Si v. Bed Bath & Beyond Corporation

Opinion

UNITED STATES DISTRICT COURT FOR THE DISTRICT OF COLUMBIA

BRATYA SPRL, Plaintiffs,

v. Case No. 1:22-cv-02541 (TNM)

BED BATH & BEYOND CORPORATION, et al.

Defendants.

MEMORANDUM OPINION

Defendant Ryan Cohen is a billionaire investor and entrepreneur who became internet famous during the “meme stock” craze. In March 2022, Cohen and his investment firm, RC Ventures, bought a nearly 10 percent stake in the struggling retailer Bed Bath & Beyond. This generated buzz in the meme stock community. That summer, amid a short squeeze of Bed Bath’s stock (“BBBY”), Cohen tweeted a “to the moon” emoji responding to a negative Bed Bath article. But within a week, Cohen sold his entire stake in BBBY, and its price collapsed.

Investors who bought BBBY during that August frenzy now move to certify a class of victims, alleging fraud and securities law violations. Cohen argues that class certification is inappropriate because BBBY did not trade in an efficient market during the Class Period, so reliance cannot be presumed on a class-wide basis. He also argues that the Lead Plaintiff, Bratya, cannot represent the class because it made investment decisions supposedly at odds with those of the typical class member. The Court finds that Bratya proves typicality. But because the market for BBBY stock was not efficient during the Class Period, the Court will deny Bratya’s motion to certify.

I. BACKGROUND

A. Factual Background Before it went bankrupt in April 2023, Bed Bath & Beyond was a multinational retail chain that sold an assortment of home goods. See Expert Report of Matthew D. Cain, Ph.D. (Cain Report) ¶ 15, ECF No. 108; Expert Report of Daniel R. Fischel (Fischel Report) ¶ 6, ECF No. 117-1. Along with its Bed Bath & Beyond-branded stores, the company operated the retail chain buybuy BABY, which sold products for newborns and their parents. Fischel Report ¶ 6. Bed Bath & Beyond’s common stock traded on the NASDAQ exchange. Id.

By 2022, Bed Bath & Beyond was facing major setbacks. Its sales were dropping, its losses mounting, and it was losing ground in a crowded and competitive industry. See id. ¶ 7. But Defendant Ryan Cohen and his investment firm RC Ventures promised a turnaround. In March, they filed a “Schedule 13D” form with the Securities and Exchange Commission indicating that they had acquired a 9.8 percent equity stake in Bed Bath. Id. ¶ 7. And in a letter attached to the Schedule 13D, Cohen lobbied Bed Bath’s Board of Directors to explore strategic alternatives including selling the company, selling or spinning off the buybuy BABY brand, or selling part of its business or other assets. Id.

Later that month, Cohen and Bed Bath announced a Cooperation Agreement. Fischel Report ¶ 8. Under this agreement, Bed Bath added three members of Cohen’s choosing to its Board. Id. In exchange, Cohen and RC Ventures agreed not to acquire more than a 19.9 percent stake in the company and to vote for directors nominated and proposals recommended by the Board. Id.

But these efforts did not turn the company around. In April, Bed Bath announced a poor fourth-quarter performance. Second. Am. Compl. (SAC) ¶ 103, ECF No. 66. The company

3

blamed lack of inventory, supply chain issues, and other macroeconomic factors for its disappointing results. Id. But analysts attributed the cause to the company’s antiquated supply and distribution networks, weak e-commerce platforms, and failure to resonate with younger customers. Id. Within a week, Wells Fargo and Loop Capital analysts reported that Bed Bath was “highly unlikely” to sell off buybuy BABY and questioned the company’s cash balance. Id. ¶¶ 104–05. The other shoe fell in June, when Bed Bath disclosed a quarterly loss of $2.83 per share, significantly larger than the expected loss of $1.39 per share. Fischel Report ¶ 9.

In early August, analysts began chattering about a possible short squeeze on BBBY stock. Id. ¶ 10. These rumors drummed up excitement among “meme stock” investors. Id. And high trading volume from these investors likely contributed to a bump in BBBY’s stock price. Id.; see also Cain Report ¶ 105 (discussing how “retail and other investors considered BBBY’s status as a potential meme stock and short squeeze candidate”). As one analyst reported, “With some 46% of the shares shorted, and a seemingly endless stream of bad news souring sentiment on Wall Street, Bed Bath & Beyond looks like many other unloved meme stocks that have been adopted by retail investors posting online in places like the Wall Street Bets subReddit.” Id.

Despite the ebullient mood of meme stock traders, analysts cautioned that the stock’s “current valuation . . . is disconnected from the company’s fundamentals.” Edward Helmore, Winning Bets? Meme Stock Frenzy of 2021Makes a Return, The Guardian (Aug. 13, 2022), https://www.theguardian.com/business/2022/aug/13/meme-stock-frenzygamestop-bed-bath- beyond-amc. From the July 29 to August 11, BBBY’s price jumped from $5.03 to $10.63—a 111 percent increase. Fischel Report ¶ 23. For comparison, the S&P Specialty Retail Index gained only four percent during that same interval. Id. BBBY’s price spike coincided with a swell in average weekly trading volume to 1,346.6 percent of all shares outstanding. Cain

4

Report ¶ 37. And all this giddiness occurred without “any press releases, . . . conference calls, or [other] new value-relevant information” divulged to the market. Fischel Report ¶ 25.

On August 12, as short squeeze rumors continued to swirl, Cohen retweeted a CNBC article critical of Bed Bath with his own rejoinder:

Id.

Cohen’s “full cart” tweet attracted attention from meme stock investors. SAC ¶ 149. On Reddit and other social media platforms, Cohen’s followers interpreted the smiling moon emoji

5

as “a rallying cry to buy Bed Bath stock.” Id.; see also id. ¶ 48 (explaining that in meme speak the moon emoji can suggest a stock is going “to the moon”). By the end of the day, BBBY was trading at $12.95, up from $10.65. Id. ¶ 153. And by August 15, the stock had jumped to $16 on a trading volume of over 164 million shares—double the number traded the day before. Id. ¶ 156.

After the markets closed on August 15, Cohen and RC Ventures filed an SEC Form 3 that echoed their March Schedule 13D, indicating that they owned 7,780,000 common shares of BBBY and beneficially owned 1,670,100 shares underlying call options. Fischel Report ¶ 11. Before the market opened the next morning, they filed an amendment to their March Schedule 13D reflecting that their equity stake in BBBY had increased to 11.8 percent “solely due to a change in the number of outstanding Shares of the Issuer.” Id. That evening, the stock peaked at $26.60 per share. SAC ¶ 163.

But on August 17, evidence surfaced that Cohen had pivoted. The SEC’s website published RC Ventures’ Form 144—which was dated August 16—indicating a “proposed sale” of its entire BBBY position. Fischel Report ¶ 11. And before the market closed on August 17, Cohen and RC Ventures had sold all their BBBY stock. SAC ¶¶ 167–68. The sale became public after trading hours on August 18 when Cohen and RC Ventures filed a Schedule 13D amendment disclosing that on August 16 and 17, they had sold their entire position (both stock and options) in BBBY. Fischel Report ¶ 11; Cain Report ¶ 18; see SAC ¶¶ 165, 167–68. All told, Cohen earned a profit of over $68 million based on the purchase prices of his holdings in March 2022. SAC ¶ 168. But BBBY’s price plunged following Cohen’s exit. Id. ¶ 178. By August 23, the stock was trading at $8.78—less than a third of its peak price. Id.

6

B. Procedural Background In August 2022, a pro se plaintiff brought this lawsuit against Cohen, RC Ventures, Bed Bath, and Bed Bath’s CEO, Sue Gove. See generally Compl. It alleged various claims of fraud and violations of federal securities laws. Id. ¶¶ 64–104. Then, with counsel, the plaintiff amended, alleging violations of Sections 10(b) and 20(a) of the Securities Exchange Act of 1934 and SEC Rule 10b-5; Section 20A of the Exchange Act; and Sections 9(a)(2)–(4) and 9(f) of the Exchange Act. Am. Compl. ¶¶ 197–233. The Court allowed this action to proceed against Cohen, Bed Bath, and RC Ventures based on three of Cohen’s alleged misstatements: (1) the August 12 tweet, (2) the August 16 amended Schedule 13D, and (3) the Form 144 dated August 16 (but released August 17). See In re Bed Bath & Beyond Corp. Sec. Litig., 687 F. Supp. 3d 1, 11–15 (D.D.C. 2023). The Court dismissed the plaintiff’s claims against Gove. Id. at 21. And it stayed the claims against Bed Bath because it declared bankruptcy. Id. at 6. 1 The Court appointed Bratya Lead Plaintiff in November 2022 after two other movants dropped out of the running. See Si v. Bed Bath & Beyond Corp., No. 1:22-cv-2541 (TNM), 2022 WL 22625243, at *2 (D.D.C. Nov. 16, 2022). Bratya is a family investment office incorporated in Belgium. Decl. in Support of Lead Plaintiff’s Mot. (Bratya Decl.) ¶ 2, ECF No. 14-7. It is managed by David Coti, who owns 76 percent of the firm. Transcript of Evidentiary Hearing on August 2, 2024 (Hr’g Tr.) 166:22–167:2. His brother, Edouard Coti, owns the remaining 24 percent. Id.

Before the start of the Class Period, Bratya already held 150,000 shares of Bed Bath stock and 1,131 open options contracts. Mot. for Appointment as Lead Plaintiff at 11 (Lead Pl.

1 The initial Complaint also brought claims against JP Morgan Securities, LLC, which Bratya later voluntarily dismissed. Notice of Voluntary Dismissal re JP Morgan Securities, ECF No. 67.

7

Mot.), ECF No. 14-1. 2 During the Class Period, Bratya spent several million dollars to purchase a further 201,711 shares of BBBY. Defs.’ Supp Mem. in Opp’n to Class Cert. (Defs.’ Supp. Mem.), Ex. 2, ECF No. 128-2 (showing Bratya’s portfolio transactions). 3 And because of Cohen’s alleged misstatements, Bratya claims it has suffered losses of over $4.3 million. Lead Pl. Mot. at 11.

As these large numbers suggest, Bratya is no ordinary retail investor. Both before and during the Class Period, Bratya pursued an active and sophisticated strategy to hedge its bets and navigate the heightened volatility. For one, Bratya sold naked call options for BBBY stock on August 12, the day of Cohen’s “full cart” tweet. Defs.’ Opp’n, Ex. 10, ECF No. 117-2. Then on August 15, when Cohen reaffirmed his stake in Bed Bath with a Form 3, Bratya sold even more naked call options. Id. Not until August 17 did Bratya begin purchasing Bed Bath common stock and long call options. Id. And it doubled down on its long positions after Cohen’s exit was confirmed, purchasing call options on August 22 and 26, and common stock on August 29 and 30. Id.

In February 2024, Bratya moved to certify under Rule 23(b)(3) a class of investors in BBBY common stock and long call options who acquired the securities between August 12 and August 18, 2022. Pls.’ Mot. at 5, ECF No. 109. After the motion was fully briefed, the Court held an evidentiary hearing. Bratya presented testimony from its expert, Dr. Matthew Cain, and

2 The Court’s page citations refer to the pagination generated by CM/ECF.

3 Simple though the accounting may seem, the parties regularly quote different numbers for Bratya’s purchases during the Class Period. Compare Lead Pl. Mot. at 11 (asserting Bratya purchased a total of 227,311 shares during Class Period) with Pl.’s Reply at 20 (asserting purchase of 226,611), and Defs.’ Opp’n at 39 (asserting Bratya sold $14,000 in BBBY stock on August 12 when it in fact sold 11,500 shares for a far greater price). The Court therefore relies on readouts from Bratya’s brokerage account and its own math to arrive at these figures.

8

from Bratya co-owner David Coti. Cohen presented testimony from his expert, Daniel Fischel. The Court found both experts well qualified. Having heard this testimony and reviewed the parties’ briefing, the Court will rule on the motion.

II. LEGAL STANDARDS

To obtain class certification, the proposed class must satisfy the dual burdens of Federal Rule of Civil Procedure 23. First, it must clear the “prerequisites” of Rule 23(a)—numerosity, commonality, typicality, and adequate representation. See In re Rail Freight Fuel Surcharge Antitrust Litig.-MDL No. 1869, 725 F.3d 244, 249 (D.C. Cir. 2013). And second, the proposed class must fit one of the three categories of Rule 23(b). Id. Plaintiffs seek certification under Rule 23(b)(3), meaning they must show that “questions of law or fact common to class members predominate over any questions affecting only individual members, and that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy.” Fed. R. Civ. P. 23(b)(3).

“Class certification is far from automatic.” In re Rail Freight Fuel Surcharge Antitrust Litig., 725 F.3d at 249. “A party seeking class certification must affirmatively demonstrate his compliance with [Rule 23]—that is, he must be prepared to prove that there are in fact sufficiently numerous parties, common questions of law or fact, etc.” Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 350 (2011). Prove is the operative word here: “Rule 23 does not set forth a mere pleading standard.” Id. This means that class certification determinations can “resemble[] an appraisal of the merits,” since a court will often need to “probe behind the pleadings.” Id. (cleaned up). The point of this examination is to confirm that “absent class members are protected and that no one has his rights trammeled on by a stranger suing on his behalf.” Gur- Ravantab v. Georgetown Univ., 345 F.R.D. 1, 3 (D.D.C. 2023).

9

Still, courts must not “engage in free-ranging merits inquiries at the certification stage.”

Amgen Inc. v. Connecticut Ret. Plans & Tr. Funds, 568 U.S. 455, 466 (2013). Merits questions may be considered only to the extent that “that they are relevant to determining whether the Rule 23 prerequisites for class certification are satisfied.” Id.

III. ANALYSIS

A. Rule 23(a) Requirements Bratya seeks to certify a class consisting of:

All persons or entities who purchased or otherwise acquired Bed Bath’s common stock and long call options between August 12, 2022, and August 18, 2022, both dates inclusive (hereafter the “Class Period”). Excluded from the Class are Defendants, the current and former officers and directors of Bed Bath and RC Ventures LLC, any person or entity that had or currently holds a controlling interest in Bed Bath or RC Ventures LLC, members of the Defendants’ immediate families and their legal representatives, heirs, successors or assigns, and any entity in which the Defendants or Bed Bath have or had a controlling interest.

Pls.’ Mot. at 5.

Bratya contends that class certification is appropriate under Rule 23(b)(3) because common questions of law and fact predominate. Pls.’ Mot. at 6. Cohen does not contest numerosity and commonality, so the Court addresses those factors briefly. It then turns to the thornier issues of typicality and adequacy.

1. Numerosity

To satisfy Rule 23(a)(1)’s numerosity requirement, plaintiffs must establish that “the class is so numerous that joinder of all members is impracticable.” See Fed. R. Civ. P. 23(a)(1). The numerosity requirement “imposes no absolute limitations,” but rather “requires examination of the specific facts of each case.” Gen. Tel. Co. of the Nw. v. EEOC, 446 U.S. 318, 330 (1980).

In securities fraud class actions, numerosity “may be satisfied by a showing that a large number of shares were outstanding and traded during the relevant period.” In re Vivendi

10

Universal, S.A., 242 F.R.D. 76, 84 (S.D.N.Y. 2007) (cleaned up); see also In re Deutsche Telekom Ag Sec. Litig., 229 F. Supp. 2d 277 (S.D.N.Y. 2002) (“Class certification is frequently appropriate in securities fraud cases involving a large number of shares traded publicly in an established market.”).

According to Bed Bath’s Form 10-K filed on April 21, 2022, the company had around 1,600 shareholders of record, who in turn represented additional beneficial owners. Pls. Mot. at 6. Based on this figure, joinder of all class members is clearly “impracticable.” Fed. R. Civ. P. 23(a)(1); cf. Consol. Rail Corp. v. Town of Hyde Park, 47 F.3d 473, 483 (2d Cir.1995) (observing that “numerosity is presumed at a level of 40 members”). So numerosity is satisfied.

2. Commonality

Rule 23(a)(2) requires plaintiffs to show that “there are questions of law or fact typical to the class.” To satisfy this requirement, plaintiffs’ “claims must depend upon a common contention” that is “capable of classwide resolution—which means that determination of its truth or falsity will resolve an issue that is central to the validity of each one of the claims in one stroke.” Wal-Mart Stores, Inc., 564 U.S. at 350. “The touchstone of the commonality inquiry is ‘the capacity of a classwide proceeding to generate common answers apt to drive the resolution of the litigation.’” Coleman through Bunn v. District of Columbia, 306 F.R.D. 68, 82 (D.D.C. 2015) (quoting id. at 350).

Bratya’s lawsuit raises several common questions of law and fact. These include whether Cohen and his co-Defendants knowingly made materially false or misleading statements and manipulated the market for BBBY’s securities; whether Defendants’ alleged misconduct artificially inflated the price of BBBY securities during the Class Period; and the extent of class members’ damages. See Pls.’ Mot. at 7. In securities class actions, commonality often exists

11

“where investors sue for misrepresentations or omissions that had an impact on stock price.” Howard v. Liquidity Servs. Inc., 322 F.R.D. 103, 118 (D.D.C. 2017). While Cohen contests other requirements that overlap with commonality—namely, predominance—he does not dispute that Plaintiffs satisfy Rule 23(a)(2). Cf. id. (finding commonality where element was uncontested). So Bratya’s claims are common.

3. Typicality

Rule 23(a)(3) requires the representative parties’ claims or defenses to be “typical of the claims or defenses of the class.” Typicality is satisfied so long as “each class member’s claim arises from the same course of events that led to the claims of the representative parties and each class member makes similar legal arguments to prove the defendant’s liability.” Trombley v. Nat’l City Bank, 826 F. Supp. 2d 179, 192 (D.D.C. 2011) (cleaned up). This does not mean that the facts and claims of all class members must be “identical.” In re APA Assessment Fee Litig., 311 F.R.D. 8, 15 (D.D.C. 2015). There just needs to be “reason to believe that the plaintiff’s claims are ‘so interrelated’ with those of the class that the plaintiff is a good stand-in for the class members in their absence.” Gur-Ravantab, 345 F.R.D. at 3 (quoting Gen. Tel. Co. of the Sw. v. Falcon, 457 U.S. 147, 157 n.13 (1982)).

On the flip side, “class certification is inappropriate where a putative class representative is subject to unique defenses which threaten to become the focus of the litigation.” Howard, 322 F.R.D. at 119 (quoting Baffa v. Donaldson, Lufkin & Jenrette Secs. Corp., 222 F.3d 52, 59 (2d Cir. 2000)). In most cases, “the presence of a unique defense” alone will not “destroy typicality.” Meijer, Inc. v. Warner Chilcott Holdings Co. III, 246 F.R.D. 293, 302 (D.D.C. 2007) (cleaned up). But class certification is inappropriate if a unique defense threatens to “skew the

12

focus of litigation” or preoccupy the class representative to the detriment of absent class members. Id. (cleaned up).

Like the putative class, Bratya was “net long” on BBBY. Between August 12 and 18, Bratya purchased at least 201,711 shares of BBBY while selling 128,211. See Defs.’ Supp. Mem., Ex. 2. This left Bratya with a net gain of 73,500 during the Class Period. Id. Nonetheless, Cohen argues that Bratya is atypical because it traded in options and other derivative instruments in a manner that proves it never relied on the veracity of Cohen’s alleged misstatements. See Defs.’ Opp’n at 8. But the Court views the evidence differently. Where Cohen sees Bratya betting against BBBY, the Court sees Bratya engaging in short-term hedges to raise revenue as part of an overall long strategy in the company.

Start with Bratya’s trading record. Bratya alleges that Cohen duped investors with a “pump-and-dump scheme.” SAC ¶ 31. But Cohen claims that Bratya “dumped” BBBY stock during the alleged “pump” by initiating new short positions and betting against an increase in BBBY’s share price. Defs.’ Opp’n at 2. Indeed, Cohen asserts that Bratya was “net short” on BBBY during the Class Period. Defs.’ Supp. Mem. at 10, ECF No. 128. If this were in fact the case—if Bratya were betting against BBBY when other class members were going all in—then it likely would be subject to the unique defense that it “believed [Cohen’s] statements were false,” Basic, 485 U.S. at 249, and did not rely on them in executing its trades.

But Cohen’s claim that Bratya was “net short” during the Class Period buckles under scrutiny. As evidence of Bratya’s short strategy, Cohen points to its sale of “naked” call options before and during the Class Period. For the uninitiated, a call option is a promise to sell a share at a certain price, called the “strike price.” See 3A Harold S. Bloomenthal & Samuel Wolff, Sec. & Fed. Corp. Law § 2:91 (2d ed. 1998) (defining call options). A “naked” call option is a

13

promise to sell a share that you do not own. Id. The call option buyer pays a premium for this contractual price freeze, thereby generating revenue for the seller. Id. Selling naked call options is similar to a “short” position because if the stock’s price rises above the strike price, the naked seller must cover his shares for a loss. See Miller v. Prudential Bache Sec., Inc., 884 F.2d 128, 129 (4th Cir. 1989) (explaining the risk in selling “naked” call options). That is, the naked seller must purchase the stock at a price greater than what he promised to sell it. So in this sense, a naked options seller is not necessarily betting that the stock price will fall, but that it will not rise above a set level.

David Coti conceded that Bratya sold “naked” call options. See Hr’g Tr. 181:10-17. In March 2022, following Cohen’s announcement that he bought a 9.8 percent stake in BBBY, Bratya sold call options for 265,300 shares. Hr’g Tr. 174:3-4; Defs’ Supp Mem., Ex. 2. Then on August 8th, following Cohen’s tweet that said “[a]sk not what your company can do for you–ask what you can do for your company,” SAC ¶ 139, Bratya sold call options for over two million BBBY shares. Hr’g Tr. 181:10-17; see also Defs’ Supp Mem., Ex. 2 at 2. And again, after Cohen’s “moon tweet” on August 12, Bratya sold call options for an additional 160,000 BBBY shares, while purchasing no new BBBY shares—despite the supposed enthusiasm generated by the tweet. Hr’g Tr. 187:9–11, 23. Last, on August 15, Bratya sold further call options for 280,000 BBBY shares. Defs’ Supp Mem., Ex. 2 at 2.

But Bratya’s option sales do not mean, as Cohen alleges, that Bratya was “net short” on the company. Bratya asserts that it was selling call options not to bet against BBBY but to generate premiums. Id. at 166:11-15 (“[T]he advantage of selling options is to get a premium . . . to generate cash.”). Indeed, during the Class Period, Bratya generated thousands of dollars in premiums from selling BBBY options. See Defs.’ Supp. Mem., Ex. 2.

14

In taking this position, Bratya was betting that BBBY’s stock price would not rise above the strike price. But it contends that this was not inconsistent with its long term bet in the company. At the evidentiary hearing, Coti testified that Bratya never bought or sold any BBBY security “with the hope that the price of that security would go down.” See Hr’g Tr. 162:7-10. Rather, its options trading was a hedge consistent with the view that BBBY’s price would rise— just not above the strike price. See id. at 161:11-15 (Coti explaining that “the advantage of selling options is to get a premium . . . and it’s not consuming as much collateral as selling some put, especially when you sell calls with a higher strike price than the stock quote at that moment”). In this case, Bratya’s strike price was high and the duration of its call option short—a strategy consistent with raising revenue from premiums while limiting risk. See Defs.’ Supp. Mem., Ex. 2 at 2 (showing that many call options had short-term expiration dates); see also Hr’g Tr. 162:25 (Coti testifying that the strike price for call options sold on March 7, 2022, was $30). So the Court is persuaded that Bratya’s sale of call options does not undermine Coti’s claim that Bratya was “net long” during the Class Period.

Next, Cohen argues that Bratya is atypical because it sold common stock right after Cohen’s “full cart” tweet. More, it did not buy common stock until five days later, and even then, it offloaded a bundle of shares that same day. Defs.’ Supp. Mem., Ex. 2 at 2–3.

Coti conceded this—but with a caveat. To start, Bratya sold 11,500 shares on August 12 to cover the exercise of an option, see Hr’g Tr. 156:1-5, not because it was bearish on BBBY. As to why Bratya did not buy shares on August 12, following Cohen’s “full cart” tweet, Coti explained that Bratya did not have the cash. See id. at 157:11-18. It was only on August 17 when Bratya had enough liquidity to purchase of 201,711 BBBY shares. Id. at 158:24–159:4. But Bratya bit off more than it could chew, and its bank requested that it sell shares to mitigate

15

risk. As Coti put it on the stand, “I bought too much, so I had to resell a part.” Id. at 159:20-21. These trades may seem all over the map, but Bratya is not required to invest on scripted cue any more than the other putative class members. It is enough that Bratya’s overall strategy aligns with reliance on Cohen’s commitment to BBBY. Cf. Loc. 703, I.B. of T. Grocery & Food Emps. Welfare Fund v. Regions Fin. Corp., 762 F.3d 1248, 1255 (11th Cir. 2014) (“That the [lead plaintiff] purchased its shares late in the Class Period presents no reason to consider the District Court’s finding of typicality to be an abuse of discretion.”). The Court credits Coti’s explanations on these points.

Bratya also sold naked call options in March, after Cohen announced his stake in BBBY.

This is not what one might expect from a trader excited about a possible turnaround strategy. But again, these trades do not mean that Bratya was betting that BBBY’s price would go down. And in any case, Bratya need only show that it relied on Cohen’s alleged misstatements during the Class Period—and not before. Cf. Halliburton I, 563 U.S. at 810 (explaining that the reliance inquiry focuses on whether a plaintiff was “aware of a company’s statement and engaged in a relevant transaction—e.g., purchasing common stock—based on that specific misrepresentation.”). So these pre-Class Period trades do not defeat typicality.

Last, Cohen argues that Bratya is subject to unique reliance defenses because it only bought BBBY stock after it realized that Cohen had sold his entire stake. To be sure, text message exchanges between the Coti brothers show that, on August 17, they knew that Cohen had filed his SEC Form 144, which disclosed the “potential sale of up to 7,780,000 common stock.” Pl.’s Reply re Mot. to Certify Class at 22 (Pl.’s Reply), ECF No. 119. Late that night, David sent Eduardo a SeekingAlpha article indicating that Cohen had sold his shares. Opp’n Def.’s 16. Yet the ensuing discussion reveals that the brothers were unsure what Cohen’s filing

16

meant. At one point, Eduardo suggested “it’s just the right to sell within 6 months,” and the following morning, wrote that Cohen “completed the filing to be able to sell.” Defs.’ Opp’n, Ex. 3 at 29. David speculated that Cohen’s Form 144 was a sign that Bed Bath was poised to sell buybuy BABY, after which Cohen would liquidate his position. Eduardo responded, “that is good logic.” Id. In other words, the Cotis thought that Cohen’s Form 144 disclosure did not mean he was getting out that very day, only that he was reserving the right to get out once he executed his plan—a plan from which they hoped to profit.

In short, the Court finds that Bratya’s trades during the Class Period align with a “long”

strategy on BBBY. So Bratya satisfies typicality.

4. Adequacy

Last, Rule 23(a)(4) requires a putative class representative to “fairly and adequately protect the interests of the class.” Courts apply two criteria to determine the adequacy of representation. “(1) the named representative must not have antagonistic or competing interests with the unnamed members of the class, and (2) the representative must appear able to vigorously prosecute the interests of the class through qualified counsel.” Hoyte v. District of Columbia, 325 F.R.D. 485, 490 (D.D.C. 2017) (quoting Twelve John Does v. District of Columbia, 117 F.3d 571, 575 (D.C. Cir. 1997)). The adequacy requirement is “not . . . stringent.” Id. Any conflict “must be fundamental and go to the heart of the litigation in order to preclude certification.” Id. (cleaned up).

Cohen alleges that the Coti brothers’ WhatsApp messages reveal antagonism to the retail investors who they alleged were harmed by Cohen’s statements. At various points, the brothers referred to retail investors as “ape mongols,” “barbares,” and “scum bags,” who “put[] in their minimum wage at robinhood.” Defs.’ Opp’n, Ex. 3; Ex 4 (second and third quote); Ex. 14

17

(fourth quote). But these privately shared pejoratives reveal little. At his deposition, Edouard Coti explained that meme stock investors used the term “ape mongols” to refer to themselves in jest. Defs.’ Opp’n, Ex. 43 at 48. And David Coti used the term “scum bags” to refer to investors in GameStop, not BBBY. Defs.’ Opp’n, Ex 4. Though in poor taste, these remarks hardly show the kind of antagonism that would be required to deny class certification on adequacy grounds— especially considering that Bratya’s interests are otherwise aligned with those of the class. 4 B. The Basic Presumption

To certify a class under Rule 23(b)(3), plaintiffs must show that “questions of law or fact common to class members predominate over any questions affecting only individual members.” Fed. R. Civ. P. 23(b)(3). This inquiry “tests whether proposed classes are sufficiently cohesive to warrant adjudication by representation.” Amchem Prod., Inc. v. Windsor, 521 U.S. 591, 623 (1997).

When considering whether “questions of law or fact common to class members predominate,” courts begin “with the elements of the underlying cause of action.” Erica P. John Fund, Inc. v. Halliburton Co. (“Halliburton I”), 563 U.S. 804, 809 (2011). In securities claims, a common element is whether plaintiffs relied on defendants’ material misrepresentations or omissions. And the predominance analysis in these cases “often turns on the element of reliance.” Id. at 810. This case is no exception.

4 Beyond Rule 23(a)’s requirements, some courts also require that the class be (1) “adequately defined,” and (2) “clearly ascertainable.” DL v. District of Columbia, 302 F.R.D. 1, 16 (D.D.C. 2013). Here, the class is limited to investors in BBBY stock and long call options who acquired these securities during the Class Period, excluding Defendants and Bed, Bath & Beyond officers. The Court is satisfied that these parameters are “administratively manageable,” and will enable “counsel and putative class members [to] easily ascertain whether they are members of the class.” Pigford v. Glickman, 182 F.R.D. 341, 346 (D.D.C. 1998).

18

Plaintiffs generally demonstrate reliance in securities fraud claims by “showing that [they were] aware of a company’s statement and engaged in a relevant transaction”—i.e., purchasing common stock—“based on that specific misrepresentation.” Amgen Inc., 568 U.S. at 461. But given the individualized nature of the inquiry, reliance would be a difficult common element to prove in a class action.

The Supreme Court addressed this issue in Basic, Inc. v. Levinson, 485 U.S. 224 (1988).

It observed that “[r]equiring proof of individualized reliance from each member of the proposed plaintiff class” would effectively prevent plaintiffs “from proceeding with a class action, since individual issues [would] overwhelm[] the common ones.” Id. at 242. To alleviate these reliance issues, Basic held that “securities fraud plaintiffs can in certain circumstances satisfy the reliance element . . . by invoking a rebuttable presumption of reliance, rather than proving direct reliance on a misrepresentation.” Halliburton Co. v. Erica P. John Fund, Inc. (“Halliburton II”), 573 U.S. 258, 268 (2014).

The Basic presumption, though nominally singular, in fact “incorporates two constituent presumptions.” Id. at 279 (emphasis added). The first presumption is based on the “fraud on the market” hypothesis. It posits that “the market price of shares traded on well-developed markets reflects all publicly available information, and, hence, any material misrepresentations.” Id. at 268. And the second presumption is that an “investor who buys or sells stock at the price set by the market does so in reliance on the integrity of that price.” Id. So for Basic to apply, a plaintiff must show the following: “(1) that the alleged misrepresentations were publicly known, (2) that they were material, (3) that the stock traded in an efficient market, and (4) that the plaintiff traded the stock between the time the misrepresentations were made and when the truth was revealed.” Id. Plaintiffs bear the burden on each element. See id. at 276.

19

Once plaintiffs have satisfied these requirements, a defendant can still rebut the Basic presumption by making “any showing that severs the link between the alleged misrepresentation and either the price received (or paid) by the plaintiff, or his decision to trade at a fair market price.” Id. at 268–69. “So for example, if a defendant could show that the alleged misrepresentation did not, for whatever reason, actually affect the market price, or that a plaintiff would have bought or sold the stock even had he been aware that the stock’s price was tainted by fraud, then the presumption of reliance would not apply.” Id. at 269.

Bratya argues that it has satisfied the Basic requirements, so reliance can be presumed.

Meanwhile, Cohen argues that reliance cannot be presumed because BBBY stock did not trade in an efficient market during the Class Period. And even if it did, Cohen argues that the evidence shows that his alleged misrepresentations did not sway the market price, so the presumption is rebutted. The Court addresses each argument in turn.

1. Indirect Indicators of Market Efficiency Start with market efficiency. For purposes of the Basic presumption, a stock trades in an efficient market if that market is informationally efficient. This means that “prices respond so quickly to new information that it is impossible for traders to make trading profits on the basis of that information.” In re PolyMedica Corp. Sec. Litig. (“PolyMedica II”), 432 F.3d 1, 14 (1st Cir. 2005). Determining informational efficiency therefore requires “analysis of the structure of the market and the speed with which all publicly available information is impounded in price.” PolyMedica II, 432 F.3d at 14; see also In re PolyMedica Corp. Sec. Litig. (“PolyMedica III”), 453 F. Supp. 2d 260, 272 (D. Mass. 2006) (“Underpinning the notion of information efficiency is the assumption that stock prices can reach a new equilibrium every time new information is released.”). This includes factors such as whether a stock “is actively traded, and whether it is

20

followed by analysts.” PolyMedica II, 432 F.3d at 14 (quoting Daniel R. Fischel, Efficient Capital Markets, the Crash, and the Fraud on the Market Theory, 74 Cornell L. Rev. 907, 912 (1989)).

Informational efficiency contrasts with “fundamental value efficiency,” which “focuses on the price of the stock as a function of its fundamental value.” Id. at 15. A market for a stock is fundamentally efficient if it “mirror[s] the best possible estimates, in light of all available information, of the actual economic values of securities in terms of their expected risks and returns.” Id. Determining fundamental efficiency “is a much more technical inquiry than determining informational efficiency.” Id. It involves “the assessment of various factors, including present operations, future growth rates, relative risk levels, and the future levels of interest rates.” Id. (cleaned up).

Proof of fundamental efficiency is not required under Basic. Bratya need not show that a market absorbed all information about a stock rationally, so that the stock reflects fundamental value. Rather, Bratya must show that the market reacted—rationally or otherwise—to value- relevant information. 5 The focus is on “whether a particular market has absorbed all available information . . . such that an ordinary investor cannot beat the market by taking advantage of unexploited profit opportunities.” PolyMedica II, 432 F.3d at 16.

This does not mean that evidence of fundamental value is irrelevant. See PolyMedica III, 453 F. Supp. 2d at 273 (“Th[e] Court will . . . tie itself to the mast of information efficiency, but loosen the bindings when considerations of fundamental value efficiency proves beneficial to the

5 As Fischel explained in his testimony, “The difference between [informational efficiency and fundamental value efficiency] is that informational efficiency requires that . . . prices react to new value-relevant information. And fundamental efficiency means that when the market reacts, it has to do it accurately.” Hr’g Tr. 118:8-12.

21

analysis.”). Information and fundamental efficiency often “go hand-in-hand.” Id. at 15. For example, “evidence that arbitrageurs are not trading in the market” might suggest both that the stock price does not reflect fundamental value but also that its price does “not fully reflect all publicly available information.” Id. Still, information and fundamental efficiency are distinct concepts, and a market could be “information efficient without also being fundamental value efficient.” Id. at 16.

Market efficiency under Basic is not “binary, ” but a “matter of degree,” and so “a matter of proof.” Id. Thus, to support a presumption of reliance, plaintiffs need not show a perfectly efficient market. Id.; see also Eckstein v. Balcor Film Invs., 8 F.3d 1121, 1130 (7th Cir. 1993) (Easterbrook, J.) (“‘Efficiency’ is not an all-or-nothing phenomenon.”). Nor must defendants show a wholly inefficient market to defeat it. IBEW Loc. 90 Pension Fund v. Deutsche Bank AG, 2013 WL 5815472, at *20 (S.D.N.Y. Oct. 29, 2013). “Instead, they must demonstrate that plaintiffs’ proffered proof of market efficiency falls short of the mark.” Id.

Because the market efficiency analysis often involves technical knowledge outside courts’ bailiwick, parties typically present expert reports and testimony supporting their positions. Both parties presented well qualified experts in this case. Plaintiff’s expert, Matthew Cain, has a Ph.D. in Finance from Purdue University and has held several research and teaching appointments. Cain Report ¶¶ 5–10. His research has been published in several leading journals. Id. ¶ 11. And he has provided expert analysis in several dozen cases, including on market efficiency. Id. ¶ 12. Defendants’ expert, Daniel Fischel, is a professor emeritus at the University of Chicago Law School, where he was previously Dean. Fischel Report ¶ 1. He has considerable expertise in law and economics. Id. ¶¶ 2–4. His research has also been published in leading journals, and he is coauthor, with Judge Easterbrook, of The Economic Structure of

22

Corporate Law (Harvard University Press). Id. ¶ 2. While both experts presented persuasive testimony, the Court found Fischel’s testimony particularly responsive to the unique issues here.

a. The Indirect Cammer and Krogman Factors Indicate BBBY Market was Efficient Before the Class Period

In applying Basic, courts have no mandatory rules or standards for evaluating market efficiency. See In re Petrobras Sec., 862 F.3d 250, 276 (2d Cir. 2017). But there is consensus among courts that certain market indicators are probative of informational efficiency. The indicators of efficiency most widely used by courts are the so-called “Cammer factors.” See Cammer v. Bloom, 711 F. Supp. 1264, 1286–87 (D.N.J. 1989) (articulating five factors); Binder v. Gillespie, 184 F.3d 1059, 1064 (9th Cir. 1999) (applying Cammer factors); In re Xcelera.com Sec. Litig., 430 F.3d 503, 511–18 (1st Cir. 2005) (same); Freeman v. Laventhol & Horwath, 915 F.2d 193, 199 (6th Cir. 1990) (same).

The first four Cammer factors “examine indirect indicia of market efficiency for a particular security.” In re Petrobras Sec., 862 F.3d at 276. These indirect factors include (1) the security’s weekly trading volume; (2) the number of analysts following and reporting on the security; (3) the number of “market makers” trading the security and the potential for arbitrage activity; and (4) the company’s eligibility to file simplified SEC forms. Id. The fifth Cammer factor looks to direct empirical evidence “showing a cause and effect relationship between unexpected corporate events or financial releases and an immediate response in the stock price.” Cammer, 711 F. Supp. at 1287. Most often, this direct evidence consists of “regression analyses that seek to show that the market price of the defendant’s stock tends to respond to pertinent publicly reported events.” Halliburton II, 573 U.S. at 280.

23

Beyond the Cammer factors, courts routinely consider three other indirect indicia of market efficiency: (1) the company’s market capitalization; (2) the bid-ask spread for stock sales; and (3) the public float. Krogman v. Sterritt, 202 F.R.D. 467, 478 (N.D. Tex. 2001); see also In re Petrobras Sec., 862 F.3d at 276 (noting that these factors “are commonly included in Cammer analyses”). Like the Cammer factors, these characteristics are potential indicators of whether the price of a company’s stock reflects all publicly available information.

Still, the Court is mindful that these factors are not a “mandatory analytical framework for market efficiency inquiries.” Regions Fin. Corp., 762 F.3d at 1255. Nor are they a guarantee of market efficiency. Rather, district courts must “consider the nature of the market on a case- by-case basis to decide whether the totality of the circumstances supports a finding of market efficiency.” Id.

Start with the “indirect” Cammer/Krogman factors. For Cammer, those are volume, number of analysts, market makers, and simplified SEC form eligibility. Cammer, 711 F. Supp. at 1286–87. For Krogman, those are market capitalization, bid-ask spread, and public float. Krogman, 202 F.R.D. at 478. These factors do not measure the market’s reactivity to new information directly—that falls under Cammer five, and typically involves dueling expert event studies. The indirect factors, in contrast, focus on characteristics that imply market efficiency, but cannot prove it.

In this case, there is no dispute that BBBY technically satisfies most of the indirect Cammer factors. Bratya dutifully ticks off each one. During the Class Period, the weekly trading volume for BBBY stock was 1,346.6 percent of total shares outstanding—far exceeding the 2% average that would justify a presumption of efficiency. See Cain Report ¶ 37, Ex. 2; Cammer, 711 F. Supp. at 1293. BBBY also had a high level of analyst coverage, both leading up

24

to the Class Period and during it. See Cain Report ¶¶ 40–44, Ex. 3. And at least 58 market makers “made markets” in BBBY’s common stock during the Class Period. Id. ¶ 52. Last, Bed Bath was eligible to file a short-form registration statement on Form S-3. Id. ¶ 55. This means that the company filed financial reports with the SEC for twelve months and had over $75 million in equity held by non-affiliates—further markers of a stock that generally trades efficiently. Id.

Bratya also breezes through the three Krogman factors. BBBY’s market capitalization averaged $1.46 billion over the Class Period, placing it in the 50th to 75th percentile of all companies listed on the NASDAQ and NYSE. Id. ¶¶ 79–82, Ex. 8. The bid-ask spread for BBBY stock was below the median for a set of comparable firms, which suggests low transaction costs for buying and selling the security. Id. ¶¶ 83–87, Ex. 9. And the “public float”— the number of shares held not by insiders but by the public—was more than 90 percent of all outstanding shares during the Class Period. Id. ¶¶ 88–91, Ex. 10. This means that, in theory, most of the company’s shares were available for trading.

b. Short Squeeze Dynamics Undermine Relevance of Cammer and Krogman Factors Under normal circumstances, the indirect Cammer and Krogman factors would be strong indirect evidence that BBBY operated in an efficient market. Defs.’ Opp’n at 26; Fischel Report ¶ 20 (conceding that the Cammer factors suggest BBBY was efficient at points before the Class Period). After all, these factors largely reflect the obvious. Bed Bath was a major corporation with a significant market capitalization traded on NASDAQ, a national exchange. And stocks traded on national exchanges are often “presumed to be traded on an efficient market.” Regions Fin. Corp., 762 F.3d at 1257 (quoting Thompson v. RelationServe Media, Inc., 610 F.3d 628, 693–94 (11th Cir.2010) (Tjoflat, J., concurring in part and dissenting in part)). This is because

25

national exchanges “are generally populated by stocks that are closely watched by analysts and that trade at a high volume.” Id. So on the whole, the prices of stocks traded on national exchanges are much more likely to rapidly incorporate value-relevant information. Unger v. Amedisys Inc., 401 F.3d 316, 322 (5th Cir. 2005) (“In many cases, where heavily-traded or well known stocks are the target of suits, market efficiency will not even be an issue.”).

But these were not normal circumstances. Cohen’s expert, Fischel, marshals considerable evidence showing that the market for BBBY stock was in wild flux in the weeks before and during the Class Period. Fischel’s expert analysis is credible and clarifying. But even the lay observer unfamiliar with trading concepts like “short constraints” or “the public float” can still grasp the gist of his argument: the market for BBBY securities was so volatile and subject to such unusual market dynamics during the Class Period that it cannot possibly have “reflected all public, material information” including Cohen’s alleged material misstatements.

Fischel’s overarching theory is that, just before and during the Class Period, BBBY stock was undergoing a “short squeeze,” which so distorted the stock’s price that it no longer efficiently reflected “value-relevant” information. Fischel Report ¶ 32. A short squeeze typically occurs when the price of a highly shorted stock unexpectedly rises. The more a stock price rises, the more money short sellers stand to lose. So when the price of a highly shorted stock unexpectedly rises, short investors will eventually start buying the stock to close their short position. This influx of demand drives the price even higher, triggering more short sellers to close their positions. See In re Jan. 2021 Short Squeeze Trading Litig., 76 F.4th 1335, 1343 (11th Cir. 2023) (explaining short squeeze conditions). Because of this dynamic, a short squeeze can cause share prices to rise to “seemingly crazy levels” “otherwise thought unachievable.” Fischel Report ¶ 33 n.67.

26

But the short squeeze fever is fleeting—typically lasting around 12 days from onset to peak. Id. ¶ 33. Once the squeezed short sellers close their positions and are no longer boosting demand, the normal rules of market efficiency typically pull the stock price back down. Following the peak, the price typically declines by half within a few days. Id. ¶ 33.

Fischel presents convincing evidence that BBBY was undergoing a “short squeeze”

during the Class Period. Between July 29 and August 11, BBBY stock rose from $5.03 to $10.65—an increase of 111 percent—without the revelation of any new “value relevant information.” Fischel Report ¶ 28. This price change “was far in excess of the return experienced by either the market or BBBY’s benchmark industry over the same time period.” Id. Day-to-day price fluctuations were similarly dramatic and befuddling. On August 8, for example, BBBY stock increased nearly 40 percent, again without any market-moving news. Id. ¶ 34. And by the end of the Class Period on August 18, BBBY stock had shot up more than 400 percent from its July lows. Id. ¶ 27.

Analysts watching BBBY noted that these price increases were not attributable to “any significant fundamental catalyst.” Id. ¶ 26. After all, Bed Bath had “been mired in a deep sales slump” and had no “corporate development[s] that may [have] seemingly move[d] shares.” Id. ¶ 25. Instead, market analysts pegged the rise in price to “the meme stock craze” driven by “loosely organized traders on Reddit and elsewhere on the Internet.” Id. ¶ 27. Analysts attribute varying motivations to these retail investors: some hoped to “drive up the share price in order to force hedge funds who have shorted the stock to either cut their losses and sell or buy up more shares[,] thus inflating the stock price further” and others acted out of simple “fomo” or “fear of missing out.” Id.

27

According to Fischel, the high volatility and short squeeze dynamics of BBBY stock during the Class Period render the Cammer factors largely irrelevant as indicators of market efficiency. See Hr’g Tr. 108:23-25 (noting that Cammer factors “other than Factor 5 . . . don’t . . . shed any light one way or the other” on market efficiency during a short squeeze). This is because most of the Cammer factors—analyst coverage, market makers, institutional ownership, public float, and SEC Form S-3 eligibility—are “static” indicators that say nothing about whether a market is undergoing a temporary period of inefficiency. Fischel Report ¶ 52 n.105.

The one indirect Cammer factor that is not static and that does register BBBY’s volatility is trading volume. Cain quickly concluded that volume weighed in favor of efficiency because it was 1,346.6 percent of total shares outstanding. Cain Report ¶ 37. And this “exceeds both the 1% and 2% thresholds” expected in an efficient market. Id. Cain also notes that BBBY had an average daily share turnover rate of 269.3 percent during the Class Period, compared to an average of 15.3 percent for the preceding year. Id. ¶ 39. This put BBBY above the 95th percentile for trading volume and, according to Cain, further shows that it traded in an efficient market during the Class Period. Id.

Yet this “explosion in volume during the class period” is more likely “an indicia of market inefficiency rather than efficiency.” Hr’g Tr. 109:6-10 (Fischel testimony). In general, a healthy trading volume is relevant to market efficiency because it “implies a likelihood that many investors are executing trades on the basis of newly available or disseminated corporate information.” Cammer, 711 F. Supp. at 1286. But the astronomical volume and share turnover rates during the Class Period do not signal market efficiency any more than tachycardia signals a healthy heart. A weekly trading volume of 1,346.6 percent means that, on average, each

28

outstanding share of BBBY stock changed hands over 13 times during the Class Period. Cain also buried the lede when discussing BBBY’s average daily share turnover rate. During the Class Period, that rate was 269.3 percent, which Cain noted, is in the “95th percentile” of stocks traded on the NASDAQ and NYSE. Cain Report ¶ 39. But in a footnote, Cain admits that the 95th percentile daily share turnover rate was only 2.95 percent. Id. ¶ 39 n.36. So BBBY stock was changing hands around ninety times more often than the most frequently traded stocks.

BBBY’s hyperactive trading volume seems less an indication that traders are responding to new value-relevant information than that they are reacting to (or participating in) market manipulation. As Fischel explains, short squeeze traders use coordinated buying strategies to increase a stock’s trading volume and price. Fischel Report ¶ 44. This puts pressure on short sellers who need to buy the stock to close out their positions, which, of course, increases volume further. See id. In that event, a spike in trading volume might mean that prices are “less informative” and that the transaction costs for non-short squeeze traders are high. Id. So while active trading is generally a sign of market efficiency, the “astronomical” volume of BBBY stock during the Class Period is more likely “aberrational market behavior . . . demonstrating inefficiency.” Hr’g Tr. 87:11-13 (Fischel testimony).

Bratya’s response to this argument is not compelling. At the evidentiary hearing, Cain suggested that “volume can still be relevant” for assessing informational efficiency during a short squeeze. Id. at 37:9-12. Maybe so, especially if volume is within normal bounds. But Cain does not address the concern that abnormally high volume might be a sign of market manipulation rather than a sigh of health. It is far more plausible that the high trading volume was not incorporating the market’s view of the stock, but the opposite: pumping up the stock’s price without any regard to value-relevant information.

29

In the end, the Court agrees with Fischel that BBBY was gripped by a short squeeze during the Class Period—a conclusion Bratya itself does not deny. Hr’g Tr. 8:8-9 (“[W]e’re not specifically disagreeing that there was [a short squeeze].”). And the Court cannot simply ignore these dynamics—as Bratya does—in determining whether BBBY’s market was efficient. That said, the Court rejects Cohen’s suggestion that the Cammer factors are irrelevant in analyzing market efficiency in a period of high volatility. The Cammer factors aim to measure efficiency by looking to indicators that reflect how quickly material information is reflected in the market price. And this is a distinct inquiry from whether a stock’s price is subject to other conditions (like a short squeeze) that might affect its volatility but not its sensitivity to new material information. After all, just because a stock price “may be inaccurate does not detract from the fact that false statements affect it, and cause loss,” which is “all that Basic requires.” See Halliburton II, 573 U.S. at 272 (cleaned up).

The indirect Cammer factors provide some evidence of baseline efficiency. The high level of analyst activity, the number of “market makers” trading the stock, and BBBY’s eligibility to file SEC forms all indicate that information about BBBY was readily available to public traders who could translate it into the stock’s price. But at root, “[t]he relevant inquiry . . . is not whether trading in [BBBY] stock met the Cammer factors; rather, the test is whether its stock price was information efficient.” PolyMedica III, 453 F. Supp. 2d at 271. The Cammer factors “are no more than tools in arriving at that conclusion.” Waggoner v. Barclays PLC, 875 F.3d 79, 98 (2d Cir. 2017). And as far as tools go, the indirect Cammer factors are about as useful here as a parasol in a monsoon: better than nothing but not nearly enough.

30

c. Short Sale Constraints Likely Impeded Informational Efficiency The first four Cammer factors and Krogman factors are not the only indirect evidence of market inefficiency. Courts have also looked to constraints on short selling as “indirect evidence of market inefficiency.” Smilovits v. First Solar, Inc., 295 F.R.D. 423, 434 (D. Ariz. 2013). Short sale constraints are particularly relevant here, considering the short squeeze dynamics at play during the Class Period.

Constraints on short selling undermine market efficiency by disrupting the ability of arbitrageurs from trading on their negative views of the company. Recall that the relevant inquiry for market efficiency is whether the market price for a stock “fully reflects all publicly available information.” PolyMedica II, 432 F.3d at 9. And “[o]ne way information gets absorbed into the market and reflected in stock price is through arbitrageurs.” Id. These are sophisticated investors who “obtain and analyze information about stocks from a variety of sources” and “immediately attempt to profit from such information.” Id. at 9. Through their hair-trigger trading, arbitrageurs can cause a stock “to move to a price which reflects the latest public information” such that “it is no longer possible for [ordinary investors] to generate profits.” Id.; see also Eckstein, 8 F.3d at 1129 (stating that “[c]ompetition among savvy investors leads to a price that impounds all available information”).

Arbitrageurs often attempt to profit from new information “through short sales.”

PolyMedica II, 432 F.3d at 9. Because short selling involves borrowing a stock and selling it on the open market, Fischel ¶ 29 n.58, it places downward pressure on stock prices, Hr’g Tr. 97:14- 18. In other words, if a stock price is “rising dramatically,” short selling can “operate as a brake.” Id. at. 31:14-19. Short sellers thereby “play a role in aligning prices with information

31

under . . . the efficient capital market hypothesis.” Schleicher v. Wendt, 618 F.3d 679, 685 (7th Cir. 2010).

In a well-functioning market, there are no constraints on short selling, and “the price of a stock reflects the consensus view of market participants about the value of the security.” Fischel Report ¶ 29. Constraints on arbitrageurs’ ability to short sell a stock, however, can impede efficiency since traders cannot easily incorporate their negative view of a stock into its price. Id. ¶ 30. These constrains can arise when there is a “lack of sufficient shares available to borrow” or “the costs of borrowing” are prohibitive. Id. When these constraints are binding, “the underlying stock will be over-priced, and subsequent returns will be low.” Id.

Cohen argues that the market for BBBY stock was not informationally efficient because it was subject to short sale constraints. During the Class Period, the short utilization rate was at or near 100 percent. Id. ¶ 31. This means that nearly all shares of BBBY stock that were available for short sellers to borrow already had been lent. Id.; Hr’g Tr. 31:14-19 (Cain testimony on short utilization); id. at 124:24–125:4. The short utilization rate approached 100 percent at the same time as “huge price increases [in] BBBY stock that were not explained by disclosure of value-relevant information.” Hr’g Tr. 96:15-9. This means that traders were less able to short the stock as its price was heating up.

Bratya claims Fischel’s short utilization data does not tell the whole story. In his expert report, Cain presented evidence from an alternative data source that there were hundreds of thousands of shares available to short on any given day during the Class Period. Cain Reply ¶ 24. Fischel concedes as much. Hr’g Tr. 100:7-10 (explaining that a 100 percent short utilization rate does not mean “there’s not a single share available for short selling”). But he argues that, viewed in context, Cain’s numbers prove little. Though hundreds of thousands of

32

shares sounds like a big number, it is only around one percent of the daily float for BBBY during the Class Period. See Defs.’ Sur-Reply re Mot. at 6, ECF No. 121-1. And regardless which data set is more precise, both demonstrate that “there was a very limited ability to sell short” during the Class Period, which is “characteristic . . . of a short squeeze.” Hr’g Tr. 100:20-24. Indeed, Cain’s own data source rated BBBY stock a “a 100/100 ‘squeeze score’ based on the crowded nature of the short.” Kevin P. Curran, Bed Bath & Beyond Shares Soar 30%+ Amid Suspected Short Squeeze, Seeking Alpha (Aug. 5, 2022, 11:02 AM), https://perma.cc/PNE5-TAXS.

In any case, whether there were five or 5000 shares available to short, the number must be “scaled to volume.” Hr’g Tr. 99:1. So Cain was “incorrect” to give the impression that “there was no constraint on short selling that could have any effect on prices” simply because there were some shares available to short. Hr’g Tr. 98:21–99:1. That there were around 500,000 shares available to short on any given day is unlikely to be a “constraint on the explosion of volume” seen during the Class Period. Hr’g Tr. 99:2-10.

Nor is availability of shares the only constraint on short selling. Even if there are shares to short, traders might be constrained because short selling is expensive. Traders “have to pay an interest rate” to borrow available shares. Hr’g Tr. 101:2-10. And during the Class Period, the fee to borrow shares reached 50 percent. Connor Smith, Bed Bath & Beyond Stock Is Pricey to Borrow. Short Sellers Are Doing It Anyway, Barron’s (Aug. 17, 2022). Even if this did not place an absolute limit on short selling, it still limited arbitrageurs from translating negative information into BBBY’s price.

More, when a stock’s price is highly volatile, short selling can be extremely risky. A trader might believe that a stock is in an “artificial bubble” that will burst, and prices will decline. Hr’g Tr. 101:16-19. But he does not know if that will happen “tomorrow” or “a week

33

from now.” Id. And if the traders “go[es] short and the price goes up dramatically,” he could still have to cover the share even if he is “ultimately right that the price is too high.” Id. at 101:21-25; 102:24–103:3 (Fischel explaining that to sell short, “you have to be right about the timing of when prices will fall and you have to . . . have enough capital to . . . stick it out if the prices start to rise”).

Few courts have analyzed the effect of short sale constraints on market efficiency. But those that have generally conclude that impediments to short selling “constitute[] some indirect evidence of market inefficiency and thus should be weighed against the indirect evidence of market efficiency” that plaintiffs present. Smilovits, 295 F.R.D. at 434; see also PolyMedica III, 453 F. Supp. 2d at 273 (explaining that short selling constraints inhibit arbitrage); In re Groupon, Inc. Sec. Litig., No. 12 C 2450, 2015 WL 1043321, at *9 (N.D. Ill. Mar. 5, 2015) (finding that high short selling costs during the class period “may suggest that the market for Groupon shares was not perfectly efficient.”). This Court agrees.

Bratya argues that none of these cases conclude that a high short utilization rate alone is sufficient to disprove efficiency. For example, the court in PolyMedica III concluded that a high short interest pointed to market inefficiency when the stock was subject to “put-call parity,” 453 F. Supp. 2d at 275, which Bratya claims is not present here. But Cohen need not prove that the facts here fit those cases like a glove. Given the near 100% short utilization rate, the Court is persuaded that there was at least some limitation on short sellers translating their pessimism into BBBY’s price. And—while not definitive—this weighs against Bratya’s indirect evidence showing market efficiency under the Cammer factors. See id. at 271 (finding market inefficiency based partly on “indirect evidence that short selling PolyMedica’s stock . . . was difficult” even though Cammer factors were satisfied).

34

2. Direct Indicators of Market Efficiency Indirect evidence—the first four Cammer factors—can provide a useful baseline for evaluating market efficiency. But direct “[e]vidence of a cause-and-effect relationship between unexpected news and market price . . . is the critical factor—the sine qua non of efficiency.” In re Fed. Home Loan Mortg. Corp. Sec. Litig., 281 F.R.D. 174, 182 (S.D.N.Y. 2012). In Cammer, direct evidence is the fifth factor, which courts have described as “in many ways, the most important[] Cammer factor.” In re Xcelera.com Sec. Litig., 430 F.3d at 512. Direct evidence of market efficiency “may be more critical” in situations like this one, where “the other four Cammer factors (and/or the Krogman factors) are less compelling in showing an efficient market.” Waggoner, 875 F.3d at 98.

Cain conducted two studies to prove BBBY traded in an efficient market during the Class Period. But this direct evidence does not take Bratya far. Fischel finds several loose threads in Cain’s methodology, and his results unravel. In the end, Cain’s direct evidence of market efficiency is no more conclusive than the indirect factors. So Bratya fails to demonstrate market efficiency under Basic.

As is typical in Basic cases, Cain attempted to demonstrate a cause-and-effect relationship between material information and BBBY’s stock price using an “event study.” Cain Report ¶ 60. Broadly speaking, an event study uses stock market data to predict an “expected” return for a stock on a given day, and then compares that to the actual return. The study then looks to whether days with “abnormal” returns—where the actual return goes against the model’s prediction—correspond to the release of value-relevant information. If so, the market is likely efficient.

35

A crucial element in constructing an accurate event study is selecting an appropriate analysis period. The analysis period establishes the baseline for a stock’s behavior. An analysis period that is too short might not provide an adequate sample size. And “small sample sizes may limit statistical power, meaning that only very large-impact events will be detectable.” In re Petrobras Sec., 862 F.3d at 278. On the other hand, an analysis period that is too long might not account for changes in the market that shift baseline expectations for a stock’s performance. Further complicating matters, a properly calibrated analysis period is even more essential when the class period is short. In this sense, predicting stock prices is like predicting the weather. A meteorologist might confidently predict it will snow in Washington next year. But his predictions as to whether it will snow on any given day in February are far less certain.

Cain selected a year-long Analysis Period, starting in August 2021 and going through the Class Period in August 2022. Cain Report ¶ 30. Using price data from each trading day in this period, Cain “performed a regression analysis to measure the relationship between BBBY’s stock price returns and (1) changes in market-wide factors that would be expected to impact all stocks, and (2) changes in industry-wide factors that would be expected to impact stocks in BBBY’s industry.” Id. ¶ 63. From this analysis, Cain determined that price movements in the market and industry correlate with—and thus help explain—movements in BBBY’s stock price. Id. ¶ 66.

Next, on each trading day in the Analysis Period, Cain performed a regression analysis using an estimation window of 120 trading days. Id. ¶ 64. In an event study, the estimation window is the period used to calculate the expected returns of a company before an event. It is used to establish what “normal” returns for a stock look like. An estimation window of 120

36

trading days means that, for each trading day in the Analysis Period, Cain looked to the previous 120 days to determine the expected return.

Using data from the 120-day estimation window, Cain constructed a model to predict the expected daily return of BBBY stock during the Analysis Period, controlled for the day’s market and industry returns. Id. ¶ 67. Cain then subtracted the actual return from the predicted return to get the “abnormal” return, “which represents the component of the return that is not attributable to market-wide or industry-wide movements.” Id. ¶ 68. In other words, within this model, a large abnormal return would suggest that something besides broader market trends moved the stock price that day.

Cain then used this model to assess whether large abnormal returns correlated with the release of BBBY related news. For this part, Cain focused on days when Bed Bath released its quarterly earnings reports, because these “announcements represent a potential opportunity for the public release of new value-relevant Company information to investors.” Id. ¶ 73. It is therefore expected—though not guaranteed—that the release of quarterly reports will produce abnormal returns and higher trading volume compared to normal trading days.

Cain’s analysis period included four days on which Bed Bath issued its quarterly report.

He compared the stock returns and trading volume for BBBY on these “news days” with those metrics for “no news days”—days with no Dow Jones headlines or SEC filings. Id. ¶ 75. At this point, Cain found that three out of the four “news days” caused stock movements that were statistically significant at the 95 percent level. Id. ¶ 77. In comparison, only 4.6 percent of the “no news days” correlated with statistically significant stock price movements. Id. According to Cain, “these results establish a clear cause-and-effect relationship between the release of new company-specific information and BBBY Common Stock price movements.” Id. ¶ 78.

37

In short, Cain’s study provides convincing evidence that BBBY’s stock traded in an efficient market in the year just before the Class Period. Fischel appears to agree. See Hr’g Tr. 113:19-20. But Cain’s event study does not resolve concerns that the short squeeze dynamics prevalent during the Class Period may have rendered BBBY’s market temporarily inefficient. See Defs.’ Opp’n at 26, 34–35.

First, Cain’s event study does not show that material information impacted the price of BBBY stock for any day within the Class Period. The study analyzed four “news days” on which BBBY released quarterly earnings. But the last of these “news days” occurred on June 29, 2021—a full 51 days before the Class Period began. Cain Report, Ex. 6b. And at this point, BBBY stock was not showing the same high trading volume and price volatility that it had starting in August. See Fischel Report ¶ 24 (showing correspondence between daily trading volume and stock price). So the fact that Cain was able to demonstrate that BBBY stock traded in an efficient market during an earlier period does not prove that it traded efficiently during the Class Period, especially when there is substantial evidence that the stock was undergoing a short squeeze.

Cain sought to correct for this flaw in a follow-up report. Using the same 120-day estimation window as before, Cain analyzed the effect of disclosures on BBBY’s stock price for four “Alternative News Days” that occurred during, and just following, the Class Period. Cain Reply Report ¶¶ 37, 40. These days were August 12 (Cohen’s “full cart” tweet), August 16 (Cohen’s filing of SEC Form 3 and Schedule 13D), August 18 (BBBY’s filing of SEC Form 8-K stating that it had reached a “constructive agreement” with Cohen regarding his stake in the company), and August 19 (Cohen’s filing of Form 4 and Amended Schedule 13D reflecting his complete divestment from BBBY). Cain Reply Report, Ex. 3A. For all four days, Cain found

38

abnormal price movements that were statistically significant at the 95 percent level or greater. Cain Reply Report ¶ 43. Meanwhile, Cain found that only one out of the six other days during the analysis period (the Alternative No News Days) was linked to a statistically significant abnormal price movement. Id.

At first glance, Cain’s study appears to address Fischel’s criticisms: it shows that the market for BBBY was efficient based on data points from within the Class Period. But Cain’s methods in this second study are suspect. For one, Cain’s second study did not use objective criteria in selecting these “Alternative News Days.” In his first study, Cain used earnings reports for his “news days.” And even Fischel agrees that earnings releases are an appropriate criterion for “news days” because they contain value-relevant information that would be expected to affect stock price in an efficient market. Hr’g Tr. 112:15-21. But for his new study, Cain’s “Alternative News Days” were “days when news was released regarding Defendant Cohen’s stake in BBBY.” Cain Reply Report. ¶ 38.

The Court has little confidence in this malleable definition. As Fischel explains, the very “premise of something being a news day” is that there is “news on that day that contains value- relevant information.” Hr’g Tr. 113:7-11. But Cain’s selection of news days was not based on “standard methodology.” Id. at 113:23. Indeed, at the evidentiary hearing, Cain even admitted that he did not assess whether Cohen’s August 12 tweet contained “value-relevant” information, id. at 73:16-19, even though his own report stated that the fifth Cammer factor “relates to whether a company’s stock price quickly responds to and incorporates new value-relevant information,” Cain Report ¶ 57 (emphasis added). These inconsistencies reinforce Fischel’s assessment that the criteria for Cain’s second study were “subjective and arbitrary.” Hr’g Tr. 113:7-11. Since he did not assess whether his alternative news days were value-relevant, Cain

39

had no objective reason to expect that they would cause movement in stock price. In other words, he had no hypothesis to test. So his selection of these news days, “without any analysis of the materiality of information released on those dates . . . undermines [his conclusions].” IBEW Loc. 90 Pension Fund, 2013 WL 5815472, at *21.

Cain’s second event study also erred in using a 120-day estimation window. Under normal circumstances, a 120-day estimation period would be appropriate. See Hr’g Tr. 147:5-7. Yet here, Cain used a 120-day rolling estimation period with low volatility and “to . . . project prices in a period of high volatility.” Id. at 115:8-15; Defs.’ Supp. Mem. at 23. Because Cain’s study compared price movements during the Class Period—a time of high volatility—to a baseline of expected prices derived from a period of low volatility, it is hardly surprising that it flagged the news days during the Class Period as “abnormal.” After all, these news days were abnormal relative to the previous 120 trading days. For that matter, so was at least one of Cain’s alternative non-news days during the Class Period. See Cain Reply Report, Ex. 4. But this does not prove that these news days (or no news days) were abnormal relative to the period of extreme volatility in which they occurred.

All in all, Cain’s two event studies do not help Bratya under Cammer factor five. The first event study is sound, but not particularly relevant. Indeed, Cohen does not contest that BBBY stock traded in an efficient market before the Class Period. Defs.’ Opp’n at 20; Fischel Report ¶ 20. The second study, meanwhile, is relevant, but flawed. Cain’s selection of news days and his use of a 120-day estimation period cast doubt on the accuracy of his conclusions. So Cammer factor five does not point to market efficiency.

40

3. Price Impact

Bratya has not shown that it is entitled to the Basic presumption. But even if it had, the presumption is rebuttable. “[A]ny showing that severs the link between the alleged misrepresentation and either the price received (or paid) by the plaintiff . . . will be sufficient to rebut the presumption of reliance.” Halliburton II, 573 U.S. at 268–69 (2014).

One way to rebut Basic is demonstrating that the alleged material statements lacked price impact. Id. at 282 (“[Basic] does not require courts to ignore a defendant’s direct, more salient evidence showing that the alleged misrepresentation did not actually affect the stock’s market price and, consequently, that the Basic presumption does not apply.”). To demonstrate lack of price impact, Fischel conducted his own event study that includes data from days falling within the Class Period. And based on this study, Fischel finds that Cohen’s allegedly misleading tweet had no impact on BBBY’s price.

The volatility of BBBY stock during the Class Period poses a challenge in conducting a proper event study—especially in selecting an estimation window. Recall that researchers use the estimation window to calibrate the baseline volatility of a stock’s residual returns. Fischel Report ¶ 38. When measuring price impact during a period of high volatility, a longer estimation window—say, Cain’s 120 days—risks showing false positives. Random movements in price disconnected from value-relevant information might nonetheless appear to be statistically significant when compared to a less volatile baseline. Still, researchers must select an estimation period that is long enough to provide a solid sample of the stock’s movements. Id. To balance these factors, Fischel used a 40-day estimation period, starting on August 1, 2022.

Fischel’s event study concluded that the residual return on August 12—the day of Cohen’s “full cart” tweet—was not statistically significant. Id. ¶ 39. In other words, Fischel

41

found no statistical evidence that the stock price movement on this day was distinguishable from zero once one accounts for background volatility. Id. This means that Fischel found no evidence that Cohen’s tweet artificially “pumped” BBBY’s stock price. Id. To check his work, Fischel conducted a second study using the 20 days before the Class Period and the 20 days after. And it returned consistent results. Fischel Report ¶¶ 36, 36 n.77.

Bratya offers two rebuttals to Fischel’s study. First, it argues that Fischel’s own data refutes his claim of lack of price impact. See Cain Reply Report ¶ 44. On August 19—the day the market learned that Cohen had liquidated his entire position in BBBY—the price of BBBY nosedived. And in his analysis, Fischel finds that this price drop was “abnormal” using his own metrics. Id. ¶ 48. Bratya makes this out to be a damning omission: Fischel’s evidence of a statistically significant price drop following Cohen’s exit from BBBY must mean that his earlier statements suggesting he was still “all in” affected the price.

Courts at times look to corrective disclosures to determine price impact. In some cases, “[t]he best way to determine the impact of a false statement is to observe what happens when the truth is finally disclosed and use that to work backward.” Glickenhaus & Co. v. Household Int’l, Inc., 787 F.3d 408, 415 (7th Cir. 2015). This is particularly so when plaintiffs contend that a defendant’s materially false or misleading statements propped an already inflated price. In these cases, a corrective disclosure revealing the falsehood of earlier statements is the best way to determine the impact of false statements that have long been impounded in the price. See In re Vivendi, S.A. Sec. Litig., 838 F.3d 223, 259 (2d Cir. 2016) (“[S]ecurities-fraud defendants cannot avoid liability for an alleged misstatement merely because the misstatement is not associated with an uptick in inflation.”).

42

The value of the supposed “corrective disclosure” here is less clear. Sure, Cohen’s exit from BBBY might reveal he was pessimistic about the company’s prospects back when he tweeted about the “full cart.” But that does not mean that the price collapse following Cohen’s exit is the result of the “truth” about Cohen’s “full cart” tweet finally surfacing. More likely, it was his liquidation of his substantial stake in BBBY that caused the stock to plunge. So while Fischel did not show absence of price impact for Cohen’s departure, the Court does not find that this fatally undermines Fischel’s finding of no price impact for Cohen’s alleged misstatements.

Second, Bratya argues that price impact from the August 12 tweet is visible to the naked eye. Pl.’s Reply at 17. A graph of BBBY’s price that day shows the price steadily rising from the market opening until 10:41 am, when Cohen sent out his tweet. Fischel Report ¶ 40. And it continues to rise at a steady rate before rising more dramatically around 12:30 pm. Id. Fischel claims that this steady, continuous rise does not register any price bump following Cohen’s tweet. Id. ¶ 40. To the Court’s eyes, however, the graph in fact does show a bump in BBBY’s price in the hour following the tweet, with a more dramatic price increase occurring just two hours later, between 12:30 pm and 1:15 pm. But neither Fischel nor Cain offer analysis on whether these blips are significant or noise. And having been lectured by both parties on the importance of statistical rigor, the Court will not make any conclusions about price impact by eyeballing one graph.

Finally, Cain critiques Fischel’s selection of an estimation window. Cain contends that Fischel’s first study, using the 40-day estimation window is “contaminated by the alleged misconduct” that according to Bratya “caused the increase in BBBY’s volatility during the Class Period.” Cain Reply Report ¶ 45. But anticipating this concern, Fischel conducted his second event study using an estimation period of 20 days before the Class Period and 20 days after.

43

Fischel Report ; Hr’g Tr. 117:3-6. 6 And this second study reached the same conclusion: Cohen’s alleged misstatements did not correlate with abnormal returns compared to the background volatility in the market. Fischel Report ¶ 39 n.83. Taken together, these studies provide further evidence that Cohen’s alleged misstatements did not have a statistically significant price impact.

IV. CONCLUSION

This case presents a novel question: How should courts analyze market efficiency for stocks undergoing extreme and temporary periods of volatility? Rather than face the question head on, Bratya defaults to the standard Cammer factors. Though some of these factors weigh in Bratya’s favor, Cohen undercuts their relevance at every turn. Bratya simply does not muster enough relevant evidence to trigger Basic. But even if it had, Cohen refutes the presumption by showing lack of price impact. So the putative class cannot satisfy Rule 23(b)(3)’s predominance requirement. 7 That said, nothing here should suggest that short squeezes offer a green light to market fraudsters to engage in pump-and-dump schemes without fear of class actions suits. Nor does the Court hold that Cammer is inapplicable to short squeezes. The Court merely finds that Bratya has not carried its burden to prove market efficiency on this record.

6 Cain himself has used such short estimations periods when “an abrupt shift in the volatility of stock returns,” would render a longer, 120-day estimation window an inapt comparator to the time being studied. Defs.’ Opp’n, Ex. 49 (Dr. Cain Vaxart Depo Excerpt); see also Ex 5, Cain Dep., at 136:21–138:25. 7 Cohen further argued in his Opposition that certification is improper under Rule 2(b)(3) because Bratya has not shown that damages are “susceptible of measurement across the entire class for purposes of Rule 23(b)(3).” Comcast Corp. v. Behrend, 569 U.S. 27, 33, 35 (2013). Because the Court finds that Basic does not apply, it need not determine whether Bratya has a reliable method for proving damages. But the Court expects that Bratya would face similar difficulties in establishing reliable damages calculations as it did in proving an efficient market.

44

A corresponding order will issue today.

2024.09.27

17:08:57 -04'00'

Dated: September 27, 2024 TREVOR N. McFADDEN, U.S.D.J.