Bartlett v. BP West Coast Products LLC

Opinion

□ PERSIAN GULF INC., Individually CASE NO. 15cv1749-JO-AGS and on Behalf of All Others Similarly Situated, ORDER GRANTING DEFENDANTS’ MOTIONS Plaintiff, FOR SUMMARY JUDGMENT Vv. BP WEST COAST PRODUCTS LLC, al., Defendants. RICHARD BARTLETT, et al., Lead Case No. 18-cv-1374-JO- Individually and on Behalf of All AGS (consolidated with No.18-cv- Others Similarly Situated, 1377-JO-AGS) Plaintiffs, □ V. BP WEST COAST PRODUCTS LLC, 77 et al., Defendants. 15cv1749-JO-AGS and 18cv1374-JO-AGS

In this putative class action for antitrust conspiracy, Defendants Chevron U.S.A. Inc. (“Chevron”), Exxon Mobil Corporation and ExxonMobil Refining & Supply Co. (“Exxon”), Phillips 66, BP West Coast Products LLC (“BP”), Tesoro Refining & Marketing Company LLC (“Tesoro”), Equilon Enterprises LLC (d/b/a Shell Oil Products US) (“Shell”), Valero Marketing and Supply Company (“Valero”), and Alon USA Energy, Inc. (“Alon”) (together, “Defendants”) filed motions for summary judgment. Dkts. 615, 619, 625. Defendants also filed motions to exclude the expert testimony of Plaintiffs’ proffered experts: Robert McCullough, Dr. Paul Hanouna, and Dr. Michael Williams. Dkts. 613, 616. Plaintiffs have similarly filed motions to exclude the testimony of Defendants’ proffered experts: Andrew Lipow, Dr. Janusz Ordover, and Dr. Richard Bergin. Dkts. 622, 626. For the reasons stated below, the Court grants Defendants’ motions for summary judgment. Dkts. 615, 619, 625. The Court also grants in part Defendants’ motion to exclude the expert testimony of Dr. Williams and Dr. Hanouna on the issue of causation. Dkt. 616. The parties’ remaining motions to exclude expert testimony, including Defendants’ motion to exclude Dr. Williams’ and Dr. Hanouna’s testimony on issues outside of causation, are dismissed as moot. Dkts. 613, 622, 626. Plaintiff Persian Gulf Inc. (“Persian Gulf’), the operator of a retail gas station, filed antitrust lawsuit on behalf of retail stations in California on July 7, 2015. See Dkt 1.' June 21, 2018, individual consumers Joshua Ebright, Paul Lee, and David Rinaldi (the ©. ! Unless otherwise noted, citations to “Dkt.” refer to Persian Gulf, Inc. v. BP West Coast Products LLC, et al., 1Scv1749-JO-AGS. 15cv1749-JO-AGS and 18cv1374-JO-AGS

“Consumer Plaintiffs”) filed two separate lawsuits on behalf of consumers who purchased gasoline in California. These lawsuits alleged that eight current and former gas refiners in California—Defendants Chevron, Phillips 66, BP, Tesoro, Shell, Valero, Exxon, and ; Alon—conspired to fix gas prices in California from 2012 to present in violation of § 1 of Sherman Act, Cartwright Act, Cal. Bus. & Prof. Code § 16700, et seg., and § 17200 of the Cal. Bus. & Prof. Code, commonly known as the UCL. See Dkt. 76; Bartlett, Dkt. 44. On July 25, 2018, the Court consolidated the two Consumer Plaintiffs’ cases into action. Bartlett, Dkt. 37. Thereafter, the Court ordered the coordination of Persian Gulfs and Consumer Plaintiffs’ cases for discovery and motion briefing because the allegations were nearly identical. See Dkt. 143. Accordingly, the Court set a single briefing schedule governing both Persian Gulf’s and the Consumer Plaintiffs’ cases which /included deadlines for motions for summary judgment and motions to exclude expert testimony. See Dkt. 589. After exhaustive discovery proceedings, Defendants Chevron, Shell, Valero, and Phillips 66 filed a joint motion for summary judgment, arguing that Plaintiffs did not have evidence to support a reasonable inference of conspiracy or causation. See Dkt. 625 (“Joint MSJ”). The remaining Defendants joined the Joint MSJ, and Defendants Alon and Tesoro also filed separate motions for summary judgment. See Dkts. 615, 619, 630, 632, 634, 636. In addition, the parties moved to exclude one another’s expert reports under Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993). See Dkts. 613, 616, 622, 626. © ? See Bartlett et al v. BP West Coast Products LLC et al., 18cv1374-JO-AGS; Rinaldi et al. v. BP West Coast Products LLC et al., 18-cv-1377-J ace 15ev1749-JO-AGS and 18cv1374-JO-AGS

A. Background Information on the Gasoline Market in California Plaintiffs’ price fixing allegations are best understood within the larger context of California’s gas market and its unique supply-chain challenges. The Court, therefore, provides the following brief overview of the players in the market, how the market is supplied, and how gas is sold in the state. In California, a small group of refiners control the entirety of gas production in the See, e.g., Dkt. 622, Ex. 2 (“Lipow Report”) 7 31; Dkt. 647, Ex. 1 (“McCullough Report”) Ff 20, 66.2 This highly concentrated market is comprised of the eight Defendants this case, including gas giants like Exxon and Chevron,’ plus additional non-Defendant /refiners with varying market shares.” Because Defendants do business in a highly concentrated market with few players, they are admittedly conscious of one another’s pricing and actions in the market, as the actions of any one refiner can substantially impact other refiners. See, eg., J. Hodgson Declaration §§ 11-15, 19-20; P. Brooks —————-- 3 The Court declines to rule on the admissibility of the expert opinions contained in the McCullough and Lipow reports as moot. Where there is no dispute, however, the Court has referenced these reports as sources of background information about the gas industry. 4 For instance, evidence in the record suggests that Exxon may have accounted for 8% of California gas supply in 2015, see Dkt. 629-1 (“Defs. Exs.”) Defs. Ex. 1 at 144:17-145:9, and Plaintiffs’ expert suggests that Chevron may have accounted for 18% of California gas production during the class period. See McCullough Report § 67. > The evidence in the record also indicates that additional non-Defendant refiners may have contributed substantially to California gas production at various points during the class period. See, e.g., id. (noting that non-Defendant PBF accounted for 12% of production and non-Defendant Marathon accounted for up to 30% of production). 15cv1749-JO-AGS and 18cv1374-JO-AGS

Declaration J 96; K. Archambault Declaration 4 59; J. Harris Declaration { 6; M. O’Neal Declaration ¥ 3.° The California gas market largely depends on this small group of in-state refiners because California is a “gasoline island” isolated from other sources of supply. Dkt. 629- (“Defs. Exs.”) Defs. Ex. 2; McCullough Report J] 37-39. California suffers from a lack of direct pipeline connectivity to other major refinery centers, such as the Gulf Coast and the Pacific Northwest. See McCullough Report 4{ 37-39; Dkt. 722, Joint Statement of Undisputed Facts (“Joint Statement”) 9 12. Thus, California can only receive imports by which is costly and requires weeks of lead time. See, e.g., Joint Statement J 12; K. Archambault Declaration [] 22—26, 34; H. Henderlite Declaration 24-26; Dkt. 699 }(“Opposition”) at 29. Given refiners’ finite production capacity and the limited options for external supply, prices in the California gas market are sensitive to events such as refinery shutdowns. See, e.g., Opposition at 1; Joint Statement 9 15. According to the Attorney General of California, because of these factors, “California’s gasoline market has been characterized by high gas and diesel prices and recurrent price spikes.” See Defs. Ex. 2 at 1. The California gas market also faces another unique limitation: only a specific gas formulation called CARBOB may be sold in the state. California refiners produce a variety gas products, such as diesel, jet fuel, and different formulations of gas, including {| ———x“« Unless otherwise stated, citations to declarations in this opinion refer to the declarations in 93 Defendants’ Joint Appendix (hereinafter, “J.A.”) in support of their Joint Motion for Summary Judgment. Dkt. 629-2. Similarly, citations to Chevron’s, Shell’s, Exxon’s, Phillips 66’s, Valero’s, and BP’s exhibits (e.g, Chevron Ex. 1) refer to the exhibits to the declarations in the Joint Appendix at Dkt. 629-2. 15ev1749-JO-AGS and 18cv1374-JO-AGS

CARBOB. Joint Statement 7 17. Within California, however, gas must meet the California Reformulated Gasoline Blendstock for Oxygenate Blending standard, known as CARBOB. Joint Statement FF 11, 13; 220 CCR § 2266.5. Refiners sell CARBOB directly to retail stations either (1) “at the rack,” i.e., distribution terminals where retailers can pick up gas; or (2) via “Dealer Tankwagon,” i.e., delivery by truck from the rack to the retail station. Joint Statement § 18. Refiners set the wholesale prices for gas sold to retailers at the rack and by Dealer Tankwagon. See C. Yates Declaration { 6; D. Smith Declaration □□□□ C. Dickson Declaration 7 27; J. Hodgson 9 4-5; M. O’Neal Declaration 42; P. Brooks Declaration ff 93-94; K. Archambault Declaration 55-56; Dkt. 621-1—2, Tesoro’s Appendix (“T.A.”), W. Eckard Declaration { 21. 1] Once gas reaches retail stations, it is sold by retail stations directly to consumers at the pump. Retail stations in California are owned either by refineries or by independent parties. Joint Statement f{ 19-20. In the latter scenario, independent retail stations can enter licensing agreements with refiners that give them the right to sell gas under a refiner’s brand. Jd. § 20. For example, Plaintiff Persian Gulf purchased gas from Phillips 66 via Deal Tankwagon and licensed the right to sell under the Phillips 66 brand. □□□ Independent owners, like Plaintiff Persian Gulf, autonomously set prices at their retail stations, while refiners set prices at their corporate-owned retail stations. See, e.g.,C. Yates Declaration § 7; P. Brooks Declaration 4 9. California refiners, like many of the eight Defendants in this case, also buy and sell CARBOB and other gas products to cover their supply shortages or dispose of excess supply. Joint Statement § 21; C. Yates Declaration § 8; M. Perez Declaration □□ 4—5; L. Lockhart Declaration § 4. In order to buy and sell, as described above, refiners like the Defendants employ gas traders to trade on the “spot market,” a trading market for gas on 15ev1749-JO-AGS and 18cv1374-JO-AGS

the West Coast. Joint Statement § 21. Gas traders perform two primary functions for their respective refineries, (1) purchasing gas to cover production shortages, and (2) selling gas to compensate for excess production. See, e.g., Lipow Report F920, 22, 26. When executing spot-market transactions, West Coast traders either communicate directly with another or through independent brokers on a bid-ask basis, typically negotiating within a range of the current spot-market price, reflected by pricing agencies such as the Oil Price Information Service (“OPIS”). See J.A. 562—76, Phillips Ex. 17; M. O’Neal Declaration 1194; C. Dickson Declaration § 59; L. Lockhart Declaration §§ 12-13; H. Henderlite Declaration { 14. West Coast traders can refer to OPIS prices when trading because OPIS publishes the daily high, low, and average West Coast spot prices based on a sampling of actual trades executed in the spot market that day. See id. B. Plaintiffs’ Conspiracy Case 1. Plaintiffs’ Initial Conspiracy Allegations In Plaintiffs’ complaints, they initially alleged that Defendants entered a price-fixing conspiracy on or around February 2012. See Dkt. 76 at 64; Bartlett, Dkt. 44 at 46. While complaints included allegations that Defendants manipulated supply to raise prices and }entered into exchange agreements in furtherance of a conspiracy,’ Plaintiffs’ original theory of the case centered on allegations that Defendants utilized the cover of refinery maintenance to raise prices and take advantage of the reality that the California gas market is sensitive to refinery outages. See generally id. Plaintiffs claimed that Defendants ;planned and synchronized unnecessary maintenance in order to raise prices. See id. 7 See Dkt. 76 at 32, 43, 67; Bartlett, Dkt. 44 at 26, 35-41. 15ev1749-JO-AGS and 18cv1374-JO-AGS

Further, Plaintiffs alleged that certain of Defendants’ outages were not outages at all and emissions data demonstrated that Defendants continued to produce despite claiming to shut down. See id. Based on the evidence and arguments presented to the Court on summary judgement, Plaintiffs appear to have wholly abandoned these “refinery maintenance” allegations following discovery. See generally Opposition. 2. Plaintiffs’ Current Evidence of Conspiracy Plaintiffs now argue and seek to prove that the eight Defendants conspired to fix gas prices by engaging in multiple coordinated actions designed to reduce supply and raise gas prices in California. Plaintiffs describe a conspiracy where Defendants blocked imports, increased exports, and lowered production levels to keep gas supply low in California, all while manipulating public facing gas-market prices to increase their profits. See generally Opposition. This conspiracy was purportedly made possible by Defendants’ systemic exchanges of sensitive information and cooperation to cover each other’s supply shortages. See id. As set forth in their discovery responses, Plaintiffs maintain that Defendants entered into, and began participating in, this illegal agreement in 2011: “Defendants, each and all, agreed, and entered into an agreement . . . by no later than August 1, 2011, to fix, maintain, or make artificial prices for gasoline sold in California.” Dkt. 455, Ex. C at 12; Dkt. 455, Ex. D at 13. In their summary judgment opposition, Plaintiffs point to the following in support lof their allegations of a price fixing conspiracy: (1) Defendants entered into exchange agreements with one another throughout the class period to lend each other barrels of gas; (2) Defendants systematically exchanged confidential information with one another; (3) Defendants restricted gas supply in California, including by running their refineries below capacity, preventing gas imports, and unnecessarily exporting gas out of California; 15ev1749-JO-AGS and 18cv1374-JO-AGS

(4) In 2015, Defendants cooperated with Exxon after its refinery explosion instead of leveraging the explosion to gain market share; (5) Defendants used various strategies to manipulate public facing gas prices, including through a “gentleman’s agreement,” wash \ trades, selective price reporting, and false public statements. Ill. LEGAL STANDARDS A. Summary Judgment Standards Summary judgment is appropriate under Rule 56 of the Federal Rules of Civil Procedure if the moving party demonstrates the absence of a genuine issue of material fact and entitlement to judgment as a matter of law. Celotex Corp. v. Catrett, 477 U.S. 317, (1986). A fact is material when, under the governing substantive law, it could affect the outcome of the case. Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248 (1986). A dispute as to a material fact is genuine if there is sufficient evidence for a reasonable jury /to return a verdict for the nonmoving party. Id. at 248-50. A party seeking summary judgment always bears the initial burden of establishing absence of a genuine issue of material fact. Celotex, 477 U.S. at 323. The moving can satisfy this burden in two ways: (1) by presenting evidence that negates an essential element of the nonmoving party’s case; or (2) by demonstrating that the /nonmoving party failed to establish an essential element of the nonmoving party’s case on which the nonmoving party bears the burden of proof at trial. Jd. at 322-23. The court must view all inferences drawn from the underlying facts in the light most favorable to the nonmoving party. Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 587— 88 (1986). At the summary judgment stage, the parties have the burden to support their motion and opposition with evidence and specific references to the record that they wish the Court 15ev1749-JO-AGS and 18cv1374-JO-AGS

to consider. Fed. R. Civ. P. 56(c); Carmen v. San Francisco Unified Sch. Dist., 237 F.3d 1026, 1031 (9th Cir. 2001). It is not the Court’s task “to scour the record in search of a genuine issue of triable fact. [The Court relies] on the nonmoving party to identify with reasonable particularity the evidence that precludes summary judgment.” Keenan v. Allen, 91 F.3d 1275, 1279 (9th Cir. 1996); Schneider v. TRW, Inc., 938 F.2d 986, 990 n.2 Cir. 1991) (“district court is under no obligation to mine the full record for issues of triable fact”); Asset Mktg. Sys. Ins. Servs., LLC v. McLaughlin, 2007 WL 3232507, at *1 Cal. Nov. 1, 2007) (failure to cite to evidence “is a complete failure of proof”). Pointing to an admissible expert opinion is one way that a party can create a triable of material fact. Nevertheless, courts need not defer to an expert when the evidence clear and conflicts with the expert’s testimony. In re Apple Computer Sec. Litig., 886 F.2d 1109, 1116 (9th Cir. 1989); Stephens v. Union Pac. R.R. Co., 935 F.3d 852, 856-57 (9th Cir. 2019) (“[e]xpert testimony cannot create a genuine issue of material fact if it rests /on assumptions that are not supported by evidence”). Moreover, “an expert report cannot be used to prove the existence of facts set forth therein.” Jn re Citric Acid Litig., 191 F.3d 1090, 1102 (9th Cir. 1999); see also Rebel Oil Co., Inc. v. Atl. Richfield Co., 51 F.3d 1421, 1440 (9th Cir. 1995). Specifically, in the antitrust context, “[e]xpert testimony is useful as a guide to interpreting market facts, but it is not a substitute for them.” Brooke Grp. Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209, 242 (1993); Matsushita, 475 U.S. /at 594 n.19. Standards Governing Antitrust Cases On summary judgment in an antitrust conspiracy case, a party can move to show that is no genuine dispute of material fact on any of the essential elements of a plaintiffs price fixing claim: (1) conspiracy to fix prices in violation of antitrust law; (2) injury—or a 10 15¢v1749-JO-AGS and 18cv1374-JO-AGS

“impact”—resulting from that violation; and (3) damages. 15 U.S.C. § 15; Olean Wholesale Grocery Coop., Inc. v. Bumble Bee Foods LLC, 31 F.4th 651, 665-66 (9th Cir. 2022); see also Big Bear Lodging Ass’n v. Snow Summit, Inc., 182 F.3d 1096, 1101-02 (9th Cir. 1999). Injury in the antitrust context is “injury of the type that the antitrust laws were intended to prevent and that flows from that which makes the defendants’ acts }unlawful.” Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 489 (1977). Damages are measured only after antitrust impact has been demonstrated. Olean, 31 F.4th at 666 (9th Cir. 2022). On the first element, conspiracy, a genuine issue of material fact can be established either by direct evidence that Defendants agreed to fix prices or circumstantial evidence which a reasonable factfinder could infer that Defendants entered into such an agreement. Citric Acid, 191 F.3d at 1093. When the evidence is circumstantial, the “crucial question” for the Court is “whether all the evidence considered as a whole can reasonably support the inference that [Defendants] conspired” to fix prices. Jd. at 1097. The Ninth Circuit applies the following two-step framework when a □□□□□□□□□□□ conspiracy allegations are based solely on circumstantial evidence. At step one, “the defendant(s] can ‘rebut an allegation of conspiracy by showing a plausible and justifiable reason for its conduct that is consistent with proper business practice.’” Jd. at 1094 (quoting Richards v. Neilsen Freight Lines, 810 F.2d 898, 902 (9th Cir. 1987)). Defendants may satisfy this burden by showing that the allegedly conspiratorial action “was in each defendant’s independent self-interest.” Jd. at 1095. At step two, “[t]he burden then shifts back to the plaintiff to provide specific evidence tending to show that the defendant was not engaging in permissible competitive behavior.” Jd. at 1094. Once Defendants have met their burden at step one, Plaintiffs must “come forward with specific factual support il 15cv1749-JO-AGS and 18cv1374-JO-AGS

for its allegations of conspiracy” to avoid summary judgment. Barnes v. Arden Mayfair, 759 F.2d 676, 680 (9th Cir. 1985) (citation omitted). Plaintiffs can meet this “step two” burden by offering proof of parallel conduct lamong alleged conspirators alongside “plus factors’—that is, additional circumstantial evidence that when combined with parallel-conduct evidence, reasonably supports an inference of conspiracy. In re Musical Instruments & Equip. Antitrust Litig., 798 F.3d 1186, 1193-94 (9th Cir. 2015). Conscious parallelism—parallel conduct by competitors who adopt similar policies around the same time for the same reasons—is neither “in itself unlawful” nor uncommon. /d. at 1193; Stanislaus Food Prods. Co. v. USS-POSCO Indus., F.3d 1084, 1092 (9th Cir. 2015); Brooke Grp., 509 U.S. at 227; Theatre Enters., Inc. v. Paramount Film Distrib. Corp., 346 U.S. 537, 540-41 (1954). In fact, conscious parallelism or parallel conduct is a common occurrence in concentrated, interdependent markets among players who “recogniz[e] their shared economic interests and their interdependence with respect to price and output decisions.” Bell Atl. Corp. v. Twombly, 550 U.S. 544, 553-54 (2007) (quoting Brooke Grp., 509 U.S. at 227). Parallel conduct alone, therefore, is not sufficient to support an inference of conspiracy, but “it is a relevant factor to be considered along with the evidence as a whole” in determining whether the facts give rise to a reasonable inference of conspiracy. Citric Acid, 191 F.3d at 1102. In order to distinguish between lawful conscious parallelism and unlawful conspiracy, courts require that additional “plus factor” evidence that “tend[s] to rule out the possibility that the defendants were acting independently.” Twombly, 550 U.S. at 554 \(citing Matsushita, 475 U.S. 574); see also Monsanto Co. v. Spray-Rite Serv. Corp., 465 U.S. 752, 763-64 (1984). “[P]lus factors are economic actions and outcomes that are largely inconsistent with unilateral conduct but largely consistent with explicitly 12 15cv1749-JO-AGS and 18cv1374-JO-AGS

coordinated action.” Musical Instruments, 798 F.3d at 1194. Plus factors may include circumstantial evidence demonstrating “a common motive to conspire, evidence that shows that the parallel acts were against the apparent individual economic self-interest of the alleged conspirators, and evidence of a high level of inter-firm communications,” among things. Ross v. Citigroup, Inc., 630 F. App’x 79, 82 (2d Cir. 2015), as corrected (Nov. 24, 2015) (citation omitted). Whatever form this “plus factor” evidence takes, it must support the inference that the Defendants’ conduct is more consistent with conspiracy than with unilateral decision. Musical Instruments, 798 F.3d at 1194. “[C]onduct as consistent with permissible competition as with illegal conspiracy does not, standing alone, support an inference of antitrust conspiracy” at the summary judgment stage. Matsushita, 475 U.S. at 587-88. Before turning to the substance of the summary judgment and expert exclusion motions, the Court briefly addresses the parties’ evidentiary objections. Plaintiffs made ;over 350 objections to Defendants’ declarations, while Defendants, in turn, objected to over 100 of Plaintiffs’ exhibits. See Dkt. 699-1; Dkt. 719-2. The bulk of the objections on both sides consisted of “boilerplate” objections on various grounds unsupported by explanation—e.g., “Hearsay,” “Lack of foundation and personal knowledge regarding sale of Carson refinery,” and “Best Evidence Rule.” See id. Defendants also objected to the admissibility of all three of Plaintiffs’ expert reports because they were unsworn and unaccompanied by a declaration. See Dkt. 719-2. Given the high volume of boilerplate objections, the Court will only rule on the objections to evidence that the Court considers in ruling on the motions currently before it. See Doe v. Starbucks, Inc., 2009 WL 5183773, at *1 (C.D. Cal. Dec. 18, 2009) (“it is often 13 15cv1749-JO-AGS and 18cv1374-JO-AGS

unnecessary and impractical” to rule on each evidentiary objection at summary judgment, “especially when many of the objections are boilerplate recitations of evidentiary principles blanket objections without analysis applied to specific items of evidence”); Pinson v. Prieto, 2014 WL 7339203, at *4 (C.D. Cal. Dec. 19, 2014) (declining to rule on objections to evidence that was immaterial to the summary judgment decision). The Court therefore addresses below the categories of evidence relevant to its rulings. Plaintiffs’ Objections to Declarations Submitted by Defendants Defendants submitted the declarations of over twenty company executives and traders to testify about specific actions the company took and to explain the business reasons behind those actions. See Dkt. 629-2. Plaintiffs objected to these declarations on grounds of hearsay, lack of personal knowledge, the best evidence rule, and improper legal conclusions. See Dkt. 699-1. I. Personal Knowledge Objections to Defendants’ 30(b)(6) Declarations First, the Court declines to reject declarations submitted by Defendants containing Rule 30(b)(6) testimony on the grounds that the corporate witnesses lacked personal knowledge. See Opposition at 57-60. By definition, corporate witnesses (also called “30(b)(6) witnesses”) testify regarding the company’s knowledge, not the individual’s personal knowledge. Fed. R. Civ. P. 30(b)(6). A 30(b)(6) witness designated to testify to matters on behalf of the company has a duty to prepare by reviewing documents, speaking with witnesses, and otherwise gathering the information available to the company on the topics the witness has been designated to testify. See id; see also Bd. of Trs. of Leland Stanford Junior Univ. v. Tyco Int’l Ltd., 253 F.R.D. 524, 526 (C.D. Cal. 2008) (noting that Rule 30(b)(6) explicitly requires a company to prepare designees to testify on its behalf as to all matters reasonably available to the company; personal knowledge of the deponent is 14 15cv1749-JO-AGS and 18cv1374-JO-AGS

“of no consequence”). Thus, a 30(b)(6) witness’s testimony is determined by the limits of the company’s knowledge, not the individual’s personal knowledge. See id.; Cooper v. United Air Lines, Inc., 82 F. Supp. 3d 1084, 1096 (N.D. Cal. 2015) (overruling objections based on personal knowledge with respect to 30(b)(6) declarations on summary judgment). Because 30(b)(6) witnesses may testify on behalf of the company at trial, they are similarly allowed to do so at summary judgment. Fraser v. Goodale, 342 F.3d 1032, 1036-37 (9th Cir. 2003) (finding that evidence that can be presented in an admissible form at trial is admissible on summary judgment); see also Univ. Healthsystem Consortium v. UnitedHealth Grp., Inc., 68 F. Supp. 3d 917, 921 (N.D. Ill. 2014). In response to Plaintiffs’ objections for lack of personal knowledge, Defendants explained that the following declarations (or portions thereof) contain designated 30(b)(6) corporate testimony: K. Archambault (BP), C. Yates (Chevron), C. Dickson (Exxon), R. Sharum (Phillips 66), J. Harris (Shell), and P. Brooks (Valero). See Dkt. 719-2. Defendants have made a showing that the above corporate witnesses were designated to testify on broad aspects of Defendants’ businesses, including trading strategy, imports and exports, maintenance, public communications, trade associations events, exchange agreements, refinery production, trading, and price setting. C. Yates Declaration 7 5; J. Harris Declaration {J 4-5; C. Dickson Declaration 4; R. Sharum Declaration □ 1; P. Brooks Declaration J 5 & n.1; K. Archambault Declaration 75. Plaintiffs did not argue otherwise; while they raised blanket “personal knowledge” objections to these declarations, they did not argue that any testimony was outside the scope of the topics for these 30(b)(6) witnesses. See Opposition 57-60; Dkt. 699-1. The Court, therefore, overrules Plaintiffs’ personal knowledge objections: K. Archambault Declaration JJ 23-— 34, 39, 66, 68-70, 72, 74; C. Yates Declaration §§ 35-36, 46, 54, 56-67; C. Dickson 15 15ev1749-JO-AGS and 18cv1374-JO-AGS

Declaration ff 13, 23; R. Sharum Declaration 10-17, 19, 20-21, 23, 25-28, 30-33; J. Harris Declaration 17, 21, 24; P. Brooks Declaration {J 22, 26, 32, 36, 38, 47-49, 54, 58-59, 72-75, 82, 85, 87-88, 90, 103, 105. 2. Personal Knowledge Objections to Defendants’ Non-30(b)(6) Declarations Second, the Court overrules Plaintiffs’ objections to Defendants’ non-30(b)(6) declarations to the extent that the declarants’ personal knowledge can be inferred from the declarations themselves. Federal Rule of Civil Procedure 56(c)(4) requires declarations to be based on personal knowledge, but that requirement “imposes only a ‘minimal □□□□□□□□□ Strong v. Valdez Fine Foods, 724 F.3d 1042, 1045 (9th Cir. 2013) (citation omitted). In determining whether this requirement is met, the Court can infer personal knowledge from /the declaration itself, including the declarant’s role in the company, the declarant’s participation in certain matters, and the declarant’s statements that her declaration is based on personal knowledge. Barthelemy v. Air Lines Pilots Ass ’n, 897 F.2d 999, 1018 (9th Cir. 1990) (personal knowledge of declarants can be “inferred from their positions and the nature of their participation in the matters to which they swore”); Sea-Land Serv., Inc. v. Lozen Int’l, LLC., 285 F.3d 808, 819 (9th Cir. 2002). After a thorough review of the declarations from Defendants’ corporate executives and traders, the Court concludes that the personal knowledge requirement is satisfied with regard to the following: J. Yomtoob Declaration ff 13, 17, 24-25; H. Henderlite Declaration 9] 16, 24-27, 30; S. Roveda Declaration §J 10-11, 13; M. Perez Declaration 4 10; J. Hodgson Declaration ff 4, 9-12, 15, 20, 23, 27-28; N. Weinberg-Lynn Declaration 1199 10-14, 18-21, 24-27, 31, 32; D. Smith Declaration ff 3, 10; J. Marino Declaration □□□ 3— 4; S. Rodrick Declaration J§ 9-10; L. Lockhart Declaration [J 4, 7-8, 21, 29-30, 32, 43, 47; E. Pestano Declaration §§ 43-45, 50; R. Plumier Declaration 7 13; G. Johnson 16 15cv1749-JO-AGS and 18cv1374-JO-AGS

Declaration J 7-8, 10; M. Pais Declaration J 13. Each of the above non-30(b)(6) declarants was either a high-level employee for a Defendant and intimately involved with that Defendant’s business practices or was a trader offering information relevant to trading. See Dkt. 629-2. In addition, every declarant swore that his or her declaration was based on personal knowledge and that each could competently testify with respect to the information described. See id. The foregoing allows the Court to infer personal knowledge from the declarations themselves, and thus, Plaintiffs’ personal knowledge objections to the above testimony are overruled. See Strong, 724 F.3d at 1045; Barthelemy, 897 F.2d at 1018. 3. Objections to Non-Hearsay Testimony in Defendants’ Declarations Third, the Court overrules Plaintiffs’ hearsay objections to Defendants’ declarations to the extent that Plaintiffs objected to evidence that does not contain an out-of-court statement or is not being offered for its truth. Hearsay is defined in Federal Rule of Evidence 801 as an out-of-court statement offered to prove the truth of the matter asserted the statement. Fed. R. Evid. 801(c); United States v. Lopez, 913 F.3d 807, 826 (9th Cir. }2019). While all declarations are technically out-of-court statements, declarations offered in support of summary judgment are not hearsay if the testimony could be presented in an admissible form at trial. See Fraser, 342 F.3d at 1036-37. Plaintiffs frequently objected to portions of Defendants’ declarations as hearsay without explanation, but a review of those portions reveals that this testimony, if presented live at trial, would not be out-of- statements. Rather, the testimony comprised statements of the declarant’s personal ;knowledge of business practices, the market, annual meetings, efc. See, e.g., K. Archambault Declaration § 27; C. Yates Declaration { 36. In other instances, where the testimony incorporated an out-of-court statement, the statement was not being offered for the truth of the matter asserted. See, e.g., P. Brooks Declaration {48 (Homeland Security’s 17 15cv1749-JO-AGS and 18cv1374-JO-AGS

statement offered to show effect on Valero). Based on the foregoing, the Court overrules Plaintiffs’ hearsay objections to the following testimony: K. Archambault Declaration 27; J. Yomtoob Jf 16, 22; C. Yates Declaration J 36; H. Henderlite Declaration JJ 24— 30; P. Brooks Declaration Jf 48, 72, 74; R. Pluimer Decl. § 6; R. Sharum Declaration /4 10; S. Rodrick Declaration ¥ 10; L. Lockhart J 47, 50. The Court does sustain Plaintiffs’ hearsay objections where the evidence takes the form of out-of-court statements offered for their truth and no hearsay exception appears to apply. Fed. R. Evid. 801-803. In accordance with these principles, the Court sustains the following hearsay objections: J. Yomtoob Declaration J 17 (the Court does not consider statements to the extent they are being offered for truth rather than to demonstrate effect on the listener); H. Henderlite Declaration { 26 (the Court does not consider BPWCP-6 to the extent it is being offered for the truth); N. Weinberg-Lynn Declaration § 32 (the Court not consider PSX-4—5); R. Sharum Declaration □ 30 (the Court does not consider PSX-12 or PSX-14). 4. Objections to Defendants’ Business Records as Hearsay Fourth, over Plaintiffs’ objections, the Court will consider documents that meet the business records exception to hearsay. The business records exception provides that a writing is admissible if, (1) it is “made or transmitted by a person with knowledge at or the time of the incident recorded,” and (2) “is kept in the course of regularly conducted business activity.” United States v. Miller, 771 F.2d 1219, 1237 (9th Cir. 1985); Fed. R. /Evid. 803(6). Here, Plaintiffs objected to certain of Defendants’ exhibits consisting of emails, slide decks, and internal business reports as hearsay without further explanation. See Dkt. 699-1. Defendants responded that these documents were admissible because they were records kept in the ordinary course of business. See Dkt. 719-2. The Court agrees 18 15cv1749-JO-AGS and 18cv1374-JO-AGS

that where the documents appear to have been created regularly pursuant to Defendants’ business activities (e.g., export reports or financial presentations) and the accompanying declarations indicated that they were created at or near the time of the event, the business records exception applies. For the above reasons, the Court overrules Plaintiffs’ hearsay objections to the following exhibits because it concludes that the business records exception to the hearsay rule applies: K. Archambault Declaration § 68; J.A. □□□□□□□ Exhibit BPWCP-1; S. Roveda Declaration Jf 10-11; J.A. 52-67, Chevron Exs. □□□□ J.A. 428-48, Phillips Exs. 1-3; J.A. 649-88, Phillips Ex. 24; P. Brooks Declaration □□□□ 26, 79; 806-13, Valero Ex. 15; G. Johnson Declaration J 7—8, 10. 5. Objections to Defendants’ Testimony Discussing Documents Fifth, the Court overrules Plaintiffs’ objections to testimony discussing documents to the extent those objections are based on a misapprehension that the best evidence rule imposes a blanket prohibition on such testimony. That is not the case. The best evidence provides that “[iJn proving the contents of a writing,” the original is required unless certain exceptions apply. Fed. R. Evid. 1002. For the most part, the declarants’ references to documents were not attempts to prove the contents of those documents but were comments upon state of mind or beliefs about the contents. In addition, most of the documents referenced are already in the record. In other cases, declarants were simply testifying to their personal knowledge and do not reference documents, let alone attempt to prove their contents. For these reasons, the Court overrules Plaintiffs’ best evidence objections to the following pieces of evidence: K. Archambault Declaration {{] 35, 66, 67, 69-70; J. Yomtoob Declaration ff 16-17, 22, 24; C. Yates Declaration {J 25, 35, 56; H. Henderlite Declaration §¥ 23, 25, 30; J. Hodgson Declaration Jf 9, 13-14; N. Weinberg- Lynn Declaration 10, 13-14; R. Sharum Declaration ff 11, 17, 19, 20, 23, 25—28, 30— 19 15cv1749-JO-AGS and 18cv1374-JO-AGS

33; D. Smith Declaration § 3; J. Marino Declaration {fj 4, 10-15; J. Harris Declaration 199 19, 21-24, 26; S. Rodrick Declaration {{ 7, 9-10; P. Brooks Declaration {J 36, 88; L. Lockhart Declaration J§ 42, 43-45, 47; R. Pluimer Declaration §§ 6, 11; G. Johnson Declaration § 7-8, 10. 6. Objections to Declarations Containing Improper Legal Conclusions Finally, turning to Plaintiffs’ objections regarding improper legal conclusions, such objections are unnecessary because the Court only considers facts contained in declarations, not legal conclusions or argumentative statements. “[S]tatements in declarations [containing] improper legal conclusions, or argumentative statements . . . will not be considered on a motion for summary judgment. Objections on any of these grounds simply superfluous in this context.” Burch v. Regents of Univ. of Cal., 433 F. Supp. 2d 1110, 1119 (E.D. Cal. 2006). Plaintiffs objected to Defendants’ declarations containing statements such as, that a decision was “in furtherance of individual businesses interest,” a sale was in a Defendant’s “economic interest,” that an action was “economically rational,” that a trade was “not a wash trade,” or that “Plaintiffs are wrong.” See, e.g., K. Archambault Declaration § 27; R. Sharum Declaration § 12; S Rodrick Declaration { 10. The Court agrees that these types of statements are not facts and therefore, the Court does consider them. Nonetheless, where a paragraph containing an improper legal conclusion is otherwise bookended by admissible statements, as is often the case here, the Court has considered the admissible factual evidence but disregarded the improper legal conclusions and argumentative statements. B. Defendants’ Evidentiary Objections Plaintiffs submitted over 200 exhibits in support of their oppositions to summary judgment. See Dkts. 619-3—16, 689-2—4, 697-1, 781-2 (“Pltfs. Exs.”). Defendants objected a 20 15cv1749-JO-AGS and 18cv1374-JO-AGS

many of these exhibits on grounds of relevance, hearsay, and Plaintiffs’ failure to identify certain documents in their interrogatory responses. See Dkt. 719-2. Defendants also argued that Plaintiffs’ expert reports were inadmissible because they were unsworm and unaccompanied by an affidavit. See id. I. The Court Considers Only Relevant Evidence First, the Court declines to rule on Defendants’ objections on the grounds of relevance because it is unnecessary to resolve the motions before the Court. Relevance is duplicative of the summary judgment standard. Sandoval v. Cnty. of San Diego, 985 F. 3d 657, 665 (9th Cir. 2021). “[I]f evidence submitted on summary judgement could create a genuine dispute of material fact, it is, by definition . . . relevant,” and if it cannot create a genuine dispute of material fact, “there is no need for the court to separately determine whether it is relevant.” Jd. Accordingly, the Court will not further address Defendants’ relevance objections. 2. Defendants’ Hearsay Objections to its own Statements Second, the Court overrules Defendants’ hearsay objections to the extent that Defendants objected to evidence of their own statements, which is not hearsay. An out-of- court statement offered for its truth is not hearsay if a party’s own statement or their agent’s statement is being offered against them. Fed. R. Evid. 802(d)(2). A statement is admissible this context when an employee of a defendant makes a statement in the scope of her }employment. Jd. at 802(d)(2)(D). Plaintiffs submitted evidence consisting of emails and instant messages among Defendants’ employees discussing logistics, deals, and trades. See Pitfs. Exs. 1-198. These statements appear to have been made in the context of {;employment, and Defendants did not offer any basis for the Court to conclude otherwise. 15cv1749-JO-AGS and 18cv1374-JO-AGS

Accordingly, the Court overrules Defendants’ hearsay objections to Plaintiffs’ exhibits that contain communications made by Defendants or their agents.® 3. Defendants’ Objections to Admitting Professor Severin Borenstein’s Blog Post Third, the Court sustains Defendants’ objection to the blog post of Professor Severin Borenstein on the ground that it is inadmissible hearsay. Pltfs. Ex. 2. In ruling on Defendants’ objection, the Court considers whether the out-of-court statements contained \in the blog post are being offered for their truth and, if so, whether there is an applicable hearsay exception. See Fed. R. Evid. 801, 803. In the blog post, Professor Borenstein opined on the causes of gas prices in California. See Pltfs. Ex. 2. In opposition to summary judgment, Plaintiffs cited to Professor Borenstein’s blog and stated that he “demonstrated that factors internal to refiners, not externalities, caused . . . higher prices,” and described his conclusions to support their own claims of anticompetitive conduct. See Opposition at 1-2; Pltfs. Ex. 2. Because Plaintiffs offered Professor Borenstein’s blog statements for their truth to bolster their own claims, the exhibit is inadmissible hearsay unless a hearsay exception applies. Although Plaintiffs did not argue that any hearsay exception applied, the Court confirms that there is no applicable exception. The learned treatise exception to hearsay permits treatises, periodical, or pamphlets if, (1) an expert relies upon it on direct examination or it is called to the expert’s attention on cross-examination, and (2) the publication is established as a reliable authority. See Fed. R. Evid. 803(18); Diodem, LLC v. Lumenis Inc., 2005 WL 6220667, at *6 (C.D. Cal. Jan. 10, 2005) (discussing the dual j___— SSS 8 Except where the element of conspiracy is met with respect to Defendants in this opinion, the Court finds that the co-conspirator exception (allowing statements made by a party’s co-conspirator during and in furtherance of the conspiracy to be cdmited agains the party) does not apply. 15cv1749-JO-AGS and 18cv1374-JO-AGS

requirements of Fed. R. Evid. 803(18)). This exception has not been established here; therefore, the Court will not consider the blog post for its truth. 4. Documents Not Identified in Interrogatory Responses Fourth, over Defendants’ objections, the Court will consider Plaintiffs’ evidence not cited in their interrogatory responses because the Court does not have the necessary information to exclude this evidence. Rule 26(e) requires a party to supplement discovery responses “in a timely manner” when it learns that the response is incomplete or incorrect and the information has not otherwise been made known to the opposing party. Fed. R. Civ. P. 26(e). Failure to comply with Rule 26(e) can result in the exclusion of evidence unless the failure to supplement was substantially justified or harmless. Fed. R. Civ. P. 37(c). Whether supplementation is timely under Rule 26(e) hinges on facts such as when the original disclosure was made, when a party discovered its response was incomplete, when the supplementation was made in relation to those events, etc. None of those facts are in the record, let alone facts that would aid the Court in determining whether a late disclosure was justified or harmless under Rule 37(c). Accordingly, on the present record, Court overrules Defendants’ objections to exhibits not previously disclosed in Plaintiffs’ interrogatory responses. 5. Objections to Plaintiffs’ Expert Reports Lastly, the Court overrules Defendants’ objections to Plaintiffs’ expert reports as unverified because the reports have been sufficiently verified through deposition testimony. Liebling v. Novartis Pharm. Corp., 2014 WL 12576619, at *1 (C.D. Mar. 24, 2014) (collecting cases). “[F]or an expert opinion to be considered summary judgment, it must be accompanied by a proper affidavit or deposition testimony; courts in the Ninth Circuit have routinely held that unsworn expert reports are 15ev1749-JO-AGS and 18cv1374-JO-AGS

inadmissible.” FNBN-RESCON I LLC v, Ritter, 2014 WL 979930, at *5 (D. Nev. Mar. 12, 2014) (citation and internal quotations omitted); King Tuna, Inc. v. Anova Food, Inc., 2009 WL 650732, at *1 (C.D. Cal. Mar. 10, 2009) (expert report can be verified by deposition summary judgment purposes). While these expert reports are unaccompanied by declarations, the parties have submitted excerpted deposition testimony of Plaintiffs’ experts that authenticates their reports. Defs. Exs. 3, 8, 9, 13-14, 22, 25-27. Based on the deposition testimony in the record, the verification requirement is met. A. Plausibility of Allegations As a threshold matter, the Court considers whether Plaintiffs allege a plausible theory of conspiracy that makes practical economic sense for the alleged conspirators. Matsushita, 475 U.S. at 596. Matsushita requires courts to engage in this analysis because “‘[t]he absence of any plausible motive [for Defendants] to engage in the conduct charged is highly relevant to whether a ‘genuine issue for trial’ exists.” Jd.; Stanislaus, 803 F.3d at 1090 (“[w]e first analyze whether the alleged conspirators would have had a rational motivation to conspire”). The question is not whether Defendants would have a motive to conspire generally but whether Defendants would have a motive to conspire specifically in the way alleged by Plaintiffs. Matsushita, 475 U.S. at 577-82 (finding no rational motive under plaintiffs’ theory that 21 defendants conspired over a 20-year period to cut prices in hopes of expanding their market share in the future—defendants had no incentive to suffer losses based on speculative future profits); see also Eastman Kodak Co. v. Image Tech. Servs., Inc., 504 U.S. 451, 468 (1992). This plausibility inquiry does “not introduce a special burden on plaintiffs facing summary judgment in antitrust cases,” but merely 24 15cv1749-JO-AGS and 18cv1374-JO-AGS

articulates “that the nonmoving party’s inferences be reasonable in order to reach the jury.” Kodak, 504 U.S. at 468. Here, Plaintiffs allege that the eight Defendants engaged in a wide ranging conspiracy ranging from 2011 to the present to fix prices at supracompetitive levels. See generally Opposition. According to Plaintiffs, Defendants orchestrated a multi-faceted conspiracy that operated as follows: Defendants agreed to reduce supply, by cutting production and imports and increasing exports, in order to raise prices. See id. at 26-36. Pursuant to Plaintiffs’ theory of conspiracy, Defendants also engaged in deceptive acts— issued false statements about refinery maintenance and manipulated gas trades—to create a public perception of demand and further drive supracompetitive pricing in California. id. at 45-49. Defendants effected this conspiracy through frequent exchanges of isupply-related information concerning refinery maintenance, production, imports, and exports. See id. 9-16. Defendants also cooperated to cover each other’s supply shortages by trading gas on the spot market and utilizing exchange agreements to trade barrels of gas with each other, instead of purchasing elsewhere at market rates. See id. at 9-16, □□□□□□ 37—45, 49-50. Through these arrangements, Defendants consistently collaborated rather than competed with each other in times of need, such as after the Exxon refinery explosion. See id. at 22-24, 49-52. The Court concludes that Defendants would have a rational economic motive to into the type of conspiracy alleged by Plaintiffs. Limiting supply and manipulating public facing information are economically rational ways to increase demand and, thus, prices—even more so, because Defendants themselves would remain protected by a high of collaboration (exchange agreements, coordinated spot-market trading, and information sharing) that enabled them to anticipate market conditions, meet their own 25 15cv1749-JO-AGS and 18cv1374-JO-AGS

supply needs, and avoid purchases at the high prices they created. In sum, Plaintiffs’ theory that Defendants kept supply low and drove prices high, while implementing protections that kept their own costs and supply risks low, makes economic sense. The Court cannot, therefore, conclude that Plaintiffs’ conspiracy theory is economically implausible such that should require Plaintiffs to “come forward with more persuasive evidence to support claim than would otherwise be necessary.” Matsushita, 475 U.S. at 587. While Defendants would certainly have an economic motive to act in the manner lalleged by Plaintiffs, that alone cannot establish an antitrust violation. Antitrust wrongdoing consists of concerted action pursuant to an illegal agreement, not independent, profit maximizing actions based on market conditions. See Citric Acid, 191 F.3d at 1105— 06. Thus, the Court turns to whether the evidence supports the inference that Defendants pursuant to an agreement. Under Matsushita, unless Plaintiffs’ evidence tends to exclude the possibility of independent action, it cannot raise a reasonable inference of conspiracy. 475 U.S. at 588. B. Plaintiffs Have Not Submitted Direct Evidence of a Price-Fixing Conspiracy While both direct and circumstantial evidence can serve to create a triable issue on conspiracy, it appears that Plaintiffs have only submitted circumstantial evidence in their opposition. Direct evidence “is explicit and requires no inferences to establish the proposition or conclusion being asserted,” while circumstantial evidence requires one or more inferential steps to reach a conclusion of conspiracy. Citric Acid, 191 F.3d at 1094 (quoting Jn re Baby Food Antitrust Litig., 166 F.3d 112, 118 (3d Cir.1999); Toscano v. Pro. Golfers Ass’n, 258 F.3d 978, 983 (9th Cir. 2001). For instance, conversations among Defendants explicitly confirming their agreement would constitute direct evidence, while lambiguous conversations among Defendants that require an inference to conclude 26 15cv1749-JO-AGS and 18cv1374-JO-AGS

}agreement would be circumstantial evidence. Compare Markson v. CRST Int'l, Inc., 2022 WL 790960, at *13 (C.D. Cal. Feb. 24, 2022) (“conversations between Defendants confirming their agreements not to hire truckers” was direct evidence), with In re Dynamic Random Access Memory Antitrust Litig., 2007 WL 9752971, at *6—-8 (N.D. Cal. Feb. 20, 2007) (emails referencing “secret meetings” and comments regarding productions cuts required inferences to conclude agreement and thus was not direct evidence). Here, Plaintiffs do not purport to submit evidence of conversations, emails, or other documents that would directly establish that Defendants entered into an agreement with one another.’ Instead, their briefing focuses on circumstantial evidence of parallel conduct and plus factors to establish a genuine dispute as to conspiracy. See Opposition at 8—9; Toscano, 1/258 F.3d at 985. As Plaintiffs have submitted no direct evidence, the Court will turn to whether the circumstantial evidence submitted by Plaintiffs is sufficient to raise an inference of conspiracy. C. Plaintiffs Have Raised a Genuine Dispute of Material Fact as to Conscious Parallelism In considering the circumstantial evidence submitted by Plaintiffs, the Court first examines whether Plaintiffs have raised a triable issue of fact that Defendants engaged in conscious parallelism, before turning to an analysis of “plus factors.” Defendants argue that there is no genuine dispute of material fact as to conscious parallelism because Plaintiffs have failed to submit evidence of parallel pricing or other parallel conduct. See ” The only mention Plaintiffs make of the concept of direct evidence is that “[e]ven if Plaintiffs’ evidence is not considered direct evidence of the conspiracy, it nevertheless tends to exclude the possibility” of independent conduct. Opposition at 9 (citation omitted). 27 15cv1749-JO-AGS and 18cv1374-JO-AGS

719 (“Joint Reply”) at 7-8. As discussed above, while parallelism is not alone sufficient to show conspiracy, it is a relevant factor in addition to plus factors. Parallelism exists when “competitors adopt[] similar policies around the same time response to similar market conditions.” Musical Instruments, 798 F.3d at 1193. Plaintiffs can point to parallel prices to demonstrate conscious parallelism. Brooke Grp., U.S. at 227 (“[C]onscious parallelism, describes the process, not in itself unlawful, by which firms in a concentrated market might in effect share monopoly power, setting their prices at a profit-maximizing, supracompetitive level.”). Plaintiffs can also point to similar business behaviors to demonstrate conscious parallelism. Musical Instruments, 798 □□□□ 1193 (defining parallel conduct as the adoption of “similar policies”); Zoslaw v. MCA Distrib. Corp., 693 F.2d 870, 884 (9th Cir. 1982) (considering whether defendants’ “account classifications, pricing structures and promotional policies” were sufficiently similar to demonstrate parallel conduct). For instance, in Petruzzi’s IGA Supermarkets, Inc. v. Darling-Delaware Co., the Third Circuit found that, although “defendants did not the same prices,” they “acted similarly by refraining from competing” on certain accounts and retaliating against competitors that did compete on accounts. 998 F.2d 1224, 1243 (3d Cir. 1993). Plaintiffs here point to a mix of similar pricing and other parallel business conduct demonstrate conscious parallelism. With respect to pricing, Plaintiffs point to a combination of evidence obtained in discovery and from publicly available documents suggesting that, (1) in 2015 and 2016, Shell based its “at the rack” prices in part on the prices of Exxon, BP, and Chevron; (2) in 2016, Phillips 66 based its Dealer Tankwagon prices in part on a brand average of the prices of Chevron, Shell, Tesoro, and non- Defendant Mobil; and (3) from January 2015 to May 2017, the Dealer Tankwagon prices 28 15ev1749-JO-AGS and 18cv1374-JO-AGS

of Chevron, Shell, and Valero were within approximately twenty cents of one another in Richmond, California. See Dkt. 781-2 (“Surreply”) at 2, 4-5; Pltfs. Ex. 211 at PSXPGI00089880; Pltfs. Ex. 212 at slide 25; Ex. Shell 1 at SOPUS_PGI_00002699. In addition, some of the evidence submitted by Defendants allows for an inference of follow- the-leader pricing, one common form conscious parallelism. See, e.g., Musical Instruments, 798 F.3d at 1195. Five of the eight Defendants admitted that they regularly monitored one another’s prices through services such as OPIS and factored that information into their own pricing. See, e.g., J. Hodgson Declaration 11-15, □□□□□□ P. Brooks Declaration § 96; K. Archambault Declaration ¥ 59; J. Harris Declaration 6; M. O’Neal Declaration 3. In addition to the above evidence of parallel pricing, the evidence in the record also suggests that Defendants engaged in other parallel actions. Defendants had common reactions to market events, including selling gas to Exxon following its 2015 explosion, increasing exports of gas out of California over the class period, and limiting gas inventories at their respective refineries. See Pltfs. Ex. 200; C. Yates Declaration □□ 35— 36; N. Weinberg-Lynn Declaration 15-16; S. Roveda Declaration § 13; R. Sharum Declaration 9] 30-33; C. Dickson Declaration JJ 11—12; M. Perez Declaration □□ □□□□□□ Lockhart Declaration 430; P. Brooks Declaration 38-40, 90; T.A., E. Pestano Declaration J 42. The Court finds that the foregoing evidence sufficiently raises a genuine dispute of material fact that Defendants engaged in conscious parallelism. Both conscious parallelism—where players in a concentrated market independently mirror one another’s prices or take similar profit maximizing actions—and illegal conspiracy to fix prices can result in high prices, but only the latter violates antitrust law. Because conduct as consistent 29 15ev1749-JO-AGS and 18cv1374-JO-AGS

legal conscious parallelism as with illegal conspiracy is insufficient to support a reasonable inference of conspiracy, Matsushita, 475 U.S. at 587-88, the Court turns to Plaintiffs’ ten categories of plus-factor evidence to determine if any evidence “tends to ; exclude the possibility of independent action.” Jd. at 588. Plaintiffs’ Plus Factors After concluding that Plaintiffs have introduced circumstantial evidence of parallel conduct among the Defendants, the Court turns to an examination of the “plus factors” that Plaintiffs contend demonstrate Defendants’ illegal price fixing conspiracy. In all, Plaintiffs argue that ten categories of Defendants’ behavior evidence their wide-ranging, price fixing conspiracy: (1) refineries cooperated with one another through exchange agreements to cover each other’s supply shortages against self-interest; (2) traders frequently exchanged confidential supply-related information against Defendants’ self-interest; (3) refineries restricted production levels and maintained low inventories to reduce overall gas supply; (4) refineries prevented gas imports into California to reduce overall gas supply; (5) refineries exported gas out of California to reduce overall gas supply; (6) refineries protected Exxon after its refinery explosion in 2015 by selling it gas rather than competing for its market share; (7) refineries manipulated the market through a “gentleman’s agreement” to limit trading hours; (8) traders manipulated the market by entering “wash trades” to create a false sense of market demand; (9) traders manipulated the market by selectively reporting trades to OPIS to raise public facing gas prices; and (10) refineries manipulated the market by making false public statements about refinery maintenance to create a fabricated sense of market scarcity. The Court will examine each category of allegedly conspiratorial conduct under the Ninth Circuit’s two-step framework for determining whether circumstantial evidence can 30 15cv1749-JO-AGS and 18cv1374-JO-AGS

icreate a reasonable inference of conspiracy. Citric Acid, 191 F.3d at 1094. For each category, where Defendants meet their Step One burden to offer plausible and justifiable lreasons for their conduct consistent with proper business practice, the Court will consider whether Plaintiffs have pointed to any evidence at Step Two that tends to exclude the possibility of independent conduct. Jd. Finally, the Court will consider the totality of Plaintiffs’ evidence across all ten categories to determine whether they have successfully raised an inference of conspiracy. Id. at 1097, 1106. 1. Cooperation Through Exchange Agreements Plaintiffs contend that Defendants widely used exchange agreements to further their price fixing conspiracy. See Opposition at 37-45. The record shows that Defendants, with the exception of Alon, entered into multiple bilateral exchange agreements with one another over the past twenty years or more to exchange fuel on a barrel-for-barrel basis. Joint Statement § 22; Pltfs. Exs. 114-28.!° Plaintiffs explain that exchange agreements allowed for geography-based exchanges of fuel, where a refinery provides fuel to another one location and receives those barrels back in different location. See Opposition at 38. Exchange agreements also allowed a refiner to provide barrels today to have them returned jat a later date. See id. Plaintiffs argue that Defendants used exchange agreements to ] 10 For instance, Exxon executed exchange agreements with Chevron in 2005, Phillips 66 in 2007, 77 Tesoro in 2012, and Shell in 2013. Pltfs. Exs. 114-15, 118, 128. In addition, Chevron executed exchange agreements with Valero in 2006 and 2014, Tesoro in 2002 and 2015, and BP in 2013 (Pltfs. Exs. 116-17, 119-20, 124); Phillips 66 executed exchange agreements with Shell in 2011 and 2012, Valero in 2012, and BP in 2015 (Pltfs. Exs. 121-23, 125); Tesoro executed exchange agreements with Shell in 1999 and Valero in 2013 (Pltfs. Exs. 126-27). Many of these agreements were amended throughout the years. See Pltfs. Exs. 114-28. 31 15cv1749-JO-AGS and 18cv1374-JO-AGS

transfer large amounts of product to one another with little formality and that this evidences }conspiracy because such behavior is inconsistent with competition. See id. at 39-40. Defendants rebut these allegations by explaining that they independently entered exchange agreements for self-interested business reasons—because they are convenient, cost-saving, and allow for geographical competition. Defendants explain that these agreements enabled them to meet supply needs in specific geographical areas without physically transporting gas to locations where they did not have a refinery. See, e.g., Joint Statement J 22; C. Yates Declaration J 51.!! By enabling companies to receive gas in one location and return the gas in kind in another location, exchange agreements allowed /refineries to expand the geographic reach of their operations at lower costs or in areas where they otherwise would not be able to compete. See id Defendants also state that used exchange agreements to quickly cover shortages and meet supply obligations, without having to buy barrels at market prices. See, e.g., K. Archambault Declaration 63, 66-70; C. Yates Declaration {§ 50—-52.'? They provide testimony that convenience- based exchange agreements helped them prevent supply shortages and disruptions without /incurring the costs of maintaining large inventories. See id. According to the terms of Defendants’ exchange agreements, any imbalances remaining after the parties swapped barrels back and forth were usually settled monthly in cash. See, e.g., Pltfs. Exs. 114-28. By explaining that exchange agreements allowed them to save time and money transporting fuel, expand geographical reach, and cover supply shortages at lower cost and lower risk,

1! See also J. Harris Declaration §§ 21-22; R. Sharum Declaration § 20; P. Brooks Declaration See also R. Sharum Declaration § 23; J. Harris Declaration §{ 21—24; P. Brooks Declaration 82-83, 86-88; T.A. 338-342, E. Pestano Dep. at 32:13-15, 38:17-19, 39:7—-14, 46:3-11. 32 15¢v1749-JO-AGS and 18cv1374-JO-AGS

Defendants have offered plausible, justifiable, and self-interested business reasons to enter exchange agreements. Plaintiffs argue that these agreements are inherently against self-interest because a refiner has no incentive to loan to competitors in times of need instead of leveraging that need to gouge them on prices.'? In support of this argument that exchange agreements were against self-interest, Plaintiffs highlight two instances of swaps performed at inequivalent values (e.g., Phillips 66 loaned barrels when the spot market price was $1.79, Shell returned the barrels a few days later when the spot price was $1.72). See Pltfs. Exs. 147-50. While this evidence could support the conclusion that these two transactions in isolation were not profit maximizing for one of the parties, see Pltfs. Exs. 147—50,"4 the value of any single, in-month transaction is not sufficient to raise an inference that exchange agreements, which enable dozens of transactions between two parties over a year or more, are not profitable for Defendants overall. See, e.g., Pltfs. Ex. 145. Plaintiffs’ argument would hold true only if exchange agreements went one way and benefited only party. The central premise of these exchange agreements, however, is that they ———— 3 Plaintiffs point to an instance of a Shell employee explaining that “[t]he whole point of the convenience exchange is to help each other out volumetrically without adverse financial impact.” □□□□□□ Ex. 139. This statement, made in the context of a price negotiation to settle remaining balances, captures the mutually beneficial essence of exchange agreements. See id. 4 In Exhibit 147, Shell requested product from Phillips 66 under an exchange agreement and Phillips 66 agreed to supply up to a certain amount and be paid back five days later. Plaintiffs highlight 77 that Shell received product on a day when gas prices were $1.79/gallon and ultimately returned product to Phillips 66 five days later when gas prices were $1.72/gallon. See Pltfs. Exs. 148-49. Plaintiffs ignore the fact that Defendants entered into an agreement not knowing what the price would be five days later. Exhibit 150 shows that Shell used an exchange agreement with Valero to cover a shortage and noted that the value of the swap was “29k.” See Pltfs. Ex. 150. 33 15¢v1749-JO-AGS and 18cv1374-JO-AGS

facilitate reciprocal exchanges over extended periods of time: the party giving one day will be the receiving party the next day, when its time of need arises. The Court therefore finds that Defendants have offered sufficiently plausible, legitimate, and self-interested business reasons to enter into mutually beneficial, long-term exchange agreements to reduce their and costs. Accordingly, Defendants meet their Step One burden. While these types of agreements can be misused as part of an unlawful conspiracy, without more, this Court cannot conclude that exchange agreements, in and of themselves, raise a reasonable inference of conspiracy. This Court is mindful that other courts have emphasized the procompetitive features of exchange agreements. In Aguilar v. Atl. Richfield Co., a consumer brought a class action alleging that gas companies had conspired restrict the output of CARBOB and pointed to exchange agreements as a plus factor. 25 4th 826 (2001).!° In concluding that these agreements did “not even imply collusive, rather than independent, action,” the court stated that: “exchange agreements have long recognized as procompetitive in purpose and effect, enabling or facilitating companies to compete in product and/or geographical and/or temporal markets in which otherwise could not or would not compete as efficiently or at all.” Jd. at 834; see also Blue Bell Co. v. Frontier Ref. Co., 213 F.2d 354, 359 (10th Cir. 1954); Thomas v. Amerada Hess Corp., 393 F. Supp. 58, 72 (M.D. Pa. 1975); Indep. Iron Works, Inc. v. U.S. Steel Corp., 322 F.2d 656, 666-67 (9th Cir. 1963) (business transactions between competitors, without more, do not allow for an inference of conspiracy). Absent evidence of misuse, courts have uniformly rejected the argument that exchange agreements in the gas industry 234 ©) 'S The Court in Aguilar addressed the plaintiff's § 1 Cartwright Act claim, which it noted “is analogous to section | of the Sherman Act.” 25 oe at 838.

15ev1749-JO-AGS and 18cv1374-JO-AGS

are per se vehicles for coconspirators to exchange favors. See, e.g., Aguilar, 25 Cal. 4th at 863-64. The Court therefore turns to whether Plaintiffs have provided evidence of such misuse. The additional evidence that Plaintiffs offer regarding exchange agreements at Step does not tend to exclude the possibility of independent action. First, Plaintiffs offer no evidence to substantiate their claim that “enormous” quantities of gas changed hands pursuant to these agreements. See Opposition at 39. The only evidence Plaintiffs submit in support of this statement is a Phillips 66 document with three numbers on it but no indication of what the numbers signify. See Pltfs. Ex. 130. Moreover, Phillips 66 and Valero provide additional context explaining that their exchange volumes were less than }3% and 1% of their annual volumes, respectively. See R. Sharum Declaration 26; P. Brooks Declaration 82. Second, Plaintiffs contend that the friendly and informal nature of exchange /transactions, as seen in the following, raise questions about their non-competitive and /collusive nature: Defendants (1) often amended agreements to accommodate different or additional product; (2) on eight occasions, settled exchange balances years after the original exchanges; and (3) referred to performance under the agreements as “help” and “favors” j/or “borrowing” gas, which was “appreciated.” See Pltfs. Exs. 157-60 (referring to performance as “help” and “favors”; see also Pltfs. Exs. 161—64 (referring to performance as “borrowing” gas).!° The Court agrees this evidence underscores the cooperative and 22) ©

'6 See also Ex. 133 (2011: Valero and Exxon informally add sub-octane to agreement); Ex. 134 (2014: Chevron and Tesoro amend to extend time frame); Ex. 144 (2012: BP and Phillips 66 “liquidate balances from old contracts” from prior five years); Ex. 145 (2012: BP and Valero settle exchange balances); Ex. 141 (2013: Shell and Exxon “liquidated all aged balances... both active and inactive” 35 15cv1749-JO-AGS and 18cv1374-JO-AGS

interdependent nature of exchange agreements but does not find that it adds dimension that supports an inference of conspiracy. Given that the purpose of exchange agreements is to mutually benefit the parties by meeting each other’s supply needs as they arise, allowing flexible contract amendments between participants could be consistent with furthering that }common goal. Furthermore, the language of “help” and “favors” could simply be consistent with collegial language used in the context of the cooperative partner-supplier relationships created by these agreements. Plaintiffs also point to evidence of eight instances where Defendants did not settle imbalances in the thousands of barrels until years after the swaps were made as evidence /of collusion. In light of the evidence that Defendants commonly produced hundreds of millions to billions of barrels per year, these are de minimis amounts. See, e.g.,C. Dickson || Declaration {{ 6, 21 (Exxon refinery capacity was over | billion gallons per year); R. Sharum Declaration { 28 (Phillips 66 refinery typically produced over 3 million gallons per day); see also Plitfs. Exs. 136-37, 139-41, 143-46. As these actions could be consistent with Defendants’ self-interest in flexible administration of mutually beneficial exchange agreements or delayed bookkeeping of de minimis amounts, they do not tend to exclude the possibility of independent action. | from 2012 totaling approximately 18,000 gallons); Ex. 146 (2014: Exxon tells Tesoro it cannot deliver 79 product to settle exchange balance and so pays); Exs. 136-37 (2014: Phillips 66 and Chevron settle “inactive outstanding balances”); Ex. 142 (2015: Chevron and Valero amend to transfer “old 93 product .. . to the active product”); Ex. 143 (2015: Phillips 66 and Exxon “clean up inactive balances” of approximately 25,000 barrels from prior three years); Ex. 135 (2015: Valero and Shell negotiate amendment); Exs. 139-40 (2017: Chevron and Shell settle balances totaling $154,949.51 from prior three years). 36 15ev1749-JO-AGS and 18cv1374-JO-AGS

Because the behavior that Plaintiff points to is “as consistent with permissible competition as with illegal conspiracy,” it does not support an inference of antitrust conspiracy. Matsushita, 475 U.S. at 588. Based on the above, the Court cannot conclude that Plaintiffs’ evidence concerning exchange agreements raises a reasonable inference of conspiracy. 2. Exchanges of Information Among Traders Plaintiffs point to interfirm exchanges of information among Defendants’ traders as the main driver of Defendants’ alleged conspiracy. See Opposition at 9-16. California refineries employ traders to buy and sell gas products on the spot market to meet refinery isupply needs. Joint Statement 921; McCullough Report §22 (describing trader transactions as “like a farmers’ market where farmers buy and sell potatoes to each other”). These traders buy and sell with traders at other refineries and third-party trading firms on behalf of the refineries that employ them. Joint Statement J 21; Lipow Report □ 20, 31; McCullough Report 722. For example, traders will buy or sell in response to their refineries’ supply conditions and either buy to ensure the refinery has enough product to its contractual obligations (such as in the case of an outage), or sell in the market to }ensure that any excess product does not slow down refinery production. See, e.g., Lipow Report Jf 20, 22, 26; K. Archambault Declaration 28-30; C. Yates Declaration □ 54. Neither party disputes that the California trading market is characterized by a handful of }companies, and thus, traders are repeat players that are familiar with one another and buy and sell from each other on a regular basis. See Lipow Report 7 31; McCullough Report 37 15ev1749-JO-AGS and 18cv1374-JO-AGS

99 20, 66.'’ Traders, therefore, regularly communicate with each other at trade association events, as well as by phone, email, and messaging platform in the course of performing their duties. See McCullough Report ¥ 50. Although the parties agree that traders must communicate to execute transactions, Plaintiffs contend that Defendants’ traders consistently and inappropriately shared confidential information regarding refinery maintenance, production, and supply in furtherance of a conspiracy. See Opposition at 9-16. For instance, Plaintiffs take issue with the fact that traders exchanged information like the following: In one conversation on April 17, 2012, BP’s trader asked if Shell’s trader had “27kb of pnw premium,” explaining he needed the product because BP’s refinery was running a unit “at min rates until may in order to build ls vgo inventory back to safety stock level.” Pltfs. Ex. 43. According to Plaintiffs, rather than just simply asking for the amount of gas needed, this trader exchanged gratuitous and confidential supply information about BP’s refinery against self-interest. Dkt. 809 at 42-44. Plaintiffs point to various trader exchanges like the above that occurred at trade association events and during trading communications as evidence of conspiracy. See, e.g., Pltfs. Exs. 6, 8-12, 19-20, 35-44, 46-50, 52, 192-93. In order to meet their Step One burden of producing plausible and justifiable reasons for their actions, Defendants offer evidence that traders exchanged information, while buying and selling gas products to each other, as customers and suppliers. Defendants explain that although the refineries’ overall business model was premised on competition, traders had very different roles: they acted as customers and suppliers that needed to © | 17 See also C. Yates Declaration § 56; R. Pluimer Declaration § 6; M. Perez Declaration 10; L. Lockhart Declaration 4, 36-37 38 15cv1749-JO-AGS and 18cv1374-JO-AGS

cooperate because they conducted repeat business with one another. See, e.g., R. Pluimer Declaration § 6; M. Perez Declaration { 10.'® Defendants assert that, in this context, traders shared information for strategic advantage—to build rapport and actively foster additional trading opportunities. See, e.g., H. Henderlite Declaration § 16; R. Pluimer Declaration 1/99 5-7.'° Defendants also note that traders were communicating already public information about refinery maintenance and import and export schedules to explain their needs for buying or selling, rather than divulging confidential information. See, e.g., P. Brooks Declaration {§ 32, 36, 75; J. Yomtoob Declaration {J 15, 21 & n.4. Defendants ; also submit that traders attended trade association events to stay informed on the industry explore potential business opportunities rather than for conspiratorial purposes. See K. Archambault Declaration J] 72, 74; J. Yomtoob Declaration 430.2? For example, traders used these opportunities to discuss market conditions and develop business relationships with other traders. See id. Based on the above, the Court is satisfied that Defendants produced plausible and justifiable business reasons for traders to share information with one another. Defendants provide evidence that traders needed to buy and sell gas from each other to meet their refineries time-sensitive supply needs, and that they did so in a concentrated market with few players, meaning that the same traders conducted repeat transactions with each other. In this context, a reasonable juror could infer that traders shared information about their 18 See also P. Brooks Declaration 72; T.A., E. Pestano Declaration J 48, 50—51. '9 See also J. Yomtoob Declaration § 13-15, 21, 24; C. Yates Declaration 956; R. Sharum Declaration J 19; L. Lockhart Declaration {J 4, 29, 36-37, 39-45; J.A. 587-93, Phillips Exs. □□□□□□ 2° See also H. Henderlite Declaration J 19; C. Yates Declaration J 57; R. Pluimer Declaration □ 13; L. Lockhart Declaration §] 49-50; P. Brooks Declaration §{ 103, 105; C. Dickson Declaration § 71; M. 45 Perez Declaration § 17. 39 15cv1749-JO-AGS and 18cv1374-JO-AGS

respective refineries for self-interested reasons—in order to facilitate transactions, cultivate /relationships, and glean information for future trading opportunities. Armed with knowledge of another trader’s supply needs, that trader is arguably better positioned to sell /to the other. In the same vein, disclosing one’s forward looking supply needs arguably better positions a trader to identify potential sources of product coverage. The Court also agrees that incidental sharing of already public information to explain the need to buy or a without more, is consistent with legitimate business conduct. Overall, Defendants’ evidence of the context and purpose of traders’ exchanges is consistent with what the Citric Acid court described as the legitimate practice of “[g]athering information about pricing and competition in the industry.” 191 F.3d at 1098 (gathering information about other }producers did not allow for inference of conspiracy).”! Therefore, Defendants’ explanations meet their burden at Step One. At Step Two, Plaintiffs submit dozens of communications among traders to attempt to demonstrate that Defendants’ traders regularly exchanged confidential information, and that such information was “widely dispersed” and reached the “highest echelons of management.” See Opposition at 10-16. The Court has reviewed the communications submitted by Plaintiffs and finds that they generally fall into three categories: (1) traders two different companies exchanged refinery information in the context of a transaction, (2) traders within one company shared information they obtained from an ——_—_—_——__ 21 See also Musical Instruments, 798 F.3d at 1196 (“participation in trade-organization meetings 93 where information is exchanged and strategies are advocated does not suggest an illegal agreement”); see also Pitfs Ex. 23 ((30(b)(6) witness for trade association testifying that traders discussing pricing at informal meetings during a trade association event would not be problematic “if it’s a supplier and a customer kind of a conversation’). 40 15ev1749-JO-AGS and 18cv1374-JO-AGS

outside trader about another refinery, and (3) only one exchange where traders ultimately passed information up the chain-of-command as Plaintiffs allege.** The Court summarizes these categories of evidence below before considering whether Plaintiffs meet their Step Two burden in light of these communications as a whole. In the first category, Plaintiffs highlight communications between traders discussing refinery maintenance, imports, and production that explicitly occurred in the context of executing or seeking to execute a transaction. For instance, on February 23, 2015, a Shell trader asked whether Chevron could supply it in light of Shell’s “ongoing structural demand.” See Pltfs. Ex. 41. Similarly, on December 2, 2015, BP’s trader told Shell’s trader that he had “a fairly steady stream of barges coming in thru the month to both gatx shell carson,” apparently in an effort to see if Shell had an interest in purchasing. Pltfs 11 at row 194801; see also Plitfs. Exs. 6, 11, 19-20, 40, 43-44, 52. In all of the foregoing, a trader from one company received information from a trader at another company regarding refinery maintenance, trading, or production in the context of securing potential deal with that trader. In the second category, Plaintiffs submit evidence of information exchanges between traders at different refineries that do not immediately appear to have occurred in the context of potential business transactions. The majority of these documents show traders internally sharing information that they obtained from a trader at a different refinery. While it is / unclear whether the information was originally obtained in the context of a sale, it appears traders circulated this outside information internally to other traders, presumably to

22 The Court notes that Plaintiffs also submit communications in which traders speculate regarding market conditions or third parties as discussed in detail at infra Section V.D.3. 41 15cv1749-JO-AGS and 18cv1374-JO-AGS

facilitate their own work of buying and selling (i.e., identifying sources of supply or potential buyers). For example, in one internal conversation between Tesoro traders on March 17, 2017, a trader noted that he was “running out of options on who to buy from,” the traders discussed that they could not buy from Valero because it was still undergoing maintenance and “traders indicat[ed] [Valero was] still struggling.” See Pltfs. Ex. 39; see also Pltfs. Exs. 10, 47, 192—93 (internally discussing the severity of other refineries’ maintenance events). The remaining conversations in the second category that Plaintiffs point to involve two traders from BP and Phillips 66; unlike the other trader communications submitted by Plaintiffs, these traders appear to also have exchanged potentially confidential export, import, and supply information, in addition to refinery maintenance information. For /instance, on January 23, 2018, BP’s trader wrote an internal market update, noting that “P66 also has a reformer TAR in Feb/March (pnc) that is unknown to the market.” Pltfs. Ex. 48 at BPWC-PG-00039851; see also Plitfs. Exs. 8, 49-50 (discussing potentially confidential information regarding maintenance, exports, and imports). The direct exchanges between BP and Phillips 66 may suggest something more nefarious, given that the information shared was not clearly public. Where that information is then circulated internally, however, it appears to be circulated solely to other traders, presumably for the purpose of aiding buying and selling. See, e.g., Pltfs. Ex. 48. In the third and final category, Plaintiffs point to a single communication that demonstrates that the information exchanged between traders was passed up the chain-of- command to individuals responsible for production and pricing at Defendants’ refineries. Opposition at 16. In an email dated March 7, 2015, Tesoro’s trader responded to an larticle that was sent to a large listserv regarding Exxon’s maintenance following an 42 15ev1749-JO-AGS and 18cv1374-JO-AGS

explosion at Exxon’s refinery in February 2015. See Pltfs. Ex. 35. Tesoro’s trader wrote: “[g]lood news—this report is erroneous.” Jd. The email does not clearly indicate the source of his knowledge.”’ Plaintiffs also submit evidence showing that the recipients of this email included high-level managers at Tesoro. See Pltfs. Exs. 36-38. Plaintiffs argue that, based this exchange, a factfinder could reasonably infer that other information exchanges among traders were similarly passed up the chain-of-command. Plaintiffs argue that these communications between traders prove that Defendants shared confidential information about their refinery operations and import/export decisions against self-interest and in furtherance of a price fixing conspiracy. Turning to the inferences that this evidence permits, the Court finds the information exchanges in the }record would not allow a reasonable juror to infer the wide ranging, eight-Defendant conspiracy Plaintiffs seek to prove. The case law is replete with discussions on whether exchanges of information are a plus factor tending to support conspiracy, and although no bright-line rules or tests emerges, the Court finds the following factors are relevant to the facts before it: (1) who is communicating and how frequently, see, e.g., Baby Food, 166 F.3d at 121-22; (2) the proximity of the communications to pricing decisions, see, e.g., In re Flat Glass Antitrust Litig., 385 F.3d 350, 368-69 (3d Cir. 2004); and (3) the context and content of the communications, see e.g., Krehl v. Baskin-Robbins Ice Cream Co., 664 F.2d 1348, 1357 (9th Cir. 1982); In re Dairy Farmers of Am., Inc., Cheese Antitrust Litig., 60 F. Supp. 3d 914, 951 (N.D. Ill. 2014), aff'd 801 F.3d 758 (7th Cir. 2015). *3 In their opposition, Plaintiffs claim that Tesoro also wrote that the information was “confirmed via several sources including the USWC XOM crude trader,” but that language is not reflected in Exhibit 35 and thus, is misquoted. 43 15cv1749-JO-AGS and 18cv1374-JO-AGS

The Court first addresses who is exchanging information and how often those }exchanges occurred. Exchanges of information among high-level executives who have pricing authority bolster an inference of conspiracy, although “there must [also] be evidence that the exchanges of information [actually] had an impact on pricing.” Stanislaus Food Prod. Co. v. USS-POSCO Indus., 2013 WL 595122, at *12 (E.D. Cal. Feb. 2013), aff'd, 803 F.3d 1084 (9th Cir. 2015) (quoting Baby Food, 166 F.3d at 125); In \|\re Publ’n Paper Antitrust Litig., 690 F.3d 51, 65 (2d Cir. 2012) (holding circumstantial evidence supported conspiracy where presidents of two companies with pricing authority shared intentions to increase prices before those decisions had been publicly announced). On the other hand, exchanges of information among low-level employees without pricing /authority are generally not probative of conspiracy. Baby Food, 166 F.3d at 121-22. Nevertheless, exchanges of information between low-level employees can be probative of conspiracy where there is evidence that those employees routinely reported information up the chain-of-command. Jn re Static Random Access Memory (SRAM) Antitrust Litig., 2010 WL 5138859, at *6 (N.D. Cal. Dec. 10, 2010). Distinguishing Baby Food, the court in SRAM found that where exchanges between low-level employees were regularly reported individuals with pricing authority, such conduct was more consistent with conspiracy with unilateral conduct. Jd. at *7. With these factors in mind, the Court turns to the trader conversations in the record. The volume of evidence that Plaintiffs submit suggests relatively frequent communications between Defendants’ traders. See, e.g., Pitfs. Exs. 8, 10-11, 19-20, 35, 39, 41-44, 46-47, 44 15cv1749-JO-AGS and 18cv1374-JO-AGS

64-65, 82, 169.24 However, frequency is not particularly probative of conspiracy in the context of gas traders in California. Because these traders must communicate with one another to do their jobs and because there are a limited number of traders in the California market, one would expect to see frequent communications among this small group. Moreover, the bulk of the communications took place between traders, low-level employees that did not have the authority to set rack prices or production levels. See In re Citric Acid Litig., 191 F.3d at 1094. With regard to six of the Defendants—Chevron, Shell, Tesoro, Valero, Exxon, and Alon—Plaintiffs can only point to one instance*> where information gained through trading was sent to higher-level employees and this evidence is ambiguous at best. See Pltfs. Ex. 35. In this instance, a Tesoro trader “replied all” to an article originally circulated to a large listerv, stating, “this report is erroneous.” See id. Plaintiffs submit evidence to show that the recipients of this “reply all” included higher-up decisionmakers, although such decisionmakers appear to be included in the “reply all” because they were recipients of the original email circulating the article. See id.*° Plaintiffs’ evidence regarding communications between BP and Phillips 66 present different picture. The traders of these two companies appear to have repeatedly 24 Some of the communications Plaintiffs submit are with unidentified third parties that may be non-Defendants, which suggest little about an agreement between Defendants in this case. See, e.g., Pltfs. Ex. 9 (Valero trader and “dan_rnr”); Pitfs. Ex. 65 (Shell trader and “toddktk@YAHOO”). 25 Although the Court is not obligated to mine the record for evidence Plaintiffs do not point to in ! their briefing, the Court has identified an additional instance where information may have been passed up the chain-of-command. See Pltfs. Ex. 65 (Valero trader to Shell trader: “go tell your supply group that gas in the bay is getting tight”; no further context or reply). This single message without a reply does not indicate whether information actually was reported up the chain-of-command. See id. 26 The Court also notes that Plaintiffs’ Exhibit 35 is the only piece of evidence that Plaintiffs have supported by submitting additional evidence showing that the recipients of the email held particular job roles. See Pltfs. Exs. 36-38. 45 15ev1749-JO-AGS and 18cv1374-JO-AGS

}communicated potentially confidential information to each other and then passed that information to production decision-makers. See, e.g., Pltfs. Ex. 8 (discussing imports, future projections, and production decreases); Pltfs. Ex. 48. For example, on December 18, 2017, BP’s trader asked, “you guys seein run cut incentive yet in LA,” and Phillips 66’s trader responded, “yes indeed. Bay and LA.” Pltfs. Ex. 8 at rows 5832-33. On the same day, BP’s trader wrote an internal marketing update to other BP traders, stating in part that does confirm to us that they are trimming runs.” Pltfs. Ex. 48 at BPWC-PG- 00039857. Plaintiffs also emphasize that BP’s trader testified at his deposition that he considered “run cuts” to be private information at BP. Pltfs. Ex. 74 at 228:9-229:19. These documents reflect that Phillips 66’s trader disclosed the company’s internal, and perhaps confidential, decision to decrease production levels (referred to as cutting or trimming “‘runs”) with BP’s trader. This evidence of confidential exchanges is coupled with declarations from BP’s and Phillips 66’s traders admitting that, at times, they shared such /information with their respective production departments for the purpose of taking advantage of predicted market shortages.”’ Based on the above, the evidence does not suggest the eight-Defendant conspiracy Plaintiffs urge. Given that frequency alone is not particularly probative of conspiracy this context, as described above, the Court focuses on whether Plaintiffs have pointed to evidence to support their argument that traders “widely dispersed” information gleaned during trading to the “highest echelons of management.” Opposition at 16. Plaintiffs have

27 See J. Yomtoob Declaration 25 (information “may sometimes be transmitted to BP’s supply and refinery planning teams to determine whether there is an opportunity to supply counterparties”); L. Lockhart Declaration § 47 (‘on occasion, I discussed with our California refineries whether they could increase production of CARBOB” based on trader (emphasis omitted).

15ev1749-JO-AGS and 18cv1374-JO-AGS

pointed to no evidence that Chevron, Shell, Valero, or Alon passed trading information up chain-of-command. Plaintiffs have pointed to one ambiguous instance involving Tesoro and Exxon where this may have occurred. See Pltfs. Ex. 35. No reasonable juror icould conclude based on this scant and ambiguous evidence that traders for each of the leight Defendants regularly passed information up the chain-of-command. Even considering BP’s and Phillips 66’s admissions that they occasionally shared trading information with their refineries’ supply teams (evidence that Plaintiffs do not point to but the Court has read and considered), this evidence does not implicate any of the six remaining Defendants such that a reasonable juror could infer an eight-Defendant conspiracy. Next, the Court considers the proximity of the information exchanges to pricing decisions or other acts in furtherance of the alleged conspiracy. Courts have consistently found that information exchanges are sufficient circumstantial evidence of conspiracy where the exchanges are closely followed in time by price increases. See, e.g., Flat Glass, 385 F.3d at 368-69 (holding that evidence of parallel price increases shortly following information exchanges took the exchanges “outside the realm of ‘mere possession’” and allowed for an inference of conspiracy); Jn re Blood Reagents Antitrust Litig., 266 F. Supp. 3d 750, 778 (E.D. Pa. 2017) (holding that “the close temporal link between the transfer [of price information] and announcement of the 2001 price increases raise an inference of conspiracy”); In re Platinum & Palladium Antitrust Litig., 2017 WL 1169626, at *13 }(S.D.N.Y. Mar. 28, 2017) (information exchanges shortly followed by price drops 47 15ev1749-JO-AGS and 18cv1374-JO-AGS

supported inference of conspiracy).”® Here, Plaintiffs do not point to any evidence of specific price increases, supply reductions, or other concerted activity in proximity to Defendants’ exchanges of information. Plaintiffs, therefore, have not created an inference conspiracy by showing the proximity of information exchanges to specific collusive actions in furtherance of the conspiracy. Lastly, the Court examines the context in which the communications occurred, and the nature of the information exchanged. Context is particularly important given that “the mere exchange of price information, without more, is not per se illegal,” Krehl, 664 F.2d 1357, and information exchanges between competitors “can in certain circumstances increase economic efficiency and render markets more, rather than less, competitive.” United States v. U.S. Gypsum Co., 438 U.S. 422, 441 n.16 (1978). Indeed, where defendants are competitors in a market but also customers and suppliers, the range of permissible inferences to be drawn from information exchanges must be determined in part by the context in which the exchanges occur. See, e.g., Dairy Farmers, 60 F. Supp. 3d at (“having repeated communications with a supplier, who also is a competitor in certain markets, does not a conspirator make”); In re Polyurethane Foam Antitrust Litig., 152 F. Supp. 3d 968, 984-85 (N.D. Ohio 2015) (finding certain exchanges that occurred in the context of a vendor-customer relationship did not support an inference of conspiracy). 28 See also, e.g., Inre Domestic Drywall Antitrust Litig., 163 F.Supp.3d 175, 197 (E.D. Pa. 2016) (“Opportunities to conspire may be probative of a conspiracy when meetings of Defendants are closely followed in time by suspicious actions or records.”); Richard A. Posner, Jnformation and Antitrust: Reflections on the Gypsum and Engineers Decisions, 67 Geo. L. J. 1187, 1199 (1979) (“[I]f the effect of information exchange were to raise the level [of] prices, one could infer that the motive was price fixing.”). 48 15ev1749-JO-AGS and 18cv1374-JO-AGS

l Here, Defendants are competitors in the wholesale and retail gasoline markets in California, but their traders operate as customers and suppliers on the spot market where buy and sell gasoline. See, e.g., M. Perez Declaration § 10; L. Lockhart Declaration 4, 36-37. The information exchanges in the first category of Plaintiffs’ evidence discussed above confirms this market reality. See Pltfs. Exs. 43-45, 78. Although many of the trader chats submitted by Plaintiffs appear excerpted, leaving the reader without the context of the conversation, many of the conversations explicitly occur in the context of a current or potential transaction. See, e.g., Pltfs. Ex. 43 (Shell and BP traders discussed product availability in context of potential transaction); Pltfs Ex. 44 (Tesoro internally discussed information about fire at BP and noted that “[BP] might be able to deliver us bbls”); Pltfs. Ex. 45 (Chevron discussed inventory when explaining an error in its delivery to Phillips 66); Pltfs. Ex. 78 (Shell trader told other traders that he was talking to Phillips 66 about one of their imports and the possibility of selling them a blending component). Regarding the nature of the communications, the Court considers whether the {information shared is the kind reasonably shared in the context of legitimate business Jinteractions. Citric Acid, 191 F.3d at 1094; Musical Instruments, 798 F.3d at 1194. Here, most of the information relates to the timing and nature of refinery maintenance, although there are also some discussions regarding imports, exports, and the state of the market. See Pltfs. Exs. 6-12, 35, 40-50. These are the types of information that are relevant to a trader’s need to buy or sell on the spot market. In fact, many of the communications submitted by Plaintiffs demonstrate that traders used the information they gained from traders at other refineries for just that purpose—to help them locate buying and selling opportunities. See, Pltfs. Ex. 19 (BP noted Exxon’s supply issues in context of potentially being able to sell to Exxon); Pltfs. Ex 20 (Chevron noted Tesoro’s storage issues in context of potential 49 . 15cv1749-JO-AGS and 18cv1374-JO-AGS

with Tesoro); Pltfs. Ex. 40 (Chevron internally discussed Exxon’s delayed maintenance in context of timing its delivery to Exxon). With respect to the six Defendants aside from BP and Phillips 66, the context and content of the trader communications in the record renders the exchanges as consistent with conspiracy (or less so) as with customers and suppliers exchanging useful information in the context of present and potential business transactions. See Pltfs. Exs. 10-11, 19-20, 39, 41-44, 46-47, 65, 169. Plaintiffs emphasize that an inference of conspiracy is created by the fact that traders shared refinery information not strictly necessary to transact purchases. See Dkt. 809, 43:444:4. In this buyer-supplier context, however, the evidence supports the inference that traders had legitimate business reasons to share refinery-related information while trading. For example, a trader might share information that its refinery will be down for maintenance in order to purchase product or gain information about who might be able to sell. Traders also have a self-interested reason to share their own information so that they will receive valuable information in return—information that will them locate either supply or parties seeking to buy. While such extensive exchanges of refinery maintenance and other supply information could have been exchanged for conspiratorial purposes, the evidence that Plaintiffs point to does not tend to rule out the possibility that Defendants shared this information for the self-interested motives set forth above. As a whole, therefore, the exchanges of information in the record do not permit an inference of conspiracy with respect to most Defendants. Defendants’ traders are low-level employees that regularly exchanged information in the course of their job duties, but there not sufficient evidence that traders reported information up the chain-of command. See Plitfs. Exs. 35, 65; Opposition at 16. Moreover, these exchanges are not linked to other 50 15cv1749-JO-AGS and 18cv1374-JO-AGS

questionable events, occurred in the context of plausibly legitimate customer-supplier relationships, and did not necessarily consist of confidential information. In light of these factors and based on a thorough analysis of the evidence in the record, the Court finds that Plaintiffs’ evidence of information exchanges is not sufficient to permit a finding of conspiracy with respect to Chevron, Shell, Valero, Exxon, Tesoro, and Alon. The Court also concludes that the above evidence of information exchanges is sufficient to permit an inference of conspiracy with respect to BP and Phillips 66. The evidence concerning BP’s and Phillips 66’s exchanges reflects that their traders both shared potentially confidential information and passed information up the chain-of-command to people who made production decisions. See, e.g., Pltfs. Exs. 8, 64, 75, 82-84; J. Yomtoob Declaration J 25; L. Lockhart Declaration J 47. BP’s and Phillips 66’s traders testified that shared information up the chain-of-command for the self-interested reason of enabling refineries to better compete in the market. See id. From this evidence, however, a /reasonable juror could infer that BP’s and Phillips 66’s traders shared information based on the understanding that each would use the information shared outside of trading. See, Pltfs. Exs. 8, 46, 48. Accordingly, the Court finds that the nature of the communications between BP’s and Phillips 66’s traders, coupled with their admissions that reported information up the chain-of-command, is more consistent with conspiracy than with independent action—or, at least, that a reasonable juror could conclude so. 3. Supply Restrictions: Running Refineries Below Capacity Plaintiffs argue that Defendants colluded to keep gas supply low in California to supracompetitive prices and create conditions that enabled price spikes. Plaintiffs /contend that Defendants restricted supply against self-interest in three main ways: by 51 15cv1749-JO-AGS and 18cv1374-JO-AGS

limiting production of CARBOB, discouraging and rerouting imports, and exporting gas when it would have been more profitable to sell in-state. See Opposition at 26-37. In support of this plus factor, Plaintiffs point to Defendants’ increased CARBOB production after the 2015 Exxon explosion as proof that Defendants could have produced more CARBOB prior to that point.”? See Opposition at 26; Pltfs. Exs. 66-67. They do not, however, offer any proof regarding optimal CARBOB production levels nor whether any Defendant’s decision to produce less than that amount was not profit maximizing, and therefore, against self-interest. Instead, Plaintiffs point to trader communications which, they claim, demonstrate that Defendants shared information for the purpose of cooperatively reducing supply. See Opposition at 26-34. 1] To meet their Step One burden, Defendants explain that each refiner independently production levels to maximize its profit. Defendants explain that their refineries produce primarily CARBOB but also a variety of other products, such as jet fuel, diesel, non-CARBOB gasoline. See, e.g., J. Harris Declaration 4 7.°° Through detailed planning processes, each Defendant decided how much of its refinery capacity to devote to producing CARBOB versus other products to maximize profit levels. See, e.g., K. Archambault Declaration J§ 8, 10, 15-21, 27; N. Weinberg-Lynn Declaration □□□□□□□□ These decisions factored in considerations such as the relative demand levels for CARBOB compared to their other products and the relative costs of producing each type of product. 79 29 Plaintiffs also point to an email by a Chevron employee stating that he expected production levels at Exxon’s refinery to increase after the refiner was sold to a third party. See Pltfs. Ex. 69. 30 See also K. Archambault Declaration J 15; N. Weinberg-Lynn Declaration {J 7, 10; M. Bodziak Declaration J 7; R. Sharum Declaration 4 5; P. Brooks Declaration {| 13-14; C. Dickson Declaration § 16. 3! See also M. Bodziak Declaration {J 8-10; R. Sharum Declaration 4 5; J. Harris Declaration 7-13; P. Brooks Declaration §{{ 13-23; C. oe Declaration Jf 14-21.

15¢v1749-JO-AGS and 18cv1374-JO-AGS

See, e.g., K. Archambault Declaration Jf 8, 10, 15—21, 27; N. Weinberg-Lynn Declaration 49 10-14.°2 When prices for CARBOB rose, such as following the Exxon explosion, Defendants produced more CARBOB relative to other products because it was profitable to do so. See K. Archambault Declaration § 38; H. Henderlite Declaration 29-32.*° Defendants further explain that their inventory decisions needed to factor in the financial costs and logistical constraints of maintaining inventories. From a financial standpoint, Defendants explain that they had little incentive to hold inventory in storage because it was costly and gas inventories had limited life spans and declining values. See, J. Harris Declaration 17-19; C. Yates Declaration 435; N. Weinberg-Lynn || Declaration {| 15-16. From a logistical standpoint, Defendants explain that refineries stored a variety of raw materials and finished products, each of which required its own storage tank. The fact that refineries only had a finite number of these storage tanks necessarily limited how many tanks could be dedicated to CARBOB. See, e.g., N. Weinberg-Lynn Declaration {ff 15-16; P. Brooks Declaration {§ 38-40; C. Dickson }Declaration J§ 11-12. In light of these financial and logistical constraints, certain Defendants state that they preferred to cover unexpected shortages through exchange agreements or purchases on the spot market, rather than building up inventories. See J. {| ——_——_—__——_ 32 See also M. Bodziak Declaration J 8-10; R. Sharum Declaration 4/5; J. Harris Declaration 7-13; P. Brooks Declaration § 13-23; C. Dickson Declaration 9] 14-21. 33 See also S. Roveda Declaration J 11; N. Weinberg-Lynn Declaration §¥ 32-33, 35; M. Bodziak Declaration J 21; P. Brooks Declaration 22, 26; J.A. 733-38, Valero Exs. 1-2. 53 15cv1749-JO-AGS and 18cv1374-JO-AGS

} Harris Declaration § 19; P. Brooks Declaration J 33; K. Archambault Declaration {J 45, Based on the above, the Court finds that a reasonable juror could conclude Defendants made CARBOB production decisions based on their own profit maximizing calculus and logistical constraints rather than pursuant to a conspiracy to reduce supply. Especially because exchange agreements and spot market trades acted as insurance against short-term shortages, Defendants’ decisions to produce profit maximizing amounts of CARBOB while maintaining low inventories to reduce costs could have been driven by plausible and legitimate business objectives. The Court therefore concludes that Defendants have met their Step One burden with regard to these allegations. At Step Two of its analysis, the Court examines whether Plaintiffs offer any evidence tends to exclude the possibility that Defendants acted independently and in their own self-interest in setting CARBOB production levels. Plaintiffs’ raw assertion that Defendants could have and, therefore, should have produced more CARBOB is not a sufficient basis for the Court to conclude that Defendants acted against self-interest. Antitrust law does not compel Defendants to produce the maximum amount of CARBOB possible; it only requires that Defendants’ production decisions be made independently for legitimate business reasons. At the least, information about the comparative costs and profitability of producing more CARBOB versus other products would be necessary to evaluate whether foregoing additional CARBOB production was not in Defendants’ financial self-interest. Plaintiffs submit no proof that any Defendant’s production or 54 15¢v1749-JO-AGS and 18cv1374-JO-AGS

/inventory decisions were not profit maximizing given demand, costs, and logistical constraints.*4 See generally Opposition. Instead of offering evidence about the economics of Defendants’ CARBOB supply decisions, Plaintiffs instead point to communications between Defendants’ traders as circumstantial evidence that the companies colluded to depress supply and shared confidential information to implement that scheme. Opposition at 27-37. Because Plaintiffs offer largely overlapping examples of trader communications as circumstantial evidence regarding Defendants’ other purported efforts to restrict supply, the Court considers all of this evidence together below. After reviewing the communications submitted by Plaintiffs as “plus factor” }evidence of an agreement to “strangle” supply, see id, the Court finds that the trader }communications generally fall into two categories: (1) communications in which traders different companies generally expressed a preference for low-supply/high-price conditions while speculating about how third-party actions might impact the market; and (2) communications, which only occurred between BP and Phillips 66, constituting an actual exchange of import, export, or production information between traders from different companies. The Court examines each of these categories of communications in turn for their tendency to exclude the possibility that Defendants acted independently in making supply decisions. The first category features emails between traders, speculating regarding competitors’ actions and bemoaning actions that could bloat supply and lower prices. For * Plaintiffs do not appear to argue that Defendants’ refineries sat idle or that Defendants forewent production opportunities. Rather, Plaintiffs appear to argue that, to the extent Defendants produced other products, they should have produced CARBOB instead. See Opposition at16. 55 15ev1749-JO-AGS and 18cv1374-JO-AGS

example, one BP trader and one Phillips 66 trader made various comments of this nature to one another over the years, including that “[i]f the refineries don’t cut runs it could really get bad”; “[I] think we will structurally see more exports going forward”; “hopefully some of the current PMI exports will show up in the next week stats”; and, “[d]ec could be a S__ storm if exports don’t clean us up.” See Pltfs. Ex. 8. One internal BP email among its traders noted “chatter” that “AOT” and Shell were importing CARBOB to the West Coast stated that “everyone is becoming very cognizant that high prices will attract oil and slowed down their buying as a result.” Pltfs. Ex. 48 at BPWC-PG-00039839; see also Pltfs. Ex. 8 at rows 852-53 (message between BP and Phillips 66 also acknowledging that traders may slow down buying to discourage imports.) Another internal BP email among traders noted that Valero was importing a cargo to LA and that “Noble” rerouted a cargo headed to the Pacific Northwest. See Pltfs. Ex. 50 at BPWC-PG-00023780; see also □□□□□□ 8 at rows 8329-9520; Pltfs. Ex. 83 (excerpted messages lamenting supply conditions without any context or evidence of a reply). Some of these trader messages are written in unguarded and colorful language but, similarly, do no more than show competitors scrutinizing one other’s actions or expressing displeasure at high supply conditions. In one set of emails, traders referred to market conditions that attracted imports as the “bug light” being on, presumably referring to the fact that “bug lights” attract undesirable nuisances. Pltfs. Ex. 11 at row 134864; Pltfs. Ex. 48 at BPWC-PG-00039827; Pltfs. Exs. 79-82. In one of these conversations, BP and Phillips 66 traders—apparently discussing what would happen if imports arrived—stated that “everyone will poop in our back yard, kill our market and leave us the dingle berries.” Pltfs. Ex. 8 at 592-95. In another email, an unidentified person asked Phillips 66’s trader 56 15cv1749-JO-AGS and 18cv1374-JO-AGS

he had any alkylate needs, and Phillips 66’s trader responded, “no and don’t bring it the west coast.” Pltfs. Ex. 84. The second category of communications, the ones which demonstrate an exchange of potentially confidential import/export information between traders, all occurred between BP and Phillips 66 and are discussed above. See supra Section V.D.2. Because }communications of this kind only occurred between these two traders at these two refineries, they do not create an inference of conspiracy for all eight Defendants. They may, however, create an inference of conspiracy as to BP and Phillips 66, which the Court discusses separately supra. Together, these communications and internal documents paint a picture of a small group of competitors who go out of their way to obtain information about each other’s /actions to inform their own. See Krehl, 664 F.2d at 1357 (“the mere exchange of price /information, without more, is not per se illegal”). They also portray Defendants’ hyperawareness of how competitors’ actions impact supply and price conditions for the }whole market. The microcosm that these communications illustrate is consistent with Plaintiffs’ characterization of the California gas market as highly concentrated and interdependent. See McCullough Report 66; Defs. Ex. 8 at 44:15-22. Such conscious parallelism in a market of this nature is not illegal, see supra Section V.C, and, without more, does not give rise to an inference of conspiracy. The Court next considers whether the fact that traders expressed preferences for lower-supply and higher-price conditions constitutes the “more” that tends to exclude the possibility that Defendants acted independently in making supply decisions. The Court finds that these trader discussions may establish that the eight Defendants, in general, had motive to collude to maintain low-supply conditions. Motive alone, however, does not 57 15ev1749-JO-AGS and 18cv1374-JO-AGS

permit an inference of agreement among the Defendants. Reserve Supply Corp. v. Owens- Corning Fiberglas Corp., 971 F.2d 37, 51 (7th Cir. 1992) (quoting 6 Phillip E. Areeda, Antitrust Law § 1411 (1986)); In re Late Fee & Over-Limit Fee Litig., 528 F. Supp. 2d 953, (N.D. Cal. 2007), aff'd, 741 F.3d 1022 (9th Cir. 2014) (“[a]s one court put it, if ‘a ;}motive to achieve higher prices’ were sufficient, every company in every industry could be accused of conspiracy”) (quoting Baby Food, 166 F.3d at 133). With the exception of BP and Phillips 66, these communications do no more than illustrate that Defendants watched each other closely and desired lower supply conditions to generate higher profits. With regard to the six Defendants not including BP and Phillips 66, the Court concludes Plaintiffs’ proffered evidence does not tend to exclude the possibility of independent decision making with regard to gasoline production and supply. To }summarize, Plaintiffs offer evidence that these six Defendants, (1) had the ability to increase CARBOB supply but did not, (2) shared a desire to maintain lower-supply market conditions, and (3) gathered information and closely monitored each other’s market behaviors. From this, a reasonable juror could infer conspiracy but could equally infer that the six Defendants independently made profit maximizing production decisions while watching each other closely to inform those decisions. While it is illegal to act pursuant to a conspiracy, it is not illegal to act pursuant to oligopolistic self-interest. Theatre Enterprises, 346 U.S. at 540-41 (“this Court has never held that proof of parallel business behavior conclusively establishes agreement” without more). With regard to BP and Phillips 66, the additional evidence that their traders shared confidential information and passed that information to their production departments tends to exclude the possibility that these two refiners made independent decisions about gasoline production. the remaining six refiners, however, the above evidence submitted by Plaintiffs does 58 15ev1749-JO-AGS and 18cv1374-JO-AGS

tend to exclude the possibility of independent action and, thus, is insufficient to create inference of conspiracy. 4. Supply Restrictions by Preventing Imports Plaintiffs also contend that Defendants’ actions to prevent gas imports into California is “plus factor” evidence of an eight-Defendant conspiracy to fix gas prices in California. See Opposition at 28-33. In addition to the trader conversations discussed above, Plaintiffs point to the following specific instances where Defendants rerouted imports originally intended for the West Coast as circumstantial evidence of conspiracy: (1) Chevron diverted two cargoes of unidentified product in August 2014, (Pltfs. Ex. 85); (2) Valero “cancelled another 80mb barge of Al to LA” in 2014 or 2015,* (Pltfs. Ex. 86 }at VMSC_003193); (3) Shell diverted a cargo originally intended for Los Angeles to New York sometime in late 2015, or early 2016,*° (Pltfs. Ex. 87 at p. 602; Pltfs. Ex. 88 at SOPUS PGI 00223181); and (4) BP diverted a cargo originally intended for California to New York in January 2016, (J.A. 915—16, BP Ex. 6). To meet their Step One burden of providing plausible and justifiable reasons for their actions, Defendants explain that importing gas into California is a slow and costly proposition and by the time imports actually arrive, market conditions may have changed such that importing is no longer profitable. In general, Defendants scheduled few imports because they were often logistically and economically unfeasible. See, eg, K. 2) ©.

35 Plaintiffs do not submit any evidence regarding what “A1” is or the dates of these cancellations, although the document appears to be dated in March 2014 or 2015. □□ Plaintiffs do not identify the date of this diversion, although the documents appear to indicate that it occurred sometime in late 2015 or early 2016. 15cv1749-JO-AGS and 18cv1374-JO-AGS |

Archambault Declaration 22-26, 34; H. Henderlite Declaration § 23.7’ Because California is a “gasoline island,” imports could only be transported by sea. See id. And due to regulatory requirements, domestic imports could only be delivered by the limited number of available Jones Act-compliant vessels (U.S.-assembled, -owned, and -crewed) generally cost $10—-20 million per vessel, per year. See, e.g., K. Archambault Declaration {J 22—23. Defendants Chevron, Valero, and BP also explain that the reroutes Plaintiffs highlight were necessary because by the time the import arrived weeks later, price margins had decreased, and Defendants determined that selling elsewhere would be more profitable. See, e.g., C. Yates Declaration J] 38-41; P. Brooks Declaration 4 47-48, 51— 57-59; see also Joint MSJ at 32; H. Henderlite Declaration 23-27; J.A. 915-916, BP Ex. 628 The Court notes that some of Plaintiffs’ evidence supports, rather than ——_ ———————— 37 See also C. Yates Declaration {J 38-41; J.A. 30-32, Chevron Ex. 3; R. Sharum Declaration 15-17; L. Lockhart Declaration J 8; P. Brooks Declaration 47-48, 54; C. Dickson Declaration 52-56; M. Perez Declaration {| 6-8. 38 BP also submitted evidence that its alleged reroute in January 2016 was in fact a complex transaction with Valero that benefitted BP. See Joint MSJ at 32; H. Henderlite Declaration {{ □□□□□□ J.A. 94 5-916, BP Ex. 6. Valero had a cargo shipping from the United Kingdom to its Bay Area refinery but realized that it could not get its cargo into its harbor because the cargo was too heavy to get through the “Pinole Shoals,” a channel that had to be crossed to reach Valero’s Bay Area refinery. See id. At the time, BP had its own cargo coming to its Los Angeles refinery but realized that the cargo was unlikely to arrive in time to be used there because it was stuck in the Pacific Northwest due to weather. id Thus, BP sold its cargo to Valero since it was not going to reach BP’s Los Angeles refinery in time and because BP was already at storage capacity at its Bay Area refinery and could not use the gas See id. BP also purchased Valero’s shipment coming from the United Kingdom. See id. BP ultimately determined that selling Valero’s cargo into New York was more profitable than bringing it to the West Coast—at that time, Valero’s shipment was near Jamaica and thus was “over 1,000 nautical miles closer to the East Coast” than to the West Coast, and East Coast prices at that time were better than what West Coast prices were predicted to be by the time Valero’s cargo would have arrived at one of BP’s West Coast refineries. See id. 60 15cv1749-JO-AGS and 18cv1374-JO-AGS

}undermines, Defendants’ explanation that imports were rerouted because market conditions changed while shipments were en route. For example, Plaintiffs submit an internal Shell presentation to prove that Shell’s reroute occurred, but the presentation itself explains that, “[w]ith the LA market weakening, STUSCO was able to divert the cargo to NYH and take $1.5mln of value on the cargo for the West Coast.” See Pltfs. Ex. 87 at p. 602. The Court finds that based on the above, Defendants have offered plausible and legitimate explanations consistent with the limitations of a “gasoline island” where imports generally cannot arrive quickly, and gas prices are prone to shifts such that prices may be different by the time an import arrives. See Defs. Ex. 2; McCullough Report J 37-39. On the whole, the evidence demonstrates that Defendants did reroute imports in the four instances identified by Plaintiffs (see Pltfs. Exs. 85, 86—-89),*? but Plaintiffs do not demonstrate that these reroutes were common or against Defendants’ self-interest. Plaintiffs argue that the fact that these reroutes occurred is sufficient to support a reasonable /inference of conspiracy because it does not make economic sense to sell CARBOB in a lower-priced market than California. See Opposition at 32—33. Plaintiffs, however, do not point to any evidence that selling in California would have been more profitable for the handful of rerouted imports they identify. || _—_vU0——_| Defendants concede on summary judgment that BP rerouted a cargo in January 2016 as 77 described above, see supra note 38, and thus, the existence of this reroute is not disputed. In opposing summary judgment, Plaintiffs point to Defendants’ admission that the reroute occurred but do not point to evidence in response to Defendants’ Step One explanation; instead, Plaintiffs merely state, without supporting evidence, that: “[this was] obviously against BP’s self-interest” because the East Coast is a “lower priced market.” Opposition at 33. 61 15cv1749-JO-AGS and 18cv1374-JO-AGS

With regard to the six Defendants other than BP and Phillips 66, the Court concludes Plaintiffs’ proffered evidence regarding import decisions does not tend to exclude the possibility of independent decision making. To summarize, Plaintiffs offer (1) four instances of rerouted imports across a seven-year period from a pool of eight Defendants; (2) no evidence that any of these reroutes were against economic interest; and (3) trader communications discussing imports and expressing a preference for low-supply conditions. As discussed above, the communications that tend to show that Defendants closely monitored each other’s actions and generally preferred low-supply conditions do not, without more, support a reasonable inference of conspiracy. Considered together with the infrequent nature of rerouted imports and the lack of any evidence suggesting these reroutes were not profitable, a reasonable factfinder could infer that Defendants made independent decisions to disfavor imports in general and reroute imports when it was profitable to do Plaintiffs’ evidence, therefore, does not tend to exclude the possibility that Defendants acted independently for self-interested reasons. 5. Supply Restrictions Through Exports Plaintiffs also argue that Defendants exported gas out of California against self- interest when it would have been more profitable to sell in-state. They contend that these j|actions provide further “plus factor” evidence that the eight Defendants in this case conspired to fix gas prices in California. See Opposition at 33-37. To support these arguments, Plaintiffs submit three categories of evidence. First, Plaintiffs contend that Defendants increased their exports over the class period even though \their production capacities did not correspondingly increase. See id. at 34; Pltfs. Ex. 200. Second, Plaintiffs point to certain Defendants who dual certified their gas to meet both CARBOB specifications and other state requirements. See Pltfs. Exs. 94, 97-98; see also 62 15ev1749-JO-AGS and 18cv1374-JO-AGS

Opposition at 34-35. Third, Plaintiffs identify two specific exports that they claim were against self-interest: (1) Phillips 66 exported 5,003 gallons of CARBOB from California to Nevada in 2017, (Pltfs. Ex. 96 at 183:17—25 at p. 602); and (2) Tesoro began exporting gas to Mexico in 2017, (Opposition at 37). To meet their burden at Step One, Defendants explain that they independently made necessary, profit maximizing business decisions to export non-CARBOB and to dual certify gas. With respect to exports of non-CARBOB, Defendants start by explaining that manufacturing CARBOB necessarily creates byproducts that cannot be blended into CARBOB and thus cannot be sold in California. See, e.g., C. Yates Declaration 46.” Defendants therefore sold all product that did not meet CARBOB specifications outside of California to make a profit from it. See id. In addition to exporting product that could not be sold in California, certain Defendants also utilized their California refineries to meet supply obligations in nearby geographies (e.g., Arizona) that were most conveniently supplied from California. See, C. Yates Declaration ff 43-48." Defendants explained that they did so to meet supply obligations in nearby states like Arizona. See P. Brooks Declaration J 24—25, 55; Pltfs. Ex. 94; T.A., M. Pais Declaration J 13. For instance, Valero had contractual obligations to sell gas in Arizona, which requires “AZRBOB.” See P. Brooks Declaration Jf 24, 55. Because Arizona is connected by pipeline to both California and Texas, Valero usually supplied Arizona through its Texas refinery but supplemented from their California refinery when 40 See also N. Weinberg-Lynn Declaration {| 18-20; R. Sharum Declaration 10-12, 14; J. Harris Declaration {| 26-27; J. Marino Declaration § 3; P. Brooks Declaration 4 61-63. 41 See also P. Brooks Declaration 4 24-25, 63; J.A. 805, Valero Ex. 11. 63 15cv1749-JO-AGS and 18cv1374-JO-AGS

Texas refinery could not supply enough product to meet Valero’s contractual }commitments in Arizona. See id. Dual certifying its gas as CARBOB and AZRBOB ;enabled this flexibility. See id. Finally, with respect to specific exports, Phillips 66 submitted evidence that it exported 5,003 gallons of CARBOB to its Nevada site in 2017—“an extraordinarily unusual occurrence” likely following a supply disruption—which “constituted less than □ of 1 percent of one day’s production.” R. Sharum Declaration §.13. As to Tesoro, it explained that exports to Mexico in 2017 did not impact its supply commitments in California and were part of a long-term strategy to develop in a market with growing demand when California demand was forecasted to decline. See T.A. 348-352. The evidence discussed above satisfies Defendants’ Step One burden to show that Defendants exported gas consistent with proper business practice. For one, a reasonable could find that Defendants justifiably exported non-CARBOB that they could not sell California. In addition, a reasonable juror could find that the practice of dual certification is consistent with competition. Dual certification produces a more fungible form of gas, which would have allowed BP and Valero flexibility as market needs arose. For instance, it would be reasonable for Valero to dual certify gas so that it could supply Arizona from California when its Texas refinery could not meet supply needs. Finally, based on Phillips 66’s and Tesoro’s explanations regarding their exports of CARBOB, a reasonable juror could find that these specific exports were justified by proper business practice. Phillips 66 explained that its one-time export was de minimis and justified by an unusual supply need. Tesoro explained that its 2017 exports to Mexico were performed for the purpose of establishing market presence in Mexico according to its long-term business strategy. Because Defendants meet their burden at Step One, the Court turns to 64 15cv1749-JO-AGS and 18cv1374-JO-AGS

whether Plaintiffs meet their Step Two burden to identify evidence that tends to exclude possibility that Defendants exported gas independently. Plaintiffs attempt to meet their Step Two burden to show that Defendants’ exports were inconsistent with unilateral conduct by first pointing to a general increase in export volumes between 2000 and 2020. Plaintiffs point to government data reflecting yearly “finished gasoline” exports from the West Coast. See Pltfs. Ex. 200. This aggregated data, however, does not contain any information on (1) the extent to which the state of California accounts for those exports, (2) the extent to which Defendants specifically account for those exports, or (3) how much of those exports were CARBOB versus non-CARBOB. See id. Plaintiffs infer from this overall trend that some of Defendants’ exports were j/unnecessary, but this generalized evidence does not shed light on whether the eight Defendants in question engaged in unnecessary exports or whether any of their exports were against their financial self-interest. Plaintiffs also reiterate that certain Defendants certified gas and argue that this practice was unprofitable but offer no evidence to support that argument. See Opposition at 34-35. Next, Plaintiffs highlight two specific instances of exports against self-interest by Phillips 66 and Tesoro. First, Plaintiffs point to a Phillips 66 export from California to }Nevada in 2017 as evidence showing that Defendants “routinely dumped” CARBOB outside of California. See Opposition at 34-35. The only evidence that Plaintiffs point to regarding Phillips 66’s 5,003-gallon CARBOB export to Nevada in 2017 is deposition testimony that merely confirms the existence of that export. See Pltfs. Ex. 96 at 183:17— In the face of Phillips 66’s explanation that this was a one-time export of a de minimis amount of CARBOB, representing less than “2 of 1 percent of one day’s production,” the fact that the export occurred does not tend to exclude the possibility that Philips 66 65 15cv1749-JO-AGS and 18cv1374-JO-AGS

)independently made this export to cover its own supply shortage, rather than pursuant to a conspiracy. See R. Sharum Declaration 7 13. As to Tesoro, Plaintiffs do not specify whether those exports were CARBOB versus non-CARBOB, nor do they submit any evidence to support the argument that those exports were unprofitable.*” Aside from the bald assertion that gas is cheaper in Mexico, Plaintiffs have submitted no evidence that Tesoro’s exports were not profitable, either in the short term or long term. Instead, Plaintiffs submit two documents: (1) a Chevron document noting that certain exports to Mexico would be uneconomic in 2014; and (2) data showing that California gas prices in late 2020 and early 2021 were above $3.00. See Pltfs. Exs. 100, 106. Whether exports to Mexico would have been uneconomic for Chevron in 2014 has little bearing on whether such exports would be equally uneconomic for Tesoro in 2017, especially given Tesoro’s /long-term business plan to expand to that geographic market. See Pltfs. Ex. 100.% Similarly, Plaintiffs’ data that gas was selling for $3.00 per gallon in California during 2020 and 2021 has little bearing on the overall lack of profitability for Tesoro’s long-term growth plan. See Pltfs. Ex. 106. Without more concrete evidence, a factfinder should not second guess Tesoro’s long-term business strategy “where the evidence concerning the rationality of the challenged activities might be subject to reasonable dispute.” Citric Acid, 191 F.3d at 1101. Therefore, the Court finds that Plaintiffs have not pointed to evidence of exports that tends to exclude the possibility of independent conduct. —_ —_——_ 42 Despite citing to certain exhibits, Plaintiffs fail to attach those exhibits to the record, which are marked “purposefully omitted” and thus are not considered by the Court. See Pltfs. Exs. 101-04. 43 This Chevron document appears to discuss exports to Mexico being uneconomic, at the time, because of a variety of factors, including production costs, Chevron-specific refinery considerations, and other then-present market factors. 66 15cv1749-JO-AGS and 18cv1374-JO-AGS

6. Cooperation After the 2015 Exxon Explosion Plaintiffs also point to Defendants’ behavior following an explosion at the Exxon {refinery in 2015 as indicative of improper cooperation among supposed competitors. See Opposition at 22-24. Plaintiffs allege that instead of competing for Exxon’s market share when its refinery was down, as one would expect competitors to do, Defendants sold gas Exxon to help it cover its supply shortage. See id. at 23. Plaintiffs also argue that when Defendants sold gas to Exxon, they should have done so at higher prices to take maximal advantage of Exxon’s crisis. See id. at 22—23. Plaintiffs claim that these actions were quid pro quo favors in furtherance of the conspiracy. See id. To satisfy their Citric Acid Step One burden, Defendants explain their responses to Exxon explosion in 2015 as follows. Defendants offer evidence showing that each refinery promptly responded to the explosion by maximizing in-state supply to take advantage of the overall shortage in the market and attendant price increases. See, e.g., K. Archambault Declaration 38-40; H. Henderlite Declaration □□ 29-32.% Defendants Shell, Tesoro, Chevron, Valero, and Phillips 66 also explain that they sold gas to Exxon at prevailing market rates rather than trying to circumvent the long-term contracts Exxon maintained station-by-station with its retailers. See, e.g., C. Dickson Decl. 49 24—25; S. Roveda Declaration § 13; R. Sharum Declaration □□ 31-33; J.A. 339-53, Exxon Ex. 1.% Defendants explain that pursuing Exxon’s retail customers would have entailed convincing each local retailer to break its Exxon contract, and thus selling gas to Exxon directly □□□ °°

“4 See also S. Roveda Declaration {J 10-12; N. Weinberg-Lynn Declaration § 32-33; M. Bodziak Declaration J 21; P. Brooks Declaration §f 26, 46; J.A. 733-38, Valero Exs. 1-2. 45 See also M. Perez Declaration {J 13-16; L. Lockhart Declaration 4 30; P. Brooks Declaration 90; E. Pestano Declaration § 42-45. 15cv1749-JO-AGS and 18cv1374-JO-AGS

market rates was the more attractive and profitable alternative. See id.; see also Joint MSJ 44. Because it is plausible that pursuing individual negotiations with Exxon’s retail stations could be costly and time consuming, the Court finds that Defendants have produced a legitimate reason for preferring to sell gas to Exxon directly instead. Accordingly, Defendants meet their Step One burden. At Step Two, Plaintiffs respond that it nonetheless would have been better for Defendants to pursue Exxon’s retail market share or extract higher prices from Exxon. See Opposition at 22-24. The bare contention that pursuing Exxon’s market share was the better economic course does not create a genuine dispute for the purpose of summary judgment. Plaintiffs do not point to any specific evidence which would allow an inference that sales were made to Exxon at below-market prices or that such sales were favors to Exxon rather than profit maximizing decisions. Moreover, based on evidence that multiple refiners sold gas to Exxon, the inference that no one Defendant had the leverage to extract higher than market-rate prices from Exxon is as plausible as the inference that Defendants conspired to help Exxon against self-interest. Plaintiffs point to one email on February 24, 2015, from a Chevron employee in an unidentified role, questioning why Chevron would sell alkylate, a gasoline blending component, to Exxon “from a competitive standpoint.” See Pltfs. Ex. 187. But without any evidence in the record regarding the economics of this sale, one employee questioning a sale cannot support an inference that this was an unprofitable transaction, much less a collusive action pursuant to a conspiracy. The evidence in the record concerning Defendants’ reactions to the Exxon explosion therefore does not suggest behavior that tends to exclude the possibility of independent conduct, and does not permit a reasonable inference of conspiracy. /// 68 15ev1749-JO-AGS and 18cv1374-JO-AGS

7. Market Manipulation Through a “Gentleman’s Agreement” Plaintiffs allege that a “gentleman’s agreement” between West Coast traders to limit trading hours provides further circumstantial evidence of collusion between Defendants. See Opposition at 49. In an email exchange in August 2014, the West Coast trading community—consisting of refiners, brokers, and large purchasers—voted and agreed to change the daily trading cut-off time from 3:30 p.m. Pacific Coast Time to 2:15 p.m. Pacific Coast Time, thereby aligning trading hours with OPIS“ reporting hours. See id.; see also 583-86, Phillips Ex. 20. Plaintiffs state that the 2014 agreement to “institute arbitrary trading hours” was intended to increase cooperation among Defendants and deter traders other parts of the world from participating in the market. See Opposition at 49. Defendants explain that the proposal to move up the trading cut-off to 2:15 p.m. was driven by a desire to (1) better facilitate pipeline scheduling, (2) match OPIS’s cut-off time that transactions were more likely to be reported, and (3) accommodate industry personnel in the Central time zone. See, e.g., L. Lockhart Declaration □□ 31-32. Defendants also submit OPIS reports from 2014 that document these reasons and show that 27 entities participated in the vote. See McCullough Report 297-300 (“the vote was dominated by independent traders as opposed to majors who own refineries on the West Coast”); see also J.A. 562-76, Phillips Ex. 17; J.A. 583-86, Phillips Ex. 20. A reasonable factfinder could infer from this evidence that Defendants agreed to the adjusted trading hours for legitimate business reasons related to convenience and transparent price reporting. Defendants therefore meet their burden at Step One. ; —__ 46 OPIS is a price-reporting agency for the petroleum fuel supply chain, including rack, spot, and prices. See Joint MSJ at 48. In the West Coast market, OPIS collects trading data and publicly 45 reports average pricing information. See J.A. 562-76, Phillips Ex. 17. 69 15ev1749-JO-AGS and 18cv1374-JO-AGS

In response, Plaintiffs do not point to any specific evidence tending to show that the agreement was not legitimate, self-interested behavior. Plaintiffs acknowledge that the agreement “was supposedly meant to align trading hours with... OPIS” but argue that “‘[t]his informal agreement to institute arbitrary trading hours could act as a deterrent to traders in other parts of the world that may not operate at the hours identified.” Opposition 49. The same could be said about any market’s trading hours. The possibility that trading hours could be inconvenient for unspecified actors “in other parts of the world” is not evidence and does not tend to rule out the possibility of independent action. Accordingly, the Court finds that the “gentleman’s agreement” does not permit a reasonable inference of conspiracy. 8. Market Manipulation Through Wash Trades Plaintiffs further contend that Defendants engaged in deceptive “wash trades” to /create a false sense of demand and raise prices in the market. Opposition at 47-49. || According to Plaintiffs’ expert, a “wash trade” occurs when party | trades a quantity of product to party 2 and, around the same time, party 2 trades a similar quantity of the same product back to party 1, at the same price. See McCullough Report 401-06. Trading activity in the form of a “wash trade” has zero net economic effects for the trading parties creates a falsely inflated sense of demand in the market. See id.; see also McCullough Report § 403. To provide circumstantial evidence of an eight-Defendant conspiracy, Plaintiffs identify a handful of what they allege are wash trades performed by Chevron, Shell, and Valero. See Plitfs. Exs. 170, 176-70. At Step One, Defendants offer explanations to demonstrate that the alleged “wash trades” were, in fact, legitimate transactions. Citric Acid, 191 F.3d at 1094. Each Defendant accused of executing a wash trade points to evidence showing that the trade in 70 15¢v1749-JO-AGS and 18cv1374-JO-AGS

question does not fit Plaintiffs’ definition of a wash trade—that is, parties trading similar quantities of gas, at similar prices and times, and without economic benefit. First, Chevron submits evidence that its alleged wash trades did not occur on the same day (and thus are not similar in time), were executed by different brokerages (and thus were not direct trades with other Defendants), and closed at different prices. See R. Pluimer Declaration § 11; J.A. 92, Chevron Ex. 9. Chevron argues that because these trades were not similar in time price and were not direct trades with other Defendants, they do not fit Plaintiffs’ allegations of two trades that cancel one another out. Second, Shell submits evidence that for each alleged wash trade, more than one of the following circumstances was present: the /trades were (1) for different products, (2) performed at different locations in a location- swap, (3) closed at different prices, (4) separately facilitated by unrelated brokers, (5) performed at different times, (6) structurally distinct (e.g., different delivery times), or (7) one-way transactions, not trades. See J.A. 164-81, Shell Exs. 7-10; J.A. 201—20, Shell 13-17; J. Marino Declaration {J 6, 9-11, 13-14; S. Rodrick Declaration □□ 7, 9; R. Pluimer Declaration § 11; J.A. 92, Chevron Ex. 9. Third, Valero submits evidence that its /alleged wash trades were beneficial to Valero, carried economic risk, and were for materially different products exchanged on different dates. See P. Brooks Declaration § 78; 804-13, Valero Exs. 14-15. Valero argues that this evidence shows that the trades not have cancelled one another out as Plaintiffs theorize. The Court finds that Defendants meet their burden at Step One to rebut Plaintiffs’ evidence of wash trades with evidence that would permit a reasonable juror to conclude that the trades in question □□□□ in fact illusory “wash trades” with no other benefit than manipulating public perceptions in the market. 71 15¢v1749-JO-AGS and 18cv1374-JO-AGS

l In opposing summary judgment at Step Two, Plaintiffs discuss only one set of /transactions between two Defendants (Chevron and Shell) as evidence tending to exclude {independent action. Plaintiffs’ evidence shows that in 2015, a Chevron trader suggested in an internal email that he would sell Chevron’s “swaps” after his coworker was “in buying for a couple of days.” See Pltfs. Ex. 176. This trader testified at his deposition that he was hoping that his coworker’s buying “would push the mark - he would have the market impact, and the market would possibly move up” so that prices would be higher by the time sold Chevron’s “swaps.” See Plitfs. Ex. 170 at 216:9-17; see also Pltfs. Ex. 176. In the days after this communication, his coworker did in fact execute several transactions, including two with Shell on April 20—21, 2015. See Plitfs. Ex. 177. Plaintiffs argue that evidence permits the inference that these transactions between Chevron and Shell on April 20-21, 2015, were wash trades performed pursuant to a conspiracy. Plaintiffs fail to meet their burden at Step Two because Plaintiffs’ evidence does not suggest any concerted action between companies, as opposed to within one company. 15 U.S.C. § 1; In re Dynamic Random Access Memory (DRAM) Indirect Purchaser Antitrust Litig., 28 F.4th 42, 48 (9th Cir. 2022) (“[o]nly concerted activity is actionable under Section 1” of the Sherman Act). Even if Plaintiffs’ evidence at Step Two permitted the inference that this trader’s coworker traded in the market for the sole purpose of raising market prices to benefit that trader’s later “swap” sales, it remains unclear how this would allow for a plausible inference that Chevron and Shell entered into an unlawful agreement, without more. Thus, the Court concludes that Plaintiffs have not met their Step Two burden to point to specific evidence that tends to exclude the possibility of independent conduct. /// /// 72 15ev1749-JO-AGS and 18cv1374-JO-AGS

9. Market Manipulation Through Selective Price Reporting Plaintiffs argue that their evidence demonstrates that traders at Chevron, Shell, Tesoro, and Phillips 66 selectively reported only higher-priced trades to OPIS to manipulate market prices. See Opposition at 45-47. OPIS is a third-party pricing agency publishes market information, including average daily spot prices. See J.A. 562—76, Phillips Ex. 17; M. O’Neal Declaration J 4; C. Dickson Declaration § 59. OPIS collects a sample of transactions and publishes the high, low, and average prices in the Los Angeles and San Francisco markets. See J.A. 562—76, Phillips Ex. 17; L. Lockhart Declaration 12-13. Defendants priced their gas at the rack and Dealer Tankwagon, in part, on price indices such as OPIS.” 1] Plaintiffs and Defendants dispute whether there is any evidence in the record of selective reporting. At Step One, Defendants argue that Plaintiffs have failed to point to evidence of selective reporting, and thus that there is no genuine dispute of material fact with respect to this plus factor. See Dkt. 719 (“Defendants’ Joint Reply”) at □□□□□□ Step Two, Plaintiffs point to communications where traders asked brokers to report completed trades to OPIS and where traders questioned whether trades would be reported to OPIS. See Opposition at 45-47; see also Pltfs. Exs. 11, 76-77, 169, 171, 174. Accordingly, the Court turns to whether Plaintiffs’ evidence is sufficient to create a genuine dispute of material fact on whether any Defendants engaged in selective reporting. 2) .

73 “7 See, e.g., K. Archambault Declaration 59, 61, 63; M. O’Neal Declaration 4; J. Hodgson Declaration {J 11, 13-14, 19-20, 23, 27; D. Smith Declaration § 3; P. Brooks Declaration ff 10, 91, 94; C. Dickson Declaration □□ 27-29; T.A. 3, W. Eckard Declaration 9 10-11. 73 15cv1749-JO-AGS and 18cv1374-JO-AGS

Examination of Plaintiffs’ evidence shows that what Plaintiffs deem “selective reporting” does not allow for an inference of conspiracy. With respect to Shell and Chevron, the evidence, at most, shows that their traders consistently reported all trades. See Pltfs. Ex. 11 at line 61501 (Shell trader: “Every deal I do needs to go to opis”); Pltfs. Ex. 77 at 87:7-14 (“Q: In other words . . . you as the trader had your own requirement that each one of your deals be reported to OPIS in some fashion? A: I wanted my deals to be reported to OPIS, yes.”); Pltfs. Ex. 171 (Chevron trader: “I would like that deal reported as /normal’”). The evidence with respect to Phillips 66 similarly falls short. Plaintiffs point to document where a broker asked a Phillips 66 trader whether he should “report this to ... or keep p and c,” and the Phillips 66 trader replied, “report please.” Pltfs. Ex. A reasonable juror could not infer from the above conversations that Defendants worked together to selectively report OPIS trades. The Court turns to the remaining evidence consisting of testimony from Tesoro’s trader that he tried to hide certain trades from being reported. This trader testified that he would typically avoid trading through a broker when he was covering an unexpected shortfall, so that the broader market would not “see a buyer looking for 500,000 barrels of gasoline.” See Pltfs. Ex. 76 at 222. This evidence that Tesoro’s trader sought to keep certain information from the market when it did not benefit Tesoro may raise an inference of inappropriate conduct by Tesoro’s trader. It does not, however, create an inference that he did so pursuant to an agreement with other Defendants rather than pursuant to what he deemed to be in Tesoro’s unilateral interest. Because evidence that is equally consistent with conspiracy and independent conduct does not tend to exclude the possibility of independent conduct, Plaintiffs’ evidence of selective reporting does not permit an inference of conspiracy. 74 15cv1749-JO-AGS and 18cv1374-JO-AGS

10. Market Manipulation Through False Public Statements Plaintiffs’ complaints originally contended that Defendants publicly made false or misleading statements about refinery outages as part of an agreement to manipulate California gas prices. See Dkt. 76 at 18-20, 28-31 (detailing Defendants’ “claimed reason[s]” for maintenance and alleging these reasons were false); Bartlett, Dkt. 44 at 10— 11, 17-20 (same). On summary judgment, however, Plaintiffs appear to be pursuing a slightly different theory. In their opposition, Plaintiffs argue that market information reported by the media was “laughably incorrect,” and that Defendants had “superior information” but made no efforts to correct inaccurate media reports. See Opposition at }20—21. Plaintiffs point to evidence like the following: In 2012, BP internally discussed an article reporting on a fire at BP’s refinery, and one employee stated that the article was “off several facts, which is good.” See Pltfs. Ex. 54. In addition to evidence like the foregoing, Plaintiffs argue that Defendants “exploit[ed] this [information] gap” by “present[ing] information contrary to known facts.” Opposition at 21. In support of this sweeping allegation, Plaintiffs submit just one instance where Valero publicly announced its maintenance schedule but may have concluded maintenance earlier than announced. See Pltfs. Ex. 56. Defendants address their Step One burden by submitting evidence tending to show that the alleged misleading statements Plaintiffs pointed to in their complaints were accurate, boilerplate, and not misunderstood by the industry. Exxon submits evidence to that its brief response to a media inquiry regarding its October 1, 2012 power outage was factually accurate. See Joint MSJ at 39-40; J. Kechkian Declaration J 8; J.A. 393-96, Exxon Ex. 9. Chevron and Shell similarly submit evidence that the statements each made were factually accurate. See Joint MSJ at 40; C. Yates Declaration J] 25-28; G. Johnson 75 15ev1749-JO-AGS and 18cv1374-JO-AGS

Declaration 4-10. Valero also argues that public statements about how long a refinery unit is estimated to be offline, made soon after an incident, are based on circumstances that are subject to change and often do change. See P. Brooks Declaration 4102. Valero contends it is widely understood in the industry that such initial estimates are usually revised as additional facts become known, and thus no one is misled when refineries give /initial maintenance estimates. See id. Defendants’ evidence would allow a reasonable juror to find that Defendants issued these statements in good faith and consistent with typical business practices. Accordingly, Defendants meet their burden at Step One to offer plausible and justifiable reasons for their conduct. Plaintiffs’ evidence is insufficient to raise a reasonable inference of conspiracy because it does not tend to exclude the possibility that Defendants acted independently and justifiably with regard to their public statements. Plaintiffs point generally to incorrect news reports, but Plaintiffs neither argue nor offer evidence that Defendants played any in supplying or proliferating the false information contained therein.** Further, Plaintiffs point to only one specific instance where a Defendant made an allegedly misleading public statement: documents establish that Valero said publicly that its machinery “should be at planned rates by mid-August” 2012 but by August 8, internally noted that “we’re looking at late this week/early next week.” See Pltfs. Ex. 56. This evidence may suggest that Valero intentionally misled the public about the duration of its maintenance and concealed the fact that maintenance actually concluded a few days earlier ©. 48 To the extent that Plaintiffs are arguing that Defendants had an affirmative duty to correct news articles they played no role in creating, Plaintiffs have not articulated any legal theory under which Defendants would have such a duty, and the Court declines to locate one for them. 76 15ev1749-JO-AGS and 18cv1374-JO-AGS

publicly announced. See id. Given Valero’s evidence that refinery maintenance estimates are subject to change, the above evidence could also support the inference that Valero publicly announced a good faith estimate of its maintenance, but that maintenance simply took less time than originally estimated. See P. Brooks Declaration J 102. Given that the evidence submitted by Plaintiffs supports both inferences, it does not tend to exclude the possibility that Valero made a good-faith public announcement rather than intentionally deceived the public pursuant to a conspiracy. Accordingly, this plus factor does not support a reasonable inference of conspiracy. 11.The Evidence as a Whole Having addressed each of the categories of circumstantial evidence offered by ) Plaintiffs in support of their conspiracy claim, the Court now considers the crucial question whether the record, viewed as a whole, supports an inference of conspiracy. Citric Acid, 191 F.3d at 1106 (courts must consider “whether the evidence considered as a whole can support an inference of conspiracy”) (emphasis in original). In doing so, the Court views the evidence in the light most favorable to Plaintiffs. Matsushita, 475 U.S. at 587-88. If the evidence produced by Plaintiffs does not “reasonably tend[ ] to prove that [Defendants] had a conscious commitment to a common scheme designed to achieve an unlawful objective,” and does not “tend[ ] to exclude the possibility that [Defendants] were acting independently,” then summary judgment must be entered against them. Monsanto, 465 U.S. at 764 (internal quotation marks and citation omitted). Before examining the inferences permitted by the evidence, the Court starts with a summary of the circumstantial evidence that Plaintiffs have marshaled to prove the wide ranging, eight-Defendant conspiracy that they allege. With respect to Plaintiffs’ arguments that Defendants manipulated public facing gas prices, Plaintiffs were not able to create a 77 15cv1749-JO-AGS and 18cv1374-JO-AGS

triable issue of fact that Defendants selectively reported trades or conducted wash trades, they point to only one purportedly false public statement by Valero which misstated the duration of maintenance. The Court is therefore largely left to consider Plaintiffs’ arguments that Defendants acted to “strangle” supply in the following ways: (1) produced than the maximum amount of CARBOB;* (2) maintained low inventories; (3) rerouted approximately four imports between eight Defendants over seven plus years; (4) exported gas out of the state of California. See, e.g., Pltfs. Exs. 66-67, □□□□□□ 89, 200. The Court will consider all of the foregoing actions against the backdrop of (1) communications showing that traders desired low-supply/high-price conditions; (2) trader communications sharing refinery information while buying and selling gas from }one another; (3) collaboration among Defendants to cover each other’s supply needs, as exemplified by exchange agreements and the refiners’ reactions to the Exxon explosion; and (4) the undisputed fact that Defendants joined an agreement to limit West Coast trading hours. In considering the entirety of Plaintiffs’ evidence, the Court finds that it does not support a reasonable inference of the eight-Defendant conspiracy alleged by Plaintiffs. Absent from Plaintiffs’ evidence that Defendants acted against self-interest to “strangle” supply is any proof that Defendants’ actions were not profit maximizing or that they lowered supply despite market demand. In fact, the evidence suggests that Defendants increased supply in California as demand rose. See, e.g., K. Archambault Declaration ©—ti<‘i‘~™SCS

49 As noted at supra Section V.D.3, Plaintiffs’ arguments in this regard are based on the argument that Defendants had the capacity to, and therefore should have, produced more CARBOB than they did during the class period. These arguments are unsupported by any evidence suggesting that it would have 3g, || been profitable for Defendants to do so. 15cv1749-JO-AGS and 18cv1374-JO-AGS

38-40; H. Henderlite Declaration J] 29-32; S. Roveda Declaration J] 10—12;>° see also McCullough Report 4 36 (“[o]ver the past twenty years, California refinery out-put has always been greater than California gasoline demand.”). In the absence of any evidence to support Plaintiffs’ theory that Defendants’ supply decisions were not profit maximizing, the undisputed facts that Defendants had the capabilities to produce more CARBOB, maintained low inventories, rerouted a handful of imports, and exported gas out of California, does not tend to exclude the possibility that Defendants made those decisions independently rather than pursuant to an agreement. Citric Acid, 191 F.3d at 1094. The trader communications expressing their preference for low-supply/high-price conditions and the backdrop of general collaboration among Defendants does not change this picture to one of conspiracy. Players in concentrated markets are often highly ;conscious of their “shared economic interests” in making “output decisions” and act accordingly to maximize profits. Brooke Grp., 509 U.S. at 227. While it is illegal to act pursuant to a conspiracy, it is not illegal to act pursuant to oligopolistic self-interest, even when those actions drive supracompetitive prices. Jd. (“[C]onscious parallelism, describes the process, not in itself unlawful, by which firms in a concentrated market might in effect share monopoly power, setting their prices at a profit-maximizing, supracompetitive level.”). The trader communications Plaintiffs highlight evidence a shared preference for low-supply/high-price conditions and a mutual desire to act accordingly. See supra Sections V.D.2-3. Aside from providing a motive (which courts have held is, alone, 23], 5° See also N. Weinberg-Lynn Declaration §§ 32-33; M. Bodziak Declaration § 21; P. Brooks Declaration 11 26, 46; J.A. 733-38, Valero Exs. 12. 15cv1749-JO-AGS and 18cv1374-JO-AGS

insufficient), these emails illustrate the reality that Defendants engaged in conscious parallelism while operating in a highly concentrated market. Plaintiffs’ other evidence similarly portrays Defendants as interdependent and highly responsive to what their competitors are doing: Defendants admit to being conscious of one another’s pricing and factoring that information into their respective pricing decisions.*! In the spot market, their traders gather as much information as possible about their counterparts refinery conditions to explore opportunities for their own future trading.°? And while Plaintiffs have shown Defendants cooperate with each other in certain ways, the evidence supports the inference that self-interest and interdependence drove this purported collaboration. Defendants entered into mutually beneficial exchange agreements to trade barrels so they could maintain low inventory levels and cover shortages without having to pay high market rates. See, e.g., Pltfs. Exs. 114-28. By selling barrels to Exxon after their 2015 refinery explosion, it could be that Defendants sought to benefit themselves the least troublesome and most profitable way, given the long-term contracts that Exxon with retailers. The above evidence may support an inference that the six Defendants, aside from and Phillips 66, engaged in conscious parallelism because each independently recognized it benefitted from low-supply conditions and each sought the maximum price advantage. While such oligopolistic maneuvering may be an undesirable market for See, e.g., J. Hodgson Declaration ff 9-15, 19-20, 23, 27-28; P. Brooks Declaration { 96; K. Archambault Declaration § 59; J. Harris Declaration J 6; J.A. 115, Shell Ex. 1; M. O’Neal Declaration % See, e.g., Pitfs. Exs. 6, 8-12, 19-20, 35-44, 46-50, 52, 192-93. 53 See, e.g., Joint Statement § 22; K. Archambault Declaration {J 63, 66-70; C. Yates Declaration 9g || 1950-2. 20 15ev1749-JO-AGS and 18cv1374-JO-AGS

consumers, such a market is not prohibited by antitrust law. Because none of the above to exclude the possibility that each Defendant took independent actions to maximize profits, Plaintiffs have not created a reasonable inference of the wide ranging, eight- Defendant conspiracy that they allege. With regard to BP and Phillips 66, the evidence supports a reasonable inference of conspiracy. In addition to the above, the traders for these Defendants exchanged supply information that they admitted they considered confidential and reported information to the departments responsible for setting production levels. J. Yomtoob Declaration □ 25 (information “may sometimes be transmitted to BP’s supply and refinery planning teams determine whether there is an opportunity to supply counterparties”); L. Lockhart Declaration § 47 (“on occasion, I discussed with our California refineries whether they increase production of CARBOB” based on trader information) (emphasis omitted). This evidence permits an inference that information was exchanged pursuant to an agreement. Unlike other traders who exchanged information in the course of buying and {selling to further their own future trading opportunities, BP’s and Phillips 66’s traders admitted to using information outside of trading to influence refinery decisions. While evidence of low-level traders exchanging information to benefit their buying and selling be consistent with independent decision making in a concentrated market, evidence that traders shared confidential information up the chain-of-command does tend to exclude the possibility of independent conduct. Accordingly, a reasonable juror could find that BP ; and Phillips 66 agreed to share information pursuant to a conspiracy. Because causal antitrust injury is an essential element of Plaintiffs’ antitrust conspiracy claim, the Court turns to whether a genuine dispute of material fact exists with 81 15cv1749-JO-AGS and 18cv1374-JO-AGS

respect to causation. In their opposition, Plaintiffs point to the opinions of their experts, Dr. Williams, Dr. Hanouna, and Mr. McCullough, as evidence that the conspiracy described above caused gas prices in California to be higher than they would have been absent the anticompetitive conduct. See Opposition at 53-56. Defendants seek to exclude Dr. Williams’ and Dr. Hanouna’s expert opinions on causation as inadmissible under Daubert and Federal Rule of Evidence 702. See Dkt. 616. Defendants also argue that neither Dr. Hanouna nor Mr. McCullough offer opinions on causation. See id.; see also Dkt. 613. In the next section, therefore, the Court first examines the admissibility of Williams’ and Dr. Hanouna’s expert opinions with respect to causation. The Court then turns to whether Mr. McCullough’s expert opinions or any other facts in the record create a genuine issue of material fact on the element of causation. A. Daubert Challenges to Dr. Williams and Dr. Hanouna on Causation Federal Rule of Evidence 702 provides that expert opinion evidence is admissible if “(a) the expert’s scientific, technical, or other specialized knowledge will help the trier of fact to understand the evidence or to determine a fact in issue; (b) the testimony is based sufficient facts or data; (c) the testimony is the product of reliable principles and methods; and (d) the expert has reliably applied the principles and methods to the facts of case.” Fed. R. Evid. 702. Plaintiffs bear the burden of establishing that the admissibility requirements are met by a preponderance of the evidence. Fed. R. Evid. 104(a); Advisory Committee Notes to Rule 702, 2000 Amendments (quoting Jn re Paoli R.R. Yard PCB Litig., 35 F.3d 717, 744 (3d Cir. 1994), and explaining that proponents “only have to demonstrate by a preponderance of evidence that their opinions are reliable’); 82 15cv1749-JO-AGS and 18cv1374-JO-AGS

also Lust v. Merrell Dow Pharm., Inc., 89 F.3d 594, 598 (9th Cir. 1996) (“It is the proponent of the expert who has the burden of proving admissibility.”). In Daubert v. Merrell Dow Pharmaceuticals, Inc., the Supreme Court held that Rule 702 requires the district court to act as a gatekeeper to “ensure that any and all scientific testimony or evidence admitted is not only relevant, but reliable.” 509 U.S. 579, 589 (1993). In its role as a gatekeeper, the trial court must ensure that an expert report is reliable “relevant to the task at hand” before it reaches a jury. Primiano v. Cook, 598 F.3d 558, (9th Cir. 2010), as amended (Apr. 27, 2010) (quoting Daubert, 509 U.S. at 597). “Expert opinion testimony is relevant if the knowledge underlying it has a valid connection to the pertinent inquiry. And it is reliable ifthe knowledge underlying it has a reliable basis in the knowledge and experience of the relevant discipline.” Jd. at 565 (citation and internal quotation marks omitted). Ultimately, “[t]he objective” of the Court’s gatekeeping role “is make certain that an expert... employs in the courtroom the same level of intellectual that characterizes the practice of an expert in the relevant field.” Kumho Tire Co. v. Carmichael, 526 U.S. 137, 152 (1999). Thus, an expert’s scientific conclusions must be supported by “good grounds for each step in the analysis,” meaning that “any step that renders the analysis unreliable under the Daubert factors renders the expert’s testimony inadmissible.” Paoli R.R. Yard, 35 F.3d at 745 (emphasis omitted); accord Hardeman v. Monsanto Co., 997 F.3d 941, 962 (9th Cir. 2021) (“expert evidence is inadmissible where the analysis is the result of a faulty methodology or theory”) (citation omitted). When evaluating the admissibility of an expert’s testimony, the Court focuses on the soundness of the expert’s methodology, not the correctness of the expert’s conclusions. Daubert v. Merrell Dow Pharms., Inc., 43 F.3d 1311, 1318 (9th Cir. 1995). Thus, “[s]haky but admissible evidence is to be attacked by cross examination, contrary evidence, and 83 15cv1749-JO-AGS and 18ev1374-JO-AGS

attention to the burden of proof, not exclusion.” Primiano, 598 F.3d at 564 (citation ;omitted). The judge is “supposed to screen the jury from unreliable nonsense opinions, but exclude opinions merely because they are impeachable.” Alaska Rent-A-Car, Inc. v. Avis Budget Grp., Inc., 738 F.3d 960, 969 (9th Cir. 2013). 1. Dr. Williams Dr. Williams, Consumer Plaintiffs’ causation and damages expert, used a forecasting regression methodology to determine the price overcharges resulting from the alleged conspiracy—i.e., the increase in gas prices attributable to the alleged conspiracy as opposed to other market factors. See Opposition at 56; see also Dkt. 608, Ex. A (“Williams Report”). Based on his analysis, Dr. Williams opined “that Defendants’ alleged conduct caused actual wholesale prices to substantially exceed but-for wholesale prices,” with total damages to consumers equaling $15 billion. Williams Report 63, 70. For the reasons set forth below, the Court agrees with Defendants’ arguments that Dr. Williams’ expert opinion on causation is inadmissible under Federal Rule of Evidence 702. While a forecasting regression analysis is a generally accepted econometric approach to determining causation and damages in the antitrust context, see, e.g., United States ex rel. Brown v. Celgene Corp., 226 F. Supp. 3d 1032, 1042 n.11 (C.D. Cal. 2016), the Court must examine whether Dr. Williams’ specific application of that methodology in case meets Rule 702 and Daubert standards for reliability and relevance. Forecasting regression methodology requires an expert to first identify a benchmark period free of anticompetitive conduct and a damages period when the allegedly anticompetitive conduct occurred. See Halbert White, Robert Marshall & Pauline Kennedy, The Measurement of Economic Damages in Antitrust Civil Litigation, 6 Econ. Comm. Newsl. 17, 17-22 (2006); Williams Report {J 44-48, 50. The expert then calculates the relationship between various 84 15cv1749-JO-AGS and 18cv1374-JO-AGS

factors impacting prices (e.g., the cost of raw materials) and the resulting prices during the benchmark, to establish a baseline for how those factors influenced prices during a “clean” period without anticompetitive behavior. See White, Measurement, supra, at 18-20; Williams Report 4 52. Using the data from the benchmark period, the expert then predicts what prices would have been in the damages period without the defendants’ allegedly anticompetitive conduct. See White, Measurement, supra, at 20-21; Williams Report 45-46, 56. The difference between the predicted prices in the “no conspiracy” world the actual prices in the damages period constitutes the class damages. See Williams Report 944. By isolating and quantifying the effects of the allegedly anticompetitive conduct in this way, a valid regression model can also tend to show that the price differential is attributable to the alleged anticompetitive conduct and serve as evidence of causation. See generally White, Measurement, supra. The fundamental premise of a forecasting regression approach is that prices in the “clean” period were influenced only by lawful economic variables (raw materials, production costs, efc.), and prices during the conspiracy were influenced by the same lawful economic variables plus the conspiracy’s anticompetitive conduct. Once the ;}economist understands how conspiracy-free prices are created by lawful economic variables, he can reasonably assume that higher prices in the conspiracy period were caused solely by the conspiracy’s anticompetitive conduct. Thus, a sound econometric approach to selecting these periods is essential to the validity of the entire analysis. An expert selecting a benchmark period must have a reasonable basis to assume that the period is free of anticompetitive conduct. White, Measurement, supra, at 18 (“The [non-collusive] benchmark may be a period of time, like the pre-plea period [o]r . . . where it is reasonable believe that collusion was absent”); Justin McCrary & Daniel L. Rubinfeld, Measuring 85 15cv1749-JO-AGS and 18cv1374-JO-AGS

Benchmark Damages in Antitrust Litigation, 3 J. of Econ. Methods, 63-74 (2014) (noting that anticompetitive conduct in the benchmark period risks producing “a biased damages estimate, which is inappropriate”); David L. Faigman et al., Modern Scientific Evidence: Law and Science of Expert Testimony, at § 43:35 (Dec 2021 Update) (noting that {regression analyses should not be used “if the allegedly unlawful behavior has always been present.”). Dr. Williams himself appears to agree with the foregoing, as he relied on some of these same sources in his opening report. Williams Report § 45 n.37 (citing White, Measurement, supra and McCrary, Measuring Benchmark Damages, supra). Moreover, Williams specifically criticized Defendants’ regression analysis on this basis, arguing that the benchmark period “should be free from the effects of the alleged conspiracy,” in part, because “if one measures cartel prices against cartel prices, those prices will no longer appear to be the result of the cartel.” Dkt. 608, Ex. G (“Williams Rebuttal Report”) □ 58. Here, Dr. Williams’ methodology disregarded the basic econometric principle of selecting a benchmark period that is reasonably free of misconduct so that it can serve its fundamental purpose—to act as a baseline such that conclusions can be drawn about how anticompetitive conduct impacted prices in the damages period. At the outset of his regression analysis, Dr. Williams selected the “clean” benchmark period as January 2005 to January 2015, but provided no justification for this selection. Williams Report { 50; Dkt. 608, Ex. C (“Williams Dep.”) at 224:3—-4 (“I was not instructed by counsel as to the definition of the benchmark period.”). Dr. Williams then defined the damages period where price fixing occurred as February 2015 to December 2019, “based on instruction of counsel.” Williams Report J 50. In contrast, Consumer Plaintiffs’ complaint originally alleged that the price fixing conspiracy first occurred in 2012, not February 2015, and Persian Gulf’s complaint similarly alleged that the Defendants’ price fixing conspiracy 86 15cv1749-JO-AGS and 18cv1374-JO-AGS

began in 2012. See Bartlett, Dkt. 44 at 46; Dkt. 76 at 64. Based on Plaintiffs’ initial jallegations, Dr. Williams’ benchmark period contained more than three years of anticompetitive misconduct, when it is supposed to be “clean.” Such incongruity between the pleadings and the expert’s analysis would not pose a methodological problem if Dr. Williams chose a benchmark period based on post- discovery evidence instead. As the In re Packaged Seafood Prod. Antitrust Litig. court noted, an expert’s benchmark period selection based on post-discovery evidence, rather pleading allegations, only strengthens the relevance of an expert’s analysis. 332 F.R.D. 308, 326 (S.D. Cal. 2019) (refining of conspiracy period based on the expert’s “analysis of the record evidence in the case . . . bolster[ed] the reliability of the [regression] model”). Here, however, Dr. Williams’ benchmark and damages periods are also inconsistent with Plaintiffs’ contentions after years of gathering evidence in discovery. At close of fact discovery, Plaintiffs asserted in binding interrogatory responses that the conspiracy—effected by information exchanges, supply restrictions, and market manipulation discussed supra—began no later than August 1, 2011. Dkt. 455, Ex. C at 12; Dkt. 455, Ex. D at 13. Therefore, based on Plaintiffs’ post-discovery theory of the case, Williams’ benchmark period contained more than four years of anticompetitive misconduct, when it is supposed to be “clean” of misconduct. Dr. Williams’ identification of the period from 2005 to 2015 as “clean” is also inconsistent with the evidence that Plaintiffs present to the Court on summary judgment. their opposition to Defendants’ motions, Plaintiffs ask the Court to consider a conspiracy /implemented by acts beginning in at least 2011 that caused inflated prices starting in “at 2012.” See, e.g., Opposition at 1, 39. For example, Plaintiffs identify exchange agreements that have been in place since 2005. See Pltfs. Exs. 114-128. Plaintiffs also 87 15cv1749-JO-AGS and 18cv1374-JO-AGS

point to information exchanges, the focal point of Plaintiffs’ arguments, beginning in 2012. See, e.g., Pltfs. Exs. 43-44 (trader exchanges in 2012); Pltfs. Exs. 19, 46, 65 (trader exchanges in 2013); Pltfs. Exs. 11, 34, 47 (trader exchanges in 2014). Plaintiffs similarly argue that Defendants’ market manipulation and supply restrictions began in 2012. See ||McCullough Report {| 404—05 (describing wash trades in 2012 and 2014); Pltfs. Ex. 56 (sole evidence of false public statements occurred in 2012); Pltfs. Ex. 85 (Chevron diverted cargoes in August 2014); Pltfs. Ex. 100 (Chevron considering exports to Mexico in 2014); J.A. 583-86, Phillips Ex. 20 (“gentleman’s agreement” on trading hours in 2014). Mr. McCullough, Plaintiffs’ liability expert, also pointed to anticompetitive conduct starting in 2011 to establish liability. See generally McCullough Report. Nearly half of the conduct Mr. McCullough identified took place during Dr. Williams’ benchmark period. See id. Thus, based on Plaintiffs’ current evidence of conspiracy and anticompetitive acts starting in 2011, Dr. Williams’ benchmark period, which is supposed to be “clean,” contains more than four years of anticompetitive misconduct. Dr. Williams’ 2005 to 2015 benchmark period, therefore, is not an approximation of a “clean period” under any metric—neither the allegations in Plaintiffs’ complaints, their binding interrogatory responses, nor the evidence they present to the Court on summary judgment supports the selection of 2005-2015 as a period reasonably “clean” of misconduct. Dr. Williams’ methodological flaw in framing his benchmark period renders his /analysis irrelevant to Plaintiffs’ case. To be admissible, an expert report must be “sufficiently tied to the facts of the case that it will aid the jury in resolving a factual dispute.” Daubert, 509 U.S. at 591 (citation omitted); Mission Viejo Florist, Inc. v. Orchard Supply Co., LLC, 2019 WL 13045054, at *2 (C.D. Cal. Feb. 28, 2019) (“Courts reject or exclude expert testimony that fails to connect damages calculations to the theory a 88 15cv1749-JO-AGS and 18cv1374-JO-AGS

of liability”). Dr. Williams’ analysis is based on the assumption that there was no conspiratorial conduct from 2011 to 2015, whereas all of the allegations and evidence before the Court suggests the opposite. Because his opinion on causation flows from this inapposite assumption, it is divorced from the reality of the case that Plaintiffs are actually trying to prove, and thus is not sufficiently tied to Plaintiffs’ liability case to help the factfinder. Dr. Williams provided no methodological justification grounded in econometrics for choosing a benchmark period so significantly misaligned with the case that Plaintiffs are currently trying to prove. Reed Constr. Data Inc. v. McGraw-Hill Cos., Inc., 49 F. Supp. 3d 385, 400 (S.D.N.Y. 2014), aff'd, 638 F. App’x 43 (2d Cir. 2016) (“[T]o be admissible, a regression analysis must examine an appropriate selection of data... some passably scientific analysis must undergird the selection of the frame of reference.”). In fact, in his opening expert report, Dr. Williams provided no explanation for why he selected 2005— as the benchmark period. Williams Report § 50. At his deposition, Dr. Williams stated that that he selected the benchmark period based on his “reading” of the complaint, even though he apparently understood that Plaintiffs “alleg[ed] that there had been anticompetitive conduct in the benchmark period.” Williams Dep. at 219:24—220:5; 222:1-8; 228:14—229:16. In his rebuttal report, Dr. Williams argued that the evidence shown to him supported the selection of 2015 as the start of the damages period, again with no econometric justification of a benchmark period that is not “clean.” Williams Rebuttal Report □□ 40-41. Dr. Williams’ selective reading of the complaint is not a methodological explanation grounded in econometrics that can cure his faulty selection of the benchmark period. Gen. Elec. Co. v. Joiner, 522 U.S. 136, 146 (1997) (“[N]othing in either Daubert or the Federal Rules of Evidence requires a district court to admit opinion evidence that is 89 15ev1749-JO-AGS and 18cv1374-JO-AGS

connected to existing data only by the ipse dixit of the expert”); McGlinchy v. Shell Chem. 845 F.2d 802, 807 (9th Cir. 1988) (upholding exclusion of conclusions in expert report with only “scant basis” in the record). Accordingly, Dr. Williams’ selection of a benchmark period that contains four years of alleged anticompetitive misconduct, based on Plaintiffs’ post-discovery contentions and evidence on summary judgment, renders his analysis fundamentally flawed and methodologically unsound. The Court acknowledges that perfect lines demarcating benchmark periods and damages periods, such that all the misconduct falls neatly into the damages period (rendering the benchmark period perfectly clean), will rarely, if ever, exist; econometric principles do not require such precision to pass scientific muster. While some of inaccuracy is bound to exist, the selection of a benchmark period needs to be grounded in econometric principles and bear some logical relationship to the evidence in the case. Where the selection of the benchmark period is neither grounded in econometric principles nor the record, all of the econometric assumptions that flow from the /identification of a “clean” period are rendered unsound. If the “clean” period contains years of allegedly anticompetitive conduct, the idea that prices in the “clean” benchmark period can be fairly traced to variables impacting price, and not the anticompetitive conduct, falls apart. When the fundamental principles that a methodology is based on are disregarded in this way, it is not “shaky but admissible evidence” with conclusions that should be tested by cross-examination. Rather, the underlying methodology is so flawed the analysis is rendered unreliable and cannot be cured by the adversarial process at Plaintiffs raise various arguments in opposition to Defendants’ motion to exclude Dr. Williams’ report. First, Plaintiffs argue that the error in Dr. Williams’ methodology 90 15ev1749-JO-AGS and 18cv1374-JO-AGS

can be remedied by an assumption that any such error would only make the damages figure more conservative. Dkt. 698 (“Daubert Opposition”) at 19. Dr. Williams similarly surmised at his deposition that a benchmark period containing some misconduct does not invalidate the regression model but merely renders the damages analysis more lconservative. Williams Dep. at 217:6-22. This theory, however, was not tested by Dr. Williams despite his acknowledgment that “[w]hether or not an alternative benchmark period would increase or decrease damages is an empirical question.” Id. at 216:24— 217:5.4 Plaintiffs argue that a few district courts have admitted flawed regression models based on the assumption that a tainted benchmark period would only result in an underestimation of damages, but the cases they point to do not stand for that proposition. Daubert Opposition at 15, 19-20; Dkt. 809 at 18-21; Packaged Seafood, 332 F.R.D. 308; In re Linerboard Antitrust Litig., 497 F. Supp. 2d 666, 675, 683-84 (E.D. Pa. 2007). These cases do not hold that benchmark periods divorced from econometric principles and record evidence can be cured by an assumption that tainted benchmark periods render damages figures more conservative. Rather, these cases emphasize that benchmark periods need to be supported by econometric principles and record evidence to be sound under Daubert. Packaged Seafood, 332 F.R.D. at 326 (selection of time periods based on “analysis of the —C~—tt——T.. KX’ *4 The Court notes that Dr. Janus Ordover, Defendants’ damages expert, submitted rebuttal evidence that when 201 1-2014 is moved out of the benchmark period and into the damages period, damages drop from $15 billion to negative $11.1 billion. Dkt. 605, Ex. C (“Ordover Report”) 4 317; Dkt. 617, J. Ordover Declaration § 7 & Fig. 1. The Court does not credit Dr. Ordover’s opinion or analysis over Dr. Williams’. Rather, Dr. Ordover’s findings merely counsel against overlooking the fundamental unreliability of Dr. Williams’ methodology based on an untested assumption. 91 15cv1749-JO-AGS and 18cv1374-JO-AGS

record evidence in the case”); Linerboard, 497 F. Supp. 2d at 683-84 (expert identified “a more competitive period, presumably free of unlawful conduct”) (citation omitted). Those courts noted that some anticompetitive conduct in the benchmark periods would only make the estimates more conservative, but only after finding that the experts selected reasonable benchmark periods based on econometric principles and the evidence in the case. See id. Here, as discussed above, Dr. Williams did not provide an explanation for his selection of the benchmark period and acknowledged that it contradicts Plaintiffs’ conspiracy case. Thus, there are fundamental unreliability and irrelevance issues here that were not at issue Packaged Seafood and Linerboard. The flaws in Dr. Williams’ analysis cannot be cured by the assumption Plaintiffs urge. Plaintiffs also argue that a conspiracy beginning in 2015 fits the facts of the case because the statute of limitations precluded Plaintiffs from seeking damages prior to June 2014. Daubert Opposition at 11-12. In the Court’s prior order partially granting Defendants’ motion to dismiss, the Court ruled that Plaintiffs could not seek damages prior June 21, 2014. See Bartlett, Dkt. 108 at 3—7. Plaintiffs argue that Dr. Williams’ damages period, therefore, had to “coincide with the Court’s ruling defining the damages period,” and thus is consistent with their case. Daubert Opposition at 12. The fact that Plaintiffs were foreclosed from collecting damages for the years 2011—2014 does not render those years conspiracy free, especially when Plaintiffs seek to prove that Defendants operated a conspiracy during those years. Accordingly, the statute of limitations may impact available damages, but it does not excuse an expert from following econometric principles, and it cannot erase the voluminous evidence Plaintiffs have placed before the Court to establish a conspiracy beginning in 2012, at the latest. a 92 15cv1749-JO-AGS and 18cv1374-JO-AGS

Finally, at oral argument, Plaintiffs contended that even though the alleged conspiracy was in effect starting in 2011 or 2012, the conspiracy did not impact prices until 2015, when the conspiracy was at its most active. See Dkt. 809 at 27-39. While conspiracies theoretically might have delayed price impacts, the evidence before this Court does not support Plaintiffs’ eleventh-hour theory that the conspiracy had a four-year delay in impact. First, Plaintiffs’ delayed-impact theory does not appear in their opposition to summary judgment and is not reflected in the opinions of their experts. Second, even attempting to interpret the record evidence to fit Plaintiffs’ delayed-impact theory, the conspiratorial conduct Plaintiffs identify is largely identical from 2011-2014 and 2015— present. As discussed in more detail at Section V.D supra, Plaintiffs point to particular mechanisms that Defendants used to restrict supply, manipulate public facing gas prices, lexchange information, and protect one another from shortages. Plaintiffs’ evidence demonstrates that these same mechanisms were being utilized in 2011 up through the end of the conspiracy period—in other words, Plaintiffs do not identify some behavioral change the part of Defendants in 2015 that would explain why the exact same market behaviors would not have impacted prices in 2011 but somehow would have impacted prices in 2015. Third, Plaintiffs admit that the price fixing conduct they describe had rapid price impacts, ranging from hours to, at most, months. See Dkt. 809 at 34-39 (admitting that efforts to restrict supply and manipulate public facing gas prices would take hours to months to impact gas prices). Based on the evidence in the record, there is no reasonable basis for a to conclude that price fixing mechanisms beginning in 2011 and occurring across the class period did not impact gas prices until 2015. Accordingly, this argument does not render Dr. Williams’s regression model applicable to the facts of this case and his analysis remains inadmissible. 93 15cv1749-JO-AGS and 18cv1374-JO-AGS

2. Dr. Hanouna Plaintiff Persian Gulf’s expert on damages is Dr. Paul Hanouna, who performed dummy variable and forecasting regression analyses to determine the overcharges to wholesale customers at the rack. See Dkt. 608, Ex. B (“Hanouna Report”). Because Plaintiffs argue in their opposition to summary judgment that his damages analysis also evidences causation, the Court will examine the extent to which Dr. Hanouna offers a causation opinion and whether such an opinion would admissible under Federal Rule of Evidence 702. To begin his analysis, Dr. Hanouna identified the cost of Brent Crude Oil, one type of blended oil that is refined to make gas, as the primary factor impacting gas prices from January 1, 2003 to February 29, 2020. Jd. § 6.a. Based on the relationship between this single variable—the cost of Brent Crude Oil—and gas prices, Dr. Hanouna concluded that there were inflated gas prices in May and October of 2012, as well as the period of January to March 2020, that could not be explained by the price of Brent Crude Oil and thus were caused by some other, unidentified variable. Id. § 6.a—6.b. Based on the above, Dr. Hanouna defined the damages periods as those periods where he saw price spikes in May and October of 2012 and January 2015 to March 2020. Id. §§ 6.b—6.c. He identified the benchmark period as being January 1, 2003 to February 29, 2020, excluding May and October of 2012 and January 2015 to March 2020. Id. § 25. Using the benchmark period, Dr. Hanouna predicted what prices would have been in the damages periods if Brent Crude Oil had continued to predict gas prices. Id. □□ 6.a—6.c, 25. Based on the difference between predicted and actual prices during the damages periods, Dr. Hanouna determined that the total damages to wholesale purchasers of gas a 94 15¢v1749-JO-AGS and 18cv1374-JO-AGS

jjequals $287,542,483 in May of 2012, $410,244,180 in October of 2012, and $9,046,092,774 from January 1, 2015 to March 2020. Id. ¥ 6.c. Although Plaintiffs argue on summary judgment that Dr. Hanouna’s opinions provide evidence of causation,* Dr. Hanouna expressly denied having any such opinion. At his deposition, Dr. Hanouna testified more than once that he did not have any opinion the causes of inflated prices in the damages periods, and further, that the lack of correlation between Brent Crude Oil and gas prices in the damages periods “[a]bsolutely” did not imply collusion. See Dkt. 608, Ex. D (““Hanouna Dep.”) at 48:23-49:1 61:14-23. Dr. Hanouna also stated that he was not an expert in antitrust economics nor the gas industry. Hanouna Dep. at 13:12—14:17; 36:3-8; 40:20—41:22; 95:12-24. Thus, he testified that he was not offering any opinion on Plaintiffs’ theory of anticompetitive conduct but was solely calculating damages assuming Plaintiffs could prove the existence of a conspiracy. Hanouna Dep. at 15:23—16:4; 44:9-20; 47:18-48:3; 61:14—23; Hanouna Report §] 5-6. As Dr. Hanouna disavows having an expert opinion on causation, the Court precludes Plaintiffs from offering his regression analysis for causation purposes. See, e.g., Mitchell v. Geo Grp. Inc., 2022 WL 874287, at *5 (D. Ariz. Mar. 24, 2022) (“when an expert affirmatively repudiates or disavows an opinion expressed in the expert’s report, courts have not hesitated to conclude that the expert should be precluded from offering □□□□ ——_—_—_—_- °° Plaintiffs themselves have made conflicting arguments regarding whether Dr. Hanouna is being offered on the issue of causation. On opposition to summary judgment, Plaintiffs point to Dr. Hanouna’s opinions as evidence of causation. Opposition at 55-56. Then, at oral argument, Plaintiffs stated that his opinion was being offered solely on the issue of damages and was not being offered to show causation. See Dkt. 809 at 32 (stating that the “experts aren’t on a causation analysis.”). 15cv1749-JO-AGS and 18cv1374-JO-AGS

/opinion at trial”); Gallagher v. S. Source Packaging, LLC, 568 F. Supp. 2d 624, 634 (E.D.N.C. 2008) (finding that where an expert denied having an opinion on an issue, he was not qualified under Rule 702 to offer an opinion on it); Devito v. Smithkline Beecham Corp., 2004 WL 3691343, *4 (N.D.N.Y. Nov. 29, 2004) (“if [the expert] has not formed the opinions which plaintiff is ascribing to him, necessarily he has no foundation, scientific otherwise, for same”). Despite Plaintiffs’ arguments that Dr. Hanouna’s regression analysis does provide information about causation, the Court finds that the expert economist himself would know best the limits of his own study and expertise. His report and his testimony clearly set forth those limits: his single-variable analysis provided no information as to what caused the price spikes he identified. See Hanouna Dep. at 15:23— 16:4; 44:9-20; 47:18-48:3; 48:23-49:1; 61:14—23; Hanouna Report {{ 5—6. Moreover, Plaintiffs—who carry the burden to demonstrate the admissibility of Dr. Hanouna’s opinions—have not pointed to any evidence that Dr. Hanouna has the “scientific, technical, or other specialized knowledge” to offer an opinion on causation. See Fed. R. Civ. P. 702. Therefore, the Court finds that Plaintiffs have not met their burden of establishing that Dr. Hanouna is qualified to do so under Rule 702. Finally, Dr. Hanouna’s analysis is further inadmissible to prove what caused inflated prices because his regression model suffers from the same flaws as Dr. Williams’ regression analysis. Dr. Hanouna’s regression analysis also utilizes a benchmark period Plaintiffs allege contains years of anticompetitive conduct. See supra Section VI.A.1; Hanouna Report Jf 6.a—6.c, 25. Like Dr. Williams, he includes most of 2011—2015 (except for two months) in the benchmark period, despite the fact that Plaintiffs seek to prove that Defendants conspired during that period. See id. Accordingly, for the same reasons 96 15cv1749-JO-AGS and 18cv1374-JO-AGS

Dr. Williams’ analysis is inadmissible with respect to causation, the Court finds that Hanouna’s analysis is inadmissible with respect to causation. B. Whether the Remaining Evidence Creates a Genuine Dispute of Material Fact on Causation Because the Court has excluded Dr. Williams’ and Dr. Hanouna’s expert opinions on causation,” the Court turns to whether Plaintiffs have otherwise provided evidence which creates a genuine dispute of material fact with respect to causation. Plaintiffs point to the expert opinion of Mr. McCullough as evidence of causation and further argue that causation can inferred from the facts in the record regarding the conspiracy. Causal antitrust injury is “an essential element of any remedy under the Sherman Act.” Catlin v. Wash. Energy Co., 791 F.2d 1343, 1347 (9th Cir. 1986) (citation omitted). While Plaintiffs need not rule out “all possible alternative sources of injury,” they must show that the alleged anticompetitive conduct was “a material cause” of the injury. Dolphin Tours, Inc. v. Pacifico Creative Serv., Inc., 773 F.2d 1506, 1509 (9th Cir. 1985) (quoting Zenith Radio Corp. v. Hazeltine Research, Inc., 395 U.S. 100, 114 & n.9 (1969)). The trier of fact must be able to ascertain causal antitrust injury “without engaging in speculation.” Jd. at 1510. Where a party fails “to proffer any evidence as to causation, they fail to raise a genuine issue of material fact, and . . . summary judgment against them be granted.” Jn re Optical Disk Drive Antitrust Litig., 2017 WL 6451711, at *3 (N.D. Dec. 18, 2017). // °° | > Given the Court’s ruling, it does not consider whether Dr. Williams’ and Dr. Hanouna’s expert opinions are admissible on the issue of damages. 99 15cev1749-JO-AGS and 18cv1374-JO-AGS

1. Mr. McCullough’s Report Does Not Create a Genuine Dispute on Causation In order to evaluate Plaintiffs’ argument, the Court first examines the expert report Plaintiffs’ liability expert, Robert McCullough. Mr. McCullough described his assignment as offering opinions “on the structure of the California gasoline market, whether it was susceptible to market manipulation, and whether and how market manipulation did occur.” McCullough Report § 2. Mr. McCullough’s report is largely descriptive: he performed a review of documents produced in discovery and concluded, among other things, that the California gas market is geographically isolated, highly concentrated, and inelastic, which makes it “susceptible to collusive activity.” See, e.g., 18-20. Mr. McCullough also reviewed Plaintiffs’ alleged plus factors and agreed those plus factors, such as information exchanges, exchange agreements, and exports, were highly unusual and suggestive of market manipulation. See id. {§ 22-31. He did not, however, opine on the existence of a conspiracy nor whether Defendants’ anticompetitive conduct identified in his report had any particular impact on prices. Plaintiffs argue that a genuine dispute of material fact on causation is created by Mr. McCullough’s general conclusion that the record evidence suggests collusive activity }among Defendants, but this alone does not provide evidence of causation. See Opposition 55. Mr. McCullough’s report, like Dr. Hanouna’s, does not offer any opinions on causation. See generally McCullough Report. While Mr. McCullough repeatedly stated /throughout his report and in his deposition that the gasoline market was susceptible to ““market manipulation,” he defined that concept more broadly than illegal conspiracy under antitrust laws and testified that he did not have an opinion about whether illegal price fixing occurred. Defs. Ex. 8, McCullough Dep. at 23:12—24:13; 31:13-—23; 79:19-80:12; 92:4-21. a 98 15ev1749-JO-AGS and 18cv1374-JO-AGS

Further, Mr. McCullough explicitly stated that he did not have an opinion on whether gasoline prices increased because of Defendants’ conduct. Jd. at 44:7—13; 83:17-84:7; 84:23-85:18; 196:14-21. This was, apparently, not for a lack of attempting to identify /causal evidence. Mr. McCullough testified that although his assignment included analyzing whether Defendants’ exchanges of purportedly confidential information caused higher gasoline prices in California, he “did not easily find an immediate impact. Much of this information, though valuable and confidential, did not lead to a specific trading decision.” Jd. at 83:17-84:7. Mr. McCullough’s opinions that the market was “susceptible” to manipulation, that Defendants and their traders “could” have manipulated prices, or that Defendants’ conduct is “suspicious” and “unusual” would not permit a reasonable juror to infer causation. See McCullough Report. Thus, his opinions do not create a genuine dispute of material fact as to causation.*” 2. The Remaining Record Evidence Does Not Create a Genuine Dispute on Causation Given that Mr. McCullough’s expert report does not go to causation, the Court examines whether the facts in the record regarding conspiracy support an inference of causation. Because the Court has concluded above that the evidence does not support a reasonable inference of the eight-Defendant conspiracy alleged by Plaintiffs, it will consider only whether facts in the record create a triable issue of fact that a conspiracy between BP and Phillips 66 caused Plaintiffs to suffer antitrust injury. °° °° °° © 57 Given the Court’s rulings in this order, it does not consider whether Mr. McCullough’s expert opinion is admissible on the issue of liability. 90 15cv1749-JO-AGS and 18cv1374-JO-AGS

Plaintiffs’ evidence must allow an inference, “without engaging in speculation,” that alleged anticompetitive conduct was “a material cause” of the injury. Dolphin Tours, 773 F.2d at 1509-10; 2 Areeda & Hovenkamp, Antitrust Law § 338a, at 317 (2d ed. 2000) (plaintiff must show that “the antitrust violation contribute[d] significantly to [its] injury”) (interpreting Zenith, 395 U.S. 100). While courts have chosen to infer causation where a conspiracy had in fact occurred or where evidence made a causal link plausible, those are not the facts here. Cf Packaged Seafood., 332 F.R.D. at 322 (noting defendants had already entered criminal pleas admitting to the price fixing conduct). The conspiracy evidence with respect to BP and Phillips 66 would not allow a reasonable juror to infer that an agreement between them caused higher gas prices. See supra Section V.D. There is no evidence that BP or Phillips 66 entered wash trades, selectively reported trades, or issued false public statements. See id. The evidence with respect to these two Defendants consists almost entirely of information exchanges between their traders. See, e.g., Pltfs. Exs. 8, 48, 74, 83-84. Nevertheless, Plaintiffs have not pointed to evidence that BP’s or Phillips 66’s traders executed any trade based on the information the traders exchanged. Defs. Ex. 8, McCullough Dep. at 83:17—84:7 (“[m]uch this information, though valuable and confidential, did not lead to a specific trading decision.”).~° || >8 Were a jury to attempt to infer causation from the trader exchanges between BP and Phillips 66, Jury p they would necessarily have to devolve into speculation. Plaintiffs would have a jury infer that these traders did in fact enter trades based on the information exchanged and that those trades were then reported to OPIS absent any evidence of same. This chain of inferences is especially tenuous given that OPIS reports an average price taken from the high and low trades in its survey. See Opposition at 47; J.A. 562— 76, Phillips Ex. 17 (OPIS pricing methodology). Accordingly, a juror would also have to infer that the trades executed and reported were the high or low trade at the time OPIS averaged the price information. 100 15cv1749-JO-AGS and 18cv1374-JO-AGS

l Moreover, the causal link between these traders’ exchanges and any resulting supply restriction is absent from the record. The only evidence in the record suggests that BP and Phillips 66 used the information exchanged to increase supply. J. Yomtoob Declaration 25 (information used to increase production); L. Lockhart Declaration {47 (same). Neither do Plaintiffs submit evidence that BP’s and Phillips 66’s exchanges occurred in proximity to a supply restriction or reduction. Accordingly, a reasonable juror could not find causation based on the mere possibility that BP’s Phillips 66’s information exchanges could have raised spot-market prices or reduced supply in the absence of any evidence supporting that possibility. Even assuming that a reasonable juror could infer that the foregoing could have caused higher gas prices, there is no evidence that would allow a jury to determine whether BP’s and Phillips 66’s conduct was a material cause of Plaintiffs’ /injury. Therefore, in the absence of causal evidence, there is no genuine dispute of material fact on causation and Defendants’ motions for summary judgment should be granted. Vil. STATE LAW CLAIMS Because the Court has found that Defendants are entitled to summary judgment on Plaintiffs’ Sherman Act conspiracy claim, they are also entitled to summary judgment on Plaintiffs’ remaining state law claims under California’s Cartwright Act and California’s UCL based on the same allegations and evidence relevant to their Sherman Act claim. “The analysis under California’s [Cartwright Act] mirrors the analysis under federal law because the Cartwright Act...was modeled after the Sherman Act.” Cnty. of Tuolumne v. Sonora Cmty. Hosp., 236 F.3d 1148, 1160 (9th Cir. 2001). Plaintiffs’ UCL } There is no evidence in the record of any of the foregoing, and a juror would have no evidence from which to determine that these exchanges had any impact on spot-market prices. 101 15cv1749-JO-AGS and 18cv1374-JO-AGS

claim is also derivative of their antitrust claims. See Dkt. 86 at 20; Clear Connection Corp. Comcast Cable Commc'ns Mgmt., LLC, 2020 WL 6742889, at *8 (E.D. Cal. Nov. 17, 2020) (“The UCL’s ‘unlawful’ prong ‘borrows violations of other laws and treats them as unlawful practices that the [UCL] makes independently actionable.”) (quoting Cel-Tech Comme’ns, Inc. v. Los Angeles Cellular Tel. Co., 20 Cal. 4th 163, 180 (1999)); Chavez v. Whirlpool Corp., 93 Cal. App. 4th 363, 375 (2001) (“If the same conduct is alleged to be an antitrust violation and an ‘unfair’ business act or practice for the same reason...the determination that the conduct is not an unreasonable restraint of trade necessarily implies that the conduct is not ‘unfair’ toward consumers.”); see also Troyk v. Farmers Grp., Inc., 171 Cal. App. 4th 1305, 1350 (2009) (causation is an element of standing to bring a UCL claim). Because the Cartwright Act mirrors the Sherman Act, the claims rise and fall together. In addition, Plaintiffs’ UCL claim is based entirely on the existence of a conspiracy that caused consumer losses, and Plaintiffs do not identify any other conduct as a basis for that claim. See Dkt. 76; Bartlett, Dkt. 44. Thus, the UCL claim also rises and falls with Plaintiffs’ Sherman Act claim. Because the Court grants summary judgment for Defendants on Plaintiffs’ Sherman Act claim, it also grants summary judgment on Plaintiffs’ remaining state law claims. Vill. CONCLUSION AND ORDER For the reasons set out above, the Court GRANTS Defendants’ motions for summary judgment [Dkts. 615, 619, 625 in Persian Gulf, 15cv1749-JO-AGS, and Dkts. 457, 461, 470 in Bartlett, 18cv1374-JO-AGS]. The Court also GRANTS IN PART Defendants’ motion to exclude the expert testimony of Dr. Michael Williams and Dr. Paul Hanouna (Dkt. 616 in Persian Gulf, 15cv1749-JO-AGS, and Dkt. 458 in Bartlett, 18cv1374-JO- AGS] on the issue of causation. The remaining motions to exclude expert testimony [Dkts. 102 15ev1749-JO-AGS and 18cv1374-JO-AGS

622, 626 in Persian Gulf, 15cv1749-JO-AGS, and Dkts. 455, 464, 467 in Bartlett, 18cv1374-JO-AGS] are DISMISSED as moot. The Clerk of the Court is instructed to close Persian Gulf, 15cv1749-JO-AGS, Bartlett, 18cv1374-JO-AGS, and Rinaldi, 18cv1377- JO-AGS. C Dated: September 30, 2022 g on. Jinsook Ohta United States District Judge 103 15cv1749-JO-AGS and 18cv1374-JO-AGS