In re Vale S.A. Securities Litigation

Opinion

UNITED STATES DISTRICT COURT EASTERN DISTRICT OF NEW YORK ------------------------------------x

IN RE VALE S.A. SECURITIES MEMORANDUM & ORDER LITIGATION. 19-CV-526(EK)(VMS)

------------------------------------x ERIC KOMITEE, United States District Judge:

Following the catastrophic collapse of an iron-ore tailings dam in Brazil, plaintiff brought this securities-fraud action against the dam’s owner, Vale S.A., and five of its executives. Judge Dearie certified a class of purchasers, finding that plaintiff had successfully invoked the presumption of reliance under Basic v. Levinson, 485 U.S. 224 (1988). Defendants now move to decertify the class. They advance new expert analysis, which they claim conclusively establishes that Vale’s securities operated in an inefficient market and, therefore, that the presumption of reliance is inapplicable. The Court disagrees, and for the following reasons, denies defendants’ motion for decertification. Background A. Reliance and the Basic Presumption To prevail on a Section 10(b) claim, a plaintiff must prove, among other things, that he relied on a defendant’s misrepresentation or omission. Goldman Sachs Grp., Inc. v. Arkansas Tchr. Ret. Sys., 594 U.S. 113, 117-18 (2021).1 The plaintiff can do so in two ways: (1) he can offer direct evidence that he “was aware of a defendant’s misrepresentation

and engaged in a transaction based on that misrepresentation”; or (2) he can invoke the rebuttable presumption, first adopted by the Supreme Court in Basic, that an investor “relies on a misrepresentation so long as it was reflected in the market price at the time of his transaction.” Id. Because individual reliance is not susceptible to proof by common evidence, the Basic presumption “has particular significance in securities- fraud class actions.” Id. at 118-19. Basic effectively “import[ed] fraud on the market theory from economics into securities litigation.” Id. at 130 (Gorsuch, J., dissenting). The fraud-on-the-market theory posits that, in an efficient market, a security’s price will incorporate all publicly available information and change only in response to that information. In re Petrobras Sec., 862 F.3d

250, 276 (2d Cir. 2017). To invoke the Basic presumption, plaintiffs must prove market efficiency by a preponderance of the evidence. Id. at 275 (plaintiffs’ burden); Teamsters Loc. 445 Freight Div. Pension Fund v. Bombardier Inc., 546 F.3d 196, 204 (2d Cir. 2008) (preponderance standard).

1 Unless otherwise noted, when quoting judicial decisions this order accepts all alterations and omits all citations, footnotes, and internal quotation marks. To assess market efficiency, district courts typically rely on the “Cammer” and “Krogman” factors — so named because they were first articulated in Cammer v. Bloom, 711 F. Supp. 1264 (D.N.J. 1989), and Krogman v. Sterritt, 202 F.R.D. 467 (N.D. Tex. 2001).2 The Krogman and first four Cammer factors

examine “indirect” evidence of market efficiency,3 while the fifth Cammer factor — share-price response to news that is unexpected and material — analyzes more “direct” evidence. Waggoner v. Barclays PLC, 875 F.3d 79, 94, 98 (2d Cir. 2017). This is usually done via an event study, which is a “regression analysis that seeks to show [whether] the market price of the defendant’s security tends to respond to pertinent publicly reported events.” In re Waste Mgmt. Sec. Litig., 775 F. Supp. 3d 742, 760 (S.D.N.Y. 2025). “[D]irect and indirect evidence” are not “distinct requirements.” Petrobras, 862 F.3d at 277.

Rather, courts should review these factors in “a holistic analysis based on the totality of the evidence presented.” Id. B. Prior Proceedings in This Case In 2021, plaintiff moved to certify a class of purchasers of two types of securities: Vale’s “American

2 The Second Circuit has not adopted the Cammer and Krogman factors, nor any other test for market efficiency. See Petrobras, 862 F.3d at 276. 3 Those factors are the market capitalization, bid-ask spread, float, average trading volume, number of sell-side analysts who follow and report on the stock, existence of market makers and arbitrageurs, and ability of the company to file Securities and Exchange Commission Form S-3. See Cammer, 711 F. Supp. at 1283-87; Krogman, 202 F.R.D. at 478. Depositary Shares” (“ADSs”) and several of Vale’s “Guaranteed Notes” (“Notes”). See Pl.’s Class Cert. Br. 1, ECF No. 96-1. Judge Dearie — relying on a Report and Recommendation (“R&R”)

from then-Magistrate Judge Bulsara — held that plaintiff made the necessary showing to invoke the Basic presumption on behalf of purchasers of Vale ADSs and Notes. See Order Adopting R&R 5- 10, ECF No. 116. The R&R concluded that the Cammer and Krogman factors, taken as a whole, pointed to market efficiency. R&R 13-39, ECF No. 110. Judge Bulsara agreed with defendants that plaintiff’s event study — which purported to establish the efficiency of the ADS and Notes markets — had serious “methodological deficiencies” and was therefore “entitled to no weight.” Id. at 38. But because (1) defendants’ expert offered no event study affirmatively proving that the ADS and Notes markets were

inefficient and (2) the weight of the other, “indirect” factors favored plaintiff, he determined that the markets for both securities were efficient — though the Notes market was a closer call. Id. at 39. Defendants raised several objections to the R&R, including — as relevant here — to Judge Bulsara’s conclusion that plaintiffs had satisfied their burden of establishing market efficiency. See Obj. to R&R 4-11, ECF No. 111. Judge Dearie was unpersuaded. Regarding the market for ADSs, he concluded that even if defendants had presented “some direct evidence of inefficiency,” plaintiff had still proven market efficiency on “the strength of the seven unrebutted indirect

factors.” Order Adopting R&R 7. Defendants argued that the same could not be said for the Notes market because not all the “indirect” factors pointed to efficiency. Id. at 9. Judge Dearie disagreed, concluding that because “the Cammer and Krogman factors were designed to test market efficiency for equity securities,” it is even more important for courts to be holistic in their analysis of debt securities. Id. (emphasis added). And while plaintiff’s evidence was less overwhelming when it came to the Notes market, it still weighed toward a finding of efficiency. Id. at 9-10. Vale sought leave to file an interlocutory appeal of the certification decision, which the Second Circuit denied.

ECF No. 119. C. Defendants’ Motion for Decertification A year later, defendants submitted the instant motion for decertification, which relies on a new expert report by Dr. Sumon Mazumdar — a finance professor and economic consultant. Mazumdar Rpt. ¶¶ 1-2, ECF No. 132-3. Defendants argue that Dr. Mazumdar’s event study provides “direct and unassailable evidence that the markets for Vale Securities were inefficient.” Defs.’ Decert. Br. 7, ECF No. 132-1. Plaintiff maintains that indirect evidence is sufficient to establish the efficiency of the Vale ADS and Notes markets. Pl.’s Decert. Opp’n 17-19, ECF No. 132-4. But plaintiff also responds with a new event study of its own, see generally Rebuttal Rpt. of David I. Tabak, PhD, ECF No. 132-6,

and argues that Dr. Mazumdar’s report should be disregarded because — among other reasons — he employed “a subjective approach in identifying value-relevant news.” Pl.’s Decert. Opp’n 19-22. Discussion “In opposing the decertification motion, [plaintiff] retain[s] the burden to demonstrate” that Rule 23’s requirements are satisfied. Mazzei v. Money Store, 829 F.3d 260, 270 (2d Cir. 2016). But the court must still “find that a previously satisfied requirement of Rule 23 is now lacking.” Jin v. Shanghai Original, Inc., 990 F.3d 251, 262 (2d Cir. 2021).

Here, defendants argue that “newly adduced expert evidence definitively resolves the question of market efficiency.” Defs.’ Decert. Br. 3. Defendants fail to adduce conclusive, “direct” evidence that the Vale securities traded in inefficient markets. In the absence of such evidence, the Court’s earlier ruling must stand: the weight of the remaining Cammer and Krogman factors supports a finding of efficiency. A. Mazumdar and Tabak Event Studies As with plaintiff’s initial motion for class certification, the Court is confronted with a “battle of the

experts.” The parties submitted dueling event studies. And each party levels meaningful criticisms against the other’s expert, namely regarding his methodology for selecting the “news days” used to test the Vale securities’ reactions to unexpected material information. “[I]dentifying news [and] categorizing which news is material . . . are [both] subjective determinations.” In re Barclays Bank PLC Sec. Litig., No. 09-CV-1989, 2017 WL 4082305, at *24 (S.D.N.Y. Sep. 13, 2017), aff’d, 756 F. App’x 41 (2d Cir.), as amended (Nov. 20, 2018); see also DoubleLine Cap. LP v. Odebrecht Fin., Ltd., No. 17-CV-4576, 2024 WL 1115944, at *11 (S.D.N.Y. Mar. 14, 2024) (same).4 As such, this process is highly susceptible to “cherry-picking.” In Re Waste Mgmt., 775 F. Supp. 3d at 760-61; In re Teva Sec. Litig., No. 17-CV-558,

2021 WL 872156, at *21-22 (D. Conn. Mar. 9, 2021). And here, there is legitimate cause for concern that both experts’ methods of identifying news days led to imperfect results. But Dr.

4 It is also a logically circular process: experts select material news to see if the price of a security reacts — as it should in an efficient market — but what makes news material is the fact that it produces a price reaction. Tabak at least employed defined and replicable criteria; Dr. Mazumdar did not. To identify all news days during the class period, Dr.

Mazumdar began with Vale’s Forms 6-K, its credit rating announcements, and articles about Vale identified in a Factiva search of “five well-known [news] sources” and “five others that . . . consistently provided news about Vale.” Mazumdar Rpt. ¶ 58, App’x C ¶ 4. He then removed Forms 6-K and news articles that “provided stale information or only provided information that was not value-relevant,” before aggregating the data to remove duplicative news days. Id. ¶¶ 58-59. Dr. Mazumdar’s method of identifying “value-relevant,” or material, news is both subjective and overinclusive. He provides examples of news that he deemed “not value-relevant” — e.g., articles that “simply reported Vale’s share price change

or mentioned Vale among other companies as part of broader discussion on market and industry trends,” id. ¶ 58 — but he does not identify a full set of criteria with which another person could replicate his results. That is because no such criteria exist. Instead, he began with the (doubtful) presumption that all news derived from his Form 6-K and Factiva searches was material, and then relied on his own case-by-case judgment to eliminate news that was (to him) clearly immaterial. While there will always be some subjective judgment involved in the process of selecting news days, see In re Barclays Bank, 2017 WL 4082305, at *24, “the events for study should be selected using criteria that are as objective as possible.” In re Countrywide Fin. Corp. Sec. Litig., 273 F.R.D. 586, 618 (C.D.

Cal. 2009). Dr. Mazumdar’s study does not meet that standard. The result is that Dr. Mazumdar identifies 214 news days out of 572 total days in the class period, id. ¶ 21 — a number that strains credulity. And, on review, many of his purported “news days” did not see the release of information that would be reasonably expected to move the price of a security. While defendants may quibble with Dr. Tabak’s examples of days Dr. Mazumdar should not have counted as “news days,” see Defs.’ Decert. Reply 5-6, ECF. No. 132-8, the Court can identify others: January 5, 2017 (“designation of the vice [chairman] and relocation of alternate members of the board of

directors”); March 10, 2017 (“Vale informs about executive recruiting firm to support the selection of new CEO”); June 27, 2017 (shareholder meeting minutes, with no further details). See Mazumdar Rpt. Ex. 1. This overinclusion biases his event study because he treats days with “events not expected to shift the market one way or another” as days on which such a price impact should have been observed. In Re Waste Mgmt., 775 F. Supp. 3d at 761 (“primary problem with Defendants’ event study [was] its overly expansive selection of news days”). Dr. Mazumdar’s event study therefore “does not disprove market efficiency.” Id. Dr. Tabak’s method for selecting news days, while also

flawed, at least rests on defined and replicable criteria. Dr. Tabak treated news days as “days with stories” about Vale “published by the Dow Jones Newswires.” Tabak Rpt. ¶ 39. He then sorted the list of news days by the number of stories published about Vale on each — treating repetition as a proxy for materiality — and focused on the top 25% and 10% of news days. Id. ¶¶ 41, 45-46. Dr. Tabak’s selection of news days may well be underinclusive. He only identifies forty-one news days in total, a number that he reduces further by concentrating on days with the highest number of news stories about Vale. See Tabak Rpt. Ex. 6a. His set of news days also excludes days that

surely saw the release of material, unexpected news — the day of the dam collapse and eight of Vale’s nine earnings-announcement days. Mazumdar Reply Rpt. ¶ 50, ECF No. 132-10. Moreover, repetition is an imperfect proxy for materiality. Particularly in today’s media environment, a piece of news could go viral not because it is material to investors, but because it is entertaining, politically charged, or tied to broader world events. Accordingly, while Dr. Tabak’s is the better of the two methodologies, it still does not provide unimpeachable evidence of market efficiency. B. Indirect Evidence of Market Efficiency

Even setting aside both experts’ event studies, we are still left with persuasive “indirect” evidence of market efficiency. Defendants do not attempt to rebut Judges Bulsara and Dearie’s conclusions on this point. Instead, they note only in passing that “academics have cast substantial doubt on whether the indirect factors . . . shed any meaningful light on market efficiency.” Defs.’ Decert Br. 12. “[H]ow courts should measure and evaluate market efficiency has spurred vigorous debate since almost the time Basic was decided.” In re Teva, 2021 WL 872156, at *7 & n. 16 (collecting academic critiques). And defendants are correct that the Cammer and Krogman factors have come under considerable academic criticism. See, e.g., Alon Brav & J.B. Heaton, Event

Studies in Securities Litigation: Low Power, Confounding Effects, and Bias, 93 Wash. U. L. Rev. 583, 601 (2015) (“There is no generally accepted method in financial economics for determining whether a security trades in an efficient market, and the so-called Cammer factors that are important in the courts have no general acceptance — indeed, are not even used — in academic research for that purpose.”). But so too have courts recognized the limitations of using event studies to establish “direct” evidence of market efficiency. E.g., Petrobras, 862 F.3d at 278 (“Event studies offer the seductive promise of hard numbers and dispassionate truth, but

methodological constraints limit their utility in the context of single-firm analyses.”). Accordingly, flawed though they may be, the “indirect” Cammer and Krogman factors are sufficient on their own to establish market efficiency. Waggoner, 875 F.3d at 97-98; see also R&R 32-33, 39; Order Adopting R&R 6-10. Conclusion Defendants’ motion for decertification is denied.

SO ORDERED.

/s/ Eric Komitee ERIC KOMITEE United States District Judge

Dated: March 10, 2026 Brooklyn, New York