In re Apache Corp. Securities Litigation, No. 4:21-cv-00575

• Court: District Court, S.D. Texas
• Decided: February 9, 2024
• Docket: 4:21-cv-00575
• Status: Unknown

Opinion

UNITED STATES DISTRICT COURT February 09, 2024 SOUTHERN DISTRICT OF TEXAS Nathan Ochsner, Clerk HOUSTON DIVISION

§ IN RE APACHE CORP. § CIVIL ACTION NO. 4:21-cv-00575 SECURITIES LITIGATION § §

MEMORANDUM AND RECOMMENDATION Pending before me is Lead Plaintiffs’ Amended Motion for Class Certification and Appointment of Class Representatives and Class Counsel (“Motion for Class Certification”). Dkt. 101. On December 6, 2023, I held a hearing on the Motion for Class Certification during which both sides presented expert testimony and voluminous exhibits. I afforded Lead Plaintiffs Plymouth County Retirement Association and the Trustees of the Teamsters Union No. 142 Pension Fund (collectively, “Plaintiffs”) the opportunity to submit supplemental case law after the hearing, which Plaintiffs did. Having considered the parties’ briefing, oral arguments, the record, and the applicable law, I recommend that the Motion for Class Certification be GRANTED in part and DENIED in part. BACKGROUND This is a securities class action brought by Plaintiffs on behalf of those purchasers of the common stock of Apache Corporation (“Apache”) during the period from September 7, 2016 through March 13, 2020 (the “Class Period”). The defendants are Apache, an exploration and production company headquartered in the Houston area, and three of its top executives: (1) John J. Christmann IV (“Christmann”), Apache’s President and Chief Executive Officer; (2) Timothy J. Sullivan (“Sullivan”), Apache’s former Executive Vice President – Operations Support; and (3) Stephen J. Riney (“Riney”), Apache’s Executive Vice President and Chief Financial Officer (collectively, “Defendants”). Plaintiffs bring claims under §§ 10(b) and 20(a) of the Securities Exchange Act of 1934 (“Exchange Act”) and Rule 10b-5 promulgated thereunder by the Securities Exchange Commission. Plaintiffs allege that, over the course of three years, Defendants misrepresented the prospects of a hydrocarbon play in the Permian Basin known as Alpine High. In particular, Plaintiffs allege that Defendants “repeatedly told investors that, even under Apache’s ‘conservative’ models, Alpine High held over three billion barrels of oil and significant amounts of ‘really rich gas,’” and “that years of rigorous testing and analysis had ‘confirmed’ and ‘proven’ that Alpine High held extremely high-quality oil and gas, and that multiple ‘successful’ test wells showed the play was ‘prolific.’” Dkt. 101 at 6–7. For the uninitiated, “rich gas” or “wet gas” is “[n]atural gas containing liquid hydrocarbons in solution.” Dkt. 65 at 6. “Wet gas” is contrasted with “dry gas,” which is “[n]atural gas that does not have a significant content of liquid hydrocarbons or water vapor” and is less profitable than oil or wet gas. Id. at 5. Among the misrepresentations that Plaintiffs allege Defendants made were Sullivan’s statement “that Alpine High’s best wells would ‘generate positive returns even at 0 [dry] gas price,” and Christmann’s statement that “this thing’s going to really hum below $2 on the gas side.” Id. at 38, 41. According to Plaintiffs, “the truth about Alpine High was gradually revealed through a series of corrective disclosures,” the first of which occurred on October 9, 2017, and the last of which occurred on March 13, 2020. Dkt. 101 at 7. Plaintiffs allege, however, that in between these partial corrective disclosures, “Defendants continued to relentlessly promote Alpine High as a transformative discovery for Apache, while consistently waving off doubts about the play’s oil or wet gas potential and profitability.” Dkt. 65 at 42. Plaintiffs seek certification of (1) a class “on behalf of themselves and all other persons or entities who purchased or otherwise acquired Apache common stock from September 7, 2016, through March 13, 2020, inclusive (the ‘Class Period’), and were damaged thereby (the ‘Class’)”; (2) appointment of Plaintiffs as Class Representatives; and (3) appointment of Saxena White P.A. (“Saxena White”) and Kessler Topaz Meltzer & Check, LLP (“KTMC”) as Class Counsel, and Ajamie LLP as Liaison Class Counsel. Dkt. 101 at 7–8. Defendants do not contest class certification for the start of the Class Period, September 7, 2016, through February 22, 2018. The only dispute is whether the Class Period will span an 18- month period, a 42-month period, or something in between. LEGAL STANDARD A. CLASS CERTIFICATION Federal Rule of Civil Procedure 23 governs the inquiry of whether a proposed class should be certified. “[T]he Rule 23 class-action device was designed to allow an exception to the usual rule that litigation is conducted by and on behalf of the individual named parties only.” Califano v. Yamasaki, 442 U.S. 682, 700– 01 (1979). “To come within the exception, a party seeking to maintain a class action must affirmatively demonstrate [its] compliance with Rule 23.” Comcast Corp. v. Behrend, 569 U.S. 27, 33 (2013) (quotation omitted). Rule 23(a) requires that any purported class meet four “prerequisites”: (1) numerosity (a class so large that joinder of all members is impracticable); (2) commonality (questions of law or fact common to the class); (3) typicality (named parties’ claims or defenses are typical of the class); and (4) adequacy of representation (representatives will fairly and adequately protect the interests of the class). Madison v. Chalmette Refin. L.L.C., 637 F.3d 551, 554 (5th Cir. 2011) (cleaned up). These prerequisites—numerosity, commonality, typicality, and adequacy—are necessary but not sufficient conditions for class certification. Rule 23(b) specifies three class types and sets out requirements—beyond those articulated in Rule 23(a)—for each. The putative class here seeks certification under Rule 23(b)(3), which permits class certification where “questions of law or fact common to class members predominate over any questions affecting only individual members, and that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy.” FED. R. CIV. P. 23(b)(3). Rule 23(b)(3) gives four considerations that are “pertinent” to whether predominance and superiority have been established: (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. Id. In considering a motion for class certification, I must “must rigorously consider both Rule 23(a)’s prerequisites and the Rule 23(b) class type.” Chavez v. Plan Benefit Servs., Inc., 957 F.3d 542, 546 (5th Cir. 2020). This rigorous analysis requires me “to go beyond the pleadings to determine whether the requirements of Rule 23 have been met: a court must understand the claims, defenses, relevant facts, and applicable substantive law in order to make a meaningful determination of the certification issues.” Cole v. Gen. Motors Corp., 484 F.3d 717, 724 (5th Cir. 2007) (quotation omitted). “Merits questions may be considered to the extent— but only to the extent—that they are relevant to determining whether the Rule 23 prerequisites for class certification are satisfied.” Amgen Inc. v. Conn. Ret. Plans & Tr. Funds, 568 U.S. 455, 466 (2013). My “obligation . . . to conduct a rigorous analysis of Rule 23’s requirements . . . is not dispensed with by the parties’ stipulation to certification or failure to contest one or more of Rule 23’s requirements.” Ward v. Hellerstedt, 753 F. App’x 236, 244 (5th Cir. 2018). “[T]he court [is] bound to conduct its own thorough . . . inquiry.” Stirman v. Exxon Corp., 280 F.3d 554, 563 n.7 (5th Cir. 2002) (emphasis added). B. THE SECURITIES AND EXCHANGE ACT Section 10(b) of the Securities and Exchange Act of 1934 makes it unlawful for any person, directly or indirectly, . . . [t]o use or employ, in connection with the purchase or sale of any security registered on a national securities exchange . . . any manipulative or deceptive device or contrivance in contravention of such rules and regulations as the [Securities and Exchange] Commission may prescribe as necessary or appropriate in the public interest or for the protection of investors. 15 U.S.C § 78j(b). Rule 10b–5 implements § 10(b) by prohibiting, among other things, the making of any “untrue statement of a material fact” or the omission of any material fact “necessary in order to make the statements . . . not misleading.” 17 C.F.R. § 240.10b–5(b). Plaintiffs may recover damages under § 10(b) by showing: “(1) a material misrepresentation or omission by the defendant; (2) scienter; (3) a connection between the misrepresentation or omission and the purchase or sale of a security; (4) reliance upon the misrepresentation or omission; (5) economic loss; and (6) loss causation.” Erica P. John Fund, Inc. v. Halliburton Co. (“Halliburton I”), 563 U.S. 804, 810 (2011) (quotations omitted). At issue here is the intersection between Rule 23(b)(3)’s predominance requirement and § 10(b)’s reliance requirement. Ordinarily, a plaintiff establishes reliance by showing “that he was aware of a defendant’s misrepresentation and engaged in a transaction based on that misrepresentation.” Goldman Sachs Grp. Inc. v. Ark. Tchr. Ret. Sys., 141 S. Ct. 1951, 1958 (2021). This individualized inquiry is impractical in a class action. So, the Supreme Court has permitted class-action plaintiffs to invoke a presumption of reliance as established in Basic Inc. v. Levinson, 485 U.S. 224 (1988). Under Basic, district courts presume that stock trading in an efficient market incorporates into its price all public, material information—including material misrepresentations—and that investors rely on the integrity of the market price when they choose to buy or sell that stock. See Halliburton I, 563 U.S. at 813. To establish the Basic presumption, Plaintiffs must prove: “(1) that the alleged misrepresentations were publicly known, (2) that they were material, (3) that the stock traded in an efficient market, and (4) that the plaintiff traded the stock between the time the misrepresentations were made and when the truth was revealed.” Halliburton Co. v. Erica P. John Fund, Inc. (“Halliburton II”), 573 U.S. 258, 268 (2014). Once established, Defendants can rebut the Basic presumption by proving “that an alleged misrepresentation did not actually affect the market price of the stock.” Id. at 284. Defendants “must carry that burden by a preponderance of the evidence.” Goldman, 141 S. Ct. at 1963. “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.” Basic, 485 U.S. at 248. Courts “should be open to all probative evidence on that question—qualitative as well as quantitative—aided by a good dose of common sense.” Goldman, 141 S. Ct. at 1960 (quotation omitted). “The 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.” Id. at 1963. The class-certification question of whether Defendants have rebutted the Basic presumption with a preponderance of evidence of no price impact overlaps with merits questions like materiality and loss causation. See id. at 1961. In considering such evidence, I must “resist[] the temptation” to draw merits conclusions. Id. at 1961, n.2 (quotation omitted). Even so, “to maintain the consistency of the presumption with the class certification requirements of [Rule] 23, defendants must be afforded an opportunity before class certification to defeat the presumption through evidence that an alleged misrepresentation did not actually affect the market price of the stock.” Halliburton II, 573 U.S. at 284. When plaintiffs’ theory is that defendants’ misrepresentations or omissions kept their stock artificially inflated, price impact may be shown on the back end. See Goldman, 141 S. Ct. at 1961. Front-end price impact may be inferred from a back-end price drop when the stock price falls after a corrective disclosure, demonstrating that Defendants’ previous statements were untrue or that Defendants failed to disclose the truth. See id. A back-end price drop supports this inference when the corrective disclosure “matches” the earlier misrepresentations or omissions. See id. A corrective “disclosure need not precisely mirror an earlier misrepresentation.” Alaska Elec. Pension Fund v. Flowserve Corp., 572 F.3d 221, 230 (5th Cir. 2009) (cleaned up). The two need only be “related” or “relevant” to one another. Pub. Emps. Ret. Sys. of Miss., P.R. Tchrs. Ret. Sys. v. Amedisys, Inc., 769 F.3d 313, 321 (5th Cir. 2014). However, if there is a “mismatch” between the contents of the corrective disclosure and the misrepresentations or omissions, the inference of price impact is weaker. See Goldman, 141 S. Ct. at 1961. The Supreme Court has given the example of “when the earlier misrepresentation is generic (e.g., ‘we have faith in our business model’) and the later corrective disclosure is specific (e.g., ‘our fourth quarter earnings did not meet expectations’).” Id. “Under those circumstances, it is less likely that the specific disclosure actually corrected the generic misrepresentation, which means that there is less reason to infer front-end price inflation—that is, price impact—from the back-end price drop.” Id. With these principles in mind, I turn to the Motion for Class Certification. ANALYSIS A. THE UNCONTESTED ISSUES Defendants concede that Plaintiffs have satisfied each of the four Rule 23(a) requirements, as well as Rule 23(b)’s superiority requirement. The only dispute is the length of the Class Period. Even so, I must conduct my own inquiry. See Stirman, 280 F.3d at 563 n.7. I have done so, and I also have no trouble finding that Plaintiffs satisfy the requirements of numerosity, commonality, typicality, adequacy, and superiority. 1. Numerosity The putative class is numerous. “During the Class Period, Apache common stock traded on the Nasdaq, NYSE, and Chicago Stock Exchange . . . and had between 375.4 and 382.5 million shares outstanding with an average weekly trading volume of over 21.5 million shares.” Dkt. 101 at 15. Numerosity “is generally assumed to have been met in class action suits involving nationally traded securities.” Zeidman v. J. Ray McDermott & Co., 651 F.2d 1030, 1039 (5th Cir. Unit A July 1981); see also Rooney v. EZCORP, Inc., 330 F.R.D. 439, 445 (W.D. Tex. 2019) (certifying a class where defendant corporation “had more than 50 million shares of Class A common stock outstanding during the class period, and the average weekly trading volume on the NASDAQ Stock Market during the class period was roughly 2.7 million shares”). 2. Commonality There are common questions of law and fact between Plaintiffs and the proposed class, including whether Defendants misrepresented or omitted material facts, scienter, materiality, economic loss, and loss causation. See Dkt. 101 at 16. These questions are “capable of classwide resolution—which means that determination of [their] truth or falsity will resolve an issue that is central to the validity of each one of the claims in one stroke.” Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 350 (2011). Thus, the requirement of commonality is satisfied. 3. Typicality “Like commonality, the test for typicality is not demanding. It focuses on the similarity between the named plaintiffs’ legal and remedial theories and the theories of those whom they purport to represent.” Mullen v. Treasure Chest Casino, LLC, 186 F.3d 620, 625 (5th Cir. 1999) (quotation omitted). “Here, Plaintiffs and all other putative Class members purchased Apache common stock during the Class Period and assert the same Section 10(b) claims, based on the same misstatements and omissions by Defendants, and the same Section 20(a) claims.” Dkt. 101 at 16. Thus, the proposed class is typical. See Marcus v. J.C. Penney Co., No. 6:13-cv-736, 2016 WL 8604331, at *3 (E.D. Tex. Aug. 29, 2016) (“Because the [Plaintiffs] and the Class members both allege that the same misrepresentations and omissions caused the same result (artificially inflating the value of securities), the [Plaintiffs’] claims are typical of the claims of the Class members.”). 4. Adequacy When assessing adequacy, I must consider: (1) the zeal and competence of the representatives’ counsel; (2) the [willingness] and ability of the representatives to take an active role in and control the litigation and to protect the interests of absentees; and (3) the risk of conflicts of interest between the named plaintiffs and the class they seek to represent. Slade v. Progressive Sec. Ins. Co., 856 F.3d 408, 412 (5th Cir. 2017) (cleaned up). Saxena White and KTMC have served as lead counsel in multiple securities class actions, have devoted substantial time and resources to this case, and have demonstrated a willingness to take this case to trial. See Dkts. 101-4, 101-5. Ajamie LLP has served as liaison counsel in numerous securities litigations in this district. See Dkt. 101-6 at 2. Plaintiffs, as “large institutional investors[,] are the preferred named plaintiffs in securities fraud litigation because they generally have the same interests as the plaintiff class.” In re Anadarko Petroleum Corp. Sec. Litig., No. 4:20-cv-00576, 2022 WL 4544235, at *4 (S.D. Tex. Sept. 28, 2022). Plaintiffs’ interests “are also aligned with other class members, as all seek to maximize recovery stemming from the alleged misconduct by [Defendants].” Id. Finally, I am not aware of any conflicts of interest between Plaintiffs and the putative class. Accordingly, the requirement of adequacy is satisfied as to Plaintiffs, Lead Counsel, and Liaison Counsel. 5. Superiority Defendants do not contest “that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy.” FED. R. CIV. P. 23(b)(3). [C]ourts readily agree that class actions are a superior method for resolving security fraud claims for publicly traded stocks. There is little interest in individual investors controlling their own claims, and a class action is more efficient than entertaining a multitude of suits. Finally, Defendants have presented no arguments suggesting the superiority requirement is not satisfied. Lumen v. Anderson, 280 F.R.D. 451, 462 (W.D. Mo. 2012); see also In re Netbank, Inc. Sec. Litig., 259 F.R.D. 656, 676 (N.D. Ga. 2009) (“As a general rule, class action treatment presents a superior method for the fair and efficient resolution of securities fraud cases.” (quotation omitted)). B. RULE 23(b)’S PREDOMINANCE REQUIREMENT Defendants concede that Plaintiffs may avail themselves of the Basic presumption to demonstrate class-wide reliance, and that class-wide issues predominate from the period of September 7, 2016 through February 22, 2018. Defendants contend, however, that they have rebutted the Basic presumption for the time period of February 23, 2018 through March 13, 2020 (the “Focus Period”). Defendants advance a three-pronged attack on the rebuttable fraud-on-the-market presumption, arguing that: (1) there was no front-end price impact attributable to the 15 alleged misrepresentations made during the Focus Period; (2) there was no back-end price impact attributable to the three corrective disclosures alleged during the Focus Period; and (3) Defendants’ purported rebuttal of the Basic presumption is “buttressed” by the fact that “the market’s perception of Alpine High’s oil and gas prospects was essentially unchanged during [the Focus Period].” Dkt. 117 at 28. I will address each argument in turn. 1. The 15 Misrepresentations Alleged During the Focus Period Did Not Cause Any Front-End Price Impact Plaintiffs allege that Defendants made 15 misrepresentations on 13 trading days during the Focus Period. See Dkt. 65 at 112–19, 133–35. Defendants contend the analysis of their own expert, Lucy Allen (“Allen”)—as well as the analysis of Plaintiff’s expert, Dr. Zachary Nye (“Dr. Nye”)—demonstrates that “(1) there was no statistically significant increase on 12 of the 13 trading days following the alleged misrepresentations, and (2) for the one trading date with a statistically significant price increase, all sources attributed the price increase to an EBITDAX and production beat, not to the alleged misrepresentation.” Dkt. 117 at 16–17; see also id. at 18; Dkt. 101-3 at 621–22; Dkt. 117-2 at 24. Plaintiffs do not contest these facts. Rather, Plaintiffs argue that, having conceded price impact prior to the Focus Period, “the law requires Defendants to prove that their alleged misstatements had no price impact whatsoever during the entire Class Period.” Dkt. 120 at 8–9.1 For the reasons discussed in the next section, that is not the law.

1 This is the redacted version of Plaintiffs’ reply brief; the unredacted version has been filed under seal. See Dkt. 121. Plaintiffs’ brief has been redacted where Plaintiffs cite internal Apache documents for the proposition that “Defendants have already produced documents that support both front-end and back-end price impact for the entire Class Period.” Dkt. 120 at 9. The first step in establishing the Basic presumption is demonstrating “that the alleged misrepresentations were publicly known.” Halliburton Because there is no evidence of front-end price impact for the 15 misrepresentations alleged during the Focus Period, I turn to Defendants’ arguments that there also was no back-end price impact attributable to the three corrective disclosures alleged during the Focus Period. 2. Whether the Three Corrective Disclosures Alleged During the Focus Period Demonstrate Price Impact Plaintiffs allege three corrective disclosures during the Focus Period: (1) an April 23, 2019 Press Release; (2) the October 25, 2019 resignation of the geologist who led the Alpine High project, Steven Keenan (“Keenan”); and (3) a March 16, 2020 article by Seeking Alpha, a crowd-sourced content service that publishes news on financial markets. In opposing class certification as to the Focus Period, Defendants argue that none of these three disclosures corrected any alleged misrepresentation—in other words, that there is a “mismatch.” Defendants also argue that there was either no statistically significant price impact following these disclosures, or that any statistically significant price impact following these disclosures is attributable to something other than the disclosure. In reply, Plaintiffs contend that “undisputed evidence of front-end price impact [in September 2016] renders each of Defendants’ back-end price impact arguments irrelevant.” Dkt. 120 at 16–17. In support of this extraordinary argument—which would render Goldman essentially meaningless—Plaintiffs cite In re Chicago Bridge & Iron Co. N.V. Securities Litigation (“Chicago Bridge & Iron”) for the proposition that “‘when Plaintiffs are able to show an alleged misrepresentation had a statistically significant front-end price impact, Defendants are not entitled to rely on [] additional back-end arguments to rebut

II, 573 U.S. at 268 (emphasis added). “[U]ndisclosed information cannot drive down the market price of a stock. Only information known to the market can cause a loss. For this reason, only information known to the market is relevant under the fraud-on-the-market theory of class wide reliance.” Alaska Elec. Pension Fund, 572 F.3d at 230. By definition— and evidenced by Plaintiffs’ redaction of the information contained therein—internal documents are not public. Thus, for the public’s convenience, I will cite to Plaintiffs’ redacted brief. I have not relied on the redacted portions of Plaintiffs’ brief because non- public documents are simply not relevant evidence in assessing price impact. the Basic presumption.’” Dkt. 120 at 17 (emphasis omitted) (quoting No. 17-cv- 1580, 2020 WL 1329354, at *3 (S.D.N.Y. Mar. 23 2020)). But this quote implies only that Defendants are not entitled to rely on back-end arguments to rebut the Basic presumption for the misrepresentations that Plaintiffs have shown had a statistically significant front-end price impact. Yet, Defendants are not contesting front-end price impact for the alleged September 2016 misrepresentations, and they do not seek to rebut the Basic presumption as to those misrepresentations. Rather, Defendants seek to rebut the Basic presumption as to the 15 misrepresentations—made between 18 and 40 months later—that Plaintiffs allege in order to extend the Class Period. “In the case of a securities fraud class action, . . . a class period ends when the truth has been disseminated to the market.” Carpenters Pension Tr. Fund of St. Louis v. Barclays PLC, 310 F.R.D. 69, 97 (S.D.N.Y. 2015) (emphasis added) (cleaned up). Despite this straightforward proposition, Plaintiffs argue that, to limit the Class Period, Defendants must “prove by a preponderance of the evidence[] if, when, or how [the] admitted positive price impact [from September 2016] was fully removed before the start of [the Focus Period].” Dkt. 120 at 9 (emphasis added). Plaintiffs again cite Chicago Bridge & Iron for the proposition that “an absence of back-end price impact does not rebut the statistically significant front-end responses to the alleged misrepresentations.” Id. at 10 (quoting 2019 WL 5287980, at *4 (S.D.N.Y. Oct. 18, 2019) (cleaned up)). But the fact that the September 2016 front-end price impact cannot be rebutted does not require Defendants to affirmatively demonstrate total dissipation of that price impact before they may contest the impact (if any) of the 15 misrepresentations that Plaintiffs allege Defendants made 18 to 40 months later. The standard, stated plainly in Basic, is that “[a]ny 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.” 485 U.S. at 248. Courts do not certify class periods based on the total dissipation of front-end price impact. Rather, “courts are required to cut off the class period on the date of a statement or event that cures the market.” Carpenters, 310 F.R.D. at 97 (cleaned up).2 With this principle in mind, I will consider, in reverse chronological order (as that will determine the Class Period’s end point), whether Defendants have rebutted the Basic presumption by demonstrating mismatch or lack of back-end price impact for the three corrective disclosures alleged during the Focus Period. a. March 16, 2020 Seeking Alpha Post It bears repeating that “a class period ends when the truth has been disseminated to the market.” Id. According to Plaintiffs’ own allegations: 106. On February 26, 2020, Apache issued a press release announcing fourth quarter and full-year 2019 results, reporting a $3 billion impairment for Alpine High, a staggering amount that more than wiped out the Company’s profits for the prior two years, and resulted in the Company reporting a loss of $3 billion during the fourth quarter and $3.6 billion for the full-year 2019. The Company further announced that, as of the end of 2019, “there were no rigs drilling at Alpine High.” 107. Leading media outlets specifically highlighted how the massive $3 billion write down and cessation of all further exploration at Alpine High was “in stark contrast” with Defendants’ statements during the Class Period “vehemently defending the play’s prospects for about three years.” Bloomberg, for example, issued a scathing article on February 27, 2020, titled “Apache Calls It Quits on Alpine High After $3 Billion Write down,” which stated: Apache Corp. is officially calling it quits on a highly publicized but disappointing shale discovery in

2 Maybe revelation of the truth removes any price impact; maybe it doesn’t. See Dkt. 126- 2 at 14 (“Simple logic dictates that . . . determining whether any alleged inflation came out before the Focus Period is irrelevant. . . . [E]ither i) the pre-Focus Period alleged misrepresentations had no impact at all, ii) the alleged inflation from the pre-Focus Period alleged misrepresentations came out of Apache’s stock price before the Focus Period, or iii) the alleged inflation was still in the stock after the alleged Class period and is therefore not relevant to Plaintiffs’ allegations in this case.”). Either way, the question is when the market learned the truth about Alpine High, not when the September 2016 front-end price impact fully dissipated. West Texas after vehemently defending the play’s prospects for about three years. The Houston-based company posted a roughly $3 billion write down on its Alpine High project, a find from 2016 that fizzled when it turned out to hold more natural gas than oil . . . The discovery was announced in September 2016 to much fanfare and claims the field held 3 billion barrels of crude and 75 trillion cubic feet of gas. . . . Until recently, Apache executives defended the Alpine High, saying in May that investors didn’t yet “have an appreciation for the potential cash flow generation from the liquids play at Alpine High.” 108. The same day, February 27, 2020, the Company held its fourth quarter earnings webcast. At the outset, Defendant Christmann acknowledged that 2019 was a year full of “challenges,” the “most significant” being “Alpine High.” Christmann explained the Company was throwing in the towel on any further development of Alpine High, stating that “further testing is not warranted at this time” and Apache “dropped the remainder of our drilling rigs in the fourth quarter and chose to defer some previously planned completions.” Where analysts previously viewed the capital allocation Apache was committing to Alpine High as a reason for optimism, Defendant Christmann highlighted the Alpine High failure, explaining that in “terms of capital allocation, Alpine High will receive minimal to no funding.” Dkt. 65 at 51–52 (emphasis omitted and added). Despite Plaintiffs’ own extensive allegations regarding public news coverage of the failed Alpine High project in February 2020, Plaintiffs nevertheless contend the Class Period should extend through “March 16, 2020, [when] a Seeking Alpha article disclosed that, due to lack of production from Alpine High amidst declining oil prices, Apache was severely financially strained and not competitive, with the highest debt-to-equity ratio of all large-cap independent oil producers.” Dkt. 120 at 29. Plaintiffs also contend—for the first time in this litigation—that in addition to the Seeking Alpha article, a Susquehanna analyst report issued the same day “revealed the falsity of Defendants’ representations that Alpine High was, among other things, a world class resource that ‘put Apache in one of the most exciting and competitive positions in the industry’ and would drive shareholder value for years to come.”3 Id. (alterations omitted). But the market had known for weeks that these statements were false because Apache had already taken a $3 billion write down and ceased all exploration and funding of Alpine High. See Dkt. 65 at 51. Moreover, four days earlier, “on March 12, 2020, [Apache] announced that it was forced to slash its vaunted dividend by a staggering 90%.” Id. at 52. All the Seeking Alpha article and Susquehanna report demonstrated was that “Apache finds itself relative to its peers in the E&P industry in a laggard position at the back of the pile, at the back of the race.” Dkt. 156 at 26. But revealing the depths of Apache’s financial struggles did not shed any new light on Apache’s alleged misrepresentations about Alpine High. Accordingly, the Class Period cannot extend beyond February 2020.4 b. October 25, 2019 Keenan Resignation The next disclosure I must address came on October 25, 2019, the date that Keenan resigned, and “the biggest intraday drop [of Apache’s stock] since January 2016.” Dkt. 65 at 50. Plaintiffs argue that “[a]nalysts and the mainstream media easily linked Keenan’s departure to Alpine High troubles.” Id. Defendants do not contest that there was a statistically significant price impact to Apache’s stock following the news of Keenan’s resignation. Rather, Defendants retort that (1) Plaintiffs fail to “explain how the October 25, 2019 article announcing Keenan’s resignation was corrective of any alleged misrepresentation [regarding Alpine High],” and (2) “analysts attributed the October 25, 2019 decline in Apache’s stock price to concerns about Apache’s Suriname exploration—not to any new news about Alpine High.” Dkt. 117 at 24. Specifically, Defendants note that “all three

3 Neither Plaintiffs’ Complaint, nor their Motion for Class Certification, nor their expert’s report alleged that the Susquehanna report was a corrective disclosure. 4 Because any price movement that may have occurred after the May 16, 2020 Seeking Alpha article and Susquehanna report cannot be linked to any alleged misrepresentations regarding Alpine High, I do not reach the question of whether there was a statistically significant price reaction following these publications. [analyst reports issued on October 25, 2019] directly attributed Apache’s stock price decline to the market’s concerns about Apache’s Suriname exploration in light of Keenan’s resignation, not Alpine High.” Id. at 24–25. Additionally, Defendants point out that “no analyst changed its reserves estimate for Alpine High following this announcement [of Keenan’s resignation].” Id. at 25. In reply, Plaintiffs argue that any discussion of “how the announcement of Keenan’s resignation corrected any alleged misstatements” is an “improper loss causation argument.” Dkt. 120 at 26. As discussed above, that is not the law. Indeed, Plaintiffs’ very next sentence demonstrates that such arguments are permissible at the class certification stage, because Plaintiffs go on to note that “courts regularly sustain as corrective disclosures resignations of corporate executives.” Id. In other words, courts routinely entertain arguments about whether corporate executive resignation announcements are corrective of prior alleged misrepresentations. See, e.g., Ferris v. Wynn Resorts Ltd., No. 2:18-cv- 00479, 2023 WL 2337364, at *9–11 (D. Nev. Mar. 1, 2023) (discussing defendants’ mismatch argument at length, despite plaintiffs’ argument that mismatch was an improper loss-causation argument). “Consistent with Goldman Sachs, the Court considers all evidence of price impact for each of the . . . alleged Corrective Disclosures.” Ramirez v. Exxon Mobil Corp., No. 3:16-cv-03111, 2023 WL 5415315, at *15 (N.D. Tex. Aug. 21, 2023). So, the real question is whether the news of Keenan’s resignation was corrective of any alleged misrepresentations regarding Alpine High. Recall that a disclosure is considered corrective only when it “reveals new facts that, taken as true, render some aspect of the defendant’s prior statements false or misleading.” Ferris, 2023 WL 2337364, at *10 (emphasis added) (quotation omitted). Plaintiffs assert that “Keenan’s resignation plainly ‘render[ed] some aspect of the defendant[s’] prior statements false or misleading.” Dkt. 120 at 27 (quoting Ferris, 2023 WL 2337364, at *10). Yet, Plaintiffs do not elaborate on what aspect of the defendants’ prior statements was rendered false or misleading that was not already known by the market. Plaintiffs cannot point to any new information revealed by the news of Keenan’s resignation. Rather, the news sources that Plaintiffs cite in the live pleading regarding Keenan’s resignation were clearly repeating old news. See Dkt. 65 at 50 (“Credit Suisse described how Keenan ‘oversaw the discovery of the Alpine High play, which has been an economic disappointment for investors[,]’ . . . [n]oting that ‘the outcome of results from Alpine High have not met high expectations.’”). More importantly, Defendants have presented compelling evidence that the market reacted to the news of Keenan’s departure for a reason wholly unrelated to Alpine High: Apache’s work in Suriname. The title of the Credit Suisse report issued on October 25, 2019 says it all: “Resignation of Exploration Head Highlights Suriname Risk to Share Price.” Dkt. 117-10 at 2. Credit Suisse went on to note that “[t]oday’s sell-off . . . highlights the high expectations for the well already baked into [Apache]’s stock price.” Id. Similarly, RBC Capital Markets stated: “We think [Apache] share weakness is a reaction to investor concern that the resignation is related to the outcome of [Apache]’s Maka-1 exploration well in Suriname.” Dkt. 117-11 at 2. Likewise, SunTrust Robinson Humphrey stated: “Apache’s stock underperformed this morning . . . on investor speculation that a SVP’s resignation . . . is linked to an upcoming unsuccessful Suriname Maka-1 exploration well.” Dkt. 117-12 at 2. The abundance of analyst speculation about what Keenan’s resignation meant for Suriname suggests that Apache’s prospects in Suriname are what the market considered significant. Plaintiffs retort that because [a]n analyst report [from RBC Capital Markets] published at 10:19 a.m.—approximately 35 minutes after the market first learned of Keenan’s resignation—disclosed to the market that the resignation was unrelated to Suriname . . ., any Suriname-related price impact was removed well before the market closed on October 25, 2019, and could not have caused the statistically significant decline . . . found in Apache’s stock price at market close on October 25. Dkt. 120 at 28 (emphasis omitted). But Plaintiffs overlook that at the same time RBC Capital Markets disabused the public of the notion that Keenan’s resignation concerned Suriname, it also speculated—in the very same report—that “[Apache] share weakness is a reaction to investor concern that the resignation is related to the outcome of [Apache]’s Maka-1 exploration well in Suriname.” Dkt. 117-11 at 2. In fact, Plaintiffs’ own expert testified that “after [Apache] confirmed that [Keenan’s] resignation had nothing to do with Suriname,” there was “a really quick rebound in the market price.” Dkt. 156 at 108. The price rebound following confirmation that Keenan’s resignation was unrelated to Suriname shows that it is more likely than not that concerns about Suriname are what moved the market. Plaintiffs go on to argue that “Apache’s own documents belie Defendants’ claim that Keenan’s departure was unrelated to the alleged misstatements.” Dkt. 120 at 28. There are two problems with this argument. First, internal company documents are entirely irrelevant to determining whether the market was reacting to public information that revealed some truth about Defendants’ alleged misrepresentations. See Alaska Elec. Pension Fund, 572 F.3d at 230 (“Only information known to the market can cause a loss.”). Second, Defendants have never argued that Keenan’s departure was unrelated to the alleged misstatements about Alpine High because that too is irrelevant. The news of Keenan’s resignation is not automatically a corrective disclosure simply because Alpine High was the driving force behind Keenan’s resignation. That is not the test. The test is whether “the disclosure reveals new facts that, taken as true, render some aspect of the defendant’s prior statements false or misleading.” Ferris, 2023 WL 2337364, at *10 (quotation omitted). I conclude that Defendants have rebutted the Basic presumption by a preponderance of the evidence by showing that the news of Keenan’s resignation did not reveal anything new about Alpine High and that the market moved out of concern about Apache’s prospects in Suriname. c. April 23, 2019 Press Release The last disclosure to address is an April 23, 2019 press release—issued before the market opened—announcing the temporary deferral of Alpine High natural gas production. Apache’s stock price declined 11% over the four days following this press release. Plaintiffs contend this disclosure was corrective because it showed that Apache “was scaling back natural gas production efforts.” Dkt. 65 at 134. Defendants counter that the press release does not support an inference of price impact for three reasons: (1) there was no statistically significant decline in Apache’s stock price thereafter, according to both Nye’s event study model and the alternative event study model tested by Allen; (2) the market expected such news given “extremely” low regional gas prices; and (3) the disclosure did not correct any alleged misrepresentation. Dkt. 117 at 20. I will address each argument in turn. i. Statistical Significance of the Price Reaction The parties agree that there was no statistically significant—where “statistically significant” means a 95% confidence level—stock price decline on the first, second, third, or fourth days after the April 23, 2019 press release. Nor was there a statistically significant price drop using a 2-day window of April 23 and April 24. Dr. Nye is only able to show a statistically significant price drop at a 95% confidence level or higher by extending his event window out to the third and fourth days. Specifically, “Nye’s analysis shows that the 2-, 3-, and 4-day returns for this disclosure are significant at the 92.58%, 98.96%, and 99.01% confidence levels, respectively.” Dkt. 120 at 21. “Allen’s event study yields similar results, with 3- and 4-day returns significant at the 94.33% and 96.56% confidence levels, respectively.” Id. at 22. Defendants contend Dr. Nye’s method is impermissibly results-driven. Plaintiffs retort that (1) “[c]ourts routinely hold that multi-day event windows are appropriate to evaluate stock price declines, including where defendants argue there was no statistically significant price decline on the first day of the window,” and that (2) “numerous courts have explicitly rejected” the argument that “statistical significance at or above the 95% confidence level is required to show price impact.” Dkt. 120 at 21–22. I will address these arguments in reverse order. As to the confidence level, Plaintiffs do not seriously contest that a 95% confidence level is the scientific and legal standard. See, e.g., Federal Judicial Center, Reference Manual on Scientific Evidence 381 (3d ed. 2011) (“Traditionally, scientists adopt the 95% level of confidence, which means that if 100 samples of the same size were drawn, the confidence interval expected for at least 95 of the samples would be expected to include the true population value.”). Rather, Plaintiffs argue that lack of statistical significance alone cannot rebut the Basic presumption. I agree. After all, Apache’s stock did decline in the four days after the April 23, 2019 press release. I also agree that “Defendants’ authorities [on this point] are readily distinguishable.” Dkt. 120 at 22 n.12 (“In Erica P. John Fund, Inc. v. Halliburton Co., 309 F.R.D. 251, 270 (N.D. Tex. 2015) (“Halliburton III”), the plaintiffs’ expert conceded that 95% statistical significance was required to demonstrate price impact, and the court in In re Intuitive Surgical Sec. Litig., 2016 WL 7425926, at *15 (N.D. Cal. Dec. 22, 2016) simply cited, without analysis, Halliburton III as the basis for requiring 95% statistical significance.”). As a matter of law, “[t]he lack of statistically significant proof that a statement affected the stock price is not statistically significant proof of the opposite, i.e., that it did not actually affect the stock price.” Di Donato v. Insys Therapeutics, Inc., 333 F.R.D. 427, 444 (D. Ariz. 2019) (emphasis omitted). That said, a fact can be relevant without being dispositive. Accordingly, the fact that none of the four days following the press release individually, or the first two days combined, showed statistically significant price declines—while not dispositive of price impact—is nevertheless a relevant fact to consider alongside Defendants’ other evidence concerning price impact. Another relevant fact to consider is that, to get to a 95% confidence interval, Dr. Nye extended his event window to the third and fourth days following the April 23, 2019 press release. Allen’s event study only showed statistical significance using a four-day event window. At the hearing, I gave Plaintiffs an opportunity and additional time to submit “any cases . . . where the Court approves a . . . multi-day window [where] none of the days are [individually] statistically significant.” Dkt. 156 at 258; see also id. at 255 (“And I guess I’m just curious, are there any cases anywhere finding a price impact using, say, a three-plus-day window when no one day is statistically significant, right?”). Following the hearing, Plaintiffs directed me to six cases that they contend approved “event windows of three or more days.” Dkt. 155 at 1. None are availing. For example, Plaintiffs’ best case is In re Groupo [sic] Televisa Securities Litigation, No. 18-cv-1979, 2020 WL 3050550 (S.D.N.Y. June 8, 2020). This is the only case to consider a four-day event window where no individual day was statistically significant. But the facts of Grupo Televisa are especially unique because the four-day period corresponded to an executive’s four days of testimony, which the court treated as a single event. See In re Grupo Televisa Sec. Litig., No. 18-cv-1979, 2022 WL 2829253, at *2 (S.D.N.Y. July 20, 2022) (“Burzaco’s testimony was the ‘single event’ which conveyed information to the Televisa investors severing the link between when alleged misrepresentations and the stock price.”). Dr. Nye offers no comparable reason for extending the event window here. In re Twitter, Inc. Securities Litigation, No. 16-cv-05314, 2020 WL 4187915 (N.D. Cal. Apr. 17, 2020), also involved a four-day event window. But unlike here, in Twitter, the one-day price decline following the corrective disclosure was statistically significant. See Report on Loss Causation and Damages at 65, In re Twitter, Inc. Sec. Litig., No. 16-cv-05314 (N.D. Cal. Sept. 18, 2019), ECF No. 352- 9. Thus, Twitter is inapposite. In Sjunde AP-Fonden v. General Electric Co., the plaintiffs’ expert “provide[d] a full explanation for [a three-day event] window based on an unusually positive price reaction on the announcement day followed by a statistically significant two-day decline.” No. 17-cv-8457, 2023 WL 6314939, at *16 (S.D.N.Y. Sept. 28, 2023) (emphasis added); see also Expert Report of David I. Tabak, Ph.D. at 35–36, Sjunde AP-Fonden v. Gen. Elec. Co., No. 17-cv-8457 (S.D.N.Y. Sept. 6, 2022) (Dkt. 355-12) (“Notably, however, the three-day price movement was statistically significant, not just considered on its own, but after ‘penalizing’ the measure of statistical significance for examining six trading days (i.e., the six original alleged corrective disclosure dates listed in Exhibit 4) and further ‘penalizing’ the measure for considering two analyses of October 20, 2017, a full-day analysis and an intraday analysis.”). The takeaway from these cases is that “where Plaintiffs seek to extend the window, they must show a sound conceptual and evidentiary basis for that extension.” Allegheny Cnty. Emps.’ Ret. Sys. v. Energy Transfer LP, 623 F. Supp. 3d 470, 485 (E.D. Pa. 2022). Dr. Nye’s reason for extending the event window in this case is that “the market was having a tough time reading through that deferral announcement and how it affected gas production for the full year of ’19 and down the road in 2020 as well.” Dkt. 156 at 105. This is certainly a reason for extending the event window, but it does not strike me as a compelling reason on-par with those offered in the cases discussed above. Even so, I do not consider the lack of statistically significant price declines on each of the four days following the April 23, 2019 press release; or the unusually long event window required to find a statistically significant price decline; or the questionable basis for that protracted event window dispositive of whether Defendants have rebutted the Basic presumption. Stated differently, the lack of statistical significance cannot, standing alone, rebut the presumption. But these are relevant points that I will consider alongside Defendants’ other arguments regarding price impact. ii. The Market’s Expectations and Correctiveness In addition to questioning the statistical significance of the price reaction following the April 23, 2019 press release, Defendants contend that “the market expected the news in the April 23 press release given the extreme pricing environment at the Waha Hub, a major hub for natural gas flowing out of the Permian.” Dkt. 117 at 21. Defendants also contend that “the press release was not corrective of any alleged misrepresentation.” Id. at 23. For reasons that will soon become clear, these issues are intertwined and best addressed together. Plaintiffs counter that any argument regarding the market’s expectation “is a truth-on-the-market defense, which is a ‘materiality’ argument that ‘has no bearing on the predominance inquiry of class certification.’” Dkt. 120 at 22 (quoting Marcus v. J.C. Penney Co., 2016 WL 8604331, at *8 (E.D. Tex. Aug. 29, 2016)). Similarly, Plaintiffs argue that whether the press release corrected any alleged misrepresentation “raises a premature loss causation challenge.” Dkt. 120 at 25. Plaintiffs notably fail to cite a post-Goldman authority for these arguments that, as previously discussed, are incorrect. “[D]efendants must be afforded an opportunity before class certification to defeat the presumption through evidence that an alleged misrepresentation did not actually affect the market price of the stock.” Halliburton II, 573 U.S. at 284. Plaintiffs’ more compelling argument is this: Defendants’ argument that the market fully expected the deferral announced on April 23, 2019 is premised on just three analyst reports (Opp. 16-17), in which each analyst speculates that investors may have expected the announced deferral. . . . Further undermining Defendants’ claim, at least eight analysts and several news outlets characterized this Corrective Disclosure as unexpected, negative news. . . . Tellingly, neither Defendants nor Allen identify any other news that caused Apache’s stock price to decline between April 23 and April 26, 2019 (Nye Reb., ¶20), further demonstrating Defendants’ failure to prove a complete lack of price impact. . . . Defendants’ fallback suggestion that “common sense” supports that the market expected the Alpine High deferral (Opp. 17-18) is also meritless. Apache did not announce the deferral until well after the Waha Hub natural gas spot price was negative, at which point it had turned positive. Opp. 18 (graphic). In fact, on April 23, 2019, the Waha Hub natural gas spot price was roughly the same as it had been in November 2018 and February 2019, and Apache did not defer gas extraction at Alpine High during either period. Id. “Common sense” instead dictates that, in the face of Defendants’ repeated reassurances that Alpine High would perform well even at low commodity prices, investors viewed the April 23, 2019 Corrective Disclosure as new, negative news. Dkt. 120 at 23–24. Defendants retort that Plaintiffs have mischaracterized the analyst reports that they say regarded the April 23, 2019 press release “as unexpected, negative news.” Id. at 23 (emphasis added). Defendants highlight that the reports “say nothing about whether the news was unexpected.” Dkt. 126 at 21. To the contrary, investors “likely expected” deferral. Id. (emphasis omitted) (quoting Dkt. 117-8 at 2); see also Dkt. 156 at 75 (demonstrating that five days prior to the April 23, 2019 press release, at least one investor remarked that he was “hearing more chatter in the investment community about Alpine High gas shut-in risk,” which Dr. Nye conceded was “probably used synonymously [with deferral]”). Defendants also contend that while three analysts lowered their Alpine High production estimates after the press release, . . . two of the analysts termed the reduction as having “minimal” impact, and the other issued an industry report that was updating estimates for many other E&P companies and linked the Alpine High “shut-ins” to the low Waha pricing. . . . No analysts, including the three Dr. Nye claims lowered their production estimates, [or] lowered their price targets of Apache after the press release, which indicates that the deferral was expected. Dkt. 126 at 22. As for Plaintiffs’ contrasting Apache’s (in)actions in November 2018 and February 2019 with Apache’s April 2019 deferral—when prices were roughly the same—Defendants note that, unlike November 2018 and February 2019, “starting in March of 2019, Waha prices actually went negative,” meaning a seller is “paying the buyer to take . . . gas.” Dkt. 156 at 52. Finally, Plaintiffs are correct that Defendants do not point to any other news that caused Apache’s stock price to decline following the April 23, 2019 press release. But as with Keenan’s resignation, Defendants argue the news was significant for reasons unrelated to Alpine High. Specifically, Defendants contend that uncertainty about how long prices would remain low caused Apache’s stock price to decline. See id. at 69–70. I appreciate both sides’ arguments. But common sense tells me that investors expected the deferral, and any decline in stock price was due to uncertainty about low gas prices generally, not Alpine High specifically. Moreover, Defendants may have repeatedly reassured investors that Alpine High would perform well even at low commodity prices, but they never said it would perform well at negative prices. In other words, news of the deferral was not corrective of any alleged misrepresentation. Thus, I conclude that Defendants have rebutted the Basic presumption by a preponderance of the evidence by showing that the April 23, 2019 press release did not reveal anything new about Alpine High and that Apache’s stock price decline was more likely than not due to uncertainty about the historically low Waha Hub gas prices.5 * * * There was no front-end price impact attributable to the 15 alleged misrepresentations made during the Focus Period, and Defendants have rebutted the Basic presumption as to each of the three corrective disclosures alleged during the Focus Period. Accordingly, the Class Period should be limited to the time period of September 7, 2016 to February 22, 2018. CONCLUSION For the reasons explained above, I recommend that the Motion for Class Certification (Dkt. 101) be GRANTED in part and DENIED in part. Specifically, I recommend that the Court: (1) certify a class consisting of Plaintiffs and all other persons or entities who purchased or otherwise acquired Apache common stock from September 7, 2016, through February 22, 2018, and were damaged thereby6;

5 Because I have already concluded Defendants have rebutted the Basic presumption, I do not reach the parties’ arguments regarding whether Defendants’ rebuttal of the Basic presumption is “buttressed” by the fact that “the market’s perception of Alpine High’s oil and gas prospects was essentially unchanged during [the Focus Period].” Dkt. 117 at 28. 6 Excluded from the Class are Defendants, the officers and directors of Apache, members of their immediate families and their legal representatives, heirs, agents, affiliates, successors or assigns, Defendants’ liability insurance carriers, and any affiliates or (2) appoint Plaintiffs as Class Representatives; and (3) appoint Saxena White and KTMC as Class Counsel and Ajamie LLP as Liaison Class Counsel. The parties have 14 days from service of this Memorandum and Recommendation to file written objections. See 28 U.S.C. § 636(b)(1)(C); FED. R. CIV. P. 72(b)(2). Failure to file timely objections will preclude appellate review of factual findings and legal conclusions, except for plain error. SIGNED this 9th of February 2024.

______________________________ ANDREW M. EDISON UNITED STATES MAGISTRATE JUDGE

subsidiaries thereof, and any entity in which Defendants or their immediate families have or had a controlling interest.