th Cir. 2017); Tongue v. Sanofi, 816 F.3d 199, 209–10 (2d Cir. 2016); Nakkhumpun v. Taylor, 782 F.3d 1142, 1159 (10th Cir. 2015).

We join that consensus: Omnicare’s framework for evaluating opinion falsity applies to claims under § 10(b) for violations of Rule 10b-5. Although differences exist between § 11 and Rule 10b-5, 3 the two provisions use almost identical language in prohibiting misrepresentations and omissions. Compare 15 U.S.C. § 77k(a) (creating liability for registration statements that “contained an untrue statement of a material fact or omitted to state a material fact required to be stated therein or necessary to make the statements therein not misleading”), with 17 C.F.R. § 240.10b-5(b) (making it illegal 3 Unlike Rule 10b-5, § 11 applies only to registration statements, and it lacks scienter and loss-causation requirements. See Obasi Inv. LTD v. Tibet Pharms., Inc., 931 F.3d 179, 182 (3d Cir. 2019) (observing that § 11 “imposes near-strict liability for untruths and omissions made in a registration statement”).

20 “[t]o make any untrue statement of a material fact or to omit to state a material fact necessary in order to make the statements made, in the light of the circumstances under which they were made, not misleading”). That textual congruence strongly suggests that the SEC, in promulgating Rule 10b-5, intended to employ the same standard for falsity as Congress used when enacting § 11. Cf. Antonin Scalia & Bryan A. Garner, Reading Law: The Interpretation of Legal Texts 323 (2012) (“[W]hen a statute uses the very same terminology as an earlier statute – especially in the very same field, such as securities law . . . – it is reasonable to believe that the terminology bears a consistent meaning.” (emphasis added)). Relatedly, this Circuit has already held that § 11 and Rule 10b-5 share the same standard of materiality for misleading statements. See In re Merck & Co., Inc. Sec. Litig., 432 F.3d 261, 273–75 (3d Cir. 2005) (reaffirming that “[s]ections 11 and 10(b) share the materiality element and the [same] materiality definition”); In re Donald J. Trump Casino Sec. Litig.-Taj Mahal Litig., 7 F.3d 357, 369 (3d Cir. 1993) (observing that § 10(b), Rule 10b-5, § 11, and § 12(2) all share the same standard of materiality). And in prior precedent, this Circuit prefigured the Omnicare framework by recognizing that, for purposes of Rule 10b-5 violations, opinion statements are misleading when “issued without a genuine belief or reasonable basis.” In re Merck & Co., Inc. Sec., Derivative & “ERISA” Litig., 543 F.3d 150, 166 (3d Cir. 2008) (quoting Herskowitz v. Nutri/System, Inc., 857 F.2d 179, 185 (3d Cir. 1988)); see also City of Edinburgh Council v. Pfizer, Inc., 754 F.3d 159, 170 (3d Cir. 2014).

In sum, Omnicare is best viewed as a more developed articulation of that principle so that for claims under § 10(b) and Rule 10b-5, an opinion statement is misleading if it: (i) was not sincerely believed when made; (ii) contains an expressly embedded, untrue factual assertion; or (iii) reasonably implies untrue facts and omits appropriate qualifying language.

21 2. None of the challenged opinion statements in this case fit within any of Omnicare’s categories of false or misleading opinions. For the first Omnicare category, the amended complaint does not contain plausible allegations that Prudential did not sincerely hold its opinion about the adequacy of its reserves. At most, the amended complaint relies on reports by three confidential former employees for the proposition that one subset of Individual Life’s portfolio, the Hartford Block, experienced negative mortality during 2018 and 2019. But even assuming, for the sake of argument, the credibility of those sources, see Avaya, 564 F.3d at 263, their accounts fail to show that Prudential, at least prior to conducting its second- quarter actuarial assumptions review, believed the Hartford Block’s problems had affected Individual Life to the point that the reserves for that entire business segment were deficient. See Williams, 869 F.3d at 246 (“[A]ctual knowledge that sales from one source might decrease is not the same as actual knowledge that the company’s overall sales projections are false.” (emphasis added)). And because the challenged statements were made before the alleged second-quarter review of actuarial assumptions could be reasonably inferred to have discovered any problems, the Warren Retirement System does not provide a basis for plausibly concluding that Prudential did not sincerely believe the adequacy of its reserve amounts as it reported them on its SEC filings.

For the second Omnicare scenario, the challenged statements have no expressly embedded factual assertions that are untrue. The statement of a likelihood of being over- reserved has only two embedded factual statements: that interest rates were low, and that Prudential held reserves at all. The truthfulness of both of those statements is undisputed. Nor are the challenged statements misleading under the falsity-by-omission scenario described in Omnicare. The Warren Retirement System’s omission argument rests on

22 information from confidential former employees that the Hartford Block had a consistently negative mortality experience, and the Warren Retirement System contends that Prudential should have disclosed that issue. Even without examining whether that confidential-source information should be discounted, the recognition that the alleged negative mortality in the Hartford Block would tend to increase the amount of needed reserves would be, at most, only one of many factors that Prudential considered in setting its reserves. And because “[r]easonable investors understand that opinions sometimes rest on a weighing of competing facts,” an opinion statement is “not necessarily misleading when an issuer knows, but fails to disclose, some fact cutting the other way.” Omnicare, 575 U.S. at 189–90. Thus, without an accompanying allegation that the negative mortality in the Hartford Block was so great that it would, for a reasonable investor, eclipse the balance of the numerous other considerations used to set reserves for all of Individual Life, the omission of that fact from Prudential’s Forms 10-K and 10-Q does not make the challenged opinions in those filings misleading. See id. at 190 (“A reasonable investor does not expect that every fact known to an issuer supports its opinion statement.”); Chubb, 394 F.3d at 156 (explaining that “anecdotal examples” of individual policies’ poor performance did not demonstrate that the broader business was failing).

For these reasons, the Warren Retirement System does not allege circumstances under which Prudential’s statements concerning the adequacy of its reserves were plausibly false or misleading opinions under any of the scenarios identified in Omnicare. Because the Warren Retirement System does not propose any allegations that would bring the company’s statements regarding the adequacy of its reserves within any of the three Omnicare scenarios, the District Court did not abuse its discretion in denying a second opportunity to amend the complaint in this respect.

23 C. The Reference to Falzon’s No-Systemic-Issues Statement in the Credit Suisse Analyst Report The Warren Retirement System also premises its securities- fraud claims upon a statement allegedly made by Prudential’s Vice Chairman, Robert M. Falzon. As required by Rule 9(b) and the PSLRA, the amended complaint identifies the circumstances surrounding that statement with particularity. According to a March 31 analyst report from the Credit Suisse investment bank, Falzon stated three days earlier, at a meeting with analysts, that “there are no systemic issues with underwriting or mortality assumptions.” App. at 67–68 (Am. Compl. ¶ 43) (emphasis removed). The amended complaint also provides the Warren Retirement System’s reason for the falsity of Falzon’s statement. That rationale depends on information from three confidential witnesses, who as former Prudential employees, assert that the Hartford Block was “not priced to cover the adverse mortality trends being experienced” and that Prudential was powerless to increase the premiums on those policies. Id. at 73–74 (Am. Compl. ¶ 53(b)). Based on that information, the Warren Retirement System contends that, contrary to Falzon’s assurances, there were systemic issues with Prudential’s underwriting practices and mortality assumptions.

The District Court rejected Credit Suisse’s report of Falzon’s statement as a basis for a Rule 10b-5 claim against Prudential. See In re Prudential, 2020 WL 7706860, at *14. It reasoned that because the statement, which was a paraphrasing, not a direct quotation, appeared in an analyst report, it was not ‘made’ by Falzon or Prudential for purposes of Rule 10b-5. See id.; see also 17 C.F.R. § 240.10b-5(b) (“It shall be unlawful for any person, directly or indirectly, . . . [t]o make any untrue statement of a material fact . . . .” (emphasis added)). In reaching that conclusion, the District Court relied on Janus Capital Group, Inc. v. First Derivative Traders, 564 U.S. 135 (2011), for the proposition that Falzon could not have made the statement because he did not “control[] the content of the report.” In re Prudential, 2020 WL 7706860, at

24 *14 (citing Janus, 564 U.S. at 142 (“Without control, a person or entity can merely suggest what to say, not ‘make’ a statement in its own right.”)). That reasoning is incorrect, but inconsequential here. See TD Bank N.A. v. Hill, 928 F.3d 259, 270 (3d Cir. 2019) (recognizing that this Court “may affirm on any basis supported by the record, even if it departs from the District Court’s rationale”). Janus announced two important points of law regarding the maker of a statement for purposes of Rule 10b-5. First, it held that for a person or entity to ‘make’ a statement, that person or entity must have “ultimate authority over the statement, including its content and whether and how to communicate it.” Janus, 564 U.S. at 142. Second, Janus emphasized the role of attribution in determining a statement’s maker: “attribution within a statement or implicit from surrounding circumstances is strong evidence that a statement was made by – and only by – the party to whom it is attributed.” Id. at 142–43. Applying those principles to allegedly false statements in a mutual fund’s prospectuses, the Supreme Court concluded that an investment advisor for the mutual fund – even if it were “significantly involved in preparing the prospectuses” – did not ‘make’ the challenged statements because the prospectuses did not attribute the statements to the fund advisor and because only the mutual fund, not the advisor, had the statutory obligation to issue the prospectuses. See id. at 146–48.

Those two Janus principles lead to a different conclusion here: Prudential, through Falzon, made the no-systemic-issues statement. The allegations about the contents of Falzon’s statement and its attribution to him are not based on information and belief, and they do not rely on confidential sources. So even under the PSLRA’s heightened pleading standard for falsity, see 15 U.S.C. § 78u-4(b); Avaya, 564 F.3d at 252–53, nothing more is needed to accept those allegations as true at this stage. From that perspective, the challenged statement is sufficiently attributed to Falzon – Credit Suisse identifies him by name as the speaker. See ESG Cap. Partners,

25 LP v. Stratos, 828 F.3d 1023, 1033 (9th Cir. 2016) (explaining that “attributing a statement to another party generally indicates that party as the ‘maker’ of the statement” for Rule 10b-5 purposes under Janus). And with Falzon’s position as Vice Chairman and the surrounding context of the statement – that it was made at a meeting where Prudential’s management discussed the company’s financial condition with outside analysts – the amended complaint allows the reasonable inference that Falzon (not Credit Suisse’s analysts) had “ultimate authority” on behalf of Prudential to speak about the company’s underwriting practices and mortality assumptions. Janus, 564 U.S. at 142. Under Janus, then, the amended complaint plausibly pleads that Prudential, through Falzon, made the challenged statement. See id. at 147 n.11; see also Basic, 485 U.S. at 227 & n.4 (treating a paraphrased statement by a company’s president that was published in a newspaper as having been “made” by the company for purposes of Rule 10b-5); Nursing Home Pension Fund, Loc. 144 v. Oracle Corp., 380 F.3d 1226, 1235 (9th Cir. 2004) (explaining, pre- Janus, that “when statements in analysts’ reports clearly originated from the defendants, and do not represent a third party’s projection, interpretation, or impression, the statements may be held to be actionable even if they are not exact quotations”). In reaching a contrary conclusion, the District Court misapplied Janus. It examined Credit Suisse’s control over the analyst report and not Prudential’s ultimate authority over the statements within the report that were attributed to Falzon. But Janus distinguishes between the act of ‘making’ a statement and the act of republishing it, such that “publishing another’s statement does not make someone the ‘maker’ of the statement.” Stratos, 828 F.3d at 1033 (quoting Janus, 564 U.S. at 142–43); see also Janus, 564 U.S. at 147 n.11 (“[A]s long as a statement is made, it does not matter whether the statement was communicated directly or indirectly to the recipient.”); Basic, 485 U.S. at 227 & n.4. In this case, because the report attributed the statement to Falzon and the context of the

26 statement indicates that he exercised control over its content and the decision to communicate it to Credit Suisse, the statement cannot, at least at the pleading stage, be considered to have been ‘made’ by Credit Suisse for purposes of Rule 10b-5.

Nevertheless, the amended complaint does not plausibly allege the falsity of Falzon’s statement. The information used to justify its falsity – underwriting problems and negative mortality in the Hartford Block – comes from confidential former employees. Even without accounting for any potential discounting of that information, see Avaya, 564 F.3d at 263, the statements support, at most, the conclusion that there were flaws in the underwriting practices and mortality assumptions for the Hartford Block that could not be remedied by raising premiums for the Hartford Block. But Falzon’s no-systemic- issues statement concerned Individual Life as a whole – not just the Hartford Block. Thus, his statement cannot be interpreted as disavowing any issues related to underwriting or mortality assumptions in the Hartford Block. Rather, the statement communicated that there were no problems with underwriting or mortality assumptions that would jeopardize the performance of all of Individual Life. And without allegations about the relative size of the Hartford Block compared to all of Individual Life, or about the magnitude of the problems in the Hartford Block relative to the full Individual Life portfolio, the amended complaint does not allow the inference that any problems with the Harford Block were systemic when Falzon made his statement.

Accordingly, Falzon’s alleged statement reproduced in the Credit Suisse analyst report is not plausibly false or misleading. Also, because the Warren Retirement System did not identify any additional information that would change this conclusion, the District Court did not abuse its discretion in denying leave to amend the complaint in this respect.

27 D. Tanji’s Investor Day Comments The most recent of the allegedly misleading statements were made on June 5, 2019, at Prudential’s Investor Day conference. The amended complaint identifies those statements and the circumstances surrounding them with the particularity required by Rule 9(b) and the PSLRA. During a presentation, Prudential’s CFO, Kenneth Y. Tanji, assured investors that Individual Life’s “recent [mortality] experience has been in between [the] range of what we’d expect[,] normal volatility, but net it has been below our experience.” App. at 76–77 (Am. Compl. ¶ 54 (emphasis removed)). Then, in response to a follow-up question from an analyst, Tanji further described Individual Life’s mortality experience as “very quarter-to-quarter, both positive and negative,” or, at worst, only “slightly negative.” Id. (emphasis removed). The amended complaint also specifies two reasons for those statements’ falsity. The first rationale relies entirely on information from a confidential former employee, referred to as ‘FE1.’ According to FE1, Prudential discussed in May 2019 that its reserves would need to be significantly increased as a result of negative mortality experience in the Hartford Block:

[A]s early as May 2019, it was discussed in forecast meetings that Individual Life was performing poorly due to negative mortality experience in the legacy Hartford [B]lock and that [Prudential] would need to take a significant charge to Individual Life adjusted operating income.

Id. at 79 (Am. Compl. ¶ 59). Tanji’s statement on June 5 that Prudential’s recent mortality experience had been within a normal range (or perhaps slightly negative) does not reconcile easily with these allegations that the company, the month before, discussed taking a significant reserve charge as a result of negative mortality experience. If credited, the information

28 from FE1 would go a long way toward establishing the plausible falsity of Tanji’s statement. The second rationale for falsity is the combined effect of the temporal proximity of Tanji’s assurances, made eight weeks before Prudential’s corrective disclosures, and the magnitude of the corrective actions – a one-time $208 million reserve charge followed by a $25 million per-quarter reduction in earnings for the foreseeable future. Those allegations increase the likelihood that, contrary to Tanji’s statements, Individual Life’s recent mortality experience could not have been within a normal range, or at worst slightly negative, on Investor Day. 4 Consistent with that conclusion, the amended complaint alleges that a sensitivity analysis published by Prudential in a Form 8-K on December 6, 2018, indicated that a one-standard-deviation change in expected mortality would decrease Individual Life’s income by $55 million to $80 million. From that reference point, it is a reasonable inference from a reserve charge of $208 million that the company’s mortality experience was not in the normal range, or at worst slightly negative, eight weeks before the charge. These allegations, if credited, along with the reasonable inferences that may be drawn from them, plausibly plead that the mortality experience for Individual Life was not within a normal range or just slightly negative as of June 5. Indeed, analysts were surprised by Prudential’s disclosure of the $208 million reserve charge so close in time to the company’s 4 Tanji’s Investor Day comments related to mortality experience, and Prudential’s corrective disclosures attributed the reserve charge to updated mortality assumptions. Those concepts are not the same, and unexpected mortality experience does not always compel updated mortality assumptions. But here, it is a reasonable inference, in light of the magnitude of the reserve charge and the information from FE1, that Prudential’s updated mortality assumptions resulted from negative mortality experience.

29 Investor Day conference. See id. at 81 (Am. Compl. ¶ 61 (citing a UBS report questioning why Prudential failed to “reset the bar” during Investor Day (emphasis removed))); id. at 82 (Am. Compl. ¶ 62 (citing a Wells Fargo report predicting investor surprise because the announcement was “so close” in time to the Investor Day conference)). Thus, FE1’s account of Prudential’s internal discussions, combined with the magnitude of the reserve charge and its close temporal proximity to Investor Day, suffice to plead that Tanji’s statements regarding the company’s recent mortality experience were plausibly false or misleading when made. 5 To avoid that outcome, Prudential attacks both alleged rationales. It contends that the confidential-source information from FE1 is unreliable and should be steeply discounted. It also disputes the inference of falsity from the combination of the temporal proximity and the magnitude of the reserve adjustment. Neither of those contentions has merit, so we will vacate and remand the dismissal of the claims, including the companion claims under § 20(a), predicated on Tanji’s Investor Day remarks. See Avaya, 564 F.3d at 252, 280 (noting that “liability under Section 20(a) is derivative of an underlying violation of Section 10(b)”).

5 Because, when taken together, the information from FE1 and the circumstances of the reserve charge demonstrate the plausible falsity of Tanji’s statements, it is not necessary to assess whether either allegation would, on its own, cross the plausibility threshold. By contrast, the allegedly misleading statements made earlier in the class period lack the same close temporal connection to the reserve charge. That attenuation, along with the lack of “mutually reinforcing” allegations and the other shortcomings identified above, makes unreasonable any inference that those prior statements were false. Avaya, 564 F.3d at 266.

30 1. The Information Supplied by the Confidential Former Employee, ‘FE1,’ Should Not Be Discounted. Prudential argues that the information provided by FE1 should be discounted to the point of insignificance. Under this Circuit’s PSLRA jurisprudence, allegations based on information from a confidential witness must be “steeply discounted” if the source is not credible or if the information is unreliable. Avaya, 564 F.3d at 262–63. Several factors guide that determination: the dependability of a confidential witness’s basis of knowledge; the level of detail provided by the witness; the degree to which other testimony or evidence corroborates the witness’s account; and the internal consistency of the information provided. See Chubb, 394 F.3d at 147; Rahman, 736 F.3d at 244. Under those factors, the information from FE1 related to Prudential’s discussions in May 2019 about taking a significant reserve charge cannot be discounted at the pleading stage.

As to the dependability of FE1’s basis of knowledge, the amended complaint provides “sufficient particularity to support the probability that a person in the position occupied by the source would possess the information alleged.” Rahman, 736 F.3d at 244 (quoting Novak, 216 F.3d at 314). Such a showing – that a confidential source, by virtue of his or her position, would have access to the information alleged – can be made through a description of the duration of the confidential witness’s employment along with explanations of how and when the confidential witness learned the information. See Avaya, 564 F.3d at 263; Chubb, 394 F.3d at 148 (requiring “allegations regarding how or why [the confidential sources] would have access to the information they purport to possess”). The amended complaint describes FE1’s basis of knowledge in the required degree of detail. FE1 was an Associate Manager in Prudential’s Planning and Analysis group between November 2016 and February 2020. In that position, FE1 regularly attended Individual Life forecast meetings with the actuarial, capital, and financial teams. And,

31 as early as May 2019, FE1 learned that Prudential was discussing, at those forecast meetings, taking a significant reserve charge due to the Hartford Block’s negative mortality experience.

The amended complaint also provides an appropriate degree of detail about the information provided by FE1. FE1 explained that, internally, Prudential understood that Individual Life as a whole was “performing poorly” and that the company attributed that poor performance to “negative mortality experience in the legacy Hartford [B]lock.” App. at 79 (Am. Compl. ¶ 59). Also according to FE1, those developments caused Prudential to discuss as early as May 2019 that it “would need to take a significant charge to Individual Life adjusted operating income.” Id. This information contains enough detail to call into question the veracity of Tanji’s Investor Day remarks that Prudential’s mortality experience was within a normal range, or at worst only slightly negative. Cf. Rahman, 736 F.3d at 245 (discounting information from a confidential witness that consisted of “little more than generalized allegations with few specifics”).

In addition, other allegations in the amended complaint corroborate the information from FE1. The explanation reported by FE1 for Prudential’s internal discussions aligns with the company’s own justification for the $208 million reserve charge – updated mortality assumptions related to Individual Life’s longer-dated vintages. See Avaya, 564 F.3d at 264; cf. generally Illinois v. Gates, 462 U.S. 213, 244–45 (1983) (noting “that corroboration through other sources of information reduce[s] the chances of a reckless or prevaricating tale” (quoting Jones v. United States, 362 U.S. 257, 271 (1960))). Similarly, FE1’s report that the relevant discussions began as early as May 2019 is consistent with the timing of Prudential’s annual, actuarial assumptions review, which occurs during the second quarter of each fiscal year. And although they did not attend the forecast meetings, two

32 other confidential witnesses with dependable bases for their more limited knowledge 6 corroborate FE1’s report of negative mortality experience in the Hartford Block. See Avaya, 564 F.3d at 266 (examining the dependability of a confidential witness’s basis of knowledge before relying on information from the witness for corroboration at the pleading stage). Consistent with FE1’s information, those witnesses reported that the Hartford Block was not adequately priced to cover its negative mortality experience and that its actuarial and data administration systems were subpar.

For similar reasons, the information from FE1 fits within a coherent narrative. FE1’s report that the Hartford Block had problems with underwriting and with consistently negative mortality does not undermine the plausibility of other allegations in the amended complaint. To the contrary, FE1’s information reconciles with the possibility that Prudential initially viewed the underwriting and negative mortality issues as localized to the Hartford Block, but that by the time of the company’s second-quarter assumptions review in May 2019, the problems with the Hartford Block began to affect Individual Life as a whole, prompting Prudential to take the $208 million reserve charge.

In sum, the factors used to evaluate the overall reliability of information from confidential sources do not reveal a basis to steeply discount the information from FE1. Consequently, the 6 As alleged, one of the confidential former employees, referred to as ‘FE2,’ worked at Prudential from April 2001 to July 2019. During that time, FE2 worked as a manager in the company’s Actuarial Project Management Office with responsibility for assessing model risks. Another confidential former employee, referred to as ‘FE3,’ worked at Prudential from 2011 to June 2018. During that time, FE3 served as an associate manager in Individual Life Insurance Financial Planning and Analysis and was responsible for financial forecasting.

33 allegations premised on the information from FE1 must be taken as true at the pleading stage. See id. at 263. 2. Inferences About the Falsity of Tanji’s Investor Day Statements Are Not Impermissible Fraud by Hindsight. Prudential separately argues that reliance on the $208 million reserve charge eight weeks after Tanji’s statements is an impermissible attempt to plead fraud by hindsight, and therefore those allegations should not receive any weight in the plausibility analysis. The fraud-by-hindsight prohibition has its greatest potency in the context of otherwise deficient allegations of scienter, 7 but for purposes of allegations of falsity, it operates as a corollary of the rule that “[t]o be actionable, a statement or omission must have been misleading at the time it was made.” In re NAHC, Inc. Sec. Litig., 306 F.3d 1314, 1330 (3d Cir. 2002); see also Williams, 869 F.3d at 244 (noting that allegations of falsity “must be sufficient to show that the challenged statements were ‘actionably unsound when made’” (quoting In re Burlington Coat Factory, 114 F.3d at 1430)). Thus, while the PSLRA forbids reliance on “speculative fraud by hindsight” allegations, In re Rockefeller, 311 F.3d at 225 (emphasis added), later developments may allow a reasonable inference that prior statements were untrue or misleading when made. See In re Merck, 432 F.3d at 272 (“[A]ny information that sheds light on whether class period statements were false or materially misleading is relevant.” (quoting In re Scholastic Corp. Sec. Litig., 252 F.3d 63, 72 (2d

7 See 15 U.S.C. § 78u-4(b)(2)(A) (requiring particularized allegations “giving rise to a strong inference that the defendant acted with the required state of mind” (emphasis added)); see also Tellabs, 551 U.S. at 320 (“The ‘strong inference’ [of scienter] formulation was appropriate, the Second Circuit said, to ward off allegations of ‘fraud by hindsight.’” (quoting Shields v. Citytrust Bancorp, Inc., 25 F.3d 1124, 1129 (2d Cir. 1994))).

34 Cir. 2001))); Plotkin v. IP Axess Inc., 407 F.3d 690, 698 (5th Cir. 2005) (recognizing that “allegations of later-emerging facts can, in some circumstances, provide warrant for inferences about an earlier situation”). And an inference of falsity is easier to justify for statements that are followed shortly by corrective disclosures of significant dimension. See Neiman v. Bulmahn, 854 F.3d 741, 751 (5th Cir. 2017) (“[T]he fact that a business files for bankruptcy on ‘Day Two,’ may, under the right surrounding circumstances, provide grounds for inferring that the business was performing poorly on ‘Day One.’” (quoting Plotkin, 407 F.3d at 698)); see also Emps.’ Ret. Sys. of Gov’t of the V.I. v. Blanford, 794 F.3d 297, 307 (2d Cir. 2015) (reasoning that “a significant gap in fourth quarter sales tends to support [a] claim that inventory was misleadingly characterized throughout the Class Period”); Novak, 216 F.3d at 312–13 (inferring from a company’s “significant write-off of inventory directly following the Class Period . . . that inventory was seriously overvalued at the time the purportedly misleading statements were made”). This case illustrates the application of those principles. Prudential’s corrective disclosures were momentous (a $208 million charge plus a $25 million quarterly impact on earnings for the foreseeable future) and close in time (eight weeks later) to Tanji’s statements. Thus, at least at the pleading stage, when Prudential has not yet had an opportunity to present its own evidence, the conclusion that Tanji’s statements were untrue is not impermissible fraud by hindsight, but instead a reasonable inference. See Plotkin, 407 F.3d at 697–98 (finding that events occurring months after the challenged statements were “so temporally connected that they shed light on the financial condition of the companies at the time” the statements were made). IV. CONCLUSION For these reasons, the Warren Retirement System plausibly pleaded falsity only with respect to CFO Tanji’s statements on June 5, 2019, regarding Prudential’s mortality experience.

35 Accordingly, we will affirm the District Court’s judgment except for a vacatur with respect to the claims premised on those statements, recognizing that this “disposition entails a shorter class period” that can begin no earlier than the date of Tanji’s statements. Avaya, 564 F.3d at 280.