Opinion
Civil Action 20-200
08-23-2022
ALLEGHENY COUNTY EMPLOYEES' RETIREMENT SYSTEM, et al. v. ENERGY TRANSFER LP, et al.
MEMORANDUM
Gerald Austin McHugh United States District Judge
I. BACKGROUND ..................................................... 2
II. LEGAL STANDARD ..................................................... 3
III. DISCUSSION ..................................................... 4
A. Predominance under Rule 23(b)(3) ..................................................... 4
1. The Rebuttable Presumption of Reliance under Basic v. Levinson ..................................................... 6
2. Defendants' Persuasive Burden to Overturn the Presumption ..................................................... 7
3. Legal Standards Relevant to the Evidence at Issue ..................................................... 8
a. The Relevant Time Frame for Assessing Price Impact ..................................................... 10
b. The Impact of the Supreme Court's decision in Goldman Sachs ..................................................... 16
c. The Remand in Halliburton II ..................................................... 17
4. The Misrepresentations and Corrective Disclosures ..................................................... 19
a. The August 2017 Front-End Misrepresentations ..................................................... 19
b. The August 2018 Corrective Disclosures ..................................................... 25
c. The October 2018 Corrective Disclosures ..................................................... 32
d. The December 2018 Corrective Disclosures ..................................................... 37
e. The August 2019 Corrective Disclosures ..................................................... 41
f. The November 2019 Corrective Disclosures ..................................................... 44
g. The December 2019 Corrective Disclosures ..................................................... 48
5. The Effect of Finding No Price Impact as to Specific Corrective Disclosures ..................................................... 49
B. The Remaining Elements of Class Certification under Rule 23(a) and Rule 23(b)(3) .................................50
1. Numerosity under Rule 23(a) ..................................................... 50
2. Commonality under Rule 23(a) ..................................................... 51
3. Typicality under Rule 23(a) ..................................................... 52
4. Adequacy under Rule 23(a) ..................................................... 52
5. Superiority under Rule 23(b)(3)..................................................... 54
IV. CONCLUSION ..................................................... 54
Plaintiff retirement systems and pension plans brought this action for themselves and on behalf of a class of shareholders in Energy Transfer LP against Energy Transfer, its subsidiary, and certain executives. Having survived a motion to dismiss, Plaintiffs now seek class certification of their claims against Defendants for publicly making statements that omitted material information and made material misrepresentations in violation of the Exchange Act. The parties have submitted extensive briefing on the motion, and a hearing was held where both parties vigorously advocated their positions before the Court. The core dispute in the motion is over the predominance element of Rule 23(b)(3) in the context of securities litigation. Plaintiffs have established that they are entitled to a presumption of reliance under Basic v. Levinson such that presumptive class members can be presumed to have relied on material public misrepresentations made by Defendants by means of a market-efficient share price of the security. After this presumption is established, the burden shifts to the Defendants to prove by the preponderance of the evidence that the alleged misrepresentations did not in fact have an impact on the price of the security. Defendants attempt to argue controlling legal principles in a way that would narrow the scope of relevant evidence and thereby erode the presumption established by Basic, but their arguments are unconvincing. Following consideration of all the relevant evidence-quantitative and qualitative alike-Defendants fail to meet their burden to rebut the presumption of market impact, with the result that Plaintiffs can show predominance. I will therefore grant Plaintiffs' motion for class certification.
I. Background
Plaintiffs' allegations center around a series of allegedly false or misleading statements and omissions made in reference to Defendants' construction of three natural gas pipelines related to the Mariner East system across Pennsylvania: the ME2, the ME2x, and the Revolution Pipeline. These statements related to the anticipated timeline of the pipeline's construction; pipeline capacity; the safety of the pipeline and its compliance with various government orders related to safety; Defendants' adherence to a code of ethics; and Defendants' general compliance with the law. After Defendants moved to dismiss, I issued a memorandum and order that reviewed in detail the various misrepresentations and omissions alleged by Plaintiffs in order to determine whether they were actionable. I dismissed statements from twenty-two paragraphs of the Complaint for various reasons, finding that the remaining statements were actionable. ECF 64, 65; 532 F.Supp.3d 189 2021 (E.D. Pa. 2021). In denying the motion to dismiss, I rejected Defendants' arguments that the corrective disclosures regarding the project timeline, the pipeline capacity, and government investigations were legally insufficient to establish loss causation for the misrepresentations alleged.
II. Legal Standard
For a class to be certified, plaintiff must satisfy the requirements of Rule 23(a), along with the requirements for a particular type of class set forth in Rule 23(b). See Johnston v. HBO Film Mgmt., Inc., 265 F.3d 178, 183 (3d Cir. 2001); Fed.R.Civ.P. 23(b). Rule 23(a) requires that (1) the class is so numerous that joinder of all members is impracticable (numerosity); (2) there are questions of law or fact common to the class (commonality); (3) the claims or defenses of the representative parties are typical of the claims or defenses of the class (typicality); and (4) the representative parties will fairly and adequately protect the interests of the class (adequacy). Id. Plaintiff seeks certification here under subsection (b)(3), which further requires that the questions of law or fact common to class members predominate over any questions affecting only individual members (predominance), and that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy (superiority). Id. Class certification requires that the Court make factual determinations and resolve factual disputes. See In re Hydrogen Peroxide Antitrust Litig., 552 F.3d 305, 320 (3d Cir. 2008), as amended (Jan. 16, 2009). “Class certification is proper only if the trial court is satisfied, after a rigorous analysis, that the prerequisites of Rule 23 are met.” Id. at 309.
III. Discussion
A. Predominance under Rule 23(b)(3)
The parties' principal dispute is whether Plaintiffs satisfy the predominance requirement. “Predominance tests whether proposed classes are sufficiently cohesive to warrant adjudication by representation, a standard far more demanding than the commonality requirement of Rule 23(a), requiring more than a common claim. Issues common to the class must predominate over individual issues. Because the nature of the evidence that will suffice to resolve a question determines whether the question is common or individual, a district court must formulate some prediction as to how specific issues will play out in order to determine whether common or individual issues predominate in a given case.” In re Hydrogen Peroxide, 552 F.3d at 310-11. “Rule 23(b)(3) requires a showing that questions common to the class predominate, not that those questions will be answered, on the merits, in favor of the class.” Amgen Inc. v. Connecticut Ret. Plans & Tr. Funds, 568 U.S. 455, 459 (2013).
The predominance inquiry in securities cases, such as this one, alleging a fraud-on-the-market theory “often turns on the element of reliance.” Erica P. John Fund, Inc. v. Halliburton Co., 563 U.S. 804, 810 (2011) (“Halliburton I”). The element of reliance here goes to whether there was a causal connection between the public misrepresentations or omissions at issue and the purchase of the security. The Supreme Court, recognizing that proving reliance in the context of investors who purchase securities on impersonal, public exchanges presents certain evidentiary hurdles, has permitted several ways for a plaintiff to establish a rebuttable presumption of reliance. The relevant doctrine here was originally set forth in Basic Inc. v. Levinson, 485 U.S. 224 (1988), and applies in class action securities cases, see, e.g., Goldman Sachs Grp., Inc. v. Arkansas Tchr. Ret. Sys., 141 S.Ct. 1951, 1957 (2021) (applying Basic).
The core Rule 10b-5 securities fraud claim requires a showing of “(1) a material misrepresentation (or omission); (2) scienter, i.e., a wrongful state of mind; (3) a connection with the purchase or sale of a security; (4) reliance, often referred to in cases involving public securities markets (fraud-on-the-market cases) as ‘transaction causation'; (5) economic loss; and (6) ‘loss causation,' i.e., a causal connection between the material misrepresentation and the loss.” In re Aetna, Inc. Sec. Litig., 617 F.3d 272, 277 (3d Cir. 2010) (cleaned up). In reviewing the case law, it is clear that much of the Supreme Court jurisprudence over the last decade dealing with securities class actions has focused on parsing how and at what stage of the litigation the elements of securities fraud must be shown, particularly who bears the burdens of production and persuasion as to the showings. This is because there is significant conceptual and evidentiary overlap as to these elements. For example, in Amgen Inc. v. Connecticut Ret. Plans & Tr. Funds, 568 U.S. 455, 465-70 (2013), the Supreme Court held that the materiality of a misrepresentation is a merits determination which does not risk individualized issues and therefore need not be proven at the class certification stage. But event studies assessing the price impact of corrective disclosures will often be used at the class certification stage as evidence of the materiality of misrepresentations, because such evidence bears upon whether Plaintiffs class-wide can be presumed to have relied on the misrepresentations. Likewise, in Halliburton I, the Supreme Court held that “loss causation” need not be proven at the class certification stage in order to establish a rebuttable presumption of reliance. “Loss causation” refers to whether the decline in the price of the security was owing to the correction of the prior misleading statement. Halliburton I, 563 U.S. at 810. Such an inquiry goes to the merits of the putative class's claims, which is separate from certification. At this stage, the issue is whether individualized outcomes predominate, and class-wide reliance remains the core focus. As stated by the Seventh Circuit in In re Allstate Corp. Sec. Litig., 966 F.3d 595, 606-610 (7th Cir. 2020):
We are obliged to follow all three cases, and we must read them together. A district court deciding whether the Basic presumption applies must consciously avoid deciding materiality and loss causation. Halliburton I and Amgen require that much. At the same time, a district court must be willing to consider evidence offered by the defense to show that the alleged misrepresentations did not actually affect the price of the securities. Halliburton II requires that. And yes, the same evidence is likely to have obvious implications for the off-limits merits issues of materiality and loss causation. Halliburton II teaches, however, that a district court may not use the overlap to refuse to consider the evidence. The court must still consider the evidence as relevant to price impact (also known as transaction causation).966 F.3d at 608. The Supreme Court later cited In re Allstate's formulation of the controlling test with approval. Goldman Sachs Grp., Inc. v. Arkansas Tchr. Ret. Sys., 141 S.Ct. 1951, 1961 n.2 (2021).
Plaintiffs also argue that a presumption of reliance is applicable under Affiliated Ute Citizens of Utah v. United States, 406 U.S. 128 (1972), applied in the securities context in Hoxworth v. Blinder, Robinson & Co., 903 F.2d 186, 202 (3d Cir. 1990). Because I find below that the Energy Transfer securities operate in an efficient market such that the presumption of reliance is applicable through Basic, I will not address whether or how Affiliated Ute is applicable here.
1. The Rebuttable Presumption of Reliance under Basic v. Levinson
“Under the fraud on the market theory of reliance, the court presumes (1) that the market price of the security actually incorporated the alleged misrepresentations, (2) that the plaintiff actually relied on the market price of the security as an indicator of its value, and (3) that the plaintiff acted reasonably in relying on the market price of the security.” Semerenko v. Cendant Corp., 223 F.3d 165, 178-79 (3d Cir. 2000). In sum, these elements represent a theory of efficient markets that assumes that all public information is incorporated into the price of a security which creates a causal link between the purchase at that price and the alleged misrepresentations. Therefore, “to invoke the presumption of reliance, plaintiffs must show they traded securities in an efficient market, and the misrepresentations at issue became public.” In re DVI, Inc. Sec. Litig., 639 F.3d 623, 633 (3d Cir. 2011), abrogated on other grounds by Amgen Inc. v. Connecticut Ret. Plans & Tr. Funds, 568 U.S. 455 (2013). There is no dispute here that the alleged misrepresentations were public, and so the burden on the Plaintiffs in this motion is to establish that Energy Transfer's securities were sold on an efficient market.
The Third Circuit, and many other federal circuit courts of appeal, have found the factors identified in Cammer v. Bloom, 711 F.Supp. 1264 (D.N.J. 1989), as instructive to determining whether the market for a security is efficient. See In re DVI, 639 F.3d at 634 n.16 (listing circuits where application of these factors has been approved). These factors include: (1) the security's average weekly trading volume; (2) significant reporting on the stock by securities analysts; (3) the existence of market makers for the security; (4) the company's eligibility to file S-3 registration statements with the SEC; and (5) a cause-and-effect relationship between unexpected corporate events or financial releases and an immediate response in the market. Cammer, 711 F.Supp. at 1285-87. Plaintiffs' expert has comprehensively addressed these factors and found that under the Cammer factors the Energy Transfer securities were sold in an efficient market. See Coffman Report ¶¶ 26-67, ECF 79-9. The Defendants concede in their brief that “evidence at the certification stage shows an efficient market,” Defs. Br. at 29, ECF 93, and Defendants' expert assumed Plaintiffs' claim of market efficiency for the purposes of her own analysis, Allen Report ¶ 22, ECF 93-3. Having thoroughly reviewed the empirical analysis of Plaintiffs' expert with respect to the Cammer factors, I am satisfied that Plaintiffs have met their initial burden under Basic to show that the Energy Transfer securities were traded in an efficient market, and that they have therefore established a rebuttable presumption of reliance.
Plaintiffs' expert identifies six other factors that support a finding of market efficiency: (1) market capitalization; (2) the bid-ask spread; (3) public float; (4) institutional ownership; (5) autocorrelation; and (6) a developed options market. Coffman Report ¶¶ 68-79. The first three of these factors were utilized by the court in Krogman v. Sterrit, 202 F.R.D. 467 (N.D. Tex. 2001), and the latter three appear to be based on economic scholarship related to efficient markets. I conclude that these factors, as analyzed by Plaintiffs' expert, also support a finding of an efficient market.
2. Defendants' Persuasive Burden to Overturn the Presumption
With Plaintiffs having established a presumption of reliance, the persuasive burden shifts to the Defendants “to prove a lack of price impact by a preponderance of the evidence.” Goldman Sachs, 141 S.Ct. at 1963. Defendants must “show that the misrepresentation in fact did not lead to a distortion of price.” Basic, 485 U.S. at 248 (emphasis added). “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, will be sufficient to rebut the presumption of reliance.” Id. However, “[t]he defendant must ‘in fact' ‘sever the link' between a misrepresentation and the price paid by the plaintiff-and a defendant's mere production of some evidence relevant to price impact would rarely accomplish that feat.” Goldman Sachs, 141 S.Ct. at 1962. Defendants attempt to “sever the link” here principally by showing that no statistically significant declines in price occurred within the window of time immediately following a corrective disclosure or that, where such declines did occur, that the corrective disclosures themselves weren't “new” news such that they would have an impact on the price in an efficient market.
3. Legal Standards Relevant to the Evidence at Issue
The core argument running through almost all the disclosures at issue is tied to the nature of the type of evidence used to determine whether the Defendants have successfully severed the link and demonstrated that there was no impact between the alleged disclosures and the security's price. Both Plaintiffs' and Defendants' economic experts use a profession-standard regression model that attempts to parse out whether specific “events” (here the corrective disclosures) were responsible for declines in the security's price. Event studies are “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. When conducting an event study, an economist “measures the movement in a stock's price after an event or public announcement, typically adjusting for the movement in the overall market and/or industry.” Allen Report ¶ 23. “The core contribution of the event study is its ability to differentiate between price fluctuations that reflect the range of typical variation for a security and a highly unusual price impact that often may reasonably be inferred from a highly unusual price movement that occurs immediately after an event and has no other potential causes.” Jill E. Fisch, Jonah B. Gelbach, & Jonathan Klick, The Logic and Limits of Event Studies in Securities Fraud Litigation, 96 Tex. L. Rev. 553, 555 (2018).
Here the parties' experts differ slightly in the industry index they use as a control variable for market movements that would generally be expected of Energy Transfer, independent of events specific to Energy Transfer. “In particular, while both models use the S&P 500 index to control for market movements, the Coffman Report model uses the S&P Oil, Gas & Consumable Fuels Index, an off-the-shelf index, to control for industry movements, while the Allen Report model uses a custom index of the peer companies discussed in Energy Transfer's SEC filings.” Allen Report ¶ 25. This choice only appears to create nominal differences in the significance values obtained by the parties. However, I note the choice because I do utilize the S&P Oil, Gas & Consumable Fuels Index in calculations below, as it is publicly available and I was therefore able to take judicial notice of it for the purposes of a minor detail of factual analysis. In re Merck & Co. Sec. Litig., 432 F.3d 261, 264 n.3 (3d Cir. 2005) (noting that the Court “can take judicial notice of [publicly traded] stock prices”).
The regression analysis utilized in an event study tests a null hypothesis. That is, it attempts to disprove the assumption that an event did not have an impact on the security price, such that a “stock's return should be the expected return [based on market conditions], together with normal variation. ... Statistical significance testing focuses on whether this deviation-the actual excess return on the event date-is highly unusual.” Id. at 574. Significance is “calculated as the excess return divided by the standard error of the regression over the sample period.” Allen Report ¶ 28 n.1. The parties here do not dispute that it is appropriate to use a 95% confidence interval in determining whether the null hypothesis is disproven for any given event study. The key disputes focus instead on how long a time period following an event the Court should consider in determining if there was a price impact, taking into account the significance of intervening news or events that might either confound the market, and what is reasonably necessary to comprehend the significance of a disclosure.
Although the 95% confidence interval is commonly used by courts analyzing price impact in the context of class certification reliance determinations, some courts have accepted a 90% confidence interval. That is not at issue here.
a. The Relevant Time Frame for Assessing Price Impact
The first issue is the proper window within which to consider price impact. Defendants argue that price impact under an efficient market presumption must take place almost instantaneously with the disclosure of the new information to the market. Defendants maintain that if they can show that any decline in price following a disclosure took place a day or two later, rather than on the same day, then they have met their burden of severing the link between the disclosure and a price impact. The second issue is the statistical significance of any price impact, and the persuasive value of the presence or absence of a statistically significant event study within the relevant window. Defendants argue that Plaintiffs must be able to show a statistically significant price decline within a same-day or next-day window. Plaintiffs respond that market impact can be shown by statistically significant declines that occur more than a day after an initial disclosure.
Although I agree that temporal proximity is an important factor in analyzing price impact under an efficient market presumption, that one factor cannot be as rigidly applied as the Defendants argue for throughout their papers in light of the actual mechanisms of market clearing required to establish the “true” price of a security. At the same time, where Plaintiffs seek to extend the window, they must show a sound conceptual and evidentiary basis for that extension. The scholarly literature allows for such variability. See, e.g., Mark Grinblatt & Sheridan Titman, Financial Markets and Corporate Strategy 679 (2d ed. 2002) (“With efficient markets, one expects to see only insignificant returns after the announcements. However, as discussed below, there is evidence that the market underreacts to some information events, and, consequently, some researchers also analyze returns on the days following the event.”)
Such an approach is consistent with the controlling precedent. In Halliburton II, the Supreme Court held that “[t]o recognize the presumption of reliance [is] not conclusively to adopt any particular theory of how quickly and completely publicly available information is reflected in market price.” 573 U.S. at 271-72 (2014) (quoting Basic, 485 U.S. at 248 n.28). The Third Circuit has also recognized a need for flexibility. In In re Merck & Co. Sec. Litig., 432 F.3d 261, 269 (3d Cir. 2005), it referred to “the period immediately following disclosure,” but added “this does not mean instantaneously of course.” Discussing the first prong of the Basic presumption of market efficiency in In re DVI, 639 F.3d at 635, it noted that “[w]e have addressed the speed with which information is incorporated into market price and explained that because a perfectly efficient market is not attainable, we do not require that public information be absorbed ‘instantaneously.'” There, the Court of Appeals further observed “[t]hat some information took two days to affect the price does not undermine a finding of efficiency.” Id. at 635. Similarly, in Merck the Third Circuit cited price increases over a period of five days after a disclosure as relevant to whether there was a price impact. 432 F.3d at 269-70.
The most recent case from this district concluded that that a two-day window was relevant. Pelletier v. Endo Int'l PLC, 338 F.R.D. 446, 486-87 (E.D. Pa. 2021), stating that “[w]hile most public information should be absorbed into an efficient market quickly, the related price impact may occur more slowly where clarifying or contextualizing information is disclosed later.” Id. at 486; see also In re EQT Corp. Sec. Litig., No. 19-cv-754, 2022 WL 3293518, at *18 (W.D. Pa. Aug. 11, 2022) (using a two-day window for a disclosure that “involved the filing of voluminous proxy materials after the market closed on June 17 and also an eventual press release from Rice and a denial by EQT on June 18, 2019”). Other courts applying the Third Circuit's standard to materiality determinations at the motion to dismiss stage have similarly permitted two-, three-, and even five-day windows to assess whether a disclosure had an impact on a security's price. See SEB Inv. Mgmt. AB v. Endo Int'l, PLC, 351 F.Supp.3d 874 (E.D. Pa. 2018) (considering two- and three-day windows); In re Merck & Co., Inc. Vytorin/Zetia Sec. Litig., No. CIV.A.08-CV-2177, 2009 WL 2855601 (D.N.J. Sept. 2, 2009) (considering three- and five-day windows).
Courts in other circuits have also allowed plaintiffs to show price impact beyond a one-day or same-day window. See, e.g., In re Xcelera.com Sec. Litig., 430 F.3d 503, 513 n.11 (1st Cir. 2005) (rejecting argument that a two-day event window is inconsistent with an efficient market); Dougherty v. Esperion Therapeutics, Inc., No. 16-10089, 2020 WL 6793326 (E.D. Mich. Nov. 19, 2020) (using two-day event window); Monroe Cnty. Employees' Ret. Sys. v. S. Co., 332 F.R.D. 370 (N.D.Ga. 2019) (using both one- and two-day event windows); In re Diamond Foods, Inc., 295 F.R.D. 240, 249 (N.D. Cal. 2013) (citing defense expert's assertion that a “proper test of market efficiency” required analyzing whether the market price reacted to new information over more than one trading day in order to “assess[ ] whether there are delayed price responses to the events of interest”); Carpenters Pension Tr. Fund of St. Louis v. Barclays PLC, 310 F.R.D. 69 (S.D.N.Y. 2015) (finding that “[a] two-to three-day window is common in event studies”); In re Vivendi Universal, S.A., Sec. Litig., 634 F.Supp.2d 352, 372 (S.D.N.Y. 2009) (using a three-day window for analysis); Fogarazzo v. Lehman Bros., 263 F.R.D. 90, 104 (S.D.N.Y. 2009) (accepting event study with three-day event window and rejecting argument that this rule only applied “where the timing of discrete events was [not] ascertainable”); Aranaz v. Catalyst Pharm. Partners Inc., 302 F.R.D. 657, 669 (S.D. Fla. 2014) (citing significant two-day stock decline as “strong empirical evidence of market efficiency”).
These decisions acknowledge the difference between the hypothesis of efficient markets and the real world. As the Third Circuit recognizes, real world markets are not perfect but only “nearly perfect.” See Peil v. Speiser, 806 F.2d 1154, 1161 n.10 (3d Cir. 1986) (“The ‘fraud on the market' theory rests on the assumption that there is a nearly perfect market in information, and that the market price of stock reacts to and reflects the available information.”) (emphasis added). First, to operate efficiently, real world markets must include participants who are able to quickly quantify the impact that specific information will have on the price of a security. See In re Merck, 432 F.3d at 270-71 (discussing calculations necessary to quantify disclosures). Second, the security must be subject to a sufficient number of transactions such that, through the push and pull of competing valuations of buyers and sellers, an “efficient” market clearing price can be established that is understood to reflect all the information available about the security. See Cammer, 711 F.Supp. at 1276 at n.17 (quoting 5 Bromberg & Lowenfels on Securities Fraud § 7:484 (2d ed.)). The efficient price is not set by an invisible hand that instantly reflects new information in a security's price, but through the dynamic, high-volume exchange of a security over an appropriate window of time.
Defendants' expert notes that “most of the stock price response to company announcements of earnings or dividends occurs within five to ten minutes of the announcement.” Allen Report ¶ 22 (citing Zvi Bodie, et al., Investments (10th ed. 2014)). But as quantitative information coming directly from the company, earnings and dividends announcements represent the kind of objective “new information” which can, with minimal or no human interpretation, be processed by brokers and-especially-trading algorithms into large-scale immediate, and price redefining transactions. As discussed in detail below, the nature and timing of the disclosures here, the lack of transparency, the attempts by Energy Transfer to cast damaging information in a positive light, and the obvious confusion in the market at certain points require a more precise and nuanced analysis of when the market would have absorbed relevant information.
The statistical significance of a given event study is another relevant factor in assessing whether price impact can be severed for a given disclosure. Thus, as will be discussed below, in the four corrective disclosures where I agree there was price impact, I find it appropriate to consider that the Plaintiffs' expert found statistical significance in the relevant window, which goes largely unchallenged by the Defendants' expert. Coffman Rebuttal ¶ 48, ECF 97-2. But it should be noted that even a lack of statistical significance is not invariably fatal to a plaintiff's case. Event studies have an inherent rate of error, and where the statistical evidence does not disprove a null value to the requisite confidence level, numerous courts have recognized that as a matter of logic such absence of proof is not proof of absence. “In recognition of this basic truism of statistics, courts routinely reject the argument that a non-statistically significant stock price decline proves an absence of price impact.” Monroe Cnty., 332 F.R.D. at 394; see In re EQT Corp. Sec. Litig., 2022 WL 3293518, at *16 (noting that failure of both parties' experts “to reach a statistically significant conclusion as to price impact is not a basis to deny class certification.”); Rooney v. EZCORP, Inc., 330 F.R.D. 439, 450 (W.D. Tex. 2019) (“The converse, however, is not true - the absence of a statistically significant price adjustment does not show the stock price was unaffected by the misrepresentation.”); City of Sterling Heights Gen. Emps.' Ret. Sys. v. Prudential Fin., Inc., No. 12-5275, 2015 WL 5097883, at *13 n.8 (D.N.J. Aug. 31, 2015) (“[I]t also does not necessarily follow from the mere absence of a statistically significant change in the stock price that there was no price impact.”); West Palm Beach Police Pension Fund v. DFC Global Corp., No. 13-6731, 2016 WL 4138613, *14 (E.D. Pa. Aug. 4, 2016) (quoting City of Sterling Heights verbatim); Pirnik v. Fiat Chrysler Autos., N.V., 327 F.R.D. 38, 46 (S.D.N.Y. 2018) (holding that non-statistically significant stock price decline following corrective disclosure “does not prove the absence of price impact”) (emphasis in original); Carpenters Pension Tr. Fund of St. Louis v. Barclays PLC, 310 F.R.D. 69, at 95 (S.D.N.Y. 2015) (“The failure of an event study to find price movement does not prove lack of price impact with scientific certainty.”); Bing Li v. Aeterna Zentaris, Inc., 324 F.R.D. 331 (D.N.J. 2018) (same), aff'd sub nom. Vizirgianakis v. Aeterna Zentaris, Inc., 775 Fed.Appx. 51 (3d Cir. 2019); see also Pelletier v. Endo Int'l PLC, 338 F.R.D. 446, 486 (E.D. Pa. 2021) (“And viewing the totality of their evidence regarding the two-day window, Defendants' lack of a statistically significant finding cannot rebut the Basic presumption.”) Given this precedent, Defendants' emphasis on a lack of statistical significance immediately after disclosure has even less force. This is borne out by Goldman Sachs, discussed in more detail below, where the Supreme Court stated: “In assessing price impact at class certification, courts should be open to all probative evidence on that question-qualitative as well as quantitative-aided by a good dose of common sense.” Id. at 1960 (cleaned up).
At oral argument, Defendants attached great significance to language from an Eleventh Circuit case, Meyer v. Greene, 710 F.3d 1189 (11th Cir. 2013), interpreting it as limiting the window of time within which events can be considered without undermining the presumption of an efficient market. Specifically, the Meyer court stated: “A plaintiff in the Investors' situation must take the bitter with the sweet, and if he chooses to embrace the efficient market theory for purposes of proving one element of a § 10(b) claim, he cannot then turn around and contend that the market is not efficient for purposes of proving another element of the very same claim.” Id. at 1199. But I do not see an irreconcilable inconsistency between the efficient markets hypothesis and extending the relevant window where there is evidence to justify such an extension. As discussed below, a determination of the relevant window depends upon the specific evidence as to each disclosure.
It is far from clear what the Meyer court meant by this statement. By way of context, the Court was addressing the fact that seemingly quantitative information in a power point presentation on which the plaintiffs placed great significance was already known to the market from public sources. I do not see how that addresses what constitutes the appropriate window within which to consider the impact of disclosures.
b. The Impact of the Supreme Court's decision in Goldman Sachs
Besides arguing that Plaintiffs must demonstrate a statistically significant price impact within a narrow window, Defendants also argue that the Supreme Court's recent decision in Goldman Sachs has in some respect modified the Basic presumption and shifted the burden of persuasion back to the Plaintiff. Defendants focus upon a single paragraph near the end of Goldman Sachs to suggest that Goldman Sachs changed the law:
The [Goldman Sachs] Court noted that, at class certification, “[t]he district court's task is simply to assess all the evidence of price impact-direct and indirect-and determine whether it is more likely than not that the alleged misrepresentations had a price impact.” . Thus, while the burden of persuasion technically lies on Defendants, the Supreme Court took pains to “emphasize” that the location of this burden “should rarely be outcome determinative.” Id. at 1958. Indeed, the burden of persuasion “will have bite only when the court finds the evidence in equipoise-a situation that should rarely arise.” Id. at 196[3].
Defs. Sur-reply at 3 (emphasis in original). They characterize this language as confirming a descriptive footnote from the Second Circuit opinion that was under review, which merely provides a background explanation of what an “event study” is and seem to argue from this that unless a plaintiff can show an immediate price change that is statistically significant then the Basic presumption becomes irrelevant. Id. at 4 (quoting Arkansas Teachers Ret. System v. Goldman Sachs Group, Inc., 879 F.3d 474, 481 n.5 (2d Cir. 2018)). As the briefing proceeds, it becomes clear that Defendants' strategic purpose in trying to construe Goldman Sachs as breaking new ground is to dismiss the substantial body of case law allowing multi-day windows on the ground that it is irrelevant “pre-Goldman” precedent. See, e.g., Defs. Br. at 20 n.21; Defs. Sur-reply at 23, 19.
The language on which Defendants rely is plainly dicta addressed to the dissent. The holding in Goldman Sachs does not undermine but rather reaffirms the Basic standard, and, if anything, supports a broad view of the record. The Goldman Sachs Court specifically noted that the “best” reading of its precedents is that “the defendant bears the burden of persuasion to prove a lack of price impact,” “mere production of evidence” is not sufficient to rebut the presumption, and “defendant must carry that burden by a preponderance of the evidence.” Id. Indeed, the Court reiterated that its holding in Halliburton II did not reduce the defendant's burden under Basic into a burden of production of some evidence relevant to price impact, with the plaintiff assigned the burden of persuasion. “If, as [the dissent] urge[s], the defendant could defeat Basic's presumption by introducing any competent evidence of a lack of price impact the plaintiff would end up with the burden of directly proving price impact in almost every case. And that would be nearly indistinguishable from the regime that Halliburton II rejected.” 141 S.Ct. at 1963.
The continued importance of the Basic presumption is reflected by Justice Gorsuch's dissent: “The Court's suggestion that the burden of persuasion will ‘rarely' make a ‘difference' misses the point too. The whole reason we allocate the burden of persuasion is to resolve close cases by providing a tie breaker where the burden does make a difference.” Id. at 1970 (Gorsuch, J., dissenting) (emphasis in original).
The rule re-affirmed by Goldman Sachs is that “[t]he district court's task is simply to assess all the evidence of price impact-direct and indirect-and determine whether it is more likely than not that the alleged misrepresentations had a price impact.” 141 S.Ct. at 1963. Under this rule, statistical evidence must be considered in combination with other forms of evidence. “In assessing price impact at class certification, courts should be open to all probative evidence on that question-qualitative as well as quantitative-aided by a good dose of common sense.” Id. at 1960 (cleaned up).
c. The Remand in Halliburton II
Defendants rely heavily on the opinion of the district court following remand from the Supreme Court's decision in Halliburton II. Erica P. John Fund, Inc. v. Halliburton Co., 309 F.R.D. 251 (N.D. Tex. 2015) (“Halliburton II Remand”). As an initial matter, I note that a district court's interpretation of an appellate decision following remand is not entitled to special weight; it is persuasive authority like any other decision from a district court. The remand opinion accurately concludes that it remains the defendant's burden to rebut the Basic presumption of reliance, but it is less clear to me that the district court always followed that standard. For example, the court's statement that “Halliburton must ultimately persuade the Court that its expert's event studies are more probative of price impact than the Fund's expert's event studies,” id. at 260, appears somewhat inconsistent with the defendant's burden as recognized by the Supreme Court. And the remand court appears to deem statistical evidence as preeminent by stating that it was “necessary” for the plaintiff to prove price impact at ¶ 95% confidence level, id. at 270, when numerous district courts have held otherwise, and Goldman Sachs specifically recognized both “qualitative” and “quantitative” evidence as pertinent. Finally, in justifying this approach to statistical evidence, the court cited a specific law review article, Merritt B. Fox, Halliburton II: It All Depends on What Defendants Need to Show to Establish No Impact on Price, 70 Bus. Law. 437, 442 n.17 (2015)), but the supporting precedent relied on in that article consists of a series of cases addressing a plaintiff's burden to establish loss causation at summary judgment. Id. at 262.Yet in Halliburton I, the Supreme Court expressly ruled that plaintiffs do not need to prove loss causation at the class certification stage, 563 U.S. at 815, and although Halliburton II required district courts to consider defendants' evidence of price impact at the class certification stage, it did not advance the stage at which loss causation should be considered, 573 U.S. at 279-280.
One case cited by the Fox article considered loss causation at the class certification stage, Fener v. Operating Eng'rs Constr. Indus. & Miscellaneous Pension Fund (Local 66), 579 F.3d 401, 409 (5th Cir. 2009), but it predated Halliburton I and its continued validity is doubtful.
Factually, the conclusion of the remand court in Halliburton depended to a great extent on its narrowing of the relevant window to the period immediately after the corrective disclosures at issue. As stated above, in my view the appropriate window is a fact-specific inquiry, and in conducting my analysis I proceed from that premise.
4. The Price Impact of the Misrepresentations and Corrective Disclosures
Having set out the relevant legal standards the must be applied to the predominance inquiry under Basic's rebuttable presumption, I will now address Defendant's efforts to rebut the presumption by proving the lack of price impact as to one inflationary front-end misrepresentation and six corrective disclosures.
a. The August 2017 Front-End Misrepresentations
Defendants identify and statistically analyze twenty-six alleged misrepresentations that, according to Plaintiffs, are later corrected by disclosures that reveal the truth. In a fraud-on-the-market theory of legal liability, the misrepresentations either inflate or maintain the overvaluation of the security price until the corrective disclosure pierces the fraudulent bubble and allows the security to assume its “true” price. Because of this interlinking of misrepresentation and disclosure, Plaintiffs must establish predominance through reliance on the front-end misrepresentation as demonstrated by the existence of a back-end corrective disclosure. Where Plaintiffs only allege that misrepresentations maintained the price of a stock, they do not need to make any showing as to front-end price impact at the time of the misrepresentation because the price impact can be measured solely by the decline in price at the time of the corrective disclosure. See Glickenhaus & Co. v. Household Int'l, Inc., 787 F.3d 408, 418-419 (7th Cir. 2015); FindWhat Inv. Grp. v. FindWhat.com, 658 F.3d 1282, 1314 (11th Cir. 2011) (“We therefore hold that confirmatory information that wrongfully prolongs a period of inflation-even without increasing the level of inflation-may be actionable under the securities laws.”) (emphasis in original). Defendants concede this point. “Thus, the Court need only consider ‘back-end' price impact ... and need not consider ‘front-end' price impact in its analysis.” Defs. Br. at 24.
Plaintiffs nonetheless contend that one misrepresentation, that of August 9, 2017, regarding the expected completion dates of the pipeline projects to be in the fourth quarter of 2017, did not just maintain the price but actually increased and fraudulently inflated the price of the security.Coffman Rebuttal ¶¶ 58-59, ECF 97-2. Although Defendants acknowledge that there was a statistically significant price increase on that date, which Defendants' own expert identifies as a 5.69% excess return, Defs. Br. at 25; Allen Report ¶¶ 24, 28, 33, they argue that (1) the information contained in the misrepresentation wasn't new information, (2) that the report may actually have been more negative than prior representations anticipating third quarter 2017 as the completion date, and (3) that the price increase was actually owing to a contemporaneous, better-than-expected earnings announcement that also included investor-applauded representations that Energy Transfer would not pursue major acquisitions in the short term, Defs. Br. at 26-27.
Defendants also express concern that Plaintiffs seek to allege some inflationary impact from an August 2019 statement. Defs. Br. at 27-29. Plaintiffs have not responded, and I therefore consider August 2017 as the only instance where Plaintiffs contend that the stock price was inflated.
Defendants point out that, as to the Revolution Pipeline, Energy Transfer had provided the same timeline three months earlier, disclosing on May 4, 2017, that the pipeline was “still on schedule to be in service in the fourth quarter of 2017.” Compl. ¶ 334; see also id. at ¶¶ 336, 338 They further point out that Energy Transfer reiterated this timeline during May 31, 2017, and June 27, 2017, investor conference calls. They also contend that the August 9, 2017, announcement actually represented a delay in Mariner East 2's completion schedule, because it stated that it would be online “sometime in the fourth quarter” of 2017,” Compl. ¶ 343, whereas in prior public statements, Energy Transfer had stated that it “continue[d] to target an end of the third quarter completion” for Mariner East 2. See Energy Transfer Q1 2017 Earnings Conference Call (May 4, 2017), ECF 93-30; see also Compl. ¶¶ 336, 338. Plaintiffs respond by arguing that even if Energy Transfer simply reiterated a pipeline's timeline, it was new information to the extent that it represented to the market that, as of August 9, 2017, Energy Transfer would still supposedly meet that timeline. And, further, that the modified timeline for the ME2 pipeline of being in-service by “sometime” in the fourth quarter of 2017 instead of an “end of the third quarter completion” deadline was not significantly different, given that the third quarter and the fourth quarter are separated by only a single day.
Defendants have not persuaded me that updates to timelines have no meaning to the market. In general, updates and estimates for project timelines are almost always new information, especially when spread out over months, because they translate the complex progress of a project in the interval time into a simple, quantifiable date value. If a project timeline remains stable between two updates, the speaker communicates by implication that physical progress on the project has advanced as previously anticipated. If a project timeline is moved forward or backward, the speaker communicates, respectively, that progress on the ground has occurred at a faster or slower rate than anticipated. Both reports are therefore “new” news for purposes of assessing price impact. Defendants' argument as to the ME2 projection being a pessimistic report compared to previous projections is more compelling. And it raises a question, which I will address below, as to whether this disclosure of some degree of delay can support the 5.69% price inflation that Plaintiffs argue for here. But a partial disclosure of negative information does not necessarily relieve a company of liability if the disclosure to the public was still more optimistic than the truth privately known by the Defendants. It is the gap between the representation and the truth that is relevant to price impact, not a gap between two purported misrepresentations.
Here Plaintiffs argue that the importance assigned to the information about the projection completion deadlines by Energy Transfer executives and third-party analysts demonstrates both that it is material and that it had an impact on price. The Third Circuit has turned to the common law understanding of materiality when addressing misrepresentations in a securities context, looking to whether “a reasonable investor might have considered them important in the making of [their] decision.” Rochez Bros. 491 F.2d at 410; see also United States v. Saybolt, 577 F.3d 195, 202-03 (3d Cir. 2009) (addressing materiality in context of representations made to government agencies). At this juncture, it is important to restate that class certification is not adjudication on the merits. In acknowledging that “materiality and price impact are overlapping concepts,” the Supreme Court cautioned that a “district court must use the evidence to decide the price impact issue while resisting the temptation to draw what may be obvious inferences for the closely related issues that must be left for the merits, including materiality.” Goldman Sachs, 141 S.Ct. 1961 n.2. Thus, I address materiality here only insofar as it represents evidence of price impact and do not make any determinations as to a finding on the merits where the burden of persuasion will return to Plaintiffs. See In re Allstate Corp. Sec. Litig., 966 F.3d 595, 608 (7th Cir. 2020) (“The court must still consider the evidence as relevant to price impact.”).
Energy Transfer's investor conference call on August 9, 2017, began with an update on the Company's “growth projects,” including the Revolution and ME2 pipelines. Energy Transfer Q2 2017 Earnings Conference Call (Aug. 14, 2017), ECF 97-7. During the call, Kelcy Warren, Chairman and CEO of Energy Transfer's general partner, emphasized that Energy Transfer was focused on bringing its own pipelines into service by specific dates, and that Energy Transfer was focused on getting “these projects online” for the purpose of “creating distributable cash flow for our unitholders.” Id. at 6. During his initial remarks, Thomas Long, CFO of Energy Transfer's general partner, also stressed that, “[o]n our Revolution project, construction is scheduled to be completed in the fourth quarter of 2017.” Id. at 2.
Michael Jacob Blum from Wells Fargo asked, “First question I wanted to ask just back on Mariner East 2. Can you give us a sense of kind of #1, your latest estimate on timing for getting that project into service?”; and, “[D]o you have any update on the Revolution project?” Id. at 67. Brian Joshua Zarahn from Mizuho Securities inquired, “And then my follow-up on Mariner East 2. Where do you stand?” Id. at 5. In responding to those questions, Michael Ramsey, COO of Energy Transfer's general partner, stated, “Well, as we said in our opening statements, we do expect to have [ME2] in service sometime in the fourth quarter;” and Marshall McCrea, President and CCO of Energy Transfer's general partner, said, “[t]he only update is, [Revolution] will be up and ready for service in the fourth quarter.” Id.
The analyst reports issued after the Company's August 9, 2017, earnings call further show that securities analysts viewed the Company's timeline for bringing its growth projects online as important to investors. On August 9, 2017, analysts from Credit Suisse reported that, at Energy Transfer, “Near-term focus remains on resolving regulatory issues and completion of ETP's growth project backlog to drive EBITDA expansion.” Credit Suisse Report (August 9, 2017), ECF 97-9. Credit Suisse also wrote that Energy Transfer had a “clear path” to deleveraging through 2019 that was supported by “EBITDA from growth projects being placed in-service.” Separately on August 9, 2017, under the heading, “DAPL/Rover/Revolution/ME2 execution in sight,” a Morgan Stanley analyst report referred to the fact that “Management indicated that Revolution remains on track to be completed in [4th quarter 2017],” specifically mentioning the statements Plaintiffs allege were false. Morgan Stanley Report (Aug. 9, 2017), ECF 97-10. It also stated that Energy Transfer “[r]esults were ahead of MSe [Morgan Stanley] and consensus as ETP/ETE execute on a large organic backlog” and “[w]e see the catalyst path over the next 6 months as constructive for the complex.” Id. Then, on August 11, 2017, Wells Fargo reported positively on the near completion of the pipelines, including in the very title of the report, “ETE/ETP, Q2 Beat -- Projects Nearly Complete -- Positive.” Wells Fargo Report (Aug. 11, 2017), ECF 97-11. In short, completion sooner rather than later, even on a schedule moderately delayed as compared to prior projections, was of contemporaneous interest to the market, at the same time as a statistically significant increase in the price of the stock.
Defendants argue that the increase was the product of a better-than-expected earnings report, and the announcement that the company would not pursue major acquisitions. Allen Report ¶¶ 34-36. Plaintiffs reply that Ms. Allen failed to disaggregate potentially confounding factors such as the earnings announcement in her analysis. Coffman Rebuttal ¶ 63. See In re Allergan PLC Securities Litigation, 2021 WL 4077942, *10 (S.D.N.Y. Sept. 8, 2021) (“[M]erely suggesting that another factor also contributed to an impact on a security's price does not establish that the fraudulent conduct complained of did not also impact the price of the security.”). Plaintiffs also argue that the promised moratorium on mergers favors the finding of reliance related to the pipelines because, according to analysts, it correlated with Energy Transfer's focus “on project execution and guiding [Energy Transfer's] sizable project backlog to completion rather than M&A.” Wells Fargo 2017 Report at 11. This is where the overlap between merits issues and certification issues asserts itself. My role as to factual determinations at class certification is to decide whether the Plaintiffs have met their burden in establishing the elements of Rule 23. In re Hydrogen Peroxide, 552 F.3d at 309. I am not persuaded that Defendants have carried their burden to sever the link between the alleged misrepresentation and a potentially inflationary impact on price, and I conclude that a more fulsome analysis of the August 2017 front-end representations is an issue for resolution on the merits as to the loss causation.
b. The August 2018 Corrective Disclosures
Turning now to the corrective disclosures, in August 2018, Plaintiffs allege that Defendants disclosed that they would only be able to meet construction deadlines for the ME2 Pipeline by temporarily utilizing old, already-in-use 12-inch pipelines for a segment of the pipeline instead of the planned-for, new 20-inch pipelines. Compl. ¶¶ 401-409. The disclosure in question was made on an earnings call with investors on Thursday, August 9, where CFO Thomas Long announced the use of the 12-inch pipelines without specifying how long the stopgap, 12-inch pipeline would be used or quantifying the expected reduction in throughput capacity. Earnings Call Transcript (Aug. 9, 2018), ECF 97-8. In response to follow-up questions, CCO Mackie McCrea confirmed that they “haven't shared capacities,” only that the “12-inch more than provides the necessary capacity to move the volumes that we've contracted.” Id. The next day, Friday, August 10, two third-party market analysts published reports that revised the anticipated earnings of the company in light of the disclosures. Wolfe Report (Aug. 10, 2018), ECF 99-15; Wells Fargo Report (Aug. 10, 2018), ECF 99-16. By end of day on Monday, August 13, the share price of Energy Transfer had declined 5.6% from its close on Wednesday, August 8, before the disclosures were made, which Plaintiff's expert avers is an abnormal return that is statistically significant to the 95% confidence level. Coffman Rebuttal ¶ 48, ECF 99-1. Defendants' principal argument to sever the link between the disclosure and the share price focuses on timing. First, Defendants argue that there were not statistically significant decreases in the price of the security on either Thursday, August 9 or Friday, August 10, and that the gap between earnings call in the morning of August 9 and the statistically significant share price drop on Monday, August 13 is too distant to support a finding of price impact under efficient market conditions. Separately, Defendants argue that the use of the 12-inch pipe had already been disclosed a month earlier, on July 3, 2018, as pleaded in Plaintiffs' Complaint, Compl. ¶¶ 171-172, such that the information was not new and therefore unable to be the cause of the decline in the price. Plaintiffs respond that neither the July 3 reference to the use of the 12-inch pipe nor the August 9 earnings call provided a clear enough disclosure to impact the market. Rather, it was the analysts' interest and report of August 10 that clarified the disclosure sufficiently enough for the market to quantify its impact on price and, from there, cause the drop in price. Looking at all the evidence surrounding the August 2018 disclosures, I conclude that Defendants have not rebutted the presumption of reliance.
Plaintiffs' expert does not explicitly quantify this decline in his rebuttal report, but merely notes that both he and Defendants' expert agree that “the abnormal returns on August 13, 2018 ... are statistically significant at the 95% confidence level,” with a citation to Bates stamped materials not available to the Court, which may have been produced by Defendants' expert. Coffman Rebuttal ¶ 48. Regardless, I take judicial notice of the fact that the security closed at $18.45 on August 8 and $17.41 on August 13 which represents a 5.6% decline. In re Merck, 432 F.3d at 264 n.3.
In their Complaint, Plaintiffs acknowledge that on July 3, 2018, “the press reported that Energy Transfer and Sunoco were planning to pump NGLs through an existing 12-inch pipeline that was being used to flow petroleum products in some sections of Delaware and Chester counties, where the new Mariner East pipelines were still under construction.” Compl. ¶ 171. But they contend that Energy Transfer actively “hid from investors the true impact of using the 12-inch pipe with ME2 until August 2018.” Compl. ¶ 181. Defendants' counsel argued strenuously at oral argument that these July reports represented the original disclosure to the market of the use of the 12-inch pipeline, and that Plaintiffs therefore are precluded from treating the August earnings call or analysts' reports as “new” disclosures. Hearing Trans. 54:20-56:3, ECF 108. In response, I invited counsel to submit to the Court the relevant news articles for consideration. Defendants submitted two articles and also pointed to a Bank of America analyst's report from July, which they submitted at ECF 102-2. Having reviewed these documents, I find that the use of an existing smaller pipeline was disclosed, but Energy Transfer made the disclosure by describing it in positive terms that obscured its impact on timely completion and overall capacity. Both the news articles begin with an Energy Transfer spokesperson quote to the effect that the use of the 12-inch pipeline wasn't a downgrade from the planned 20-inch pipeline but was instead “an existing pipeline that will allow us to meet our customer obligations to [transport natural gas liquids] while work on the Mariner East system continues.” In the Hurdle article, the company spokesperson represented that “a minimum amount of work will be required which can be done quickly and safely.” A local township official quoted Energy Transfer having described the pipeline as “in service, it's tested and permitted, and apparently all they have to do is notify PHMSA.” Similar assurances appear in the Rettew article. The Bank of America Report adopted Energy Transfer's positive spin, with the general observation that “[w]e view the potential for expediting the project positively as it could alleviate investor concerns and allow ETP to achieve its current 3Q18 in-service target, albeit perhaps not all full volume given the capacity difference on ME2 and the existing pipe.” ECF 102-2. The specific disclosures of a dramatic reduction in capacity, significant delay, and likely lower profitability-disclosures that resulted after analysts sensed Energy Transfer had not been entirely forthcoming-did not occur until August.
The two articles include Jon Hurdle, Sunoco wants to use older pipeline to pump NGLs over unfinished sections of ME2, StateImpact Pennsylvania (July 3, 2018), and Bill Rettew, Mariner East 2 pipeline may take alternate route, Daily Local News (originally published July 5, 2018) (updated Aug. 25, 2021).
Even when Energy Transfer discussed the use of the 12-inch pipeline on their call with investors on August 9, 2018, the disclosure was artfully crafted and required additional clarifying information. During its August 9, 2018, earnings call, Energy Transfer informed investors that in order to “avoid any delays,” it intended to incorporate an “already-constructed” 12-inch pipeline into the ME2 pipeline, rather than wait to construct the new 20-inch line “in areas where [authorities] had halted construction.” ECF 97-8. Later that same morning, at 10 a.m. on August 9, a reporter for Argus Media contacted Energy Transfer spokesperson Lisa Dillinger seeking to “clarify some remarks that were made on ME2 and ME2X.” Argus Emails (Aug. 9, 2018), ECF 97-14 (emphasis added). The reporter explained that he had understood from the call that Defendant “[McCrea] said y'all expect the full 275,000 b/d capacity on ME2 to be up in 3Q 2019,” and asked: “Is that correct? Can you give detail for the push back in time for when the lines will be completed?” Id. (emphasis added). Dillinger responded that “we will have ME2 and ME2X in service using a combination of 12-, 16- and 20-inch pipe by Q3 2019,” and added that this “will allow us to flow the majority of the capacity.” Id. Later disclosures made clear that the true starting capacity was not a “majority” of the previously claimed 275,000 barrels per day, but rather 100,000 barrels per day or approximately 36% of the projected capacity.
On August 10, 2018, following what was described as a “private discussion” with Energy Transfer, an analyst report from Wolfe Research provided further clarification to the market about what Energy Transfer had disclosed on August 9 disclosure. Wolfe Report, ECF 97-15, Compl. ¶ 405. It stated:
Mariner East is messier than expected but consistent with our conservative modeling. Discussion of ME 2 on the call was confusing. We talked to the company afterward. The “interim solution” with re-purposing of pipes is large enough to satisfy contracts plus some walk-up volumes. We estimate this could provide $150M - $250M of EBITDA starting in Q4. However, ME 2 may not reach full capacity (previously disclosed as 275 mbd) until 4Q19 (one year delay). Due to construction / regulatory issues, the company is using a hodgepodge of different sized pipes that will then be reworked / upsized over time. A final connection involving a small amount of ME 2 / ME 2x capacity won't be completed until 4Q20. We assume an incremental $100M contribution from the project starting in 4Q19 ($300M total). While transparency and disclosure on this project are very low, we are comfortable with the conservative contribution we
assume in our outlook.Id. (emphasis added). Of note, the Wolfe Report criticized Energy Transfer for its “confusing” discussion of ME2 on the second quarter 2018 earnings call the prior day and characterized Energy Transfer's “transparency and disclosure on this project” as “very low.” Id.
Also on August 10, 2018, Wells Fargo analysts released their own report to make sense of Energy Transfer's earnings call, similarly noting that Energy Transfer's pipe combination statements “confuse[d] the market,” which necessitated follow-up from their analysts “to gain clarity.” Wells Fargo Report, ECF 97-16; Compl. ¶ 406. (the “Wells Fargo Report”). After further inquiry of Energy Transfer to “gain clarity” on the topic, the Wells Fargo Report concluded that (1) ME2's initial capacity would be a nearly 65% reduction in the capacity Energy Transfer had told investors throughout the Class Period; (2) the projected cost of the ME2 project ($6 billion) would be double what Energy Transfer had previously represented ($3 billion); and (3) “full ME2 service” would not start until two years later than previously expected. Id. Specifically, the Wells Fargo Report projected “almost full” ME2 service to begin in “late Q3' [2019]” and full service in “Q3' [2020],” while their “previous model . . . had assumed full ME2 service by the end of Q'4 [2018].” Id. at 2. The report specifically tied these developments to lower profitability: “Given recent delays, we now estimate the project to cost $6B and assume a 12x EBITDA multiple return in our model. We are assuming a higher multiple (i.e. lower return) on this project given the limited visibility into the contract structure.” Compared to an analyst report issued a year earlier, this projected cost was significantly higher, when analysts expected construction costs for ME2 to be around $1.4 billion and for ME2x to be around $1.0 billion. Credit Suisse Report (Aug. 9, 2017), ECF 97-9. And although the Wells Fargo Report expressed a note of optimism, the wording of the report revealed a degree of speculation: “Our best guess at this juncture is that only ~100 MBbls/d of the 525 MBbls/d project is currently secured under long-term contracts. While the contracted amount is low, we have a high degree of confidence that contracts/volumes will increase rapidly as capacity becomes available given ... .” Wells Fargo Report (emphasis added).
Given the confusion described on the earnings call, and the need for journalists and analysts to follow up, a delay in the market absorbing the news does not on its own suffice to sever the link. It bears emphasis that the critical information was not announced by the company itself, but through analysts' reports, a factor that courts have deemed relevant in looking to time periods beyond a day. See Pelletier, 338 F.R.D. at 486; In re CenturyLink Sales Practices and Securities Litig., 337 F.R.D. 193, 211 (D. Minn. 2020) (disclosure by third party supports event window longer than one day). Moreover, the clarifying information was delayed until Friday, August 10 such that Monday, August 13 was the next market day on which the decline occurred. Pelletier, 338 F.R.D. at 486 (“[P]rice impact may occur more slowly where clarifying or contextualizing information is disclosed later.”). And the Wolfe Report included a notation stating that it was intended solely for the use of Wolfe Research clients and was not to be disseminated, which would limit the speed of its public dissemination. ECF 97-15.
Defendants accurately observe that as to these two analysts, their overall outlook did not change. Wolfe Research concluded that its analysis was “consistent with” its “conservative” prior models for Mariner East 2 and that its overall “[i]investment thesis [remained] solidly on track”. Wolfe Report, ECF 97-15. As to Wells Fargo, they concluded that “[b]y and large, our proforma ETE estimates do not materially change post the release of Q2 earnings.” Wells Fargo Report, ECF 97-16. But reporting entities are not entirely disinterested chroniclers of the market, and the conclusions and recommendations they draw are distinct from the facts and figures that they report. Investors, particularly sophisticated investors, will make their own judgments based on the information made available through the investigation and quantification work of the analysts. The information reported here, independent of the analysts' conclusions, would have been objectively concerning to market participants because it quantified delays and identified an impact on profitability. Moreover, the surrounding context was not reassuring, with its description of the project as “messier” than anticipated and afflicted by a lack of transparency. To underline that point, Energy Transfer refused to provide a quantification of how the use of the 12-inch pipeline would reduce capacity, stating “We really haven't shared capacities. We can look at maybe doing that in the future ...” ECF 97-8. And it continued to do so through the next day in direct response to journalists' questions, only acknowledging that by Q3 2019 they would be able “to flow the majority of the capacity,” again without quantifying the impact of the 12-inch pipe in the meantime. Argus Emails, ECF 97-14.
The price decline on August 13 was measurable and statistically significant, and Defendants offer no alternative explanations for it. See Coffman Rebuttal ¶¶ 40-43, ECF 97-2. Ms. Allen testified that she did not analyze the unit price reaction on August 13, 2018. Allen Dep. 191:3-8 (March 30, 2022), ECF 97-3 (“I did not do an event study -- I do not include an event study for that date.”). Instead, they rely on their baseline premise that because the (Friday) August 10 abnormal price decline was not statistically significant at the 95% confidence level, there was no price impact. Although I reject Plaintiffs' contention that a failure to explain the decline on August 13 is “fatal” to the defense position, it is nonetheless a factor to consider, especially because Defendants bear the burden of persuasion. See Wallace v. IntraLinks, 302 F.R.D. 310, 317 (S.D.N.Y. 2014). The justifications advanced by Plaintiffs' expert in his deposition for applying a multi-day window are sensible given all the surrounding circumstances. Coffman Dep. 76:1721; 87:16-24 (Nov. 18, 2021), ECF 97-12. Under Amgen, the issue of when information “credibly entered the market” is “a matter for trial.” 568 U.S. at 481-82.
But even if I were to limit my consideration of price impact to August 9, Plaintiffs' expert has produced analysis that suggests that the actual change in the price of the security that day was substantially below what might be expected based on the unexpectedly strong earnings reported on the call. Coffman Rebuttal ¶¶ 40-43, ECF 99-1. In an appendix table, he shows that the difference between the reported and expected EBITDA, i.e., earnings, multiplied by the analysts' enterprise value multiplies implies a $0.87 (4.71%) or $0.94 (5.09%) security price increase for August 9. Id. ¶ 41 (citing Ex. 5 to Coffman Rebuttal). However, the actual change in price was a decline of $0.17 (0.9%). Defendants' expert did not in her calculations account for the positive earnings announcement masking or muting the negative price impact of the corrective disclosures related to the pipeline.
According to historical data about the security price from Yahoo! Finance.
Thus, for present purposes, given all the evidence, Defendants have not rebutted the Basic presumption as to the August 2018 disclosures.
c. The October 2018 Corrective Disclosures
Plaintiffs allege another back-end disclosure occurred in October 2018 related to a stop work order that followed some weeks after a line explosion. Plaintiffs' allegations turn on the republication of an article on Saturday, October 27, 2018, by the Associated Press, and then the issuance of a stop work order by the Pennsylvania Department of Environmental Protection on Monday, October 29, 2018. There was a statistically significant drop in share price between market close on Friday and market close on Monday. Allen Report ¶ 79; Coffman Rebuttal ¶ 48. Defendants respond that the article was originally published on October 21 in the Pittsburgh PostGazette and contend that the stop work order wasn't published until Tuesday, October 30, 2018, and thus neither of Plaintiffs' alleged “disclosures” can be linked to the Monday price decline.
Turning first to negative press coverage, on Sunday, October 21, 2018, the Pittsburgh PostGazette published an article reviewing some of the controversy surrendering the ME2 pipeline. Anya Litvak, Pipeline ruptures bring new scrutiny to Pennsylvania geology, Pittsburgh PostGazette, ECF 93-15. Most significantly, the article addressed a September 10th underground earth movement that ruptured a newly activated gas line, caused an explosion, and destroyed a house. ECF 93-16. Homeowners near the pipeline expressed their concern that the pipeline was proceeding through unstable areas subject to subsidence, while Defendants spokespersons and consultants replied with assurance that all risks were being competently evaluated and addressed. The article did not purport to resolve the disagreement but concluded with a nearby family's resolve to sell their home and move away from the pipeline. The Associated Press's October 27threpublication of the article from the Pittsburgh Post-Gazette necessarily broadcast the controversy over the September explosion to a wider audience: the Gazette is a regional publication in Western Pennsylvania which as of October 2018 published print editions three days a week, while the Associated Press is a leading national wire service with broad distribution.
Defendants argue that the article was originally published on October 21, 2018, but the printed web article they attach in support of this, ECF 93-15, isn't actually dated, nor does the active webpage of the article, https://newsinteractive.post-gazette.com/mariner-east-2-pipeline-subsidence/, contain metadata related to the date. However, the ProQuest database indicates that the article was in fact published in the print edition on October 21, 2018, on page A-1. I may take judicial notice of news articles in order “to indicate what was in the public realm at the time.” Benak ex rel. All. Premier Growth Fund v. All. Cap. Mgmt. L.P., 435 F.3d 396, 401 n.15 (3d Cir. 2006).
Defendants argue that the Associated Press actually picked the Pittsburgh Post-Gazette piece up on October 23, 2018, and attach a copy of the AP story retrieved from the Dow Jones Factiva database with that date. ECF 93-16. AP's public, online version of the story identifies a date of October 27, 2018. https://apnews.com/article/2e0005ec7db342a290199a4d8464b5a0. WestLaw's news service also identifies the AP republication as October 27, 2018. The LexisNexis database contains a version of the article dated October 23, 2018, but identifies the source as “Associated Press State & Local Wire.” Based on these datapoints, I understand October 27 to be the relevant date as to when the original October 21 story was republished to the AP's wider audience.
On Monday October 29th, two days after the republication by the Associated Press, the PA DEP issued a Field Order to Energy Transfer's subsidiary ETC Northeast Pipeline specifically related to the September 10th explosion of Energy Transfer's Revolution pipeline. The order required it to “immediately cease all earth disturbance activities” other than essential repair and stability activities. Compliance Order (Oct. 29, 2018), ECF 99-5. Energy Transfer internal documents show that the DEP issued the stop work order to Energy Transfer on October 29th, by means of an email sent at 1:07 p.m., with a mail copy to follow. Cover Email to Compliance Order (Oct. 29, 2018), ECF 99-5. The PA DEP website also reflects that the PA DEP issued the Revolution Compliance Order to Energy Transfer “[o]n October 29, 2018,” and the website links to the Compliance Order, which is dated October 29, 2018. Printout of https://www.dep.pa.gov/ Business/ProgramIntegration/Pennsylvania-Pipeline-Portal/Pages/Revolution-Pipeline-.aspx, accessed April 11, 2022, ECF 97-19.
The parties dispute when news of this order was publicized to the market. With respect to publication of the DEP's action, Plaintiffs have submitted a document retrieved from Westlaw originating with the State News Service, dated October 29, 2018. DEP Issues Compliance Order to ETC for Violations on Revolution Pipeline, State News Service (Oct. 29, 2018), ECF 97-20. Defendants submit a DEP email disseminating a press release sent on October 30, 2018, at 12:59 p.m. ECF 93-18. The defense further points out that according to the Westlaw Data Base, the States New Service document was not posted until the 30th and argues that it could therefore not have affected the stock price on October 29th.
Defense counsel reached this result by using the “AD” (also known as “ADDEDDATE”) search function on Westlaw. This function allows users to search for documents by the date they were first made available on (i.e., “added” to) Westlaw. Compare ECF 93-19 (results of Westlaw news search for: adv: AD(10/30/2018) & “DEP ISSUES COMPLIANCE ORDER #TO ETC”, including the document in question, indicating it was added to Westlaw on Oct. 30, 2018), with ECF 93-20 (results of Westlaw news search for: adv: AD(10/29/2018) & “DEP ISSUES COMPLIANCE ORDER #TO ETC”, not including the document in question, indicating this document was not added to Westlaw on Oct. 29, 2018); “Westlaw User Guide” at 37, ECF 93-21 (explaining that the “AD” search function allows users to search documents “By Date Added to Westlaw”).
Neither side's analysis of the timing issue is particularly illuminating. But my close consideration of the record leads me to conclude that Defendants have not rebutted the presumption of market reliance. Focusing first on the DEP's action, the parties discuss what they consider to be competing versions of a press release. This is incorrect. Westlaw/Thomson Reuters carries State News Service as a source in its wire service databases. The State News Service incorporated the content of the DEP press release but reported, as an item of news, the issuance of a stop work order. The indexed report of the State News Service bears a date of October 29, 2018, and a Harrisburg dateline, and reports as a fact that “[t]he following information was released by the Pennsylvania Department of Environmental Protection (DEP),” followed by the content of the press release. ECF 97-20. The fact that Westlaw, which is a research database and not a news organization, did not index the article until October 30th does not undermine the fact that a wire service that follows state news was in possession of the press release on October 29th and published the contents of the release that same date as an item of news. It should also be noted that the State News Service report of October 29th bears a dateline of Harrisburg, while the October 30thversion submitted by Defendants bears a dateline of Pittsburgh. It would be logical for DEP to undertake a specific effort to communicate news of its action to media outlets in Western Pennsylvania given that the explosion occurred in that section of the state, separate from the initial release that originated in Pennsylvania's capital where the DEP is based.
The defense submitted a Pennsylvania Right-to-Know request seeking “any and all records regarding any release, dissemination, publication, or other external communication of the Press Release, including but not limited any communication of the Press Release to State News Service, prior to October 30, 2018.” See Pennsylvania Right-To-Know Request Form, ECF 93-22. In response, the DEP stated that “[t]he DEP office covered by this final response possesses no records responsive to your request ....” PA DEP Response to Right-to-Know Request (Oct. 21, 2021), ECF 93-23. I attach little weight to this argument given the multiple channels through which information reaches the public. Most notably, the current DEP website posts press releases upon their date of issue on both its home page, https://www.dep.pa.gov/, and a page dedicated to news releases, https://www.ahs.dep.pa.gov/NewsRoomPublic/. A review of archived versions utilizing the Wayback Machine, https://archive.org/web/, reflects a similar architecture on the DEP website as of 2018 with press releases featured both on the home page where searchers land and separately in the newsroom section.
Turning next to the news stories, Defendants are correct that, by definition, because the AP story was a republication, all the information contained in both articles was known to the market since the Post-Gazette first published the story on October 21st. The Defense further argues that neither of the October articles represented “news” to the market, because the explosion at the center of the stories occurred on September 10th. That is also true to some extent, except the initial stories set forth both sides of the controversy-homeowners and pipeline company-and concluded without attributing blame. The issuance of the stop work order broke that stalemate on the side of homeowners, prohibiting further work based upon the DEP's investigation, and, as such, linked the explosion to regulatory non-compliance with potential consequences for future construction and the ability to meet timelines. In effect, the DEP rejected Energy Transfer's explanations to the press: the articles had alerted the market to there being smoke; the stop work order confirmed the existence of fire.
It should be noted that Defendants' expert identified a statistically significant negative return on Energy Transfer the next trading day after its publication, Monday, October 22, 2018, albeit not a drop as dramatic as the one on Monday October 29. Allen Report ¶ 81 n.131.
This also explains why Defendants' argument based on the market's failure to react to news of the September 10 explosion lacks merit. Defs. Br. at 16 n.17. Energy Transfer did not concede a link between the explosion and its work up to and including the Post-Gazette article in October. The DEP's stop work order cast those denials in a different light. See also Allergan PLC Sec. Litig., 2021 WL 4077942, at *12 (S.D.N.Y. Sept. 8, 2021).
In my view, the Associated Press reprint has significance in three respects. First, the fact that the story was deemed worthy of reprint would give it greater significance. Second, as noted above, it would bring the story to a wider, national audience. Finally, it would heighten awareness that the project was one to watch closely, increasing the likelihood that concerned investors would more closely monitor project news and the DEP website. Furthermore, the issuance of a stop work by state authorities is a significant objective event that does not require nuanced analysis, especially coming on the heels of press coverage suggesting serious issues with the project.
On the record as it stands, the same day a stop work order was publicly issued and on the first trading day after a problematic story was widely distributed, Energy Transfer's stock decreased by a substantial amount relative to the energy index. As with the August 2018 disclosure, Defendants do not offer a competing explanation sufficient to sever the link between the disclosures and the alleged misrepresentations.
Based on the available evidence, I conclude that Defendants have failed to rebut the presumption.
d. The December 2018 Corrective Disclosures
In December 2018, the District Attorney of Chester County announced an investigation into Energy Transfer regarding their involvement in causing sinkholes in Pennsylvania. On Wednesday, December 19th and Thursday, December 20th, a number of media outlets reported on the announcement through various fora including Twitter, AM radio, news wires, trade news publications, and daily newspapers. By market close on Friday, December 21st, the security had experienced a decline from its closing price on December 18th. Defendants argue that the news was already widely disseminated as of Wednesday, December 19th, and that on both December 19th and December 20th, the security traded up relative to the energy sector as a whole. Defs. Br. at 18-19. Defendants further argue that December 21st is too distant in time from the December 19th corrective disclosure to be the product of an efficient market. Id. at 19. The factual nuances surrounding this disclosure are complex, and the issue is a close one, but on balance there is a substantial and unexplained drop on December 21st that corresponds to Energy Transfer's internal expressions of concern over traffic about the story on social media. On that basis I conclude it has failed to rebut the presumption.
On December 19, 2018, at 11:36 am, the District Attorney for Chester Count, Thomas Hogan, announced an investigation into Energy Transfer, “following numerous reports of sinkholes appearing within feet of residences.” DA Press Release, ECF 93-24 (attached to @deannadurante tweet). News outlets disseminated the story to the public beginning in the morning hours of December 19, 2018. See Deanna Durante (@deannadurante), Tweet Sharing Chester DA Press Release (Dec. 19, 2018, 11:36 AM EST), ECF 93-24 (Durante is an NBC10 Philadelphia reporter) (tweet shows 25 retweets, 4 quote tweets, and 31 likes). Then, after trading hours on December 19, the Associated Press ran an article on the subject. Marc Levy, Prosecutor opens investigation on Mariner East pipeline work, Associated Press (Dec. 19, 2018), https://apnews.com/article/6966207bad89479cb54e93e010fa6c14, ECF 93-25. As noted above, coverage by the AP would reach a wider, national audience. Trading on December 20th reflected mixed results, with Energy Transfer trending down at the opening, recovering briefly, reaching its nadir mid-afternoon, but then rallying late in the day. The Complaint alleges that “Reuters picked up the story of DA Hogan's investigation of Energy Transfer, and further reported on and disseminated it at 3:35 p.m. on December 20, 2018, just before market close, in an article entitled ‘Pennsylvania Prosecutor Opens Criminal Investigation of Mariner East Pipe.'” Compl. ¶ 415. While both parties discuss the Reuters story, neither has submitted a copy. Regardless, an internet search confirms its content and date of posting. Energy Transfer stock briefly dipped and then closed higher, though still down for the day.
According to the Complaint ¶ 414, KYW News Radio reported on the story on December 19, however neither party addresses this reporting nor points to evidence on the record to support its inclusion as part of my determination here.
Defendants argue that the AP article would have had an impact during the day of December 19, but metadata available from the website indicates an online publication time of 23:19 GMT (or 6:19 p.m. EST). See Benak, 435 F.3d at 401 n.15.
Plaintiffs include the following chart in their brief, which suggests two major declines in those three days for the Energy Transfer security, one from about 12pm on December 19th to 12pm on December 20th, and one from open to close of market on December 21st.
(Image Omitted)
On its face, the chart shows declines, but the variability of the market makes it difficult to determine cause and effect. More fundamentally, Defendants contend that the chart is deceptive because it fails to control for the broader market. After using Mr. Coffman's industry index (the S&P Oil, Gas, & Consumable Fuels Index, see Coffman Report ¶ 50) to control for these effects, the data shows that Energy Transfer units experienced a relative increase, compared to the industry index, of 0.05% from 11:36 AM to the close of trading on December 19, 2018, and, as well, a relative increase, compared to the index, of 1.78% from the closing price on December 19th to the closing price of December 20th. Defs. Reply at 15-16, ECF 102. Plaintiffs have not responded to this critique, and Defendants' argument as to the 19th and 20th is persuasive.
Defendants do not, however, offer similar calculations for the precipitous drop on December 21st, Allen Report ¶¶ 68-69, which came the day after Reuters re-distributed the story late the day before. Plaintiffs' expert avers that the decline is statistically significant, Coffman Rebuttal ¶ 37, and it seems unlikely that Defendant's failure to address this drop is simply an oversight. Although Defendants' expert contends that both December 20th and December 21st are too distant from the December 19th disclosure to reflect price impact in an efficient market, she only states that “there was no statistically significant decline in Energy Transfer's stock price on December 20, 2018.” Allen Report ¶ 68. She is notably silent on the statistical significance of the price movement on December 21st, only restating her contention as to distance in time. Allen Report ¶ 69. The averment of Plaintiffs' expert is borne out by the data: the stock declined a substantial amount relative to the relevant index relied on by the parties on December 21st when the Energy Transfer security declined 5.12% from open to close, compared to only a 0.74% decline in the price of the sector.
According to the historical prices available at Yahoo! Finance (https://finance.yahoo.com/), the index opened at $459.59 and closed at $456.19, where the security opened at $12.90 and closed at $12.24. (There appear to be minor discrepancies of a few cents among various sources for the historical value of the index, identified by the symbol SP500-101020, such as those found on Bloomberg-used by the parties, Marketwatch.com, and Yahoo! Finance. Such discrepancies only account for hundredths of percentages of the index's value and do not impact my rougher calculations here. I have chosen Yahoo! Finance as my primary source because it is free and available to the public.)
The question then becomes whether the drop on December 21st is too remote in time from the initial announcement of the investigation, and whether wide dissemination of the news on the 19th and 20th would have blunted the impact of the disclosure. Plaintiffs argue that the link is not severed, based upon a flurry of social media posts being tracked internally by Energy Transfer showed widespread interest in the news on December 21st. An Energy Transfer email from 11:29 a.m. ET on December 21, 2018, collected Energy Transfer-related news reports. ECF 99-6. It noted that the December 20, 2021, Reuters “article was circulated by a number of other outlets including The New York Times, CNBC and Shale Directories.” Id. A lengthy section of the email summarized the December 21st Energy Transfer-related social media posts and stated that social media activity “remained focused on circulating news of DA Hogan's investigation in the Mariner East 2 pipeline project.” Id. The email included a list of several Energy Transfer-related tweets dated December 21, 2018, almost all of which circulated forms of the article that Reuters had published late in the trading day on December 20, 2018. Id.
The Reuters story went online less than thirty minutes before the close of trading on December 20th. Defendants are correct that the market reacts almost immediately to certain types of information, but immediate reaction is more likely to occur when a listed company itself releases information or the information is quantitative in nature such as earnings. The decline on December 21st occurred contemporaneously with substantial online dissemination and discussion of the Hogan investigation. Defendants offer no plausible alternative explanation for the declines that day in comparison to the sector as a whole.
Considering all the evidence, Defendants have failed to rebut the presumption.
e. The August 2019 Corrective Disclosures
On August 8, 2019, the Chester County DA arrested and charged two Pennsylvania Constables “in connection with their work as private security guards in connection with the construction of the Mariner East pipeline project,” and Energy Transfer denied a connection to the constables. Compl. ¶¶ 418, 422. A local, online news outlet reported the news in the early afternoon trading hours of August 8th. Justin Heinze, State Constables Charged with Oppression at Mariner East Pipeline, Patch (Aug. 8, 2019) (“Posted Thu, Aug 8, 2019 at 1:12 pm ET”), https://patch.com/pennsylvania/westchester/state-constables-charged-oppression-mariner-east-pipeline, ECF 93-26. According to experts for both sides, Energy Transfer traded up, both during the day as a whole and during the portion of the day after the news was released, on August 8, 2019. Allen Report ¶ 71. On Friday, August 9, 2019, Energy Transfer common units fell, but the decline was not in excess of the relevant index to a statistically significant degree. Allen Report ¶ 73. However, on Monday, August 12, 2019, Energy Transfer common units fell by an additional 3.7% which Plaintiffs' expert contends is statistically significant. Coffman Rebuttal ¶ 54.
While the post to Patch.com, an online news service and aggregator that attempts to offer “hyperlocal” coverage, was posted at 1:12 p.m. on August 8, it was updated on Saturday, August 10 at 9:36 a.m., with no indication of what precisely was updated.
Defendants' key argument is that the price actually increased on August 8th and only declined a non-statistically significant amount on August 9th, such that the statistically significant decline on August 12th cannot be connected to the disclosures. Plaintiffs argue a multi-day window is appropriate, first, because any negative effect on August 8th was muted by a positive earnings statement that same day and, second, because Energy Transfer obfuscated about the nature of the relationship between the constables and Energy Transfer when the news originally broke such that it took longer for the real news to make it to the market. As I view the record, Defendants have successfully shown that there was no price impact on either August 8th or August 9th, and Plaintiffs' position therefore relies on extending the relevant window to Monday, August 12th. Here, however, I find that Plaintiffs fail to show that the market was continuing to absorb the information such that an extended window for assessing the price impact would be warranted.
It is true that Energy Transfer issued positive earnings on August 8, 2019, that caused a statistically significant increase in Energy Transfer's unit price that day. Allen Report ¶¶ 71-72. And Plaintiffs are correct that Ms. Allen can be criticized for acknowledging that there was confounding positive information that entered the market without attempting to evaluate its impact. But there was a minimal downturn in Energy Transfer stock on the August 9th, and a distinct lack of growing media interest over the intervening weekend. The initial press was local, and there is no evidence of any wire service or national media outlet picking up the story. Nor can Plaintiffs point to interest from analysts or social media activity circulating the story. Plaintiffs are correct that on August 8th, Defendants responded to the disclosure of the constables' arrests with an arguably misleading denial that Energy Transfer had any direct connection to them. The article reporting the arrests stated that, “Sunoco was adamant that the pair were not employed by Sunoco or Energy Transfer directly.” ECF 93-26. And I acknowledge that Energy Transfer answered a media inquiry in a manner intended to limit its connections to the constables. See Emails between Law360 and ET PR Representative (Aug. 8, 2019), ECF 99-7. But the weekend news coverage still appears to have been only local, with an updating of the Patch website, ECF 93-26, and a story in a local paper, The News-Item, published on August 10th, identifying the constables arrested as Sunoco subcontractors, ECF 97-23. In short, the evidence available does not show increasing or geographically expanding interest in the story.
Plaintiffs also reference in their Complaint several other local news sources that reported on the arrests including 3 CBS Philly, phillymag.com, State Impact, and WHYY. Compl. ¶¶ 419, 420. It is not clear why Plaintiffs did not put any of these sources into the record. But regardless, they do not disturb a finding that the coverage was local in nature and did not appear to give rise to a concern in the wider market.
Qualitatively, I note that this disclosure was different in nature than those discussed above. It cast Defendants in an unflattering light and called into question their business and civic ethics, but it did not involve potential disruption of Defendants' ability to reach capacity, complete the project in a timely way, or operate profitably. Although matters of ethics and corruption may be material to investors, for the limited purpose of assessing price impact they do not carry the same weight in the absence of corroborating evidence that market participants were aware of or otherwise concerned about their impact on the value of the security.
More importantly, Plaintiffs' presentation as to market performance on August 12th is lacking. In their briefing, they simply cite to paragraph 421 of their Complaint, incorporating an item from News Bites reporting on August 12th that the price of Energy Transfer units had “sunk [] 4.6% over the past two days” on higher-than-average trading volume. They do not even submit the article: their expert merely states his conclusion that the abnormal returns on August 12, 2019, are statistically significant. Coffman Rebuttal ¶ 48. And although Plaintiffs and their expert criticize Defendants' expert for failing to account for potentially confounding positive earnings news, in neither their brief nor their expert report do they draw an effective link between the price decline on August 12th and the initial disclosure and its ensuing coverage. As to this market drop, the Basic presumption does not suffice, as I do not see an evidentiary basis for extending the window to consider price impact beyond the two-day window.
f. The November 2019 Corrective Disclosures
In November 2019, Plaintiffs allege that the disclosure of an FBI investigation into corruption in the permitting process for the pipelines led to declines in the price of Energy Transfer securities. On November 12th, the Associated Press published an article reporting that “FBI agents have interviewed current or former state employees in recent weeks about the Mariner East project and the construction permits” and that “[t]he focus of the agents' questions” included “the permitting of the pipeline” and “whether Wolf and his administration forced environmental protection staff to approve construction permits.” Compl. ¶ 423; Coffman Rebuttal ¶ 23, ECF 991. The same day, the Philadelphia Inquirer also covered the investigation, noting “[t]he FBI's investigation ... has focused on whether Wolf administration officials forced staff of the Pennsylvania Department of Environmental Protection to ignore shortcomings and approve the pipeline's construction permits.” Compl. ¶ 424. On November 12th and November 13th, Energy Transfer posted losses of 2.6% and 4.3% respectively, which Plaintiffs attribute to the news of the investigation. Compl. ¶¶ 423-426. Defendants' expert concedes that “there was a statistically significant price decline over this period” but argues that the decline is not evidence of price impact following a corrective disclosure.
Plaintiffs do not include either article in the record, but instead rely on quotes of the article in their Complaint and the rebuttal report of their expert. Having confirmed the existence and substance of these articles independently, I am considering them only insofar as they reflect the entry of news into the market and not for the truth of any matter contained within. See Benak, 435 F.3d at 401 n.15.
The decline on November 12 is significant to a 90% confidence interval, and the decline on November 13 is significant to a 99% confidence interval. Coffman Rebuttal ¶ 27, ECF 97-2. In some cases, parties contest what constitutes the appropriate confidence interval for establishing or disproving price impact when using event studies. As to this disclosure, the Defendants do not attempt to do so. That is likely because the November 12 interval measures to the whole trading day, even though the disclosure did not occur until half an hour before the market closed. What followed was a clear and precipitous decline that carries over into November 13 as reflected in the price chart. Ex. 1 to Coffman Report ¶ 27.
Defendants first argue that all the front-end misrepresentations concerning the investigation into permitting were excluded from the case by this Court's order on the motion to dismiss, and therefore the back-end corrections cannot form the basis for liability. Defs. Br. at 2021. Second, Defendants contend that the focus of the investigation was the Governor's office, rather than Energy Transfer, and no charges resulted. They further argue that the investigation wasn't new because rumors had been swirling about political interference in the permitting process well prior to the FBI's investigation into whether such political interference was illegally motivated. Defs. Br. at 21-22. Defendants develop this point further in their Sur-reply, where they suggest that the news of the investigation itself was not “new” because “it does not speak to the existence or quality of existing evidence.” Defs. Sur-reply at 19.
Addressing Defendants' first argument, I note that the paragraphs dismissed from the Complaint were statements that the permits were “fully protective of the environment,” Compl. ¶ 373, that all required permits had been or were in the process of being obtained, Compl. ¶ 389, and that the “permits were properly and lawfully issued,” Compl. ¶ 391. These statements were dismissed as unactionable statements of opinion regarding the legality of the permits following the Supreme Court's holding in Omnicare, Inc. v. Laborers Dist. Council Const. Indus. Pension Fund, 575 U.S. 175 (2015). See Energy Transfer, 532 F.Supp. at 225. It was Defendants' characterization of the permitting process that I struck as a basis for imposing liability, not the underlying facts alleged.
My opinion specifically described the misrepresentations alleged in the Complaint to include misleading investors about how “significant geological risks were ignored, such that .. the timelines provided were impossible to achieve.” See Energy Transfer, 532 F.Supp.3d at 205 (citing Compl. ¶ 333). Multiple portions of the Complaint that were not stricken set forth the pertinent allegations as to this corrective disclosure. See, e.g., Compl. ¶¶ 343-44 (claims about ME2 coming online “sometime in the fourth quarter” of 2017 alleged to be false and misleading “because, undisclosed to investors ... the Energy Transfer Companies had obtained the ME2 permits impermissibly, and Energy Transfer had failed to conduct the planning and study necessary to construct, and bring in service, the ME2 pipeline at the claimed time”); see also Compl. ¶¶ 352, 359. Elsewhere, the Complaint pleads that Energy Transfer's statement of December 19, 2018-that “[t]he safety of all those who live and work along our [Mariner East] pipeline is our first priority and this project was planned and implemented based on that fact”- were false and misleading because the “Energy Transfer Companies knowingly sought to shortcircuit important safety-related regulatory oversight.” Compl. ¶ 381. The Complaint further set forth Energy Transfer's purportedly false representation that “[w]e are committed not only to following the strict guidelines set forth in our permits but to employing the highest levels of construction expertise and to preserving and protecting the environment in which we conduct our work.” Compl. ¶ 377. Media reports that state regulators were pressured “to ignore shortcomings” represent stories that would directly relate back to the misrepresentations alleged. The November 2019 disclosures are therefore relevant to allegations that remain in the case.
As to Defendants' second series of arguments, I am persuaded that news that the FBI was investigating improprieties surrounding the issuance of permits was significant and had implications for Energy Transfer even if the focus of the investigations was on regulators and the Governor's office, and even if some of the underlying contentions were already known to the market. Defendants are correct that there had been press coverage of deficiencies in the permitting process and allegations of political pressure dating back as far as March of 2017. Susan Phillips & Jon Hurdle, DEP approved Mariner East 2 permits despite deficiencies, documents show, StateImpact Pennsylvania (Mar. 10, 2017), ECF 93-27. I have further reviewed the other articles cited by the Defendants, Defs. Br. at 21-22, from January of 2018 and April of 2019, ECF 28, 29, and the series of excerpts collected by Defendants' expert, Allen Report ¶ 52. But the fact that the FBI placed sufficient credence in such allegations so as to open an investigation was a significant new development in its own right, and one that bore upon multiple prior representations made by Energy Transfer.
This is demonstrated by the reactions of analysts. On November 13th, a daily analysis provided by Steve Fleishman at Wolfe Research was sent to Defendant McReynolds that included a large section devoted to the investigation and its impact on the stock. ECF 99-10. The analysis stated that “[t]he FBI investigation creates an overhang for the stock,” and that, while “Mariner is already under investigation by the state AG and local county DAs[,] some investors viewed this as political given the controversial nature of the project, but the FBI involvement now could imply something more.” Id. (emphasis added). It continued: “[T]he worst case scenario we could envision at this initial stage assuming foul play would be for a court to pull back ME 2's permits, suspending operations of the pipeline.” Id. On November 12, 2019, the investment bank Stifel issued a “flash note” in response to the breaking news, in which it disclosed it was “removing ET from the Select List” based on the investigation and noted that it “could potentially place ME2X in jeopardy of further delays and postpone its corresponding cash flows.” ECF 99-9. UBS, Wells Fargo, and RBC Capital Markets also took note of the investigation as impacting prospects for Energy Transfer, with RBC identifying the stories as a “key driver of recent ET unit price weakness.” Coffman Rebuttal ¶ 29, ECF 99-1. In sum, Defendants' argument that the subject matter of the investigation was not “new” to the market is refuted by the contemporary evidence of the market's reaction. “The existence of a price decline and analyst commentary highlighting the negative news is, ‘of course ... evidence of price impact.'” Monroe Cnty., 332 F.R.D. at 396 (quoting Loc. 703, I.B. of T. Grocery and Food Employees Welfare Fund v. Regions Fin. Corp., CV 10-J-2847-S, 2014 WL 6661918, at *7 (N.D. Ala. Nov. 19, 2014)).
Defendants do not dispute that the price impact was substantial and statistically significant and have failed to rebut the presumption of reliance by showing that the corrective disclosures did not have a price impact.
g. The December 2019 Corrective Disclosures
Finally, in December 2019, Plaintiffs allege that Energy Transfer's prices were impacted when the District Attorney of Chester County filed criminal bribery and conspiracy charges against Energy Transfer's head of security for the Mariner East project. The news was made public on December 3rd coinciding with a same-day decline of 2% in the price of the security. Compl. ¶¶ 427-431. Defendants' primary argument to defeat the presumption of price impact is that the 2% decline was not statistically significant, Defs. Br. at 23, and in their Sur-reply and Expert Report, Defendants also point out that there is no evidence of market interest in the story. Defs. Sur-reply at 20-21; Allen Report ¶ 77.
Both experts evaluated the drop in price as compared to the energy sector. According to Plaintiffs' expert, the excess negative return was 0.85%; according to the Defendants' expert, 0.45%. Both experts agree that this difference in performance was not statistically significant. Unlike other corrective disclosures, Plaintiffs cannot show further reporting or inquiries from press or analysts, or further drops in the stock price in the days that follow. As the defense points out, the District Attorney had already brought criminal charges against two Pennsylvania constables for related conduct which would further diminish the novelty of the disclosure.
In short, there was no meaningful market reaction to this disclosure, with the result that it cannot form a basis for class-wide relief.
5. The Effect of Finding No Price Impact as to Specific Corrective Disclosures
The parties engage in a limited argument over what effect, if any, a finding of no price impact as to certain disclosures has on the scope of the order granting class certification. Specifically, Defendants object to Plaintiffs' parenthetical quotation of Monroe County to the effect that “the existence of a statistically significant stock price decline following one corrective disclosure ‘dooms Defendants' attempt to rebut the presumption of reliance because the inquiry is whether Defendants have proven a complete lack of price impact during the period.'” Pltf. Reply at 20 (quoting Monroe Cnty., 332 F.R.D. at 395). I agree that Plaintiffs' formulation is too expansive.
Defendants correctly point out that class certification is claim specific and here the alleged misrepresentations define the Rule 10b-5 claims raised. But a finding that there was no price impact for a specific back-end disclosure does not necessarily render front-end misrepresentations related to that disclosure irrelevant. If a given misrepresentation is corrected by more than one disclosure, such a misrepresentation remains relevant so long as at least one of the corresponding corrective disclosures survives.
These inquiries are necessarily fact-intensive and may well be impacted by expert analysis. I will reserve their determination for the loss causation phase on the merits. This is in line with what other courts have done in similar circumstances. See, e.g., Strougo v. Tivity Health, Inc., No. 3:20-CV-00165, 2022 WL 2037966, at *9 (M.D. Tenn. June 7, 2022) (citing cases); Pelletier v. Endo Int'l PLC, 338 F.R.D. 446, 477 (E.D. Pa. 2021) (“During class certification, however, a class period should not be cut off if questions of fact remain as to whether the disclosures completely cured the market.”) (cleaned up).
B. The Remaining Elements of Class Certification under Rule 23(a) and Rule 23(b)(3)
Although the parties' principal dispute is over whether the predominance requirement of Rule 23(b)(3) is met, it is still necessary for the Court to conduct a “rigorous analysis” of the evidence and arguments presented” in order to determine, by a preponderance of the evidence that “the requirements of Rule 23 are met” and that “the plaintiffs' claims are capable of common proof at trial.” In re Lamictal Direct Purchaser Antitrust Litig., 957 F.3d 184, 190-91 (3d Cir. 2020).
1. Numerosity under Rule 23(a)
Rule 23(a) requires that a “class is so numerous that joinder of all members is impracticable.” “This calls for an inherently fact-based analysis” and does not rely on meeting some specific number, although classes greater than 40 are typically found to suffice. In re Modafinil Antitrust Litig., 837 F.3d 238, 249-50 (3d Cir. 2016), as amended (Sept. 29, 2016). The proposed class here includes “all persons who purchased or otherwise acquired common units of Energy Transfer LP” in the roughly thirty-two-month relevant time period. Pltfs. Mot. ¶ 1, ECF 79. Over the course of those thirty-two months, Energy Transfer stock was traded on the New York Stock Exchange. More than 1,220 institutional investors purchased or acquired shares during the Class Period. Millions of shares of the stock were traded every day, averaging over 5 million trades per day regularly and over 12 million trades per day following earnings announcements. These trades exchanged some of the 1.079 billion shares outstanding of the company. All these factors suggest that the prospective class numbers are at least in the thousands if not tens of thousands and that joinder of all the individual members would be impractical here, and I therefore conclude that the numerosity element is satisfied.
2. Commonality under Rule 23(a)
“Commonality requires the plaintiff to demonstrate that the class members have suffered the same injury.” Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 349-50 (2011) (cleaned up). Class members' “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.” Id. at 350. “Commonality is satisfied when there are classwide answers.” Reyes v. Netdeposit, LLC, 802 F.3d 469, 482 (3d Cir. 2015).
Here, the legal and factual questions all turn upon a common course of conduct by the Defendants as to representations made to class member shareholders about the status of the Defendants' pipeline projects. The factual and legal questions as to whether such representations were false or misleading, Defendants' state of mind in making such representations, and whether such representations violated the securities laws are all common to the putative class. And, further, questions of the impact these representations had on Defendants' securities and alleged damages caused to class members as holders of these securities are also susceptible to common resolution. Thus “the answers to questions about [Defendants'] alleged misconduct and the harm it caused would be common as to all of the class members, and would thus inform the resolution of the litigation ... .” Sullivan v. DB Invs., Inc., 667 F.3d 273, 299-300 (3d Cir. 2011). Commonality is therefore satisfied here.
3. Typicality under Rule 23(a)
Typicality, along with commonality, serves as a guidepost to determine “whether under the particular circumstances maintenance of a class action is economical and whether the named plaintiff's claim and the class claims are so interrelated that the interests of the class members will be fairly and adequately protected in their absence.” Marcus v. BMW of N. Am., LLC, 687 F.3d 583, 597-98 (3d Cir. 2012). Typicality “screen[s] out class actions in which the legal or factual position of the representatives is markedly different from that of other members of the class even though common issues of law or fact are present.” Id. at 598 (quoting 7A Fed. Prac. & Proc. Civ. § 1764 (4th ed.)). Typicality requires the court to address “three distinct, though related, concerns: (1) the claims of the class representative must be generally the same as those of the class in terms of both (a) the legal theory advanced and (b) the factual circumstances underlying that theory; (2) the class representative must not be subject to a defense that is both inapplicable to many members of the class and likely to become a major focus of the litigation; and (3) the interests and incentives of the representative must be sufficiently aligned with those of the class.” In re Schering Plough Corp. ERISA Litig., 589 F.3d 585, 599 (3d Cir. 2009)). “Even relatively pronounced factual differences will generally not preclude a finding of typicality where there is a strong similarity of legal theories or where the claim arises from the same practice or course of conduct.” In re Nat'l Football League Players Concussion Injury Litig., 821 F.3d 410, 428 (3d Cir. 2016) (cleaned up). The claims of the named Plaintiffs and class members are virtually identical, and all arise out of the same alleged unlawful course of conduct by the Defendants that artificially inflated the price of securities owned by class members. The requirement for typicality is therefore satisfied.
4. Adequacy under Rule 23(a)
“The principal purpose of the adequacy requirement is to determine whether the named plaintiffs have the ability and the incentive to vigorously represent the claims of the class.” In re Cmty. Bank of N. Va. Mortg. Lending Practices Litig., 795 F.3d 380, 393 (3d Cir. 2015). “The adequacy requirement has two components: (1) the interests and incentives of the representative plaintiffs; and (2) the experience and performance of class counsel.” Dewey v. Volkswagen Aktiengesellschaft, 681 F.3d 170, 181 (3d Cir. 2012).
The Named Plaintiffs here are five retirement and pension funds of various public entities and unions. Agents of all the Named Plaintiffs have submitted affidavits attesting to their diligent attention to the case and their understanding of their duties as named plaintiffs to provide fair and adequate representation. ECF 79-3, 79-4, 79-5, 79-6, 79-7. All Named Plaintiffs have a financial interest in the outcome of this litigation that is representative of class members, as they were all holders of Energy Transfer securities during the Class Period and were allegedly injured by the same course of conduct by the Defendants. I thus conclude that the interests and incentives of the Named Plaintiffs are representative and adequate of the class at large. I further conclude that class counsel are adequate to legally represent the class in this action. Both firms have extensive history litigating securities class actions and, through this litigation, have competently briefed, argued, and survived the motion to dismiss, and have provided competent briefing on the present class certification motion.
Plaintiffs' counsel also includes with the filing a copy of an order from Judge Alsup in the Northern District in California which addressed the appearance of impropriety in an action brought by Bernstein, Litowitz, Berger & Grossman, LLP (BLBG), where in-house counsel for one of the named plaintiffs joined BLBG mid-litigation. Seb Inv. Mgmt. AB v. Symantec Corp., No. C 18-02902 WHA, 2021 WL 1540996 (N.D. Cal. Apr. 20, 2021), ECF 79-8. While Judge Alsup did not conclude that there were in fact improper motives in the hiring, he ordered disclosure of the conflict to the class in that action, and he ordered disclosure of the order itself to named plaintiffs and courts in filings made over the next three years. Named Plaintiffs here, in their affidavits, indicate that they have received and reviewed the order from that case. BLBG therefore appears in compliance with that order, and I do not find that it presents any impediment to BLBG's ability to adequately represent the class here, particularly where they are co-counsel with attorneys at another highly qualified firm.
5. Superiority under Rule 23(b)(3)
In addition to predominance, plaintiffs seeking certification under Rule 23(b)(3) must show that “a class action is superior to other available methods for fairly and efficiently adjudicating the controversy.” Rule 23(b)(3) identifies the four factors that courts must look to in determining whether the class action is the superior procedural method for adjudication: “(A) the class members' interests in individually controlling the prosecution or defense of separate actions; (B) the extent and nature of any litigation concerning the controversy already begun by or against class members; (C) the desirability or undesirability of concentrating the litigation of the claims in the particular forum; and (D) the likely difficulties in managing a class action.” These factors favor class certification here, where the complexity and costs of the litigation outweigh the possible recoveries of individual plaintiffs, there are no known parallel suits by individual plaintiffs, judicial and administrative efficiency favor centralization, and the case presents no untoward difficulties in management in the form of a class action.
IV. Conclusion
Based on the analysis above, I conclude that the Plaintiffs have proven that class certification is appropriate here. They have established all the Rule 23(a) elements, as well as the necessary elements of Rule 23(b)(3). Accordingly, Plaintiffs' motion for class certification will be granted. An appropriate order follows.