In re Commodity Exch., Inc., Gold Futures & Options Trading Litig.

328 F. Supp. 3d 217

Opinion

VALERIE CAPRONI, United States District Judge:

Plaintiffs in these consolidated cases allege a conspiracy to fix the price of physical gold and gold-denominated financial instruments from 2004 to 2012. Until November 2014, the price of physical gold was set twice daily through a private auction involving some of the largest bullion banks in London. Plaintiffs allege that the afternoon "Gold Fixing"-also known as the "PM Fixing"-was a cover for a price-fixing conspiracy among the entity charged with operating the Gold Fixing, defendant London Gold Market Fixing Ltd. ("LGMF"), and the participant banks: The Bank of Nova Scotia ("BNS"), Barclays Bank plc ("Barclays"), Deutsche Bank AG ("DB"), HSBC Bank plc ("HSBC"), and Société Générale SA ("SocGen") (collectively, the "Fixing Banks").¹ Plaintiffs have also named as a defendant UBS AG and its affiliates (together "UBS"). Although UBS was not a member of the Gold Fixing at any point during the alleged class period, Plaintiffs contend UBS conspired with the Fixing Banks to suppress the price of gold as determined by the PM Fixing.

Plaintiffs are individuals and entities that sold physical gold, gold futures traded on the Commodity Exchange, Inc. ("COMEX") market, shares in gold exchange-traded funds ("ETFs"),² or options on gold ETFs during the Class Period. Seeking to recover alleged losses suffered as a result of Defendants' alleged manipulation and *220suppression of the price of gold through the gold "fixing" process, Plaintiffs bring putative class action claims for (1) unlawful restraint of trade in violation of Section 1 of the Sherman Act, 15 U.S.C. § 1 et seq. ; (2) market manipulation in violation of the Commodity Exchange Act ("CEA"), 7 U.S.C. § 1 et seq. and CFTC Rule 180.2; (3) employment of a manipulative or deceptive device and false reporting in violation of the CEA, 7 U.S.C. § 1 et seq. and CFTC Rule 180.1; (4) principal-agent liability under the CEA, 7 U.S.C. § 1 et seq. ; (5) aiding and abetting manipulation in violation of the CEA, 7 U.S.C. § 1 et seq. ; and (6) unjust enrichment.

Before the Court is UBS's motion to dismiss the Third Amended Complaint (Dkt. 266) (the "TAC"). In brief, UBS contends that its participation in a scheme to suppress the PM Fixing is implausible. According to UBS, the Fixing Banks, with their ready-made forum for collusion and substantial market power, had no reason to involve UBS in their alleged conspiracy. UBS also moves to dismiss for lack of personal jurisdiction over a Swiss entity, UBS AG. For the reasons that follow, the Court agrees that Plaintiffs' allegations against UBS are implausible. The motion to dismiss is granted.

BACKGROUND

The Court assumes familiarity with the prior proceedings in this case and the Court's prior opinion in In re Commodity Exch., Inc., Gold Futures and Options Trading Litig. , 213 F.Supp.3d 631 (S.D.N.Y. 2016) (" Gold I ") and the Court's memorandum order granting leave to amend, Dkt. 258 ("Gold II "). In recent years, the Gold Fix auction (which originated in 1919) has been conducted during a conference call among representatives of the five Fixing Banks. TAC ¶¶ 87-92. No third parties participated in the call, making it an almost perfect forum for collusion among competitors. The market-clearing price in the auction (the "Fix Price") was published as a benchmark price for physical gold. TAC ¶¶ 83-84. Because of a near-perfect correlation between prices for physical gold and gold derivatives, Plaintiffs allege that the Fix Price also effectively set the market price for gold futures, options, and forwards. TAC ¶¶ 112-22.

In support of their claims, Plaintiffs allege that there has been a persistent downward bias in gold prices immediately before and during the PM Fixing call. TAC ¶¶ 9, 27. From 2004 through 2012 (Plaintiffs' proposed class period), the spot price of gold decreased during the PM Fixing on between 60% and 80% of trading days, TAC ¶¶ 27, 123-25, even though, it is alleged, an efficient market would be equally likely to move upwards or downwards on a given day. TAC ¶¶ 128-29. The PM Fixing also registered as one of the most volatile periods in the trading day, TAC ¶¶ 145, 156-60, a phenomenon not seen around the morning or "AM" fixing, TAC ¶ 153. Plaintiffs' analysis further shows that the downward movement in the price of physical gold consistently began in the minutes before the PM Fixing call began, evidence, according to Plaintiffs, of coordinated trading based on foreknowledge of the Fix Price that would emerge from the auction. TAC ¶¶ 125, 145.

Plaintiffs also identified alleged patterns in the Defendants' price quotes, which, Plaintiffs contend, link them to anomalous pricing behavior observed around the PM Fixing. According to Plaintiffs, Defendants consistently quoted similar (or "bunched"), below market prices around the PM Fixing, a trend that was more pronounced on so-called "down days," i.e. , on days when the Fix Price was lower than the spot price of gold before the Fix Price was announced. TAC ¶¶ 209, 257-60, 262-65. On certain days identified by Plaintiffs, Defendants *221quoted prices that appeared to either cause or anticipate downward movement in the PM Fixing. TAC ¶¶ 268-74 & App'x. I. Plaintiffs also allege that Defendants employed manipulative trading tactics, like "spoofing," "wash sales," and "front-running" client orders, in order to further manipulate the price of physical gold and gold-denominated assets before and during the PM Fixing. TAC ¶ 10 & n.4. Notwithstanding the fact that UBS was not a member of the fixing panel at any point during the class period, Plaintiffs contend that UBS facilitated the conspiracy by using its large trading position to move the market in tandem with suppression of the Fix Price. See TAC ¶ 15 ("A single actor could not and would not have attempted to move the market so consistently. There would not have been enough 'ammo' to do so, and the risk (and cost) would have been too high."). Trading to create artificial downward momentum in the price of gold assisted the Fixing Banks by "altering the starting price, ..., and giving cover to an auction-rate that would otherwise have stood out like a sore thumb." TAC ¶ 10. Defendants supposedly benefited from this scheme by using their foreknowledge of the PM Fixing to make profitable trades in the physical gold and gold derivatives markets, trigger "stop-loss" orders, and time margin calls to accrue favorable cash flows. TAC ¶¶ 18, 237-39.

In Gold I , the Court denied the Fixing Banks' motion to dismiss and granted UBS's motion to dismiss. See 213 F.Supp.3d at 642, 682. The Court concluded that Plaintiffs plausibly alleged, "albeit barely," a conspiracy among the Fixing Banks to suppress the PM Fix Price, id. at 659-60, and that Plaintiffs had antitrust standing, id. at 653, 656-57.³ UBS, on the other hand, was not a member of the fixing panel during the class period, and Plaintiffs' statistical analyses did not connect UBS to suppression of the PM Fixing. UBS was included in the "bunching" analysis of spot market quotes described above,⁴ Second Amended Complaint ("SAC") (Dkt. 44) ¶¶ 202, 250-53, 255-58; see also SAC ¶¶ 261-67 & App'x I ("Defendants" quotes anticipated downward movement in prices around the PM Fixing, while other market participants quoted higher prices during the same period), but that analysis lumped the Defendants together, and it was unclear what the coefficient of variation was for UBS's quotes in particular. Likewise, while Plaintiffs claimed the Defendants were net short COMEX gold futures during the class period, SAC ¶¶ 171, 209-216, 228, 272-suggesting a motive to suppress gold prices-they did not specify which banks were short and by how much. See also SAC ¶¶ 201-08 (stating UBS had large positions in gold products during the class period, but with no indication whether UBS was long or short, or what percentage of those positions were held on behalf *222of clients). On certain days, identified in Appendix I to the SAC, UBS and one or more of the Fixing Banks were alleged to have offered spot quotes showing a reversion in prices at the time of the PM Fixing. See SAC 253-58 & App'x I. But it was impossible to tell whether these quotes reflect legal, parallel pricing behavior (e.g., matching a competitor) or something more nefarious. See Gold I , 213 F.Supp.3d at 660. As the Court has previously explained, "the data does not plausibly support an allegation that any particular bank was net short at any particular time." Gold I , 213 F.Supp.3d at 663. The charts in Appendix I were based on "all available data," meaning Plaintiffs cannot say how the quotes offered by UBS during these windows compare to the universe of market participants.⁵ See SAC ¶¶ 260, 262. In short, the Court wrote, Plaintiffs allegations that UBS quoted prices that were "lower than market averages" were "simply inadequate to create a plausible inference of conspiracy." Gold I , 213 F.Supp.3d at 663.

Plaintiffs filed the TAC to provide additional support for their claims against UBS. The centerpiece of the TAC are fragments of electronic chat messages, sixteen in all, between a precious metals trader at UBS and a precious metals trader at Deutsche Bank.⁶ TAC ¶¶ 358-73. Plaintiffs contend that these chat messages are evidence that "UBS understood, participated in and benefited from [ ] collusive activities" in the gold markets. TAC ¶ 11. At a minimum, according to Plaintiffs, they show a high degree of interfirm communications and are therefore circumstantial evidence of an antitrust conspiracy. See Pls.' Opp'n (Dkt. 301) at 13-14. The chats describe brazen efforts to manipulate the gold markets through coordinated trading. TAC ¶¶ 364 ("[UBS]: okay when gold pops 1430 ... we whack it ... u sell your 50k ... I sell my 20k ... then we double that up and produce our on [sic] liquidity too ... that should be enough to cap it on a holiday"), 366 ("[UBS] im gonna sell more silver and gold [Deutsche Bank]: k ... i really think we are on the right side today, being short"), 372 ("[UBS]: im buying gold [Deutsche Bank]: seems like we buy ... [UBS]: [we should] try to stay together today"); see also TAC ¶¶ 367, 369. UBS and Deutsche Bank also shared proprietary pricing information and coordinated quotes to customers.⁷ See TAC ¶¶ 368 (UBS and Deutsche Bank traders discussed "giv[ing] each other a heads up" when "boc," presumably a reference to the Bank of China, was buying gold), 369 ("[UBS]: how much in offers u got from 68-69 in gold? [Deutsche Bank]: 10k 69 ... 5k 68 [UBS]: just match me up 5k at 68 pls [Deutsche Bank] done"); see also TAC ¶¶ 370-71.

Three of the chats reference the Gold Fixing, but none references an agreement among UBS and the Fixing Banks to suppress *223gold prices.⁸ In March 2011, the Deutsche Bank trader told his counterpart at UBS that he would be participating in "the fix" when he was "there." TAC ¶ 360. The UBS trader responded that it was "not rocket science" but cautioned his friend that he had "seen fixings go real wrong before ... like-300k pnl [personal net loss]." TAC ¶ 360 The Deutsche Bank trader asked, "wrong side?" to which the UBS trader responded "nope basically bad timing ... push too early ... run out of ammo at the end." TAC ¶ 360 One month later, the Deutsche Bank trader said he had been "prop [proprietary] trading on the fix" and characterized "the fix," somewhat nonchalantly, as a "free option." TAC ¶ 361. In response, the UBS trader recounted the same story. Lastly, on May 11, 2011, the UBS trader told the trader at Deutsche Bank that "we smashed it [the fix] good." TAC ¶ 363.

The TAC also includes two new statistical analyses, which Plaintiffs contend support an inference UBS participated in a scheme to suppress the PM Fixing. According to Plaintiffs, UBS, on average, quoted below-market prices beginning in the ten minutes before the PM Fixing and continued to quote below market prices until immediately before the London market closed. See TAC ¶¶ 355-56. Plaintiffs also allege that "UBS's prices at the time of the PM Fixing fell in the bottom 5% and 10% for prices of the day far more often than they fell into the top 5% and 10%." TAC ¶ 357. This is evidence, according to Plaintiffs, that UBS's prices "were not in line with normal market expectations ...." TAC ¶ 357.

UBS has moved to dismiss the TAC on the grounds that the chat messages do not alter this Court's earlier conclusion that Plaintiffs have not plausibly alleged UBS was involved in a conspiracy with the Fixing Banks to suppress the Fix Price. As UBS points out, the Gold Fixing was a self-contained process, involving only the Fixing Banks. Because the Fixing Banks had complete control over the process, UBS's involvement was not necessary to the scheme and would not have been to the conspirators' benefit. UBS Mem. (Dkt. 298) at 12-13. The chat messages, UBS contends, are perhaps evidence of order manipulation or collusive trading-what UBS calls euphemistically " 'episodic' coordination," UBS Mem. at 2-but not of an agreement to suppress the fix: time-stamps on the chats, annoyingly omitted by Plaintiffs, show that the communications occurred primarily in the middle of the night London time and never close in time to the PM Fixing. See UBS Mem. at 16-17; see also Declaration of Eric J. Stock ("Stock Declr.") (Dkt. 299) Ex. 1 (chronological chart of chat messages cited in the TAC including time stamps and locations for the participants). Moreover, none of the chats references an agreement to manipulate the PM Fixing; the only references to a fixing describe incidents during which UBS admitted to "smashing" an unspecified fix and the story described above in which the UBS trader recounted (twice) a failed attempt to "push" an unspecified fix by an unspecified trader at an unidentified bank.⁹ See UBS Mem. at 17-18 (citing TAC ¶¶ 360-61). Hedging its bet, UBS also moved to dismiss on the grounds that Plaintiffs lack antitrust standing because *224UBS was not the proximate cause of their injury and that the Court lacks personal jurisdiction as to the Swiss entity UBS AG because the TAC does not allege any suit-related, in-forum conduct by UBS AG in particular.¹⁰

DISCUSSION

In evaluating a motion to dismiss, the Court must "accept all factual allegations in the complaint as true and draw all reasonable inferences in favor of the plaintiff." Meyer v. JinkoSolar Holdings Co., 761 F.3d 245, 249 (2d Cir. 2014) (quoting N.J. Carpenters Health Fund v. Royal Bank of Scotland Grp., PLC, 709 F.3d 109, 119 (2d Cir. 2013) ) (alterations omitted). Nonetheless, in order to "survive a motion to dismiss, a complaint must contain sufficient factual matter ... to 'state a claim to relief that is plausible on its face.' " Ashcroft v. Iqbal, 556 U.S. 662, 678, 129 S.Ct. 1937, 173 L.Ed.2d 868 (2009) (quoting Bell Atl. Corp. v. Twombly, 550 U.S. 544, 570, 127 S.Ct. 1955, 167 L.Ed.2d 929 (2007) ). "Plausibility" is not certainty; Iqbal does not require the complaint to allege "facts which can have no conceivable other explanation, no matter how improbable that explanation may be." Cohen v. S.A.C. Trading Corp., 711 F.3d 353, 360 (2d Cir. 2013). But "[f]actual allegations must be enough to raise a right to relief above the speculative level," Twombly , 550 U.S. at 555, 127 S.Ct. 1955, and "[courts] 'are not bound to accept as true a legal conclusion couched as a factual allegation,' " Brown v. Daikin Am. Inc., 756 F.3d 219, 225 (2d Cir. 2014) (quoting Twombly, 550 U.S. at 555, 127 S.Ct. 1955 ) (other internal quotations marks and citations omitted).

I. Sherman Act Claims¹¹

Horizontal price fixing is per se illegal. United States v. Socony-Vacuum Oil Co., 310 U.S. 150, 223-24, 60 S.Ct. 811, 84 L.Ed. 1129 (1940). Claims for bid rigging, on the other hand, typically involve competitors conspiring to raise prices for purchasers-often, but not always, government entities-who acquire products or services by soliciting competing bids. See, *225e.g. , Gatt Commcn's, Inc. v. PMC Assocs., LLC , 711 F.3d 68, 72-74 (2d Cir. 2013) ; State of N.Y. v. Hendrickson Bros. , 840 F.2d 1065, 1068-69 (2d Cir. 1988). With regard to unlawful restraints of trade, "[b]ecause § 1 of the Sherman Act does not prohibit [all] unreasonable restraints of trade ... but only restraints effected by a contract, combination, or conspiracy, ... [t]he crucial question is whether the challenged anticompetitive conduct stem[s] from independent decision or from an agreement, tacit or express." Twombly , 550 U.S. at 553, 127 S.Ct. 1955 (alterations in the original) (internal quotations and citations omitted). Regardless of whether Plaintiffs' allegations are evaluated in terms of price fixing, bid rigging or an unlawful restraint of trade, an unlawful agreement must be pleaded with respect to each antitrust claim brought under Section 1. See, e.g. , In re Elevator Antitrust Litig. , 502 F.3d 47, 50 (2d Cir. 2007) ("To survive a motion to dismiss ... a complaint must contain enough factual matter ... to suggest that an agreement ... was made.") (internal citations and quotations omitted).

To allege an unlawful agreement, Plaintiffs must allege either direct evidence (such as an express agreement among competitors to fix prices) or "circumstantial facts supporting the inference that a conspiracy existed." Mayor & City Council of Baltimore v. Citigroup, Inc. , 709 F.3d 129, 136 (2d Cir. 2013) (emphasis in original). Because conspiracies "nearly always must be proven through inferences that may fairly be drawn from the behavior of the alleged conspirators," the fact that Plaintiffs have no direct evidence does not mean there was no conspiracy. In re Foreign Exch. Benchmark Rates Antitrust Litig. , 74 F.Supp.3d 581, 591 (S.D.N.Y. 2015) (" FOREX I ") (quoting Anderson News, L.L.C. v. Am. Media, Inc. , 680 F.3d 162, 183 (2d Cir. 2012) ). A plaintiff may plausibly allege an antitrust conspiracy by alleging facts that evidence "conscious parallelism, when such interdependent conduct is accompanied by circumstantial evidence and plus factors," Mayor & City Council of Balt. , 709 F.3d at 136, such as "(1) 'a common motive to conspire'; (2) 'evidence that shows that the parallel acts were against the apparent individual economic self-interest of the alleged conspirators'; and (3) 'evidence of a high level of interfirm communications.' " Gelboim v. Bank of Am. Corp. , 823 F.3d 759, 781 (2d Cir. 2016) (quoting Mayor & City Council of Balt. , 709 F.3d at 136 ) (internal quotation marks and additional citations omitted). At the pleading stage, Plaintiffs "need not show that [their] allegations suggesting an agreement are more likely than not true or that they rule out the possibility of independent action ...." Id. at 781 (quoting Anderson News , 680 F.3d at 184 ). Instead, " 'a well-pleaded complaint may proceed even if ... actual proof of those facts is improbable, and ... a recovery is very remote and unlikely' as long as the complaint presents a plausible interpretation of wrongdoing." FOREX I , 74 F.Supp.3d at 591 (quoting Twombly , 550 U.S. at 556, 127 S.Ct. 1955 ) (emphasis in original); see also Gelboim , 823 F.3d at 781 ("At the pleading stage, a complaint claiming conspiracy, to be plausible, must plead 'enough factual matter (taken as true) to suggest that an agreement was made ....' " (quoting Anderson News , 680 F.3d at 184 ) ).

The TAC does not include any direct evidence that UBS was involved in a scheme to suppress the PM Fix Price. Three of the chat messages included in the TAC reference "fixes," but none describes a scheme among UBS and the Fixing Banks persistently to suppress the Fix Price. In messages on April 1, 2011, and May 1, 2011, the UBS trader described an occasion on which he observed a fixing "go *226real wrong" resulting in a personal net loss of "300k". TAC ¶¶ 360, 361. The Deutsche Bank trader responded by asking if the trader-whose identity is unspecified and whose employer is open to speculation-had been on the "wrong side," to which the UBS trader responded "nope," he had "push[ed] too early" and "run out of ammo." TAC ¶ 360. Even assuming the unidentified trader was employed by UBS, a single failed attempt at "pushing" the PM Fix is inconsistent with there being a conspiracy with the Fixing Banks to persistently suppress the PM Fix Price; given the Fixing Banks' complete control over the Fix Price, "pushing" the PM Fixing itself would be unnecessary. Moreover, the context of the discussion was to warn the Deutsche Bank trader that it is possible to lose money on the PM Fixing-it is "not always fun and g[a]mes." TAC ¶ 360. That warning suggests that the UBS trader did not have inside knowledge that the Fix Price was being controlled by a conspiracy. The other references to the "fix" describe proprietary trading that has no obvious connection to a fix-suppression conspiracy, and a time UBS "smashed" an unspecified "fix." TAC ¶¶ 361, 363. Like "pushing," "smashing" the PM Fix-assuming that is what UBS did-is inconsistent with a broader conspiracy involving the Fixing Banks. Neither of the chats indicates the traders acted in concert with others. The other chat messages describe coordinated trading and sharing of order flow information and proprietary pricing information, but they do not discuss manipulation of the PM Fixing. See TAC ¶¶ 364-73.

Plaintiffs do not dispute that the chat messages are not direct evidence of a price fixing conspiracy. Instead, according to Plaintiffs, the chat messages are a "plus factor" or circumstantial evidence of a conspiracy to be evaluated in the context of Plaintiffs' statistical analysis that shows anomalous pricing behavior around the PM Fixing. The Court does not find Plaintiffs' theory of parallel pricing and circumstantial evidence to be sufficient to nudge its allegations across the line to "plausible." In Gold I , the Court rejected Plaintiffs' argument that they had plausibly linked UBS to a conspiracy to suppress the PM Fix through circumstantial evidence of parallel pricing during and around the PM Fixing. See 213 F.Supp.3d at 678-79. "At best," and accepting Plaintiffs' group-pleading, the statistical analysis in the SAC was evidence that UBS quoted below-market prices at the same time as a downward reversion in prices occurred around the PM Fixing. Id. at 678. But unlike the Fixing Banks, UBS did not quote these prices while it was engaged in a private teleconference with the other alleged conspirators to set a benchmark price for gold. See id. at 662 ("[T]he Court finds significant Plaintiffs' allegations that, as a group, Defendants were apparently pushing gold prices down ... around the same time that they were sharing order information via the PM Fixing call on days when the PM Fixing ultimately resulted in a significant downward price swing."). As the Court explains below, the chat messages are an inadequate substitute for similar allegations to show that UBS shared relevant information with the Fixing Banks at a relevant time. And the TAC includes no factual allegations that provide context to suggest UBS was a member of the alleged price fixing conspiracy.

Plaintiffs' statistical analysis draws only a weak connection between UBS and suppression of the PM Fixing. With limited exceptions, Plaintiffs' analysis refers to "the Defendants" and does not distinguish between UBS and the Fixing Banks, as a group or individually. See TheECheck.com, LLC v. NEMC Fin. Servs. Grp., Inc. , No. 16-CV-8722 (PKC), 2017 WL 2627912, at *2 (S.D.N.Y. June 16, 2017) (dismissing claims that lumped defendants together and did not allege facts about the conduct *227of individual defendants); see also Sullivan v. Barclays PLC , No. 13-CV-2811 (PKC), 2017 WL 685570, at *24 (S.D.N.Y. Feb. 21, 2017) (rejecting "conclusory assertions that fail to distinguish between the conduct of the individual defendants"). Plaintiffs' analysis of net short COMEX futures positions-intended to show that UBS and the Fixing Banks were incentivized to suppress gold prices-refers generally to the "Defendant Banks." TAC ¶¶ 176-77; see also TAC ¶¶ 235-40 (alleging the "Bank Defendants" profited from fix-suppression through large positions in "Fix price-denominated derivatives," "digital options," and similar gold-denominated financial products). Plaintiffs have no evidence UBS specifically was short gold during the class period. Plaintiffs' coefficient of variation, or "bunching," analysis also refers to the "Defendants." See TAC ¶¶ 257-65. Because the analysis does not specify the coefficient of variation for any individual bank, the Court cannot determine whether UBS quoted "bunched" prices or whether variance in UBS's quotes is obscured by the group-wide analysis.¹² Those analyses that are UBS-specific show that UBS had a substantial position-long or short-in the gold markets, TAC ¶¶ 210-15, and that on specific days during the class period UBS quoted spot prices for physical gold that were in parallel with a downward trend in prices around the PM Fixing, see TAC ¶¶ 267-74 & App'x I. But UBS's status as a major market-maker for gold is not evidence that UBS was involved in a conspiracy to suppress the price of gold, and without information as to whether UBS specifically was long or short, the Court cannot infer a motive to suppress gold prices. The charts in Appendix I are as consistent with parallel, market-following behavior, or a legitimate expectation prices would fall, as they are with participation in a price-fixing scheme.¹³ To the extent the charts in Appendix I are probative, they are limited to approximately a dozen days over an 8-year class period.

The TAC adds two new, UBS-specific analyses. According to Plaintiffs, UBS quoted below market spot prices for physical gold throughout the class period, beginning shortly before the PM Fixing and persisting until shortly before the close of the London market at 4:30 p.m. London time. TAC ¶¶ 355-56. Normalizing the market price to 1.000, at the peak of the variance, shortly after the PM Fixing began, UBS quoted prices that were, on average 0.04% or 0.05% (4 or 5 basis points) below market.¹⁴ It is notable that this underpricing *228begins before the PM Fixing, an indication that UBS may have anticipated a drop in prices. But the analysis also appears to show that from approximately 2:50 p.m. to 4:25 p.m. (i.e., from 10 minutes before the beginning of the Fixing Call until five minutes before the London market closed, which is approximately 85 minutes after the Fix Price was announced), UBS was underpricing everyone, including the Fixing Banks, its alleged co-conspirators. Plaintiffs do not explain why UBS continued (on average) to quote below-market prices for more than an hour after the PM Fixing. Because the Fix Price was also the market price, Plaintiffs' analysis appears to show that UBS consistently quoted prices below the Fix Price in afternoon trading in London. A market participant with foreknowledge of the price to be set in the PM Fixing would be able to accurately predict the market price, not consistently undershoot it. Plaintiffs also present an analysis to show that UBS's prices around the time of the PM Fixing were more frequently in the bottom 5% to 10% of the market that day than they were in the top 5% or 10%. TAC ¶¶ 357. Plaintiffs do not specify why this fact supports their theory that UBS was involved in a fix-suppression conspiracy.¹⁵

The chat messages incorporated into the TAC do not meaningfully add to the mix of circumstantial evidence. Plaintiffs contend that these chat messages show a "high-degree of inter-firm communications," which is a recognized plus factor. See In re Currency Conversion Fee Antitrust Litig. , 773 F.Supp.2d 351, 369 (S.D.N.Y. 2011) ; Todd v. Exxon Corp. , 275 F.3d 191, 198 (2d Cir. 2001) ("Information exchange is an example of a facilitating practice that can help support an inference of a price-fixing agreement."). But analysis of inter-firm communications is not mechanical, and the probative value of such evidence depends on the participants, the information exchanged, and the context-specifically, the connection between the content and the price-fixing conspiracy alleged. For example, communications of pricing information among senior executives around the time of observed parallel pricing behavior tends to make it more likely that the parallel prices are the product of an unlawful agreement rather than legitimate, conscious parallelism. See In re Publ'n Paper Antitrust Litig. , 690 F.3d 51, 67 (2d Cir. 2012) ("[T]he causal link is presumed to be particularly strong when, as alleged here, the agreement is between executives at rival companies, each of whom has final pricing authority."). Information sharing among competitors without a legitimate purpose is also good evidence of a conspiracy. See In re Currency Conversion Fee Antitrust Litig. , 773 F.Supp.2d at 369 (interfirm communications were a "plus factor" where the defendant "failed to offer a compelling rationale for why it would disclose such information").

In some recent benchmark fixing cases in this district chat messages have been persuasive evidence of a conspiracy. In FOREX I , for example, the plaintiffs plausibly alleged a conspiracy to manipulate foreign exchange benchmark rates by alleging the defendants had participated in *229chat rooms with incriminating names like "The Cartel" and "One Team, One Dream." FOREX I , 74 F.Supp.3d at 587. The complaint included allegations that the chat room participants shared information about pricing and order flow information before the fixing, including the "types and volume of trades they planned to place," and agreed in advance to trading strategies to manipulate the foreign exchange fix. Id. The chats in FOREX I were evidence of information sharing close in time to the alleged manipulation and involved participants with direct responsibility for the benchmark submissions. Because the foreign exchange benchmarks were set through open market quotations, rather than a private fixing call, the chats were also evidence of the primary mechanism through which the conspiracy operated. Another recent benchmark fixing case built on chat messages included similar evidence of information sharing with direct relevance to the fix-setting process. See Sullivan , 2017 WL 685570, at *23-24 (identifying bilateral chat messages, primarily involving a trader at Deutsche Bank, as evidence of a broader conspiracy to manipulate the Euribor benchmark); see also Sonterra Capital Master Fund Ltd. v. Credit Suisse Corp. AG , 277 F.Supp.3d 521, 556 (S.D.N.Y. 2017) (" CHF LIBOR ") (concluding that chat messages were adequate to state conspiracy claim against the bank quoted in the chat messages); In re Libor-Based Fin. Instruments Antitrust Litig. , No. 11-MDL-2262 (NRB), 2015 WL 4634541, at *41 (S.D.N.Y. Aug. 4, 2015) (sustaining complaint where Plaintiff identified "sporadic" examples of rate manipulation).

Here the communications and their context do not suggest a fix-suppression conspiracy. While Plaintiffs unhelpfully excised the time-stamps and locations from the chat messages in the TAC, they have not disputed the accuracy of UBS's more complete reproductions.¹⁶ Without exception, the time stamps show that these communications occurred in the middle of the night London time. See Stock Declr. Ex. 1. None of the chat messages records sharing of pricing information around the time of the PM Fixing. See In re Flat Glass Antitrust Litig. , 385 F.3d 350, 369 (3d Cir. 2004) ("exchanges of information" that "are more tightly linked with concerted behavior" are "more purposive"). Both traders who participated in the chats were based in Singapore at the time of the messages, and Plaintiffs do not allege that the Deutsche Bank trader was a fix-submitter for Deutsche Bank at that time-although one chat suggests that he expected to have some involvement in the fix in the future when he would be in London. See TAC ¶ 360. Unlike in FOREX I and Sullivan , the chat messages in the TAC do not evidence sharing of information that would have been necessary to a fix-suppression scheme, or sharing of that information close in time to the fixing itself. See Sullivan , 2017 WL 685570, at *23-24 ; FOREX I , 74 F.Supp.3d at 587 ; see also In re Flat Glass Antitrust Litig. , 385 F.3d at 369 (affirming that inter-firm communications that could be used to implement collusive price increases are probative of a conspiracy); In re Currency Conversion Fee Antitrust Litig. , 773 F.Supp.2d at 369 (plus factor established through evidence *230of sharing information necessary to effectuate price-fixing conspiracy). That the PM Fixing was set through a daily private auction, rather than open market quotes-unlike the fixes in Sullivan and FOREX I -further diminishes the relevance of the chat messages to a fix-suppression conspiracy.

The Court also is not persuaded by Plaintiffs' characterization of certain of these chat messages as "discuss[ing] [ ] tactics and efforts to manipulate the PM Fixing," Pls.' Opp'n 13, or as evidence of UBS conspiring to suppress gold prices "around the PM Fixing," Pls.' Opp'n at 11. As discussed above, the conversations occurred "around" the PM Fixing only in the most general sense that the middle of the night is "around" 3 p.m. And the chats Plaintiffs describe as sharing "tactics" to manipulate the PM Fixing refer to an unidentified trader-not the UBS trader or the Deutsche Bank trader in the chats-who lost money attempting to "push" the PM Fix price. See TAC ¶¶ 360, 361. The chat does not indicate in which direction the trader attempted to push the market (prices can be "pushed" up or down). The fact that the trader was unsuccessful in his attempt to push the PM Fix Price suggests he was not a part of a conspiracy involving the Fixing Banks. Likewise, the context of this anecdote was a warning from the UBS trader to his confederate at Deutsche Bank that the PM Fixing "was not always fun in [sic] g[a]mes," i.e., there was risk involved. TAC ¶ 360.

The rest of the chats describe coordinated trading and sharing of order flow information that is either inapposite to or inconsistent with a fix-suppression scheme. On April 10, 2011, the UBS trader told the Deutsche Bank trader to "push it higher." TAC ¶ 367; Stock Declr. Ex. 1 ¶ 6. On June 28, 2011, the Deutsche Bank trader asked if "gold should be higher then?" TAC ¶ 362; Stock Declr. Ex. 1 ¶ 9. And on July 26, 2011, the Deutsche Bank trader said, "i [sic] really think we are on the right side today, being short." TAC ¶ 366. Attempting to "push" gold higher is inconsistent with a conspiracy to suppress the Fix price and could, in fact, have made it more difficult to profit from foreknowledge of the Fix Price. See In re Zinc Antitrust Litig. , 155 F.Supp.3d 337, 345 (S.D.N.Y. 2016) (granting dismissal when alleged conspirators would be counter to the goals of the alleged conspiracy). The Deutsche Bank trader's uncertainty regarding prices is also inconsistent with an insider's knowledge of a price-fixing conspiracy. As Plaintiffs appear to concede, the relevance of chat messages showing market manipulation and order flow sharing depends on an inference that if these two traders were willing to share information and collude in Singapore, then they or others at UBS and Deutsche Bank must have also been sharing information in London leading up to and during the PM Fixing. That is not quite a "if it happened here, it could've happened there" argument, see In re Elevator Antitrust Litig. , 502 F.3d 47, 52 (2d Cir. 2007), but it is very close. Cf. Sullivan , 2017 WL 685570, at *25 (S.D.N.Y. 2017) (recognizing that "horizontal activity to fix the price of Euribor-based derivatives on a transaction-by-transaction basis" does not necessarily "overlap" with the "fixing of the Euribor" benchmark rate); CHF LIBOR , 277 F.Supp.3d at 546 n.11 (rejecting inference of an overarching conspiracy to manipulate markets in Swiss-denominated LIBOR because "a group of defendants could have agreed to fix bid-ask spreads regardless of the CHF LIBOR rate, and vice versa, and there is no indication that the two conspiracies were part of one interwoven plot, as opposed to two separate sets of misconduct allegedly committed by the same entities."). Allegations that two traders at Deutsche Bank and UBS episodically coordinated positions and engaged *231in market manipulation in Singapore does not make it plausible that they or others employed by the same banks engaged in a much larger, much more ambitious multi-year conspiracy with four other financial institutions to persistently suppress the PM Fixing in London.

Plaintiffs do not allege facts that support other "plus factors" as to UBS. In Gold I , the Court rejected Plaintiffs' reliance on a FINMA report from 2015 which described manipulation of precious metals markets by UBS traders. See Gold I , 213 F.Supp.3d at 661-62, 678 ("FINMA's findings that UBS shared order information with 'third parties' and engaged in front-running and other conduct against its clients' interests, does not support Plaintiffs' allegation that UBS conspired with the Fixing Banks (or others) to manipulate the Gold Fixing."). The FINMA report does not mention the PM Fixing or gold in particular. Plaintiffs also rely on the fact that the Deutsche Bank trader involved in the chat messages has since pleaded guilty to market manipulation, and that a U.S.-based UBS gold trader has been charged with spoofing and market manipulation. But neither proceeding ties UBS to a price fixing conspiracy, as Plaintiffs' concede. See Pls.' Opp'n at 18 (arguing that these proceedings are "circumstantial evidence that plausibly suggest[s] that Defendants' traders engaged in collusive conduct affecting the gold market.").

The Court has evaluated Plaintiffs' allegations as to UBS as a whole and considered their "combined character and effect," and has done so in the light most favorable to Plaintiffs. In re Zinc Antitrust Litig. , 155 F.Supp.3d at 373. Nevertheless, the TAC fails to allege a plausible link between UBS and the price fixing scheme alleged against the Fixing Banks.¹⁷ Plaintiffs' Sherman Act claims against UBS are dismissed.

II. CEA claims

Section 9(a)(2) of the CEA makes it unlawful for "any person to manipulate or attempt to manipulate the price of any commodity in interstate commerce." 7 U.S.C. § 13(a)(2). There are four elements to a manipulation claim. See In re London Silver Fixing, Ltd., Antitrust Litig. , 213 F.Supp.3d 530, 566 (S.D.N.Y. 2016) (" Silver I "). "Plaintiffs must allege that: (1) Defendants possessed an ability to influence market prices; (2) an artificial price existed; (3) Defendants caused the artificial prices; and (4) Defendants specifically intended to cause the artificial price.' " Id. (quoting In re Amaranth Nat. Gas Commodities Litig. , 730 F.3d 170, 173 (2d Cir. 2013) ) (additional citations omitted). Market manipulation claims sounding in fraud must be pleaded with particularity in accordance with Rule 9(b) of the Federal Rules of Civil Procedure. See In re LIBOR-Based Fin. Instruments Antitrust Litig. , 935 F.Supp.2d 666, 713 (S.D.N.Y. 2013) ; Silver I , 213 F.Supp.3d at 565.

The parties agree that Plaintiffs' CEA claims depend on the same theory of a conspiracy as Plaintiffs' Sherman Act claims. See UBS Mem. at 34 ("the CEA claim against UBS fails for the same reasons the Section 1 claim fails.") (emphasis in original); Pls.' Opp'n at 24 ("Plaintiffs have alleged plausible Section 1 claims, as explained above, and the CEA claims are viable for largely the same reasons."). Accordingly, the Court's conclusion that *232Plaintiffs' Sherman Act claims against UBS are implausible applies equally to Plaintiffs' CEA claims. Plaintiffs' CEA claims are dismissed as to UBS.¹⁸

III. Unjust Enrichment

For the reasons stated in Gold I , Plaintiffs' unjust enrichment claim is DISMISSED.

IV. Leave to Amend

Under Rule 15(a) of the Federal Rules of Civil Procedure, "[t]he court should freely give leave" to a party to amend its complaint "when justice so requires." Fed. R. Civ. P. 15(a)(2). "Leave may be denied 'for good reason, including futility, bad faith, undue delay, or undue prejudice to the opposing party.' " TechnoMarine SA v. Giftports, Inc. , 758 F.3d 493, 505 (2d Cir. 2014) (quoting McCarthy v. Dun & Bradstreet Corp. , 482 F.3d 184, 200 (2d Cir. 2007) ) (additional citation omitted). Ultimately, "the grant or denial of an opportunity to amend is within the discretion of the District Court." Foman v. Davis, 371 U.S. 178, 182, 83 S.Ct. 227, 9 L.Ed.2d 222 (1962).

Plaintiffs have not requested leave to amend, and they have not attached a proposed, fourth amended complaint for the Court's review. Given that Plaintiffs have already amended three times, including based on discovery from Deutsche Bank, and that Plaintiffs have not requested leave to amend, the Court denies leave to amend. Plaintiffs are represented by competent, experienced counsel. If they had the facts necessary to plug the obvious holes that exist in the TAC, the Court is confident those facts would have been included in the pleadings filed to date.

CONCLUSION

UBS's motion to dismiss is GRANTED. Plaintiffs' claims against UBS are DISMISSED WITH PREJUDICE. The Clerk of the Court is directed to close the open motion at docket entry 297 and terminate defendants UBS Securities LLC and UBS AG.

The remaining parties are directed to appear for a status conference with the Court at 11:00 a.m. on August 24, 2018 . By August 17, 2018 , the parties must submit a joint letter of not more than 5 pages setting forth a proposed schedule for discovery in this action.¹⁹ The parties are forewarned that the Court will not accept dueling letters; the parties are required to work together to produce a joint letter.

SO ORDERED.

1 The Third Amended Complaint names what appear to be the parent corporations of each of the Fixing Banks. Also named are U.S.-based affiliates of each, including ScotiaMocatta Depository (BNS), Barclays Capital Inc. (Barclays), Deutsche Bank Securities, Inc. (DB), HSBC Securities (USA) Inc. and HSBC Bank USA (both, HSBC), and Newedge USA, LLC (SocGen).

Deutsche Bank has settled the claims in this case, as is discussed further below.

2 Gold exchange-traded funds invest solely in gold bullion and issue shares that are directly linked to spot gold prices and can be traded via exchange. TAC ¶¶ 109-10. The Court has dismissed Plaintiffs' claims with respect to gold ETFs and gold ETF options.

3 The Court granted the motion as to the period from 2000 through 2005. Plaintiffs' supplemented their allegations as to 2004 and 2005 in the TAC, and the Fixing Banks have not renewed their motion to dismiss as to those years.

4 UBS was not included in Plaintiffs' original "underpricing" analysis, which purports to show that the Fixing Banks quoted below market prices on days they allegedly manipulated the PM Fixing. See TAC ¶¶ 262-64. The fact that UBS was not included in that analysis is at odds with Plaintiffs' claim that UBS's quotes were moving in "relative unison" with the Fixing Banks on days the PM Fix Price marked a reversion in the market. See TAC ¶¶ 257-61. Plaintiffs present a version of this underpricing analysis as to UBS using different metrics at TAC ¶¶ 355-56. The underpricing analysis at TAC ¶¶ 262-64 is measured in terms of absolute price, whereas the underpricing analysis at TAC ¶¶ 355-56 is normalized to a market price of 1.000. The difference in metrics makes comparison difficult.

5 The TAC also included allegations based on an enforcement proceeding against UBS instituted by the Swiss regulator FINMA. See TAC ¶¶ 303-11. FINMA's report primarily concerns manipulation of the foreign exchange markets. The report also states, however, that FINMA uncovered evidence that UBS traders manipulated precious metals benchmarks and engaged in manipulative trading in the precious metals markets. TAC ¶¶ 309-10. The FINMA report does not specify whether the Gold Fixing was included in this alleged manipulation.

6 The Deutsche Bank trader involved has since pleaded guilty to wire fraud and "spoofing" charges arising out of manipulation of the precious metals markets, including the gold futures market. See Dkt. 261 Ex. A.

7 On January 29, 2018, the CFTC sanctioned UBS for manipulative trading in the precious metals markets, based in part on chat messages between the same traders as are quoted in the TAC. See Dkt. 304 Ex. A at 4-6.

8 A third chat references the "dip" at 4 p.m. See TAC ¶ 362. Although Plaintiffs present this as a reference to the PM Fixing, the chat message does not specify a time zone. The Court doubts that the traders were referring to the PM Fixing; the PM Fixing occurred at 3:00 p.m. in London, which was 11:00 p.m. in Singapore, where the traders were based at the time.

9 Context suggests that this story involved a trader at a Fixing Bank.

10 UBS's motion presents alternative arguments on the assumption that Plaintiffs intend to bring claims for a conspiracy to manipulate the gold markets more generally on an episodic basis through coordinated trading and tactics such as spoofing, wash sales, and front-running. See UBS Mem. at 22-28 (Plaintiffs lack antitrust standing to bring such claims), 28-29 (Foreign Trade Antitrust Improvements Act, 15 U.S.C. § 6a, bars Plaintiffs' claims), 31-37 (Plaintiffs' CEA claims are untimely and impermissibly extra-territorial, and Plaintiffs lack CEA standing). Plaintiffs have disavowed this theory. See Pls.' Opp'n at 2 ("UBS relies on the false premise that Plaintiffs have shifted the theory of their case."), 9 (characterizing Plaintiffs' claims as a "Fix-suppression conspiracy theory"), 31 ("such a theory does not exist"). Limiting Plaintiffs' allegations to a fix-suppression theory narrows the questions presented by this motion to whether Plaintiffs' claims are plausible and whether the Court has personal jurisdiction over UBS AG. The Court addressed the balance of Defendants' arguments (with the exception of whether Plaintiffs' CEA claims are extra-territorial), as they relate to a fix-suppression theory, in Gold I .

11 Under the circumstances, the Court exercises its discretion to address UBS's motion to dismiss for failure to state a claim before turning to personal jurisdiction. See Sullivan v. Barclays PLC , No. 13-CV-2811 (PKC), 2017 WL 685570, at *11 (S.D.N.Y. Feb. 21, 2017) ("In cases such as this one with multiple defendants-over some of whom the court indisputably has personal jurisdiction-in which all defendants collectively challenge the legal sufficiency of the plaintiff's cause of action, we may address first the facial challenge to the underlying cause of action and, if we dismiss the claim in its entirety, decline to address the personal jurisdictional claims made by some defendants." (quoting Chevron Corp. v. Naranjo , 667 F.3d 232, 247 n.17 (2d Cir. 2012) ) ). The Court need not address personal jurisdiction because Plaintiffs' claims fail on their merits.

12 Plaintiffs do not dispute that they have access to the raw data necessary to conduct the same analysis on a defendant-by-defendant basis. TAC ¶ 257. Thus, this is not a case in which the information necessary to distinguish the actions of the individual defendants is uniquely within the defendants' control. Cf. In re Foreign Exchange Benchmark Rates Antitrust Litig. , No. 13-CV-7789 (LGS), 2016 WL 5108131, at *20 (S.D.N.Y. Sept. 20, 2016) (excusing Plaintiffs' failure to plead information concerning the "specific dates, times, currency pairs and customers discussed" in chat messages between traders because that information was "exclusively in Defendants' control").

13 Reviewing the charts is an inexact science because Plaintiffs have not provided data on the variance of each quote from the market trend-line. As far as the Court can tell, the charts do not show that UBS quoted prices that were at a greater variance from the trend line than other market participants. It is also unclear whether the charts are based on a complete universe of quotes from the spot market for physical gold or a limited subset of market participants.

14 There is no information in the TAC as to how this variance compares to other market participants. It is unlikely that any market participant would quote prices perfectly in line with the average market price, even averaged over many years. Nonetheless, it is interesting that UBS's prices moved downward immediately before the PM Fixing.

15 Insofar as the PM Fix Price more often marked a downward trend in prices, it seems logical that quotes from UBS (and apparently every other market maker) around the PM Fixing would be more likely to be in the bottom 10% of quotes for the day than the top 10%. Plaintiffs initially presented a version of this analysis as evidence of anomalous pricing around the time of the PM Fixing. The Court agrees that this analysis, currently included at TAC ¶¶ 139-42, is evidence of a downward reversion in prices at the time of the PM Fixing. But Plaintiffs leave unexplained why quoting prices consistent with a market-wide trend is evidence of participation in a fix-suppression conspiracy.

16 Because the chat messages are incorporated into the TAC, the Court may consider the complete messages as provided by the Non-Fixing Banks. See In re Nokia Corp. Sec. Litig. , No. 96-CV-3752 (DC), 1998 WL 150963, at *7 n.4 (S.D.N.Y. April 1, 1998) ("The Second Circuit has held that courts may consider the full contents of documents partially quoted in the complaint where the documents are integral to the complaint." (citing San Leandro Emergency Med. Grp. Profit Sharing Plan v. Philip Morris Cos., Inc. , 75 F.3d 801, 809 (2d Cir. 1996) ) ).

17 Because Plaintiffs have failed to plausibly allege a conspiracy the Court does not consider UBS's alternative argument that Plaintiffs lack antitrust standing. As noted supra note 10, Plaintiffs do not allege a conspiracy to episodically manipulate the gold markets. Accordingly, the Court does not consider whether such a claim is plausibly alleged or whether Plaintiffs would have antitrust standing to pursue such a claim.

18 As noted supra note 10, Plaintiffs have disavowed an episodic market-manipulation theory, in favor of their fix-suppression theory. Accordingly, the Court need not address whether Plaintiffs could state a CEA claim for episodic manipulation. The Court also does not address UBS's argument that Plaintiffs' CEA fix-suppression claims are untimely and extraterritorial as to UBS.

19 The parties are encouraged to coordinate discovery with the parallel silver fixing case pending before the Court.